University of Ghana http://ugspace.ug.edu.gh
THE DETECTION AND MOLECULAR 
CHARACTERISATION OF 
BACTERIAL SYMBIONTS OF 
ANOPHELES GAMBIAE S.L. 
by 
CHARLES ADDOQUAYE BROWN 
This thesis is submitted to the University of Ghana 
in partial fulfilment of the requirements for the award of 
Ph.D Biochemistry degree. 
SEPTEMBER, 2003 . 
University of Ghana http://ugspace.ug.edu.gh
DECLARATION 
This thetis is the ~suh of research work undertaken by CharleJ A. Brown in the 
Department of Bk>chemiSlry. Uni ... mity of Gtwla. under the supervision of Prof. M.D. 
Wilson and Prof. F. N. Gyang. 
(C harlesA.Brown) 
~ I 
(Prof. M.D. Wilson) 
University of Ghana http://ugspace.ug.edu.gh
DEDICATION 
This thesIS IS dcdic:ared to Joaa, for aU thai she went throusJ1 to enable me complete: II, 
C110ch and alJ theNii "Atians". 
University of Ghana http://ugspace.ug.edu.gh
ACKNOWLEDGEMENTS 
1 am glad 10 IYve this opponuaily of exprcsiin& my deepest gratilude to my supervisors. 
Prof. M.D. Wilson and Prof. F.N. Gyana. who provided paidance and encouragement 
tbrou&bout the: period of thi:; work. neir invaluable commcolS greatly helped to improve 
lb. quo1ily or .... work Spec.a! 1lwIk. go to Dr D.A. Boakye (Noguc:hi), Prof. T. 
Unnuc::b and Dr Tariq Higazl (both of the Unh'CfSllY of Alabama at Binningham, USA) 
for tbeir divene contributions. 
My sincere thanks goes to Nkem Okoyc, Mrs. Anitli Ghansah. Mrs Bridgette-Marian 
Ogoe, Nancy Duah. Janet Midega. Mrs Benedicta Kuivi. Fred Aboagye--Antwi, S. Dadzie, 
Enns Glah, Helena Baidoo. Naiki Puplampu, Awo Osafo-Addo, F.e. Mills-Robertson. 
Hury A>mah. Shelly, Tol., L)'Ii. . Abena, BB IIId MIs S .... Ad. .A mankwab whose: 
support kept me Fin&- I am equally grateful to Mr Nana M. A. Appawu. Dr K. 
Boaompan IOd Dr. K. Koram for their encouragement. A word of appreciation goa to 
all manbcrs of the PansilOIogy Unit (NMIMR) and lecturer.; of the BiochemiJtty 
Department for their oontinucd interest and encouragement. 
Finally, but by no means the least, I thank my ... cry good friendl Hasl80 H855m. David 
Mmab,. Akosua Boafo. 5.B. orea, Hetty and Esi Colecraft fOf their CDCOUI8.JCIl1COI, moral 
support and prayersdunng timcsoffrustration. 
God Bias You All. 
flus work wa, supported in .,.n with a grant from the UNDPlWorld BanklWHO Special 
Pmgnmmc for Rcscan:h and Training in Tropical Diseases (TOR) 
University of Ghana http://ugspace.ug.edu.gh
ABSTRACT 
The aim of the prcseot study was to identify and characterise bacterial symbionts in 
An. g"m/JiM s.l. mosquitoes. and ultimately select one that occurs in all its life 
stases. Mosquito larvae and pupae samples were collected &om six locations in the 
GlUIer Accra region of GbaAa and some reared 10 adults in the laboratory. Wild Mlult 
All, ,iUllbUle mosquitoes from Navronso and Dodowa (Ghana) and Jaribuni (Kilifi 
District, Kenya) and laboratory colonies of adult An. gambiae mosquitoes from 
KihmanJaro (Tanzania), Suakoto (Ubcria) and Kisumu (Kenya) and An. lU'abknsu 
(Wqcningen SU'ain) were alw studied. Each specimen was surface sterilized before 
DNA extraction was carried out The peR method of Scott el al. (1993) was used 
on each 1p«lmcn to determine the species oflhe An. gambiae complex. PCRs usin, 
univawl cubactcria 16S and 2JS rONA oligonucleolide pnmers were first used to 
detect the presence of the microorganism's DNA sequences in lhe mosquito. then peRs 
using WOLB 16SF liWOLB16SR I aJXJ /tsZI !ft~Z1 primers were carried out on posItive 
reactions to detamine wbecber or not they were Wolbachia sp. /" sllico 
(computational mole-cular biology) analyses of DNA sequences of Esderlclria coli 
and PallloQ(" agglo",uatu were perfonned WlinS DIGEST software and the results 
compared with those obtained by restriction analysis of the amplified 16S and 23S 
rONAl with six enzymes. The peR products were cloned and sequenced using an 
ABI )77 automated sequencer. Two conJCnsus DNA scqucm:n were generated from 
Che sequence data by pajrwise sequence similarity and the phylogram methods. 
Pbyloametic relatIOnships of the consensus sequcnca to homololOus acquenc:cs in (he 
databases were inferred using nelghbour·joining (NJ). muimum likelihood (Ml), and 
maximum pvsimony (MP) methods. Of Ihe 4)2 molquilO spa:imcns studied, 37) 
(94.4°·.) WC1'C idC1'lhfied IS An. gambiae s.s., 20 (5.''''0) U All. lU"dbiellJis. and one 
(0.2SV.) each of All. ",enu and All. melas_ 295 An. gambiae (74.7".) adults. 29 (7.3%) 
University of Ghana http://ugspace.ug.edu.gh
pupae and 71 (18%) larvae were studied. DNA fragmenls of Ihe predicted sizes 
were successfully amplified usiPl the 16S rONA md 23S rONA primers in 85.1% 
(3361395) and 795% (314/395) of the specimens. respeaively. Bactcnal DNA sequences 
were amplified from aU the sibling species. which consisted of 54 (76.1 %) larvae. 26 
(89.70/.) pup« and 256 (86.r...-) adults from both wild and laboratory reared specimens, 
lIT~vcofthe geographical origin. Out of281 specimens, peR positives for 16S and 
23S rONA primers. 94 (33.S%) were positive (or the WOLl6S rONA primers but none 
for the ft;sZ primers. Failure of PCR with /t3Z primers indicated _ possible absence of 
Wolbachia infection. The restriction studies revealed that none of the amplified peR 
products could be either E. col; or P. agglom~raflS. Seven sequences. namely AgAl 
(AY24716S), AgA2 (AYJ2S810), AsU (AY247160), AgL2 (AY247161), AgLJ 
(AY247162), AgL4 (AY24716J) and AgLS (AY247164) were obtained, four and two 
with high homology to each other and one, a stand alone. CONSEN4 and CONSEN2 
were hishly homoloSOUS to 16S rONA sequences of Par/JCOCCUJ and Rhodobact~r 
species. The sequence AgLS was highly homologous to the 16S rONA sequences of the 
Cylophaga~Flexibacter~Bacteroidcs group (CFB) group. The phylogenetic trees 
constructed. indicated that CONSEN4 and CONSEN2 probably belong to the a 
subdivision of the Proccobeclcria whilst AgLS is _ member of the CFB group. Since 
PIllDCOCC14 and RJtodoboct~r 11'. are Gram-negalive aquuic bacteria and members of the 
CFB group are _]so aqualic bacteria. it illikcly Ihal they occur in the breeding habitats of 
DX*Iuitocs. Further research is nccdcd; (I) 10 ddenninc if lhcy are in the breeding 
habitats and are inaatcd by larvae. (2) 10 cultuR and isolale them. and (3) 10 decermme 
which part of the adult IlU*plitothcy reside in. 
University of Ghana http://ugspace.ug.edu.gh
TABLE OF CONTENTS 
DECiARAT ION 
DEDICATION 
AKNOWJ.EDGEMENTS 
ABSTRACT 
TABLE OF CONTENTS vii 
LIST OF TABLES x; 
LIST OF FIGURES ]l;ii 
ABBREVIA nONS 
C'IIAPTER ONE 
1.0INTRotn'(TION 
1.1 Objectives JO 
CHAPTER TWO 
LITERATURE REVIEW II 
21 MosquitoCl II 
2.1.1 GeneraJintroduction 
2.1.2 Classificationandidcntification II 
2.1.3 Spccicscomplael 16 
2.1.4 MethodJ for diltinpishing between cryptic species 17 
2.1.-4.1 'Traditional tools' 18 
i) Morpbologica1identification 18 
Ii) Rcproductiveincompalibility 19 
iii) Cytogcnelicsludies 20 
iv)CuticuJarhydrocarbonanalysis 21 
v) Allozymcfisozytne anaI)"is 22 
vi) Hybridizationauays 23 
2.1.4.2 Classicalgeoeticmarkers 24 
i) Restriction &agment length polymorphisms (RfLPs) 24 
ii) MicochondrialDNAanalysil 25 
iii) The polymcrue chaia reaction IIpplied to nuclear DNA 26 
(nONA) 
2.1.4.3 Hishly polymorphiC marker.; 28 
i) Miaosatdlite DNAs 28 
;;) RlUIdomly ..p hficd polymorphic DNA (RAPOs) 30 
2.I.S Distribulionandccology JI 
2.1.6 Mc:dicalimportance 32 
2.2 The Genus Anopheles 34 
2.2.1 The life cycle of Anopheles mosquitoes 14 
2.2.2 ARopIwlrs species of medical importance 40 
2.J 1bc """"MIn gaM_ Gila Complex. 
University of Ghana http://ugspace.ug.edu.gh
2.3.1 DistnbutlOn and importance or members of the Anopltelu 41 
gaINbUJe sibling species complex 
2.3.1.1 Anopheles lnwnrtbae 47 
2.3.1.2 AnopltelesqlladriQlfnlltahlS 49 
2.3.1.3 Anopltele3lMlas 50 
2.3.1.4 AftOplwks m~1i45 51 
2.3.1.5 ArtOpIre/~s orab"~1lSis 52 
2.3.1.6 Al'tOf'Iwles gombioe sensu stricto 
2.4 Symbiotic Associations 58 
2.4.1 Type$ ofsymbiOlic associations 59 
2.4.1.1 Phoresis 59 
2.4.1.2 Commensalism 59 
2.4.1.3 MutuaJism 60 
2.4.1.4 Parasitism 60 
2.4.2 Modes of bow symbionts get together 60 
2.4.3 Batterialsymbionts 62 
2.5 Molecular Genetic Analysis 64 
2.S.1 Extraction and purifiCilion of genomic DNA 64 
2.5.2 Restrictionendonuclcasc:s 66 
2.5.3 Restriction n.1l"''''' Imath pol)1Jlorphism (RFLP) 68 
2.5.4 AgaroscgeJclcclrophorcsis 70 
2.S.S Determination of ONA  frlgment sizes 73 
2.5.6 Pol)1Jlcras< Choin R_on (PCR) 74 
2.6 MolecularPhylopneticAnaiysis 80 
2.6.1 Molcculorphylogmymclltods 81 
2.6.1.IPhylogenctict:rec:!. 81 
2.6.1.2 Searchalgoritluns 83 
2.6.1.3 Heuristics 85 
2.6.2 Tree eonsttuctioa methods 86 
2.6.2.1 Discretccharactcrmctbods 87 
i) Muimum ~imony (MP) and weishled par!>imony (WP) 87 
ii) Maximwn likclihood (ML) 89 
2.6.2.2 Distancematrixmethods 90 
i) Transformation of sequence data to distances 90 
ii) Unweightcd pUr group method with arithmetic mean 92 
(UPGMA) 
2.6.2.3 Trwnsfonncddistancemethods 92 
i) Neighbourrdalionmethods 93 
ii) Fi,ch ond M_liuh (FM) method 94 
iii) Distance Wagner method (OW) 94 
iv) Minimwn evolution (ME) 95 
University of Ghana http://ugspace.ug.edu.gh
2_6_3 AccountIng for superimposed events (nucleotide substitution in 9S 
a DNA sequence) 
1.6.4 AJ,scssing the reliabilityofa tree (confidence in phylogenetic 97 
estimate<;) 
CIIAPTER THREE 99 
MATERIALS AND METHODS 99 
3.1 Chemicals. Reaacnts, Eql. .. pmenl and Software 99 
3.2 Biological Specimeol aod Sample Collection 99 
3.3 I..aboraWry Rearing of Mosquitoes 103 
3." IdcntificationofAnophelesspecies lOS 
3.4.1. Morpholoaica1 identification lOS 
3.4.2. Molecular identification of sibling species oftbe AnopIwlu lOS 
gambiae complex 
3.4.2.1 Genomic ONAexlr8Ction 107 
3.4.2.2 peR amplification 107 
3.4.2.3 AnaiysisofPCRproducts 108 
3.S Dctectioo of Bacterial Symbionts Using PCR 109 
3_5.1 EWmationoftheooncentrationofPCRproduCIS 112 
3.6 Amplified Ribosomal DNA Restriction Analysis (ARDRA) 113 
3.6.lln$ilk:o(computItionalmolecularbiology)rcsbictionanalysis 113 
3.6.2 Restriction endonuclease digestion and analysis 114 
3.7 CIonin& of Amplified Bactaial Sequences 115 
3.7.1 dAtoilinaofPCRprodueu 116 
3.7.2 Ligation 116 
3.7.3 Transfonnationexperiments 117 
3.7.4 Plasmid DNA ilOlltion 118 
3.8 IdcntificatiOllof Amplified Bactaial Symbiont Sequences and 120 
Phy~cAnalysis 
3.8.1 Sequencing ofamplificd bacterial DNA sequences 120 
3.8.1.1 Scq""""cditing 120 
3.8.1.2 VecSerccnandehimcradetcction 122 
3.8.1.3 ScquencesimiiariryandDNAdarabasesc811:hcs 123 
123 
124 
128 
128 
129 
SequcnccAa:cssionNumbcrs 130 
III 
Identification III 
University of Ghana http://ugspace.ug.edu.gh
4.2 PCR Dc:tectioa of Bactcri. DNA5Cqueoces inAn. galllbiM s.J 13S 
4.3 Amplified Ribosomal DNA Restriction AnaJysis (ARDRA) 142 
4.4-.1Scqucoc:<sinl6S,DNA 152 
4.4. 1. AsAl (GenBank acccssion number AY24716S) 152 
4.4.2 AgA.2 (GenBank acc"";on number AY3258 10) 152 
4.4.) AgLl (Get!Bank accession nwn"'" AY247160) 162 
4.4.4 AgL2 (GaJ.Bank accession number AY247161) 162 
4.4.5 AgLJ (GenBank ac~C5$lon nwnb« AY247162) 162 
4.4.6 AgL4(GcnBanltocccssionnwnberAY247163) 162 
4.4.7 AsL5(Gc:n8&nk accession nwnberAY247164) 163 
4.' Comparative Analysis of the 16S rONA Sequences 164 
4.5.1 Consensussequence(CONSEN4) 161 
' .S.2 Con. ................... (CONSEN2) 168 
4.6 IdcntifiCMion of Amplified Baderial Symbiom Scquc:nca Ind 171 
PhylogcnelicAnalysis 
4.6.1 Identification of homoJoaous sequences in DNA 01&1 171 
rcpo5llones. 
4.6.2PhylogoDcbcanalysi' 172 
4.6.2 .1 Phy~c""ysi.ofCONSEN4 172 
4.6.2.2 Ph~""ysis ofCONSEN2 In 
·Ul.2.1 Phylogmeti~ analysis of AgL5 J8 2 
CHAPTER FIVE 187 
DISCUSSION A~D £.:ONCLUSfONS 187 
51 0iscusslon 
S.2 ConclusioRi 197 
REFERENCES 198 
APPENDICES 241 
University of Ghana http://ugspace.ug.edu.gh
LIST OF TABLES 
Table 1 .• Anoplrelaspccics of medical importance. 
41 
Tablel.2 Malaria vectors of the world. 
42 
Table 3 •• Sequence details and melting lem~ (Tm) of 106 
oligonucleotide primcn used for the peR identification of the 
An. gambfDe species complex (Scott ~I 01 .• 1993). 
Details of oligonucleotide primer sequences UIOd for peR 110 
detection of blctaial symbiona in An. gambiae ..I . and lheir 
mellin, tempcralurcs 
s.ctai. species and strains used for lhe phylogcoetic analysis 125 
ofCONSEN4. 
TableJ.3b Bacteria specie:sand strains used forlhe phytogeneticauiysi5 126 
ofCONSEN2 
Table3.3c Bacteria species and strains used for the phyloJCflctic analysis 127 
ofAaLS· 
Table 4.1 Distribution of 5peCics identified among the different life stage 1).4 
formJ of AIt. gGIIfbiM complex species using the peR method 
of Scott er a/. (1993). 
Table 4.2 peR amplification of bacteria 16S and 235 rONA gene 136 
sequences in All. glUllbitJtlLI. mosquitoes. 
Table 4.3 PCR amplification of Wo/bachill jtsZ and 16S rONA gene 141 
sequences in spccimcm of An. gambitU 1.1. mosquito. 
Observed and expected fragment sizes for amplioona. E. coli 145 
and P. agg/omrroM rDNAs after diption with the various 
rcstric:tionenrymes 
Overall similarities between lhe bacterial 165 rRNA sequences 165 
obtained. 
Table 4.6 Similarity index. of CONSEN4 and homologous 16S rONA 
lOQuences o(close1y related organisms retrieved from DNA 
da!abucs. 
Table 4.7 Similarity index of CONSEN2 and homologous 16S rONA 178 
sequences o(closely related organi.snw retrieved from DNA 
dotabases 
T.bIe 4.8 Simil.nly index of AgLS and homologous 16S rONA 183 
~ucnces of closely relaled organisms retrieved &om the DNA 
databases 
University of Ghana http://ugspace.ug.edu.gh
LIST OF FIGURES 
Fi~.2.1 0lII1 showing the main differences ~ecn ~uilOeS oftbe 
13 
subfamilies Anophelinae (left) and Cuhclnae (nghl). 
Fig.!.2 An anopheline wing which is identified by its dW8Cleristic 35 
pattern. 
36 
I;i~. 2.3 uCe cycle of AMpltel£f mosquitoes. 
Fi,.2.4 Map of the malaria cpidemiologi~ zones (after MacDonald. 44 
1957). 
Flo. 2.5 Schematic diagnm of the peR process. 76 
FiB·l.la MOIquHo collection site at Adenta (open drain with stagnant 100 
water exposed to sunlight). 
F'K- 3.lb Mosquito collection site at Madina (slow HowinS pollulcd 100 
.m"""from ........... in&houses). 
,iiK·3.lc Moequito collection site at Acrumola (n4lTOw stretch of water 101 
flowing from a broken water pipe), 
FiC·3.1d Mo-.uito collection site a' Dzorwulu (small shallow pool of 101 
stapanlwalcr). 
Flg.3.la Plastic basins used for rearing mosquito larvae. 104 
Fi&-l.lb Cages used for the rearing of pupae and the mamtenance of 104 
emerged ~uh mosquitoes 
FiK-3.3 Mapoftbc linc:arizcd vC'CtorpCR~.1 121 
Fi14.1 Eth.dlUnt bromide--slained 0.8 -I. agarose gel electrophoregram 132 
of genomic DNA extracted from An. gambiae 5,1. mosquitoes. 
FiK4.2 Ethidium bromidc-staincd 2.0 % agarose gel electrophorqram 133 
of DNA ...... produced by the rONA-peR identification 
method for members of the An. gambiae complex. 
Fix·4.J Elhidiwn bromide-stained 1.0% _prose gel e1ectrophorcgram 137 
of amplified bKtaia 165 rONA sequences &om mosquito 
specimens using rD2IrP2 pnmers 
10-12· .. •. . Elhidium bromide-stained 1.0% ~ gel electrophorcgram 138 
of amplified bacteria 23S rONA sequcoces from mosquito 
.,..."... ..... WAIIWA3primc:rs. 
Fil-O Ethtdium bromide--ttained 1.0% agaroee gd eIcctropoorcgram 139 
of amplified bacteria 16S rONA ICIqUI:nI.:eS from mosquito 
specimens usiq. WOLl6S primen. 
t-·ia· 4.6 Elhidi~ bromide-stained 1.0% agarose gd dedrophorcgnm 140 
ofamphficd bac:teria 16S rONA scqucnc(..""S ulingjbZpnmers 
..1 t,.47 Ethidiwn bromide-stained 2.0% uprose ael elcctrophorcgrmn 143 
of rcstnchon enzyme digests of 165 rONA peR products. 
F1g.U Elhldium bronude-ltaincd 2.0% IIplOse FI clcctrophoregram 144 
University of Ghana http://ugspace.ug.edu.gh
of restriction enzyme dlgC'lb of 23S rONA PCR products. 
Fig. 4.9 Ethidium bromide-stained 2.0% .garoee ael elec:trophoregram 
141 
of HindiU restriction enzyme digest! of WOLl6S rONA PCR 
products. 
Ethidium bromide-stained 2.0% .prose gel elec:trophoregram 148 Fig. 4.10 
of Apol restriction enzyme digests of WOLl6S rONA PCR 
products. 
Fig. 4.11 Ethidiwn bromide-stained 2.0% agarose gel electrophoregram 
of R.tG1 and Nspl restriction enzyme digests ofWOL16S rONA 
PCRprodUCU. 
Fig. 4.12 Ethidiwn bromide-stamed 2.0% .garosc gel elc:ctrophorqram 150 
of !,'coRl restriction enzyme digests of WOLl6S rONA peR 
prod ..... 
Fig. 4.13 Ethidium bromide-slaincd 2.00.4 aprose gel elc:ctrophoregram 151 
of BamHI and Hi"jI restriction enzyme digests of WOL 16S 
rDNAPCRproducts. 
Fig. 4.14 Ethidium bromide-stained 1.0% aprn-.c: gel elcctrophoregram 153 
showiDgtbepresencelabscnceofinserts(-IOOObp). 
Fig.4.IS Ethidiwn bromide-stained 0.8% agarose gel electrophoregram 
ofilOlatcd plumid DNAs. 
Fig. 4.16 Details of bactenal DNA sequence AgA I (GenBank accession 155 
number AY247165). 
F~4.17 Details ofbactaial DNA sequence AgA2 (GenBank accession 156 
number A Y32581 O. 
Fig. 4.18 Details ofbKteriaJ DNA sequt.'ncc: AgLl (GenBank Kccssion 157 
number AY247 160) 
Fig. 4.19 Details of bacterial DNA sequence AgL2 (GenBank accession 158 
number AY247161). 
Fig. 4.20 Details of bKtaiai DNA ~ucnce AgL3 (GenBank accession IS. 
number AY247162). 
FiJI. 4.21 Details of bacterial ONA JCqucncc AglA (GenBank accession 
numbc<AY247163). 
Fig. 4.22 Details of bacterial DNA sequence AgL5 (GcnBank lCCCSSion 161 
nwnbcrAY247164) 
FIg. 4.23 Examples of dot plot paiT¥llse sequence comparisons between 166 
thebacterialsequencesobtamed. 
1'i~.4.24 U_pby\o&aletic_construcIed bylhe neighbour- 167 
JOuling method. sbowmg the phylogenetic reJationWpI of 1bc: 
bactcnal6SrDNAsequences. 
""ig.4.2S Consensus scqucnoc (CONSEN4) generated from AsA2, 16~1 
AgLl,AsJ.2.andAgAl. 
~1c-4.26 C..-..., sequence (CONSEN2) gcncn'cd from Ag1.3 It1d 
~\  
"'''''-
University of Ghana http://ugspace.ug.edu.gh
AglA. 
fl3. 4.Z7 Phylol!J8fl1 (ne>ghbour-joining) oflhe CONSEN4 and selected 
174 
membcn of the Q_Proleobacteria infared from 16S rONA 
sequcnceoompansons. 
Fig",2R PhytoglWll (maxlffiwn pllS1mony) of the CON5EN4 and 
175 
selected members of the Q-Proteobacteria inferred from 16S 
rONA scqucncc comparisons. 
FIs-4.Z9 Phytogram (maximum likelibood) of the CONSEN4 and 176 
selected members of the a-Proteobacteria inferred from 165 
rONA sequence oompari!OOS. 
fl3.4.JO Phylo. .... (oeighbout-joining) of the CONSEN2 and selected 179 
membcn of the a_Pro/~cteT;a inferred from 16S rONA 
sequenceoompansons 
Fie. 4.31 Phytognun (maximum parsimony) of the CONSEN2 and 180 
selected members of the a-Proleobacteria inferred from 16S 
rONA sequence comparisons 
Fig. 4.32 Phytogram (maximum likelihood) of the CONSEN2 and 181 
selected members of the a-Proteobaclerla infelTctI from 16S 
rONA sequeoce comparisons. 
F1c.4.33 Phytosram (neighbour-joining) of AgLS and .mainly. selected 184 
memben of the CFB group infemd from 165 rONA sequence 
compc;tonS 
fl" 4.34 Phylogram (m&Ximwn parsimony) of AgL5 and, mainly, 185 
xlccted membcn of the CFB group inferred from 16S rONA 
scquencccompan5OfU. 
Fig.4.JS Phylogram (maximum hkelihood) of AgL5 and, mainly, 186 
selected members of the CFB group inferred from 165 rONA 
scquc:noec:ompll13OflS. 
University of Ghana http://ugspace.ug.edu.gh
ABBREVIATIONS 
bp b.:lScpalt"S 
dATP dcoxyadenosine triphosphate 
dCTP deoxycytidinetriphosphate 
de-ionizeddistilledwlter 
dGTP dcoxyguanosinetriphosphate 
DNA dcoxyribonucleicacld 
dNTP dcoxyribonw::leotidephosphate 
cbDNA double-strandcdDNA 
.nTP dc:oxythymidinetripoosphlte 
EDTA etbylenediaminetetraacetale. 2HlO 
EtRr ethidiumbrotnide 
ethanol 
GPS g1obalpositioningsyslcm. 
HBI hwnanbloodindex 
H,O 
kb kilobasc 
KOH potISSiumhydroxide 
l.B Luria Bcrtani broth 
molar (moles per litre) 
"I miaolitre 
I'M micromolar 
ML Maximumlikc:lihood 
millilitre 
Maximum panlmony 
mRNA messcngcrRNA 
Mw molccularwClghl 
N.OH sodium hydroxIde 
Neighbour-JoIning 
openllonalwooomic urul \0 
University of Ghana http://ugspace.ug.edu.gh
PAUP Phylogenetic AnalYSIS USing Parsimony 
PeR polymerase cham reaction 
pll hydrogen-ion exponent 
PHYllP pHYlogeny lnference Package 
rONA ribosomal DNA 
RNA ribonucleic ecid 
nbonuclease 
revolutionsperminutc 
ribosomal RNA 
sddHlO stenle double distilled water 
5.1. $~nsulalo 
ssDNA single-stramiod DNA 
TAE Tris·ac:eta1eEDTA 
mchingtemperature 
2 -amino-2-(hydroxymethyl)-t,3 propanediol 
University of Ghana http://ugspace.ug.edu.gh
If you would see 01/ of Nature gathered up at one point, in all of her 
Iotleliness. and her skill, and her deadliness, and her sex, where would you 
lind a more exqUfSf te symbol than the Mosquito? 
. HAVE LOC K tiLlS, 1920 
L_ ____ _ 
University of Ghana http://ugspace.ug.edu.gh
CHAPTER ONE 
INTRODUCTION 
Vector-borne diseases are major causc!> uf death and morbidity worldwide. Not only do 
they craie publtc:: health problems. they also preSl.'nt serious obstacles for socia-economic 
developmc:ntin tropicaJ countries. 
Of the disease-causing H:Clors, mosquitoes transmit some of the world's most life-
threatenin& and debilitating parasitic and viral diseases. including malaria. Venezuelan 
equint' encephalitis. yellow fever. filariasis and dengue/dengue haemorrhagic fever 
(Miller. 1992; Priest. 1992; Monath. 1994). Of these, malaria is by far the most pn:\'alcnt 
uopicaIVCClor-bomcdisease. 
Malaria. the vector of which is the female Anophe/e$ mosquito, is estimated to rep~SC'nt 
2.)% of the overall global disease burden and 9% of Africa's; il ranks third after 
pnaunoc:tXCaiacutere!ipiratory tract infections (3.5o/e) and lubcreulosis(2.8".) (WHO. 
1999). An estimated 2.3 billion people. almost oncHhird of the world's population. are at 
risk of infection with the malaria parasite (WHO, 1999). The annUIII incidence of the 
dlscase is estimated to be 300- SOO million clinical cases with mortality accounting for 
between 1·3 million deaths among children under five years. In sub.Saharan Africa it 
accounlS for more than ~/. of the total global incidence: and mortality (Breman, 2001; 
W}10fUNICEF.2003). h is a major cause of dealh in pn=gnant women (Lindsay el aI. • 
2000) and also an imponant calIX of sti ll births and low binh weights (Menendez, 1995; 
Stcketee ~, ai. • 2001). These figures are probably undereS1imates a' the statistics used in 
their colkction may vary by a fxwr of tJuec dependi"K on the method of estimation 
(WIIO. 1999). The 21 million n:ported cases of malaria in Africa an: believed 10 
University of Ghana http://ugspace.ug.edu.gh
represent only 5-10% of the total malaria incidence on the continent (Hamoudi & Sachs. 
1999). 
Globally. the nwnbcrs of malaria cases ace increasing and the rate of increase is 
accelerntina. This pattern is illustrated by multi fold increases in malaria rales since 1979 
in Soulh America (Roberts el aI.. 1997) accompanied by a rise in the proportion of 
populations at high risk to the disease. Malaria is nrpearing in urban areas and in 
rountries lhal had previously succeeded in eradicating the disease. for example:. in urben 
arasofthe Amazon Basin (Roberts & laughlin. 1(99), South and North Korea (Feighner 
el at.. 1998). ,\nnerua. Azerbaijan. and Tajikistan (Curtis, 1999). In Africa. the 
pn:vaIencc or malaria has been escalating al an alarming rate. within the last decade. 
Between 1994 and 1999.ma1ariaepidemicsin 14 sub-Saharan Africa countries caused an 
unacceptable high number of deaths. many in areas previously free of the disease (OAU. 
1997). Additionally. the increase in cases and the altered geographic distribution of 
malaria is underestimated beausc: acCUnltc information on global incidence is difficult to 
obWn, since trplrts arc gcncnlly fragmentary and in'Cgular (Roberts el 01 •• 2000). 
Air travel bas brought the threat of the disease to the doorsteps of industrialised countries 
with an increasing incidence of imported cases and deaths of visitors to endemic areas. 
IleaJth authorities in many countries are bewming mucasingly concerned about the 
poICntiaily dc8dIy risks of malaria auricd inaothcirterrito!')' by "jet-setting" mosquitoes 
that bavcl on international Righls and spread the disease (Gratz et aI., 2000). Between 
1969 and t 999. 12 counbies reported a total of 81 cases of m.laria in people livins near 
• airport. France headcd the Jig, with 26ca'iC5. foliowedby Belgium, 16. and the United 
Klngdum. 14 cases (WHO. 2(){K)). TIlt occurrmcc of lhe relatively large number of cases 
of airport malaria in Pwis and Brussels reflects the large number of Rights arriving from 
University of Ghana http://ugspace.ug.edu.gh
Central and WcstAfrica. lbe3e-airportmalaria" CIlSCS, occurring in or near airports. are 
distinguished from other ca5eS of imported malaria among persons who contJ'Kt the 
infection during a stay in a malarious area :md ~\lhscqucntJy faU ill (Lusina el aI., 2000; 
Gratz el aI., 2000). The DlIjority of the cases were caused by P. /o/clptrUM. At least five 
deaths ha\"C rcsuhcd; all cateS occurred among non-immune individuals, accounting for • 
relatively hish mortality of 6% (Gratz et aI., 2000). 
The prevalence of malaria bas an enormous impact on a country's economy 
incapKitaIinR part of the labour force and thus affecting productivity. The number of 
DAL Ys (disability adjusted life YCIIS) lost in 2000 to malaria has been estimated to be 42. 
280. 000 (WHOIUNICEF. 2003). II is also estimaled thal a bout of malaria. depending OD 
the severity, typic:aIly entails a loss of four or more ",orking days, followed by additional 
days with reduced work capacity (Shepard tlal., 1991; Picard &. Mills. 1992; Hempel A 
Najera, 1996). In Africa, where malaria IK:Counts for up to • third of all bospital 
admissions. this costs. sum equivalent 10 over 10 working dlys, adding to the continent's 
economic burden. Malaria attacks are also • major cause of school abJenteeism and 
appear to rteplive!y impact on long tenn leaming capacity (McLlonald, 1950; 
Wemsdorfer &: McGregor, 1')8)(; Holding & Snow, 2001). Endemic malaria also reduces 
the growth potentiaJ for some industries. notably lOw-ism and transportacion. and sharply 
raises the cost ofinfrutructure projec:ts and Olhcr coll«tive enlcrpriscs(Gallup& s.:hs, 
2001; s.m. & Ma1aaey. 2002). Some raeaR:hers have eslimated the economic burden 
of malaria .. 0.6-1.0% of GOP in Africa. although recent reports indialle tbal the 
economic impact of the disease on national income is likely to be much hi&her (Branan, 
20(H; Sachs & Malaney. 2002). The economic loss due 10 malaria in Africa is in excess 
of two bilhon llS dollan per year (Okenu. 1999). Malaria thus has ~iaI consequences 
and isaheavyburdcnoneronomicJe\'dopment 
University of Ghana http://ugspace.ug.edu.gh
o;_~ "" .. __ lOconuoI malari .. ThoCOOlrOI methods include lilt 
usc of inicctic::idts. chemotBcnpy and rNftIIICIIlCnt of.he environment for more th&n 10 
years, ins«tK:ides. most notably dictdorodiphcnyltricbJorocthane (DOn. have been onc 
of the prim.")' ~ of controlling imect-bome diseases_ However. resistance to 
imectK:~s has appeared in the major insect vectors from evcr)' genus (Brogdon &: 
McAllister. 1998). According 10 the World Health Organization approximuely 125 
Irtbropod species an: resistant to at least one, and often two or more, insecticides (WHO. 
1992). Rcsiscance has also developed to every chaaic:aJ class of ins.«cicide. includina 
microbiAl dn"gs and iMCCt growth regulators (Brogdon &: McAllister, 1998). In many 
parts of lhc wortd where insect-borne di:IeUCS cause illness and death. insecticides are 
3\"ailable. However. the panial wjlhdnwal of available effective ituecticida due to 
resiSWlCe in target species, IDd the non-renewal or the registration or SOrM insecticides 
(l..aecy • Orr. 1994), coupled with environmcnlal safety issues and the rugh cost 
a.xia&ed willi heavy iuecticide usage (Gcorghiou.. 1986), make sustainin8 &ona term 
tmccticidc wmor control. extrcmdy CWIIy aDd pmcticaUy UNChic"aWe. 
Drup 10 tre.t malari. have heen used ror thousands of years and WHO and govenunents 
of malarial c:ounmes have turned IOward dru& treatment strategies 10 reduce malarial 
incidence rates. Owing the carly 19005. doc&on used quirune for rnaIariIi thenp:y. The 
use ofthr drug was short· lived and wa, rq:Uced wilh chloroquine, a che.p. sate, and 
effectiw anti·m.JariaJ th8t gMned wideIprcad Kcepwx:e among doclOrs in eodtmk: 
countries durins the t9SOs and f'e"$ulleci in an enormous decline ill maa.;. incidenc:c rales 
(WHO. 1997). But.fter years of usc. becau.sc it 'A'ti cilhcr misused or abused. 
chloroquine-resistant mata.;. ...... ba\"C cvoh'Cd (NIAID, 2000). From Ihe 19505 10 
the present., c:hl~uinc resdIMcc .. ptu,aJ1)" spmtd 10 nearly a.l1/okipanun malaria 
endmttc: tqtons (WIIO, 1997; NJAID, 2000). 
University of Ghana http://ugspace.ug.edu.gh
Newer drug therapin. unfortunllfely. have not been effective .pinst drug resistant stnUDS 
of the malaria paras.ite. The drug mcfloquillC' was introduced in ~ Asit in the 
mid.I98Os. bul complete drug resistance was observed only four years thereafter (Ba5U. 
2000) and is DOw almost aalhe same level as chloroquine (Nchinda. 1998). Resistance 10 
a more recent drug ato\'aquone developed SO quickJy that resistant strains were detce-Ied 
during clinical trials (Strobel, 1999). Artemisinin and its derivatives (artemcther. 
arttet.her. and artesunate). which were developed as a result of !he more serious dNa 
rWstaoce situation in Asia. are currendy the most effective of all anti-malarial drugs. 
However, it bas been shown in laboratory studies that malaria parasites can also become 
resismnt to artemisinin (MAID. 2000). Moreover. according to scientists studying drug 
~slstance.themalari"purasile frequently mUlate3ancicanlhtreforeevolveresislanc:elo 
".arly any drug (Basu, 2000) 
The ~ a1tached to disease control, especially of P. /alciparum malaria.. for 
example, bas moUvalCd retc::an:b ineo the development of efficient vaccines. Vac:cinalion 
apinst P. /oJciparum and P. vivar is seen as the mcthod of intervention with the ~atest 
poteoti&l to tnlu.;e the morbidity and mortality associated with severe malaria in areas of 
intense tnnsmiSliion (Miller a. Hoffman. 1998) and even contribute to eradicating the 
disea:(Greenwood,I997). Forthesc rea5OnS. considerahle effort and severa.l hundred 
million dollan have been ~J'l'tnt 0\'1:1" the last )0 yean 10 develop vaccines (Collins & 
l·a~k.ew11Z, 1995). UnfmtutLaldy. YJlCcine devek»pment is associated with mumrous 
Jlllicultio: such as antigenic divCr.iIlY and immune evasion of the paruite<f: difficulties 
for the ~scalc production of clinical grade rnatcria1 for human tnah, lad: of good In 
vitro com:lates of protection; .., IKk of ldequate or proven delivery systems or 
adjuvants for usc in bwnans (Alonso &: Daedgc. 1999). A lthough YKc;:ine retcan:h 
University of Ghana http://ugspace.ug.edu.gh
mnaini a high priority. it is also important to look for alternative approaches to control 
vtttor-bome tropical diseases like malaria. while vu('cine research continues. 
Significant progress has been made in the research and de,,"elopment of new tools for 
malaria control. Inseeticide treated bcdnets (lTN) and curtains have emerged in recent 
years as one promising 1001. Several studies in 6 African countries hive demonstrated the 
efficacy of ITN in reducing significantly morbidity and mortality in infants (AlOJllO el aI •• 
1991; Nevill., aJ .• 1996; Bioka el 01 .• 1996; Lengeler a: Snow, 1996; Abdull. el 01 .• 
2001). Hownrer, the costs of the nets and treatment still inhibit wide~scale use. In 
tddition. further work remains 10 be done on different insecticide-fabric:: combinations and 
it is not yet kno\\n what perc~ntagc of the beds in a community need to be equipped with 
nets to achieve effective malari. control (NIAID, 2000). There is also the lhreat of 
pbysiologicalandbehavioural~stanc::etopyrethroids(usedforimpregnalin8thenets) 
by mosquitoes. of increase in severediseasc in areas ofinlense transmission due to the 
reduction ill Ir1InImission ad the possibility that (he iniliai successes of bednell may 
disappmr dwinA IonS tenn application. because of the fading of pre-existing immWlity 
I. .... (NIA1D. 2000) 
To date, the oaly effective control programmes for mal"ria hIIvc been those that targeted 
the moJquilO vector (TDR. 2001). Funhcrmon=, sine\!' the vector remains Ihc key link in 
the transmission of malaria, it warrants a new and serious research effort. In this reprd. 
rttCJV. bioccchnological advances in the area of host-parasite relationships and \'ector 
grne1ic::s suucst the possibility of n=\'Olulionary methods for malaria control (TOR, 
2001). The uhimue .un of these biotechnological approaches i. to ute recomI:riDlllt 
DNA technology tn replace a highly competl!'nl vector population wilh an othern..  ise 
identical plpUialion enllineen:d 10 be an incompmible basi for the malarial pant5ilc 
University of Ghana http://ugspace.ug.edu.gh
(Crampton~' QI .• 1990, 1994; Collins & Besansky, 1994). Thus mosquitoes would exiSi 
in their oormaJ environment, but be unable to transmit malana. This approach is rughly 
desirable because it does not seck 10 exterminate.' mosquitoes, as 5UCh • step would be 
I.:,hr.i c:dl~ va)' difficult and the ecological consequences profound, as mosquitoes arc a 
~1t:lllfi':.Inllink in the food chain (TOR. 2001). 
Several basic problems remain (0 be soh'cd before such • biotechnology stnltegy ean be 
clearly impkmented. One such problem is finding ways of inuoducing Ile" or altered 
genes into the mosquito genome. In principle this can be achieved either by the use of 
transposable genetic: elements (Kidwell & Ribeiro, 1992; ('oates f!1 aI., 1998), viral 
transfcctioo (c.ruon d uJ.. 1995; Olson el 01., 1996) or by genetic manipulation or 
bacteria s)mhiOflIJ (Beard el al., 19938: Duravasula et al., 1997). The problem of 
bansfcctionswlthexoticvirusesisthatnoefficiententomovirusiscl.lITefItlyavailabJeand 
oooe has beeo economic.Uy produced in sufficient quantity to control dixue vectors 
(ColliruandPaskcwiv_I995). Efforts to transfonn anophelines have been pursued since 
the first rqx>ned cae of foreign DNA introduction into Aft gombiae genome in 1987 
(Miller ~I aI., 1987). These were intensified after the !UCCeSSful development of routine: 
u.nsfonnalion teetw.iques usina the Mino$ lransposable element in the Mediterranean 
fruitfly CUalil" cop4la1Q (Loukeris ~I aJ., 1995) and the Morlner and Nermes elements in 
Aer/u tU1JIPI/ (Coeles ~I aI., 1998; Juinskicne ~I uf.. 1998). The ability of the MiltO$ 
traasposabJc eLemcnt to fwlc1ionasau-ansform2tion vector in anopbelim-mosquitoc:s has 
n:cmlly bttn demonstrated (Canerucia ~I aI. . 2000) 
'The potential of using N[ur3Uy occurring symhiotic bacteria that can be geocltcally 
erlginecn.-d (or ~ control of mosquitoes., on the other hand. has re-c::ei~ much less 
artenlJOn. \\"uh thiS 3fJprur;x:h. the arthropod is not transformed. but the symbi~ bacteria 
------------------------------------------~8 
University of Ghana http://ugspace.ug.edu.gh
tha' it harbours are (Beard elol., 1993.). Such uthropods an: called paratraJ\SgeDic:. This 
approach is guided by the (o!towing observations and basic concepts: 1) Throughout the 
cnlin.! devck>pmenaal cycle many arthropods. especially tOOK thai feed on restricted food 
sources such as blood. cellulose. phloem and stored grains. harboW' bacterial symbionts; 
2) Some of these symbionts can be cultured and genetically transformed to express 8 gene 
whose product kills a pathogen that the arthropod transmits; 1) Normal arthropod 
symbtoota CaD be replaced with genetically modified symbionts, resuhing in a population 
of arthropod 'oUtOfl that can no longer transmit diseasc (13card f!laJ .. 1993a). 
A wKk variety of t.cleri.l symbionts have been idenlified in practicaJly every <>rder of 
thclnsectaandinticks. In those insects lhat feed on f]uid m41lcrial such as plantjuioesor 
baood. the symbionts appear to provide metabolic products that supplement the host 
Wec:t'. outritioo (Beard et aI., 1993.). Exampln of such symbionts include WolhDchia 
In drosophiHds Ind p3taSilic ""-asps (O'Neil tl aI .• 1992; Stouthamer el 01.. 1993). 
Rickmsia-like orpnisms (RiO) from IsclSC nies Glossl"o spp.(Beard tl al., I 993b) and 
ladybttdes (WcrTen el 01. 1994) and Rhodococcu.r rhodnii in the vector of Chaps 
di9C*te. Rhodnius prolixus (Beard dol. . 1993a). for mosquilOeS. Wolbochio has been 
reported in Ihe Aslan tiler mosquilO, Aedes olboplclU$ (Graig el al., J 994) and in Culex 
plplnu (O'NeiI el aI., 1992). Ho-.-.ner, there have been very few reports of bacteria 
symbionts in Culicids and there is no reason for symbionls not to occur in An. gambior 
If symbiotic bacteria occur in An. gambiae. then it becomes possible 10 exploit them as 
vchtcies for the introduction of foreign genes tNt can express anti-Plmmodlum proteins. 
t'1te feasibility of using transgenk cndus~ mbionts 10 alter the disease carrying ~ity of 
• d.isea9t wctor has beta demonstrated in It pro/aus (Duravasula el 01 .• 1997). 
Morc:owr,usinc bKl.c:ti.IlbIthavcspecialisedand5fX'(iflCsymbtot,cl.5.SOCiationwith 
insect hosts 10 spread ~ greatly mtuca the chance of unwanted gene spl'Qd. 
University of Ghana http://ugspace.ug.edu.gh
Ikfol"C' lhe development of assays utilizing mo~ular techniques. the quantiUltive 
docwnentation of the presence of bacteria in blood-sucking arthropods such IS lic;ks, 
mites and mosquitoes. depended upon specific slaining techniqucsand serologK:a1-based 
assays. However. biological staining may be unreliable whilst serological.based assays 
require species-specific antisera (Higgins & Azad. 1995). Moreover, very few symbionts 
have been cultured outside their host in cell free systems. thereby preventing even a 
nditional bacterial taxonomic placement (O'Neil, 1997), though in a few cues. 
symbionts have ICtUaJly been isolated and grown in artificial culture medium (Welbum ('I 
ul.. 1987: W.1bum &; Dal •• 1997). 
Atthoush &he flSlidious nature of symbionts makes their isolation and in vilw cultivation 
diffICult, the advent of polymerase chain reaction (peR) has greatly aided efforts to 
characterise these symbionts at the molecular lC\el, and also 10 monitor changes that 
occur in vector populations over a period. Ribosomal gene primers specific for WO!. 
pIpi~tl/i.J haw: been used to detect lhis symbiont in insects belonging to several different 
Orden (O'Neil el al.. 1992). Roussel el at. (1992) described the use of 16S and 23S 
ribosomal RNA (rRNA) primers to amplify segments of these gcnes from Wolbochiu in 
terrestriaJ isopods, IS ...., "ell a..s insects. There arc several published oligonucleotide primer 
seb for amplif'ying sequences of Ri~kClIsiae-)lkt: symbionts (O'Neil et aI., 1992; Roussel 
rl uf. 1992a.; Higgins and Azad, 1995) and Eubacteria (Wcisburg elol. . 1991) in several 
insccls. These can be used 10 screen initially for symbionts in An. gambl«. For 
Idcntification purposes and 10 investigatetbe phylogenetic rclalionshipsoflhescbacleria, 
lhe sequences of the peR amplified DNA fragments can be compared to those lhal have 
been published ant.! depo .. iled in DNA databanks !luch as the GENBANK. Such 
phy1ogcnc1ic shidlCC'i can ousist in dcvclopins successful culture conditions in v;Iro as well 
as in Kkntifying .. tJlI.;shk- pla."imid transformation \'eClors. 
University of Ghana http://ugspace.ug.edu.gh
1.1 Objectives 
The uJbmate goal of this study is to identify a microbial symhiont that is universal in All 
gamb;~ s.l. and which can potentially be Benetically manipulated to express anti-
PlaIIItOdilllllproteinsmadultmosquitoes. rhespet;ificobjectivc~are: 
I To use published oligonucleotide primer sequences to amplify by peR, the DNA 
sequences of eubac:teria and rickettsiae-like organisms (RLO) of An. gumbiae. 
2 To confirm the identicy of the symbiont's amplified DNA sequence using 
restriction analysis. 
3. To done and sequence the amplified DNA fragments. 
4. To idcalify the species of the symbionts by conductinB bio-informatic search for 
homologous sequences in gene database banks 
5 To determine lhe symbiont', phylogenetic relationships with related orgllflisms. 
University of Ghana http://ugspace.ug.edu.gh
CHAPTER 2 
LITERATURE REVIEW 
2.1 Mosquitoes 
2.1.1 General int roductio n 
Mosquitoes arc perhaps the muse Familiar of all blood·sucking insects. 'They are of 
tremendous significance to man from both the economic and health point of vteW, 
bc.:a~ of their role as para..  lte vectors . 1ney include the only organisms able to transmit 
human malaria. and, apart from carrying this and other diseases. are almost unrivalled as 
initatingbillngpests 
Aristotle was the first to chronicle mosquitoes, which he rt:fernd to as "cropis" in. his 
"Historia Animalwn" in 300B.C in which he documented their life cycle and 
metamorphic abilities (Floore. 2000). Although it has bc-en c:Iaimcd that the name 
""kt!oquito" is eitha a Spanish or. Portuguese word meaning "little fly", Floore (2001) 
postulltes thai lbe. word is ~lly North American, and dates back to about 1583. In 
Europe. mosquitoes were caJled -musketas" by the Spanish. "gnats" by the British. "Lcs 
moucherons" or"Les cousins" by the French, and "Stc:chmuc:ken" or "Schnaekc" by the 
2.1.2 ClOlSsificatlon and identification 
\.1·' ..... 1IJil,~ belong to the order Diplera. sub-order Nemaloccrca ofttle class Insecta wbi<:b 
" 1M most dominant group of the Phylum Arthropoda. Mosquitoes are readily 
discin~uished from ot11<.., similar-lookmg fli~ in the sub-order (Nematocerca) by their 
con:,plclIUUS forwardl) prujttting proboscis, scaJC$ on the thorax, legs. abdomen and 
"'In~ \CUb, and fringe or tealcs alona the posterior margin of the "ings (Service. \91)). 
KCllk. 1(95), AU mosquilOc:s b..:long to the famil) Culicidae. which is divided into three 
,---------------------------------------------.12 -
University of Ghana http://ugspace.ug.edu.gh
sub--families Cuilcinae, Anophdiue aod Toxorhynchites. and JS genera (While. 1996). 
The CUIH::lnac (culicines) and me AnopheJinae (anophelines) both (onwn blood fcedinS 
man-biling species that are Unportant disease vectors, and it is important to be able to 
dislinguish betwem these sub-families. The Fi&W'C 2.1 illustrates the criteria which are 
uswIlyutedtodlstinguishthcm. 
The3e two mosquito subfamilies ruso differ significantly in their genomic structures. 
Anophc:li,. have heteromorphic sex chromosomes, a small genome size, and repetitive 
C'lcrntnlS dull Il'C' distributed in _ long-period interspersion pattern. In contrast. Culicinae 
have bomomorpbic sex chromosomes, and repetitive DNA tlw is organi.led in a short-
period inlerspenion panem (Rai & Black. 1999). 
The Anophelinae contain about 450 sp«ies in three genera. Bironella (which is 
Aumalasian in distribution. ocxwrillg mainly in the Papunn subregion). Chogruio (which 
is NtotrOpicaJ) and AnopJwl,s (wbich cootairu: by far the largest number of species. 437. 
and ls newty lA'Orldwidc in distribution) [Lehane, 1991 ; Sallwn et aI., 2000j. 'The 
phylogenetic relationships among Anopheles. Bironefla. and Chagasla were evaluated by 
lIarbl:lI.:h and Kitching (1998) using morphological characters. The Anophelinae were 
fuund by the authors to be: monophyletic:. "ith Chagoslo occupying a basal position 
withm the subramily. and AnopIwlts sharing. sister-group relationship with Biro,,~l/a 
The Culic:iMt. the largest sub-ramily, is divided into two tribes the Culicini and Sabethini 
(Servicr, 1993). with almost 2500 spe'Cic5 in over )0 genera (White, 1996). The main 
medically important genera an: Atdts. Cwu and Mwuomo species 
~ ____________________________________- '3~ 
University of Ghana http://ugspace.ug.edu.gh
ANOPHELINES I CULICINES 
~ffr~f~ 
;'~.:~~~:, ~ 
egg' 
·~ · II~~ .... 
IA edes Culex MIJn$onla 
~~o,~~ 
I Aed.s Culex Mo,,,onl. 
1 (7 po~~e ~~ 
tfj) 
AnoPheles IA edes .. Culex ~son;. 
fi&- 2.1 Chart showing the main differences between mOSQuitoes of the subfamilies 
Anophclinae (len) and Culicinae (right), (From Service, 1993). 
The main characttristics of anopheJines which distinguish them from culicines arc: adult 
anopbelines rest with !heir bodies II an angle of 30450 to the surface with the head, 
thorax and abdomm in a straight line,have datk and pa1c scalcs on the veins of the winas 
arTangN in distinct bklcks and palps ,bout as long IS the proboscls in both 5eXCl; CIIt 
him: flouts and are laid singly on the surface of water; larvae lack a siphon and re .. with 
their bodtcs par.allel to the water stuface. below the swface lilm. and pupae have 
bn:attunc uumpclS wttic:h arc short with a broad openina. In conlrUl, adult culicines rest 
with tbeir bodies roughly parnllello the surf.:e. have tc.aIes on the wina veins - but not 
ananacd an blocks and palps which are much ~horttt than the proboscis in females. Eggs 
are~ptOvidcdwithnoat.sandan:eithc:rhIltJsepa.rateI)'(Aetks).orJtackedin.floetin& 
~n U 'wu) or in a ~ on ~ling \'egcl~Jllnn (Mamvm); larvae have • lana or ~hort 
Siphon and rest with then bochCl> at !lit angle to the "'Utc' surface: and PupK baYc: mort or 
long bn:alhinl:.\ trumpecs ""jlh a natT'O"" opening. 
University of Ghana http://ugspace.ug.edu.gh
A high proportion of mosquitoes belOlll to the genus AeMs in which there are mote than 
1000 species. clAssified as 40 subgenera. AMes ~,,-,,(ICS lft dislributcd throughout the 
world, especially in temperate COWltries (While, 1996). M..ny _ vector.> ofarboviruses. 
inctudintJ; SC\cral of the most important mosquito-borne human viral diseases (e.g. yellow 
fcver,J~eoccph.1litis.dengucfcver,RiftvaHeyfeverandWestNilefever). 
Representatives of three subgenera of AeMs Dre naluml veetors of BrUI(IU malay; and 
Wuchereria bancrofti that cause human disease (White. 1996). The subgenus. 
Slfgo. .y ia. is probably of the grealest medical interest. Aedes (Slegomyia) aegypli is 
the most widespread and most dangerous species in this sul-ogenus. It is the principal 
vcclOr of ehikuogunya and dengue viruses in almost evcry outb~ak. Urban ycllow 
feycr is also mainly transmiued by Ae. a~ygpIi in Central and South America and in 
WcsIAfrica(While.l996), 
About&OOspccicsofCuJuareknown.htingclassifiedin2Isubgencra,withrTW'lyspccies 
acting as me vecLon of enzootic arboviruses. protozoa and filariae ('Mlite. 1<)81). 1(96) 
1Dc rypiul subacnus Culex contains the majority of species. Japanese cncephalitis 
virus is transmitted mainly by Culex spp. in \be Oriental region. espcclal.ly by the 
followlI1& species which bn:ed prolifically in peddy and swamps: ex 1';la~niOThyrtcltll$, 
ex, fl~lldll.J and ex, v;shltll; (Vanna, 1989} Culex lh~Jlerl in southern Africa and 
membtts of the ex, piPI~lLf complex in Eg) pi are important vectors of Rift Valley fever 
virus transmitted from livestock to man. In Australia, ex. annulir(JJlr;s plays a simil. role 
In thc= rptdcmiology of \1urray Valley encephalitis (Whilc. 1996). The Cul~x pip;ens 
complex comprises xvmtl species. subspccic!\ and forms., with ~(lrescntatives in .U pans 
III' the world (Scrvke. 1993). Typical ex. pipiens occurs in temperate c;ountnes of the 
nonhnn hem.isphe~. spading through temperate highlands 10 southem Africa. In 
Egypt. lhe: l.x. pIP'~ns complex is responsibk for roncwl\ian m.nasis transmis:sion 
University of Ghana http://ugspace.ug.edu.gh
(Southgate. 1919) and has been implicated as the main vector in outbreaks of Rill 
VaUey fevc=T. CtJu quinqueja.JclallU is the main man·biting tropical member of the ex. 
pipl~fIS complc:x, distributed up to 38"N in USA. lOON in Asia, but only 24"N in Africa. 
Bancroftian filariasis is largely maintained in tropical villag~ and towns by ex. 
qumqlle/tudtJllLf alone. although West African strains of W. boncrofti do not develop in 
Ihls species of mosquito (White. 1996). 
I he: ~cnll' Mansonia has only a dozen American specics. notably the widespread pest M. 
III1JJmu "hich brced~ in swamps and marshes with floating water lettuce (Pislia) and 
WIllet hyacinth (Elcltornia) or rooced Pon/~Mriu (White, 1996). Mansonia liritlans is an 
important \'«10(" ofVenc.ruclan equine encephalitis and may contribute to transmission of 
If bancrojil. As with other troublesome MansOnia and some Coquillellidla, .\1 IIIII/ans 
disperses far from the breeding sites in search of hosts to bite. Subgenus MansonioiMs 
abo h.as a dot.tn 5peCies., endtmic to the Old World tropics. btiRg pests arising from 
breedina sites in water ..-.ilh plenty of Eichornio, Purla, rooted Isachne (swamp grass), 
ZIIZtJIUa IIbd other suitable "egelation (White, 1996). Some atboviruses are transmitted 
"} Mansonioilks spp .• but these species of mosquito are generally refractory to W. 
bancroft'- The bites are more painful than with most other mosquitoes. Mansonioitks 
spp. are uf particular ~itologjcal interest because. together with Anopheles 
(AtrOpMJ~J) 5pp., they are the \I«lors ofbrugian fi lariasis (White. 1989). In South-East 
A~il'l. sv.amJ>-forest populations of M unnuluru, .1.1. uniform/.! and especially the siblill8 
species M. hunftelJl! and .\/ Jiws share the transmission of zoonotic subperiodic 8. 1IftJIayt, 
wtUch they ttansmitto man from Icafmonkeysan:Sochc:r wild animals (Wharton, 1962) 
The third su.b-family, the Toxorh)"ochiliR3(. has unly one genu .. the Toxot'Jtyncltilu, :mJ 
contains the world's largest mosquitoes lScnoicc. 1993) lhere llf'r some 76 mainl~ 
University of Ghana http://ugspace.ug.edu.gh
b'opical ipCCies of ToxorJrynch/(t.'J bue • few occur in Japan. the maritime atC'a of the far 
eastern pans of Russia and in the eastern USA northwards towards Canada (Service. 
(993). Adults are readily recognised b), the large size (up &0 19 rnm long. and 24 mm 
wing spread), rneuIlic colouration and prominent curved probo-'Cis. Females are 
i~)eofb)oodfcedingandbothsexcl> sockrteCtarandothcrnatutall)'occurring 
supry sub5t&nces. ToxOl'hynchil~J is consequently not involved in disease transmission 
(WhiI<.I996). 
Approximatel)' 3300 spc:cies of Culicidae have been described. more being recognised 
each year (While, 1996). The most important man-biting species of mosquitoes belong 10 
tht ~nera AnoplwleJ, Cu/~x, Aedes. MaflSon;a, HlNmu~(lglIs. Sabethes and PsorQP#tora 
(Servke.1980). 
2.1.3 Species complexes 
~ mu~uitoes occur as sibling spc:cies (or cryptic species) complexes, pall's or ¥roups 
ofckllc:ly ~1a1ed species lhat are morphologiCliI indistinguishable (isomorphic) but 
rc:producti~l)'isoltllcd.andwb.ich frt!qut!nlly live in lhe same area (Le. aresympatric) 
[Senic:c.1993). me first to be recognised as such is the A.noplw/~,t lltaCuli~n"u 
complex. (Service, 1993). Simia.ty. All gambioe was aJJO later shown to be a species 
complex. (eoluzzi. 1966), and m:tny olha Amlpheln species can now be considered as 
rcpre.seotiat COItlplcxcs, for example, An Jirus (Green el al., 1992), An 1,.·IJ"cifocj~J 
(Supna~/aJ.,19U)andAn. qlJadrimaculatlJs(Narang~/aJ., 1989a). 
In many complexes. species-spccilic differences in ecology and behaviour can 
sub5tantiaU)' a!Tect both disc:a~ trarumission and the success of rDClhods IiSCd for 
coouol. Fnrexamrle. II difTCTC'ntial response Iu Lnsecticides has bttn noted in the All 
University of Ghana http://ugspace.ug.edu.gh
cwicifQC;~~ complex in India. In this case. DDT resistance is mainly associated with 
species B. which is. very poor vector of malaria (Subbarao d 0/ .• 1988). 
2.1.4 Methods for distinguishing between cryptic species 
Sptcics Vo llhm a l:oOlplex by definition cannol be distingu l~hed morphologically, or only 
wilh great difficulry . In mOSI cases the benchmark definition o f cryptic s ibling spec ies 
has been mating incompatibility (Collins & Paskewitz, 1996). 
Mating compatibiJity studies were later replaced by polytene chromosome examinations, 
whieh wert amonc: the flfSt to provide researchers with langtble genetic markers that 
could be uttd to differentiate between spc:ctes (within cryptic complexes) and 
chromosomal forms (Coluzzi & Sabatini, 1968; unui el aI. . 1977; Coluzzi d 0/., 1979; 
Toure tI aI., 1998). For routine species identification however, chromosomal markers 
have major limitatioM. A number of alternative 5pC'Cies-diagno~tic procedlUcs h.m: 
IhcTtfnre hem developed, including idenlifiCltion by electrophoretic IJUItysis of gene-
etv.ytne systems (species-specific allozymes) [Mabon~' 01 .• 1976; Miles. 1978; Narana el 
al. • 1911. 1919a,b; lanzaro el oJ .• 1995). hybridisation with DNA probes (Colhns el oJ., 
1987; Gille & Crampton, 1917; Co..:kbum &. Seawright. 198&~ Panyim el of., 1988) and 
quantitative differences in the cuticular hydrocarbon profiles (Carlson & Service, 1979, 
1980, Ilamilton& Service. 1983; Phillips&. Milligan, 1986: Phillips~'al., I'JM8) 
From these: ·traditional tools ' of cytological eXllmination of polytene chromosomes. 
namdy. genetic compatibility, immunolosical and hybridisation techniqUH and isozyme 
an31 ~~ is (Fa"ja ~I 01. . 199411; MunstCTTT\alUl & COM, 1997), identification tools have 
expanded 10 include a \ 'as( amy (If moleculllr milrkers. 11lc:sc conlcmporvy rn.rt.ers 
range frum what are no\\ rdc:rrcd to IS 'classicaJ gemtic mutte rs' such as mitochondrial 
University of Ghana http://ugspace.ug.edu.gh
DNAs (Caccone elm. • 1996: Besaasky elal. • 1997; Lehmann dol. • 1997; Thclwell elal., 
2(00). spccieHpeciflC restriction fragment length polymorphism (RFLP) analysis of 
ri~ UNA (Collins ~I ul . 1988; YasothornsrikuJ el al., 1988: Torres el al., 2000; 
FlviaetaJ. . 2001). methodsusedtodetectandidentifysinglenucleoctdepolymorphisms 
(SNP,) [De Merida " oJ .• I999J and fmally 10 highly polymorphic nwkers such as 
random ampluaed polymorpruc DNAs (RAPDs) [Wilkerson d al. • 1993: Favia et al. • 
1994aJ. microsatellite ONAs (Zbeng e, 01 .• 1996. 1997; Lanzaro et al., 1998; Conn" m., 
2001; Shankov II ul .. 200la), and amplified fragment length polymorphisms (AFLPs) 
(Bl.ack & Lan2aro, 2001). M~ recently, sequence-tagged amplified RAPDs and sing1e~ 
copy markers (Mukaba)"irc I!/ al. • 2001) have alto been used 10 distinguish between 
Onc of the areatcst advantagcs of this wide variety ofgcnelic J'TUU"kers is that researchers 
RlIY c:booee to IAilisc any combination of marfc;crs or techniq..an to addrcs.s muJtifacetcd 
quaciooIrclatinatnmalana Iransmissionbyanophclinemo5quilocs(Taylorelal. . 2001). 
1bex mol~ul;u markers have proven useful in a wide variety of applications including 
molccul;u taxonomy, evolutionary systematics. population genetics. genetic mapping, and 
I variety of molecular diagnostics. 
I) "orphotogicatldentification 
The analysis oi morphological features is silli the must widely used technique for 
taxonomic and sysccmatic studies uf anopheline mo~uiIOCs. and lbe intemal 
dauiflC&llon of the genus Anophtle~ is based largely on morphological character!> 
(H,..b.ch. 1994). The available morphological di.qnostic ctw.cttf'S. altholJlh OOt 
relLablc, an: of ddinile laXonomic \ "Iue fOf the distinction of the: saltwater and the fresh 
University of Ghana http://ugspace.ug.edu.gh
WIlier sibling species of the An. Kambi(N complex. The number of sensilll coeloconica. 
the value oflhe palpal index and the shape ofthc eggs can separate these two groups 
(Scf'\·kc, 1993). However, none of these characters appears to be completely 
discnmlllani as tht= morphological characters of adults are variable and overlap in many 
inslal1c~(Coluz.zi.I964). 
In tM field. morphology still remains a very useful tool for identifying WU'Clated species. 
and cnnt'lk .. the preliminary sorting of material prior to the applicatt<m. if necessary, of 
other identification techniques. Sucb a procedure allows a colUiderable sivina in time 
and moocy and the advantages offcted by it have ensured ilS continued use. 
Unfortunately. hOWt.OVCT.1he prulifemtioa ofs~cies groups and complexes within many 
of the anopheline taxa is rendering the use of morphological characteristics im·alid for the 
idtntificatton of important malaria vetton (Beebe &: Cooper, 2000). With these cryptic 
spt'C\es, it has been proposed that the present morphological studies have not been 
cxhlUSlivt:enougbandthat more det&iled observaLions on aJl st8gesofthe life·cycle may 
yet yicld diagnostic charactcrs (\Vhile. 1977). However. species keys involving obscwe 
chanu:ters SuchlS the scu!ptwing of the egg chorion, larval chaC'lotaxy. male genitalia or 
the ratitl of various adult female body structun:s, may not be applicable in the 
Kienlificalllln oflatge numbers offieki-collected adult femalc mor.quiloes 
II) Reprodudiveincompatibility 
ODe o( lhc: earliest tcchnKjucs for SpcciC1 identifiCalion is that involving hybridisalioa 
",. ..: rmk.·nh wncre field material o( unknown taxa is cros.-mattd with mernben of 
koo\\n speciCS. the progeny of these pairings are c'(amint"d fOf anomalies in survivaS 
ralc' . !iC"r.uiO$.gonad~andfcrti lit)· . ·lhccxlcntlo .... hichthcseanomalinoccur 
IUlhculr. . rringhasalsobc:cnuscdwiDdiate:similantiesbetweenperentspccics(Btyan, 
University of Ghana http://ugspace.ug.edu.gh
1973}. This method requires no sophisticated technology and was widely used in carty 
Sludics of l'poc:ics groups and complexes (Davidson, 19~: Bryan. 1913~ Mahon & 
Mtcthkc. I C)82). Problems with this technique an: that it is time consuming and laborious. 
and requires WI a nwnber of ref~ colonies be mainlained . These practical 
limilalions make il iruppropri:ut.' for large--scale routine idcntification. Also, the fact that 
the parents arc force mated. artificially overcomes any prcmating bamcrs that may have 
existed naturally In the field (Patterson, 1985, Baimai flol. • 1987). 
IN)Cytogenetk:studles 
Cytogenetics involving the karyotyping of polytene chromosomes was one of the earliest 
tools forthc study of anopheline genetics (Norris. 2002). This tool has proven inunensely 
useful for diffeteluiating amon, sympatric taxa and chromosomal forms., and remains as 
the only ~Iiable tool to diffc~ntialt: ~ aU of the An. gambiae s.s. chromosomal 
(orms(Coluni tloJ., 1979. 1985); and until recently it was the only toollhalcould be 
reliably uxd 10 differentiate between all nine members of the AII./UMfIUS &roup (Grttn 
&: Hunt. 1980). Sharakhov tl 01. (2001 b) developed a cytogenelic map of An. ~SIW 
and compared it wilh (he polytene chromosome organisation of An. ga",blae. In addition. 
frcqueoci,-"So(chromo~mc:in\'crsionarrangrnlenlshavebeenulilised for investigations 
of population structuring and estimates ofetfectivc population size (Taylor d 0/. • 1993; 
Petrarca ~(oJ. • 2000; Kamau ~I 0/. • 2002) 
ClvOIU.NlInttl In\c:tslons have also been utili5Cd 10 adcIr\:u phyloi;cOC:lic issues reprdin, 
tbe IJn~lIR, maintenance and introgres.sion of invcrsions between sympalric populations 
Ind l&XI. for many anophdine spc<:H:s lfOops (Coetzce ~I uJ .. 1999). Such infol'Tnation 
bctwei!n lht: two ~bling species., AIL J:DIIIbi« s.s. am An QTubl.:nsu (della Tone tl al., 
University of Ghana http://ugspace.ug.edu.gh
1997). Simil:ll' compuisonsofshared polytene chromosome: banding patterns have been 
made amana pulative anopheline species in the New World as well (Pc' rez &:. COM, 
19(2). In both ca.o;cs. n:sean;hets must ~k to other markers to come to definitive 
conclWIOO$ concerning tbe$e homologies 
Among the dLsadvantages for using this technique III AnopM/~.s mosquitoes are ahat the 
polytene chromosome Pf1:parations must be made from ovarian tissue or fourth instar 
larvae. This limitstbe srunples to either adult bloodfed female mosquiloesor late instar 
larvae. In edditioo to Iile paucity of experienced personnel trained to read polytene 
chromosome preparatioru;, markers are not abundant or particularly informative in some 
species (lounibos " Conn, 2000). Despite the limitations. this method rcmains integral 
I'or much ofthc: contemporary work. 
Iv) CvtJcu&.r hydrocarbon .nalysjs 
Inttaspreific Jiffcrcncn in thc: h)'d.roc&rbonsof1he insect cuticle has beenc8pitaJised on 
in the identification of cryptic species. Gas liquid or capillary ps chroma.tograph), is used 
10 anaiYlE the bydroc;ubons in the CUlicuJw wax extracts, and mullivariate statistics is 
then .pplicd to the chromatographic data to differentiate sibling species by quantitative 
analysis of the cuticular hydrocarbon peaks (Carlson &: Service. 1979, IlJ80 : Hamilton & 
Sen.·icc. 1983: Phillip''' Milligan. 1986; Phillips et 01. . 1988). This hu beat found 
petticularl)' tDCful for the ditrerentiation of sibling spec ies of Anopheles mosquitoes 
(Carbon &. Service 1979, 1910j Hamilton &. Service, 1983: Millipn tf oJ .. 11)86 , 
Afty"",,"u e' 01 .. 1993). Ahhough the biological implications of variations in cuticular 
b)'drocarbonsbetwecnvcryc:loselyrcJ.atedsp:c:iesha\'e not been fuji), lIWIied. PhJllips 
and ~lhlh1!9n t 1986, and Phillips e' aI. (1917) ruggeskd that within such rdatKlll\hlp .... 
sptties.-specific dill'lI:I'mCa do exist wtUch tend to highligtll lhc: effect of geographical 
University of Ghana http://ugspace.ug.edu.gh
variation and possible incipient speciation mechanisms. The species-specific fundion 
performed by insect cuticular hydrocarbons in various environments has been outlined by 
Howard and BlomquiC'it (1982) and Phillips ~I oJ, (1988). They serve mainly to prevent 
desiCAtion and to assist ill "chemical commWlication" (lockey. 1980; Howard & 
8lomquist. 1982), Milligan~' ai, (1990) argued thai the hydrocarbon discrimination is 
based on the relative concentrations of the component chemic:aJs, I'IIther than their mere 
prnmee Of abIencc. Similarly, Hamilton and Service (1983) observed that, although the 
fow1h irulan of An. gambiM and All arobiemu bad simi lar cuticular hydrocarbons, there 
wm: still differmoes in the relative kvels oC some of these chemicals. II is thOUght that 
hy<l.rocerbons~notjUS(passi\'eprolectorsagainstdcstccation; they may additionally 
play a leading role in mate selection/recognition and population divergence. 
Hydrocarbon differences among sympatric populations of An, gamhwe 5.S. (Phillips et aI .• 
1987) miaht retl«t a xmiologjcal functioo of the compounds, enabling the insect to 
recopiJe potential males in those Ioeations where 5iblin~ species coexist. tn §Ome 
inJC'Cts, ~ of the male tttognilion mecbanism has been linked with the detection of 
specific hydrocarbon compounds and other componenls of the eUlieular lipid layer. such 
as fany acids, a1cobols. sterols. and aldehydes (Jalion. 19M; lJonavita·Courgourdan d al. • 
1917; PeJClcc.1(11) 
v) AUozvme/lso:zymeJlMlysi. 
Sp...'(u• . ,·specificallozymr.:sanalyscshave bcenuscd ror sc"cral years in the ident.lieaLion 
of siblinG and CTyptic species in mosquitoes (PasIc\Ir d al.. 1981: Cianchi ~I (II .• 1985). 
This technique, N.'oCd Ull !.he electrophoretic separation of enzyme variants., allows 
quantir.catMxl of lhc: amount of gene ditTerentiation among populations, showina 
evolutionary dm!rsity up to specks Inel (Bullini &. Coluzzi. 1978): bioe~mlcal keys are 
DOW in use for thcdiffctcrlliationufscvcral speclcsor. muJtirudeoforgarusms, Early 
University of Ghana http://ugspace.ug.edu.gh
work. utililOing these biochemical markers In An. albimonws. were used (0 document 
significant amounts of polymorphism among laboratory colonies that were then UJCd in 
link3~e mapping srudic:s (Narang ~t 01.. 1981). Careful analysis and use of these 
t«hniqucs led 10 the discovery of diagnostic a1lozyme loci for wild sympatric populations 
of All. qfMMlriJftOClllDtv.r A and B ([.anzaro rl al .. 1990) and allowed broader UJe oflhc:se 
lools for the population genelics ofnarural v«lor populalions (Hii eloJ. • 198"1 This tool 
YolUaJSO fashioned iDlOa dichotomous electrophoretic taxonomic key for three species 
within the An. qllDlirlmueWaIKJ complex (Narang el cd .• 1989a. b) and was utilised to 
delineate IWO ~Ibhng SPC"IC\ \\;trun the An. min;mus complex in Thai land (Sucharit d al• • 
1988). Among tlw= Jr.moacks of isozyme analysis arc thaI specimem; must be fresh. or 
kcptfro.«nuntilanalysis(RicbardsonetaJ. • 1986)andthaltheproccdurcitsclfrequiresa 
relatively large amoWlI of sample material compared to the few nanogrnms of DNA 
rcquiml for PCR (Norris, 2002). Finally the cost of the kctuUque is relatively high. 
~i.aJ.ly. coruKkring its tediousness and. the amoWlt of time needed to pcrl'onn large· 
leak: studies as compared to other techniques. Noncthc:less, iso.l)'mc:s I\a\·e provided 
val~e data on which much of the contemporaJ')' work on anophelines is based. 
vi) Hybridlutlon assays 
MOSI of the first DNA based work for idemifying cryptic species were based on 
hybridiralionassays lhatdetectedspccies-specilicdiffercnccs inhighl)' repetitive 
~l\ICnc("~ (C('tllins rt al .. 2000). Although somewhat time conswning and technical, 
the5ea.,". .. a~\"crcmtlrccflicicntlhanpn:\iousmc:thods.5uchaspolytencchro~ 
analysis and isoc1\1)tnc gd dc-cttophon:sis. Tbe approach, bowcvct,lenendly had two 
map limitations (Collin.'i 1:1 al.. 2000)_ The spc:cin composition of the cryptic species 
complex under stud)' had 10 be kno",,, hefon: the &S.S.lI) could be developed, and each 
spectcs in the complex had to differ from lhe others in lhc abundanc:< of one or mo~ 
University of Ghana http://ugspace.ug.edu.gh
repeat sequences_ OespitethesclimitatiodS,rcpeatsequencehybridiZllionasslyswere 
developed and used for field studies of the All. p",hiat. An. dirUS and An. pvncrulatus 
oompkxe5 (Gale &. Crampton. 1987. 1988; Panyim e' 01 .. 1988; Cooper dol. • 1991 ; 
Beebe~laJ .• (994). 
2.1.4.2 Claulc.al genet:k: markers 
I) RutricUon fragment length polymorphisms (RFLPs) 
ID DNA strands, restriction fragment length polymorphisms (RFLPs) represent changes in 
the lengtlu of DNA between sJX'Cific enzyme eUHing sites brought about by sequence 
changes. DNA can be cut using one or more restriction cndonuclea.ses thai recognize 
sites on the DNA tempJltc. The resulting restriction fragments are separdtcd according to 
size by electrophoresis on agarose gels revealing diagnostic polymorphisms with distinct 
DNA fragments. Ho~ver. in most cases. the digestion products appear as a smear with 
fntcment.s r;lIIgmg up to approximately 2S kb in size (LoxdaJe &. Lushai. 1998). To gain 
useful infCtrmalll)n on these: geis. the DNA must first be transferred to a nylon or 
nitrocclluloSt membrane by Southtm blotting (Sambrook ~, 01.. 1919) and selective 
fragmmlSvisuaJiz.edbyhybrldi/ation with a suitable labeled DNA probe (Milts.. I 984}. 
Favia ~I 01. (1997) designed a polymerase chain reaction (PCR·RFLP) 3S.S8y. which 
unambig\lOU5ly xparales the An. gomlticw Mopti fonn from the Savanna and Bamako 
forms. This method. baed un lhe pl"Csence of a restriction site length polymorphism, has 
been tnted on ptt'viou.,l) karyul)pcd sympatric specimens from Mali and Hurkina !-'aso, 
and it has alrudy been applied to verify the distribution of other molecular markers. as 
for example the pyrethroid resistance gene (kJr) among these chromo!lOmaI forms 
(Cbandn:<lo/. . I999). 
University of Ghana http://ugspace.ug.edu.gh
The maIO advantage which is appticabk: to all DNA markers. is that the level of 
molecuJar variation detectable is increased. With RFLPs. lhis is because of numerous 
restriction enzymes that exist which eu1the DNA at different sites and a diverse selection 
of probes based 00 hypervariablc mot'ifs (Loxdale &, Lushai, I99B). By virtue of their 
ability 10 detect analogous DNA on both lhc members of pairs of homologous 
chromosomes. RFLP based techniques aJso yield eodominanl. non-epistatic Mendelian 
markers which provide greal genetie resolution because of the large nwnberofrestriction-
cw:yme-probc combinations available (Zraket el a1 .• 1990). Finding a probe that will be 
suiUIbIe 10 resolve the population under investigation is \he main drawback of this 
klCbnique, but this is overcome by the availabiliry of suitable probes from paraJlel studies. 
Anothtrdisadvantase is that such metbodsutilize probI.!s which need to be screcned from 
genomic libraries if not already available thus leading 10 increase in costs. Tbest markers 
are thcreforebestadoptcd if probes from parallel studies 10 those contemplated already 
H) Hltoc:hOndrtatDNAanalyais 
Mitochondrial DNA (mtDNA) has been used widely in taxonomic and population studies 
(Simon ~I oJ .• 1994) and is a valuable IDII'ker to indicale maternal gl!ll¢ now, as it is 
predominantly tnnsmincd through maternal Jines (Avise, 1991, 19')4). Because mtDNA 
is more numerous than nuclear DNA (the ratio of mitochondria 10 nuclei being far greater 
per cdl), it can be more easily detected in old samples. Nwnerous investigations have 
utillttd mtDNA to ckrive the phylogenies of All. RtI",blOr s. 1. and to address questions of 
rnt. . :I1~ :.tNCturing among members of this ~in complex (Caccont 1:1 aJ .• 1996; 
Ik"",.s.).) ~I 0/. • 1991: I.ehmann el aJ .• 1997; Thelwell el 0/. • 2000). MilOChondrial DNA 
has btt-n lolmltarly utlli~d in investigations of. wide variety of anophelines and 
anophdinc species complexes. ,"·oley ~I 01. (1998) URd the COli Bene lu deriw the 
University of Ghana http://ugspace.ug.edu.gh
phylogeny of the AustnLIasian anophelines aod De Merida el al. (1999) also utilised 
mtDNA (NOS) and SSCP analysis to examine the population snucNrt of A.n. albimanw. 
In these uamples. mtDNA has proven useful in exammatlons of phylogenetics and 
populatittn structure over large g~ scales. but it should be ~ilerated thai this is 
nollrueforallanophdmcs 
Apart from the versatility of its applicaLions, the other advantages of this marker are that 
~ucnceanalysiscan differentiate to the levelofrace-fonns. This is partly because 
mtDNA mulait's approximately 20 times Caster than nuclear DNA, so that a 2% 
dh'ergence in St.'qucnccs bctw'ccn pairs oflincages sepamleJ for less than 10 million years 
equates to I million years since their genetic separation (Brown d ai" 1919). Ho ....- ever, a 
confowuling factor noted in aphids for example. is the presence of multiple copies of 
tmain mlDNA genes (8unnucb &. Hales.. 1996). In such cases., copies of mtDNA aenes 
haYe become incorporated into the nuclear genome. In one respect. this may prove to be 
an 8dvantagr:, as tBeIe ltalUlocated genes become 'Cossilized' in the nuclear eenome, and 
can be a source tocnluate ralcsofmutation in relation 10 lhcequivalent mitoc hondrial 
fraction , 
Somc of tnc di$.Bdvantages are thai the lewl of variability bct\\ren samples depends on 
tncir <le8fCC of genetic i!Olatiun (an eslimnle ofme kJwcr end of the resolution of genetic 
isolation woWd be about 50 ycars). Mitochondrial DNA has also failed to differentiate: 
cryptic laxa within the An. maculipennis complex (Collins el 01 .. 1990). 
III) The polymerase chaIn reaction applied to nudur DNA (nONA) 
Ribusnmal 0'" (rONA) is a muhicllpy grnc compte" found III nuclear DNA (Hllli~ & 
I.>i"oo. 1991), The rONA cistron cumprilCS the 288, S.SS and ISS coding ~gion~ (hat arc 
University of Ghana http://ugspace.ug.edu.gh
interspersed with inle:maJ traMCribed sp.:er rcgioos (ITSI and ITS2, either side of the 
S.SS gene) (Hillis&: Dixon. 1991). These modWesue then repealed in tandem linked by 
the inlcrgmic spacG' region (lGS) (Avise. 1(94). Thc:re arescvcral different sites with.in 
the rONA for wruc:b universal PCR primers have been construCted (Black. 1991). 
HO"'-'evc:I', cODSCl"Y'Cd sequences flaok regions of bypc:rvariability making lhis mmer 
versatile at di{f~nt levels of taxonomy reflecting different rates of evolution (JTS2- .:!ss. 
Campbell ~r aI. • 1993; ITSI - ITS2. PaskewilZer al. • 1993; lIS, Pashley ~l ell .• 1993). 
Many invc5tigators have used thc:3c regions to distinguish between members of spedes 
populations and complexes of medically important Vtt10rs of disease (Paskewitz d oJ .• 
1993: TOWlUOn.t: Onapa. 1994; Collins cl Paslcewitz, 1996). Currmlly. the most widely 
used method in the identification of member species of the An. gambiae complex is besed 
OQ lIJrCies-specific nucleotide sequences in the ribo!Omal DNA (rONA) inlergeRie spacer 
rcgKxu (Scott et DI. • 1993). Thjs may be used 10 identify both species and inlcrspedcs of 
bybrids, reprdJess of life stage:. using either exlr1lctcd DNA or fragments of. sp«illlt!n. 
In usina this method for the Ktentification oC mosquitoes. intact portions of a specimen as 
small as an eea or the segment of one leg. may be placed directly into the PCR mixlUre' 
for amplirtCatioo and PnIlysis (Scott d aJ. • (993). 
This t~ p.! l·f d;:agnnstic tool has abo been extensively used for differentiation of species 
-.ithin the All {UIWJIW group (Hacken el a'-. 2000). All dtrus (Xu ~I al. • 1998), An. 
jllWkl1i1i.f (Marn:mmani el aJ .• 20011. and the An quodrimucululLU complex (Comel el al. • 
1996). Such srudies have readil) identified the lUoCfulnc:s.s of this molecular marker for 
difTcrmc~ at the I~el of sub-species. races and strains. Ho"c\cr. continuous 
\'lIri"ti~ of the! ln1crl::eRlc 'racer (lOS) and multiple copies of the gent within I sin~k 
I~I\ "jual can confound analyst,>. PCR amplification and sequencing results. This is "-ell 
University of Ghana http://ugspace.ug.edu.gh
demonstraltd by (GS variability within an indIvidual which has been {0W'd to be as greaa 
as thai across a sample population of the aphid, Schizqphis grammum Rondani 
(JiemipCera: AphJdidat) at various spatial scales (Shufran el 01 .. 1991). The costs 
involved with this marker are similar to any PCR based technique. In gmeral. problems 
with muhiplC' copies mHll that lhis technique, unlike others. should be used with careful 
consideration. 10 some studies to date. contaminants have ~n amplified along with the 
~udy $OUtce (Fenton el m.. P)()4); in this particular case, a novel fungal organism was 
identified. 
Apart from rONA Benes. there are a few other nuclear Benes mat have been widely 
applied to phylogcnelK: investigations. However, Friedlander (I 01. (1992) have reported 
14 nuclear lenes thai may prove useful for highcr level phylogenetic analysis over a wide 
range of taxa, e .g ENOL. enolase; G6PO. glucose-6-phosphate dehydrogenase and 
SODPUMP.Na'/K··ATI»uc. Further sites for short sequences thai are proving useful in 
\·;ariow taxonomic and population genetic SNdies have recently been investigated in 
mlrons (Lc~...a. &:. Applebaum. 1993; Palwnbi &. Baker. 1994; He & Haymer, 1997), non· 
coding regulatory regjonsclosely a.s.soc;iated with genes (Slade elul .. 1993). 
2.1.4.3 Highly polymorphic markers 
I) MlcrowteNlteONA$ 
Simple tandem repetitive DNA, more commonly known as microsalcllile DNA. has 
become a popular 1001 for ~rnctic studies ofanophdines. Micros.lelliles are described as 
DNA fragments consiscini; of simple short sequences woolly of 2~ nuckotide!l (nl). 
tandanly fq)C31C'd in man: or less unifonn tracts up to approximarcly 10' nllon, (TaulJ'~ 
1993; Chamben &. MacAvoy, 2(00). MicrosateUiles art' found in the I~nomes of just 
ilboul (vcry known organism and ~Ue. In most eukaryotic organisms.. 
University of Ghana http://ugspace.ug.edu.gh
micro5PItellites are disperxd throughoul the genome (Hamada el al .. 1982) and can occur 
as frequently as evt:r) 10 kb (T4Utz. 1989). Polymorphism in miccosatclliles leads 10 aD 
tncru.sed probability of finding heterozygous indi\'iduaJs. Mkrosattllitc kx:i are liso 
frequently hy~rvariabk. (Le. they possess several alleles 11 ~lIlively high frequency, 
improving lMlr versalillt)' and hence are tdell tools for mokcular charactcrwtk>n of 
individu.aJsandilUdiesofint.raspecific ... ariatton(Tautz, 1993;L.anzaroetai. . I 99S}. 
£Xploiwton of micros.atellite kxi has pro\'ided biologists with 8 sel of molecular tools with 
un)U~stdversatilily.Thckeytothisversatilitylicsinthehighlcvelsof\'ariability, which 
arecharactcristically(oundatsuchloc i.coupledwiththe speedandreltabilitywith which this 
in(ormo1tton can be accessed in the labol1lloT) (Chambers &: Mac:Avoy, 2000). Their 
applications range (rom estimation o(the spatial relationships between chromosome segments 
10 the elucidation of temporal ~Ialion!ohip.s bet. .... een oricins of species and genera 
Mlcros.tellite loci ...... eallO been described ti ideal mark~f!lfotmc:asuringpopuillion le .... el 
phenomena iUCh &5 population structun: due to their high polymorphi.vn, c:odominance, 
abundanl preaeN:e throughout the genome and relalive ease in sceKing (Bowcock et al.. 1994 ; 
Butt.nan ~taJ. . 1994: Scribneret ul. . 1994: lanzaro et 01.. 1995). OtMr applicatKMu include: 
ux in the analysis of lIboratory :mJ agricullul1ll organisnu and human genome div~rs ity 
projecu 10 mIp single gene trailS rapidly (Hcamc et 01.. 1992), DNA profiling (Gill ~t al. • 
1994) and .150 in fou;lIsic .... ork (Chambers ~I al .. 1997). MicrosalCliites art almost perfect 
tools for application In dcwnnining the ,,"em of ~lationship:. bctv.ecn individuals because 
!hey are capable: o(hiply discriminatinl biperentally inherited co-dominant markers. In the 
analysis of An gu",""", popubtions., microsarcllile DNA _lysis has been used 10 generate. 
genctK: map of the mosquito (Zhcng ~I aI. . 19(3), in the Ilenet ..;  difftrenlillt~ of populations 
(Lanzaro~ttJl.. 1995) and in the studyofpopulatton Mructurc (Donnelly etal. . 1999). 
University of Ghana http://ugspace.ug.edu.gh
The advantage of this approach is me ability to de~t greater levels of genetic variability 
as many microsatellite loci, often with numerous aJle&es. can potentially be screened for 
ecologjealuse (i::vans, 1993). Individuala1le~canbesc;orcdatpatticularlociand 
provide good Mrodclian markers. Tri· and tetra·nuclcotidc repealS provide much better 
signals. i.e. reduced stuIIert.nds (caused by mispriming inducing templates of smaller 
size 10 the ori&inal t.nd) and are more easily recorded for size variations (Queller tf al., 
1993). However. the disadvantages include the I't'quirementlo sen.:·cn ~veralloci for 
adequate population infonnation (a minimum of four polymorphic loci in clonal 
orpnisms and ten and above forsexuaJ populations). thus increasing development timc 
lOCI costs pcr sample. Although micro-satellite peR primers are designed to be species 
specific., cross species allele amplification his been noted (Queller ~I oJ .• 1993). 
III Randomly ampllfled polymorphle DNA (RAPO) 
RaDdomIy amplified polymocphic DNA (RAPO) marker:! ka\'e been extensively used 10 
distinguish berween members of cryptic species (Williams el al .. 1990). The technique is 
fast. lechnicaJly C81)'. aDd requires little material. Mo!rt importantly. no prt'viow; 
nucleotide SC'qlkncc iDfonnallon is needed for the construction of primers. Many markers 
can be readily identifted for a variety of laXOnOmic: kvels and, in comparison with UNA 
sequmcina. the effort and cost are modest so Ihat many individuals can be assayed. This 
method, whtch I'C'veals great genetic variabilily due to the regHms in which amplification 
takesplace(BlackelaJ.,1992),iSU5C'ful in differealiatinl closcly-related indi\'iduals, and 
there are numerous commcTCially auilable primer kits which can be used to ICTttn 
popalatiom.. The __y Iis of RAPDs provides a novel and etrccth..  c mtthod for 
di:Miapishing AttopJwles specin and other organisms accon1inC 10 the bMding panerns 
ofthrir DNA. as wet! as providing a ncw means of obtaining gcnctK: markers (Iledriek. 
1992) 
University of Ghana http://ugspace.ug.edu.gh
However. theae markers have se~raJ dnwbecks: (i) they often reveal continuous 
variation betwem sample populations: (ii) primer libraries need screening to identify 
suitable primers and stable genetic polymmphisms, both of wbich are time (:onsummg 
and costly; (iii) the dominant natw"C of the markers makes it impossible (0 diS1inguish 
between homu.lygous and heterozygous alleles (Carlson et aL. 1991); (iv) additional 
tedwques arc n:quim1 for useful Mendelian data to be: obtained (Vaughn Ie. Antolin. 
l
1998); and (v) numerous factors such as DNA qualiry and quantity. Mg ' concentrations.. 
Toq souree etc. can affect the reproducibility and standardization of reactions (Black. 
\993). 
2.1.50istributionMMl-=oI09V 
Mosquitoes havc an almost \\orldwidedistribution.. being found throughout the tropics 
and temperate f'C'gioll.'l and even well beyond the Arctic Circle; they are Ibsent onJy from 
Antarctici and a few islands (Service. 1993). They ate found at clcvations of 5.500 m 
and in mines at a depth of 1.2S0 m below sea level (Service. 1980). Their great diversity 
of habitats and IifC'-hi"ory strategies has allowed them 10 colonize many contrd.Sting 
environments. For example, mosquito larvae arc founc1 in ponds, swamps. sa]HYaler 
manbc:s,pools,tree-ho~,poliuted""'8tC1ofSCJ>'ictanks..ricefields.discardeddomcstic 
conlainers., rock pools, plant axils and pilcher plunts. and in a variety of other aquatic 
habitats. Adults are encountered in almost all types of ecological zones. from equatorial 
rain rorests. urbanised conglomerates, cuitivalCd lands ttl ~cmi-arid areas (Service, t 993 J 
Some gcr'lC13.. bo¥l'n'a. tuve a rnuiC1Cd distribution and may be confined to certain areas 
of the world. lbe gCncr:l IJolu"otfJgut and Su~thts. for example. are round in only 
Central and South Amenca. Some mosquitoes may occur in only • few counTries or 
localities. for example An. hwambtN, .....i lcreas oth..-n such as~. ,wnqueJafciOlu.f and At. 
De/OlPti arc widcspn:8d in the troptcaI region .. ofthc ¥lurid (Service. 1980) 
University of Ghana http://ugspace.ug.edu.gh
Most mosquiloes probabl) dispenc only a few hWldred meU'CS or SO from their 
emergence sites; for example A~des Qltgypll usually proMbly files only 25-100 m or so. 
Nunnally Anopht/~J do not fly more than 2 kIn, but in certain circwnstances they can 
rqulatly fly l·S km (Scrvic:e. \993). The distance mosquiloes fly is determined largely 
by the enviroomenl: if suitable hosts and breeding places are nearby, mosquitoes do nol 
have 10 di!'pcrse far, but if one or both are 1(. di~tam; ('. sreater dispersal will be 
2.1.6 Hedicallmportance 
Mosquilocs cause more bwnan sUITcring than any other organism, with over one million 
people dying from mosquito-borne diseases every year. Mosquito-bome dixases 
continue 10 cau-.e significant human health problems.. lugely in the sUbtropics and tropics, 
and lheit inCldc:m:e h. .b  increased significantly ....i thin the last 2 dcc:1des(World Bank 
Repott, 2001). Not only can mo~ulloeS carry diSCl.X'S that affitcl humans, but they also 
[nnSmit ",veral diseases and parasites that dogs and horses arc: very susceptible to. These 
include dog heartworm, Wesl Nile virus (WN) and Eastern equine encephalitis (EEE) 
LWbil~· . 19961. In addition. mosquito bites can cause severe sk.in irritalion through an 
,Ilk/gil.: n.:~llun to the mosquito's saliva causing a red bump and itching. Mosquito 
\ ••: t'lore,j d.~3$C's include prolOllOaR di.seucs, i.e. , malaria., filarial diseases such as dog 
hc:artwonn.and arthropod-bome viruses (1lJ"tKwiruses) such udenaue, enccphalilisand 
yellow fever. 
Estimales from the: World Health Orpnization indicate that three mosquilo-bome 
dilcascs an-: among the leading causes of morbidilY and mortality in developing countries 
around the world (WHO, !OOI,. Nearly 500 million clinical cases of ma.1aria caused by 
University of Ghana http://ugspace.ug.edu.gh
mfection wtthPltuwtOdi. ... ,.....,itesoccurcach yc&r, resulling in 2.7 million deaths. 
mainly in children. Malana is exclusively transmitted by AttOph~/eJ mosquitoes 
Lymphatic filariasis is caused by parasitic nematodes and is the second leading cause of 
pc1l1lIllCnt and long-term disability worldwide. with 120 million people presenting 
chmcal morbidIty (WHO, 200(8). More than half of the world' s burden of lymphatic 
filariasis (LF) is transnutted by ex. qlllnquefasciafllS and other man-biting mosquitoes of 
the Cx. p'pirtU complex which are responsible for Bancroftian filariasis transmission in 
thc:Americas. Egypt. urban East Africa, the Indian subcontinent, Indonesia and SOUthC4St 
Asia (WHO. 2000). In about 40 countries in the African regaon and Papuan sub-region. 
W. bancrofli is largely transmitted by AnoplteleJ mosquitoes that also vector mafari. in 
rural areas.. In most Pacific countries. W. bancrofti is vectored by aedine mosquitoes 
(Aedes and Od/rl'OlDtu..f) that also transmit arboviruses. notably dengue. Srugian 
filiUla.. . IS, tnnslTullcd by Man.wtlia and AMplw/eJ. i.J now limited to only 8 orientaJ 
countries (While, 1996) 
Dengue fever virus, particularly Its haemorrhagic form, is. threat to >2.5 billion people. 
with an annual incidence in the tens of millions and 324,000 death.! per )'Car (WHO, 
2002b). Dengue vit\I5CS are transmitted to bumms through the biles of infective female 
A.:Ja mosquitoes. Many mosquito spc:cu:s are abo vectors of otht:r arboviruses. 
including sevaal of the MO!illmportant mOSQuito-borne hwnan diseases., West Nile ViruS 
(WN). ca:>tem cqumc: encephalitis (EEE). western equine encephalitis (WEE). S1. Louis 
encephalitis (SlE), La Crosse (LAC) cnc:ephaIitis and Jap.anc::sc: c:nccphalitis (JE) lWh.ite. 
1996]. The: arbovl.rusc:s are the mMt divc:ne, nlJmcroUJ and senous dlsca.tel transmitted 
to SUSC\.-p(lblc vcnebrate hosts by mosquitoes and other blood· reeding arthropods. 
University of Ghana http://ugspace.ug.edu.gh
2.2 The Genus Anopheles 
Nocablc d~in&uiwog ClW'a..:teristics of anopheline mosquitoes (from other mOSQuitoes) 
include the long pe.lps tbat are present on both males and femaJes and the chatacterislic 
pattern of blocks of dark and pale scales on tht.: Wln~ veins. especially along the CO$I.II (top 
partoflhe w;ng) [Fig. 2.2]. 
Chromosomal data suggest that Anoplwlu is "primilh'C''' within the rllffiil~ CulieK:iae 
(Besansky & Collins. 1992). Evidc:nc.c. incluJinl significantly smaller chromosomes and 
lower nutle.r DNA conlcnl (Rae> & R.i, 1990), and their unique possession of dimorphic 
sex chromosomes and long-period interspersion of repetitive sequences in the genome 
(Black &. Rai, JI)8KJ. support the extensive divergence of AnoPM/~s from the other 
mosquitoes. 
"The genus AltOpIt~/es Meigen consists cunmtiy of 437 recognised species which divide 
into six su!>-gcncra; Anopheles (185 species). Celllll (200 speei~), LophopodomyJa (6 
species). K~r/~zjQ (12 species), Nyssorhynchus (29 species), and SJelhomyia (5 species) 
[SaIlwne/ul. . 2000). 
2.2.1 The life c:ycle of Anopheles mosquitoes 
The AnoPM/~s mosquito goes through four separale and distinct stage. of its life cycle; 
egg. larva. pupa. and adult (Fig. 2.3). Each of these scaaes (lUI t't easily recognized by its 
spec ..'  appcaranc(. The immahate scaaes of Anophtl('s are aquatic, "hilst Ihe aduJt is 
tenatrW. Anophc:linelarvac:cxisclnawidc\·arielyofdiffcrf:nlhabitats(Scrvi«.1993). 
University of Ghana http://ugspace.ug.edu.gh
Flr.l.l Art IIDOphelme wmg which 15 idMfied by i1s c~ ~ 11wn are 
blocks ofpale aad dad scales (red arrows) on the vein. especial Iv along me costa (tIw lOp 
... of"~. Then is also • Iiiap of rwrow. outsa.tdina SQ].. . (peM arrows) 
alona1hebonomoftbcwma 
36 
University of Ghana http://ugspace.ug.edu.gh
P'4'a 
"1.1..1 Li:fe~eofAllop#wlesmosqui1oes 'Iben 1ft r~dillinc.1 ~ of 
development (completcmetamorpbOSlsl: ega. lana. pupa. and adult. the eggs are 
LMiD ..... lDdhawepalredlalenlaar-ftlledfloll$,lJOb..a ..... _lMYhllehin 
1-3d.,.. lhelro"ali_.,...&&eItodlewttersurf'aceandhlsconspltuousmouth 
brushes 'Mae pupa IS collllDHhlped. bas afuscd head aad thor-.. The pupil SfaBe IS 
• ";'ar:l\·el~' short (two to line dayt), aonf-'ins. a.sIIJonaI .... 11'1 which the adult 
~trilhiBtbepupaicovennsBolhedullrmllllesandmaJesdnnknec:tlr1nd 
01Nrplad:Jwcesasenll'gysources : ~onl)' ranakllClb.bkIod"". tdizin8the 
...--_ .. pmd. ... -
University of Ghana http://ugspace.ug.edu.gh
Some species, such as All. ps~ucJopunclipeltlfU Theobald. prefer pools of water in drying 
streams in relatively dry. sunny areas, while Qthas prefer inundated an:as such as ric:e 
fields. AnDjJMles minimus in 'I118iland and the Philippines lives amongst bamboo roots in 
relalivcly swift·moving water. Anopheles gambiae sensu lato often lives in puddles of 
water that accwnulale in lyre ruLs. and An. stephens; may live in rnatrmade conlainers 
Anoplwles l.rvae as a whole lead 10 prefer clean. rather than polluted water (Service, 
1993). 
The A1IOphelesegg is difficult to sec .....- ith the nak.ed eye, being quite small about O.S x 0.1 
mm in length (Service, 1993). It as boaI·shapcd and. has two laleral floalS which are filled 
with air. The shape ofthesc floats. as well as the chorion ornamentation may be used to 
distinguish the subgenera and sometimes the species of IlllOphelines (Hervy el aI., 1998). 
The eggs are laid singly on the swf8Ce or on the edge of water and are attached side·to· 
side or end·to--cnd. but never md·to--side. The eggs arc white when newly laid but tum 
brown or ~k within [·2 hours. In tropical temperatures, ens hatch in 2·3 days to 
produce t.vac. These eggs cannot ""ithstantl desiccation (Service, 1993). Some 100·200 
per batch are laid every 2-4 days. Survival and egg development are mainly dependent on 
temperature and relative humidity; Wlder extreme climatic conditions, mosquitoes may go 
into hibmYtion or aestivation, which allows the survival of the species through Ihe winler 
in temperate climates. or long dry !Casons in tropical arid areas (Clements, 1992). 
l'helarvaewhichhatch from the floating egg have ahead, a thorax und nine abdominal 
segments. The larvae abo show numerous morphological pcculiaritirs., the most striking 
ofth~ lxing the IKk of a siphon on the segment vln. the presence of palmalc setae on 
the lil'!it <;c\cn abdominal segmenLs. and Ihc shape of the cephalic setae (Hervy el al. 
1M). Anopheline I. ..... arc filter (ecdcn; of paniculales at the surface of water. The 
University of Ghana http://ugspace.ug.edu.gh
buccaJ appatalUS is SllJ1'OUI1ded by bruYles which create a "''llter flow that brings food 
f*1K:les to the mouth (Service. 1993). A conspicuous distinctive eharnctcr of AnopMleJ 
is the \\ay the larval body positions ilSel(parallc:! to the water surface as against the 
obIiqoou:s st)'k exhibited by Atlks and Clilu. This position might be related to the 
abscnce. in lheA1tOpIteltsWva. ofLhc respinilOry siphoa which inCulicioelarvaebreUs 
the surflCe Of\\13Icr and takes in air into tbe trac:beae. 
The dumion of the 1Qf'\"al stage ranges between S and II days dunng which the larvae 
uoderso 3 JUCCessive moults. during which they maintain the same morphology, but 
tnC~asc in sitt from less than 1 mm for the I Sf. stage larvae to more than 0.5 em for the 
4" inscar(Service. 1993). Rates of larval growth are influenced by co\'ironmental facton: 
aach IS warer temperatwl:. photoperiodicity, food supply, degree of overcrowding and the 
5pC'Cies(Scrvicc, 1993: White, 1996). Tht fourth instar larva moults into a pupa, in which 
lhe hc.:t and thorax are (wed to form 11 common en,'elope whicb bears the respimtol) 
lrWnpelS. Thepu~lsarelativelyshor1stage,usuaIly2-4d.)·,.durin&wtUchildoesnot 
feed. The actual eclosion takes approximately 12·15 minutes (~ervice. 1993) and for 
most species emergence occurs in the early evening or late at nighl. 
After~cnce,Mluitfemales5ecksbelleramongstvegeta1ionunliJtheyarereadyfor 
mating (!xt\\lren a few bouts 10 2 days) wtuk the mouthparts harden 10 permit ~kln 
pc:nclntion and f«ding. Males arc unable to mate until their genitalia have rotated 
through U<f, a proc:essthar lakes 6·24 hours depending on Ihe spC'Ctes(Service, 1993; 
White. 19(6). Malina usually occurs in night and commonly around du..<:k. Female 
mono".3my in An.. I'JI"bll.le is briefly enforced by the mak's deposition of. matins plu[t 
mlo lhe fcm:ale's aaUul chamber .fief speml are tnnsferred. The' maling plug, formed 
from proteinaceous SlC'Ci"etK>n." of the male's acceuory ,t.nds. CffC"Cli\e1y blocks sperm 
University of Ghana http://ugspace.ug.edu.gh
(rom subsequent copulations from raching the spennatheca (Gillies & Chir t 956; 
GigHoti &; Ma-.on 1966; Kettle. 1995). 
Female anopheline mosquitoes obtain energy from sugar meals for metabolism and for 
night Flight is Deeded for mating and for finding a ho51 that will provide II blood meal 
SOW'CC: (appc=titive night). The blood meal is • protein·rich diet lhal the mosquito 
SWTOWlds, after ingestion. with the peritrophic matrix (PM). a thin structure containing 
chitin and proteins. Digestion requires secreted proteases that penmate the PM 
(Clemenl5.. 1992). The smaller digestion JWOducts are hydrolyzed by microvilli-bound 
enzymes before absorption by the midgut cells. lbe blood meal-derived nutrients are 
processed by the in5eC1 fat body (eqw\,it!entto the ]i\'C~r and Mfipox tissue in vertebrates) 
intoeu proteins (vitellogeruns) and vanous lipicis associated with tipoproteins. These 
are then exported through the haemolymph to the insect ovaries. The egg development 
process takes 1 to 3 d.ys, and no further food intake i! needed until after oviposition. 
when a new cye~ of active host finding and blood fceding. digestion, and egg 
devdopmenl begins (Clements. 1992) 
Inc adult female may live from a few days 10 wdl nvcr • month. going tlvough several 
cycle .. til' hload feeds and egg-laying. '] hI.' Juralion of the whole process. from 
o\ipnsilion In adull eclosion. illCmpcrature-<kpendcnt. At the elevated temperatures of 
th\,., Inlcnropic:aI arus,. it takC!\ a minimum of7 days (Clements, 1992) 
University of Ghana http://ugspace.ug.edu.gh
2.2.2 Anopheles species of medical importance 
Only :I :.mall number of mosqUito species (Table 2.1) ~ g(netically competent to 
tr..uumit pathogens to a \lc:rtebrate host and competenli:e varin within populattons of a 
gi\lcn species (Gul»er el aI., 1982; CUrtis & Graves. 1983: Christensen &: Se\·(rson. 
1993). 
Anopheline mosquitoes an: tht c:xclusive vectors of human malaria. A handful of spccic:s 
pmIorninate as the most notorious malaria vttton:. but the species and foons involved in 
the transmission of human malaria worldwide are very di\lerse (Coetzte rl al .• 2000; Van 
Bortel rl oJ., 200 1). Malaria \lectors are often divided inlo primary and secondary 
,·«ton. bul this is rather unsatisfactory because a species can be a so-called primary 
v«tor in onc: area and a secondary vector or even a non-vector inanorner. White (1982), 
prefcrml to classify AnopMlu as main \I('"Cto~ (widespread, dominant vecton:) and 
subsidiary vectors., i.e. incidental \leeton which, by themaelves. are incapable of 
sustaining lransmission. or secondary \lectors localised in their distribution but sble 10 be 
principal Vtt\Ol'1 within their areas. The most important Yl!ctors art listed in '''3blc 2.2 
according 10 thc epidemiological zones (Fig. 2.4) of Matdonald (1957). 
The iMJman-biting index. (H81). whkh is lhc proportioo of blood meals taken from a 
human host. has a large impact on a mosquilo species's vectorial capacity for malaria. 
This is illustrated by the fact thal the world's mljor malaria vectors all feed 
predominantly on humms (Garmt-Jones, 1964). 
University of Ghana http://ugspace.ug.edu.gh
T.blt 2.1 AnClpheJes SpeCies of medical importance (Hervy el aI .• 19(8) 
Atboviru.ses 
All hrohiui Shokwe (Ivory Coast) 
All bnmni~J west-Nile (Mildagasc.ar) 
Allchruryi Rift Valley fever (Kenya) 
All COWlanl Rift Valky fever (Zimbabwe, Madagesc.ar) 
All d(}llflcola Wessel.sbron (Senega!) 
A" jluvi<os,. Middelburg (Senegol) 
Allfretl()M1VIUH We:s:seLsbron(Senegal) 
Anfuneslus O'Nyong-Nyong(Uganda. Tanzania. Kenya) 
Wesselsbroo (Ctntral African RepubHc) 
O'Nyong-Nyong (Uganda, TIIIIZIIIlia. Keny.) 
Wesselsbron(CameroWl) 
All. mot:tIIipoJplJ Wcst-Nite(Madagascar) 
AnrttaJCarensH M~ (Madag"",ar) 
An.nJli Dagaza, T.taguinc(ScnegaI;Bangui:Burkinara50) 
A"poIwIU Gomolca (Ccnlnl Alliean Republic) 
A"patJlDnI Rift Volley fi:ver (Modag""'"') 
AocWibe(Madaaascat") 
Anphnroensu Rift Valley r. ... (Kenya); W ....l sbron (Cameroun) 
Sindbis(Egypt) 
An pnt~1Ui.r Ngari (Senegal) 
All nJipu nd'iDeJ __ Wesselsbron, ChikWlBWlya and Gomoka (Senegal) 
FIIiU'luis 
V«tDr Placn Whtrc: lacri..i;;tcd - ----
~ - MIl~~;~fW. ooncrojU in thcAfrollOpicaJ 
'"lion 
AII.~ Main vector of W. bancroft; in the Afrotropical and 
Orientalrqions 
An gambloe Maio vector of W. hcmcrofti in the AfrotropicaJ 
reilon 
An anlhropohogus Vector of Brugla ItUllay; and W huncr()jll in the 
OricntaJregion 
An.pt,nMiani SecondaryvectorofW.banc:rojJiinMadagascar 
An. albimanU! Vector of W. Ixmcroft; in Centl"1lllfMl South 
~merica. 
University of Ghana http://ugspace.ug.edu.gh
Table 2.2 Malaria vccton of the wortd. Zone nwnbers and geographical naming follow 
Macdonakl (1957). Abbreviations for ~genera given in perenlhescs are: A. Anopheles 
s.str.; C. Ct!Jlia: K. K~rtes::ja; N, .VyuorJrynchus. Main (primal}') vectors are 
distinguished from subsidiary (incidental and local) ~tors by bold!ype. The data have 
been compilcd from Service (1993). 
Zone. ~---
(A.)!n,1Iont1 A. (N.) d/b,marllS 
A.(A.)quadrimllt:uJalus 
A.(A.)azt~c", A.(N.)aJbil.,r ", 
A.(A.)ptI1fCtllftocu/a A.(N.)aqullstl/is 
A. (A.)pseudopunclipennis A. (N.)arg,.rjta,su 
A. (N.) a/bI. .." us A. (N.) da,/i",; 
J SouthAmerKan -- (A.)puIMlDpu"cI~nnis A.(N.)aI6iMnls 
A. (A.) PUMIIm«IIl. A.(N.)tlqlNlSalls 
A.(K.}beI/O/Of" A.(N.)at'1O"'itOJ'siJ 
A.(K)cnt:i; A.(N.)brtuili,.",if 
A.(K.)neiva' A.(N.)tl.,lIng/ 
A. (N.)IIll1b.nw A·lN.j"lUk:.IU"""" 
A.(N.)/rllmnulalu 
A. (A.)atrtlfMl"VlIS A.(A.)socitalvvl 
A.(A.)meueoe A.(A.)sine11Jis 
A.(C.,pollonl 
A.(A,'Ilffl,pannU A.(A.)sacliarpI'1 
A. (A.' da'f/gr:T A.(C.)hul'd",o/a 
A. (A.) Ittb,.."dlM A. (C.,supe,plclus 
A.(A.'~n~1W 
A.(C.)aJici/acirs A.(C.)multicuior 
A. (C.)jlJMQtiliJ A.{C.)pharwnsls 
A.(C.,hispanioia A.(C.,Jerpnlil 
7. AfroTropical A.(C.)a,ableIlSG A.(C.)rnrnu 
A.(C.)!UC'UUf A.(C.)mOlfChe.j 
A.(C.)gambJ,u A.(C.)niI; 
A.(C.)m".iar A.(C.)pIJorrwlUiJ 
University of Ghana http://ugspace.ug.edu.gh
Tabltl.2 (:oat'd 
Vtdon 
A. (A.) :ICxhoTTWI A. (C.) p/tiJippiMf'UU 
A. (C.) ucomtw A . (C. ,ptJclK!, r ;mw 
A. (C.) a rtrt14{aru A . (C.) SltpMfUi 
A. (C.)cuficifaci~s A. (C.) sllltdok-uJ 
A. (C.)j1l1fl'iriis A . (C.) S14/H,piullJ 
A . (C.») e)poriefUis A. (C.)/tJsdla/w 
A. (C.)nti";",,,, A . (C.)vonuta 
A. (A.)nigerrimus A.(i:.)jiIlYiQ~ 
A. (C.)annu/aris A. (C .)jeypori.nsis 
A. (C.)culidfacitJ A. (C.) macuialus 
A. (C.) dirlU A.(C.)mil1imus 
~-- A.(A.)eO"lptl/'U - A.(C.)jl",·i"')/'i.~ 
A (A.)eona{dl A. (C.)jt}!pOt'lcn$i.J 
A . (A .)dor'klt./,. A. (C.) kwoJpllynu 
A . (A.) I.·",.·, A.(C.lludloHillt 
A. (A.)w,;,· rrI1Fl'" A.(C.)IfI«IIJIIIIIJ 
A . (A),.h.Jr/" m A . (C.)mDI'IK)IOIlU 
A. (C.) accolfj,ur A.(C.)Mlnlttf'u 
A. (C.) IuU4lH1UNJU A . (C.)pJlilipplC'Hi.\ 
A.{C.,d/rur A. (C.) tubplcrus 
A . (C.)$IuultlicUJ 
A.(A.)elf,hropopharJU A . (C.)bulal>uccflm 
A.(A.)Jinell.fn A. (C .)jtyporieruis 
A.(C.)ptJtlOIti 
A. lA )b.l<:r,,/11I A . (C.)kUl"WUFI 
A.(C)/f"Qulilypel A. (C.) loliensi$ 
A. (C)/.'lJllllfJ'IJe 2 A.(C.)plllfcluinllU 
A. lC.)h;/Ii A. (C.)suhpic/w 
University of Ghana http://ugspace.ug.edu.gh
--------------------------------------~4,__ 
University of Ghana http://ugspace.ug.edu.gh
In most JWlS of Africa, several vectors tranSDlit maJana \0 each location, in some cases at 
the same lime and in other eues during different ~asons, as for example in iJietmo, 
Senegal (Fonlcnille tH oJ .• 1997). Much variation can be observed between villages 3 few 
kilometres apart . Ln Central Africa. it is not rare to capture four different vectors (e.g. An. 
gombioe. An. . junes/us. An. "iii and An. mouchen) during Lhc same nighl (Fonlenille cl 
Locho. ..... I999). 
Anophelt'.f gambiDI sensu slriclo and An. arobiensis, the two most anthropophilic species, 
an: responsible for more than 3/4 of Ihe world Plasmodium !alciparum inoculations 
(Favia & Louis, 1999) IU1d an: thus the most efficient vectors in sub-Saharan Africa. 
However. they differ in the degree of anlhrupophilly and endophily. with An. gwnbiae S.s. 
sbowina an iocft:ucd. propensity towards endophily, endophagy, and anthropophagy 
eompued 10 An. orabiensis. Anop~/es gambiae 5.5, the nominal species, is thus 
consi<kred to be thc most importantmaJaria vector.south of the Sahara (Mekuria. 1983). 
Tk impon:uX"c tlr ..:ompdence of All gambiae as a maJaria veClor is well illustrated by 
its iJhhkllt.ll IOlroduction into Bmil, where in 1938. after series of .small but intense 
local tlUtbn:-3k.-.. it caused the worst epidemic of malaria recorded there, with over 14,000 
dcaths in less Ihan I months (Soper & Wilson. 1949). 
---------------------------------------.46 
University of Ghana http://ugspace.ug.edu.gh
2.3 The Anopheles Gambiae Giles complex 
The Anopheles Komin IH complex is the world's mOSI important malaria vector complex. 
The comph.-:t is a group of morphologically indistinguishable (isomorphic) yet geneticaJly 
distinct spt!Cio;!s IRaI vary in c.heir behaviour and veclorial capacity and hence imponance 
as malaria vectors in Africa. Colleclively all the members of the complex are referred to 
as Annphetes gambial? s.l (J'ensu Jato). Tbc complex comprises six named species (Gill;es 
a. Coeuce. 1987), one unnamed spedes (Hunt ~l aJ. , 1991:1) and several incipient species 
(Colu.zz.i tt oJ .• 1979; Favia tl 01 . 1997). lbt six named specKS acc: Anoplw/e.s gamblae 
Giles. An. artJbirmis Panon, All quodriannuJolw Theobakl. An. mtrus DONIZ, An. melaJ 
Theot*d and All bwambM White. The latest addition to the complex is provisionally 
designated All. quadrwnnulalus species B because of its close similarity to this species 
(HuntelaJ.I~8) 
The rccopitioa of distinct, cryptic spec;:ics within the All gtuftbiae complex stems 
from i!JIUdie& of lbc inheritance of insecticide resistance. Crosses among some strains 
of 'gambi«' produced sterile male offspring (Davidson. 1962; 1964; Davidson 
&. Jackson, 1962; Davidson et aJ .• 1967; Davidson & White, 1972; Davidson & 
Hunt, 1913). All 30 possible reciprocal crosses of tbe presently recognized 
species have been made in the laboratory resulting in fertile FI females.. but sterile FI 
mab, with two exceptions: An. qUOtirtaN1ulatU8 females x An. gambiw.· males and 
A .... quodriaNnJatw females x An. bwamba~ males yield lome fer1i1ity in F I males 
as wt:1l as females. The genetic distinctness of the six formally named species was 
confi~ by cytogenetic examinations lhil revealed fixed inversion difTen:oces 
among taxa which we very stable also in areas of sympatry (Colu.ui, 1966; COIU7.zi 
a. Sabatini. 1967. 196&. 1969). WhBe not &.s deftnitive as the inver.;ion data. 
aJlozyme studies abo confirmeJ tbe ,tnctic distinctness of the six species (Rullini 
47 
University of Ghana http://ugspace.ug.edu.gh
a: Coluzzi. 1913. 1978: Mahon eta/.. 1976; Miles, 1978). Most recently. diagnostic 
DNA probcs have beeudeveioped (Collins etal. . 1987. 1988; 5oon .. 0I . 1993). Thus 
lhcsc are 'good species' by most definitions.. ahhough the possibiliry of occasional 
~ific CC"" (\ow (inIrogr<ssKln) exists through fertile hybrid fcmales. Hybrids 
lwve been observed 1n nature bctwcl'n An. arabiens;s and An. gambiae in Tanzania. 
Kenya, Mali . and Nigeria (\\!hile. 1970; Coluzzi el 01 .. 1979; Pellve. ef al. . 1991 ; 
Toure cl al .• 1998), betwttn An. QTGbiensis and An. quadrlunnululU$ in Zimbabwe 
(While. 1974), and bttwtenAn. Ku",biae and An, ",e/as in The Gambia (Bryan. 1979). 
However, such hybrids are rare and comprise less than 0.10/. of the samples analyzed 
(Coluzzi elol. . 1979; Petrarca el 01. . 1991; Toure et Ql~ 1998). Nevertheless, there is 
evidence that this de,ree of hybridization is sufficient to produce introgression of 
icnetic material 
Speae:s within the: A1tOf'IwJts gambiCN Giles complex differ ttighly in their vectorial 
ClpllCiry for malaria. 8c:cause the sibling species appear equally sWICeplible to P. 
faldponun Welch infection (Takken el aI., 1999), differentes in the hwnan blood index 
(HBI) or the sibling spei:ics mostly detmninc lheir sa.tus ItS malaria veclors. 
2.3.1 Distribution and Importance of members of the Anophr:les gambJae sibling 
spedes com pIe. 
2.3 .1.1 !U'Oph.'H bwamb.!le 
lbis ~ .. IC:I is ~n to occur only an the vic::ini!y of the Buranga hot !lprings in 
Uwamt-..I Cnunty. Uganda. where it breeds in brackish water from geothermal springs 
Iwhlch ~l ~ ~ geothcnnal c::.)flIjllions for the dew~nt of its IIn-acJ 
tos~,. wilh 0Ibt:r halophilM: mosquitoes (White, 1915). They were first in\'cstigated by 
Haddow (1945) who~· attention to their hiological c::onlrUb with I)'pica! If,., gamb/(~ 
, .1. but was unaware of the spec.iHc:: taxonontic distinction (White, 1985). ArrvpJr.-frJ 
48 
University of Ghana http://ugspace.ug.edu.gh
bwambtw species differ from the other :5pC('C:s of the group by the presence of. wide ,-Je 
tpicalbondonlhefemal<paIps(Wlti.e.1985). 
LAn'X of All. bwtJllllbae an: associated with 'springwater' habitats having much rug.her 
conductiviry, muc:h greater concentrations of dissolved solids and slightly IUghcr 
IcmperalUI'C ant.! pH than 'normal' fresh water sites inhabited by larvae of An. K(Jmh;~ 5.5. 
(Harbach ~t al. • 1991). l.arval habital., are generally fuJly sun--exposed. sometimes shaded 
byl~· vqnation. Egg,sareparticuJarly abundanl in ponds directly fed by these springs 
"itt. muddy ~ often mixed and polluted by wild animals. Egg.~ arc regularly laid all yellt 
Adults of An. bwamhae occur strictly in. circle of 10 km around the hot springs (White, 
1915). During da)1ime, adult males and females rest mainly in the Semliki Fore!! on the 
buttrt'S\.C,j h.:a..~., ofliUJc trees, on faUen Joss and sticks, and under loose dry leaYeS 
(Vlhitc . 19&5) FcmaJC"5 also rest among 1M extensive plantations of bananas fringing the 
forest.lDd iDdoonat villages situated near the forest (White, (973). Theadultpopul3lkln" 
~~aJlyearrowxl. 
nus OppotWnlllt ~ is :uwa)'s very agpasive. Adull females al1ack people in roads.idc 
villqcs and fomt at tile caslem edge of the Semlilti Valley (lladdow el aI .• 1947. 1951 ; 
Lumsden. 1951, 1952; While. 1971;Gillett. 1985). In\"il~ female5emcrlnsidehouses 
- fcal on htmans. However. in the same locality. they can abo be' aur.cted by domestic 
pip.. Their activity is tlleDtiaJly nocturm.l, even lhougb m the fon.:st. females bile hurnms 
duringthcday(Lwntden,1952). 
49 
University of Ghana http://ugspace.ug.edu.gh
In nannJ conditions some An. /1waIItb« eduits have been found with both sporozoites and 
filariae (Whilt. 1985). Ho""'ever, this species is not suspected to play an important 
qJi~"'101ocical ro6c in the transmission ofpathogeruc agents of humans or domestic animals 
becaltlCofitsveryre:str'icteddistribution. 
2.3.1.2 AnopltttlftSquadriannu/.tus 
Anophll~R qllOliriannu/olus is found restricted 10 eastern ond southern Africa. It is, unlike 
to the other spcc~ relatively tolerant of the highland cold conditions and has been found 
in the hi(thI:UlJ~ of Ethiopia and extensively in southC'rn Africa (White. 1974), The 
Ethiopian population was recently recognised as a distinct species and has been 
dr:signlled as An. qlKKll'iannuJOItu species 8 (Hunt ~I al. • 1998). 
The cU-layina; sites an: lIimilar to those of the other frnh ...... lcr species of the RQmbi« 
tomplclt (White. 1974), The sites ate always sunny and consist ITI05I often of residua! 
puddles or ternpont)' or scmi-pcnnanent ponds. 
Abo. unlike the other member spec:ics, An. quoariannulurus i:. aggressive during the first 
half of the wahL Females of the spccin are considered to be strictly zoophilic (Wbite. 
1974; Gibson. 1996; Dekker & TaUen. 1998). Although house-resting adult populations 
oftht spttics have been obscrwd in South Africa and Zimbabwe (Hunt & Mahon. 1986), 
they "'ne found more frequently in buildings with canlc only. or with human and cattle 
togethcr(WhilC. I974). However,anthropophagic bcha\'iour notprc\iously recorded in 
thc=speoclrs A has been reponcd (Patcs rtol. . 2001). 
AnopMle$fllDdrimtnlllUllU has nnCf been implicaled in maJarill tntnsmission, althou&h 
n:ccnt laboratory o:pcrimcnls have shown that it i .. susccptibk: to P. julClparllfft (Tak.Jcm 
-----------------------------------------50---
University of Ghana http://ugspace.ug.edu.gh
dol. . 1(99). II is a150 susccpcibk to me hwnan filariaJ worm W. bancrofti (HUlII &: 
<lundeN. 1990). 
2.],I.J Anophe/esmelas 
The distribution as well as some taxonomic characteristics of An. meltJj facilitate its 
distinction from the other members of the complex. hs distribution area is limited to the 
coauofWestcmAfrica(Coluzzi,1984), 
Ahhough An. melas larvae can withstand very high salinity they usually develop in w.ten 
with salinity which is Ihc same or lower than seawater (AkogbelO. 1995). They are only 
rarely found in freshwater poods. and always in small numbers. The larvae of this 
littoral species cu:ntially colonize estuaries, lagoons and mangro\'c swamps (Coiuzzi, 
1984; Oiop el 01 .• 2(02). However, in Gambia where ltK.: hydrogrdphic nel pennits an 
upstram tide. this species can occur up to ISO km inland (Snow, 19K3). 
AduJ, All. nwlm populations are partkularly wutable and their tknsilics may vary 
tJvoua,hout the y~ (F0DIC'Ca d aI. • 1996; Diop et 01 .• 2002). During the rainy season. 
ilSdcMUY is innuenced by botbrainsandlides. It shows(ewfeed ingpreferences:can 
be: highly aothmpophilic in the presence of man and vet)' zoophilic in hi. ebsenc:e 
(Soo ....· . 1983; Akogbeto. 2000: Diop ~I aI .• 2002; Awolola t!I al., 2002). Anopheles 
",eiGS life expectancy i. short and il does nol seem 10 live more then I S days in natural 
conditions. 
lu 10 .... anlhrupoptul)' and its reduced lifespan make it • bIId malaria vcctor. In endenUc: 
malan • .rona, where An. Rillftbiot! s .s. is the main vcctor, !.he observed spoco7oilC' 
mdiccs 1ft It least 10 times smalter in An. mrlas (Bryan. 1983). b -en when An. mrlu., 
51 
University of Ghana http://ugspace.ug.edu.gh
is the only spec~ present, the plasmodial inclices observed in man are always very low 
(Akogbeto a:. Romano. 1999). These data confirm thai this species plays. negligible 
role: in malaria transmission. except perhaps when il is Vet)' abundant (Oiop ~I al. • 2002). 
2.3.1.4 Anopheleil merus 
AnopMlt'l menu is one oftbe species found in East and Soum Africa as well as on some 
Africanislandsintbelndinn{)ccan(Muirhead.Thomson,1951;Coluzzi, 1984). However. 
the distribution. as well as some: taxonomic characteri~ics. facilitate its distinction from 
the: members of the complex. 
AttOphtlu tJWnIS larvae develop in brackish waters but Wllike An. ~/QS. lhcy are WlIblc: 
10 withstaDd rush salinity (Mosha & Mutcro. 1982). This species is {\artLcul:vty abundant 
in coastal ~ Vt'hcre the Wvac are often folmd in bradosh water expanses, such IS 
Ltaooos ponds and S'o'ownps and seav..-atcr brought by tide diluted with rain (Mosha &. Subra. 
1992). ltcanpenetraleandswviveatgrealdistanceswhcresalrywaterpondsarepresenl. 
A1W.'IpIttt/~s maw is opporrunist when in search of a blood meal, although il has a clear 
zoophilic ptc(ermoe. Thus. in domestic environment, when cattle are present. homed 
caWe are much more: altraclive lhan man (Mutero el aJ., 1984). On Ibe other hand. 
fcmaJcs can be very aggressive to man both indoors and outdoon when canle are Ibsenl 
Precipitin Iests on engorged fema1c5 confinn dJcse dill in lhal, those coJlecled inside 
houses had fcd 00 man and those found outside had fed nearly exclusively on bovids 
tMutero el Q/. . 19&4). When herds stay overnight ncar breeding pools., it may be toLally 
.lhsenl from !he villagC1 though it is both exophiltc and mdoptUlk (MUlero el at., 1984; 
Busbrod,. 1979). females ratia& outside an: usuaJly found along roots and trunks of 
University of Ghana http://ugspace.ug.edu.gh
mangrove trees as well as in natural holes (termitaries. crab-holes. etc) [Mutero Itl aI., 
19K41. TIle survival ralc of this A1tOpIwJ~$ is low. 
AnophlJe$ IMrus lJ"aIlSfIlits malaria and bancroftian filariasis perfectJy in the laboratory 
(Hunt & GUIlders. 1990). However. it has been intimated that it is. bad vt.'Ctor for 
t.ncroftian fllariasis in nann.1 conditions becaux of its shan lifespan and feeding habits 
(Southgatc &. Bryan, 1992). The sporozolic indices of An. IMnu are always smaller than 
those of An. RumbiDI &..J., although An. merus may still playa role as secondary vector in 
IoeaIiticsw~itispart.icularlyabundanl 
2.3.1.5 Anopheles ....b lensis 
A,.opltdts arublt'tlJIS is the member species of the complex with the widest geographical 
distribution. It isfouodin the Afio..tropicalregjonin lheSahel,onthe plateaux ofsouthtm 
Africa and in Madag.ascar. It isaiso presenl in thecountric:s bordering the Red Sea and Aden 
Gulf, and most oflhe islands in 1bc Indian Ocean (Abdullah & Merdan. 1995). Moreov", it 
pcnctrwtes deeply in humid savannas, but not into the forest. Where it occurs in the 
eqUiltorialrainf~trcgions,suchasinBeninCil)"inNigeria,ilisusuallyas.sociated 
with a history of CKlcn.~i\c lam! clearance (Coluzzi ~l aI., 1979), although lhiJ is not 
always the case. Anuphelts urob,~nsis shows evidence of being better adapted to less 
humid environments, and is often the predominant or only member of the complex in arid 
reeions of We~ Africa. fthiopia and Sudan (Gillies & CoelZl:e, 1987). It is al$o more 
commonly found rnting outside houses and in plates such as granaries. than An. galflhiae 
U.; this also reflects a bc11~r adaptation 10 Jowerhwnidity (Donnell) & Townson, 2000). 
This species if found mainly in the drier sa\'3nna an:as with rainfaH ofk:ssthan 1000 mm; 
• fact "hicb is in kccpinR "ith its knn"'n habital preferences (Coluzzi rl ul.. 1979. 
----------------------------------------·53 
University of Ghana http://ugspace.ug.edu.gh
Gillies a. Coctzee. 1987). Anomalies of Art. arQ/)JttltSu occurring in desert areas can be 
explained by Ihc:ir associatioo with rivcr systems. such as the Niger in Mali, and the Nile 
in Sudan. In most ca.'!es. the ecological context or the distribution. as well as some 
tlXODOmic characteristics permit its distinction from the other members of the cornplel(, 
However, these crileria by themselves make it still impossible 10 distinguish it &om An. 
gtIIfIbiMs.swhich it oct"un in SYmr-trY with over most ofilSdistributioo are&. 
Tht ega-laying sites of An arub,t'nus are simjiar 10 those or the other freshwater 5peCies 
of the complex. Tbese are temporary water collections, usually small. shallow and 
exposed to the sun, without vegetation (Gillies & DeMeilioo. 1968; Gillies & Coev.C'C 
1911). These conditions are foWld In waters accumulated dwing !he rainy season, in 
dcprnsionsor iD avitics such as ditches., ruts, foot or hoof.prinlS (Beier eta/. . 1990). 
(While ~I oJ., 1972. GiLhc:ko el aI., 1996) and garden wells (Robert et al., 1998), 
Howntr. consistent differences in habitat u.se by either An. gambiac u. or An. 
orobknsu have 1\01 beenobscrved, and both spa;iesoften have been found occupying 
the same hahitat (ScMce. 1970: While & Rosen 1973. Service tim., 1978. Charlwood & 
F.doh 1996, Minakawa tloJ. • 1999). Population density of this species varies seasonally in 
relation Co rainfall. It lncrea'ICS quickly with the firsI mns and the maximum density is 
Itachedat the end of the mny season (While tl aI .• 1972; White & Rosen. 1973) and then 
dccrt:b:saslhearmporary breeding sitcs dry up. 
fhe host p~ference of An. arabirlUu is more dlfficull to gcnt,.lisc . Depending on 
lhe ;J\·:ul:J.bility of humans relative to alternative hosts, the HRI of this specil:S is 
,"("ncf311~ hilh: but in vililges wbc-re cattlc are abundant, Ihis proportion cln drop in 
University of Ghana http://ugspace.ug.edu.gh
(avour of bovid (<<CIs (Coluzzi elol. • 1979; Gillies &: Coetzee. 1987; Githeko ~I 01 .• 
1994). For example, AIt. orobil!lUlS is caught on man and donkeys in Niger while its 
trophic preferences are the: same as those of An. gamblOe s.s. in some localities in south 
SeoegaI (Cbartwood el oJ., 1995). Although one explanation would be that All. 
orob,enslS i~ simply a more opportunistic feeder. its feeding pattern is not always 
proportiooal to host abundance. East African populations are generally less 
anthropophilic than West African ones , while in Madagascar mainly zoophilic 
populations are known to occur independently of the human-bovid ratio (Ralisoa 
Randrianasolo & Coluzzi. 1981). 
AnopJwlu ~abiefUls is one o( the main malaria and bancroRian filariasis vectors in 
the: Afro-lrOpicll RelioR. together with An. ~ambiM s.s. and An.. funeslus . It transmits 
~se diseases in most p&ft.S of the continent, on Madagascar. on Reunion and Mauritius 
islands., IS \\01:11 as on the Arabian Peninsula (White. 1996). This species is often more 
.t.oophilie. has a lower lifespan than An. gomhi« s.s. nnd thU1, it is generally a less 
efficient vector. However it resists drying better and may be the principal vector of 
malaria and barK:roftian filariasis in dry areas. 
2.3.1.6 Anopheles gambia. sensu srrlcto 
Anophe/~J 1(umhIClt' sensu str;c(o,1hc nununai taxon. is the most anthropophilic member of 
the complex and the main malaria vector in sub-Saharan Africa. Genetic and behavi0W'31 
\·aria1ions-.ithinlh, :. ,pcdcshavebecnwelldcmun.\tr.lledbyColuzzillal.(1979,198S) 
\\oM JC"iCnhcd the diffCTCnt c)'lologieal fonns (named with the: non-Linnean nomenclature 
ru~t. S,muma. . Bamaku. ~Iopli, Bissau) which show restricted or no inter·breeding in 
the f.eld (Bryan ~I aI. . 111M2: Coluzzi I!t aI., I98S; Toud et 01 .• 1983. 1994, 1Y'}8) ,!lid 
---------------------------------------5'5 
University of Ghana http://ugspace.ug.edu.gh
whose distributlOll depmds on environmental fllCtors of climace. breeding sites., etc. 
(Towe eI m.•  1994). In sympatric areas, hybrids bctw~en SaVaMI. and the other forms 
hive ~cn observed at frequencies lower than expected. but no individual, carrying 
h..-ICrozy(!O~ complements of the Mopti and Bamako inversions are seen in nature, even 
Ihough the rwo forms produce viable offspring Wlder laboratory conditions (Coluzzi tl 
01 .• 1915; Pcni&Da tl al. • 1986), Gmotyping X-linked rONA of An. gambiot 5.5. has led 
10 I~ characteri ... .1otion of two molecular fonns (M and S) that differ in both the 
transcribed and nontranscribed spacers in the rONA repeat WlJt ( ....v ia tl aI. • 2001: della 
Tom rial. • 2001; Gentile r/ol., 2002). The relationship bet .... ttn the M and S molecular 
forms and the chromosomal fonns - defmed according to nonrandom associations of 
im'enions in chromosome 2 (Coluzzi el aI., 2002) - varies according to their ecological 
and pog.rnphlc distribution. In some areas of West Africa. such as Mali and Burkina 
Fuo. there is a one-1ooOne correspondence betwttn the M molecular fonn and the Mopti 
ctwmotomat {ann. Similarly. the S moke<:ular fonn alv.'ays com:sponds 10 the Savanna 
Of Ba.mako ch:romosrllll.il fnnn (Favia ~, 01 .• 1991). In other areas of Wtsl Africa. lh..is 
dear COI"I"e!JlOftcncc breaks dOWD (della Torre et aI,. 2001,. Although inlerbrttding 
bdwcaI M.nd S forms yields fertile progeny. M-S bybrids are rarely obstrvtd in nature. 
When: ~ rOnN overlap in time and space. &he: rate of heterogamous insemination is 
31% (Tripel ettll. . 2001),elearlydemonstratinatbe exislenccofa pre-mating barrier, 
albeilaninc:omplCk'onc 
AltOJI'Nlel RQlNhw(' 5 S 15 t!xtremely versatile regarding loler.&nec to • wide: '<'ariety or 
micro- -.1 mac:ro-enllironmt:ntaJ condition". as cvidenced by its broad geographK 
dUcribubon. It I~ widespread in Africa. though better _pled to welter regions than 10 
"~nna an:as (lind5.ay~, aJ •• 1998). The rorest and Bissau chromosomal forms prevail 
In humid and coastal an:as of Wesa "rriea and could both belong 10 • single polentially 
---------------------------------------~---
University of Ghana http://ugspace.ug.edu.gh
paruntcuc unit widely dimibuled also in East Africa. The other three taxa, the Bamako, 
MOpli and Savanna chromosomal forms. arc more adapted to dry environments, are often 
sympatric with evidence of as.wrtative mating. and overlap the diwibotion of An. 
uTuh,c'IUU (Favia ~I aI .• 1991). The Bamako chromosomal form nppeaf:S 10 have a 
nn'nlX ecology found in the upper Niger river basin. from southern Mali to northern 
GUIIX'-'; it generally coc:xists with Mopt; and. less frequenIJy. with Sa·.annll. The Mopli 
chromo'SOmai form is lhe dominant taxon in the iMct' delta of the Niger river and in all 
impled areas in Southern Mali and eastwards in Burkina Faso up to the middle Niger 
river buin (Towe d al. • 19(4). lbe Savanna chromosomal form has a wider 
distribution overlapping boln the Bamako and Mopti ranges. It is generally absent from 
the Slbelian ;.ones and the inner part of large irrigation schemes where Mopt; becomes 
dominan! . The Savanna fonn has also various contact zones with Bissau-Forest 
populations and studies carried out in some of these areas showed incomplete 
inlergradation (Bryan elai. • 1982; Petrcl:aelol. . 1913). 
The Larval 1lAge5 of the Mopti form of An. gombiae 5.5. occur mainly in man~made 
habitats. such IS impled areas (even in the dry season) while Savanna and Bamako 
forms hnc • lCndeftCy to breeJ in more natwaJ sices. exploiting r.lin-dependent pools fot 
I. ...... dt:\'c~pment. These atTect their ~tiaI and 5e8SOnai distribuliun, since the Mopti 
form will breed Ihroughoutlhc year and can therefon: displace the other forms in imgaled 
areas, leadin~ to chang~ In the: patterns of malaria entomology (To ore el uJ.. 1994). 
There is C"o'idenc:e lhat margjnaJ arid envirorunents of Burkina FIlSO. where the Savanna 
~ Mopci ehrorno..nal forms correspond to lhe S and M molecular (orms rnpe<:tively 
(tnla el al. . 1997).ltlt:se h-.o tax. contrast signiftcanlly in the way they exploit limiting 
rt":w:tul't'cs.. such as lanai hr'C\.'ding and aJ\lh n:sling habitats. l'he M molecular form 
,ho.ms lh~ dosc'4 as.s.xiallon with th...· ,juffil!slic environment lDi J.-vaJ habitals created 
University of Ghana http://ugspace.ug.edu.gh
by human activities. whereas the: S form is more frequent in rain-dcpendent temporary 
bree-dingsitcs(Faviaelol., 1991). 
In nuny areas in tropical Africa An. gamhiae s.s. is the main ma.laria vector~ however it is 
responsible for only part of the transmission in Africa as I whole . Anoplwlrs xamhiae s.s. 
is highly anlhropophagic and rests predominantJy in housn(endophilic) and w;thin the 
complex. it has the highest vectorial capacity . The: ,"cclonal potency of An. gambltx s.s. 
siems from its strong association with humans, that is, its preference for biting hwnans 
'::'I:acerbalcd by its capacity to exploit changes in its natural habitat induced by Homo 
soplrns (della Tom: rl al. • 2002). It is an efficient vector of malaria. banc:roftian filariasis 
andsomtarbovifUJlCs. 
58 
University of Ghana http://ugspace.ug.edu.gh
2.4 Symbiotic Associiltions 
The teM symbiO!lois. which simply means, "living together", was first coined in 1879 by a 
German botanist. HeiTVich Anton Dc 8ary to describe the relationship betwc:en certain 
spcciesofalgM:and fungi 10 farm lichen (Ch,,·n~. 1973). It is bowcver<kfined as any two 
crpnisms living in close association, commonly one living in or on the body of the other . 
• c:ootrasted whh -free living" (Robert..  & Janovy, 1996). 
TheorilJinal defmition of symbiosis did nolinclude ajudg.mcnl on whether the partners 
bettefit or hann each other. Currently, most people usc the term symbiosis to describe 
interactions between the symbiont (the smaller organism) and the host (the larger 
organism) from which both partners benefit; this is also called mutualism (Gray. 2000). 
Ir~ is a Mptivc effect on one of the pat"tIlen. it is called a parasitic symbiosis and if 
I~ is no beneficial or negative effect it is. commcn:.al s)mbiosis (Gray. 20(0). These 
c_-cuI definitions ate not always easy to apply In nature. For example, the bacleriwn 
Ps~udo".onas aef1lgmasQ can he fmmd on the skin orhwnans and docs not cause diX'a5C. 
and hence can be called. cornJlYnsal. but ifthc penon has. severe bum. P. Q~rug"'mQ 
caD eautC infection and become a pathogen (Gray. 2000). 
SpecifICally. under the broad hetlliing of symbiosis, types of associations known as 
phornis. commensallsm, parasitism and mutuaJism can be: identified. Whether an 
a.uoci.tion is a mutualism, comlllCnSalism or pwuitism depends on the relative 
~stm1glhs" of the partncTs and the balance of po .....e r can change over lime (Gray, 2000) 
The .ssod.oofts can also differ in their !lp8.liaJ relations (Ro~rts 4 JanoV)', 1996). If one 
ofthc orpni.snu exists outside the ceUs of the other, the rd~ion is called ectosymbiMis; 
if one o(thc O{g~isms lives in.. . ide the cdls o(thc other. it isc.Ued endosymbiosis. 
----------------------------------------5'9 
University of Ghana http://ugspace.ug.edu.gh
SymbioSiSmayoccurbetweentwokindsofplanl.s(c.a.lichen.CormingalJlandfwlgUS), 
I\oH' Kinds of animals (e.g. herbivores and cc:llulosc-dignling gut microorganisms), or a 
rl.ml and an animal (e.g. fig and fig wasp). 
2.4.1 Types of symbiotic associations 
2.4.1.1 ....0 ,...,5 
I'horc1is is. fairly loose fU~ialion and is • specialized form of commensalism in which 
OQ( orpnism (phoronO. usually smaller than lhe other, uses the larxer organism as a 
transport host (Cbena. 1973). There is: no physiological or biochemical dependence on 
the part ofeither .,.nicipanl (Roberts & Jaoovy. 1996). Examples are bacteria 00 the legs 
ofaOyorf'tmgussporesoothcfeetofabcet1e. 
Com.mtnY.lism, "eating together at the same table", is an association in which one 
member. usually the smaller, derives benefit from the association. whcreu fOf' the other 
membtr. the association is neither benelkiaJ nor bannful (The lIulchinson Dictionary of 
Sdcnce. 1998). The relationship can be that of5haring space, substrate, defence, sheller, 
transpOrt or food. M~ often these a.s.socia(ions arc facultnth·e. that is the commensal 
will not !fie if it does not ~nler into the associAtion as oppo~d 10 an obligate auociation. 
A claWc example or commensalism i.s the unique relationship between Ihe pilot fish 
tNa'K'u/~~ dIlclor) and the shark (Cheng, 1973). lbe pilot fish accompanies the shark in 
• (ree·swinuning momncof. eating the fragments of food thlt beccme .vailable as the shart 
feeds 
---------------------------------------.60 --
University of Ghana http://ugspace.ug.edu.gh
2.4.1.3 Mutualism 
Mutualism, as lbe named suggests. is an association in whkh both orgarusms derive 
mutual bmcfil (Cheng. 1971). These associations are frequently very intimale and 
~igllOfY.since . inmostcasestphysiok>gicaidependellCehasevolvedtosuchdegrecthat 
if the tWO organisms are separated, neilher will survive (Roberts &. Janovy. 1996). 
Macinnis (1976) defines mutualism as when each ofthc interacting species functions as 
bothahoscandaparasile. Tennites,and lheceliuJose-digcsting prOlozoansor bacteria 
tMI Jive in their inkSlines. are an excellent example of mutualism; neither organism can 
swvivc:wiitMlucthcother. 
P..sililm is an association in which an organism (parasite) living in or on another living 
orgarusm (host). obtains from the host part or all of its organic nutriment. usually to the 
detriment aCme host (Roberts &. Janovy, 1996). Parasite! may CIUse: rnechaniea.l injury • 
.sucba boring a bole into lhehostordiga.ing inlo its skin or other tissues. by stimulating 
a d.mqing iJlf1ammatory or immune resporuc. or simply by robbing the host of nutrients 
(RobcrIs & Janovy, 1996). The relationship may be permanent, as in the case of 
lIpCwonns found in the intestines of manunals, or very temporary. as during the feeding 
ofmosquilOes. leeches. and ticks on their hosts' blood. Pa.rusilism is said to be obligatory 
bccaux the parasite cannot normaJl) survive if it is preven1ed from makina contact with 
th.host(Chcna.1971). 
2.4.2 Modes of how symblonts get together 
All symbiotic auociations mu.'il meet the challetlge of maintainin. their rartncrships over 
SUCCCS.SIVC rentnbons. In The \o\cll-known associations bctv.(",;n bacteria and metazoans., 
61 
University of Ghana http://ugspace.ug.edu.gh
the bacterial symbiont must 5IOftId1ow become intimately as.soeialed with its metazoan 
host There are three major modes of getting this accomplished . 
("he firsc mode involves external transfer. Here, the metazoan host is born free of 
becterial symbiorns and becomes infected from the mvironmcnt . This appears 10 be the 
case of the tquid infections with VibrlofiscMTI . lbesquid is bom without mest OOclcri3 
and IICquires them from the water (Nuholm el aI.. 2000). The 5econd method is 
horizontaJ (in1ertaxon) transfer. Here. the hosts are born frce of symbionts. but they 
atqui~ them (rom members of the pamlW generation. The third method. venieal 
lrVISfer, involves gametes (most commooly, the oocyte) carrying the symbionts 10 the 
nexlg<nerahon 
Several types of vertical transmission mcc:hanisms have been oh~rved (Krueger et al. . 
(996). FcmaJn of the oligochaete worms lnani(/rilus /euhxiermallIS and I. planus appear 
1O infect thc:ir offspring by smearing symbionlS onlo their eggs as the eggs pass through 
symbionl--coaled genital J*Js u they leave the mother (Giere el oJ .• 19(1). C~in clams 
(such lIS SoIcM)!tJ reuii, S. velum. and CaJyplO~ftQ J'O}YJe) prod~ oocyles that contain 
the microbial symbionts already in lheir cytoplasm. Polymcra....e chain reaction techniques 
and electron microscopy have found bectcrial symbiontsin Ihe host ovaries, eggs, and 
larvac(Endo. .... &.Ohta. 1990: Cary, 1994; Krueger el al. • 1996). 
62 
University of Ghana http://ugspace.ug.edu.gh
2.4.3 hcterial ~mbionts 
11 1 ~ccl1sof..ukaryotesconstitule the sole habitlll Ora vaJt and varied arTIly ofspeci.lised 
pluk:lrypli"Iineagcs. 
!'he bacteria that inhabit animal cells can be dividtd into several groups (Moran &. 
8aumann.20(0). The most distinctive an= 'primary' symbiontsthat reside mthi" the 
specialised host cells called bacteriocytes. They have reciprocally beneficial - often 
tttiptOCaJ oblip&e • ~13tionships with hosts. and occur in many terrescrial arthropods as 
wen as some marine invertebrates. 'Seconduy' symbionlS and intntccllular patho8en~ 
aremon: sporadKally associated with host individuals and vary in the li:ssues O(cupicd. 
The effects on hosts are u.',"all), nul known. therefore there is 00 clear demarcation 
between symbionu: iUld pathogens; a wide range of interactions; certainly exists between 
non·~e associates and their 00sts (Bandi eI aJ .• 1999). Infection of hosts cln be 
strictJymatemaJ..matemaJwithoccasiona.lhorizontallT3nsfcr,orenlirelyhorizontal,and 
thert- is no absolute correspondence between mode of Irnnsmission and effects on ho~ 
fitness (Mo,," a. Bawnann, 2000). 
Close auociations of bacteria with higher organisms can either have a mUluaJistic. 
comrnt:n$I.I Of parmitic chanc:ter (Hooper&. Gordon, 2001). However, although the ftnal 
outcomcofsuchassociationsiscnrirely differenL the analysis of the inte1attionsbetwttn 
these microorganisms and their host species reveals many common themes. For eltamplc, 
in all cases, the bacteria require sp:c:ifac surface structures on the host to adhere to the 
host (Gross, 2002). They also have tocompetc for nutrients, and modulateoradapc lhcir 
lenetJc~ll)thcspec:ificn:quircffil:ntsonorinsidclht-hostorganisrns 
MOf'C'OVCr. symbioors md parasite-scan colonize: ~m..i1ar niches. either in extracellular 
locations or inside _ cells. Accordingly, many s~mhiont5 ha\C close relalives kno~l1 
University of Ghana http://ugspace.ug.edu.gh
to be UnportanC. ~tbogeos for man or animals. For example. many obligate intracellular 
symbionlS of insects. helminths or trypanosomatides an: closely related 10 pathogens such 
as Rieul/sill spp .. Coxiella burnetii, F'rtmclsello Ihulariemis. IkNdetello bronchiJeplico 
or the ENerohot:ler~(W (Gross. 2(02). 
MolccuJar phytogmetic analyses have revealed the rtlationships ofcndosymbionts 10 
several groups of t.cteria (Mann & TeJang. 1998). Several, including lhe associates of 
aphids (Baumann el aI. • 2000). tsetse flies (Chen el 01. . 1999), psyUids (Spaulding & von 
DohIc:n. 1998). anLS (SchrGder elaJ. • 1996)andsomebivalves(P«kerol.,II)<)M; Distel. 
1998). are relac.ed to the: enteric bacteria. within the "(-Proleobaclerlu. Other groups of 
bacteria have also given rise to endosymbionts, such as the Flavobacteria in cockroach~ 
(Dandi eloJ. • I99S). Amon~ the y-Proteoboct~riQ, endosymbionls of different host groups 
hI~ evolved as independent lineages from nonsymbiotic bacteria (Moran 8£ TeJang. 
19'j~J 
64 
University of Ghana http://ugspace.ug.edu.gh
2.5 Molecular Genetic Analysis 
2.5.1 Extraction and purification of genomic ONA 
The exlracllOn ;utd purificalion of DNA I~ a k~y step for most protocols in molecular 
biology and all recombinant DNA tcchniques. MOS1 DNA in vivo is present in association 
",ith RNA and proteins. Proteins directly involved in me process of gene c:xpn.'ssion. 
such as RNA polymerase regulatory proteins, interact with DNA in vivo to form 
nudcoprOlcin complexes. DNA polymerase, DNA ligase. various unwinding and 
suptrCOiling enzymes, recombination and repair enzymes, and proteins involved in the 
initiation or maintenance of DNA replicalion are also associated with DNA In \lItH) 
(Rodriguez" Tail. 1983). VariOWi tissues are known to contain large quantities of 
proteins. cell wall materials. phenolic compoWKls. etc., which tend to contamirwle the 
DNA extract. It is. therefore neces..o;.a.ry. first to isolate crude complexes of nucleoproteins 
from cells aod then purify lhc: DNA by separating it from proteins and RNA. Because II 
plcthcn of mel hods exists for extraction and purification ofnudetc acids. researchers 
usuallychoosc tbe most suitable technique depending on; (i) Lhc 11llgC1 nucleic acid 
(ssONA, dsDNA, total RNA. mRNA. etc.), eii) source organism (mammalian, lower 
cuk.aryotes, plinlS. probryOICS and vinL~s). (iii) startinlJ material such as whole organ, 
(issue. cdl eulbft. blood, (iv) dcsiml results (yield. purit)'. pwification time required) 
and (v) downstrnm application (PCR, cloning. labelling. bloning. RT·PCR, eDNA 
s}nlhcsis, RNase protection assays, etc.) [Roche, Applied Science. 2002). The quality of 
Cl(lnCled DNA thus depends on the source material (tissue). presencel absence of 
contaminants. method of extraction and puriftcation. and the precautions taken to 
mainWn the intqrity oflhe DNA. 
The buk Sltps In\'o!\'ed in DNA extraction and purification are; (i) the release of soluble. 
high molttular \~icichJ: DNA from dlsrupc.ed cells and membranes (ii) dis.soci.llion of 
65 
University of Ghana http://ugspace.ug.edu.gh
DNA-protein compkxes by dmaruration or proteolysis and (iii) the separation of DNA 
&om other macromolecules (Marmur, 1963; Rodriguez & Tail. 1983). Often. the ideal 
lysis procedure is a compromise. It num be rigorous enough to fragment the complex 
sta/1ing malcriaJ (e.g. bk>od, lissue), yet gentle enough 10 preserve lhe target nucleic acid. 
Commoa lysis procedures include: mecharUca1 disruption. chtmicAl lrtaunent and 
enzyrnatie:digc:stioo. Mechanica.l dLomJpthlR methods include grinding. hypotortic lysis 
and freeze-thawing. Examples of chemical treatment methods include detergent lysis. usc 
of strona cbaotropic agents such as guanjdiniwn thiocyanate and cesiwn Irinuoroacctatc 
and thioJ reduction. Enzymatic digestion methods include treabnenl with either 
Proceinue K or Prona.tc. Cell membrane disruption and inactivation of intracellular 
nucleaes may be combined (Roche Applied Science, 2002). For instance. a single 
solution may contain detergents such as sodiwn dodccyl sulphate (SOS) to solubilize cell 
membranrs and strona chaotropic salts (for example. ethylene diamine letr:l IIcetate. 
EDT A) 10 inactivate intracellular enzymes. After cell lysis and nuclease inactivation, 
ccllullrdebris rnly easily bc removed by filtrationorprecipilalion. 
Methods fot purifying nucleic acids from cell extracts are often combinations of 
curoctiotvpreeipitation, chromatography. ctntrifuption. electrophoresis. and affinity 
~paration, Solvent extraction is often used In eliminate contaminant!!; from nucleic acids. 
"ficr cell lysis. the cell debris and proteins are removed by denaturation and 
centrifugatiOll. T1w:re ~ sevcraJ mc1hods for deproteinilina the lysed cell suspension 
;md ~ include shaking with a mixture of chlorofonnfiJOam),lak:ohol (Mannur. 1963), 
all)'mlltie cSt:p.btion of the proteins with Prona..-ce Of Proteinase K (flotta &. ~1. 
I96S).and shakinl with phc-nol (Kirby, 1968), The~noldisruPlScC'lIularinlegrit)'and 
dcna,u~, ptUteins. and the final extraction with chlorofonn remo\es traces of phenol 
Protein in tht I)'sa&e can also be so!ubili,cd by treabnmt '-"ith SOS, Degradation b~ 
---------------------------------------.~---
University of Ghana http://ugspace.ug.edu.gh
DNAse and divalent meta! ion contamination is prevented by the presence of cbclaling 
lIIeotsand by the actionofSDS (Mannur, 1963). The DNA is sel«lively precipitated by 
the addition of ethanol or isopropanol. Precipitation with aJcohol tervcs to concentrate 
the high mo~uJar 'Neight DNA whilst removing the small oligonucleotide! of DNA and 
RNA, dctcrgcnl. and the organic solvents used in the removal of proteins (Rodriguez &. 
Tail. 1983). A subsequent wash with 70% ethanol, followed by brief centrifugation. 
removes residual salt and moislun= . To further purify the DNA, the enzyme ribonuclease 
IRNa.c) may be used to digest !.he contaminating RNA (Marmur. 1963; Rodriguez &. 
lad. 1983). I[the amount of target nucieic acid is low. an inert carrier such as glycogen 
caabeaddedlOlhc:mixluretoincreascprecipitalionefficicncy 
2.5.2 Restriction endonucleases 
The class of enzymes known as restnction endonucleases have played a key role in the 
developmtnt of recombinant DNA technology. These bacterial enrymcs possess an 
c.ndoauckuc activity which is directed to a specifw: sequence of bases in double-stranded 
DNA. In IWure, they serve to protect bacteria from the possible incorporation of foreisn 
DNA into their genomes by digesting such material (Arber &. Duissoix, 1962). The 
battnium's o~n DNA is protected by being methylated on A or C residues. whk:h mtden 
il UI\I\'ait.ble for digestion by its own enzymes (Smith & Nathans. 1973~ Watson~' "I . 
1992). It has been sugested that these enzymes may also playa role in promoting "site-
spccifk iIIeg,itimatc m:ombiR3lion", allowinB incoming DNA to be delved mel 
incorpon&edinaoiliectwmosome(Rodriguez&' ·fait. 1983). The ItTm MrestrictK>n- arose 
beaux it wa originally found that certain bacleriophqes "uuld not grow on certain 
biKlcnal "trains; hence they were -.aid to be restricted. Invcstigation of this phenomenon 
~\"C'aled lhat it wa due to the actIOn of this class of enzymes (Arbcr, 1979). 
University of Ghana http://ugspace.ug.edu.gh
ReMc,ion enzymes have been classified inlOthree groups. Types I and III enzymescany 
modification (i.e . methylation) MId ATP-dcpendent restriction (cleavage) activities in the 
same procein (Sambrook el 0/. . 1989). Type III enzymes cut the DNA at the m:ognition 
site and then dissociate from the substrate; \\hile Type 1 enzymes bind to the recognition 
xquenc:~ hut cleave at random siles when the DNA loops bock to the bound enzyme. 
Neither Type I nor III restriction enzymes are widely used for mokc ular biology studies. 
Type II restriction/modification systems are birwy systems consi sting of • restriction 
endonuclease that: cleavcs a specific sequence of nucleotides. and a separate methylase 
that modifies the same recognition sequence (Sambrook el oJ. , 1989). These enzymes 
recogniu: specific sequences of four to eight base pairs in length (Watson ~t lIl., 1992), 
The: 5a:jucnca in the ty,'() strands of DNA that are recognized by the enzymes possess • 
Iwo·rold axis. of symmetry (Rodriguez 4 Tail, 1983; Sambrook et 0/. . 1989), The 
locatioaofcLeavage sites ....i trua the wsofa dyad syrnrnrtry differs from enzyme to 
enzyme. Some enzymes make cuts which ate exactly opposite in the Iwo DNA strnnds. • 
.,1hIllbe ends are said 10 be 'blunt', Otberscleavc each strand at similar locations on 
opposilc sides of Ihe axis of symmetry. creating fragments of DNA that carry protruding 
single-stranded termini (Rodriguez &:. Tail. 1983; Sambrook et aJ. 1999). 
Restriction enzymes thIt cut specifIC sequences havc been is.olaled (rom several hundred 
bacterial strains. and O\'cr 150 different specifIC cleavage sites havc been found (Watson 
~l oJ. . t 992). 1.0 order to simplify the naming of these enzymes, a nonk:nclature has been 
dcwlopcd that is b~d on an abbrni.ation of the RiIlIlC of the organism from Which the 
\'n~'mc was isolated (Smith &. Nathans. 1973: Smith, 1979). The first initial of the genus 
~ thc first two iMia1s of the spcciC$ form the basic name This may be foUowed by a 
~n designation. when the enT.) me is prescfU in a .specific strain, or. Roman numeral (0 
--------------------------------------.68 
University of Ghana http://ugspace.ug.edu.gh
diff'emttialc ~ from the same source. For example, HM II is one of the three 
enzymes purified from the 5b'ain H«mophtllU MgypllclU and //inf I is the enzyme 
purified from lloemophilus influelfZQlt strain f (Rodriguez 8t Tait, 1983). In general. 
diffcrrnl restriction enzymes recognize different sequences. Ilowever, there a.re many 
C'xamples of mzymes isolated from different sources that cleave wilhln the same target 
Kqucnces. T'he:te an: k.no\\ n as isoschizomeT'S (Rodriguez & Tail. 1983; S3mbrook el 01 .. 
1989). Examples of such enzymes are Hind Ul and Hsu I (Rodriguez &; Tail. 1983). 
2.5 .3 Restrict60n fragment length polymorphis m (RFlP) 
Polymorphi:w of nucleotide scqucncclii in the DNA of SCH.:ral organisms has proven 
useful In distinguishing between strains and determining then relaredness to one another 
(pancrson.tHyypia.1985). 
t 'ntil recently. polymorphisms could be detected only if they were expresxd by 
difTel'C'ras in w behaviour of • prOle in, for example, by differenc~ in enzymatic 
Ktivicy IX electrophon!tic mobi lity. This situation changed dramalicaJly with lhe 
realisation that sites recognil£d by restriction endonuckascs could be polymorphic 
(WIllSon ttl u1 .. 1992), This is because mutations would haw caused me loss of sites at 
which. panicular restriction enzyme can act. Since these sites au only present in the 
genome of certain species, they an: polymorphic. 
1'(l I~m. . )rphi$rhS dc1ccted in this way are known as restriction fragment length 
JXlI~ moqlhi!>lnS. RflPs can be u.'w.'d to characterize an organism. levels of lenotypic 
dl\<'nll" and phylogenetic rtla tlllmtups. They can arise from point mUlation.s leading to 
('Ilhcr a loss or I pin of. site at which a n:-striction mciofll.acleasc acts, In addition. 
--------------------------------------.~---
University of Ghana http://ugspace.ug.edu.gh
deleliom or insertions rcsuJtina: in variations in the nwnber o f tandemJy repeated ONA 
sequences can aller the k'ngth of fragments bctwttn two endonuclease sites (Watson" 
ai , 1992). Since DNA polymorphismscan occur in any type orONA sequence. it means 
thal an aJteration producing a resuiction frag.ment polymorphism JeQIIh can occur wilhin 
a coding ~uence of a gcne, ooncoding sequences (inlrons), sequences between genes, 
and e\lcn DNA with no known function. such as repetitive DNA. ONA polymorphisms 
can .ffect any pu1 of the genome and are thm:fore extremely valuable markers 
tRothwell.198S) 
Restriction fi'agmcnt length polymorphism (RFI .P) is most suited for taxonomic studies at 
the inlnlSpeCific kvel or among closely rdated tlUa (Avisc el al .• 1989). Prc-scncC' andlor 
abscnceoffrqmc:nts resulting from changes inrttognilion silesare used in identifying 
speclCS and in characterising poplllations . For example. if the total genomic DNA is 
digestcdtocompJetionwith8highf~u.:ncyrestriclionendonudcasc,anextremelylargC' 
number of fragments are produced that can be SCJ*IIled by electrophoresis and will show 
up as eilher weak or strong bands. When DNA (rom different species are analysed this 
~")' the pettem of blinds will bediffC!renl. These restriction spectra will be !'Ipn:ific. not 
ONY (()f' individual restriction endonucleases but liso. (or the genomc:. ofthc different 
specic:s (A\'i.se ef aI. • 1989). Thus. RfLPs arc used fO measure gcnetic divergence 
between different populations or related !'Ipecics. The restriction-sire difference is 
etTectivdy a DNA difTcmtCc. so I measure o( the tolli l nwnber of RFLP differences 
repraents. mcasureofgencllc differences., and an: therefon: important inevolulionary 
5IUdics(Griffilhs<,oI. . I9'l9) 
70 
University of Ghana http://ugspace.ug.edu.gh
2.5.4 Agarose gel electrophoresi5 
Elccuophorcsis bas abo played an cS~'utial role in the elucidation of the structure, 
!OCqucnce and function of DNA. Electrophoresis through as.rose or polyacrylamide gels 
is lhe standard method used to separate, identify and to purify DNA fragments (Sam brook 
et oJ .• 1989). The: method is used for DNA fragments generated after endonuclease 
digestion. before or after enzymatic modification., ligation with other fragments. or after 
sequendns (PerbaJ. 1988). The technique is simple, rapid and c.pable of resolving DNA 
that cannot be separated adequately by other procedures, such IS density gradient 
cCl'llnfugalion (Sambrook el 01 .• 1989) 
A~. which is extrKted from seaweed, is. linear polysaccharide lSamnrook tl 01 .. 
19&9; Pbannacia, 1989), It has larae pores and aJlows rapid run times. (t is easy to 
prt'PIItetndc.n be set into a variety o(shapes, sizes and porosities, and can be run ina 
number of different configurations. The choices \.\ithin these paramc1tn depend 
primarily on the sizes ofthc DNA fragments fa be separated. Although agarosc gels have 
lowtTI'CI01vina power than polYKr)'lamideacLs, tbey have a greater rangc of separation. 
DNA fragmrnts from 70 bp to approximately tHJO kb in length can be separated on 
qarosc gels of various concentrations (Fangman, 1978). The location of DNA within the 
gel an be dctcnnmed direetly by staining with low cOrl\:cntratitln~ of the fluorescent 
intercalating dye. ethidiwn bromide. Band~ containin, as linle as I-lOng of DNA can be 
<k1eC1td bydircct examination of the gel in ultraviolet light (Sharp etal. • 1973). 
A.:aroseaelsareuo,u.all} run inan hori:t.onlal configuration in an electric field of constant 
"rm~th :InJ JIO,.-.:hoo. When an electric 6cid is applied acroM the gel, UNA. which is 
r.'·'· 'III\d~ ..:harl!cd at neutral pH. migratc:s towards the anode, 11le me of migralion is 
---------------------------------------,71 
University of Ghana http://ugspace.ug.edu.gh
cklennined by scvual facton iDeluding the concentration of the agarose gel, molecular 
weight and confonnauon of the DNA. 
Mo~ of linear doublc·stranded DNA in an electric field lend to orientate in an end-
IXl position (Flsher &: Dingman. 1971: Aaij & Borst. 1972) and migrate through Bel 
tnIIric:cs at mes thIt arc inversely proportiorul to the 10(1;10 aftbe number of base pairs 
(Helling ~I aL 1974). L.arger molecu.les migrate more slowly than the smaller molecules 
tInuah the pores of tile eel due to gn=ater frictional drag (Sambrook. etaJ , 1989) 
The different coofonnations of DNAs, namely. superhelical circular (form I), nicked 
circular (form 11). and linear (form Ill) of the same molecular weight. migrate Ihrough 
agarose gels aI different rates (Thome., I %7). The relative mobilities of the thrtt forms 
depend primarily OD the ~ concenlr.ltion in the act, but they are also inllumctd by 
tbesuaatheJflhc 'ppliedeunenl,lhc ionic strength of the buffer, and lhedensity of the 
.superbclicaJ twillS in (be form I DNA (Johnson & <";,ossman. 1977). 
A linear DNA fragment of a given size migratH at ditTncnt rates through gels containing 
dilTerent conttntrations of agarose: (Sambrook rl 01 . 1989). There is OJ linear relalionship 
between gel contenb'atiun and 10110 of the eletlrophorttic mobility of DNA (Sambrook ~t 
01. 1989; Phannacia, 1')89, . Thus. by using gels: of different concentrations (differ~nt 
porosities). it is possible to resolve a wide range orONA molecules. 
lloth lhe liC'pU'Ition and ~lution of DNA fragmenLS are affected by the volla~c "radient 
(Suulhcm, 1979; pt.macia. 1919). At low vol..,"- the rate of migration of linear DNA 
fra~mcnls is proportional tu the vollage applied. HO\\'ever. 3.'1 the clectric field stm'Igth is 
r.,,'>Cd. the mobiliry ofhitth·moJecular-\\,cight fragmcnu of DNA incrnses differentially 
--------------------------------------~72 ---
University of Ghana http://ugspace.ug.edu.gh
Thus, the effedi~ range of separaliuo in agarose gels decreases as the , 'ollage applied is 
intreased (S::unbrook etQ/., 1989). A baiancc has to be struck berween resolution and 
SoCpanition. Low.molecular-weight fragments diffuse and arc thus best sC'J*'8lcd at fairly 
hig.h-,"oltage gradients. Large fragments. however, diffuse vcry slowly, and lhc best 
resolution IS achieved by using low-voltage gradients and running fot long times 
(McDonell elw. • 1971; Pha.macia, 1989) 
DNA molecules of larger than 50-1 00 kb in length migrate through agarose gels at the 
same rMc: if the dlrccrion of the electric field remains constant (Smith &. Cantor. 11)87: 
Sambrook et aI. . 1989). HOWf::vcr, because of Ihe sieving eff«t oflhe gel mltrix. if the 
direction of the electric field is .1tered periodicaHy. the DNA molecules are forced 10 
change coone. The lime it takes for a molec:ule 10 reorient itself in the new electric field 
depends on its length (Smith ~ Cantor. 1978). Larger moh:culcs take longer to orient 10 
!.he new ditec.tion of the facld: as a result pulse-field gel electrophoresis is uxd to 
fractionMe extremely large molecules of DNA, up 10 about 10,000 kb (Smith &: Canlor, 
1987, Son>brook"aJ. . 1989). 
The behaviour or DNA in agarose gels. in contrast 10 polyacrylamide gels (Allet et oJ. • 
1973). .. nol significandy affected by either the base composilinn of the DNA (Thomas &. 
Davis. 1975) or by the temperature at which the gel is run (Sambrook ef oJ,. 1989). Mosl 
q,&I'O$C gels arc run al room Icmperalure. because the rehuive clectrophorc:lic mobilities 
or DNA (rtgmcnts of di(ferenl si:t.cs do not vary between 4'C and JO'c. However,eels 
cQ,uainin& less than O. .sV. aprose and low-melling Icmpcralun: agarose eels are rather 
ru-y. and.llt: heSI run at 4'C, \o\o"herc they are stronscr {Samhrool ('I dl. , 1919) 
----------------------------------------~73 
University of Ghana http://ugspace.ug.edu.gh
Ethidtum bromMk. a (luoresc;:eat dye that is used to detect DNA In agaro:st and 
polyacrylamlGc gels, reduces the electrophoretic mobility of linear DNA by about I(). 
15% (PerbaJ. 1988; Sambrook eI al.. 1989). The dye intercalates between Slacked base 
peirs. extending the length of linear and nicked circular DNA and making them RlOf'e 
risid (Slmbroolt tt aI. . 1989). 
TbecompositionandionicstrmgthofeJecrrophoresisbuffendoaJso.ffecl the mobility 
of UNA (Sambrook el QI.. 1989). In the absence of ions, electrical conductance is 
minimal and the DNA migraaes slowly, if at all. In buffers of high ionic strength. 
electrical conductance i$ vny effiaem sod significant amounts of ml are gcncrukd. If 
O\.erhc:ating sboukI occur. the DNA bands will be distoned, the DNA denatured or the eel 
melts (Sambrook l!1 m..  1989). Seven! different buffers are available for e lectrophoresis 
of double-stranded DNA. These contain EO J A (PH 8.0) with either Tris-acctate (1 AE). 
or Tris-borate (TBf-:), or Tris-phospbate (TPE) at a finaJ concentration of approximately 
50 mM(PH 7.5-7.8)(Sunbrook ttal. . 1989). 
2.5.5 Determination of DNA fragment sizes 
~imalion of molecular Size of nucleic acids in either acrylam.ide or .garose gels after 
electrophoresis is important for the identifiulion and the dwacterisalion of restriction 
fragmen15and for studies of native and denaturcd DNA. 
To estimate the size of DNA fngmenlS from lhc-ir mobilitjl:S in gel electrophoresis. a 
rct.tionship is ntatKished between the mobilitil"S and the lengths of S1andard fraammlS 
(SouIhnn, 1979; Duggleby rf 01.. 1981; Elder & Southern. 1983). This ff:lationship is 
then used to c.llcWate the lengths of unknown frd.gl1lC11ts from their mobilities. The 
.xtU'K)' with whM:h fr.J.gmcnt lengths can be estimated dqJends on the lCCutacy of the 
--------------------------------------~74 
University of Ghana http://ugspace.ug.edu.gh
chosen relationship between mobility and length (Dugglcby ~I tJi" 1981; Elder " 
Southem, 1983: Rocbelleetal. . t985; Oertcr~tol. . 1990). 
Numerous methods have been proposed for grnphical and computer analyse! of the 
rclatiooships between mobiliry and length (Duggleby et 01.. 1981 ; Elder & Southern. 
1913; Rocbcll el aI. . 1985; Oerter ef 01 .. 1990). The most commonly used method is the 
ploning of the logarithm of the molecuJar weight against the mobility of standard 
frq.mcnts. and e~tlmating the lengths of unknown fragments from the resulting graph 
(Duukby ~I uf . 1981: Schaeffer & Sederoff. 1981 ; Elder &. Southern. 1983), "The 
st.andud ctJrVe:oo ohtaincd from such plots oRcn show pronounced curvulUre which may 
introduce significant subjectivity into the interpolation (Duggleby el al. . 1981: Elder & 
Southern. 1983). As a result. linear models have been developed that more or less tit with 
expenmental data (Aaij & Borst. 1972; Duggleby el 01. • 1981 ; Schaeffer &: Sederoff 
19S1 ; Eidu '" Southern. 1983). 
2.5.6 Polymerase Chain Reaction (PCR) 
Jne pol) mcra...: chain reaction (peR) is an In l·lIm tcchni411~ for the enzymatic 
,Implificatum of. specirtC DNA fragmmt of interest from small amounts of a longer 
mulccule, 1be technique was invented by KaIy Mullis (Mullis el 01., 1986: Mullis & 
Faloona, 19K7, ..1 Od was originally med to amplify human (Il)~globin DNA and for 
prerwaldia£l14.hlsofsidde-ce1l anaemia {Saiki flat .• 1'185; Saiki el al., 1')8h; Embwy et 
uJ .• 1987). 
A typical amplification reaction medium includes a thmnostal"lle UNA polyrrcrasc. 
oIiJOOuclrotidc: primm.. deox)nuclrolidc triphosphates {dNII)>J. rt"aClion buffer. 
"..~sium and optK:w.J additi\CS 3.nd a tl!mplate DNA. ~ reaction is earned OUI usins 
75--
University of Ghana http://ugspace.ug.edu.gh
an automated thennal qclcr (reactor). which takes the reaction through series of different 
&empetarurt's for varying time dunttioas. Each peR cycle thromically doubles the 
amount of wgeled template sequence (amplicoo) in the reaction: 10 cycles theoreticafly 
multiply theampHcOD by a f&clarofabout I thousand; 20 cycles. by a factor of more than 
InUllioninam.:ltterofhours(WhiletloJ. • 1989). 
Each cycle of PCR amplification consists of three phases (Fig. 2.5). The first phase is the 
dcnatucatioo or separation of the two strands of the duplex DNA molecule. This is 
,l("l.'tlmrh!'hed by brieny healing the raction mix 10 temperatures of about 92·9SoC. Each 
~mmd serves as a template on which a new strand is built. The temperature is reduced 
during tlk: s«onJ phasco so that the primer1 which are oriented with their 3' ends pointing 
towards each other can anneal to the template (White el 01 .• 1989). lbc temperature is 
then raised dwina the third phase to the optimum for the DNA polymerase to .dd 
nl)tleolides onto thI: mds of the anneaiC'd primers (cxlen.-;ion). The time or incubllion at 
Ihi:llcmperalute varies according 10 the kogth ortuget being amplified (Saiki, 1989). At 
the end oCme cyc~. which lasts ror a few minutes, the temperature is raised to 92-95°C 
for dc:nalunlion 10 commence another cycle. A typical peR of usually 25 to 30 cycles 
produces 8 sufficient amooot of DNA for further experimental procedures e.,. cloning. 
A major problem encountered in the original PCR procedure was thai the DNA 
poIymensc: (Uclwrich,o coli DNA Polymerase I, Kle:now f1ll8Jlleftt) had to be 
tqIIenished after every cycle becawe it is not stable at the high tcmpetlturc:s needed for 
denalur.uion (Saiki ~I oJ.. 11)85. ~ulli, &: Faloona. 1987). This problem was sol-.'ed in 
--------------------76 
University of Ghana http://ugspace.ug.edu.gh
::"~ y.~~~'~.,~==T=====~-
..... - ~9~§~-~...f. . 
---~ 
,..,.._+--N 
tOil· 2.S S<:hemalic diagram of the PCR process (Protocols and Application Guide, lid 
Edllton. Promep Corporation. Madison. USA). 
University of Ghana http://ugspace.ug.edu.gh
1917 with the diJCOVC1)' or. heat-sc.ble DNA polymerase called Tuq (Saiki el al. • 1(88); 
an enzyme isolated (rom the thc:rmophiHc bacterium. Th~"n&U aquulicUJ. which inhabits 
hot springs (Chien el oJ .. 1976). Taq DNA JXllymerase has an optimal c xteos;on rate 
(polymerisation rme) of 35·100 nucleOl:ides per second at between 70-lifC (Newton & 
Graham. 1997). The higher temperature optimum for the Taq polymerase extension 
allo"'Cd the use of higher temperatures for primer annealing and extension. thereby 
incmtSlrtl the ovenll stringency of the peR reaction and minimising the extension of 
primers lhaI were mismatched with the template, The increase in the specificity of Toq 
polymerase t'eR1Jts in an improved yield of amplified target fragment by reducing the 
competition by nonwgct products for enzyme and primer.; (While ~, ul. 1989). Apart 
from Taq DNA polymerase, its cloned and modified version, Tth DNA polymerase from 
TlwrmtlS Ih~r1ltOpirilw. Pfo DNA polymerase from Pyroc(xCIJS ju,iosus and a (ew others 
can also be used for PeR (Newton & Graham, 1997). 
The annealina seep is an imponanl parameter in enhancing the specificity of a PCR 
reaction , USI.I&Ily. the temperalure chosen for annealing of the primer is a compromise 
(Saiki ~I 0/. • 19&&). lbis is because at lower temperatures anncaJing is more efficienl but 
there is • significantly increased amount of mispriming. At higher tcmpenuun:s, 
however. there is lncreased sp«ifieity but the overall efficiency of amptiftCation is 
~d. Under standard conditions. the: annealing tempcr.llurc in a reaction should be 
S·C loWC( than the: mtlling tempnalUres (T,.) of the primers (Innis &. Gelfand. 1990). To 
tktcnnine the aprtu.'IOimale melting lemperatures of the primer-DNA duplex, the equation 
T", ... 4(G +C)+2(A+n 
(",,"",U " dUTP. C - dCTP. A' dATP md T· dTTP) con be used (W;nnac:k ... 1917). 
More CUC1 numbm; cun be calculated hy applying algorithms (Rychlik & Rhoads. 1919) 
rbli take inlO ac:tourI( the OC:cum:nce tlf IRtramolccular IRteraotfion. o;uch as. Mirpin 
University of Ghana http://ugspace.ug.edu.gh
str'UC'tUr'eS. Normally, a series of reactions are set up 10 determine the opcimal lI'UlIeaIing 
tempcralun:. 
A number of factors influence the specificity of the amplification reactions. The 
SIJi.n8cncY of the annealing step can be controlled lD some extent by adjusting the 
.nnealina Icmpcruture. Minimizing the incubelion lime during the annealing and 
extmsion steps will limit the opponunitics for mispriming and extension by molecules of 
otherwise idle polymerase (Saiki, 1989). Reducing primer and enzyme concentrations 
also servcs 10 limit mispriming. particularly tbe type that leads to dimerization. Finally. 
changing the: concentration I~ls of MgCh can further improve specificity. either by 
increasing (be stringmcy of the reaction or by direct effects on the polymerase itself 
(s.;ki.1989). 
The PeR method can be used with a complex template such IU genomic DNA and can 
amplify a singlt-copy ge:oecontained therein. It is aJ50Clp.lble of amplifying a single 
moleculeoft.:u-gct ina complex mixture of ON As and RNAs (Saiki erat., 1988) and can. 
undeT cena.in conditions. produce fragments up to )5-42 kbp long (Barnes. 1994; Cheng 
rt aI., 1994). The starting material, the template DNA may be single strandctJ or double 
wanded DNA. The DNA to be amplified should be free of contaminanls or non.target 
materials 10 avoid their amplification ...t her ttw. the target DNA. Nonnally, 
subnanogram quantities of the template DNA arc used for PCR. 
The PeR is '*'" in a wide "'ariety or fields including : molecular bioLoa;y. environmental 
Ktcnce. formsic: 5Cien~. medical Kienee. binh:\.'hnulogy, microbiology. the rood 
indu.\'t). di3.¥J'OSl.ic science. epidemiology, ~C1lC'tics. gene ck>nina. and many more. II 
h.u been I.l'C'd In the diagnosis of l!CI'ICIK: disorders (Saiki tl 01.. 1985; 1981). the analysis 
-----------------------------,"9 ---
University of Ghana http://ugspace.ug.edu.gh
of .UelM: xquence variation (Sambrook el aI., 1919), and analysis of mutation or any 
research that invoh'CS the rapid cloning and ~uencing of homologous DNA fragments 
(Gyllensten, I <189). II bas also been used for the analysis of individual identity in forensic 
samples by the amplifKalion of highly polymorphic DNA regions (Higuchi el 0/., 1988), 
and the examination of nucleotide sequences from ancienl preserved specimens (Paabo el 
aI., 1988: Pabo. 1990). Th: PCR has been applied to difficuh problems in 
dtvcloprnmtaJ biology. For nampk, peR of eDNA has been UKd 10 study .V-J regjon 
combinations in the T ~C'II rcreplor 5-ctWn (Loh tl 01 .• 1989) and the ('xamination of the 
mRNAs for growth rectors in smaJl nwnbns of mocrophagcs isolated from wounds 
aclivelyundetgoingbeaiing(Rappoleeeloi., 1988). 
University of Ghana http://ugspace.ug.edu.gh
2.6 Mole cular Phyloge netic Ana lysis 
\iolo.:w.r phylOKeoy is me study of evolutionary relationships among organisms by using 
tcchniques of mokcular biology. It began at the twn of the century, even before 
Mendel's I.lWS "en: disco\"cred in 1900. ImmwlOchcmicai studies had shown that 
serologIcal cross-reactions were stronger for closely relaled organisms than for distant!)' 
relakdones. The evolutionary implications of these fIndings wen: used by Nultal (1904) 
to infer phyLogenetic relationships among various groups of animals. 
In the 1960sa 1970s the study ofmolecularphyJogeny , us ing prolcin sc-qucnce data. 
prosres$Cd Il"mlCndously. Less expensive and more expedient methods, such 15 protein 
electrophoresis., DNA-DNA hybridization, and immunological methods. though less 
accurate than protein sequencing, -were used to study the: phylogenetic relationships 
amana populations or dO!Jety related ~ics. The applacation of these melhods 
stinaalalQi the dtveiopmcnt of measute!J of gC1"'lCtic distance and tree-making methods 
(Fil<:h'\ Morioliash. 1961; Nei. 1915; Fel>enstein. 1988) 
The rapid accumulation of DNA sequence data from the late 197~ has had a grrat impact 
on molecular phylott'ny" DNA sequence data are not only more abundant, but also easier 
to analyze than protein sequence data. Thus, they have ix'Cn used in a wide ranac of 
applatiOlll. R.-consttuc1ing the ancestral gene sequences from which c:..tanl genes are 
derivul; srudying the origin and epidemioJoKY of human dixaxs; inferring the evolution 
ofccoiogic.al and behavioral trait' through time; C'Stimating hiStorical biogeographic 
rd,1llonships; prioritizing the conscn:al lonofendangered spccies; and reconstructing the: 
historical relatW>nsIUps acrouaJl of life (Wocse. 1987; Hillis, 1(97). 
--------------------------------------,81 
University of Ghana http://ugspace.ug.edu.gh
Whatever the application. all phyloge:ndic analySts require a crilerion (known as an 
optimaJity criterion) for assessing how welllhe d8Ia fit candidate trees, a method for 
searcbing among possible solutions for the nee with the best fit to the data. and a method 
for assessing confidcnce in lhe results (Hillis., 1997). 
2.6.1 Molecular phylogeny methods 
MoilxuJar phylogeny methods allow. from a given set of aligned sequences.. the 
suggl:Stlono(phy~ctrtts~nichaim'lrttonstructinilhehistoryorsuccessive 
di\'crgmccs which lOOk place during the evolution betW!!I!" the considered sequences and 
1.6.1.1'hy~enetk:trees 
All life forms. both extant and extioct. have I common orisin somcy,hcre in the past. 
(rom whieh they C'\'olvcd inlO \he plethora of spec ies found today. ThUs. all animals. 
plants. and baclcriaarc rcllted by dcscent to cacb other. C losely rclalcd spe:cies share a 
fI'IOfe ft'Ccnt common ancestor than distantly related species. The objectives of 
ph)"logcnc:tic Sludiesare to: 
I) ftt(lnstruct the correct genea.logical tics bet:wecn uq;anisms, and 
Ii) estimate the time of diveraence ~l"'ft1l organisms since lhcy last shared a common 
ances.tor(Li&Gnaur.1991). 
The evolutionary ~lationships between species can be represmted by • phylogenetic tree. 
Thi.s is • gnph COI1lpOSC.'d of nodes and branches in which onl), one branch Conntf;U any 
.tj.:cm nodes. The nodes rqnscnt taxonomic units, while the branches connttlin~ 
thcm. refkct thrit R'lationships in terms of descenl and ancestry. A phylogenetic tree is 
chuxkriscd by its topology (form) and its k:ngth (sum of its branch lengths) Iti a: 
82 
University of Ghana http://ugspace.ug.edu.gh
Graut. 19911. The topology is the pattern of branches found in a tree. The branching. 
pmtem (often caJled branching order) shows lbe ,('ncalagy of the organisms. That is. it 
shows which species !bare more common ancCSIfy with which others The branch length 
is commoa.Iy used to indkatc some ronn of evolutionary distance represented by thli 
brtilCh. The BCluat still existing taxonomic units are often called operational taxonomic 
Wlics (OTUs); a generic lenn that can represent many types of comparable taxa. for 
c,\;unple .• family of organisms. individuals of a single spec~s. a set of related genes or 
even gene rqjons. are represented by nodes on the lips ofthe' branches.., called external 
nodes (Wcillcr el m..  1995). The other nodes arc called internal nodes. Inlemal nodes 
may be called hypotheTical taxonomic units (HTU) (0 emphasise that they are the 
hypothetical progC'nilOf1 orOTUs (Weiller tl ai. • 1995). 
Phylogenelic trees can be either rooted or unrooted. A tree where 8 special node 
indicating the common anceS10r to all oros is present is called a rooted tree. An 
untOOted tree le.ves the position of the common a.nl,;l,:~IOf unspecified The number of 
bir..c.t:insrootcdtrces(.Va)fornOTIJsisgivcnby 
ror,,~ 2(Li&Graur, 1(91). 
The number of blfun:ating unmuled tn:c:s (Nu) for tr ~ 3 is 
. (20-5) 
.~IJ= 2- ' (n_3)! {PeM} dul .. 1982: li & Graur. 1991) 
-------------------83--
University of Ghana http://ugspace.ug.edu.gh
A SfWcie$ rue is s phylogenetic tm: reprcxoting the cvolutionary pathway of. aroup of 
sptCte3. When a tree is construCted from one gene from each species. the inferred tree is 
somcti~ called a J1:nW trn (Nei, 1987). These two arc not necessarily the same. as the 
divergmcc of the gene can .ctually predate the divergence of the species. This is not a 
problem when k»ng-term evolution is studied. Another problem is that the topology of the 
gme ttee does not nccessarily COlnclde wilh that of the species (li & Graur, 1991). 
2.6.1.2 $earchalgorithms 
()OI;cacriterionhasbcmselectcd for evaluating the filofdalatQtrces. it is necessary 10 
scatth among the universe of possible trees for the: optimal solution. A number of 
a1gorilhms for ~1111311Og phylogeny from DNA sequences exist, and it not aJv.ays clear 
what the: sumgths and weaknesses of each method are, or which should be preferred in a 
given situation (Kuhner & felsentein. 11)')4). For a small number of taxa, there are few 
po$$.ib~ &rec topologies (branching arrangements), but the number of distinct tn:n 
lncrcaes rapidl)' as • fWX:llon of the number of wa. There art only three branch.ing 
OfJCI'l for four Wta, but there are more than two million ditTertni trees for 10 taxa. and 
mo~ than 2 x IOIIl different tr~es Cor 100 taxa (Hillis. 1997). 
For smaller data sets. there are two computOltional procedures. or algorithma, that are 
guaranteed to find the optimal1ltt(s) for a particular criterion , One is simply AD 
ex.baustive evaluation oCall distinct ~ (Saitou 4 Imanishi, IIIIN). but thi, can be \'ery 
timc-consumin& and is f"II'Cly feasible for more than about 12 taxa (Hillis. 1997). The 
' __O Dd·bound' a1gorilhm (Hendy &< Penny. 1982) is • -.mil algorithm dUll will 
fitld the exact solul ion(.) (or 20 or mure c.ua, dependina on the computational ~un::n 
and Ihc '1'IOt51ncM; uf the <SaL&.. 11 works b~ ignorin& whole c:lasIc:s of trees thai can be 
malhc:mati~lI} -=:\dudcd as suboptimal cumpared to • known solution to the problem 
--------------------------------------~~ 
University of Ghana http://ugspace.ug.edu.gh
(Hillis. 1997). This known solution may nOl be optimal. but it serves 10 define an upper 
'bowld',orlimit,fortheanaiysis. 
All exhaustive sean::b methods examine all or a large nwnber of the theoretically possible 
btt topOiogiesaod use ct!r1am criteria to choose the besI one (Sailou & lmanishi. 1989) 
Their main advantage is that they produce a large nwnber of difl'ettnt trees together with 
a relative estimatc of the likelihood that they represent the phylogeny . lbisallows the 
InvestiptOr to compare: the support for the best tree with the support for the second best, 
and thus ~itnlte the confidence in the tree obtained. Unfortullitely the number of 
possible trees and thus computing time grows ver)' quickly as the number of taxa 
increues; Ibt number of bifurcated rooted trees for nOTUs is given by (2n-3)!I(2n-2(n-
2»! [Cavalli-Sforza & Edwards. 1(67). This means that for a data set of more than 10 
OTU, only • subset of possible trees can be examined, 50 variou.s strategics are used to 
search Lhc 'tree spKC' but there is no algorithm thai le!oorunlecs thai the best pos.sible tree 
~"&5 u.amil"led. 1"he algorithms lhIl .search for oplilNlllttS CM conlAin an optional nnaJ 
st~. global n:arrangcmtnt. which, involves removing each branch in rum I!U\d trying all 
possible positions for il; this improves performance but approximalcly triples run time 
(Kuhner & Fcllentein. 1(94). 
In contrast, the stepwise c1usterinl methods avoid this problem by examining Joca.I 
suhutt, lirst (Saitou &: lmanishi. 1989; Hillis. 1997). Typically. the two clox-st relaled 
OTUs iW combined to form a eluster. 1nc elu.'~tcr is then treaIcd M.e • single oro 
rcpratnti ll8 thc ancestor of the 01115 it rep1acn. Thus the complexity of the data KI is 
mtuced by one onJ. This pnxes:s is repeMcd, elustcTinj: the next clostSt rdated oro. . 
uncil all on '50 an= comhined. The "anous stepwise clustering 31Sorilhnu differ in their 
methods of dctcmuning the ~lationship of orus and in combining OTI.Js to clu.~cts 
--------------------------------------'85 
University of Ghana http://ugspace.ug.edu.gh
(Kuhner & FelsentclD. 1994; Hillis. 1997). They are usually very fast and can 
ateommodatc large numbers of OTIJs. As they produce only one tree. the confidence 
estimIlOn of me exhaustive search methods are not available, although various other 
staliSlicaJ methods have been developed for estimating the confldcnce in the correctneSS 
ofo ... obooincd (l.i & Gouy, 1990). 
'The majority of distance matri)( methods use • stepwise clustering to compute uecs to 
reprnent their ~Ialionsbips while many character state methods adopt the exhaustive 
search approach (Sailod & imanishi. 1989: Weillcr el 01., 1995; Hillis. 1997) 
For luge data sets, it is necessary to rely on approxitTUltion techmques (also called 
t\eurislics) to rmd near optimal solutions (Hillis. 1997), Several methods have been 
devixd for aettirc a quick initial approximation (or point estimate) of the optimal !:Re. 
wbich '*' be utcd ttl a starting point for mote thorou~ analyses. Sometimes thex initial 
approximalionsaretre:ltt'dasenJ-pointsintheestimltionprocedure, butthispraclice has 
tome serious drawbach Initial point estimates can almost always be improved. CXCq>1 in 
the C&K of very small data sets. In addition, point-estimation mtthods pro\'idc no 
Uldication or how many other trees (Of Yo'bicb trea) may be JUSl IU good. or c\'cn bencr. 
esUmlittsorthepbylogcny 
r h~' t~'O moll widely used point estimation methods are stepw15C addition and neighbor-
I'llnlng (llillis., 1997). The former method adds taxa one by one 10 • vowing tree, 
""hcrn.s the lifter method starts with aU taxa in an unresolved 'sw' and adds the interNl 
hr.and-=s to the tree in a stepwise fashion. Both methods find Jimilat tc .. ults (or .. mall 
trtts. but Slepwilr addition usually finds better solutMxl.s (and is computatlonall) .. Ii~tly 
------------------------------------,~ 
University of Ghana http://ugspace.ug.edu.gh
sJower) for large data SICU. 
Once. poi.nl-estimale of a tree has been fotmd. the estimate usually can be impro\-ed in a 
procedure knoull as branch-swapping (Hillis.. 1997). There are several different types of 
branch s. ... appang. bul all of them involve rearranging the branches on .... initial tree: 10 
look for IOPOIogtca1Jy related tn.~s thai areas good or bctterphylogenetic solutions. If 
branch swappina fUlds an equally good or better tree. then the bnnch swapping is 
conlinuedonthenewtrec. 
2.6.2 Tree construction methods 
Numerous methods for constructing phylogenetic trees have been proposed (Sneath IL 
SoUl, 1913; Fclscnlein. 1981, Nei. 1987; Swofford &: Olsen, 1990). These methods can 
bcclassifiedinlotwomaintypes:distancemalflxmcthodsanddiscretc-characlcrmcthods 
(Li" Grauer, 1991), In distance malrix methods, evolutionary distances (usually the 
number of nucleoeidc or amino acid substitutions separating two taxonomic units) are 
comput.:d for all pairs of data, and a phylogenetic tree is constructed by using an 
al.orithm based on some functional relationships among the distance values. In dixrete-
cblnctCT mdhods. chanel" stales (e,g .• the nucleotide and amino acid at a site) are used. 
and tbC' shortest ,.lhWll)' leading 10 these character stales is ch05en as the phylogenetic 
tree. Allct.raccenan:analyscd sepa ...t elyand usually independently from eKh other 
(Weillereta/"I99S) 
Some types of molecular data (t'.g. , DNA hybridiZilion data) exist only as dislanC'e data; 
therefore phy~nctic trtts for these data can be consuucted only by dist1ln\:c methods, 
nycontrast.di~-characterdatacanusuaJlybecoD\·('ncdil1l{)diSlanccdalallndthU5 
Ih ...~  can he analyzrd either by distance mdhods or by discft:tc-char.Krcr methods. 
!C, 
University of Ghana http://ugspace.ug.edu.gh
There are two mljor groups of discrete character metbods. namely. maximum plU'SlPlOny 
mclhods and maximum likelihood methods. Two typeS of errors can occur in the 
construction of phy10genetic trees: topological errors and branch-length errors (TaleJlo ~I 
.1,1982). 
2.6.2.1 Discrete character methods 
I) M._lmum ~rslmony (MP) and weighted parsimony (WP ) 
The theoreltcal blSis of the MP method is William of Ockham's philosophical idea that 
the best hypothesis 10 explain "processis meont that requires the snmllcst number of 
assumplions (Sober. 1988). The method was first proposed by Camin and Sakal (1965) 
as a method for lttonstructing phylogenetic trees from morphological data. Later 
modiflCatioos edjust.ed the MP method for amino acid (Eck &:. Dayhoff. 1966) and 
nuclcocide sequences (Fitch.. 1971). The latest modification utilises dynamically modified 
chancter weightina (Williams &: Fitch, 1990) and is caUed 'y,.-cighlcd parsimony' (WP). 
A maximum parsimony tree is !.he tree lha1 requires the smallest numl'>cr of evolutionary 
cha.nies to result in the: lei OTUs under study (Li & Graur. 1991; Hillis. 1997). It is often 
possible to find more than one ~ with the same minimum number of changes.. so that no 
tJruqUC lrtt can be inferred. Parsimony considers al l sites. but nol all sites convey 
iofOl'lD8tion regarding the most parsimonious tree. Only sites thai favour some topologies 
ova-othcnare informative and. consequeotly, onlythC3C silesarc used in the 
ealcu1ations. For molecular sequence data. sites are informlfive only when [here ~ al 
least two different k.inds of residues at the site, each of which is reprexntcd in at least two 
of the srquc:nc:cs under study (li & Graur. 1991; WeHler ~I al., 1995). This usually 
I"eICI'icbdralicaUywnumbcrofposilionsuscdiothelUlaJysis 
University of Ghana http://ugspace.ug.edu.gh
InOC'der 10 fiDCl the most parsimonious tree. first . all the informative sites are identified. 
Nexl, Iht minimum number of substitutions al each informative sile is calcuJated for ali 
possible tree topologies. In the final step, the number of changes over all the informative 
5ites for each tree is summed up and me ~ associaled with the smaJlesl nwnber of 
5Ubstitutiooslschosen(Li&Graur, 1991 ) 
The tWKIard MP method weights all changes equally although for sequence data this is 
not al"'..,.5 desirable (We-iller el 01 •• 1995). Firstly it is sometimes desirable to weight 
different sequence positions differentiaUy in order, for example, to emphasise the few 
changesatconscrved positions rather than the several at variable positions. Secondly it it 
often advisable to weighl differenl types of changes differently, forexamplctransversions 
are usually morc siarurlCatlt than transition. . , and certain amino acid substitutions mon: 
tt.n otbm. Recendy. Williams and Fitch (1990) introduced the dynamically weighted 
parsimony method (WP) which uses an initial tree (seed lree) 10 assign different weights 
to different Iypes of substitution and/or sequence positions. It then uses this weighting 
scheme to generate a new tree. This process is repeated until the same best tree is 
obttinedintwocon.sc-cutiveruns. 
The: nwnber of possible topOlogies increases very sharply with the number of taxa, 
making computer time a limiting factor for calculations on more than • few sequences. 
The "branch and bound" method (Hendy and PCMY. 1982) can limit Ihe number of 
topoloa.ies that have to t'C tested. and still find the optimal tree. Even with this 
optimisation. the method is limited to a small number of sequences. For larger sets., 
hl:uristic methods must be uscd. but these cannot {!uarantee 10 find the most parsimonious 
tree. Panimony is also very sensitive 10 tDeqUal ~ of n'olution in different bnncbes 
University of Ghana http://ugspace.ug.edu.gh
in the (Fclxnstcln. 1978) Since it is not possiblc to corrcct for superimposed lrtt 
cvolutionar)' cvents (where a given nucleotide position bas changed more than once since 
thc lWO sequences diverged), tbe prescnce of distantly related lineages can cause 
distonion. 
The cffecu of the W"lCqual rates of evolution should be solved in the cvolutionary 
paDimony md.hod proposed by Lake (1987). but this method has been criticized for 
depending on a number of unreaIistie assumptions (Jin &. Nei, 1990) and for 
underestimating branch lengths (Olsen, 1991). Information on evolutionary processes 
may be incorporalCd by weighting character.;: differentially (such as frrst versus third 
positions of codons), or by weighting character state changes differentially (such as 
tr1IDSitions~1rIDSYersions)[HjJlis,1997J. 
II) Mulmum m'~ljhOocl (ML) 
The maximwn-likelihood method for recooslrUCling phylogenetic relationships has 
become wry popular because of its main advantage of application of a well-defined 
n 'olutionmodeltogj'tlendataset(Felsenstein, 1988). 
The MI. method was originally proposed by Cavalli-Sforza and Edwards (1967) for gene 
frequency clara. Later. Felsenstein (1981) developed 8 ML algorithm for nucleotide 
sc:qucnocs. Mnximwn-likelihood is similar to the MP method in that it cxamines every 
reMOnIIbletlttlOpOlogyande\'alualesthesupportforeachbyexamlningc\,eryscquence 
posItion 
Phylogenetic tree iafemx:e baed lin the method or maximum likelihood is appealing 
from both bio1ocaJ and statistIcal pcn;pecti\'es. The maximum likelihood method tries 
University of Ghana http://ugspace.ug.edu.gh
10 infer thc topOJogy that is most c:onsistent with a sci of observed dala (FeI$CnS1Cln. 1981; 
Strimmn '" H~lcr, 1996). The possibility that a liven lopolo&y will produce the 
observed sequences is calculated for aU or many possible topologies (Swofford It Olsen. 
1990. 1996). In order to calculate these possibilities, a concrete model ofthc evolutionary 
proccss that converts one sequence into another must be spcciried. For example, all 
nuckotides are assumed to be equally frequent and the probability of change of any 
nudeotide '0 any other nuclC'Olide is a.c;sumed 10 be the same in the Jukes·Canlor model 
(Mount. 1999). For each possible tree. the likelihood of finding the actual sequence 
changes ", each. column in the aligned sequences is calculated. The probabilities for each 
aligned position ate then muhiplied 10 provide 8 likelihood for each tree. The tree which 
provicks the: maximum likelihood vaJue is the most probable tree. 
Just as for lhe parsimony approach. this method requires much wmputer lime and can 
only be used on • limited nwnber ofscqucnces. However, il is possible to speed up 
calculations by paraUelizing the aJgorithm by using approximative techniques (Adachi &. 
HllSC'gawa.IQ94 ; OI~tla/. • 1994) 
I) Transrormatlonofsequencedatatodlstances 
Although st'quences are fWldamcnlally charKter data, phylolenc1ic analysis based on 
sequences IS often prtCCded by a prior reductton 10 distance values relating all pairs of 
.sequc:nce5. Many fast methods arc available '0 construct evolutionary trees from distance 
dala (Nci, 19&7; Gojobori ~I 01., 1m). Moreover, the usc of distances easily allows 
tOrm;:IH)ns for multiple mutations at the same: Mle. which are mone difficult to take into 
accowllwhen using character data (De Rijk., 1995). 
---------------------------------------91 ---
University of Ghana http://ugspace.ug.edu.gh
la the distance method. a matrix of distance scores among all of the sequences is first 
drawn. ~ distance score is a measure of dissimilarity between the sequences. so that 
the: less similar the sequences, the higher the distance score between them (De Rijk. 
1995). Another Vo'3.y of Lhinking about genetic distance is that it represents the minimum 
numberofcMnaes. including substitutions. insertions and deletions lhal are required to 
convert one scqucnce inlO the other. 
One of the most common ways of swnmarizing the relationship of t wO sequences [0 a 
number is their fractional similarity. In its simples. form. st!qmm('(~ similarity is the 
number of alignment positions containing identical re3idues divided by the number of 
alignment positions compared (De Rijk, 1995). ACcoWlting for gaps is the biggest 
problem in !.he caJcwation of similarity. Sequence comparison analyses can provide 
either .:Iimilwity or distance scores, depending on the scoring ~'ll'm used (Smith &. 
W;ucnnan. 1981; von Heijnc, 1987). SimHarity scores also may be converted to distance 
scores, baed upon assumed models of evolutionary change in sequences (Swofford & 
Olsen. 1990. 1996) 
Adcfitive distances can bc: fitted to an unrooledtree such that.1I pairwise distances are 
equailO the sum of the branch lengths that connect lhc OTUs (Dc Rijk, 1995; WeHler Itt 
oJ., 1995). Uhrametric distances arc the most constrained. They will precisely lit a tree 
so that the dislanc~ between any two taxa is equal to the sum of the: branches joining 
them. and can bc: rooted so that all taxauec:quidistant from thc rool (De Rijk. 1995). 
It) Unweighted paIr group method wtth arithmetic me,." (UPGMA) 
Ttk: Wlweighl",1 p.ur group method with arithmetic mean (UPGMA) is the simplest 
method ror In.''!: n=construclion. 11 .... '&5 originally developed for constructing taxOnomic 
------------------92 -
University of Ghana http://ugspace.ug.edu.gh
phrnograms. i.e .• U'CeS that reflect the phenotypic similaritieS between OTU, (Sakal It 
Michener, 1958), but it can also be used to construct phylogenetic trtt'S if the rates of 
evolution are approxiRllltdy constant among the different lineages, so that an 
approximately linear relation exists ~ evolutionary distance and divergence time 
(Ne;. 1975). For such a relation 10 hold. linear distance measures such as the number of 
nuclcotidc {or amino ac:id) subSlitutions shouJd be used. 
The trtt is conslructl-d by a sequential clustering algorithm (Sneath cl Sokal, 1973). 
Within (be matrix. the two OTUs with the smallest distance (i.e .• most similar) are 
clustered inlo a composite oru. which will then be treated as a new single OTU. A new 
matrix is creased, where the distances of other orus to the composite OTU are 
c3kulalN. This process is repeated until only two orus remain. These are cluslered 
togcthcr,andthcrootisplaccd.thalfthccalcuJateddistan<"cbc:twemthetwoc!usteB 
The UPGMA caJculales the dislatlces to a composite 0111 IS the arithmetic mean of the 
distances between the constiruent orus in each compo~te oru. If a simple average of 
the dislancC$ hctwecn the composiCe on;s is used, the method is called ~ig.hted pair 
group TIklhod with arithmetic mean (WPGMA) [Sneath & Sokal. 1978J. 
If the: assumption of I"IIC constancy amoag liocagcs does not hold, UPGMA may give an 
erroneous topology. The topological errors might be remedied. howevcr, by using a 
correction called the t~nsformt:d distance method (Farris. 1m; Klot~ 1I 01. • 1919; li, 
1981). "lhismethodusesanoutgroupasref~ lofMkecorrutioasforuncqual rates 
and then applies the UPGMA to the new distance INtnx 10 infer the topology or the tree. 
T1lc outgroup is an OTIJ which. hued on C'Jf.tcmaJ knowledge such a.. . tnonomic or 
93 
University of Ghana http://ugspace.ug.edu.gh
paleontologic informMion.. dearly diverges before the common ancestor 10 all other 
ow, (De Rijk. 1995~ 
11 is oftcn difficuh 10 decide which of the WI is an outgroup. To overc:omc this problem. 
Li (1981) proposed. two--stagc approach which first infers the root using UPGMA, and 
then UICS the taxa on either side of the rool LS an oulgroup to correct the distances among 
lincagcsonthcothcrsidc. 
I) Heighbour,"'tionmettlOdS 
The neighbour-joining (NJ) method (SaiIOU &. Nei. 1987; Studier & Keppler, 1988) 
constructs the tree by sequentially finding pairs of neighbours. which are the pairs of 
OTUs connected by a s;ngle interior node. 1be method is conceptually related to chISler 
analysis. but mnG\'C"5 the assumption IhII data are ultrametric (De Rijk, 1995). In other 
words. il doe:I not requin: that all lineages have diverged by equal amounts. As it does 
not ;Jnmlpt to clusttr the most closely related Ol1Js, but rather minimises the length of 
aJl intemal brlJlChes and. thus, the length of the entirc tree, it can be regarded as 
parsimony applied to distance data (WciUer tlol., 1995). 
lnoontrasl tocluslcranalysis. ncighbow-joining keeps track ofnodesoflhc tree rather 
thin taxa or clusters of tau. The NJ algoritJun SIartS by assuming a bush like Irec thai has 
no iDtemaI branches. In the fn step, it introduces the first internal branch and calcubles 
the length of the rrsulling tree. The algorithm sequetltiaJly connecb every possible OllJ. 
pair and finally joins theOTIJ·pairtlw:yiekbtheshortesl trce(WciliereloJ. • 1995). The 
length of a brancb joining a ".u of neighbours, X and Y. 10 their adjacent node, is the 
aVttagC distance ~ all OTIJs and. X as well as Y, Icsslhc: avcrqe distances of all 
94 
University of Ghana http://ugspace.ug.edu.gh
remaining OTU-pairs. This process is then repealed, always joining two OTUs 
(nc;ghboun) by ;ntrodudng the shortest possible internal branch. 
N) F'rtc:h and Hilrgoliash (FM) method 
The FM a1gonlhm (Fikh 4. Margoliash, 1967) initially uses the same clustering method 
as UPGMA and thus the initial topology is aI~ the same &S in UPGMA. but the branch 
lengths art calculated differently. AU trees with closely related topologies are then 
C'xplottd and compM:d in Irons of the so caJJed 'percent sumdard deviation', This is 
essnui.Uy • IDeIlStft that assesses how well the distances in the matrix correlate with the 
branch length obtained in the tree (palristic: di.stanc:cs). The tree wilh the smallest 
'standard deviation' is chosen. 
III) Olst:lncf! Wagner method (OW) 
As the network reconstnJ(:ted from character state data is often called a Wagner network. 
Farris (1972) caned nis method for reconstructing WlI'OOted rrecs from distance data 
"Oisunce Wqner". Like the FM method.. the OW method ~s a diS1ance matrix to 
analytc: many possible trees. Like many clUSIcrin, algorilhm.s., the OW method firstly 
combines the two mOil closely rtlated 011Js. As the network grows, it tries to fil all 
remaining oros in turn into any of the possible edaes. choosing the one that can be 
connetted with the shortest branch. This method minimises the total length of each 
subtree u it is buill (W~iller el ai., 1995). As it depencb ~ry much on correct estimation 
of e\'c:ry bnneh length. it is vel') suscC:plible to SllDplint errors, Improvements to the 
.i.lSorilhrn have beco suqcsted and these yield the modified Farris method (MF) [Talmo 
""'. • 1982]. 
--------------------------------------,~ 
University of Ghana http://ugspace.ug.edu.gh
Iv) Minimum evolution (ME) 
The nuwmum evolution (ME) method ( Edwards &. Cavalli-Sforza. 1963; Saitou & 
lmaniw. 1989; Rzhc1sky &: Nei. 1992; Edwards. 1996) is a diSlatKe-ba.sed algorithm. 
The method rust computeS each branch length ofa given tree topology by using Fitch and 
Margoliasb's (1967) procedure for branch-length estimation, and the tree that shows the 
smallc:st sumofbranch lengrns is chosen as the: best o nc. lbusthecri1erion used in this 
method is the sum of branch lengths (SBL). Rzhelsky and Nei (1993) showed thai the 
expected valuc of SBL ls smaJlesI for the true topology when unbiased estimates of 
pUwix distamC5 U'C Wtd. HO'A.'CVCT.lhis result does not mean that the topology with 
the smaticstSBL value is the most probable tree (Nei, 1996). 
1be pnnciple of minimum (',olulion, which implies search for a tree with minimum 
mutations to explain difTerences observed between sequences, was proposed by Cavalli-
Sfona and Edwards (1961), who eonsideRd II Steiner tree. With the use of Filch and 
Margoliash's prcxedwc for branch-length eslimalion. compulation becomes much simpler 
lhan the c.vaJ1i-Sforza and Edwtnls method. The ME method also seems to be relalcd to 
o.yhoff's (197&) method (Blanken el al., 1982). Although Ihis principle bears some 
~Ianoe 10 mal of maximum parsimony, in lhal it seek.!: the: tree with the lowest 
overall change in chanclers, it differs from parsimony in that 'change' is adjusted 10 
accoWlt for inferred superimposed events, using Il model ofevolulion (Hillis, 1997). The 
method ls similar 10 the NJ method bccau:te the principle of minimum evolution is also 
adopced in the NJ _ (Sail,," .It Nei. 19M7). ond because both the ME and NJ 
methodsaredLst:.ccmcthods. Simulalion results have sho""n that Ihe ME method and 
the NJ method an: indeed quite similar. However,lhc ME method is an exhaustive-KarCh 
method and examines.1I possible trees to choose the: best one . In contrast. the NJ method 
--------------------------------------'96 --
University of Ghana http://ugspace.ug.edu.gh
is a scepwise clustering method, and the computational time is much shorter in this 
lIlCtbod than in the ME mcthod 
2.6.3 Accounting for superimposed events (Nucleotide substitution in a DNA 
sequence) 
A ta\ic process in the evolution of DNA sequences is the change in nucleotides wilh 
IUnc . However. as the process of nucleotide substitutions is extremely slow, to detect 
evolutiooary changes in a DNA sequence, compAmive methods whereby a given 
sequence with which it shared a common ancestry in the past are used (Li & GraUl. 
1991). To study the dynamics of nucleotide substitution, several assumptions regarding 
the probability of substitution of one nucleotide by another are made. Numerous such 
mathcmeticalschemes have been proposed (Li&.Graur. 1991). 
The subllitution scheme of Jukes and Cantor's (1969) model is onc of the most frequently 
often uxd ones to correct for multiple mutations per site. This model swts from the 
assumptions thai all subsl:itutions are independent. that aJi 5CqU(ncc po~ilions are equally 
subject to clwlge, that substitutions occur randomly among the four types of nucleotides, 
and that no iluertions ordelctions have occurred . For example. iflhe nucleotide under 
considention is a. it will change to T. C. or G with equal probability . Since the model 
involves only one paramder. it is also called the one-parameter model (L i &. Graur, 
1991). ~ on Ihese pre-assumptions, the authors derived an flIU3tion for estimating 
evolutionary distances from observed dissimilarity. 
Sncnl otbtt equations for the estimation of evolutionary distances have been proposed. 
For e:umpit. K1mur. (19&0) has provided a method for inferring eVOlutionary distances 
------------------------------------~97---
University of Ghana http://ugspace.ug.edu.gh
based on a model of evolution in which uansitions and transvenions may occur al 
ditTerentrat(..'S. ltisthtreforeof'tenca1ledthe'twoparametcrmodel'. 
Oo)ding (1983) has shown that applying ~. Jukes-Cantor correction 10 sequences 
composed of sites with unequal evolutionary rates leads to an WlCiercstimalion of large 
evolutionary distances. As a result, distant species may seem closer than they actually Ite 
aod this can cause artifICial clustering of long branches (Olsen. 1987). This artefact can 
be avoided by con\'crtingdissimilarities intoevolurionary distances according 10 Jin and 
Nei (1990). 'IlleY asswne thai there is a gamma distribution of substitution rates over the 
sequence po5ilioM. A general equation for applying this idea in the correction of 
distances was proposed by Rzetsky and Nei (1994). This equation conlains a parameter a 
which depends on thcdalaset understudy. 
OlbcTcquationsan:bascdClnsubstitulton models in which the four different nucleolidcs 
art no( uscc:I: in equal proportions (e.g. Tajima &. Nei. 1984). or where a bias in the 
dim:tiooofchmgc is accounted for (e.g. Tamura&. Nei.I993; Zharkikh. 1994), 
An imponant drawback of most ofthesc models is that lbey do not consider dilTereocl!" in 
substitution rat~ among the sites of a molecule. 
2.6.4 Auessing the reliability of II tree (confidence In phylogenetic 
estimates) 
All methods oftrtt construction make: assumptions. \\-hen these a.'.sumptions are not 
satisfied. systematic: errors can be introduced. h'm if .U twumptions arc satisf.ed, 
random erTOrS may occur due to the use of. limited data~I , Methods are thcref<>f'r: 
ne-edcdfortestine:lhestatistkalsipi6ca:oc:eofanygiventree(Hillis. I997), 
--------------------------------------98 --
University of Ghana http://ugspace.ug.edu.gh
1berc are two different types of methods for testing the reliability of. tree constructed. 
One is latest the topological difference bern'een the tree and its ciolc:ly related tree by 
using certain qlWltiticS such as the likelihood value in the maximum likelihood method 
(Kishino &, H~gawa, 1989) and the swn of all branch lengths in the minimum evolution 
method (R2hetsky &: Nei. 1992), This type: of lest is supposed 10 examine the reliability 
ofevcry inleriorbranch of the tree. and it isgenernJly • conservative test. The pcocedure 
of the lCSI is I1SU&Jly quite complicated and requires considerably more computation. 
The other type of test is to examine the reliability of each interior branch whether it is 
significantly different from zero or DOl. If a particular interior branch is not significantly 
different (mmura. the possibility of trifurcation of the branchcsassociated or even tbe 
other'typesofbifuccatingtrees that can be generuted by changing the splining order of the 
tIuu branchel involved cannot be excluded (KWT\II el oJ., 1993). There are rwo different 
"',,),s of tcstiD& the reliability of an interior branch. One way is to compute the standard 
error of the interior branch and lest the deviation of the hranch length from 7.cro, and the 
othC'CiJlOusethebootrtraptest(Efron,1912;}'elsenstein, 1985) 
When the sequences do oot provide enough phylogenetic information (e. g. sequences are 
too short or IKking in variation), no algorithm will prOOlK:C sensible answers, One way 
loevaltmtewhethcrtbett is enough phylogenclic sianal in the data is 10 apply tests like 
';':kknifing or bootstrapping (Efron, 1981, I-dsenslein. 1985; li &: Gouy, 19(0), Both 
methods indicate \o\oht:ther smaller sample sizes would result in the same tree. 
-------------------------------------.99----
University of Ghana http://ugspace.ug.edu.gh
CHAPTER THREE 
MATERIALS AND METHODS 
3.1 Che miCAl 's, Ru ge nt5, Equipment and Software 
The sources and/or manufacturers of the chemicals and rt'tlgcnts. software and amine 
analysis servers. and equipment used for the study arc listed in Appendix I. T'he various 
bufTasand solutions used were prepared as described in Appendix II. 
3.2 Biological Specime ns and Sample Collection 
Mosquito larvae and pupae samples were collected from six locations, separated by • few 
kitometre5, in the Greater Accra region of Ghana. The areas were as fo llows: Madina 
(S'4Q'N, O'9'W), Adm. . (S'42'N, 0'9'W), Dzorwulu (S'J7'N, 0'12'W), Ach;mola 
(S-3S'N. OOI4'W), Legon (S039'N, Oog'W) and Airport Residential Area (S036'N, 
O"II'W). The &lot.J. positioning system (GPS) was used to determine the geographical 
coordillAtesofeachbm:dinisite. 
Anophr/u moequico breeding sites ",ne identified by sampling of small pools of stagnant 
water in the open and in sutters. exposed CO sunlight (Figs. l. la, b. c and d), The 
Dz:orwulu and Airpon Residential sites were characterized by small shallow pools of 
SUIgJlant rtinwalCr. At the Adenta. Madina and East Legan, the sites ..... ert mainly open 
drains containing stagnant waler exposed to sunlight. and these contained Il mixtur~ of 
bach Anopheles and Culex pre-adult stqes. The site at Ac:himota was water flowing from 
a leak.ina water supply pipc and exposed. to suaJight. 
University of Ghana http://ugspace.ug.edu.glOhll 
ftl. 3.la Mosqw.to collection sne at Ad~ (opm drain wllh ~ waaer 
...,...t., ....i sJa~ 
from 
101 
University of Ghana http://ugspace.ug.edu.gh
n,. 3.1(' Mosquito · coiltA;:ti~n site .. Ac:himot. (~ strdch or water fJOWIng 
fromabrokenwUerpipe) 
ne.3.14I Mosqwto t.ollecbon Sl1e aI Dzonrrulu (snWllhIllow pool of,..1ftI 
University of Ghana http://ugspace.ug.edu.gh
TIle Anophele$larvae were identified by their characteristic resting position, with the 
body parallel to the water surface and just be~w the surface film. The larvae and pupae 
wt'T'C carefully collected into small plastic containers by scooping gently wiili a JSO mJ 
dipper to avoid injwXs. The samples were transported to the laborntOC}' in 2-liue jars that 
were loosely capped to allow adequate ventilation. 
Wild adult An. gambiae mosquitoes were also obtained from NaVTongo (IOo30'N, 
I"OO' W) and Dodowa (soSI'N. 004 ' W) both in Ghana and Jaribuni (3°00'5, 3~OO'E) in 
the- Kilifi District. Kenya. I.aboratory colonies of adult An. Rambiae from Kilimanjaro 
(TIUlT..ania). Suakoko (Liberia) [which is the strain used as the: standard reference for An 
80mb/De 5.5,1 and Kisumu (Kenya) and An. arabiemu (Wageningcn strain) were obtained 
by courtesy ofM.D Wilson. 
University of Ghana http://ugspace.ug.edu.gh
3.3 Laboratory Rear ing of Mosqui toes 
In tilt' laboratory. the Iaf"\'ac and pupae samples were transferred into \4 x 24 " S.S em 
plastk basins (Fig.l.la). In cases where larvae of Culex and/or Aed~$ species were 
pre~-nt they wne identified by their angular position on the waler surface: and were 
removed. Using a plastic Pasteur pipelle. a few Anophelu larvae and pupae were 
tranSferred and plxc:d in 1.5 ml cppendorflubes containing I mI ofoopropanoJ and 
The remaining pupae were similarly transferred into small plu.<:lic cups and placed in 
labelled 25 cm cubic cages (Fig. 3.2b) for adull emergence. The remaining Iwvae were 
maintained in the bastns filled with waleT. from the collcction sile, to a depth oflbout 2 
em. The larvae were fed once on finely ground Nutrafin Boldfish food (Rolf Hagen, 
USA). 
The larvae and pupae were reared to adults under conditions of 27·3<f'C and 76±2% 
relative bwnidlly. The rooms had a 12 ht photoperiod. Each morning, the pupae were 
collt"Ctc:dand placcdinthtnpproprialdy labelled cages for adult emergence. 'Theadults 
that emerged \\~re transferred from the cages into small pap« cups. using an a.'ipiraIOr, 
and killed by brief refrigeration at 4°C. Adult mosquilo samples were preserved on silica 
gels at room tempe:rature and a few in isopropanoJ at 4°C. 
University of Ghana http://ugspace.ug.edu.gh
Fil-l.2b.CII,I!e:;usedror1berearingorpupuaodlbe~ofemerpdaduh 
"""'1"_ 
University of Ghana http://ugspace.ug.edu.gh
3.4 Identification of Anopheles species 
3.4.1 Morphologkal Identification 
lIIvae w.:rc identified as Anoplk/n hiUCd OD 1he fact that they Ii..: pataUel to the water 
surface and lack a siphon. Adult mosquitoes wttC' identified as AnopMles using the 
markings on the paJps, the banding and speckling on the legs and the characteristic pattem 
ofblocks of dark and paJe scaJeson the vc:in of the wings. espccially along thc costa (top 
part of the wing). Anophtler gambioe complex species were differentiated from An. 
/IIMShd spectes using the morphological key ofGill;cs 8IId de Meilloll (1968) and Hervy 
~( 01. (1998). Brieny. the species of the All gombitw complex. have 5 pale spots on the 
costaJ margin of the wings. anal vein coJoration with 3 whitesp<)ts, a dark apical f ringe 
and white speckled (or spots in the median part) tibia ornamentation. [n contrast. An.  
./uMstw and other species have 4 paJe spots on the costal mhrgin on the wings. entirely 
dark anal vein colOl1ltion. and entirely dark tibaa ornamentation. 
3.4.2 Molecular identification of sibling species of the Anopheles gambiae 
complex 
The PeR mtthod of Scott tI 01. (1993) for the species identificaLion of single specimen of 
the An. gambioe eomplex ~'a.s used. The amplification process utilised one un.i~ 
primer and fOW" species-spe-cific primers each of 20 bases rrlble 3.1). The universal 
primer UN &Meals 10 the same posiLion on the rONA of each the five species.. but the 
n:\"C"r:\C primer GA anneals specifically to An. gamb;oe s.s .• ME to both An. MerUJ and An. 
me/u.t. AR to An. arabiensis, and QD to An. quadrionnu/otus. The sizes of the amplified 
products are 153 bp for An. qllOdrionnulalflS, 315 bp for An. orabif!flJis, 390 bp for An. 
gdIrIbiat s.s. • and 464 or 466 bp for An. merwor An. mf!/o.srespeclJvely . 
University of Ghana http://ugspace.ug.edu.gh
T.ble 3.1 Sequence details and melting tempcl1ltures (Tm) of oligonucleotide primers 
used for the PCR identification or the An. KllmbuJe species complex (Scott dol. . 1993). 
I'rimcrllame Seque.tt:(5' t03') Tm("C) 
UN GTG TGC CCC nc CTC GAT GT 58.3 
GA CTG Gn TGG TCG GCA CGT n 59.3 
ME TGA CCA ACC CAC TCC cn GA 57.2 
AR AAG TGT CCT TCT CCA TCC TA 47.4 
QD CAG ACC AAG ATG Gn AGT AT 42.7 
University of Ghana http://ugspace.ug.edu.gh
3.4.2.1 Genomic DNA extraction 
Each $(X~' llIIcn wa~ Initially "ashed once in a solution of 400 IJI EDTA buffer. 10 IJI of 
25% 50S and 5 III of 10 mglmllysozyme, and incubated at 31'C. with agitation for 1 hr. 
This washing prococol """as designed to wash Iway external sources of bacterial DNA by 
l)Slng bacterial cells on the c:.:krior surfa:c of the specimen. The specimen was then 
homogeniSC"d in a I.S ml Eppendorftube containing 100 III Bender" buffer (preheated It 
65·q tWng a sterHe polypropylene rod foUowed by incubation at 6~C for )0 min. Then 
125 ",I of buffer salwaled phenol were added to the homogenate and mixed well by 
vortexing at 1000rpm and spun at 14 krpm for 10 min. The supernatant was transfened to 
a fresh tube, and 250 pi of chloroform added; it was vortcxed briefly ond spun at 14 krpm 
for 10 min. The supernatant was again transferred into a new tube and 250 III of pre-
chilled absolute ethanol and 10 JJl of 8 M poto.. . sium acetate added, followed by 
Incubation at -40"C (or I hour. The DNA was pelleted by centrifugation at to krpm for 
10 min and the supernatant was discarded. Two hwtdn:d micro litres of 70% ethanol 
~re added to the pelkot. the tuh.e gently swirled and the DNA re-pelleted by 
caltrifugatton al 10 krpm for S min. The supernatant was discarded and the lube opened 
and mvert.ed over a paper towel to dry. The dried DNA pellet was redis.o;olved in 25 j.11 of 
IE t RNue (Sj.1Wml) and incubated on ice for about an hour. it was then slored at -200C 
uotilnttdcd. 
3.".2.2 peR ampUfiQtfon 
Each I"('&clion mix of 20 .,,1 contained Ix PeR buffer (" (from the Invitrogen ')("R 
Optirni.n:r Kit), 200 ."M each of the four deoxyribonucleotide triphosphates (dNTP.s). 
0.25 ."M ~ach of oligonucleotide primen and 0.5 U or DNA raq polymcra.tc enzyme. 
One nUcrolitft of c:x.tr8c:ted mosquito DNA was used as tempiatt' in the amplifICation 
rcection. The rc:Ktion was lhoroughly mixed. centrifugcd bricfly II 10 krpm and OYft"laid 
University of Ghana http://ugspace.ug.edu.gh
with 20 ~I of minet1ll oil (nuc~ free) to avoid evapomlion and rcnu:oting during 
thennocycling. The temperature profHc for the ~actions was 940C for 3 minutes (Utili" 
mell) follo ..... ~d by 35 cycles of 940C for :W se<::oods (denatunltton). 500c for 30 seconds 
(annealing). n·c for t minute (extension). and a final cycle of 72°C for 10 minutes. for 
('OKh sct of reactions. a negative control. which contained no DNA template was 
perfonned. The amplifICation reactions were carried out using a PCR Express Thennal 
Cycler (Hyboid Ltd .. UK). 
3.4.2.3 A". . tysls of PCR products 
Eight microlitres of each peR product were added to I 1-11 of lOx bromophenol blue gel 
kMding dye and electrophoresed in 2.0% agarose gels stained with 0.5 "glml ethidium 
bromide. ·rne gels were prepared and run with I X TAE butTer, us.ina either a midi· or 
maxi-acl.,.Slem.. II IOOV for one hour and were visuaJisedand photographed over a UVP 
dual inlmSity b'anSilluminator It short wavelength using a Polaroid dirttl screen instant 
aunen. fined with an oranse filter, a hood and a Polaroid Type 667 film. The film was 
processed as recommended by the manufacturers (Polaroid Inc .• USA). lbe sizes of lhe 
PeR products were estimated by comparison with the mobility of a standard 100 bp 
1adde,(S;gma,USA). 
109 
University of Ghana http://ugspace.ug.edu.gh
3.5 Detection of Bacterial Symbionts u sing peR 
To detect the symbionl"s DNA sequences in the life stages of An. ga",bjQ~ mosquitoes.. 
PCR with general and spccific cubaclcria oligonucleotide primer sequences was used in 
the larvae, pupae and 8dult mosquito tife stages. peR reactions using general primers 
designed from sequmcesofbecteriall6S rONA (IDltrOt . fDlIrPl and fD2IrP2) and ns 
rONA (rRNA FI and rRNA R) were first carried Qui, and then another PCR with specific 
primers (WOLD 16SF I!WOl.B 16SR 1 and fbZI/ft~Z2) was carried out to dctennine if the 
symbionts dct«ted were lhose of Wolbachia sp. The details of all the primer sequences 
:lie gi\'cn in Table 3.2. The genomK: DNA extracted from larvae. pupae and adult An. 
RQ",bia~ 5.1. specimens (see Section 3.4.2.1) were used D.S template for the PCR. 
The eubacleria 165 rONA primer stts amplify sequences of a wide variety of bacteria 
taxa (WeisbW'g ~I aI., 1991). ThC' 23S rONA primers were used to confinn the initial 
resuJtsobtaincdwiththc 16S rONA gene primcrs(RollUe'teta/., 1992). Both primer sets 
ampHI'y approkimately 1.5 kb DNA fragmenLs. 
Each reaction mix of 25 J.LI contained Ix buffer C (from the Invitrogen peR Optimizer 
IGO. 1.5 mM MgC12• 0.40 mM each oC !he .. dN·rps. 0.40 ~M of taCh oligonucleotide 
primer and O.S U of TOif polymerase enzyme. One microlitre of a single mosquito DNA 
exlnlct was used as template in the amplification reaction. The reaction was thoroughly 
mixed, CUltnfugcd bndl~ at 10 kq)m IDd ovataid with about 20 ~I of nucleusc: tree 
mineral oil. 'The amphfic31wn reactions wrre carried out using the peR Ekpress Thermal 
Cycler (Hybaid Lid .• UK). 
110 
University of Ghana http://ugspace.ug.edu.gh
Details of oligonudeotide primer sequences used for peR detection of 
bacterial symbionts in An.. gambiae 5.1. and their melting temperatures 
Sequence Size T .. 
GtDelSptc::its Primernamt (5' 103') (mers) ('C) 
Eubecterial IDI AGAGTTTGATCCTGGCTCAG 20 53 
16SrRNA rDI AAGGAGGTGATCCAGCC 17 52 
rPI ACGGTT ACCTTGTT ACGACTT 21 53 
ID2 AGAGTTTGATCATGGCTCAG 20 51 
rP2 ACGGCTACCTTGTT ACGACTT 21 55 
Eubacl"ri:li 23SrRNAFI CCGAA TGGGGAAACCC 16 57 
23SrRNA 23SrRNAR CCACCTGTGTCGG1Tf 16 49 
Wolbociriasp WOLBI6SFI AGTCCTGGCT AACTCCGTGCCA 22 64 
WOLB I6SRI TCACCCCAGTCACTGATCCCAC 22 62 
j/,Z] GTATGCCGATTGCAGAGCTTG 59 
j/,D GCCATGAGTATTCACTTGGCT 21 55 
-----------------------,111 
University of Ghana http://ugspace.ug.edu.gh
The optimised PCR conditions for the 16S rONA were: an initial denaturation stage of 
94°C for 3 min. followed by)S cyclesof94OC for 60 s. SifC for 60 sand 72°C for 2 min 
mel ended ....; th 3n additional elongation step of 71!'C for 10 min. The same reaction 
conditions ....' Cre used for the 2JS rONA amplifications, except that the anneaJing 
Icmperalutt was set at 47"C. 
Samples which were positive fortbc presence o{both 16S and 23S bacterial sequences 
wet'C sclecfed and u.tcd for another round ofPCR, using pre.viously dcscribc:d Wolbochio 
primers (flolden 'Ial., 1993; W. ...l eers & 8;I\.n, 2000). The WOLBI6SFI (foruvd) 
and WOLBl6SRI (reverse) universal primer pair is highly specific for Wolbachia, and 
amplifies a 1000 bp of the small subunit (SSU) rONA of all known Wolhachia strains, 
including the reunt1y discovered CandO strains found in nematodes. but no other 
related b.cleria (WemclOCTS & Billen. 2000). A fu.rthcr peR was carried OUI using the 
JbZlJftsZl primer set wbkh is basN on sequences in a gene coding for a protein that 
initiltes cell division in prokaryotes and is broadly reactive with Wolhachia spccin in 
insects (Holden ~t aI. • I99J). The approximate prodUCI size for this set of primers is 769 
bp. 
Each rellCLion mix of 20 J.LI contained 1 X PCR buffer C (from the Invitrogen peR 
Optimil.er Kit), I.S mM MgC12. 0.40 mM each of the four dNTPs. 0.40 tAM of each 
oligonudco(ide primer. 0 .5 U of Taq polymerase: enzyme and O.S ~I of DNA as template. 
The reaction mix was treated as befon:: and the reK1ions "'-l!r'e carried out using the peR 
Expre>s Thennal Cycler (flyba;d Lid., UK). 
-------------------'112 
University of Ghana http://ugspace.ug.edu.gh
For the WOl.816SFI I WOlBI6SRI primer!. the kmpcratlR profile was as follows: an 
initLaI denalur.ltionat 94"C for 3 min, followed by 35 cyclcsof94I1C for60s. 6(/C (or 60 
s aDd 7'rC for 2 nUn for 3S cyclcs.. plus one oddilional cycle with • finaJ 10 min chain 
dongJtion step. The same reaction conditions were used for the ftsZll/ilZl primers, 
cxccpt for tbe anneaJing k'mpen.ture which was sctat 55°C. 
Template DNA from locally caught brown-banded cockroach Supello long/po/po 
(Blattaria : 8lauellidae) nymphs was used as positive control for the presence of 
Wolbach,." Negative controls. containing only double-distilled water, were abo included 
to check r.,r contamination. lbe peR products were electrophoresed in D. 1.00/. agacosc 
gel as described under section 3.4.2.3. The sizes of PCR produclS were estimated by 
comparison with the mobility ofastandard I kbladder 
3.5.1 EstimOiltion ot the concentration of peR products 
Five microli1tes of PCR product was diluted to 95 ",I with nuclease-free water and the 
absorbance reading al 260 nm wavclcngth (A2*) was m:ol'dcd. TIle conct:ntrntion of 
PeR product (X tAg/tAl) was then estitnlted by using the formula (Wilfinger el al. • 1997) 
below, 
Al6O X dilutionfactorxSOpglJ,ll- X tAg/p.1 
-------------------'113 
University of Ghana http://ugspace.ug.edu.gh
3.6 Amplified Ribosomal DNA Restriction Analysis (ARDRA) 
Reruiction analysis of enzymaticaJly amplified 16S rONA ..:an be used to identify the 
species of many bacteria genera. This exploits the differences in the DNA sequences that 
resulrsindifTt."1cntrestrictionsiles. lntbis variationofribotyping, the ribosomal DNA is 
amplifted and the product digested with • restriction enzyme. and the diagnostic DNA 
fi1IImenlprofilesare vi5U&l.ized following separation by gel electrophoresis. avoidinglhe 
need ror Southern blotting which involves cwnbersome blolling techniques. This method 
has been used to differentiate bacterial species (Vaneechoufle, 1996; Dijkshoom d al. • 
1998). 
3.6.1 In silica (computationill molecular bloJogy) restriction analysis 
Tbe most predominant bacteria species detected in An. gamblae have been E$cherichia 
coli and Punt(HJe agglomerans (Slraif et oJ. . 1998). In order to identify whether M)' of 
dleamplicons might bceitberoftbese microotganisnu. virtual restriction analyses were 
perfonned on their rONA nucleotide 5CqUCTICes. 
The complete 16S rONA scqu~nces of £ coli and P uKiJlomrruns Md the complete 23S 
rONA sequence of E.. col. were retrieved from the Ribosomal Database Project II (Cole et 
aI .• 2(03) and copied into the software "Jellyfish" version I.l (biowire.cOm). The FIND 
rWlction of Ibis programme was used to align the primer pairs used for the peR 
amplir.cations to the homologous rONA sequences, and the region thai they flanked was 
copied, saved as a new file and then imported into the software programme DIGEST 
vers.ion 1.0 (Nak..isa, 1992). InsJlico anatysis WaJ pnfonned with the DIGEST software 
using the enzymes in its database WISCONSI.920. The output included the positions of 
the DNA sequcocc:s thai each enzyme cut. the number of fragments and their slzJ:s. It 
-------------------n~---
University of Ghana http://ugspace.ug.edu.gh
also listed all enzymes that do not restrict The results w~re compared to those obtained 
by the experimental restriction p8I1tms of the amplified 165 .nd 23$ rDNAs. 
3.6. 2 Rutriction endonuclease digestion and analysis 
hom the results obtained from the in silieD analyses, the ..:nzymes that cut at less than six 
sites, and some which had no sites. were selected for this anaIyis. Dral. EcoRl. Hindlll. 
hut. Soli and XbaJ were selected for the analysis offDl/rDI. fDJ/rPI. fd2/rP2 and 238 
rRNA FIIRI amplicons and Apol. £CoRl. Hindill. &01. Nspl. Hinjl and BamHI for the 
WOlBI6SFII WOLBI6SRJ amplicons. (be digestions were carried out iii 
recommended by the manuf.cruten (Sigma-Aldrich, USA). The final reaction volume of 
20 j.iJ contained 2-6 ~1 of the amplified products. The incubation was carried out at 37'C 
for 2 hours using a heat block. The products were electrophoresed inethidium bromide-
ained 2% .garose leis using IX TAB buffer and the results were visualised by 
lrarWllwninaiion. A standard 100 bp ladder was used as roolecular weight marker. The 
sizes of the die;ested products were calculated u.slna the computer software DNAFRAG 
version 1.03 (Nash. 1992). 
115 
University of Ghana http://ugspace.ug.edu.gh
3.7 Cloning of Amplified Bacterial Sequences 
I. ciani", of PCR products allows the genention of relatively large amouftlS of the 
MIpIified teSion. which can then be con\'CflieOlly uxd. v.rt.encver needed. for 
experiments such AS sequcncina and hybridisation (Newton & Grah;un, 1997). 
The cloning process involves \be JipUon. using DNA ligase. of the amplifltd DNA into I 
suilab&c plasmid vC'Ctor to fOrm • recombinInC molecuJc. followed by transfonnation of 
Ihc recombinant molecule inlo suitable competent bacteria cell. for replication and 
multiplication. To select for transformants. "blue-white" colour scrtening, a well-
c:Slablished means for identifyins a ligation product or indk:atina the presenc:e of a DNA 
insert. is employed. The mdbod is based upon the abilily of P-plKtosida.te to hydrolyse 
S.tJromo-4..chloro-3-indolyIIJ-D-galactopyranoside (X-pi). resultina in the characteristic 
blue ....... of. coIooy .. phose pIoque <HorwOz., ul., 1964). The multiple cloning 
rqioa of NImCI'OUS plambd and phage vectors is imbedded within the amino-tenninal 
frqrnenl of .. i_I ~D-thiogalaciosiclc (IPTG)-i_'" 1I-Pf.o<I0>ida0e 8<"". 
Successful liption of a DNA fr.tgment ink> the multiple cJonina site disrupts ~ 
galactosidase expression; without insertioo. expression 110 unincmuplCd. Thus. when the 
f." mli host, ell.prcuina the carhoxylenn.inaJ portion of fS-g.daclosidase, is transformed 
with I p&a.nid vector or infected with a phaae withom I DNA insert, a "blue" colony or 
plaque wiD result (Ullmann el U/., 1967). Transformation Of infection of I vector with 
DNA fraament inxrtion results in I "white" or clear colony or pl-tue. 
For this study, the TA eloaiac method was used 10 clone tbe Implified bacteriaJ 
!llC'Qucrxa. The TA cloninS method takesadvantaccoflhc1erminal transrCT<LSe activity of 
some DNA pol)fncrases such as TOIl poIymrra.,,<. This enzyme adds a smale. ) '-A 
OV,","", \0 each end or \he PCR pn>duc1 (CIaot, I9IS, HII, 199)), _ makes il 
-------------------116--
University of Ghana http://ugspace.ug.edu.gh
possible to clone it dltectly ineo a linearized cloning vector with single, )'.T  overhangs 
eTA cloning). DNA polyrncrases with proofreading activity, such as Pfo polymerase, 
cannot be usN bcal.l3C they provide bluol-cnded peR products. 
3.7.1 dA tailing of peR products 
To enable a hlah efficiency of cloning of the PCR product in the TA Cloning system 
(Invitrogen, USA), the addition of J'A-overbangs post amplification was carried out. 
Twmty·five micmlilrt' reactions each consisting of 19.75 ~I of diluta:l peR product ( 10 
in 9.75 ~I oeWaltr), 2.5 ,.d lOX PCR buffer C (from the Invitrogen PCR Optimizer Kit), 
2.5 "at 2 mM dA TP and 0.25 "I Toq were set up. Each reaction mix was vQrtcxed, briefly 
centrifuged and incubated aI noc for 10 minutes in a thennal cycler. The products were 
used immtdiately for the ligation reaction 
3.7.2 Ligation 
To cstilTUllc the amount of PCR product needed to ligate with SO ng (20 fmoles) of 
pCR~.1 veClor, the formula (TA  Cloning Manual. Invitrogen, USA) below was used: 
(V bp peR productXSO iii pCR"2.1 v<c .or) 
XnaPCRproduct-
(size in bp ofthc pCR*2.1 vector: - )900) 
""here X ng is lhc aJ1lol.lnl of peR product of V base pairs 10 be ligated in. 1:1 (vector: 
inscrt) molar ratio 
U5ing the concentration previously delemlined for the PCR product. (see scction J.S.I). 
the volume needed to sive thl! required amount as detmnined using thi.! formula, was, if 
QCQC:5Sary, obcained by diluting with sterile water, One vial of pCR*2.1 vectors 
(Invitrogcn, USA) was centrifuged bri.:Oy to coUect aU the liquid in the bottom of the 
"'-I . The lillfion reaction mix was col1lp054:d as follows: 1-3 ~I dA-tailcd peR product. 
--------------------------------------'117---
University of Ghana http://ugspace.ug.edu.gh
1 loll of lOX I.igation nuffer. 2 JlI pCR~.1 vector (25 ns!pl), -.vim _rile water to make 
up the volume to 9 JlI and I JlI T4 DNA Ligase (4.0 Weiss units). Each tube was 
vonexedaodceatrifuged briefly. The liption mix wu incubated" 14°C overnight. The 
products of the IUclKtRS lhaI were DOl immediately used for transformation. were stored 
.-2O"Cuooil....w. 
3.7.3 Transform_lion experiments 
To perfonn transformation eXper1meRlS. the cemperature of tile water bath was set II 42°C 
and the SOC medium at room temperature. The LB pl.lltS (two pillcs for eac:h 
liptioniU'liruft.nt\illion), containing 100 mg/m1 ampicillin, were opened and dried in a 
blctrrioJQ8ical hood UDder UV light at 37°C for 30 minutes. Each plate was labelled and 
sprad with 40 III of 40 maim! X-Gal and the liquid allowed to s(lak into lhc: plates for 
.bo~ I S minutes under sterile eonditioN. The b"ansfonnation pcoccdune was carried oul 
roUawing the ~ fMftU31 supplied \\1m the TA Cloning· kit. with sligh!: 
The vials containing the ligation reactions were briefly centrifuged and placed on icc. 
One SO .. I vial of frozen One Shol~ chemically competent E. coli cells for each ligation! 
tnnsfortnltion wu thawed on ice. ImmedialeJy after the cells had thawed. 2 III of each 
lil.tion raclion was directly pipened into the vial of compdtnt «lis and mixed. by 
gendy tipping the base of the vials. They were then inc:uNled on ice for 30 minutes and 
then hea1.shod:ed 11 42ee in • "''akr t.th for euttly 30 seconds, avoidin, any .')hal..ing 
and mixina· 11lC' vials were then removed from the water bath and placed on ice for 2 
minutn. Two bundn:d and fifty microlitrn or the SOC medium "'Cf'C 8dded to each vial 
and the vi.l.ls dWten hori/..ontaUy at Jr(~ for exactly I laour 81. 225 rpm in • sblking 
incubetOf. The vials were neX1 reruovcd, placed 011 ice fOf • few minUlcs BDd ceatrifua.ed 
-------------------IT~ 
University of Ghana http://ugspace.ug.edu.gh
for a minute at 10 Iupm. The supernatants were dilt..ded w the pelkts resuspended in 
100 .,1 of SOC medium. Five microlitres and 95 JII of cells from cadi trartsfonnarion "ial 
~ spread scpara&ely on the LB a@arplaln. The pla&es were then incubau:d .t 3'tC for 
.. lea 18 holn and b'M5femd to • +4-c CI"Ivitoamenc for 2·3 hours for colour 
To dcletTl1ine the pre~nc(' 0( iasert.s, 10-15 whik colonies were randomly pid.:cJ from 
ICletted plaSH for plasmid isolation. The colonies 'A"Cre grown ovemiahl in S m1 LB 
broth containing 50 JlI'mI of ampicillin in a rotary shaker at 3'PC. Then. 100 ",I or tile 
brotb were removed and placed in O.S mJ microfuge rubes and heated at IOI1'C ror 5 
minu&cs in. thmna1 cycler and placed immediately on ice. The rubes "-ere spun at 10 
krpm in a micro(uae for 10 nUnuees and the supemMant uxd for peR. 
Each PCR reaction mix consilCCd of I6.0S", PCR _ .... 2.5 ~IIOx PCR butTer C (from 
the lnvittopaPCROptimiz.erKit), 2.5 ",1 20mMdNTPmix, O.625 J.11 eachofw 20 JIM 
Uoi ...... Mil ForwanI (-40) ond "',. • "" prim.", 0.2 ... Toq polymelMO (Su/~I) ond 
0.25 pi of supernatanl from ,he cell preparalions. 1bese were thoroughly mixed, 
cmlrif'uFd briefly and overlaid with SO fil of mineral oil. The PCR reaction cycling 
prutile was 30 cycles of I minute at 9s<'C. I minUle aI S5-c and 1 minule at 72'C. and 
ended with 7 minutes at 7rC. Ten microlitres or each rexlion was electrophoresed in 
1% aaarose iels and visual~ as described before (see X'Ction 3.4.2.3). A PeR pnxIuct 
bmd size ar approxirMlely • toOO bp was observed for vectors with ioscn. 
3 .7 .4 Plasmid DNA isolation 
Posilive colonies were pn.~scd ror plasmid DNA imlation. One mkrolift of cell 
""- - pct- by eenlrifugotion .. 10iapm r.. 2 minuIeo. ond DNA from the ceUs. 
l" 
-------------------119--
University of Ghana http://ugspace.ug.edu.gh
extracted using the QlAprep Spin Minipn:p Kit (QiAGEN. USA) protocol IS desc:ribed 
below. The bufTersand reagcntsused were partofthek.il supplied by thc manufacturer. 
The S\I~m:uanls were discarded and the pcllctcd becteriaJ cells resuspended in 250 ~I 
Buffer PI containing RNase, making sure that no cell clumps were visible after 
resuspension. Two hundred and fifty microlitres of Buffer P2 was added and the 
suspension mixed well by gently inverting the tube 4-6 times. When necessary. the 
inversion of the tube was continued until the solution became viscous and slightly clear 
but trus was not allowed to proc«d for more than 5 min. Then, JSO .-' of ButTer N3 was 
added and the tube inverted immedillely but gently 4-6 times 10 avoid locaJiled 
~cipilation. followed by centrifugation at 10 krpm for 10 min 
ThcsupemalantsobtainedweretrlDSferm.lonloaQIAprepcolumnancilhencentrifuged 
foc 1 minute and the flowthrough discarded. The QIAprep spin column wns washed with 
O.S mJ Buffer PB and cmbifuged for I minute and the nowthrough again discarded. The 
QIAprep spin column was again \\1lShed with 0.75 mt Buffer PI: and centrifuged for 
another 1 minute and the nowtlvough treated as before. An addilionaJ 1 minute 
centrifugation al the same spet'd was done to remove residual wash buffer. To elute 
DNA, tht QIAprep coIwnn was placed in a clan J.5 mI microcentrifuge tube and 50 ).II 
Buffer E8 (10 mM Tris-CI. pH &.5) added 10 the centre or each QIAprcp column. left to 
stand for I min, and centrifuged for I min. The isolated plasmid DNA was checked after 
a 0.&% agarosc gel e&ectrophoresis and visuaJisc:d Wtder UV light The expected band 
KzeortheisolatedpJasmid ....~ 8pprollimately5kb. 
-------------------120 
University of Ghana http://ugspace.ug.edu.gh
3 .8 Identification of Amplified Bacterial Symbiont Sequences and 
Phylogenetic Analysis 
3.8.1 Sequencing of amplified bacterial DNA sequences 
Tbe sequencing of the: inse" DNA in the pCR~.1 vector was earned out hi-directionally. 
The MIl reverse primer \Io"'lS used to sequence from the loe promoter end of the veclor 
and the Mil forward (-20) Primer from the (uclA end (Fig. 3.3). 
Five microHtres of plasmid DNA was diluted to 80 J1l with nuclease-free water and the 
concentration of the plasmid DNA was delennined by absorbance re.ding at 260 run 
wavelength (AltO). The conccn1lalion of plasmid DNA (Y J..lg/fJl) was estimated by using 
the (onnula (Wilfingn et al. • 1997) below: 
AltoxdilutionfactorxSOJJglt,d-Yf,lgI",,1 
lbe sequencing reaction mix contained plasmid DNA (final concentration of I J..lg/JlI), 
0.65 j..tl of 5 J.!Mcachofeilhertheforwat'dorrcvcr.;cMIJ primcr and nuclease-!ree water 
10 a final volume of 9 j.1l and lhe reaction carried out using the ASl "RJSM Big Dye 
Terminator Cycle Sequencing Read)' Rcoclion Kit (PE-Applied BiosySlcms, USA) and 
analysed with the Perkin Elmer ABI 377 automated sequence analyser (Applied 
BiO'\~ siems. USA). Raw chromatogram reads wcre base-called using PHRJ~D (Ewing t!I 
aI. • 1998). 
3.a.l.15equenceedtting 
The sequence dala obtained were copied iato the MacDNASIS (version 3.2) software 
package (Hitachi Software Enginccrina Co .• ltd., Japan) and manually edited to remove 
the EC'oRI sites and the '"OGTCC" sequences that flank the insrrted PCR amplified 
fragmcnton each sidr or the sequenced primer ends, 1'hccomplettsequenceofthc P('R 
product Wll.S gCnttatl!d from the -shotgun" fragments usina the AUTO CONNEC nON 
121 
University of Ghana http://ugspace.ug.edu.gh
:;;:gg=~a.rig~'!!g~~~~'~!~ 
~v .,..'..,.' -, ""t' ;:t' 
~!,!~~=;:==~~:!=I.o.~~OID:;!!:¥ 
~~~~;f!E!fE.hm~!~ ~~~~ 
FIC· 3.3 Map of the linearized veclor pCRtI>2.1. The: sequence of the mulliple cloning 
lite is shown willi • P('R product inserted by TA Clorung.&\ E,'oR I sites flank the 
insetted peR product on each side. The arrow indicates the start of transcription for the 
17 RNA polymerase. 
------------------------------------'122 
University of Ghana http://ugspace.ug.edu.gh
option W'Kkr the CONnG manqcr function of lhc software package An overtop 
minimum six of 20 bp and • minimum matchins percentage of 6OI'A. wen: selected. 
3 .•• 1.2 VecSc,een and chi ...... detectiotI 
Prior 10 sequcnce analysis.. sequences ~ finl saunrd to revcalllnd remo~ any o(the 
v~tOfS, using the VecStteeD programme from the BLAST webpa&e of the National 
Center for 8iotechnolo&y Information (NCBl) websile (Altschul el aI., 1990. 1997). 
VecScreen searches a quay for segments that match any s:equence in a specialized non-
redundant \'eelOr' dltabue. The search UICS BLAST paramcltrs preset for optimal 
<SelCCtion of veclot coalUnination. Thox segments of the quc!)' dwt match vector 
diJplayed. 
XvaaI studies have indtcMcld chit • signirKanI fraction of the 165 rONA scqumc:es 
obcaintd by PC1l amplification of acnomic DNAs utnclN from ft\vVonmenlal SImpleS 
could be recombinants Of ~himmc (as 3. consequence of PeR coamplificalion from mixed 
aenomn(Li~kelaJ. • 1991;Choi elaJ. • 1994; Kopczynski elui. 1994; Wanl& Wang, 
1996). Chimera fonnalion is thought to occur when a prelMt..-ely lenruMltd amplieon 
re-anneals 10 a (oreian DNA strand and is copied 10 completion in the followina PeR 
cycles. This ruultJ in • sequence compoted of two or more phylogenelically distin<:t 
parml ~ucnces and. when comparatively ...... Iy.scd with other 165 rONA sequences, 
sunc:sts the prt'scnc::c or. non.-cxistent organism (Hugenhoitz & lIuber, 2003). To check 
fOf possible chimeric sequences created durina ImpliflCation, lhe CHE<.:K_CHIMERA 
..- ve"ion 2.7 .(the R.;bo>omallJara_ Projcclll (Maidak .,01 .. 2001) _ ...... 
This PfOII'WiDI c:u help lO decmnifte if.!Cq\ICDCe is composed or two rragments that are 
------------------123--
University of Ghana http://ugspace.ug.edu.gh
similar 10 cJcarly difTereot sequences from thedatabasc and uploaded user sequences, i.e. 
if~ sequence is of chimeric origin. The default settings for the program wc~ u.ted. 
3.8.1.3 Seiluenc. similarity and DNA databas. searches 
After editing. the Sl!qlkncCS WeTe submined to the standard nuclernidc-nucleotide BLAST 
dal.l~ snrch programme (blastn 2.2.1) of the NeBI website (AJtschul el al .• 1990; 
Altschul el at.. 1991) to search for closely related sequences in the non-redundant 
GenBank database. The default parameter settings were used. Related sequences were 
acquired by using the: Batch Entrez progrvnme (NCBI). 
Percent similarity between the sequcnc:~ obtained was determined by using the maximum 
matching option under the compare function in the DNASIS software package. This 
function compares {W(l DNA/amino acid sequencn and Illigns them for maximwn 
homology. Homology piau were alJo constructed using the: Homology option Wlder the 
compce fwx:tion of the software 
3.8.1.4 Construction of consensus sequence 
Based on the resullS of the bomulogy plots. sequences showing o\"cr 80% similarity to 
oneanolhtrweregroupedandalisnedlogcneralcaconscnsussequenceusinglhe 
multiple alignmtnl funcLion in DNASIS. The automatic Higgins option was used with 
the following multiple alignment parameters: gap penalty of 300, fixed gap penally of 20, 
fbting gap pcnalty of 20, k-tuple vaJue of 2. windowsiz.e valuc of 5 and number of top 
diagonal value of S. The alignments were compldcd manually and the: consensus 
sequences 8ft'MIlCd subjected to BLAST sean;hes as described prcvtoudy. 
124 
University of Ghana http://ugspace.ug.edu.gh
3.8.2 Phylogeny con5truction 
Sequences homotogous to the consensus sequences .... erc retrieved from the ONA 
cbtabanks for comparisons. For sequences of groups of nearly identical groups only 
single representatives ~re used for phylogenetic anaJysis. Reference 165 rONAs used in 
the analyses wen: obtained from the GcnBank and are shown in Tables J.3a, 3.3b and 
J.3c. 
Inc sequence data were aligned using the CLUSTAL W package (Thom~on el oJ., 1994) 
integrated into the MacVector 7.1 software suite (Accelerys, Pharmacopeia Inc .• USA). 
For both pairwise and multiple alignmeots. the open gap pcnaJty of 30 and extended gap 
pmalty of 10 were applied. A slow alignment mode was selected for pairwise 
alisnmmts. For multiple a!ignments, I delay divcrt;cnce of 60% was selected and the 
trvlsitionswere ....' Cighted 
The re,ioru of the DNA tequmces that aligned ambiguously (base positions which ..... ere 
of indeterminate identity, insertions and deletions, and aJigrunl!nl gaps). most of which 
com:sponded to variab&e regions of SSU rRNA, ....' Cre not used for the subsequent 
dclenninalion of phylogenetic relationships. A number of algorithms, including 
neighbour-joining. maximum likelihood and maximwn parsimony. for inferring 
phylogeny were used to evaluate the sequmce relationships. 
125 
University of Ghana http://ugspace.ug.edu.gh
Tabk J.Ja Baclttia species and strains used for the phylogenetic analy~is of 
CONSEN4. 
Organism 
accts5ionllumber 
X78717 Purple bacterium (wwuned) 
AB041770 Paroco«w kuwasaki~nsis 
YI6930 P, de"itriftcQIIS 
016429 Rhodobocterblastica 
YI2703 p, mDrcusJl 
All006899 P. carOl;nifoc;~ru 
Af365994 Marine alpha proteobactcrium 
BBATJ 
X53855 R. sphaeroicks 
ABO I 7797 Rlwdobacter sp. TCRI S 
016427 R. cupsu/orns 
AYOl4179 P. ),u;;slramGJ06O 
070847 R.azotojormuflj' 
032238 P. ulkaliphifus 
AYOO5463 Ruegeriosp 
032241 P. Weurll 
Af245634 UncuJturedRO$toboclersp 
X78315 R. aJgocolus 
088523 Agrobaclerium gelollnovorum 
032240 P. aminm'Oram 
-------------------126 
University of Ghana http://ugspace.ug.edu.gh
Yabk J.3b Bacteria species and strains used for the phylogenetic analysis of 
CONSEN2. 
UenBonk Organism 
accessioflflllllfber 
AJ44075 1.1 Gram-negative bacterium MM I 
AF502217.1 Uncultured bactenum clone HPIB64 
X78717.1 Purple bacteriwn (unnamed) str3in SW:! 
Y16930.1 ParacoccusQeflilTijiCQrIS 
AF229874.\ Parococc:ussp. 4FB8 
032238.\ P.aJkallphlius 
AYO\4176. \ P.mrrinophilw 
AF527586.1 Uncultured bacterium clone LPB54 
AF1l6850.1 Kelogulooogemum robuslum 
AYOI417J.\ P ~ei;strainGt212 
AF445668.\ Uncultured alpha proleobal:tcnum done SMICll 
AFl68\83.1 Uncultured Rhodobacl~r group bacterium clone 
SBRTlS5 
AB0I7799. \ Rhodobocler sp. TCRJ 
127 
University of Ghana http://ugspace.ug.edu.gh
Table 3.3(' Bacteria ~rccies and strains used for the phylogenetic anaJysis of AgLS. 
Organism 
a('cessionnumber 
VIS83) lIymenobaclerroseosalivarius 
AY279405 Endosymbiont of Aphytis sp 
AY271)402 Endosymbiont of Aspldiocu.s "..,ii 
AY2741J9.1 Uncultured bacterium clone 0 113 
AFJ82107.1 Uncultured bactenum clone ZA2626c 
AF3J6J56.1 UacteriumWuba47 
D12658.1 CYlophogo auran/laea 
AYOJ8780.1 Uncu1tured CFB group bacterium clone 
TAF-B66 
AJ244689.1 Cyclobocler;um sp. V4.MSJ2 
AFJ61200.1 Unculrured Cytophoga/es bacterium clone 
30 
AJOl1917 F7«,obacillussp. 
AFJJ788J .2 Uncultured gold mine bacterium 028 
AY279404 Endosymbiont of Aphyfis lingnanensfs 
Af408175.1 Unculnued Hy",enobacler sp. clone KL-2-
4-9 
Al400J40.1 Uncultured marine bacterium Z00203 
YI88Jl Tateeoboc'"ocellatus 
128 
University of Ghana http://ugspace.ug.edu.gh
].8.2.1 Neighbour-Joining (NJ) analvsis 
The neighbour joining (N/) programme in the PHYLlP package version 3.Se (Felsenstein. 
1993) (or inferring phylogenelic relationships. also contains prog.rturulcarrying oul other 
rtlatcd wks U5ing different algorithms on difTerent kinds of data. 
SEQBOOT was flfSt used to bootstrap the sequences with 1000 replicates (Felsenstein, 
1985). DNADIST was then used to calculate evolutionary distances from the 
bootstrapped OUlput data with the Kimura two-parametcr model for nucleotide change 
(Kimura, 1980) using a transition-tnlnSVersion ratio. estimated from the data with the 
software PUl.ZlE 4.0.2 (Strimmer & von Haese)cr, 1996) and one category of 
subslitution rates. Phylogenetic trees were constructed from the cvolutioD8l}' sequence 
data usmC NeiGHBOR (Sajtou and Net. 1987). A conxnsus tree was constructed using 
CON SENSE with the trttfile from NEIGHBOR. 
3 ••. 2.2 Maximum IUceUhood iiln.alys's (ML) 
Phyk>tmetic lItt'5 wen: alto deriv~ from the molecular sequence data by the maximwn 
likelihood (ML) method carried out using the quartet puzzling (QP) analysis with the 
programme PUULE 4.0.2. PUZZLE is ANSI C compliant and a PHYUP compatible 
program that implements the quartet puzzlm,. method for reconstructing tree topologies 
from character state ciatL Quanet: puzzlina is. method that applies maximum likelihood 
1m: rec:onstrUCtion to all possible quan.ets of taxa and subsequently tries fA) combine most 
of tbt fow-taxa maximum likelihood trees to construct aD overull In:C. 
" 1000 puzzling slt:r~ U~II" tbt Ha.sc:sa .... a, Kishino. and Vano (HKY) substitution mode:! 
(Hascp ...... rl 01 . IQX5) with an estimated lransition-~rsion ratio. and site-to-site 
substitution rate varialion modelLed on I gamma disbibuLioo with four categories and the 
129 
University of Ghana http://ugspace.ug.edu.gh
shape pwunctc1 wen: estimated from the data. For paramctC't estimation. quarkt 
sampling (Cor suhstirution. process) plus NJ tree (for rW variation) WCfe used. 
3.8.2.3 JoIIaximum parsimony. analysis (HP) 
A MP tree requires the smaJlest number of ewlutio,*>, cblnges 10 result in the set of 
oro. under the study (Hillis. 1997). for each tree 1o be evalua&cd. the minimum 
possible number of chances for each 'CMrac\et' (nucleotide position or morphotoaical 
nit) is caJc:v1aIed, tad the minimum number of chartJcs across all characters are totalled, 
10 obc3in the parsimony score. The best tree is the one thai requires the fewest changes 
KfOSi all charatter$ (Hillis, 1997). Infonnation on evolutionary processes may be 
UlcOlpO;.ltcd by ~ightina characters differentially (such as first versus third positions of 
c:odoas).orby wcichtinacbarader state changesdifferentiaJly (such utransitions versus 
nnsvmions) 
Maximwn parsimony (MP) analysis was accomplished with the Phytoaenetic analysis 
w.ina: parsimony (PAUP) version 3.0r propun (Swofford, 1991). A slepmatrix. was used 
to give transvet'Sions twic:e the weight of tnnsitions. and pps were lreaIed as missing 
da&I. Two heuristiC -=-reh 5Irarcgies ~ UIed for each IMlysis, with ac!cheq • simple. 
and lht ICCOftd with IIddJeq • random and 10 rcplicaacs. Branch swappiOi was by tree 
bixction-reconnection. 
Consensus (50% majority rule) trees 1.Io-en: construttcd by usinS wx::orrecled NJ distances 
with 1,OOObootstnpreplicatcs. Iheseueesexc:ludedaroupings thatocc:urredinleathln 
5O'/.o(tbereplKaiesMdncgath·cbtanch k:nachswereprohibiled. 
1 30~ 
University of Ghana http://ugspace.ug.edu.gh
3.B.3 Nucleotide Sequence Accession Numbers r 
The :unplified 165 rONA sequences of the bac1eria from the mosquitoes. after !; 
iden'iticataoo, were submiued to the ONA Databank of Japan (OOJB), European 
Mol«ular Biology Laboratory (EMBL) and GenBank (at NCBI) nucleotide sequence 
University of Ghana http://ugspace.ug.edu.gh
CHAPTER FOUR 
RESULTS 
4.1 Mosquito Species Identification 
Gmunllc DNAs for peR were extracted from 432 mosquito specimens comprising of91 
IItrVK, 37 puper and 304 adult femaJes (Fig 4. 1) and these were ustd for the srudies. 
Out of the 432 mosqWIO specimens processed, PCR amplification for species 
idcntiticatton was successful (or 395 (91.40,..) while amplification failed for 37 (8.6%). 
Of the 395 peR positive specimens, 373 (94.4%) were identified as An KIJ,"b;n~ S.S., 20 
(5.1%) as An. arab/ens/s, 1 (0.25%) as An. merus and I (0.25%) An. me/as (Fig. 4.2 and 
Table 4.1). The 395 mosqwto specimens comprised 295 (74.1%) adults, 29 (7.3%) pupae 
and 71 (IS-/.) larvae. The details of the species identifted and of the life-stage forms are 
abogivco in Table 4.1. 
-----------------~Ilz_____ 
University of Ghana http://ugspace.ug.edu.gh
extrw:wd from AIL ga",biae 5.1. mosqui&oes. Lane \1 .}., DNA· Hind III diant; lane I -
oduU, _2-.14-1arwo;_3 o _ 
University of Ghana http://ugspace.ug.edu.gh
600bp 
400bp 
200bp 
Fic 4.2 Elhidi ... bromido-!iIaincd 2.0 % agam. . gel ekclrophor<gram of DNA bonds 
produced by the rONA·PCR idmtificltiun IDeIbod for members of dw: An. fOIItl1itn 
complex, Lanes I and 2 - An. RambiM s.s adults; lane ) • An. gambiur u pupa: lane 4 • 
All. p1ffbf« Sol l.w; lane 5 '" An. 1IWf1U adult; lanes 6 and 7 ., An. arabiensir adults; 
lone M - 100 bp !add" (Sigma-Aldrich, USA) 
University of Ghana http://ugspace.ug.edu.gh
Table -4.1 Distribution of species identified among the different life stlge forms of An 
xamhiCN complex species using the peR method of Scott el aL (1993) 
Number of AIL gambun complex species 
gambiots.s arabiensis melas 
Pupae 
Adults 274 20 
TOTAL(%) 313(94.4) 20 (S. I) 1(0.25) 1 (0.25) 
University of Ghana http://ugspace.ug.edu.gh
.,.. 4.2 peR Detection of Bacteria DNA sequences In An. gamblae s.l. 
The 395 An. gambiM 1.1 specimens were examined for the presence of symbiotic bacteria 
using the eubactm. 16Sand23Sprirners. CkrtofthetotaJof395spccimcnsstudied. 
DNA fragments of the predicted sizes \\--etC successfully amplified in 336 (85.1 %) and 
314 (79.5%) oflhcm for the 16S rONA and 23S rONA primers. n::-. p«tively (Table 4.2). 
Bacterial DNA sequences were amplified from all the sibling species used and also from 
Larvae. pupae and adult mosquito specimens (Figs. 4.3 and 4.4). Bacterial DNA 
sequences were also amplified from both wild and laboratory reared adult mosquitoes. 
The peR amplifications that were positive for both 168 and 23S rRNA primers produced 
single DNA bands in all cases and negative controls wilh no DNA template consistently 
eave no amplification product. Bacteria DNA sequences were amplified from specimens 
from aJi the mosquito collection sites. 
A lOCal vf 181 (11.1%) peR positi,·c specimens for both J65 and 23S rONA primcrs. 
were IItlected and screened for Wo/bochlo using the WOL J6 5 rONA and JlsZ primers. 
Using the WOL16S rONA primers, 94 (33.5%) spet:imcns yielded amplified DNA 
product of the expected J.O kb size (Fig. 4.5). The DNA fragments were amplified from 
larvae., pupae and adult mosquito specimens (Table 4.3). The peR amplifications that 
~ positive fOC' the WOl.16S rONA primers produced single DNA bands in atl cases 
and negative controls with no DNA template eonsistcntly gave no amplification product. 
None of the motquilO specimens. however, amplified DNA with the ftsl primer pair 
although aJl amplifications of the positive control were successful (Fig. 4.6). 
University of Ghana http://ugspace.ug.edu.gh
Tabl~ 4.2. PeR amplification of bacteria 16S and 235 rONA g~nc sequences in An. 
gtUIIbkMs.l. mosquitoes 
Sumber(-/_)positive ror 
LifrsllIgr Number 16SrDNA 23S rONA 
Larva< 71 54(76.1) 50(70.4) 
IJupae 29 26(89.7) 26(89.7) 
Adults 256(86.8) 238(80.7) 
395 336(85.1) 314(79.5) 
University of Ghana http://ugspace.ug.edu.gh
2.0kb 
I.Skb 
1.0kb 
O.Skb 
F" 4,3 [dUdium bn:InUde-s&amed 1.0% _garose sel electrophorcGran1 of amplified 
bIdaill6S rONA sequences from mosquilo spccimel'lS using rD2IrP2 primers. lane M 
• I kb ladder (MBI Fc:rmen13S, Gennany); Jane I -. An. ~umbiae s.s. larva; lane 2 - An. 
~ u.1uva; _] • An gulftbi« 5.S. pupa; lane 4 E An J(umbioe 5.5. &duh: lane 5 
• All lftIbJelUu aduh; lane 6 =- positive control (cockro.ch nymph: S""etlo ImrgipalptU); 
laac 7 ... ncptive control. 
University of Ghana http://ugspace.ug.edu.gh
... ~ I 
~.
2.0kb 
I,Skb --.....  
O.75kb 
O.5kb l --
f1a. 4..4 Ethidium bromide-stained 1.0-/0 agarose sci elec:trophoregram of amplified 
backN 235 rONA sequences from mosquito specimens usin& WAIIWA3 primers. lane 
M. 1 kb ladder (MSI Fc:rrnmtas, Germany); lane I "" An. gamblat s.s. larva; lane 2· An. 
gWNbIDr 1.1. P4*~ lane ) .. An. gambitlf S.s. aduh~ line 4 - positive control (coekmlch 
nymph:~IIQlfHtRIpalptU);llDeS-negativecontrol 
University of Ghana http://ugspace.ug.edu.gh
15 kb ---
I .Okb-" 
O.skb---
Plio 4.S Ethidiwn brornide-tIaIne 1.00';' agMosc ~el electrophoregram of amplified 
bKteria 16S rDNA sequences from mosquito specimens usin. WOLI6S primen. Lane M 
- I kb ladder (Sipnl-Aldrich. USA); lane I - positive control (cockroach nymph: 
s.".U.IDnglpalpus); laDe 2 - All gamblDt S.S. lar\'I; lane 3 ... An. gambiae $.!. pupa; lane 
.. -A. .. ~ u. 8CIuh.; lane S '" An. gambuw S.5. adult; IIIte 6 • ..tn. arabiensis adult; 
lMe7-acgativeeontrol 
140 
University of Ghana http://ugspace.ug.edu.gh
r .. "-, Ethidium bromide-Nined t.O% 19arose gel eledrOphore,a.ram of ampliflCd 
becteria 16S rONA sequences usin.JbZ prinxr5. Lane 1 • An. gambiut! s.s. larva; lane 2 
• ..eN. ~ u . pupa; \me 3 '" An. Rambioe 5.S. aduh; lane 4· An. gombl. u . ..tult; 
lane 5 • An. arablrmis Idult; lane 6 - positive control (cockroach nymph: SlIptlio 
""'PPa/"",); lane 7 • rqative conuol; lane M • 100 bp molecuiOl"';ght ladder (Gibeo 
BRL.VK) 
University of Ghana http://ugspace.ug.edu.gh ,4' l 
Table 4.3 PCR amplification of Wolbachia fisZ and 16S rONA gene sequences in 
specimens of An. gambiM s.l. mosqui1oes. 
Numberpositinror 
Ute.tage Numbu fslZ('!') WOLl6S(%) 
lArva< 0(0) 23(57.5) 
Pupae 20 0(0) 2(10.0) 
Adults 22' 0(0) 69(31.2) 
281 0(0) 94(33.5) 
University of Ghana http://ugspace.ug.edu.gh 142 
4.3 Amplified Ribosomal DNA Restriction Analysis (ARORA) 
AJI the amplified 165 and 235 rONA amplicons were analysed using six restriction 
enzymes Drat, £CoRI, Hindlll. hut. SaIl and Xbat For all the amplicons analy!ed as 
~h. 00 diITcreoce in restriction patterns was observed. With the exception oftbe £coRJ 
resuicliUtl enzyme, none of the othc.!r enzymes cleaved the 16S rONA amplicons (Fa, 
4.7). The EcoR1 enzyme cleaved the 16S rDNA amplicons into 2 fragments which were 
approximately 880 and 680 bpin siz.es. 
The 23$ rONA amplicons were caved by only Dral and Sail . Both CIU)'rJlC'S cleaved to 
live two fragmcnLS wbic:h were J0 20 bp and 540 bp for Drol, and 1260 bp and 340 bp for 
Sail (Fig. 4.8), The sizes ofthe restricted frngmenls obtained with each oflhe 6 enzymes 
~gi\'cninTable4.". 
<.:omparison of the results of the experimental ~ction pattemso( 16S and 23S rONA 
amplicons and those obtained from the Iff -"Iieo an.aJyscs for £ CD/i and P. agg/omeronr 
rcvetied thlt none oflhe pruduct$couJd be m..l of either E. coli or P oulomerotu(Table 
4.4). For clWnpte, both the 16S rONA amplicons and the homologous E. coli and P. 
agfDIMrans 16S rONA sequences were cleaved by EcoRJ, but only E. coli possessed 
raariction sites for HmcJIll. funhennore. none of the amplicons was cleaved with San in 
rontrasc to both E. coli and P. aggfomeram. For the 23S rONA, the amplicons and E. 
coil tequences pos5CSSCd sites for both Dral and San but only E. coli rONA was cleaved 
by £coRI. MOt'tOvcr. the sizesoftbe restriction fi'agmentsditrered in the <:ase of the San 
di,ee!l whkb (or the arnplicons resulted in fragments sizes 1260 bp and 340 bp. whilst for 
E..colltbey-.ue t340and 160bp. 
a•. AO .!) 
University of Ghana http://ugspace.ug.edu.gh 143 
Fia 4.7 Ethidiwn bromide·stained 2.0"10 qarose ael electrophoregrarn of restriction 
__ digc,," of 16S rONA peR prod_. Lane M - I kb mokculor woight raarUr 
(MBI fcnncnw, Germany); )...., I " £CoRl ; I. .. 2 - HloodIlI; .... 3 - SoIl; ..... -
DraI;_S-I'>Id;I_6-XOQ[ 
144 
University of Ghana http://ugspace.ug.edu.gh
I.Skb-
O.75tb -
O.5 kb -to 
fil 4.8 Ethidiwn bromide.scained 2.U'Y. aprose gel clectrophorcs,ram of restriction 
enzyme dilC"StS of 23S rONA PCR products. Lae M - I kb molecular weieht marker 
(MBI Fermentas, Germany); _ I - Oral; I. .. 2 - Pv.i ; lane } - XI>oi; lane 4 • £CoR!; 
lane\ = 1Ihtd111; 1ane6-Sa/I 
145 
University of Ghana http://ugspace.ug.edu.gh
Tabk-"'" Obsen'cd, and expected fragment sizes for amplicon., E. coli and P. 
UUlonwrans rDNAs after digestion with the various restriction enzymes ( - denotes no 
restriction) 
Rcstric:lioD fragmeat sizn (bp) for 16S RCSlric:tion fra281c., 
rONA primers sius (bp) for 23S 
Restriction rONA primers 
Amplicon, P.agglomf!rllns* Amplkont E.coli* 
n..1 1020 1080 
S40 
&oR! 880 870 836 840 
680 .70 619 670 
890 
l70 
80 
SDiI 820 806 1260 1340 
670 636 340 160 
lO l3 
Xbo( 
*Restriction fragments sizes sho'An are those obtained from the In 3t11"' analyses 
using the software DIGEST 
146 
University of Ghana http://ugspace.ug.edu.gh
AJt the amplic:ons obl.&ined using the WOLI6S primers were digested wilh Apol. EcoRI. 
HindlH, R.JoI. Nspl . Hinjl and BamHI. None ofthe ampltc:oos was cleaved by the Hindfll 
restriction enL}'me (Fig. 4 .9). NspJ RsQI and Apol , each cut once 10 produce the same two 
fra&meolS in all the amplicons analysed. The restrfClion fl"l!llgmcnlS sizes were estimated 
.. be 800 and 180 bp for Apol (Fig. 4.10), 524 and 418 bp for //sQ1 (Fig. ' . 11) and 585 
andJ3J bpfoc Nspl (Fig.'. II). 
Restriction with EcoRJ , BamHI and Hinjl gave multiple banding patterns. Complex 
patterns. with the sum of the restriction bands being greater than the Si7.e of undigested 
PeR.amplified WOL16S rONAs wen: obtained with the EcoRJ and Hinjl . The £CoRl 
digestion (Fig. 4.12) yielded 3 fragments of sizes 1039. 590 and 436 bp and 4 fragments 
of sizr:s 1039. 736. 590 and 436 bp. The Hinjl digcstion (Fig. 4.13) yielded rragments of 
171 bp, 871 and 693 bpand 800 bp in sizes. For the BamHI restriction. 3 distinct 
reMriction pcltterns with estimated fragment sizes of 746, 663. and 693 bp rcspeetively 
'tII1"'t'Teobsttved (Fig. 4.IJ) 
147 
University of Ghana http://ugspace.ug.edu.gh
1.00b 
O.6kb 
F.4.9 t::lhidium bromide-stainrd 2.0-/0 .garosc SCi electrophorcgram of 1Jina111 
mcrictioa eazymc digesuof WOLI6S rONA PCR producu. lAne M· 100 bp motcc. 
,«iaM nwkt'f (Sigma-Aldrich. USA); lanes I - 4 - digeskd bacterial peR products &om 
diffmnlnkl~_5-undigc:stedPCRproduct. 
148 
University of Ghana http://ugspace.ug.edu.gh
IOObp 
600bp 
400bp 
Fit 4. • ' Ethidium bromKie-stained 2.0% aaarox gel e'"tropbort'gram of Apo( 
__ enzyme digests of WOLl6S rONA PCR products. lane: M. 100 bp mol<cular 
weight nwbr (Siama-AJdrieb. USA); lanes 1-7 z:: bacterill PCR products from different 
mo,quiton di.-cl with Apol; lane 8 .. undiscstC'd PeR product. 
University of Ghana http://ugspace.ug.edu.gh
'II 4.11 "lbidium b<omide· ...i ncd 2.0% apIOS< gel elec:<rophoregram of is.1 and Nspl 
restriction cmymr: diRests of WOL 16S rONA PCR products. Lane M - 100 bp molecular 
weipc owter (Sipa-Aldrich. USA); lanes 1·3 • bacterial PCR products from difTermt 
IDOIQUi\on digested with ,bal; lanes 4-6 - bectcrial PCR products from differenl 
mosquilOes di,ested with NspJ; 1. lane 7 - undigested peR product 
University of Ghana http://ugspace.ug.edu.gh
4.4 Bacterial Sequences in 165 rONA 
Plasmjd,s isolated from 12 clones with insert (Figs4.14& 4.15)wcre partially sequenced. 
Using the VecScreen search. five were found to contain sequences idenlilied 10 be either 
ofvcclororiginorgn~poorqua1ityreadsandwerttbcn:forerejetted. Seven sequences 
with sitts ranging from 810-992 bp (excluding the eubacterial primer sequences used for 
amplification) were obtained (Figs. 4.16 - 4.22). Six of these sequences, designated 
AgAI, AgAl, AgLI. AgL2, AgL3 and AgL4 were obtained from the WOLl6S primer 
PCR products whilst tilt sequence designated AgLS was obtained from the eubacteria 16S 
rONA primer product None of the 235 rDNA PCR products was sequenced 
4.4.1. AgAI (GenBank accession number AY247165) 
The details oftbc AgAl DNA sequence are shown in Fig. 4. 16. Briefly, it is 992 bp with 
molecular weights of 306.18 and 611.05 ltD for the single and double stranded DNA, 
respectively. It is composed of 24.40/. A. 24.5Y. C. 30.0% G and 21.1% T and the 
percenlageGCis54.S%. 
4.4.2 AgA2 (GenBank accession number AY325810) 
lbe details of the AgA2 DNA sequence are shown in Fig. 4. 17. It is 991 bp with 
molecular weights of 306.51 and 610.45 kD for the singJe and double stranded DNA, 
ftSptttivcly. It is composed of 24.80/. A. 24.go;. C, 30.6% G and 19.~/e T and the 
pnccntage GC is 55.5%. 
153 
University of Ghana http://ugspace.ug.edu.gh
1.86kb 
0.93 kb 
0.38kb 
F• ..."  Ethidium bromide-stained 1.00,t. agarosc gel clectrophortgrant showing the 
pt"estflC'eJ absence of inserts (- 1000 bp). 
L.- M· pBU22 DNA BsiN I digest; l.net 1,6, 7, and 8 • dones with insert; lanes 2, 3, 
4IDdS-doneswitboulinsen. 
154 
University of Ghana http://ugspace.ug.edu.gh
M 
1.16kb 
O.J8kb 
Fit: ".15 EdUdium bromide·staincd 0.8% .prose pi ek'Ctrophonegram of isolated 
plasmidDNAJ 
Looe M · pBRJ22 DNA a"NI di ....:  I.-. I ODd 2' pl""';d DNA isolates. 
-----------------~135--
University of Ghana http://ugspace.ug.edu.gh
1 .ilgtcctgqct ,aactccgtgc cage.geege ggt ••t acgg ag999gctag 
51 cgctgttcgg aattact.g99 cgca_.gege .cgtaSSCSg accagaaag t 
101 t9999gtgaa atcceggggc tcaaccccgg .actgectee aaaactcctg 
151 gtcte.g_gtt cgagagaggt gagtggaatt ccgagtgtil9 aggtgaaatt 
201 cgtagatatt cggaggaac. ccagtggcga a99C99'ctca ctggctcgat 
251 actgacgctg aggtgeqaa. gc9t9999a9 caagcaggilt tagataecct 
301 ggtagtCCAe gecgt.gacg atg• •t gee. gtcgtcggaa egt. . taett.g 
351 tcggttgaca cacctaac9g att.aagcatt cegcctgggg agtacggtcg 
401 c._gatt• •• actc ••a agg aatttg.egg ggggcccgca e..agcggtg 
451 g4ge.C9t99 ttta. . ttc. . a agcaacgcgc agaaccttac caacccttg4 
501 catcctgatc geggttagtg aagac_ettt ccttcagttc 99ctggatca 
551 gtgillc.aagtg ctgcatt99c tgtcgtcaac ttcgttgecg ega •••t get 
601 C99gttaagt ccggcaacga agcgcaaccc cacgtcctta gttgec,-tea 
651 t.tcagttggg cactctag99 aaact.gccga tgataagtcg gag9a.S9t9 
701 [.99at9acgt caagtcct.ca t.ggcccttac 999ttgggct acacacgtgc 
751 tacaatggca gtgacaatgg gttaatccct aaaagctgtc tcagttegga 
801 tt99C]gtct.g caactcgacc ccatgaagtc 9g&at.cgcta 9taatcgcgt 
851 heagcatga c9c99tgaat aegttecc99 gcctt.gtaca caccgcccgt 
901 cacaccatgg gaattg9ttc t&cccgaeg& cgctgcgcta accctacggc 
951 a99C49gc99 ccaeggtg99 atca9tcact 999gtgaaag cc 
Fta. 4.16 Dcu.ils of t.cterial DNA ~uencc: AgAl (GenBank acc(!ssion number 
AY247165). 
156 
University of Ghana http://ugspace.ug.edu.gh
1 4gtCCtggct aactccgtgc cagcagccgc ggt.ataegg 8qggggct49 
51 cgttgttcgg a.ttactggg cgtaaagcgc Acgtaqgcgg aac~9a.a9t 
101 cagaggtgag atccca9S9c tc.acctcgg aactgccttt gaaactectg 
lSI ttcttgaggt cgagagaggt gagt.99i1att ceg8gtgt_g a9gtgaaatt 
201 cgtagatatt egg_gqaac. ccagtggcga aggeggee.:. ct99ctcgat 
251 actgacgc:tg aggtgcgaaa gc9t9999i19 caaaca99i1t t.gat.ecct 
)01 ggtagtcc.c gccgt• •a eg atg •• tgcc. gtegtcg999 ga.gcatget 
lSl attcggtgac dc.cetaa.:g gatta_gcat. tcccgc:ctgg ggggtacgcc: 
401 cqc •••g gtt ••••e tcaaa ggaattgacg ggggcccgca c ••g c99 t gg 
451 agcatgtggt tcaattagaa gea"cgcgc. ga.ecttacc ••c cettaac 
501 at99cagtga ccgttccaga gatggtcctt tctcgcaaag agacacttgc 
551 acae• •g tqe tgcatggctt gtcgtcaact tegtgtegtg _gatgttteg 
601 ggttaagtcc ggc .... cg.gc gccaacccca cgccttcagt tgccagcatt 
651 cagttgggca ctctgaagga actgccggtg at .. agccgga ggaag9tgt9 
701 gatgacgtca agtcctcatg gcccttacgg gttgggctac acacgtgcta 
751 c ... tggtggt gacaatgggg taatcccaaa aagccatctc agttcggatt 
801 gtcgtctgca actcggcg9c atgaagtcg9 aatcgctagt aatcgcgtaa 
851 cagcatgacg cggtgaa.tac gttcccgggc cttgtacaca ccgcccgtca 
901 caccatggga attggatcca cccga.ggcg gtgcgcca .. c cagcaatgga 
'51 ggcagccgac cac9gtggga tcagtcOllctg 999t90l.a.gc c 
F~ 4. 17 Details of baclcrial DNA sequence AgAl (GenBank accession number 
AYJ25110) 
IS7 
University of Ghana http://ugspace.ug.edu.gh
1 agtcct4gct aacteegtgc cagcagccgc ggtaatacgg .599999'cta9 
51 cgttgttcgg aattactggg cgtaaagcgc &C9t.99c99 at.c4gaaagt 
101 eag.99tgaa atcccagggc tcaaccttgg aactgccttt gaaactcctg 
151 atctt.gaggt <:949a9a99t 9a9t994.tt ccgagt.gt.ag aggtgaaatt 
201 cgt.agatatt c991199."c. cCIIgtggcga aggcggctca etggctegat 
251 act.gaegct.g .99t9c94• • 9c9t9999a9 caaacaggat tagat_ecct 
JOI 9gtagtccac 9c<:9t ••4 C9 atgaatgcca gtcgt.cgggt agcatgctat. 
)SI tcggtgacac acct.4e994 ttaggcatcc <:9cct.9999- gtacgcecgc 
401 .aggttaaa. etc8a499". tt9ac99999 gcccgcacaa gC99t99agc 
451 atgtggttta attag4agell acgcgcagaa ccttaccaac ccttgacatg 
501 9<:.9t94C<:9 tcceagagat ggt.cctt.tct cgeaagaa.c acttgcacac 
551 aagtgcttgc at9gct9tcg tcagctcgtg tC9t949.t9 ttcggt.taaa 
601 tccggcaacg aacgcaaccc acgccttcIIg ttgccagcat tcagttgggc 
651 actctgaa.gg aactgccggt gataagcc99 a991la99tgt ggatgacgtc 
701 aagtcctcat ggcccttacg ggttgggcta cacacgtgct olcaatg9t99 
751 tgacaatg9g gta.tcccaa aa.agccatct cagttcggat tgtcgtctgc 
801 aactcggcgg c.tgaagtcg gaa,tcgcta,g ta.tcgC9ta olc.gcatgac 
851 gcggtgaata cgttccc999 ccttgtacac accgcccgtc acaccatggg 
901 aattggatcc acccgaaggc ggtgcgccag ccagcaatgg aggcagccga 
951 ccaC99t999 atcagtcact ggggtgaa.ag cc 
FiJ. 4.18 Details of bacterial DNA sequence AgLI tGl.'nOank accession nwnber 
AY247160) 
158 
University of Ghana http://ugspace.ug.edu.gh
1 agec::ctggct aactccgtgc cage.geege ggtaatacgg ll99999cta9 
51 cgtt9ttc99 ••t tact.gg9 cgtaaagcgc aegtag9cgg atcagaaagt 
101 cagaggtg.u. atcccagggc tea.ccttg9 aactgcc:tet gaaactcctg 
151 atettgaggt cgagaga99t gagtggaatt ccgagtgtag aggtgaaatt 
201 cgtagatatt cggaggaaca ccagtggcga aggc99ctca ct99ctcgat 
251 actgacgctg aggtgcgaaa 9C9tggggag ca ••c aggat tagataccct 
301 ggtagtcc:ac: gccgta• •c g atg.atgec. 9tcgtc999t agc,;r,tgctat. 
351 tC9gt.gac.c accta.egga ttaagcatcc cgcct9999. gtac99cc9c 
401 ••g get •••• ctc• •a gg• • ttgaC99999 cccgcacailla gcggtggagc 
451 atgtggttta att_gllages acgcgcaga. ccttacc• •c  cettg_catg 
501 gcagtgac:cc gttccagaga tgg-tectttc tege._gaga cacttgcaca 
551 caagt.gctgc atggctgt.cg tcaactcgtg tegtg_gatg tttC99ttaa 
'01 gt.t.ccggcaa cgaaccgcaa ceeacgcctt. caattgceilg eattcilgttg 
651 9gcactctga aggaact9cc 99tg",t"'8ogc cggagg ••g g t9t.99ol1tgac 
101 gtc",,,,gtcct c",t.ggccctt acgggtt99'9' ctacilC,J,cgt gctaca.tgg 
151 t99tgacaat ggggt.atee ca.a.agee. tetc.gttcg gattgtegtc 
801 tgeaactcgg cggeatgaag t.cqg.atcge tagt44tege gtaaeageat 
851 gacgcggtga atacg:ttcee gggeettgta eacaecgccc gtcaeaccat 
gOl gggaattgg. t.ccaccegaa GGC99t.gcge caaccagcaa t.ggaggcagc 
951 cgaccacggt gggat.co1.gt.e aetggg9t9a aagec 
Fie- ".19 Details of baeteNl DNA stqurnce Ag12 (GenBank accession number 
AY247161j. 
University of Ghana http://ugspace.ug.edu.gh
1 ggctt.t.cacc ccagtgact.g atcccOleegt. ggtcggctge etccattget 
51 ggt.tggegea ccgcettcgg gtggatccaa ttcccatggt gtgac9ggc9 
1(11 gtgtgtacaa ggccc999aa cgtattcilcc gcgtcatgct gttacgC9at 
151 taet.agcgat tccgacttca t gccgcegag ttgcagaC9a caatccgaac 
201 tgagat9gct ttttgilgatt aceccattgt caccaeeatt gtagcaegtg 
251 tgtagceeail ceegtaaggg ccatgaggae ttgaegteat ceacacctte 
301 etacggctta teaecggeag ttcettcagil gt,gcccaact gaatgetgge 
351 aactgaaggc gt9ggt.t9C9 ctcgttgecg gacttaaccg aaaateteac 
401 aaeacgaget gacgacagec atgeageaec tgtgtgeaag tgtetcttgc 
451 gilagaaagga ecatctetgg aacggteaet gccatgtcga 999ttg9taa 
501 aggttetgcg ecgttgettc gaattaAilec acatgeteca cccgcttgtg 
551 c9g9'9ceece cgteaattcc tttgagtttt aacettgegc gecgtaetec 
601 cceaagcggg aatgetttaa tttccgttag ggtggtgtea acccgaaat. 
651 fJ,eatgetaee cgacgaetgg catteatcgt ttaeggcgtg ggctaccagg 
101 gtatetaate et.gtttgctc eceacgcttt cgeacetcag cgteagtate 
751 gagecagtg. gcegeettcg eeaet.ggtgt teeteegaat atetMgaat 
801 ttcaectetil caeteg 
Fig. 4.20 Delails of baclerial DNA sequence Agl3 (GenBank accession number 
AY247162). 
160 
University of Ghana http://ugspace.ug.edu.gh
1 99ctt.t.cacc ccagt-gact.q atcccaccgg ggtcggct.gc ctceattgct 
S1 99tt99c9C. ccgccttC99 gtggatccaa ttcccat.qgt. 9 t 9&0999C9 
101 gtgtgtacaa 99cCC999aa cgtattcacc gcgtcatgct gttacgcgat 
151 tactagcgat tccgaettca tgccgccgag ttgcagacga caattcgaac 
:201 tgagatggct ttttgggatt aecccattgt tacc.ccatt gtateacgtg 
251 t9t99cccaa c(:cgtaa999 ccatgaggac ttgacgt.cat ccaeaccttc 
301 ctccggctta tcaccggcag ttccttcana gtgcccaact gaatgctg9C 
351 iIIact.ga.agc 9t999tt9c9 etcgttgccg gaattaacC'g .acatctcaa 
fDl gacaaaaget gacgacagcc aegcage.cc tgtgtgcagt gtctcttgeg 
451 aguaggacc atctct99aa C9gt.cact.gc catgecaag9 gttggt• •a g 
501 gttctgcgcc tttgctttcg aatta• •c e. catgetccaa ccgnttgtgc 
551 cggcccccgt caaattcctt t.gagttttaa cctttgcggc cggtacttee 
'01 cca99c99ga atgcttaatt ccgttaaggt gtgtcaaccg aatagcatgc 
'51 tacccgacga ct9gcattca tcgtttacgg cgtggactac cagggtatct 
101 aatcctgttt gctccccacg ctttcgcacc tcagcgtcag tatcgagcca 
151 gtgggccgcc ttcgccactg gtgttcctcc gaatatctan gaatttcacc 
801 tctacactcg 
Fi&- 4.21 Details of bacterial DNA sequence AgL4 (GcnBank accession number 
AY24116J). 
---------------,,1~1 --
University of Ghana http://ugspace.ug.edu.gh
1 ggctntacgg ctacctntgt eacgngactt agccncncan attacctgtt 
51 etOlecctaat ccncgttctt gacgcggaac tggcttaggg tetaccagac 
101 tttcatggct caggatgaac gcgtagcg9c tag9cctaat acaagtgtcc 
151 csgaaC99ta tgaaccggge gcat• •g gca atgacacact. geege.ctag 
201 gtgC'gtaatt cdc.getatg catacgctac ctgagttgct gatecceg• • 
251 tcagaacttt 9gagnnaacg gag.tttega gataccgcat ccagtaagac 
)01 4gtqn.atcat ggctaccctc atgtaccgcc catgtgtaa.c agC9tgt9tc 
351 ggcccata99 cgatagagtg gcgcatgagt cc• •c ttat.g ttaegtctgg 
401 tccaegtaac ggcttcactc tca.gtacga tgattgc99t 49ge_ggett 
451 gatctgagag gat• •g tccc ccac4ctggc aectgtagat acg99cca9& 
501 ctuc• •c ag ggagttgeag cacgtagcgg .acattaacc caacacctcg 
551 acggollcgagt ctgaccacag ccatgcagca cgcaggatga aagt.cgtcat. 
1i0l gcagact.uC ctauctgct aaattattcc ttcacattct agcctaagaa 
lSI aaagttcctt g09a999a •• gtcaccg.ec t.taaccaca t.actccaccc 
701 gcttactcct gccaccilccc gtcggtaatt. cctttgaagg tgagaccgt.t 
751 ggc09aeggt acccccctca aatggtgcgt &gacggttaa ccgattagtg 
801 gtgaaaacgn cgtcacgatt nilgtgcattg anctgtng 
lia. 4.22 [)ecajis orbclcteriaJ DNA sequence AgLS (GenBank accession number 
AY247J~).