University of Ghana http://ugspace.ug.edu.gh PREVALENCE OF BLOOD GROUP A2 AMONG GROUP A AND AB DONORS WHO VISIT THE SOUTHERN AREA BLOOD CENTRE BY EVANS AGYAPONG OWUSU (ID: 10598294) THIS THESIS/DISSERTATION IS SUBMITTED TO THE UNIVERSITY OF GHANA, LEGON IN PARTIAL FULFILLMENT OF THE REQUIREMENT FOR THE AWARD OF MPHIL HAEMATOLOGY DEGREE JULY, 2019 University of Ghana http://ugspace.ug.edu.gh DECLARATION I, Evans Agyapong Owusu of the Haematology Department, School of Biomedical and Allied Health Sciences, University of Ghana, do hereby declare that this thesis was duly carried out by me and the results obtained therein are the true reflection of the work and supervised by the supervisors below. Student: Signature: ______________________________ Date: ____________ Evans Agyapong Owusu Supervisors: Signature: ______________________ Date: __________ Professor Joseph K. Acquaye [FWACP (LAB. MED.)] Signature: ______________________________ Date: ___________ Dr. Yvonne Dei-Adomakoh (MB BS, FWACP) II University of Ghana http://ugspace.ug.edu.gh DEDICATION I dedicate this work to God Almighty for giving me the strength throughout the period. This work is also dedicated to my wife Franklina Asenso and two kids (Jason and Anaiah) for their support. III University of Ghana http://ugspace.ug.edu.gh ACKNOWLEDGEMENT I wish to express my sincere gratitude to my supervisors Prof. J. K. Acquaye and Dr. Yvonne Dei-Adomakoh for their supervisory role with this work. I would like to acknowledge the entire management and staff of National Blood Service, Ghana for giving me the opportunity to carry out this work at the various study sites. Thanks also go to the management and staff of the Accra Area Blood Centre for making me carry out the work at their various departments. The entire laboratory staff of the Southern Area Blood Centre are also appreciated for helping me in the practical aspects. Enormous thanks also go to Dr. Benneh and the entire staff of the Haematology department, University of Ghana for their support and advice. IV University of Ghana http://ugspace.ug.edu.gh TABLE OF CONTENTS DECLARATION .................................................................................................................................... II DEDICATION ....................................................................................................................................... III ACKNOWLEDGEMENT ..................................................................................................................... IV TABLE OF CONTENTS ........................................................................................................................ V LIST OF FIGURES ............................................................................................................................. VIII LIST OF TABLES ................................................................................................................................. IX LIST OF ABBREVIATIONS ................................................................................................................. X ABSTRACT .......................................................................................................................................... XII CHAPTER ONE ..................................................................................................................................... 1 1.0 INTRODUCTION ............................................................................................................................ 1 1.1 BACKGROUND .......................................................................................................................... 1 1.2 PROBLEM STATEMENT ........................................................................................................... 3 1.3 JUSTIFICATION ......................................................................................................................... 4 1.4 AIM ............................................................................................................................................... 5 1.5 SPECIFIC OBJECTIVES ............................................................................................................. 5 CHAPTER TWO .................................................................................................................................... 6 2.0 LITERATURE REVIEW ................................................................................................................. 6 2.1 HUMAN BLOOD GROUPS ........................................................................................................ 6 2.1.1 Historical Perspective ............................................................................................................ 6 2.2 CARBOHYDRATE BLOOD GROUPS ...................................................................................... 8 2.2.1 Secretion of ABO Antigens ................................................................................................... 8 2.2.2 The H gene (FUT1) .............................................................................................................. 10 2.2.3 Secretor gene (FUT2)........................................................................................................... 11 2.2.4 Lewis gene (FUT3) .............................................................................................................. 12 2.2.5 ABO gene ............................................................................................................................. 13 2.3 STRUCTURE OF ABO, LEWIS AND H ANTIGENS ............................................................. 19 2.3.1 Glycoconjugates that express ABH antigens ....................................................................... 19 2.3.2 Carbohydrate determinants of ABH and Lewis antigens ..................................................... 20 2.4 BIOSYNTHESIS OF ABH ANTIGENS .................................................................................... 25 2.4.1 Biosynthesis of H antigen .................................................................................................... 25 2.4.2 Biosynthesis of antigens A and B ........................................................................................ 26 2.4.3 A and B transferases ............................................................................................................ 29 V University of Ghana http://ugspace.ug.edu.gh 2.4.4 H and Secretor transferases .................................................................................................. 31 2.5 INHERITANCE OF ABO BLOOD GROUPS ........................................................................... 33 2.6 BLOOD GROUP A .................................................................................................................... 35 2.7 ABO SUBGROUPS .................................................................................................................... 36 2.7.1 Major subgroups of A .......................................................................................................... 37 2.7.2 Blood Group A2 ................................................................................................................... 38 2.7.3 Characteristics of A1 and A2 Transferases ........................................................................... 41 2.7.4 A1 and A2 Determinants ....................................................................................................... 43 2.7.5 Other weaker subgroups of A .............................................................................................. 45 2.7.5.1 A3 ...................................................................................................................................... 45 2.7.5.2 Ax ...................................................................................................................................... 46 2.7.5.3 Am ...................................................................................................................................... 47 2.7.5.4 Ay ...................................................................................................................................... 47 2.7.5.5 Aend .................................................................................................................................... 49 2.7.6 Interactions between A and B genes .................................................................................... 49 2.7.6.1 Gene Suppression .............................................................................................................. 49 2.7.6.2 Allelic Enhancement ......................................................................................................... 50 2.8 H deficient phenotypes ............................................................................................................... 51 2.9 ABH ANTIBODIES ................................................................................................................... 53 2.9.1 Clinical Significance of ABH Antibodies ................................................................................ 54 2.9.1.1 HDFN ................................................................................................................................ 54 2.9.1.2 Transfusion ....................................................................................................................... 55 2.9.1.3 Transplantation ................................................................................................................. 55 2.10 METHODS OF TESTING FOR ABO BLOOD GROUP ........................................................ 56 2.10.1 Slide Method ...................................................................................................................... 56 2.10.2 Standard Tube method ....................................................................................................... 57 2.10.2.1 Forward grouping ........................................................................................................... 57 2.10.2.2 Reverse grouping ............................................................................................................ 58 2.10.3 Microplate method ............................................................................................................. 58 2.10.4 Gel Microcolumn ............................................................................................................... 59 2.10.5 Molecular determination of blood groups using PCR ........................................................ 59 2.10.6 Other methods .................................................................................................................... 61 CHAPTER THREE .............................................................................................................................. 62 3.0 METHODS ..................................................................................................................................... 62 3.1 STUDY DESIGN ........................................................................................................................ 62 3.2 STUDY SITE DESCRIPTION ................................................................................................... 62 VI University of Ghana http://ugspace.ug.edu.gh 3.3 STUDY POPULATION ............................................................................................................. 62 3.4 SAMPLE SIZE ........................................................................................................................... 63 3.5 INCLUSION CRITERIA ............................................................................................................ 63 3.6 SAMPLE AND DATA COLLECTION ..................................................................................... 64 3.7 LABORATORY ANALYSIS..................................................................................................... 65 3.7.1 ABO BLOOD GROUPING (STANDARD TUBE METHOD) .............................................. 66 3.7.1.1 Forward grouping .............................................................................................................. 66 3.7.1.2 Reverse grouping .............................................................................................................. 67 3.7.2 TESTING GROUP A CELLS FOR ABO SUBGROUP A2 .................................................... 68 3.7.2.1 Forward grouping ............................................................................................................. 68 3.7.3 Testing A2 and A2B serum for anti-A1 ..................................................................................... 69 3.7.4 Anti-A1 titre determination (Doubling dilution) ...................................................................... 69 3.8 DATA MANAGEMENT ............................................................................................................ 70 3.9 DATA ANALYSIS ..................................................................................................................... 70 3.10 ETHICAL ISSUES ................................................................................................................... 71 CHAPTER 4 ..................................................................................................................................... 72 4.0 RESULTS ................................................................................................................................... 72 4.1 Demographic Characteristics .................................................................................................. 72 4.2 ABO blood grouping results ................................................................................................... 74 4.3 Prevalence of ABO subgroups A2 and A2B among the 1140 donors ...................................... 75 4.7 Prevalence of anti-A1 among study participants ..................................................................... 79 CHAPTER FIVE .............................................................................................................................. 80 5.0 DISCUSSION ............................................................................................................................. 80 5.1 LIMITATIONS OF THE STUDY .............................................................................................. 84 5.2 CONCLUSION ........................................................................................................................... 84 APPENDICES .................................................................................................................................. 97 APPENDIX 1: Ethical approval from National Blood Service, Ghana ................................................ 97 APPENDIX 2: Ethical approval from College of Health Sciences, University of Ghana .................... 98 APPENDIX 3: participant informed consent form ............................................................................... 99 APPENDIX 4: Data collection form ................................................................................................... 101 APPENDIX 5: Washing of red blood cells ......................................................................................... 102 APPENDIX 6: Preparation of 3% red cell suspension ....................................................................... 103 APPENDIX 7: REAGENTS USED ................................................................................................... 104 VII University of Ghana http://ugspace.ug.edu.gh LIST OF FIGURES Figure 1: Organization of the ABO gene ................................................................................. 14 Figure 2: Schematic presentation of the Open Reading Frame (ORF) of five common ABO alleles ....................................................................................................................................... 16 Figure 3: Structures of the immunodominant sugars of antigens H, A and B antigens ........... 22 Figure 4: Biosynthesis of ABH antigens ................................................................................. 28 Figure 5: Structure of α-1,3-d-galactosyltransferase (GTB) showing the two domains separated by a central cleft, UDP - Gal and H antigen ............................................................ 30 Figure 6: Inheritance of ABO antigens .................................................................................... 34 Figure 7: cDNA (black line) and protein products (coloured box) of ABO alleles, showing how A2 (A201) differs from A1 (A101). .................................................................................. 39 Figure 8: ABO blood groups of the 1140 blood donors .......................................................... 74 Figure 9: Prevalence of ABO subgroups A2 and A2B among the 1140 blood donors ............. 75 Figure 10: Proportion of group A individuals with antigen A2 ............................................... 76 VIII University of Ghana http://ugspace.ug.edu.gh LIST OF TABLES Table 1: Basic concept of ABO blood group system ................................................................. 7 Table 2: ABO antigens and their gene products ...................................................................... 21 Table 3: Some characteristics of A1- and A2-transferases ....................................................... 42 Table 4: Number of antigen sites on red cells of various ABO groups .................................. 44 Table 5: Characteristics of some weak ABO Phenotypes ....................................................... 48 Table 6: Grading of agglutination ............................................................................................ 57 Table 7: ABO forward blood grouping .................................................................................... 66 Table 8: Reaction for A1 and A2 cells ...................................................................................... 68 Table 9: Description of Demographic characteristics of blood donors .................................. 73 Table 10: Agglutination reactions observed in the various samples........................................ 77 Table 11: Difference in antigen A2 among blood group A and AB donors ............................. 78 IX University of Ghana http://ugspace.ug.edu.gh LIST OF ABBREVIATIONS AIHA: auto-immune haemolytic anaemia Anti-A: antibody A Anti-B: antibody B Anti-D: antibody D Anti-H: antibody H Asp: Aspargine DIC: Disseminated Intravascular Haemolysis EDTA: Ethylene diamine tetraacetic acid Fuc: Fucose FUT1: Fucosyltransferase 1 FUT2: Fucosyltransferase 2 Gal: Galactose GalNAc: N-acetyl galactossamine GIT: Gastro intestinal tract GlcNAc: N-acetylglucossamine Gly: Glycine GTA: Glycosyltransferase A GTB: Glycosyltransferase B HDFN: haemolytic disease of the foetus and newborn HTR: Haemolytic transfusion reaction IgA: Immunoglobulin A IgG: Immunoglobulin G IgM: Immunoglobulin M Km: Michaelis constant Le: Lewis antigen X University of Ghana http://ugspace.ug.edu.gh Met: Methionine Mn2+: Manganese ion mRNA: Messenger RNA ORF: Open Reading Frame RBC: Red blood cell Rh: Rhesus SABC: Southern Area Blood Centre Se: Secretor gene Ser: Serine SeW: weak secretor gene UDP: Uridine diphosphate Val: Valine α2FucT1 α2FucT2 %: Percent ᵒC: Degree Celsius XI University of Ghana http://ugspace.ug.edu.gh ABSTRACT Introduction: The ABO blood group antigens are major antigens whose discovery paved way for the performance of safe blood transfusions and transplantations. Subgroups of the ABO antigens were discovered few years after ABO antigens discovery. A1 and A2 are the two principal subgroups of A, both of which are produced by different enzymes. Anti-A1 antibody appears as an atypical cold agglutinin in the sera of some individuals with A2 or A2B who lack the corresponding antigen. One to eight percent of group A2 individuals and 22-35% of group A2B individuals develop anti-A1 in their serum which can cause discrepancies in ABO blood typing leading to haemolytic transfusion reactions. Knowing the proportion and distribution of blood group A2 will be helpful in preventing Haemolytic Disease of the Newborn (HDN) and also help in safe blood transfusion. Knowledge of antigen A2 frequency will also improve the profile of ABO blood groups among Ghanaians. Aim: The aim of this study was to determine the prevalence of antigen A2 among group A and AB blood donors visiting the Southern Area Blood Centre (SABC), Accra. Methods: This study was carried out at all the sites operated by the SABC. It was a cross- sectional study that included all blood donors who passed the pre-donation screening. Approximately 2mls of their donated blood was used for the study. Using the tube method, forward and reverse ABO blood grouping was done with their cells and serum using anti-A, anti-B monoclonal reagents. Blood group A and AB were selected and used against Anti-A1 reagent (Dolichos lectin) for blood group A2 screening. Commercially prepared A1 and A2 red cells were used against the serum of the group A samples which served as controls. Data obtained was analyzed using SPSS version 22. Results: A total of one thousand, one hundred and forty (1,140) participants (all Ghanaians) who were made up of 942 (82.6%) males and 198 (17.4%) females were screened for the study. The ABO blood group distribution observed was; O(51.3%), A1(16.8%), A2(4.4%), B(23.5%), A1B(1.7%), A2B (2.2%). The prevalence of antigen A2 among the study participants was 6.6%. XII University of Ghana http://ugspace.ug.edu.gh Among participants with antigen A (groups A and AB individuals), 74% were A1 whiles 26% were A2. Among group A individuals, 20.66% were A2 whiles 79.34% were A1. The prevalence of antigen A2 among group A2B individuals was 55%. Only one individual with blood group A2 had developed anti-A1 antibody. Conclusion: The prevalence of blood group A2 among group A and AB individuals in the Ghanaian population is similar to that obtained from studies in several West African countries. However, unlike many African countries, antigen A2 is the most prevalent A antigen among AB individuals in Ghana. Anti-A1, which causes discrepancies in blood typing leading to transfusion reactions in group A2 and A2B individuals, was hardly seen among Ghanaians with antigen A2 hence screening for anti-A1 may not be necessary among Ghanaian donors. XIII University of Ghana http://ugspace.ug.edu.gh CHAPTER ONE 1.0 INTRODUCTION 1.1 BACKGROUND The ABO blood group antigens are clinically significant and their discovery by Landsteiner in 1900 paved way for the performance of safe blood transfusions. These antigens are characterized by their distinct carbohydrate structures attached to the glycoprotein or glycolipid membranes of red blood cells (RBCs). Their expression can also be detected on various human cell and tissue surfaces including sensory neurons, vascular endothelium and platelets (Szulman, 1971). Due to their wide expression on various human tissues and body fluids, ABO antigens have also been named histo-blood group antigens. Landsteiner’s discovery of the ABO system in 1900 was based on the detection of antigen A or B on red blood cells (RBCs) and their Mendelian inheritance pattern was discovered by Bernstein in 1924. Subgroups of the ABO antigens occur and are characterized by reduced quantities of the antigens on the RBCs and saliva of secretors (R. Race & Sanger, 1975). Variation in antigen- A expression was seen a few years after the discovery of ABO antigens (von Dungern & Hirszfeld, 1911). Landsteiner & Levine (1930) subdivided blood group A into their principal subgroups, A1 and A2. The presence of A subgroups is due to genetic variations in the coding region which leads to diminished amounts of A substance and higher amounts of H substance on red blood cells (Geoff Daniels, 2013). There are weaker subgroups than A2 which do not frequently occur but when present, they show a characteristic decreased A antigen sites on their RBC and a corresponding increased H antigen activity (Economidou, Hughes‐Jones, & Gardner, 1967). The difference between subgroups A1 and A2 is seen in their reactivity with the Dolichos biflorus lectin. The Dolichos lectin specifically agglutinates A1 cells, leaving the A2 cells unagglutinated. 1 University of Ghana http://ugspace.ug.edu.gh The ABO antibodies occur naturally in the sera of individuals who do not have the corresponding antigens and can cause haemolysis in-vivo (G. F. Springer, Horton, & Forbes, 1959). These antibodies develop after 4–6 months of age, reaching peak levels by 5–10 years and declines in the aged (Cooling, 2016). They are mostly immunoglobulin M (IgM). These antibodies through complement activation can cause severe, life-threatening transfusion reactions due to incompatible ABO blood transfusions. Anti-A1 antibody appears as an atypical cold agglutinin in the sera of few people with group A2 or A2B who do not have the corresponding antigen (Chaudhari, Misra, & Nagpal, 2008). One to eight percent of persons with blood group A2 and about 22-35% of persons with group A2B have serum anti-A1 (Cooling, 2014). These antibodies may cause discrepancies in ABO blood grouping which can further result in HTR (Cooling, 2014). Rare cases of haemolytic transfusion reaction due to anti- A1 have been reported (Chaudhari et al., 2008). The ABO gene has three alleles which occur differently among different groups of people and has resulted in variation in the frequency of the blood groups across the world (Bernstein, 1924). Allele frequency variations in the ABO gene reflects the social tendency of populations to marry, reproduce and migrate. The principal A subgroups show great variation among different populations. It is estimated that about 80% of individuals with antigen A are A1 while the remaining 20% are A2 (Cooling, 2014). In Africa, a study among Sudanese (both Arabs and Negros) revealed 8.3% of group AB individuals were A2B whiles 6.58% of group A individuals were A2 (Ahmed M. Elnour et al., 2015). A similar study among Liberians also revealed that 19.7% of blood group A individuals were A2 whiles 32.6% of AB individuals were A2B (Livingstone, Gershowitz, Neel, Zuelzer, & Solomon, 1960). Among Indians, the frequency of subgroup A2 among group A individuals is 6.58% whiles that of A2B among group AB is 8.99% (Shastry & Bhat, 2010). Studies however, show that the proportion of blood 2 University of Ghana http://ugspace.ug.edu.gh group A2B is significantly higher in blacks and Japanese although the inheritance of the A2 gene follows the Mendelian pattern (Ogasawara et al., 1998). Besides their importance in evolution, blood group prevalence studies have become very necessary due to their relation to disease susceptibility (Khan, Khaliq, Bakhsh, Akhtar, & Amin ud Din, 2009). Knowing the frequency of ABO blood group antigens at various levels is very important in effective blood bank management (Atire, 2015). Knowledge of the frequency and distribution of blood subgroup A2 will help to enhance blood compatibility testing, population genetic studies, investigation of transfusion reactions, studying disease associations and resolving some medico-legal issues (Tesfaye, Petros, & Andargie, 2015). It is therefore very necessary to study blood group distribution in any population in order to get information (Atire, 2015). 1.2 PROBLEM STATEMENT The serum of individuals with blood groups A2 and A2B may contain anti-A1 antibodies which appear as atypical cold agglutinins (Shastry & Bhat, 2010). Studies on blood groups have shown that approximately 1 to 8% of blood group A2 and close to 35% of blood group A2B individuals develop serum anti-A1 (Cooling, 2014), which becomes clinically significant by reacting at 37oC and destroy A1 red cells (Chaudhari et al., 2008). The presence of anti-A1 in serum can interfere with blood typing and may lead to wrong blood grouping (Bangera, Fernandes, & Swethadri, 2007). Although there are several reports on HTRs due to anti-A1 in group A2 and A2B individuals, the percentage of group A2 and A2B individuals affected has not been cited. Pregnant mothers who have blood group A2 can also develop hyper-immune IgG anti-A which may cross the placenta and lead to haemolytic disease of the foetus and newborn 3 University of Ghana http://ugspace.ug.edu.gh (HDFN) (Regan, 2016). Subgroup A2 is however, known to be significantly higher in blacks (Ogasawara et al., 1998), putting some Ghanaians at a higher risk of transfusion reaction associated with antigen A2 and hence necessitating this study. 1.3 JUSTIFICATION A lot of work has been done on blood groups in Ghana. However, no published information on the proportion of subgroup A2 has been cited in the country. Routine blood grouping tests done in Ghanaian blood banks and laboratories do not include blood group A2 screening. From personal observations, I have realized that most of the reports on HTR among A2 individuals with anti-A1 only come from areas where A2 screening is routinely done. This means HTR’s due anti-A1 among group A2 individuals may be occurring at our blind side in Ghana because we do not routinely screen for blood group A2. This study seeks to reveal the baseline proportion of blood group A2 among Ghanaian donors and the scope of the possible problems associated with transfusion of blood group A2 and its clinical implications. Knowing the frequency of blood group A2 will inform whether or not to add its screening to the routine blood grouping tests. This may help reduce the possible blood transfusion reaction which may be caused by anti-A1 among group A2 individuals, hence helping in safe blood transfusion. Blood groups are also used for anthropological studies. Knowledge of the frequency of A2 antigen in Ghanaians will also improve the profile of ABO in the Ghanaian population. 4 University of Ghana http://ugspace.ug.edu.gh 1.4 AIM The aim of the study is to determine the frequency of blood group A2 antigen among group A and AB blood donors visiting the Southern Area Blood Centre (SABC). 1.5 SPECIFIC OBJECTIVES 1. To determine the prevalence of antigen A2 among group A and AB donors who visit the Southern Area Blood Centre. 2. To determine the difference in frequency of blood group A2 between groups A and AB individuals. 3. To determine the frequency of production of anti-A1 by individuals who possess the A2 antigen i.e. groups A2 and A2B. 5 University of Ghana http://ugspace.ug.edu.gh CHAPTER TWO 2.0 LITERATURE REVIEW 2.1 HUMAN BLOOD GROUPS 2.1.1 Historical Perspective Apart from being discovered first, the ABO system is the most essential antigenic system to blood transfusion and organ transplantation (Ravn & Dabelsteen, 2000). Landsteiner (1900) first described the ABO system and later reported in 1901. After testing red blood cells and sera of his associates, Landsteiner identified three different patterns of reactivity with human red cells, which he termed A, B and C (later named O) (Karl Landsteiner, 1900). He observed that group A red cells have antigen A but their serum have naturally occurring antibody B. Group B erythrocytes were seen to have antigen B but, in their serum, antibody A which also occurs naturally was detected. No antigens were detected on red cells from blood group O individuals although both antibody A and B were present in their serum (table 1) (Karl Landsteiner, 1900). In 1902, Von Decastello and Sturli, through a series of experiments, discovered blood group AB which is the rarest ABO blood group (Decastello, 1902). Antigens A and B are both found on the red cells of group AB individuals but their serum lack anti-A and anti-B (Decastello, 1902). After ABO antigens discovery by Landsteiner, Epstein and Ottenberg (1908) suggested the probable inheritance of ABO antigens which was later confirmed by Von Dungern and Hirszfeld in 1910 after investigating 72 families (Dungern, 1910). They hypothesized that two independent alleles produce antigens A and B, a claim which was later debunked by Bernstein in 1924. Bernstein revealed the three alleles at the ABO locus; and this was later called “one gene three-locus model’’ (Bernstein, 1924). Since then, various advanced methods have been used to study the ABO system. 6 University of Ghana http://ugspace.ug.edu.gh Several other blood groups were discovered after the ABO discovery by Landsteiner. Three hundred and thirty-nine blood group antigens have been discovered, 297 of which fall under one of the thirty-three existing blood group systems (Geoff Daniels, 2013). Each system is made up of genetically unique collection of antigens which are controlled by one gene or a cluster of homologous genes which are closely linked. Biochemically, blood group antigens fall under two main umbrellas; 1. protein determinants (primary products of blood group systems) 2. carbohydrate determinants on glycoproteins and glycolipids (Geoff Daniels, 2013). Table 1: Basic concept of ABO blood group system ABO blood group Antigens on RBCs Antibodies in serum A A Anti-B B B Anti-A AB AB No antibody O No antigen Anti-A & Anti-B 7 University of Ghana http://ugspace.ug.edu.gh 2.2 CARBOHYDRATE BLOOD GROUPS ABO, Lewis, H, I and P blood group antigens are characterized by their carbohydrate epitopes which appear as post-translational modifications on glycoproteins, mucins and glycolipids (Cooling, 2016). They have similar complex antigen biosynthesis, immune responses and in- vivo outcome after incompatible transfusion. Their complex biosynthesis only differs by the distinct enzymes (glycosyltransferases) that catalyzes their various glycosidic linkage formations. ABH antigens are capable of stimulating B lymphocytes to produce IgM antibodies without requiring any assistance from helper T lymphocytes. Without any previous transfusion, individuals produce naturally occurring serum IgM antibodies to the carbohydrate blood group antigens lacking on their RBCs. The IgM antibodies produced by the carbohydrate blood groups can cause the agglutination of antigen-positive human RBCs even in the absence of antiglobulin reagent. In-vitro reactions by the naturally occurring ABH antibodies characteristically improves below 37oC temperature. IgM molecules can immediately cause intravascular haemolysis after incompatible transfusion of red cells because of their ability to fix complement. 2.2.1 Secretion of ABO Antigens The presence of ABH antigens are not limited to red cell surfaces but can also be found in soluble forms in body secretions (Yamakami, 1926). Some A, B and AB individuals named non-secretors do not secrete antigens A and B in the body fluids. The secretor gene (FUT2) controls ABH antigen secretion and is located on chromosome 19. The secretor gene (Se) is inherited in the Mendelian manner on a single locus and the ability to secrete ABH antigens 8 University of Ghana http://ugspace.ug.edu.gh does not depend on an individual’s ABO blood group (Schiff & Sasaki, 1932). Close to 80% of Europeans are secretors whiles 20% are non-secretors (Sambo, Yerima, Amaza, Bukar, & Gali, 2016. The frequency of the secretor gene among people of African descent is similar to that of Europeans and the secretor gene is dominant (Sambo, Yerima, Amaza, Bukar, & Gali, 2016). Although one’s ability to secrete is independent on his/her ABO blood group, ABO antigens are absent in the secretions of non-secretors due to their inability to secrete antigen H (which is the substrate for antigen A and B production) in their body fluids. Studies show that the presence of the secretor gene is needed alongside either A or B gene in the same cell in order for A or B to be expressed in body fluids (Knowles, Bai, Daniels, & Watkins, 1982). ABH antigens of secretors can be detected in their goblet cell secretions as well as secretions of their mucous glands of the genitourinary tract and respiratory tract. These antigens can also be detected in tears, milk, sweat and amniotic fluid (Morgan & van Heyningen, 1944). Although secreted ABH antigens are present as free carbohydrate chains in urine and milk, they are often carried by mucins (Kobata, 1972). Type 1, 2, and 3 structures express Secreted ABH antigens (Clausen & Hakomori, 1989). Salivary secretion of ABH antigens occurs as early as nine weeks of gestation and the antigens produced during this period are very well developed (Szulman, 1965). Research on secretors mostly using the Dolichos lectin or human anti-A has revealed higher quantities of A antigen in A1 saliva than in A2 saliva (Sturgeon, McQuiston, & Van Camp, 1973). However, relatively lower amounts of antigens A and B are found in saliva from group AB secretors (Sturgeon et al., 1973) and this has been attributed to competition between the A and B enzymes for H- antigen (Geoff Daniels, 2013). 9 University of Ghana http://ugspace.ug.edu.gh An individual’s secretor status can be determined by haemagglutination inhibition. To perform this test, an individual’s saliva is boiled to inactivate enzymes that may destroy blood group substances in it. Anti-A, -B, and -H are then appropriately diluted and mixed with the boiled saliva after which their inhibition is determined. The inability of the mixtures to respectively agglutinate A, B, and O cells determines their inhibition. 2.2.2 The H gene (FUT1) On chromosome 19q13 lies the H gene (FUT1) together with the Sec1 pseudo-gene and the secretor gene (FUT2) (Kelly, Rouquier, Giorgi, Lennon, & Lowe, 1995). Among its four exons, the active enzyme of FUT1 can be found in exon 4. The H gene shows specificity for type 2 precursor core structure and it is the only gene that controls the synthesis of antigen H on RBCs (Tsai et al., 2000). Erythroid-megakaryocytic progenitors and stem cells express FUT1 but expression of FUT1 cannot be detected in lymphoid and myeloid cells (Koda, Soejima, Wang, & Kimura, 1997). Epithelial tissues and other tissues also commonly express FUT1 but the expression is also absent in salivary and parotid glands (Cartron, 1976). α1,2-fucosyltransferase is made up of 365 amino acids and its catalytic domain contains 240 amino acids. The catalytic domain of the H antigen contains two cysteines, two conserved N- glycans and three motifs of α-fucosyltransferase (H. Clausen, S. B. Levery, E. Nudelman, M. Baldwin, & S. Hakomori, 1986; Tsai et al., 2000). GDP-fucose donor recognition is believed to be the responsibility of Motif I. On the other hand, Motifs II and III help to locate the terminal galactose (Cooling, 2016). Specifically along the catalytic domain, FUT1 and FUT2 transferases share 80 percent homology (S.-i. Hakomori, 1999). 10 University of Ghana http://ugspace.ug.edu.gh FUT1 activity and ABH antigen expression are not detected in Bombay and para-Bombay individuals. (Reid, Lomas-Francis, & Olsson, 2012). The conserved amino acids between FUT1 and FUT2 are involved in nearly all missense mutations (Kelly et al., 1995). FUT1 transcription is highly regulated with specific tissue promoters and alternative splicing. Close to four distinct mRNA transcripts can be recognized from two distinct transcription initiation sites in erythroid cell lines which are highly regulated. (Koda et al., 1997; Reid et al., 2012). Transcription of FUT1 in undifferentiated K562 cells primarily starts from exon 1A. Transcription, however proceeds from exon 2 in the bone marrow. Studies on FUT1 transcription has also been conducted in cancer and other cell lines. FUT1 is generally transcribed in epithelial cells from exon 1A, while transcription in vascular endothelium begins with exon 3 (Koda et al., 1997). The expression of FUT1 mRNA and LeY in colon carcinoma is up-regulated due to the presence of Elk-1(Taniuchi, Higai, Tanaka, Azuma, & Matsumoto, 2013). 2.2.3 Secretor gene (FUT2) The secretor gene controls an individual’s ability to secrete ABH antigens and it is located on chromosome 19 together with FUT1 and the SEC1 pseudo-gene (forming the FUT1-FUT2- SEC1 cluster) (Kelly et al., 1995). The SEC1 pseudo-gene which resides 50kb of FUT2 has 80% of its sequence identical with that of FUT2 gene but contains translation termination codons. It is believed that FUT1, FUT2 and SEC1 resulted from gene duplication. The secretor gene product, α2FucT2, catalyzes the transfer of fucose to the terminal galactose of the carbohydrate core structure in an α1-2 linkage. Whiles FUT1 shows specificity for type 2 chain precursors, FUT2 is responsible for types 1,3 and 4 ABH antigen synthesis (Chandrasekaran, Chawda, D., Piskorz, & Matta, 2002). Out of the two exons that FUT2 has, 11 University of Ghana http://ugspace.ug.edu.gh exon 2 is the only one that encodes the active enzymes (Kelly et al., 1995; Reid et al., 2012). The secretor gene shares 68% homology with the H gene and also has 343 amino acids (Kelly et al., 1995). Due to their high homology and proximity to several Alu clusters, the FUT1- FUT2-SEC1 gene cluster always becomes the object of mutation and genetic recombination (Koda et al., 1997). Homozygous FUT2-null alleles (se/se) can be found in non-secretors and Lea and/or Lec are found in their secretions and tissues (Cooling, 2016). Most null alleles result from non-sense or missense mutation leading to the synthesis of truncated proteins and se428 (Trp143stop, FUT2*01N.02) is the most common example (Reid et al., 2012). The se428 allele frequently occurs in Africans, Turks and Iranian non-secretors. Sew variant of the FUT2 gene is common in Asia and is characterized by weak activity. This variant is caused by missense mutations. Individuals with SeW/SeW and SeW/se genotype can type as Le(a+bW), Le(a+b-), or Le(a-b-) and particularly, those with Sew/se genotype may be phenotyped as non-secretors (Kudo, Iwasaki, & Nishihara, 1996). FUT2 is highly expressed in various parts of the human body including the trachea, female reproductive tract. The acquisition of FUT2 inducing bacteria occurs slowly in babies through breastfeeding, reaching normal “adult-type” intestinal microbiome by age one (Mackie, Sghir, & Gaskins, 1999). 2.2.4 Lewis gene (FUT3) The Lewis gene can be found on chromosome 19 (p13.3) and has about 90% sequence homology with FUT5 and FUT6. FUT3 forms a FUT5-FUT3-FUT6 cluster on chromosome 19 with FUT5 and FUT6 and the three genes are members of the GT10 family of α1,3 12 University of Ghana http://ugspace.ug.edu.gh fucosyltransferase (Reid et al., 2012). Exon 3 is the only exon out of the 3 exons on FUT3 gene that encodes the active enzyme. A type II glycoprotein with two N-glycan sites and five cysteines is the translated protein product of FUT3. FUT3 shares two extremely preserved α- fucosyltransferase motifs in the GDP-nucleotide donor (motif I) and acceptor domains (motif II) of the enzyme respectively (Breton, Oriol, & Imberty, 1998). Motif I includes a divalent cation (Mn2+) interacting DxD motif and the N-glycan site. FUT3 expression is limited to tissues and it is expressed strongly in the trachea, bladder and the intestine. Thirty-seven FUT3 null alleles which are associated with Le (a-b-) phenotype are currently known, most of which have a minimum of two mutations. Four or more single nucleotide polymorphisms can be observed in many of these alleles. Many of the FUT3 null alleles showed distinct distributions among some ethnic groups and geographical locations (Cooling, 2016). The predominant le alleles among Chinese and Japanese and other Asian countries are le59, le59,508 and le59,1067 whiles le202,314 predominates among whites (Cooling, 2016). Several mutations have been identified among Africans and black populations in America. The most common alleles in blacks are le13,484,667, le484,667, and le59,308. 2.2.5 ABO gene The four main ABO blood groups (A, B, AB and O) are controlled by three alleles on the ABO locus. Antigens A and B formation on red cells and secretions are under the control of the autosomal co-dominant A and B alleles while the O allele (inactive) does not yield any antigen. The ABO antigens are encoded on the ABO locus located on the terminal portion of the long arm of chromosome 9 (q34) (E. P. Bennett, Steffensen, Clausen, Weghuis, & van Kessel, 1995). This is a large gene spanning 18Kb and contains seven exons in the coding region whose sizes range from 28 to 688bp. Exons 1-5 which encodes the amino-terminal portion and the 13 University of Ghana http://ugspace.ug.edu.gh transmembrane region also contains about nine percent of the catalytic domain of the transferase (E. P. Bennett et al., 1995). Exons 6 and 7 accommodate most (77%) of ABO genes’ open reading frame (ORF) (fig 4) and close to 90% of the soluble, enzymatically-active glycosyltransferases (E. P. Bennett et al., 1995). Exon 6 and 7 alone make up 274 out of the 354 amino acids that constitute GTA/GTB and are the sites for most of the mutations affecting ABO activity. The intron sizes range from 13kb to 554bp. Organization of the ABO gene has been illustrated in figure 4. Figure 1: Organization of the ABO gene Adapted from A. Hult (2013) The numbers above represent the number of nucleotides in each exon and the numbers below correspond to the introns. Amino acid numbers are written in boxes. Seven nucleotide changes in exons 6 and 7 differentiate the A and B alleles. Four of these seven changes encode amino acid substitution. The O and B genes are believed to arise from the A1 gene which is considered as an ancestral gene (Saitou & Yamamoto, 1997). The A1 sequence is the consensus sequence that other ABO allele sequences are compared with (Geoff Daniels, 2013). 14 University of Ghana http://ugspace.ug.edu.gh Weak ABO phenotypes result from mutations in ABO genes and more than one hundred of such alleles are known (Reid et al., 2012). Most of these weak alleles result from missense mutations which affect the activity of the enzymes. The commonest mutant allele (ABO*A2) is linked to the A2 phenotype (Reid et al., 2012) which results from a deletion of one of the three cytosines located just before the stop codon in the A1 allele. This results in the abolishing of the stop codon and translation continue until a new stop codon is met. The A-transferase product formed has 21 extra amino acids (F. Yamamoto, McNeill, & Hakomori, 1992). The inheritance of an amorph ABO alleles by an individual results in ABO blood group O phenotype, which is inherited as autosomal-recessive phenotype. Although many group O alleles exist, O1 and O2 are the commonest with O1 constituting about 95% of all O alleles (Reid et al., 2012). A single base deletion (261delG) in the O1 sequence differentiates it from that of A1 (Olsson & Chester, 1996a). The O1 phenotype results from a deletion in exon 6 of the ABO gene which induces a frameshift, creating a new stop codon (nucleotides 352-354). This results in the formation of a shorter protein product (117 amino acids) with no detectable glycosyltransferase activity (O ’ Keefe & Dobrovic, 1996). The O2 allele undergoes a missense mutation at nucleotide 802 (G-A) which causes a glycine-arginine substitution at codon 268 (F. Yamamoto, McNeill, & Yamamoto, 1993). This Gly268-Arg substitution was shown to take away the protein's enzymatic activity (F. Yamamoto et al., 1993). The less common O2 allele lacks 261delG (non-deletional) but shows differences in its nucleotide (Arg176Gly and Gly268Arg) from the consensus A1 allele at exon 7 (table 5) (F. Yamamoto et al., 1993). 15 University of Ghana http://ugspace.ug.edu.gh Figure 2: Schematic presentation of the Open Reading Frame (ORF) of five common ABO alleles Adapted from A. Hult (2013) The translated non-A1 consensus is represented by light grey bars. Changes in the amino acids (dark grey squares) are compared with the corresponding residues translated by the consensus A1 allele (top white squares). Nucleotide positions are indicated by white vertical bars whiles nucleotide changes are represented by the lower white squares. The cisAB phenotype results from a hybrid ABO allele which synthesizes both antigen A and B. These individuals express the A2B phenotype with very weak B expression and frequently develop allo-anti-B (Issitt & Anstee, 1998). cisAB was formally thought to result from unequal crossover which leads to hybrid A-B gene. However, studies have shown that cisAB is caused by a mutation in the A or B gene that causes it to produce an enzyme that can transfer significant 16 University of Ghana http://ugspace.ug.edu.gh amounts of both A and B sugar to the H acceptor substrate (Geoff Daniels, 2013). This mutation mostly results from single nucleotide changes encoding amino acid substitution at position 266 (Met—Leu) or 268(Gly—Ala) (Mifsud, Watt, & Condon, 2000; Ogasawara, Tsuneyama, & Uchikawa, 2010). GTA and GTB enzyme activities are detected in cisAB alleles. ABO*BA allele also brings about the B(A) phenotype which is rarely seen in blood group B individuals. These group B individuals have in addition to their large quantity of antigen B, a small amount of antigen A which is usually detected by some monoclonal anti-A antibodies during typing. Unlike ABO*BA, red cells from the cisAB phenotype have equivalent amounts of antigens A and B (M. Bennett, Levene, & Greenwell, 1998). There is evidence of tissue-specific regulation of ABO mRNA expression. The ABO 5´ UTR which is located 5’ of exon 1 is the first region that is regulated. Methylation of the CpG rich region (82%) is believed to play an important role in regulating ABO expression in various tissues (Kominato, Hata, & Takizawa, 1999). The most frequently used transcription site is located 12-38 bases upstream of the start codon of exon 1. An alternative transcription start site (exon 1a) is also located at the 5’ end of the CpG island (Kominato, Hata, & Takizawa, 2002 ). The alternative exon 1a lacks the ATG start codon as seen in exon 1. A start codon located in exon 2 is used instead and this results in a functional but N-truncated glycosyltransferase (Kominato et., al) ABO transcription regulation may also depend on a mini-satellite containing a CBF/NF-Y transcription factor-binding motif (M. Yamamoto et al., 2003). This mini-satellite has four copies of a 43bp repeat sequence in A2, B, O 1 and O1v alleles, but only one copy in A1 and O 2 alleles (Irshaid, Chester, & Olsson, 1999). Studies conducted on gastric cancer cell lines showed a very low transcriptional activity in the A enhancer compared to the B enhancer (Twu, Hsieh, & Tu, 2006; Yu, Chang, Twu, & Lin, 2000). Unlike B, O1, O1v and O2 allele transcripts which are easily detected in peripheral blood, those from A1 and A2 alleles are not detected. 17 University of Ghana http://ugspace.ug.edu.gh Bone marrow erythroid cells express higher A1 and A2 transcript levels than those from B, O1, O1v and O2 (Thuresson, Chester, Storry, & Olsson, 2008). Hypermethylation of ABO promoter site in leukemia patients has been linked to loss of A/B expression (Bianco-Miotto, Hussey, Day, O’Keefe, & Dobrovic, 2009). There is no TATA box upstream of the ABO promoter region as seen in many glycosyltransferases. This proximal promoter region also houses transcription factor Sp-1 binding site and other transcription factor binding proteins (Kominato et al., 2002 ). Transcription is enhanced in erythroid cells by the help of two GATA regions within intron 1 (+5819 and +5890) (Sano, Nakajima, & Takahashi, 2012). In conclusion, the ABO gene which is found in humans and other species is well studied and the genetic background for its polymorphism is well understood. Most of these result from a single nucleotide polymorphism that leads to the synthesis of entirely different amino acid. The A and B genes encode different glycosyltransferases of different specificities. These transferases are type II transmembrane enzymes. The inactive O gene however, cannot produce any enzyme. Antigens A and B are weakly expressed as a result of SNPs in any of the ABO gene’s seven exons. The changes in amino acid affect the specificity and/or activity of the glycosyltransferase. 18 University of Ghana http://ugspace.ug.edu.gh 2.3 STRUCTURE OF ABO, LEWIS AND H ANTIGENS Most of the early studies on the chemical structure of ABH antigens were done on materials extracted from the RBC membrane (Lowe, 1995). Later studies revealed that these antigenic molecules are also expressed in tissues and are not exclusive to red cells (Watkins, 1980). The antigens were however, previously known to be present in soluble form in body secretions (Yamakami, 1926). The early works demonstrated that the antigenic moieties of the ABH antigens are composed of oligosaccharides molecules (Kabat, 1956). Subsequent studies using organic synthesis procedures helped to determine the constituents of antigen B and H determinants (Lemieux & Driguez, 1975). Lewis and ABH antigens consist of oligosaccharide chains which may be attached to polypeptides (glycoproteins) or may be attached to ceramides (glycosphingolipids). The synthesis of these oligosaccharides occurs in a stepwise fashion with specific glycosyltransferases catalyzing the addition of each monosaccharide. 2.3.1 Glycoconjugates that express ABH antigens The major oligosaccharide chains located on glycoproteins that express ABH antigens are O- and N-glycans. The N-glycans consist of carbohydrate structures that have been linked to the amide nitrogen of aspargine via N-acetylglucosamine (GlcNAc). O-glycans however, may consist of simple or complex structures and attached to the hydroxyl oxygen of serine or threonine through N-acetyl galactosamine (GalNAc). ABO-active oligosaccharides can also be found on glycolipids. Close to 75% of ABH antigens on RBCs are estimated to be expressed as N-linked glycans on glycoproteins, especially Band 3 (Wilczynska, Miller-Podraza, & Koscielak, 1980). Ten 19 University of Ghana http://ugspace.ug.edu.gh percent of ABH antigens are estimated to be present on O glycans whiles 5% are expressed on glycosphingolipids (Wilczynska et al., 1980). Close to 15% of ABH antigens are also expressed on polyglycosylceramides (Wilczynska et al., 1980). Being a major glycoprotein on red cells with about one million monomers, Band 3 constitute about 50% of all ABH antigens (Fukuda, Dell, Oates, & Fukuda, 1984). Lewis antigens are adsorbed from plasma hence they are not expressed on glycoproteins. Based on the nature of their carbohydrate chains, glycosphingolipids can be grouped into lacto, globo or ganglio-series. ABH and Lewis antigens originating from glycosphingolipids are predominantly present on lacto-series glycolipids, although they are also present on ganglio and globo-series glycolipids. The carbohydrate chains of most glycoproteins that bear ABH antigens and lacto-series glycolipids are extended by repeating Galβ1→ 4GlcNAcβ 1→ 3 disaccharides (poly-N-acetyllactosamine structure). ABH substances can be obtained from body secretions but abundant quantities can be obtained from the human ovarian cyst (Morgan & van Heyningen, 1944). The secreted ABH and Lewis antigens are glycoproteins that have been linked to mucins. Free oligosaccharides which have ABH and Lewis activities can also be found in milk and urine (Kobata, 1972; Lundblad, 1978 ). Plasma ABH and Lewis antigens determinants are present on glycosphingolipids. Some of these determinants however, may be attached to the membranes of RBCs. 2.3.2 Carbohydrate determinants of ABH and Lewis antigens For ABH antigens to be expressed, specific monosaccharides must be attached to the peripheral core structures at the non-reducing end of the carbohydrate chain. Five main peripheral core structures (precursor disaccharides) have been found (Clausen & Hakomori, 1989) ; Type1 Galβ1→ 3GlcNAcβ1→ R 20 University of Ghana http://ugspace.ug.edu.gh Type2 Galβ1→ 4GlcNAcβ1→ R Type3 Galβ1→ 3GalNAcα1→ R Type4 Galβ1→ 3GalNAcβ1→ R Type6 Galβ1→ 4Glcβ1→ R R indicates the carrier which can be carbohydrate, glycolipid or glycoprotein The structure of H-antigen is made up of fucose linked in an α-linkage to C2 of the terminal galactose (Weiner, Lewis, Moores, Sanger, & Race, 1957). H-activity is lost in the absence of fucose hence fucose is H-antigen’s immunodominant sugar (Geoff Daniels, 2013). Persons with Bombay phenotype lack the H-antigens on their RBCs and secretions irrespective of their ABO group (F. Yamamoto et al., 2014). Table 2: ABO antigens and their gene products ABO Antigen Immunodominant sugar Gene product Phenotype A A N-acetylgalactosamine α1,3-N-acetylgalactos- aminyltransferase (GTA) B B d-galactose α1,3-galactosyltransferase (GTB) AB A & B N-acetylgalactosamine and d- Both GTA and GTB galactose O none L-fucose Absent or non-functional GTA or GTB GTA- glycosyltransferase A, GTB- glycosyltransferase B 21 University of Ghana http://ugspace.ug.edu.gh The immunodominant sugars of antigens A and B are N-acetyl-d-galactosamine (GalNAc) and d-galactose respectively (table 2). For antigen A and B structures, GalNAc and Gal are respectively attached to C-3 of the α1→2 fucosylated Gal residue via α-linkage (fig 1). Elimination of the A and B immunodominant sugars from their common structure results in loss of reactivity with their corresponding antibodies. Blood group O RBCs lack both GalNAc and Gal, hence they do not express A or B substances but only H substance. Figure 3: Structures of the immunodominant sugars of antigens H, A and B antigens R indicates the carrier which can be carbohydrate, glycolipid or glycoprotein. Modified from Westhoff and Reid (2007) 22 University of Ghana http://ugspace.ug.edu.gh Expression of Lewis antigens, Lea and Leb, requires fucose to be linked to the GlcNAc residue of Type 1 precursor and Type1H respectively (Rege, Painter, Watkins, & Morgan, 1964; Watkins & Morgan, 1957). The Type 2 isomers of Lewis antigens (Lex and Ley) are not significantly present on RBCs (Gooi, Feizi, & Kapadia, 1981 ; Marr, Donald, & Morgan, 1968). Fucose is linked α1-4 to the GlcNAc residue of a Type 1 chain in Lea and Leb and α1-3 to the GlcNAc of type 2 chain in Lex and Ley (F. Yamamoto et al., 2014). RBCs do not synthesize type1 structures but these structures are incorporated into the RBC membranes from plasma (Marcus & Cass, 1969 ). The Type 1 precursor structures are rather synthesized by endodermally derived tissues which include, epithelial cells lining the digestive, urinary and respiratory tracts (R. Oriol, Le Pendu, & Mollicone, 1986). The major ABH molecule carriers which are detected on proteins and lipids in secretions and other body fluids are Type 1 determinants (Lowe, 1995). Lea and Leb can only be detected on type1 core structures. Type 2 chain structures form the majority of ABH-active oligosaccharides on RBCs and can be detected in secretions, ectodermally derived tissues (like the epidermis) and mesodermally derived tissues (like red cell) (Rafael Oriol, 1995; Watkins, Greenwell, Yates, & Johnson, 1988). In secretions, Type 2 structures are mostly defucosylated (Ley, ALey, BLey) (Sakamoto, Yin, & Lloyd, 1984 ). Type 3 ABH antigens consist of two types, the O linked mucin and the repetitive A-associated types. O-glycosidic bonds directly link disaccharide to a serine-threonine residue of mucin in the precursor of the O-linked mucin type (Donald, 1981 ). ABH antigens from this O-linked mucin type cannot be detected on RBCs (Le Pendu, Lambert, Gerard, et al., 1986). However, repetitive Type 3 chains can be found on the glycolipids of RBCs and secreted mucins of blood group A individuals. Their restriction to blood group A is due to their biosynthesis which 23 University of Ghana http://ugspace.ug.edu.gh involves the addition of galactose to the terminal GalNAc of A-active Type 2 chain followed by fucosylation of that gal to form Type 3H (H. Clausen et al., 1986; Le Pendu, Lambert, Gerard, et al., 1986). The Type 4 core structures can only be detected on glycolipids. Their globo-series precursor chain is made by linking the terminal Galactose to the globoside (Kannagi, Levery, & Hakomori, 1984 ). Large quantities of globo-H and -A can be found in the kidney (Breimer & Jovall, 1985) but very small quantities have been detected on red cells (Clausen, Watanabe, & Kannagi, 1984; Kannagi et al., 1984 ). Type 5 is chemically synthesized. Type 6 chains are detected as free oligosaccharides in milk and urine (Lau, Sererat, Beaty, Oilschlager, & Kini, 1990; Lundblad, 1978 ). 24 University of Ghana http://ugspace.ug.edu.gh 2.4 BIOSYNTHESIS OF ABH ANTIGENS The synthesis of ABH carbohydrate antigens requires sequential addition of monosaccharides to existing oligosaccharide chains which is catalyzed by specific enzymes, the glycosyltransferases. These enzymes specifically catalyze monosaccharide transfer from a nucleotide donor to a nucleotide acceptor in a specific glycosidic bond. Glycosyltransferases therefore represent the primary products of ABH, Secretor and Lewis genes. N-acetyl galactosaminyl transferase and galactosyltransferase are the respective enzymes for blood group A and B antigens. The H transferase is of two forms, ie α2FucT1 and α2FucT2 depending on whether it acts on cells or secretion. Α2FucT2 is the secretor transferase and it is found in the secretions of over 80% of Europeans. Biosynthesis of ABH and their related antigens involve very complex regulatory mechanisms. This involves the presence or absence of various enzymes at various stages according to the genes being expressed in tissues and the genotype of the individual. The various transferases compete for donor and acceptor substrates and this is very helpful in determining the carbohydrate chain produced (S. Hakomori, 1981). 2.4.1 Biosynthesis of H antigen Synthesis of antigen-H, which is an immediate biosynthetic precursor for the synthesis of antigens A and B, is the first step in ABH antigen synthesis. Antigen H synthesis is controlled by FUT1 (H gene) and FUT2 (secretor gene) on red cells and in secretions respectively. These closely linked genes on chromosome 19 act independently of the ABO gene (Reid et al., 2012). These genes produce two enzymes, α2FucT1 and α2FucT2, which act on different core structures in secretions and on red cells. The H transferase primarily synthesizes H substance on type 2 core structures on erythrocytes. α2FucT2 acts mainly on type 1 core structures present 25 University of Ghana http://ugspace.ug.edu.gh in secretions and plasma. FUT1 and FUT2 consist of four and two exons respectively. Only one of those exons in both genes encodes a protein product (exon4 for FUT1 and Exon2 for FUT2) (R. Oriol, Mollicone, & Cailleau, 1999; Sarnesto, Kohlin, & Hingsgaul, 1992). For H antigen to be synthesized, fucose (Fuc) is added to a terminal galactose of the core oligosaccharide structure (fig 2). This is catalyzed by α2FucT1 or α2FucT2, depending on the core structure of the oligosaccharide acceptor. The H-enzyme, α2FucT1, adds fucose to the terminal galactose of the precursor oligosaccharide to make type2 H-antigen on red cells. However, the secretor transferase, α2FucT2, adds fucose to the terminal galactose of the precursor oligosaccharide to make type 1H-antigen in secretions. Once the H antigen is formed, it becomes the substrate for A and B gene glycosyltransferases irrespective of being a product of FUT1 or FUT2 gene. 2.4.2 Biosynthesis of antigens A and B H-substance is very essential in the biosynthesis of A and B substances. The A and B substances cannot be produced in the absence of H transferase even though their respective transferases may be present. This happens in the secretions of ABO non-secretors and on RBCs of individuals with Bombay phenotype (H-deficient individuals). A and B transferases recognize H substance as their acceptor substrate and transfer Gal or GalNAc to their terminal galactose (F. Yamamoto et al., 2014). The A-transferase in blood group A individuals, α1,3-N- acetyl-d–galactosaminyltransferase (GTA), recognizes uridine diphosphate (UDP)-GalNAc to transfer an α1-3N-acetylgalactosamine (GalNAc) to the fucosylated galactose residue of the H substance. The B-transferase, α-1,3-d-galactosyltransferase (GTB) however, transfers α-1,3 galactose (Gal) to the terminal galactose of the H-substance in order to synthesize B substance. 26 University of Ghana http://ugspace.ug.edu.gh Both the A and B transferases do not transfer their respective sugars to the terminal galactose when it is not fucosylated. Antigens A and B are co-expressed in group AB individuals because they inherit both A and B alleles. Both antigen A and B are therefore synthesized in group AB individuals. The protein encoded by the O allele has no enzymatic activity, hence leaving the fucosylated galactose unsubstituted (and express H substance) (F. Yamamoto, Clausen, & White, 1990). The transfer of GTA and GTB is restricted to only acceptors that carry the terminal H- determinant. However, the A and B enzyme structures are very similar (an acetyl group differentiates them), hence the enzymes may sometimes cross-react towards the donor substrate. That is, GTA is capable of transferring a little quantity of galactose residues to the H precursor. Similarly, the B transferase is also able to transfer a small quantity of GalNAc residues to the H precursor. This cross-reactivity results in the production of hybrid ABO antigens especially under altered physiological conditions (Helmut Schenkel-Brunner, 2000). 27 University of Ghana http://ugspace.ug.edu.gh Figure 4: Biosynthesis of ABH antigens Adopted from (Lowe, 1995) Fuc- fucose, Gal–galactose, GalNAc-N-acetylgalactosamine, GlcNAc-N-acetylglucosamine, GDP- guanosine diphosphate, UDP- uridine diphosphate, R-the glycoconjugate substructure consisting of N-linked, O-linked, or lipid-linked glycoconjugates 28 University of Ghana http://ugspace.ug.edu.gh 2.4.3 A and B transferases GTA and GTB do not change nucleotide donor configuration and hence called retaining enzymes. These transferases occur naturally and differ by four amino acids at 176, 266, 235 and 268 (Yazer & Palcic, 2005). They are homologous and transfer different naturally occurring donors (Yazer & Palcic, 2005). Yamamoto et al used recombinant enzymes to demonstrate that Gly235Ser, Leu266Met and Gly268a1 are the polymorphisms that define the presence or absence of GTA and GTB activity on enzymes. Both GTA and GTB are type II transmembrane proteins which contain 354 amino acids and are located in the lumen of the Golgi apparatus. These enzymes have short cytosolic amino- terminal domains (amino acids 1-32), short protease-sensitive stem regions and large globular carboxy-terminus where the catalytic region is located (F. Yamamoto, 2004). GTA and GTB have single N-glycan sites (Asn113) and five cysteines. Advanced study of the secondary and tertiary structure of the ABO protein shows two functional domains on either side of a catalytic cleft (13A wide) with a highly conserved Glu303 (Patenaude, Seta, & Borisov, 2002). The carboxy-terminal of the enzyme serves as the site for H-antigen acceptor binding whereas the amino-terminal serves as the site for uridine diphosphate (UDP)-nucleotide sugar donor binding (fig 5). The soluble GTA and GTB have lots of similarities in their coding regions and are detected in urine (Chester, 1974), serum (Schenkel‐Brunner, Chester, & Watkins, 1972) and milk (Ginsburg, 1972). The UDP-binding pocket size and geometry appear to be the determining factors to the enzyme’s utilization of UDP-Gal or UDP-GalNAc. Unlike that of GTB, GTA- associated amino acids appear biochemically “smaller” hence they are able to contain the larger UDP-GalNAc donor (Patenaude et al., 2002). Tyr126, Glu303 and amino acids 233–245 also interact with the H antigen acceptor. The DxD motif located close to the enzyme’s nucleotide- 29 University of Ghana http://ugspace.ug.edu.gh binding pocket binds Mn2+ which helps in coordinating the UDP-phosphate group (Patenaude et al., 2002). Figure 5: Structure of α-1,3-d-galactosyltransferase (GTB) showing the two domains separated by a central cleft, UDP - Gal and H antigen Adopted from Daniels (2013) 30 University of Ghana http://ugspace.ug.edu.gh 2.4.4 H and Secretor transferases The H gene produces the H transferase, α1,2-fucosyltransferase, which further catalyzes the attachment of α-fucose to the terminal galactose of the oligosaccharide core structure during the synthesis of H substance. The secretor gene, Se/se, encodes a second α1,2- fucosyltransferase (a2FucT2) (R. Oriol, Daniiovs, & Hawkins, 1981) and also regulates H- antigen expression in secretions (Watkins, 1980; Watkins & Morgan, 1959). Both transferases are known to have distinct stereochemical differences especially in type 1 and type 2 oligosaccharide core chains (Lemieux, 1978). The C-2 hydroxyl group in the type1 chain is restricted by the acetyl group of the subterminal N-acetylglucosamine whiles that of type 2 chain is easily accessible (Lemieux, 1978). This shows that two different enzymes synthesize type 1 and 2H substances. Secretor transferase is believed to be expressed on endodermally derived tissues whiles the H- transferase is restricted to tissues originating from ectoderm and mesoderm (Schenkel-Brunner, 2000). Significant differences are observed in the kinetic properties of the transferases and also their affinity towards their respective substrates. “The secretor-transferase showed a significantly lower affinity for GOP-fucose, phenyl-13-0-galactoside, and type-2 oligosaccharide acceptors than the H-transferase, but preferentially utilized type-1 and type-3 chain acceptors” (Helmut Schenkel-Brunner, 2000) The H-enzyme however, strongly prefers type 2 chain acceptors and is less sensitive to heat (Le Pendu, Cartron, Lemieux, & Oriol, 1985). Enzymes in many parts of the human body are known to express α1,2-fucosyltransferase activity (Cheng & DeVries, 1986). The presence of α1,2-fucosyltransferase in the serum (Munro & Schachter, 1973), erythrocytes (Cartron, Mulet, & Bauvois, 1980) and bone marrow (Pacuska & Koscieak, 1974) depends on an individual’s secretor status. H deficient individuals 31 University of Ghana http://ugspace.ug.edu.gh lack the H transferase. It is also absent in the saliva (Yazawa & Furukawa, 1980) and milk (Shechter, Etzioni, Levene, & Greenwell, 1995) of ABH non-secreting individuals. These findings buttress the fact that the inability of non-secretors to secrete antigen A and B is due to the absence of H acceptor substrate but not the absence of A/B gene products (Schenkel- Brunner, 2000). 32 University of Ghana http://ugspace.ug.edu.gh 2.5 INHERITANCE OF ABO BLOOD GROUPS Epstein and Ottenberg (1908) first suggested that ABO blood group antigens were inherited, a claim which was later confirmed in 1910 (von Dungern and Hirschfeld in 1910). During this period, ABO antigens were believed to be inherited on separate genes until Bernstein demonstrated the existence of three allelomorphic genes (A, B and O) on chromosome 9 (Bernstein, 1924). Each individual can inherit only two of these alleles, one from each parent (fig 6). This results in a lot of possible genotypes combinations; AA (AO), BB (BO), AB and OO. Bernstein further revealed the co-dominant inheritance of antigens A and B over O (Bernstein, 1924). The O gene does not produce any active substance and individuals who inherit this gene do not produce any detectable antigens. Alterations in the ABO gene leads to the aberrant or weak expression of the ABO antigens resulting in subgroups. ABH antigen expression in secretions is controlled by FUT2 gene whiles FUT1 gene controls ABH antigen expression on RBCs. 33 University of Ghana http://ugspace.ug.edu.gh Figure 6: Inheritance of ABO antigens Adopted from Fox (2016) 34 University of Ghana http://ugspace.ug.edu.gh 2.6 BLOOD GROUP A ABO antigen distribution varies worldwide and their proportions vary even among different ethnic groups. In a study conducted among university students in Kumasi with 412 participants, Smith et al., (2018) revealed the prevalence of A, B, AB and O among Ghanaians to be 19.7%, 26.0%, 3.1% and 51.2% respectively. Another study conducted in the central region of Ghana revealed the frequency of blood groups A, B, AB and O as 18.9%, 19.6%, 3% and 58.3% respectively among Ghanaians (Kretchy, Doku, Annor, Addy, & Asante, 2017). In Africa, a study revealed the frequencies of blood groups A, B, AB and O among Ethiopians (Blacks) to be 32.7%, 20.9%, 4.3% and 42.1% (Alemu & Mama, 2016). A similar study among Nigerians (Blacks) revealed ABO distribution to be 23.1%, 21.3%, 2.7% and 52.9% respectively among A, B, AB and O (Iyiola, Igunnugbemi, & Bello, 2012). Forty-one percent of Caucasians in the United States are estimated to have blood group A whiles 27% was recorded among African Americans. The frequency of blood group A among western Europeans is 42% (Pramanik & Pramanik, 2000). The A antigen was found to show variation soon after ABO discovery (R. Race & Sanger, 1975) and was grouped into two principal subtypes; A1 and A2. Rare weaker subgroups of antigen A exist (A3, A4, Ax, Aend etc) and are characterized by an increased H antigen activity with a decreased A antigen sites on RBCs (Elnour et al., 2015). Erythrocytes from weaker subgroups of A weakly react or give no reaction with anti-A, an indication of their weakness in converting H substance to A. 35 University of Ghana http://ugspace.ug.edu.gh 2.7 ABO SUBGROUPS ABO subgroups are differentiated by the quantity of A and B substances on their RBCs and secretions. A and B activity is usually significantly reduced or may be undetected in these subgroups. A significant reduction in antigen A and B activity is also observed in secretions of secretor individuals with ABO subgroups. In general, there are more A subgroups than B subgroups. Von Dungern and Hirschfeld first recognized in 1911 that subgroups of A exist, with one subgroup weakly expressing antigen A than the other (von Dungern & Hirszfeld, 1911). The presence of weak ABO subtypes sometimes results in discrepancies in ABO forward and reverse typing. It is estimated that 1 to 8 percent of blood group A2 individuals and 22 to 35 percent of blood group A2B individuals have an anti-A1 antibody in their sera. A2 RBCs strongly react with Ulex europaeus (anti-H lectin) due to the inefficient conversion of H→A antigen by A2 transferase. Weaker A subgroups have been described in addition to A2 (eg, A3, Ax, Am, and Ael). These subgroups are sometimes generally categorized as Aw or Aweak, signifying their weak expression of A. The very weak A and B subgroups are rarely encountered routinely. They are usually recognized after they have caused discrepancies between the forward and reverse grouping results. Weaker subgroups of A mostly do not react with human polyclonal anti-A but shows variable reactivity with human polyclonal anti-A1, anti-A,B, and murine monoclonal antibodies. The strength of reaction with commercial murine monoclonal reagents is clone dependent; however, most commercially available anti-A agglutinates A3 red cells. Nearly all weak ABO subgroups have higher expression of antigen H due to the reciprocal relationship between H antigen and A and B antigen synthesis (C. Race, Ziderman, & Watkins, 1968). 36 University of Ghana http://ugspace.ug.edu.gh The subgroups of B are rarely encountered and unlike the A subgroups, they are mostly very difficult to classify (Geoff Daniels, 2013). The RBCs from weak B variants show weak reaction with anti-B or anti-A,B but absorb anti-B. Secretors exhibit distinct B activity as well as strong H activity in their secretions (Schenkel-Brunner, 2000). Anti-B can be found in some B variants although they do not agglutinate their own red cells under normal conditions. Bm and Bx subgroups are frequently encountered among the Japanese population (Yamaguchi, Okubo, & Tanaka, 1970). 2.7.1 Major subgroups of A The most common subgroups encountered clinically are A1 and A2. Majority of group A individuals (80%) belong to subgroup A1 which is characterized by approximately one million A antigen sites on each RBC (Schenkel-Brunner, 2000). The A1 allele seems dominant over A2 and individuals who inherit A 1/A2 and A1/A1 genotypes have the A1 phenotype when determined serologically. In routine blood testing, both A1 and A2 red cells show strong agglutination with monoclonal anti-A. The Dolichos lectin is used to distinguish A1 red cells from A2. This lectin agglutinates A1 cells due to the high quantity of A substance on their surface, leaving A2 cells which have lower amounts of A substance. Two major antibodies, Anti-A and anti-A1, are detected in blood group B serum. A1 RBCs react with both group A antibodies whiles A2 RBCs react with only anti-A antibody (von Dungern & Hirszfeld, 1911). A2 RBCs can be used to adsorb anti-A from group B serum leaving anti-A1. However, continuously adsorbing group B serum using A2 cells will remove all antibodies (Lattes & Cavazzuti, 1924 ). Serum from some group A2 and A2B individuals 37 University of Ghana http://ugspace.ug.edu.gh have been found to contain anti-A1 which is a cold reacting antibody that is capable of causing haemolysis at 37oC (K. Landsteiner & Levine, 1926; K. Landsteiner & Witt, 1926 ). 2.7.2 Blood Group A2 A2 generally occurs in 20% of all group A individuals while A1 subgroup occurs in approximately 80% of all group A individuals (Schenkel-Brunner, 2000). Anti-A1 is produced by about 1-8% and up to 35% of A2 and A2B individuals respectively and can agglutinate A1 and A1B cells (Cooling, 2014). This shows the existence of another antigen on A1 RBC’s, called A1, which is absent from A2 cells. Molecular studies on the cDNA of A2 individuals revealed a cytosine deletion among the three consecutive cytosine nucleotides at positions 1059-1061, i.e. just before the stop codon in the A1 allele coding sequence (fig.7). This results in a frame-shift and abolishing of the stop codon. Translation therefore continues until a new stop codon is encountered. The A-transferase product formed has 21 extra amino acids and decreased activity (F. Yamamoto et al., 1992). Studies had revealed a C-T exchange at nucleotide 467 leading to a Proline-Leucine substitution at position 156 in all A2 individuals (Schenkel-Brunner, 2000). This mutation has also been observed in subgroup A1 individuals in Japan (Geoff Daniels, 2013) but does not affect enzyme activity or sugar-nucleotide donor specificity (Schenkel-Brunner, 2000). 38 University of Ghana http://ugspace.ug.edu.gh Figure 7: cDNA (black line) and protein products (coloured box) of ABO alleles, showing how A2 (A201) differs from A1 (A101). A1 (A102) is the commonest A1 allele among East Asians and differs from A1 in Europeans (A101) by 467C > T encoding Pro-156Leu. A2 (A201) results from a single base deletion in A1 and has an additional 21 amino acids (Modified from Daniels, 2013) Advanced studies on A1 and A2 subgroups have revealed there are four H antigens (H1, H2, H3 and H4) which serve as the precursor for the synthesis of their respective A antigens (A a, Ab, Ac and Ad) (Fujitani, 2000). Although they have identical terminal sugars, H1 and H2 have simple and straight chains whiles H3 and H4 have highly branched and complex chain structures. A1 and A2 transferases can respectively convert H1 and H2 straight-chain glycolipids to Aa and Ab antigens, although A2 transferase does so with less efficiency (Fujitani, 2000). A1 transferase can also convert the complex H3 and H4 glycolipid respectively to A c and Ad antigens with very high efficiency. However, the A2 transferase poorly converts the highly branched H3 and H4 chains to A c and Ad antigens, resulting in more unconverted H3 and H4 antigens on A2 RBCs (Fujitani, 2000). 39 University of Ghana http://ugspace.ug.edu.gh Studies show that group A individuals who are most likely to develop anti-A1 in their serum have extremely low amounts of Ac antigens with a total absence of antigen Ad on their red cells (Harmening, F., & T., 2012). Anti-A1 has therefore been suggested to be antibody to A c and Ad determinants, which is absent in group A2 individuals with Anti-A1 (Harmening et al., 2012). Newborns with blood group A do not have the branched H3 and H4 structures and hence lack the Ac and Ad antigens. As a result, they usually type as group A2 at birth, which later returns to A1 after a few months. Adults however have the H3 and H4 structures and hence develop the Ac and Ad antigens which are seen in A1 individuals (Hosoi, Hirose, & Hamano, 2003). Laburnum alpinum, an anti-H lectin, is much useful than Ulex europaeus in distinguishing A1 from A2 in newborns (Pawlak & Lopez, 1979). A2 red cells strongly react with the Laburnum alpinum lectin than A1 cells. Also, the Dolichos lectin distinguishes the A1 and A2 infant cells better than human anti-A1, especially when the cells are enzyme-treated (Klein & Anstee, 2006). In Japan, the most common alleles responsible for A2 have different missense mutations within codon 352 but do not have 1016delC (Ogasawara et al., 1998). Another common hybrid allele in Japan is the AB-O1v. When this allele is linked with O in a heterozygote, an A1 phenotype is produced. However, when the allele is paired with B, it produces A2B phenotype. This has been attributed to competition between A-active hybrid transferase and GTB for a common acceptor (Ogasawara et al., 1998). AB-O1v allele has been cited as the cause of imbalance of A2 and A2B phenotypes in Japan. Many alleles with 1016delC are responsible for A phenotypes, most of which display the characteristics of Ax subgroup (A. K. Hult, Yazer, & Jørgensen, 2010). Abantu is a weak A phenotype that occurs in close to 4% of South Africans (Blacks) (Brain, 1966). “This results 40 University of Ghana http://ugspace.ug.edu.gh from a hybrid of the common A2 allele with 1016delC and an O 1-like allele (O1bantu), with crossover region near exon 5 (Abantu01) ” (Geoff Daniels, 2013). Other similar hybrid alleles that result in weaker A phenotypes have been discovered among people of African descent (Hosseini-Maaf, Hellberg, Rodrigues, Chester, & Olsson, 2003). A2 glycosyltransferase shows both qualitative and quantitative difference from the A1 glycosyltransferases (section 2.7.3) (F Yamamoto et al., 2014). The density of A substance on the A2 RBCs is also lower than on A1. However, A2 RBCs show increased amounts of antigen H on their surface (F. Yamamoto et al., 1992). The antigen density of A1 on RBCs is four times higher than that of A2 RBCs and the optimum pH for A1 and A2 glycosyltransferases are 5.6 and 7-8 respectively (Watkins & Morgan, 1957). 2.7.3 Characteristics of A1 and A2 Transferases Studies on the major A transferases, GTA1 and GTA2, reveal that they both perform the same function (Schenkel-Brunner, 2000). That is, they are both N-acetyl-galactosaminyl- transferases, and transfer a1-3N-acetylgalactosaminyl residues specifically onto terminal H determinants. However, significant differences can be seen in their kinetic properties. These difference include Km values for acceptor substrates, reaction velocity, optimum pH, cation requirement and isoelectric point (Schachter, Michaels, Crookston, Tilley, & Crookston, 1971). GTA extracted from the serum of group A1 individuals has proven more efficient in converting group O red cells to A1 cells than that extracted from group A2 individuals (Helmut Schenkel- Brunner, 1982). The reaction is 5 to 10 times slower when A2 enzyme is used to convert O and A2 cells to A1 phenotype and hence requires an extended incubation time (Schachter et al., 1971). Extending the incubation time with A2 enzyme may result in weak agglutination of O cells by A1 specific reagents (Schenkel - Brunner, 1982). 41 University of Ghana http://ugspace.ug.edu.gh The ability of A2 transferase to produce A1 specific red cells from A2 and O cells shows the serological differentiation of A1 and A2 cells is partly due to a quantitative influence. The A1 erythrocytes have high receptor densities and form multiple bonds with the low-affinity Anti- A1 antibodies and agglutinating the cells (Makela, Ruoslathi, & Ehnholm, 1969). However, the low antigen density of the A2 red cells permits a bond RBC which is inadequate for agglutination. This is evident by the action of the Dolichos lectin which reacts with only erythrocytes that have high densities of the A determinants (Furukawa, Mattes, & Lloyd, 1985). Table 3: Some characteristics of A1- and A2-transferases Adopted from Schenkel-Brunner H., 2000 A1 transferase A2 transferase pH optimum (serum enzyme) 5.5-6.5 7-8 Isoelectric point (serum enzyme) 9-10 6-7 (Enzyme from cyst fluid) 9.5-10 9.5-10 Km (mM) (serum enzyme) 2’-FL pH 5.8 1.15 pH 7.2 0.1 3.6 LNF-1 pH 5.8 3.6 pH 7.2 0.7 5.6 2'-FL = 2'-Fucosyllactose, LNF-1 =Lacto·N-fucopentase I Mn2 + ions are usually required to convert O cells to A by GTA in vitro. GTA1 is still active when Mn2+ is replaced with Mg2+ but GTA2 does not remain active (Schachter, Michaels, Tilley, Crookston, & Crookston, 1973). The A1 enzyme is capable of converting A2 cells to A1 42 University of Ghana http://ugspace.ug.edu.gh phenotype but the A2 enzyme only does so after prolonged incubation (H. Schenkel-Brunner & Tuppy, 1973). Both GTA1 and GTA2 are equally specific for low molecular weight acceptors. These enzymes also catalyze the synthesis of the same A determinant (Baldus et al., 1996). The activity of GTA1 is 5-10 times higher than that from GTA2 at pH 5.5 (Schachter et al., 1971). The optimum pH for serum GTA1 and GTA2 are respectively 5.6 and 7-8 (Schachter et al., 1973). The Km value of the less efficient GTA2 is about 10 times higher than that of the more efficient GTA1 at pH 7.2 (Schachter et al., 1973). 2.7.4 A1 and A2 Determinants The discovery of A subgroups sparked a lot of debate as to whether their difference is structural or as a result of the number of A determinants. A lot of work went into determining the differences between these determinants. A2 cells remove all antibodies in B serum when they are repeatedly used to adsorb Anti-A1 from the B serum, suggesting only a quantitative difference (Lattes & Cavazzuti, 1924 ). However, the production of Anti-A1 by some A2 and A2B individuals also suggests the absence of a determinant on A2 that is present on A1 cells (Landsteiner & Witt, 1926 ). An A2 red blood cell contains only one fifth (2.2 × 10 5) of the total A1 antigen sites (table 4) (Schenkel-Brunner, 2000). 43 University of Ghana http://ugspace.ug.edu.gh Table 4: Number of antigen sites on red cells of various ABO groups Modified from Klein & Anstee (2006) Antigen Phenotype Number of A antigen sites (x103) A1 adult 810 – 1170 A1 newborn 250-370 A A2 adult 240-290 A2 newborn 140 A1B adult 460- 850 AB A1B newborn 220 A2B adult 140 Higher dilutions of anti-A cells can agglutinate A1 cells, something that is not likely with A2 cells. Most of the red cells from A2 individuals fluoresce faintly in the presence of fluorescent Dolichos lectin while strong fluorescence is seen in few cells. A different result appears in the presence of A1 red cells, where majority of cells give strong florescence whiles only a few fluoresce faintly (Rochant, Tonthat, Henri, Titeux, & Dreyfus, 1976). This may be the reason for the usually observed mixed field reaction with anti-A1 lectin. The precise biochemical basis to A1 and A2 is quiet controversial. A1 red cells appear to have repetitive Type 3A and Type 4A glycolipids whiles A2 red cells either do not have both Type 3A and Type 4A glycolipid or lack only Type 4A (H. Clausen, S. Levery, E. Nudelman, M. Baldwin, & S. Hakomori, 1986; Svensson, Rydberg, de Mattos, & Henry, 2009). Studies show 44 University of Ghana http://ugspace.ug.edu.gh that A2 red cell membranes contain abundant quantities of Type 3A glycolipid (Svensson et al., 2009). This brings to the consideration that the abundance of type 4A glycolipid on A1 red cell membranes which is essentially lacking on A2 red cell membranes differentiates A1 and A2 phenotypes. GTA2 is therefore unable to utilize Type 4H as its acceptor substrate, probably due to the extension of the A2 gene by 21 amino acids (fig 7). Anti-A1 is therefore probably either specific for or has high affinity or preference for Type 4A structures. The Dolichos lectin only detects sufficient quantities of GalNAc so its use as a reagent only depends on the quantitative difference between A1 and A2. 2.7.5 Other weaker subgroups of A Numerous weaker subgroups of A with reduced amounts of A substance on the surfaces of their RBCs are known. These subtypes which are usually termed Aw (for weak expression of A) show enhanced or normal expression of H-antigen and include Ax, A3 Aend, Am etc (Geoff Daniels, 2013). Just as in the case of A2, most weaker subgroups occur as a result of rare alleles being inherited on the ABO locus. Most of the rare alleles result from missense mutations at exon 6 and 7. Most of these weaker subgroups are not agglutinated by Anti-A reagents (Table 5) due to their weaker expression of A and are wrongfully typed as O. 2.7.5.1 A3 Incubation of antisera A or A,B with A3 red cells produces a characteristic mixed field agglutination (Friedenreich, 1936). Serum from blood group A3 individuals occasionally 45 University of Ghana http://ugspace.ug.edu.gh contain anti-A1 antibodies. Antigen A3 can be detected in two out of every 180,000 Canadians and 9 out of 150,000 French group A donors (Reed, 1964). 3-4% of A3 RBCs have enough antigen sites to allow agglutination with anti-A2 (R. Oriol et al., 1981) and react at optimum pH of 6-7 (Oguchi, Kawaguchi, Suzuta, & Osawa, 1978) A3 cells have relatively lower H-antigens on RBC than A1 and A2 cells (Lau et al., 1990). These cells also have 40,600 - 118,000 A sites per RBC (Cartron, 1976). There are 3 types of A3 based on enzyme characteristics; a. Have transferase with high activity i.e. close to 50% of that of A1 and optimum pH of 6. b. Have weakly active transferase with 1-6% of A1 activity at optimum pH of 7. c. Have no enzyme activity detected (Schenkel-Brunner, 2000). 2.7.5.2 Ax RBC’s from Ax individuals are not agglutinated by anti-A1 lectin but show agglutination with anti-AB without any mixed field reaction (table 5). Ax serum mostly contains anti-A1 and on some occasions, contain antibodies that are capable of agglutinating A1 and A2 cells (Le Pendu, Lambert, Gerard, et al., 1986). Ax is a heterogenous phenotype and saliva from individuals with this phenotype contains traces of A in addition to H substance (Le Pendu, Lemieux, Lambert, Dalix, & Oriol, 1982). GTA is not easily detected on Ax serum or RBC membrane (Larson & Samuelsson, 1980). Sera from individuals with Ax phenotype are known to have low H- transferase activity (Watkins, 1978). Red cells from Ax individual have 1400-10000 A sites per RBC (Cartron, 1976). 46 University of Ghana http://ugspace.ug.edu.gh 2.7.5.3 Am Am cells absorb Anti-A (Schenkel-Brunner, 2000) but Anti-A1 or A,B does not agglutinate them (Hrubisko, Calkovska, Mergancova, & Gallova, 1966). With about 1200 A sites per RBC (Cartron, 1976), less than 5% of cells express A antigen and majority do not produce any reaction with anti-A serum (Heier, et al., 1994). Although serum from Am individuals does not contain anti-A, anti-A1 has been detected in some cases (Schenkel-Brunner, 2000). With an optimum pH of 6.0, Am transferase is about 30-50% of that of A1 with very weak activity (Cartron, Badet, Mulet, & Salmon, 1978). Am erythrocytes can also be converted to A by A1 transferase (Cartron, Gerbal, Badet, Ropars, & Salmon, 1975) and its own transferase. It is believed that a gene independent of A gene, controls the synthesis of Am antigen on RBC (Weiner et al., 1957). 2.7.5.4 Ay Cells from group Ay individuals have about 1200 A sites (Cartron, 1976) and are not agglutinated by Anti-A sera (Dodd & Gilbey, 1957). Serum from Ay individuals lack anti-A whiles saliva shows weak and distinct A activity. Glycosyltransferase A is detected in trace amounts in the serum of Ay individuals. Both H and A substances are detected in the salivary secretions of Ay secretors with A substance present in lower than normal quantities (Cartron et al., 1975). Very sensitive methods are required to detect Ay transferase activity (Koscielak, Lenkiewicz, Zielenski, & Seyfried, 1986). The Ay phenotype can be observed among siblings due to its recessive inheritance. This phenotype occurs within a family as a result of a germline mutation of the A gene (Garratty, Glynn, & McEntire, 2004). 47 University of Ghana http://ugspace.ug.edu.gh Table 5: Characteristics of some weak ABO Phenotypes REAGENTS ANTIBODIES IN SERUM OTHER phenotype Anti-A Anti-B Anti-AB Anti-H Anti-A Anti-B Anti-A1 Substances Presence of A Number of present in saliva of transferase in antigen sites secretors serum RBC X103 A3 ++mf 0 ++mf 3+ no yes sometimes A,H sometimes 35 Ax wk/0 0 2+ 4+ 0/wk yes Almost H Rarely 5 always Aend wk mf 0 wk mf 4+ no yes sometimes H no 3.5 *Am 0/wk 0 0/+ 4+ no yes no A,H yes 1 *Ay 0 0 0 4+ No Yes No A,H trace 1 *Ael 0 0 0 4+ some yes yes H No 0.7 Adapted from Harmening et al., 2012 *A specificity demonstrated only by absorption/elution procedures 0 = negative; mf = mixed-field agglutination; wk = weak 48 University of Ghana http://ugspace.ug.edu.gh 2.7.5.5 Aend Aend red cells show weak reaction with antisera -A and A,B resulting in mixed field agglutinations. About 70,000-140,000 A sites are found on the agglutinated RBCs while no A site is detected on the non-agglutinated Aend cells (Cartron, Reyes, Gourdin, Garretta, & Salmon, 1977). A activity is not detected in salivary secretions of ABH secretors with Aend blood group. Anti-A and Anti-A1 occasionally appear in the serum of Aend individuals, both reacting only at low temperature (Schenkel-Brunner, 2000). No A-transferase activity is present in serum from Aend individuals (Cartron, 1976). The Aend phenotype is believed to be a variant of the A2 gene (Olsson & Chester, 1996b). 2.7.6 Interactions between A and B genes 2.7.6.1 Gene Suppression Several studies attribute the increase in A2B genotype among the AB population to suppression of the A gene when combined with a strong B gene (Voak, Lodge, & Reed, 1970; Yoshida, 1983). It is known that some A1B individuals can type serologically as A2B (although these individuals have the A1 but not the A2 transferase) due to suppression of their A gene by a strong B gene (Voak et al., 1970; Yoshida, 1983). Serum transferase analysis among these individuals reveals the presence of a B transferase with considerably higher than normal B activity (Yoshida, 1983). This highly active B transferase is known to utilize most of the available H antigen sites, hence preventing the A1 transferase from producing more A antigens (Daniels, 2013). As a result, red cells from A2B1, A2B2, A1B2, and AintB 2 always type as A2B although they do not have the A2 transferase (Yoshida, 1983). 49 University of Ghana http://ugspace.ug.edu.gh 2.7.6.2 Allelic Enhancement There is an enhanced expression of weak A genes on some occasions when they are combined with strong B genes in an AB individual (Helmut Schenkel-Brunner, 2000). In such cases, a weak A antigen (eg Ael) may be expressed serologically as A2 due to a rise in antigen site density. This allelic enhancement is believed to result from the dimerization of the glycosyltransferase through the stem region which stabilizes the structures and thus increasing the specificity and efficiency of the weak enzyme (Hassinen, Rivinoja, Kauppila, & Kellokumpu, 2010). Studies across some families (Fisher & Cahan, 1962) showed that some Ax alleles in AxB individuals were expressed as A2 due to the enhancement of antigen Ax in the AxB. This enhanced serological activity has been attributed to a rise in antigen site density of the Ax individuals from an average 11,200 A sites to 96,000 in the AxB (Salmon & Cartron, 1977). 50 University of Ghana http://ugspace.ug.edu.gh 2.8 H deficient phenotypes These consist of rare heterogeneous ABO phenotypes with RBCs completely or partially deficient in antigen-H. This was first described in 1957 (Anstee, 1990). The occurrence of H- deficient phenotype does not depend on the secretor status of the individual. “Red cell H- deficient phenotypes result from homozygosity or compound heterozygosity for mutations in FUT1 gene that totally or partially inactivate the H-transferase in red cell progenitors” (Geoff Daniels, 2013). Twenty-three FUT1 alleles are known to be responsible for weakened H- transferase activity whiles nineteen are known to produce inactive enzymes (Daniels, 2013). Most of these alleles are characterized by substitutions in nucleotides which lead to exchanges in amino acid or creation of premature stop codons. Some alleles also show deletion of single bases within the coding region leading to changes in the reading frame and encoding truncated proteins. The H deficient phenotypes have been grouped into the Bombay phenotype which comprises all H-deficient ABH-non-secretors (se/se, h/h genotype) and the Para-Bombay phenotype which includes all H-deficient ABH-secretors (genotype h/h, Se) (Clausen & Hakomori, 1989). There is also an extremely rare H-deficient type (Hm) which is not controlled by FUT1 (Schenkel-Brunner, 2000). The Bombay phenotype occurs in 1 in 7,600 (Bhatia & Sathe, 1974) people in Marathi (Near Bombay, India) and extremely rare in Europe (1 in 312,081 cases) (Wagner & Flegel, 1997). The RBCs of Bombay (Oh) individuals show no agglutinations when reacted with anti-A, anti- B and anti-H reagents in routine blood grouping methods (Bhatia, 1977). However, pretreatment of cells with papain (Dodd & lincoln, 1978) and serum adsorption/elution techniques (Lanset, Ropartz, Rousseau, Guerbet, & Salmon, 1966) helps to detect cryptic ABH substances on red cells of Bombay individuals carrying A and/or B genes. The designations O ah , O b h and O ab h indicate the presence of A and/or B antigens on the Bombay phenotype. Bombay individuals do not secrete ABH antigens but have anti-A, anti-B and anti-H antibodies 51 University of Ghana http://ugspace.ug.edu.gh which react with both type 1 and type 2 chain substances (Le Pendu, Lambert, & Samuelsson, 1986). The H transferase for red cells and the H transferase for secretions and other cells are both absent in the Bombay Phenotypes. Although Bombay individuals have no serum H transferase, A and B transferases with normal activity can be detected in their serum (C. Race & Watkins, 1972). “Since Oh individuals represent ABH-non-secretors (genotype h/h se/se), they are not able to synthesize H determinants. Consequently, even in the presence of A and/or B gene products, A or B substances cannot be formed on erythrocytes and in secretions.” (Schenkel-Brunner, 2000) Para-Bombay (genotype h/h Se) phenotype is very rare and individuals with this phenotype lack the H-transferase, hence lack ABH antigen expression on their red cell membranes. They however secrete ABH antigens with normal activity in their saliva. In some few occasions, red cells from para-Bombay individuals show a weak ABH activity due to the presence of a slightly active H allele (Schenkel-Brunner, 2000). Weak anti-H can however be detected in the serum of para-Bombay individuals (Mak et al., 1996). Molecular studies in H allele of H-deficient individuals show a lot of mutations. Majority of these mutations cause the production of silent alleles which encodes inactive glycosyltransferases when transcribed (Schenkel-Brunner, 2000). Other alleles encode enzymes which have weak activity leading to weak antigen H expression in Bombay and para- Bombay individuals. Most of the mutations result from nucleotide substitutions in some alleles leading to changes in amino acid. Other alleles also show base deletions within the coding region leading to changes in the reading frame and encoding of truncated proteins (Schenkel- Brunner, 2000). 52 University of Ghana http://ugspace.ug.edu.gh 2.9 ABH ANTIBODIES ABH antibodies may be IgM, IgG or IgA but all the three antibody classes may be found in some sera (Klein & Anstee, 2006). These cold-reacting antibodies are predominantly IgM and they activate complement. IgM Anti-A and anti-B occur naturally against missing ABH antigens. Persons with blood group A naturally possess antibody B, whiles group B individuals have antibody A (Karl Landsteiner, 1900). Anti-H also occurs naturally in the sera of blood group A1 and A1B non-secretors. ABH antibodies develop 4–6 months after birth, reaching adult titers by 5–10 years and declining with increasing age (Cooling, 2014). ABO antibodies may sometimes be undetected in the serum of the elderly and may lead to discrepancies in the reverse grouping (Harmening et al., 2012). Environmental factors play a major role in antibody levels in an individual (Nijenhuis & Bratlie, 1962). Anti-A and anti-B formation are likely stimulated in the neonatal period by exposure to environmental antigens in the digestive tract (Wiener, 1951). Gut bacteria have also been cited to stimulate ABO antigen production in the neonatal period (Georg F. Springer, 1971). This is due to the expression of ABH-like structures on lipopolysaccharides of many bacteria which can cross-react with lectins and ABH antigens (Georg F. Springer, 1971). Immune reactions such as incompatible transfusion reaction and immunization by pregnancy change the characteristics of IgG ABH antibodies. These changes which are detected serologically include a rise in titre and strength of agglutination, elevated haemolytic activity and greater activity at 37oC (Daniels, 2013). These cold-reacting antibodies react preferentially at room temperature (20oC-24oC) but activate complement at 37oC. ABH antibodies can be detected in body fluids including saliva, tears and milk (Putkonen, 1930). Blood group B individuals have two main serum antibodies, anti-A and anti-A1. Some individuals with blood group A2, A2B, Ax and A3 may also have anti-A1 in their serum (Landsteiner & Witt, 1926; Le Pendu, et al., 1986). It is estimated that 1-8% of A2 and close to 53 University of Ghana http://ugspace.ug.edu.gh 35% of A2B individuals have anti-A1 in their serum (Cooling, 2014). This antibody causes haemolytic transfusion reaction and HDFN in some patients (Chaudhari et al., 2008). Serum from group O individuals contains anti-A,B, a cross-reacting antibody which can detect structures common to both A and B determinants (Kabat, 1956; Owen, 1954). This antibody is usually IgM but may also be IgG or IgA (Klein & Anstee, 2006). Although they may have rare occurrence, ABO auto-antibodies (auto anti-A and auto anti-B) can cause auto-immune haemolytic anaemia, (AIHA) which can end in kidney failure (Sokol, Booker, Stamps, & Windle, 1995). Auto-anti-A1 has also been reported in AIHA (Sokol et al., 1995). Most individuals have serum Auto-anti-H due to the presence of antigen H to some extent on all red cells. 2.9.1 Clinical Significance of ABH Antibodies 2.9.1.1 HDFN Pregnant mothers carrying ABO-incompatible foetus may develop antibodies that may cross the placenta and destroy the foetal erythrocytes. Severe ABO HDFN is relatively rare with an incidence of about 0.04% (Cooling, 2010). The inability of foetal red cells to cross the placenta and poor expression of ABO antigens on foetal erythrocytes have been cited as the reasons for the low incidence of severe HDFN (Cooling, 2016). “HDFN due to ABO incompatibility is mostly observed in group O mothers, especially with a history of a prior, non-O pregnancy and immune-stimulated, anti-A/B IgG antibodies” (Cooling, 2016). These mothers have anti-A and anti-B which destroy the foetal erythrocytes before they cause trouble. Studies show that HFDN caused by anti-A occurs once in every 150 births whiles lower frequencies are observed in anti-B (Murray & Roberts, 2007). 54 University of Ghana http://ugspace.ug.edu.gh 2.9.1.2 Transfusion Transfusion of ABO-incompatible cells may lead to antibody destruction of donor cells which may lead to Haemolytic transfusion reaction (HTR). Antibodies A and B are IgM antibodies that are capable of causing HTR in patients (Cooling, 2016). HTR may lead to Disseminated Intravascular Coagulation (DIC), renal failure and death (Daniels, 2013). Allo-anti-H in Bombay and para-Bombay individuals is also capable of causing HTR. Autoantibodies with I and ABH activity are frequent in cold autoimmune haemolytic anaemia (Cooling, 2016). 2.9.1.3 Transplantation Selected organs for transplantation are generally ABO compatible with the recipient. This helps to prevent hyperacute rejection of the organ resulting from immune recognition ABO antigens on the tissues. Subgroup screening is now done for group A organ donors (A2, Aweak) (Hurst, Sajjad, & Elster, 2010). Antigen A is relatively absent on endothelium and epithelium, hence solid organ transplantations from group A donors to non-A recipients can be executed successfully (Hurst et al., 2010). The presence of memory B cells and antibodies in adult tissues poses a huge challenge to adult transplantation. ABO-incompatible organs can be transplanted in neonates and infants with minimal risk of rejection because their immune system is naive (West & Platt, 2010). Unlike adults, neonates do not have ABO antibodies and memory B cells hence the use of special procedures required in adult transplantations is eliminated in neonates (Grasemann, de Perot, & Bendiak, 2012). 55 University of Ghana http://ugspace.ug.edu.gh 2.10 METHODS OF TESTING FOR ABO BLOOD GROUP Many techniques are available for blood typing which differ from each other in their time of operation, sensitivity, throughput analysis, reagents and equipment requirement. The classical methods are mostly used routinely and almost all of them involve the formation of agglutinates. Although some of the classical methods are relatively not very sensitive, they are still very important in ABO grouping. These methods include slide, tube, gel and microplate methods. 2.10.1 Slide Method Although it is the least sensitive blood grouping test, it is of high importance especially during emergency conditions because of its quick time for results. It involves agglutination of blood cells on a slide when they react with an antisera. One drop each of antisera A, B and D is dispensed on clean glass slides or tile. A drop of well- mixed 35-45% red cell suspension (cells suspended in their own serum or plasma) (Regan, 2016) is added to the reagents on each slide which is then mixed with an applicator. After continuously tilting the slide gently back and forth for two minutes, the result is read and recorded. The presence of agglutination on any of the slides indicates the presence of antigens to the antibody added. The agglutination is graded using table 6. A further test to determine blood group A2 is done when antigen A is detected. This time, a drop of anti-A1 reagent is mixed with a drop of red cell suspension on a slide and the results recorded after continuously tilting the slide gently back and forth for two minutes. The absence of agglutination indicates the presence of blood group A2. 56 University of Ghana http://ugspace.ug.edu.gh Table 6: Grading of agglutination Modified from Cooling (2014) Macroscopic findings Designation One solid agglutinate 4+ Several large agglutinates 3+ Medium-size agglutinates, clear background 2+ Small agglutinates, turbid background 1+ Mixtures of agglutinated and unagglutinated red cells (mixed field) mf Very small agglutinates, turbid background 1w Barely visible agglutination, turbid background. Confirmed w+ or +/- microscopically No agglutination. Confirmed microscopically 0 2.10.2 Standard Tube method The standard tube method usually involves forward and reverse grouping and requires incubation at room temperature for one hour. Forward grouping utilizes reagent antibodies to detect antigens on RBC surfaces (Storry & Olsson, 2009). With reverse grouping, known red cell antigens are used for the detection of serum antibodies of individuals. 2.10.2.1 Forward grouping A drop each of monoclonal Anti-A, B and D is dispensed into each tube (12x75mm) labelled A, B and D respectfully followed by a drop of 2-5% saline washed RBC suspension. The contents of the tube are mixed and incubated for 1 hour at room temperature after which the cell button is gently re-suspended and observed for agglutination macroscopically. Microscopic examination is done to confirm all negative tests. The presence of agglutination 57 University of Ghana http://ugspace.ug.edu.gh in any of the tubes indicates the presence of that particular antigen. All agglutination reactions are graded using the table below. When antigen A is detected, a drop of anti-A1 lectin is mixed with the RBC suspension in a clean test tube (12x75mm) and taken through the same procedure as above. The absence of agglutination indicates the presence of blood group A2. 2.10.2.2 Reverse grouping A drop of test serum is dispensed into each clean test tube (12x75mm) labelled A and B after which a drop of known A and B cell suspension are respectively added. The contents of the tube are mixed and incubated for 1 hour at room temperature after which the cell button is gently re-suspended and observed for agglutination macroscopically. Microscopic examination is done to confirm all negative tests. The presence of agglutination in any of the tubes indicates the presence of the corresponding antibody. The agglutination is graded using table 6. The same procedure is employed for testing for blood group A2 after antibody B is detected but A1 and A2 cells are used against the serum of the blood group A individual. 2.10.3 Microplate method This technology is faster and more sensitive than the standard tube method and also has some level of automation. The advantage of microplate technology is its fast response, low reagent volumes and high throughput analysis. Unlike the standard tube method, incubation of the plate is done at room temperature for 5 to 10 minutes after which they are centrifuged. Both forward and reverse blood groupings can be done with the microplate method. This technique also allows agglutinations to be read by automated read-out devices. 58 University of Ghana http://ugspace.ug.edu.gh 2.10.4 Gel Microcolumn This is the commonest technique for blood grouping in the UK as over 80% of their laboratories employ this technique in their routine blood grouping (Regan, 2016). “The principle of microcolumn tests is the separation of agglutinated from non-agglutinated red cells by centrifugation through a miniature filtration column” (G. Daniels, Contreras, & Allard, 2016) The columns with their reservoir usually arrive in the form of cards or cassettes depending on the manufacturer. Small quantities of serum and cells are mixed in the upper reservoir of the narrow column that contains a Dextran gel. The test containing cassettes are spun in a centrifuge after following the required incubation conditions by the manufacturer. The red cells leave their suspension medium and enter the column after centrifugation. All the agglutinated RBCs are trapped in the upper portion of the column whiles unagglutinated RBCs form a pellet at the lower portion. 2.10.5 Molecular determination of blood groups using PCR Serological techniques were solely used in blood typing until the introduction of molecular ABO blood group determination in 1990 (F. Yamamoto et al., 1990). Haemagglutination was the gold standard in ABO grouping for over a century until molecular methods were developed. Despite its benefits, haemagglutination methods for blood grouping have several limitations especially in recently transfused and multiple transfused or direct antiglobulin test DAT positive patients. Molecular methods are very important in screening fetal DNA samples, screening for rare blood groups and determining blood group polymorphisms among a population. Molecular techniques become very important in situations where serologic testing is impossible or inconclusive. The molecular methods were developed to overcome the shortfalls in the serological techniques in blood grouping. Molecular blood typing has become 59 University of Ghana http://ugspace.ug.edu.gh easier due to the successful cloning and sequencing of genes encoding twenty-nine blood group systems. The molecular bases associated with most antigens have also been determined, hence making it possible to predict ABO blood groups using molecular methods with high accuracy. Despite their importance, molecular methods in blood grouping come with their own challenges. The biggest challenge with molecular blood group determination is the high cost of testing. Also, few samples can be tested at a time hence making it unsuitable for large scale genotyping. They are therefore mostly used in reference laboratories. Several methods for ABO genotyping exist whiles many others are still in the development stages. The currently available methods include polymerase chain reaction (PCR)- restriction fragment length polymorphism (PCR-RFLP), allele-specific-PCR (PCR-ASP), PCR- Amplified product length polymorphism (APLP), multiplex PCR, real-time PCR, DNA sequencing, reverse transcriptase-quantitative (RT-q) PCR and DNA chip (Sensabaugh, 1981). The fluidics methods for blood grouping which are under development include matrix-assisted laser desorption ⁄ionization time-of-flight mass spectrometry (MALDI-TOFMS) and mini- sequencing (Monteiro et al., 2011). Some of the PCR based methods for blood typing have been discussed below; PCR-RFLP This involves a PCR-amplification of relevant sequences of blood group genes followed by restriction fragment length polymorphism. This test was designed based on the introduction or loss of restriction sites by SNP of interest. To differentiate the alleles, restriction enzymes are used to digest the products after PCR amplification. These highly specific restriction enzymes cleave only a unique sequence of nucleotides. The different alleles are clearly distinguished by electrophoresis of the PCR products amplified with allele-specific primers. Known controls are 60 University of Ghana http://ugspace.ug.edu.gh usually tested alongside the test samples. By analyzing the electrophoresis patterns, ABO genotyping is conclusively accomplished (Sensabaugh, 1981). Allele-specific-PCR The method uses the molecular size of allele-specific amplification products to identify the different ABO genotypes. Two reactions are set up for each DNA sample for AS-PCR assay; one with a gene-specific primer and the other with allele-specific primer. Annealing only occurs when the sequences present are complementary to each other and is followed by amplification. Annealing does not occur in non-complementary sequences, hence no amplification occurs in such sequences. The amplified products of the two reactions can be observed using gel electrophoresis for the identification of ABO antigens (Franchini & Liumbruno, 2013). Multiplex PCR Unlike PCR-RFLP, this method allows the use of many primer pairs to simultaneously amplify a lot of target alleles in one reaction. This method saves time and reagents due to the reduction in the number of assays performed. It is however technically challenging procedure and a limited quantity of primer pairs can be combined at a time. 2.10.6 Other methods Modern methods of blood grouping which include Natural antibody sensors are now available. These methods are highly sensitive but relatively expensive due to the equipment involved, hence they are mostly confined to research laboratories. 61 University of Ghana http://ugspace.ug.edu.gh CHAPTER THREE 3.0 METHODS 3.1 STUDY DESIGN This was a cross-sectional study, conducted at the sites operated by the Southern Area Blood Centre (SABC), National Blood Service (NBS), Ghana. 3.2 STUDY SITE DESCRIPTION The SABC has several donation sites in major hospitals within the Accra metropolis. The headquarters of the SABC is located on the premises of Korle-Bu Teaching Hospital, which is the largest hospital in Ghana. The SABC houses the largest blood storage area in Ghana and also serves as the headquarters of the national blood service. The SABC building is an ultra- modern complex made of various well-structured offices and departments which include laboratory, Research, IT and Statistics. Various professionals including medical officers, laboratory scientists, nurses, statisticians and IT officers work at the Headquarters of the SABC. SABC averagely receives about 2,783 blood donations per month (SABC, 2016). The laboratory department houses modern blood bank equipment such as blood grouping and automated cross-match analyzers. It is made up of various units including microbiology, cross- matching and serology. The laboratory adopts standard quality control measures, making their output very reliable. 3.3 STUDY POPULATION The study population included individuals who passed the pre-donation screening and were accepted to donate blood at any of the sites operated by the Southern Area Blood Centre. 62 University of Ghana http://ugspace.ug.edu.gh 3.4 SAMPLE SIZE The sample size was calculated based on data from a previous study in Liberia which revealed the prevalence of blood groups A2 and A2B among groups A and AB individuals respectively as 19.7% and 32.6% (Livingstone et al., 1960). The formula described by Naing, Winn, and Rusli (2006) was used for the calculation. 𝑍2𝑃(1−𝑃) That is, n= 𝑑2 Where; Ƶ= Reliability Co-efficient with a 95% confidence interval P= Proportion variance from previous data, q=1-p d= the desired or required size of standard error allowed The Value of a 95% confidence interval is 1.96 (Z=1.96) From the above calculations, 242 blood group A individuals were recruited. However, due to the low proportion of blood group AB individuals in Ghana (3.7%) all group AB individuals encountered during the study period were included in thi study. 3.5 INCLUSION CRITERIA Consented individuals with blood group A and AB who successfully donated their blood to the blood centre within the study period. Individuals who later withdrew their consent were excluded from the study. 63 University of Ghana http://ugspace.ug.edu.gh 3.6 SAMPLE AND DATA COLLECTION Research personnel were stationed at the blood donation rooms of the various sites of the SABC (including mobile sessions). This personnel approached and gave a thorough explanation of the study to blood donors who passed the pre-donation screening and were ready to donate their blood. These donors were taken through the consent form including an explanation of the risks and benefits of participating in the study. Eligible individuals who consented to be enrolled in the study were made to sign or thumbprint on the consent form. Prior to the donation, questionnaires were also administered to the consented participants in the donation room on the basis of acquiring information such as their demographic data and donation history such as whether they were first time or retained donors. The research personnel read and explained questions on the form to the participant whiles recording their responses in the spaces provided. Two millilitres of the donated blood from consented individuals were taken from their blood bags through the sample collection port, using Vacutainer® EDTA tubes (12x75mm). This was done by gently pushing the Vacutainer EDTA tube inside the sample collection port of the blood bag and allowing the blood to flow through the evacuated tube. The sample was then labelled with the sample ID and collection date by a well-trained phlebotomist. The blood samples collected were used for ABO blood group screening using anti-A and anti- B monoclonal reagents (Lorne Laboratories Ltd, UK). Blood group A and AB samples were used for blood group A2 and further Anti-A1 screening. 64 University of Ghana http://ugspace.ug.edu.gh 3.7 LABORATORY ANALYSIS Samples were analyzed immediately or within 24 hours of collection during which it was stored at 2-8oC (Reagent manufacturer’s instruction). 1ml of the well-mixed sample was taken into a plain tube (12x75mm) labelled with the same sample ID. The cells in the new tube were then washed three times with normal saline (0.9%) after which a 3-5% cell suspension was prepared from them. The suspension was prepared by adding 0.2mls of the washed packed cells into a test tube containing 2.8mls of normal saline (Appendix I). The remaining sample in the parent tube was centrifuged at 3500rpm for two minutes and the serum separated from the cells into a new plain tube (12x75mm). Using the tube method, forward and reverse ABO blood grouping was done with the cells and serum using anti-A and B monoclonal reagents. The forward and reverse grouping procedures are described in section 3.8. Interpretation of the ABO grouping was done using table 6. Samples which came out as blood group A and AB were used against Anti-A1 lectin for the blood group A2 screening as described in section 3.8. The reaction of the blood group A2 serum with commercial A1 cells helped to detect the number of A2 individuals with anti-A1 antibodies. The interpretation of blood group A2 screening was done using table 8. A doubling dilution of the A2 serum was made to determine the titre of anti-A1 antibodies present. The agglutinations observed in all the reactions were graded as described in table 6. Controls were done alongside each batch of tests using known cells. 65 University of Ghana http://ugspace.ug.edu.gh 3.7.1 ABO BLOOD GROUPING (STANDARD TUBE METHOD) 3.7.1.1 Forward grouping Three rows of clean labelled test tubes (12x75mm) were arranged in a rack for each batch of tests (table 7). The last three tubes in each row were used for control. A drop each of monoclonal anti-A, B and O sera was respectively dispensed into each tube in the 1st, 2nd and 3rd rows of labelled tubes. This was followed by a drop of labelled test cells (3-5% red cell suspension) into their respective tubes. Control cells (A1, B and O cells) were respectively used in place of the test cells in the last three tubes in each row. The contents of the test and control tubes were mixed and incubated for one hour at room temperature after which the cell button was gently re-suspended and observed for agglutination macroscopically. Microscopic examination was done for cell suspensions with no visible agglutination after the macroscopic examination. The presence of agglutination in any of the tubes indicated the presence of the corresponding antigen. All agglutination reactions were graded using table 6. Table 7: ABO forward blood grouping Reagent Tests cells Control cells Sera 1 2 3 4 5 6 7 A B O Anti-A Anti-B O sera 66 University of Ghana http://ugspace.ug.edu.gh 3.7.1.2 Reverse grouping Three rows each of labelled clean test tubes (12x75mm) were arranged in a rack for each batch of test and the last three tubes in each row were used for control (Table 8). A drop each of test sera was dispensed into their respective labelled test tubes. Control sera (Anti-A, anti-B and Anti-A,B) was respectively used in place of the test sera in the last three tubes in each row. This was followed by a drop of reagent A, B and O cell suspension (3-5%) into each tube in the 1st, 2nd and 3rd rows of labelled tubes respectively. The contents of both test and control tubes were mixed and incubated for one hour at room temperature after which the cell button was gently re-suspended and observed for agglutination macroscopically. Microscopic examination was done for cell suspensions with no visible agglutination after the macroscopic examination. The presence of agglutination in any of the tubes indicated the presence of the corresponding antibody to the cells. Table 8. ABO reverse blood grouping Reagent Test sera Control Sera Cells 1 2 3 4 5 6 7 Anti-A Anti -B Anti-A,B A1 cells B cells O cells 67 University of Ghana http://ugspace.ug.edu.gh 3.7.2 TESTING GROUP A CELLS FOR ABO SUBGROUP A2 After determining the ABO groups of the samples collected from the donors, blood group A and AB samples were selected to determine whether they belonged to subgroup A2. 3.7.2.1 Forward grouping Rows of clean labelled test tubes (12x75mm) were arranged in a rack for each batch of test and the last three tubes in each row were used for control. The anti-A1 lectin was dispensed, one drop in a tube, in a row of labelled tubes. This was followed by a drop of labelled test cells (3- 5% suspension) into their respective labelled tubes. Control cells (A1, A2 and O cells) were used in place of the test cells in the last three tubes in each row. The contents of both test and control tubes were mixed and incubated at room temperature for one hour after which the cell button was gently re-suspended and observed for agglutination macroscopically. Microscopic examination was done for cell suspensions with no visible agglutination after the macroscopic examination. The controls were read alongside the tests using table 8. The absence of agglutination in the test tubes indicates the presence of blood group A2. Table 8: Reaction for A1 and A2 cells Cell Reaction With A subgroup Anti-A Anti-A1 lectin A1 R R A2 R 0 R-agglutination, 0- no agglutination 68 University of Ghana http://ugspace.ug.edu.gh 3.7.3 Testing A2 and A2B serum for anti-A1 All the samples which after typing, came out as blood group A2 and A2B were selected and further tested for the presence of anti-A1 antibody in the sera. The A2 sera were dispensed, one drop in a tube, in a row of labelled tubes (12x75mm). A drop of 3-5% A1 red cells was dispensed into each of the tubes. For control, a drop of A1 red cells was dispensed into a tube containing a drop of B serum. This was done for each batch of test. The contents of the tubes were mixed and incubated at room temperature for 1 hour after which the cell button was gently re-suspended and observed for agglutination macroscopically. Microscopic examination was done for cell suspensions with no visible agglutination after the macroscopic examination. The tests were read alongside the control. The presence of agglutination in the tube indicated the presence of anti-A1 in the A2 serum. The agglutination was graded using table 6. 3.7.4 Anti-A1 titre determination (Doubling dilution) After determining the presence of anti-A1, a doubling dilution was done to determine anti-A1 titre in the sample with anti-A1. Ten test tubes (12x75mm) were arranged in a rack and labelled 1 to 10. 0.5mls of saline was dispensed into each of the tubes except tube 1 (neat) whiles 1ml of each blood group A2 serum was dispensed into the tube labelled 1 (neat). Using a graduated pipette, 0.5mls of group A2 serum was transferred from the tube labelled neat into the tube labelled 2 and the contents were thoroughly mixed. 0.5mls of the well-mixed content of the tube labelled 2 was also picked and dispensed into the tube labelled 3 and its contents were thoroughly mixed. This process continued up to the last tube. 0.5mls of the contents of the last tube was discarded. Another set 69 University of Ghana http://ugspace.ug.edu.gh of tubes labelled neat, 1 to 10, were arranged in front of their corresponding tubes. 20ul of the contents of the first tubes were transferred into their corresponding tubes arranged in front of them after which 20ul of commercial A1 cell suspension was added. The content was mixed and incubated for one and a half hours after which the reaction was read macroscopically and microscopically. Grading of the reaction was done as indicated in table 6. The anti-A1 titre was defined by the reciprocal value of the highest dilution that gave 1+ agglutination reaction. 3.8 DATA MANAGEMENT Confidentiality of data obtained for all study participants was ensured. Unique identification numbers were assigned to each person instead of names. All data were coded and doubly entered into a password-protected database in Microsoft Access. Data was then transferred into SPSS version 22 for analysis. 3.9 DATA ANALYSIS Data obtained was entered into Statistical Package for Social Sciences (SPSS) version 22. Ages of donors were summarized as mean and standard deviation while categorical variables such as donor type and residence were summarized as frequencies and percentages. To compare the prevalence of blood group A2 among blood group A and AB individuals, the Z test of proportion was used. The Z-test of proportion was also used to compare the prevalence of Anti- A1 among blood groups A2 and A2B. All tests were two-sided and a P-value less than 0.05 was considered significant. 70 University of Ghana http://ugspace.ug.edu.gh 3.10 ETHICAL ISSUES Ethical approval was sought from the Ethical and Protocol Review Committee of the College of Health Sciences, University of Ghana. Approval was also sought from the management of SABC. Written informed consent was also sought from the study participants. Study identification numbers were assigned to all respondents and data analysis was done based on these numbers without referring to any particular participant’s name. This was password protected and used by only named investigators. Only a summative report of the results of this will be given out. 71 University of Ghana http://ugspace.ug.edu.gh CHAPTER 4 4.0 RESULTS 4.1 Demographic Characteristics Table 1 shows the demographic characteristics of the participants. A total of one thousand, one hundred and forty (1,140) participants (all Ghanaians) were screened among blood donors (voluntary and family replacement donors) who donated their blood at the Southern Area Blood Centre during the study period. The participants comprised 942 (82.6%) males and 198 (17.4%) females, with a mean age of 25.64±6.68 years. The ages of the donors ranged from 16 to 52years. Six hundred and forty-seven 647(56.8%) donors were below 25 years old and only one donor was above 50 years old, representing the least age group. More than half 806 (70.7%) of the donors were first-time donors whiles 334 (29.3%) were retained donors. Majority of these donors 942(82.6%) were males most of whom are residents in the Greater- Accra region, around Achimota, Madina, Adenta, Nima, Ada, Ashaiman, etc. Majority of the blood donors 852(74.7%) involved in this study came voluntarily to donate their blood. Of those who had previous history of blood donation, 295(88.3%), 38(11.4%) and 1(0.3%) of them had donated blood one to five times, six to ten times and above ten times respectively. 72 University of Ghana http://ugspace.ug.edu.gh Table 9: Description of Demographic characteristics of blood donors Variable Frequency Percentage (%) Gender Male 942 82.6 Female 198 17.4 Age (years) Below 25 647 56.7 25 to 34 353 31.0 Above 35 140 12.3 Nationality Ghanaian 1140 100 Residence Central region 180 15.8 Greater Accra 980 84.2 Donor type Voluntary Male 693 60.8 Female 159 14.0 Replacement Male 249 21.8 Female 39 3.4 Donor history First-time donors 806 70.7 Retained donors (RD) 334 29.3 Frequency of donations for RD 1 to 5x 295 88.3 6 to 10x 38 11.4 >10x 1 0.3 RD: Retained donors 73 University of Ghana http://ugspace.ug.edu.gh 4.2 ABO blood grouping results Majority 585 (51.3%) of the donors recruited in the study were blood group O individuals. The number of Blood group A and B individuals obtained were very close, 242 (21.2%) and 268 (23.5%) respectively with group B donors slightly higher than those of A. The least blood group observed was AB 45(3.9%). These are shown in Figure 8 below. ABO blood groups of the 1140 blood donors 60 585(51.3%) 50 40 30 268(23.5%) 242(21.2%) 20 10 45(3.9%) 0 O B A AB ABO blood group Figure 8: ABO blood groups of the 1140 blood donors 74 Percent University of Ghana http://ugspace.ug.edu.gh 4.3 Prevalence of ABO subgroups A2 and A2B among the 1140 donors Figure 9 shows the distribution of A1, A1B, A2 and A2B subgroups among the donors. Out of the 242 blood group A individuals, 192 (79.3%) are A1 whiles 50 (20.7%) are A2. The 45 blood group AB individuals observed were also made up of 20 (44.4%) A1B and 25 (55.6%) A2B individuals. ABO subgroup A1 was the most common subgroup observed among blood group A individuals whiles subgroup A2 was the most common A subgroup among group AB individuals. TOTAL n=1140 (100%) A1 B O A1B n=192 A n=268 n=585 n=20 2 A 16.8% 23.5% 51.3% 1.6% 2 B n=50 n=25 4.4% 2.2% Figure 9: Prevalence of ABO subgroups A2 and A2B among the 1140 blood donors 75 University of Ghana http://ugspace.ug.edu.gh Proportion of blood group A individuals with antigen A2 50,(21%) 192,(79%) Antigen A2 Antigen A1 Figure 10: Proportion of group A individuals with antigen A2 76 University of Ghana http://ugspace.ug.edu.gh With regards to the grading of agglutination, 93.4% of A showed 3+ and 4+, and 95.2% of B showed 3+ and 4+. With the use of Dolichos lectin, the weakest agglutination was 2+ and was seen in only 14 (6.6%) samples. The rest were 3+ and 4+. This is indicative of the absence of weak forms of A. Table 10: Agglutination reactions observed in the various samples Agglutination Strength n(%) 1+ 2+ 3+ 4+ Antigen A reaction with; A cells 1(0.4) 15(6.2) 145(59.9) 81(33.5) AB cells 0 3(6.7) 32(71.1) 10(22.2) Antigen B reaction with; B cells 0 10(3.7) 94(35.1) 164(61.2) AB cells 1(2.2) 4(8.9) 34(75.6) 6(13.3) Dolichos lectin reaction with; A1 cells 0 11(5.7) 52(27.1) 129(67.2) A1B cells 0 3(15) 8(40) 9(45) 77 University of Ghana http://ugspace.ug.edu.gh There is more A2 than A1 with group AB individuals whilst the opposite is observed in group A individuals. The difference between the proportions of antigen A2 produced by group A and AB donors in this study was significant (P<0.05). This has been illustrated in Table 11. Table 11: Difference in antigen A2 among blood group A and AB donors Antigen A2 Blood type Total A2 z-value p-value A 242 50 (20.7) 4.891 0.001* AB 45 25 (55.6) *Significant at 5% 78 University of Ghana http://ugspace.ug.edu.gh 4.7 Prevalence of anti-A1 among study participants Among the study participants, only one anti-A1 antibody was recorded from a group A2 individual with a titre value of 64. This antibody only reacted at room temperature and there was no reaction at 37oC. No anti-A1 antibody was detected from any subgroup A2B individuals. 79 University of Ghana http://ugspace.ug.edu.gh CHAPTER FIVE 5.0 DISCUSSION The ABO blood group is the most widely studied blood group system in the world and its variation across different races and groups of people are well documented. Studies on ABO blood groups has improved our knowledge of its disease association, haemolytic transfusion reactions, population genetic studies, solving paternity disputes among others (Kassahun, Yohannes, & Mebeaselassie, 2015). The advent of subgroups has brought more understanding to the study of ABO blood groups since it has helped to resolve some discrepancies in blood group testing. This study sought to find out the prevalence of A2 and A2B subgroups among blood groups A and AB donors who visit the southern area blood centre in Accra, Ghana. The findings from this study reaffirm the strong dominance of blood group O individuals among the Ghanaian population. The frequencies of ABO blood groups in this study can be presented in the order O>B>A>AB which agrees with findings from earlier studies in Ghana (Kretchy et al., 2017; Smith, Okai, Abaidoo, & Acheampong, 2018). Similar ABO blood group distribution pattern has been observed among people of several black African countries and other Blacks living in America. Among Nigerians, the distribution of ABO blood groups O, B, A and AB are 51.9%, 23.50%, 20.8% and AB 3.2% respectively (Musa, Ndakotsu, Abdul- Aziz, Kilishi, & Aliyu, 2015). ABO blood group antigen distribution among the black population in America is similar to that seen in the current study. The frequencies of O, A, B and AB blood groups among black American population (non-Hispanic) are 50.2%, 25.9%, 19.7%, and 4.3% respectively (Cooling, 2014). The frequency of blood group B is known to be higher than group A among people of African and Asian origin (Ndoula, Noubiap, Nansseu, & Wonkam, 2014) whiles blood group A occurs more than B among Caucasians in Europe and America (Ndoula et al., 2014). 80 University of Ghana http://ugspace.ug.edu.gh The current study shows the frequencies of antigens A1 and A2 among the study population as 18.5% and 6.58% respectively, with antigen A1 emerging as the more common A antigen. The frequency of antigen A2 from this study is similar to data obtained from studies in some West African countries; Liberia 5.7%, Ivory Coast 7.31% and Burkina Faso 7.14% (Livingstone et al., 1960). This data also agree with data from studies in India, South Africa and Japan that reported that majority of group A individuals belong to subgroup A1 (May & du Toit, 1989; Sujata, AkankSha, Richa, & Niraj, 2017; Yoshida, Dave, & Hamilton, 1988). Higher frequency of antigen A2 (11.5%) is observed among the people of southern England (G. Daniels et al., 2016) whiles lower frequency (3.93%) is observed among Kazakh population in China (Fen Qiu, Jun, Xuemei, & Asmuguli, 2018). The frequency of ABO subgroup A2 among the Indian population is 4.6% (Bangera et al., 2007). A2 generally occurs in 20% of all group A individuals while A1 subgroup occurs in approximately 80% (Helmut Schenkel-Brunner, 2000). The proportion of subgroup A2 obtained among group A individuals from the current study (21%) compares very well with the worldwide proportion of A2 (20%) as well as data from neighboring Burkina Faso (21%), Ivory Coast (25.5%) and Liberia (19.7%) (Livingstone, et al., 1960). Lower A2 frequency was, however, obtained from Sudan (6.58%) in a study involving both blacks and Arabs (Ahmed Mohammed Elnour et al., 2015). The proportion of A2 among group A Caucasians (23%) (Reid et al., 2012) agrees with the data obtained from the current study. However, the proportion of group A Asians with subgroup A2 is extremely lower than that obtained from this study as well as that of Mexicans and Caucasians (Reid et al., 2012). The A2 proportion among group A Japanese, Indians and Chinese are 0.17%, 1.1% and 0.52% respectively (Shastry & Bhat, 2010; Sujata et al., 2017; Xiang, Liu, & Guo, 2006). The genetic differences in A1 genes between the consensus A1 (A101) and that of Eastern Asians (Japanese and Chinese Han) (A102) may have a role to play in the low frequency of A2 among Asians. The close agreement between the A2 81 University of Ghana http://ugspace.ug.edu.gh proportion among Ghanaians and that of Caucasians may be as a result of the Ghanaian A1 gene being genetically similar to that of Caucasians (Consensus A1 gene). Some other factors may be responsible for this similarity and needs to be studied. From the current study, 55% of AB individuals belong to A2B subgroup. This is relatively high comparing data from other West African Countries; 33.72% in Liberia and 37.5% in Ivory Coast (Livinstone et al., 1960). Most studies reveal a higher antigen A2 ratio among group AB individuals. The values have ranged from 35% to 53% (Livingstone et al., 1960; Morel, Watkins, & Greenwell, 1984; Ssebati, 1973). The 55% obtained in the current study is among the highest reported. There is a relatively lower frequency of subgroup A2B among AB individuals in Sudan (8.33%) whose population is made up of Blacks and Arabs (Elnour et al., 2015). The A2B subgroup is known to occur more among Blacks and Japanese (Daniels, 2013). A study involving 5000 African Americans revealed 47% of blood group AB individuals were A2B whiles 53% were A1B (Morel et al., 1984). Similar observations have been made among Caucasians and Chinese people (Voak et al., 1970). About 30% of southern English population with blood group AB are A2B (Daniels et al., 2016) whiles 8-10% of blood group AB Indians are A2B (Sujata et al., 2017; Shastry & Bhat, 2010). Several studies attribute this increase in A2B genotype among the AB population to suppression of the A gene when combined with a strong B gene. Although this may possibly be the reason for the outcome of this study, we are unable to confirm that the higher quantities of A2 among AB individuals observed are due to the presence of a strong B gene because we did not analyze the B genes of the individuals studied. Moreover, a different observation has been made among Japanese on the cause of their higher A2B frequency. Molecular studies among Japanese population revealed the recombinant A allele (R101) is expressed as A1 phenotype among individuals with A/O (R101/O) genotype. This allele is however expressed as A2 phenotype among individuals with A/B (R101/B) genotype (Ogasawara, et al., 1998). This shows that 82 University of Ghana http://ugspace.ug.edu.gh several other factors may be responsible for a higher blood group A2 frequency among specific group of people. There is an enhanced expression of weak A genes on some occasions when they are combined with strong B genes in an AB individual (Helmut Schenkel-Brunner, 2000). In such cases, a weak A antigen may be expressed serologically as A2 due to a rise in antigen site density. The high frequency of A2B individuals in the current study cannot be attributed to this allelic enhancement phenomenon due to the absence of weak forms of A in the current study. Out of the 95 A2 and A2B samples received, only one sample had developed the anti-A1 antibody. Surprisingly, none of the A2B individuals who are known to produce more anti-A1 had developed the antibody. The anti-A1 detected in this study is not clinically significant because it only reacted at room temperature. The lack of anti-A1 production among the A2 and A2B study participants may be associated with increased levels of A c and Ad determinants on their RBC surfaces. Anti-A1 is believed to be an antibody to A c and Ad determinants because A2 individuals who produce anti-A1 in their serum lack these determinants (Harmening et al., 2012). 83 University of Ghana http://ugspace.ug.edu.gh 5.1 LIMITATIONS OF THE STUDY Only one of the A2 samples had developed anti-A1 antibody, hence making it very difficult to determine its significance among the A2 and A2B subgroups. On the other hand, one can conclude that production of anti-A1 is low and therefore of no significance serologically. 5.2 CONCLUSION This study shows that 6.6% of the entire blood donor population who visit the SABC were group A2 or A2B. Group A2 constitutes 26.1% of all group A encountered at the SABC. Among group A individuals, 20.7% were group A2 whiles 79.3% were group A1. This study shows that majority of group A individuals in Ghana are A1 whiles majority of the AB individuals are A2B. The distribution of ABO blood group antigens among the donor population can therefore be arranged in the order O>B>A1>A2>A2B>A1B. Screening for anti-A1 among Ghanaian donors at the SABC may not be necessary due to its low prevalence and also, the only anti-A1 detected in this study was not clinically significant. 84 University of Ghana http://ugspace.ug.edu.gh RECOMMENDATIONS The following recommendations are made based on the study findings; a) Study of H-antigen sites in blood group A2 and A2B individuals and A1B individuals. This may help to find out whether the proportion of A2 and A2B in this study results from decreased H antigen sites or other causes. b) Molecular studies of the subgroups of A i.e. A2-Aend and A1B and A2B individuals. This may reveal the gene and gene product that controls the production of A2 among the Ghanaian population. 85 University of Ghana http://ugspace.ug.edu.gh Reference Alemu, G., & Mama, M. (2016). Assessing ABO/Rh Blood Group Frequency and Association with Asymptomatic Malaria among Blood Donors Attending Arba Minch Blood Bank, South Ethiopia. Malaria Research and Treatment J, 2016, 6. doi: 10.1155/2016/8043768 Anstee, D. J. (1990). Blood group - active surface molecules of the human red blood cell Vox Sang, 58(1), 1 – 20. Atire, F. A. (2015). Analysis of the Blood Type and Group among Undergraduate Physics Students of Dilla University, Ethiopia. Hereditary Genet, 4(140), 2161- 1041.1000140. Baldus, S. E., Thiele, J., Park, Y., Hanisch, F. G., Bara, J., & Fischer, R. (1996). Characterization of the binding specificity of Anguilla anguilla agglutinin (AAA) in comparison to Ulex europaeus agglutinin I (UEA-I). Glycoconjugate Journal, 13(4), 585-590. Bangera, I. S., Fernandes, H., & Swethadri, G. K. (2007). Prevalence of A 2 sub group in A and AB blood groups and the transfusion implications. Asian Journal of Transfusion Science, 1(2), 103. Bennett, E. P., Steffensen, R., Clausen, H., Weghuis, D. O., & van Kessel, A. G. (1995). Genomic cloning of the human histo-blood group ABO locus Biochemical and Biophysical Research Communications, 206(1), 318–325. Bennett, M., Levene, C., & Greenwell, P. (1998). An Israeli family with six cisAB members: Serologic and enzymatic studies. Transfusion, 38, 441-448. Bernstein, Felix. (1924). Ergebnisse einer biostatistischen zusam-menfassenden betrachtung über die erblichen blutstrukturen des menschen. Journal of Molecular Medicine, 3(33), 1495-1497. Bhatia, H. M., & Sathe, M. S. (1974). Incidence of 'Bombay' (Oh) phenotype and weaker variants of A and B antigen in Bombay (India). Vox sanguinis, 27(6), 524-532. Bianco-Miotto, T., Hussey, D. J., Day, T. K., O’Keefe, D. S., & Dobrovic, A. (2009). DNA methylation of the ABO promoter underlies loss of ABO allelic expression in a significant proportion of leukemic patients. PLoS ONE, 4(3), e4788. Brain, P. (1966). Subgroups of A in the South African Bantu. Vox sanguinis, 11(6), 686 – 698. Breimer, M. E., & Jovall, P. A. (1985). Structural characterization of a blood group A heptaglycosylceramide with globo - series structure. The major glycolipid based blood group A antigen of human kidney. FEBS Letters, 179(1), 165 – 172. Breton, C., Oriol, R., & Imberty, A. (1998). Conserved structural features in eukaryotic and prokaryotic fucosyltransferases. Glycobiology, 8(1), 87-94. Cartron, J. P. (1976). Etude des proprietes aN-acetylgalactosaminyltransferasiques des serums de sujets A et ‘A faible’. Revue Francaise de Transfusion et Zmmuno- hbmatologie, 19, 67-88. Cartron, J. P., Badet, J., Mulet, C., & Salmon, C. (1978). Study of the a-N- acetylgalactosaminyltransferases in sera and red cell membranes of human A subgroups. Journal of Immunogenetics, 5, 107-116. Cartron, J. P., Gerbal, A., Badet, J., Ropars, C., & Salmon, C. (1975). Assay of a-N- acetylgalactosaminyltransferases in human sera: Further evidence for several types of Am individuals. Vox sanguinis, 28(5), 347-365. Cartron, J. P., Mulet, C., & Bauvois, B. (1980). ABH and Lewis glycosyltransferases in human red cells, lymphocytes and platelets. Revue francaise de transfusion et immuno-hematologie, 3(23), 271–282. 86 University of Ghana http://ugspace.ug.edu.gh Cartron, J. P., Reyes, F., Gourdin, M. F., Garretta, M., & Salmon, C. (1977). Antigen site distribution among 'weak A' red cell populations. A study of A3, Ax and Aend variants. Immunology, 32(2), 233-244. Chandrasekaran, E. V., Chawda, R., D., Locke R., Piskorz, C. F., & Matta, K. L. (2002). Biosynthesis of the carbohydrate antigenic determinants globo H, blood group H and Lewis b: a role for prostate cancer cell α1,2-L-fucosyltransferase. Glycobiology, 12(3), 153-162. Chaudhari, C. N., Misra, R. N., & Nagpal, A. K. (2008). Transfusion in blood group A2B with anti A1 Recipient. Medical journal, Armed Forces India, 64(4), 371-372. Cheng, P., & DeVries, A. (1986). Mucin biosynthesis: enzymic properties of human-tracheal epithelial GDP-L-fucose: D-galactoside a.-(1.2)-L-fucosyltransferase. Carbohydr Res, 149(1), 253-261. Chester, M. A. (1974). The occurrence of a beta-galactosyltransferase in normal human urine. FEBS Letters, 46(1-2), 59-62. Clausen, H, Levery, SB, Nudelman, E, Baldwin, M, & Hakomori, S. (1986). Further characterization of Type 2 and Type 3 chain blood group A glycosphingolipids from human erythrocyte membranes. Biochemistry, 25(22), 7075–7085. Clausen, H., & Hakomori, S. (1989). ABH and Related Histo‐Blood Group Antigens; Immunochemical Differences in Carrier Isotypes and Their Distribution. Vox sanguinis, 56(1), 1-20. Clausen, H., Levery, S. B., Nudelman, E., Baldwin, M., & Hakomori, S. (1986). Further characterization of type 2 and type 3 chain blood group A glycosphinogolipids from human erythrocyte membranes. Biochemistry, 25(22), 7075-7085. Clausen, H., Watanabe, K., & Kannagi, R. (1984). Blood group A glycolipid (Ax) with globo - series structure which is specific for blood group A1 erythrocytes: one of the chemical bases for A1 and A2 distinction. Biochemical and Biophysics Research Communications, 124(2), 523–529. Cooling, Laura. (2010). Polylactosamines, there's more than meets the" Ii": a review of the I system. Immunohematology-Journal of Blood Group Serology and Education, 26(4), 133. Cooling, Laura. (2014). ABO, H, and Lewis blood groups and structurally related antigens Technical Manual (18 ed., pp. 291-315). Bethesda, Maryland: AABB Press. Cooling, Laura. (2016). Carbohydrate blood groups. In L. Toby, L. Edward & G. Bjarte (Eds.), Rossi’s Principles of transfusion medicine (5th ed., pp. 159-169). Oxford: Wiley Blackwell. Daniels, G., Contreras, M., & Allard, S. (2016). Red cell immunohaematology. In V. Hoffbrand, D. Higgs, D. Keeling & A. Mehta (Eds.), Postgraduate Haematology (7th ed., pp. 195-213). West Sussex: John Wiley & Sons Ltd. Daniels, Geoff. (2013). ABO, H, and Lewis Systems. In G. Daniels (Ed.), Human Blood Groups (pp. 11-68). Oxford: John Wiley & Sons, Ltd., Publication. Decastello, A. von. (1902). Uber die Iso-aggiutinine im Serum gesunder und kranker Menschen. Munch med Wchnschr, 49, 1090-1095. Dodd, B. E., & Gilbey, B. E. (1957). An unusual variant of group A. Vox sanguinis, 2(6), 390-398. Dodd, B. E., & lincoln, P. J. (1978). Serological studies of the H activity of Oh red cells with various anti-H reagents. Vox sanguinis, 35(3), 168-173. Donald, A. S. R. (1981 ). A-active trisaccharides isolated from A1 and A2 blood-group- specific glycoproteins. European Journal of Biochemistry, 120(2), 243 – 249. Dungern, E. V. (1910). Ueber Vererbung gruppenspezifischer Strukturen des Blutes, II. Z. Immun. Exp. Ther., 1, 284-292. 87 University of Ghana http://ugspace.ug.edu.gh Economidou, J., Hughes‐Jones, N. C., & Gardner, B. (1967). Quantitative measurments concerning A and B antigen sites. Vox Sanguinis, 12(5), 321-328. Elnour, Ahmed M., Ali, Nada Y., Hummeda, Sulafa A., Alshazally, Wigdan Y., Elderdery, Abozer Y., & Omer, Nawal E. (2015). Frequency of the A2-subgroup among blood group A and blood group AB among students of faculty of medicine and health sciences at Alimam Almahadi University, White Nile, Sudan. Hematology & Transfusion International Journal, 1(4), 6. doi: 10.15406/htij.2015.01.00022 Elnour, Ahmed Mohammed, Ali, Nada Y, Hummeda, Sulafa A, Alshazally, Wigdan Y, Elderdery, Abozer Y, & Omer, Nawal E. (2015). Frequency of the A2-subgroup among blood group A and blood group AB among students of faculty of medicine and health sciences at Alimam Almahadi University, White Nile, Sudan. Fen Qiu, G., Jun, W., Xuemei, T., & Asmuguli, Hui P. (2018). A survey of phenotype distribution of subgroup A2, blood groups MNS, P, and Kell in the Kazak nationality living in northern Xinjiang of China. Asia-Pacific Journal of Blood Types and Genes, 3(2), 159 -163. Fisher, N., & Cahan, A. (1962). An addition to the family in which Ax is transmitted through a person of the blood group A2B. Vox sanguinis, 7(4), 484 Fox, Patrick Keolu Ozer. (2016). Next-generation ABO Genetics and Genomics. (Doctor of Philosophy Genome Sciences Doctoral Dissertation), University of Washington. Franchini, M., & Liumbruno, G. M. (2013). ABO blood group: old dogma, new perspectives. Clinical Chemistry and Laboratory Medicine, 8(51), 1545-1553. Friedenreich, V. (1936). Eine bisher unbekannte Blutgruppeneigenschaft (A3). Z Immun Forsch, 89, 409 – 422. Fujitani, N. (2000). Expression of H type 1 antigen of ABO histio–blood group in normal colon and aberrant expressions of H type 2 and H type 3/4 antigens in colon cancer. Glycoconjugate Journal, 17(5), 331–338. Fukuda, M., Dell, A., Oates, J. E., & Fukuda, M. N. (1984). Structure of a branched lactosaminoglycan, the carbohydrate moiety of band 3 isolated from adult human erythrocytes. Journal of Biological Chemistry, 259(13), 8260–8273. Furukawa, K., Mattes, M. J., & Lloyd, K. O. (1985). A1 and A2 erythrocytes can be distinguished by reagents that do not detect structural differences between the two cell types. Journal of Immunology, 135(6), 4090-4094. Garratty, G., Glynn, S. A., & McEntire, R. (2004). ABO and Rh (D) phenotype frequencies of different racial/ethnic groups in the United States. Transfusion, 44(5), 703–706. Ginsburg, V. (1972). Enzymatic basis for blood groups in man. Advances in Enzymology and Related Areas of Molecular Biology, 36, 131-149. Gooi, H. C., Feizi, T., & Kapadia, A. (1981 ). Stage-specific embryonic antigen involvesα 1→ 3 fucosylated type 2 blood group chains. Nature 5819(292), 156 – 158. Grasemann, H., de Perot, M., & Bendiak, G. N. (2012). ABO-incompatible lung transplantation in an infant. American Journal of Transplantation, 12(3), 779 – 781. Hakomori, S. (1981). Blood group ABH and Ii antigens of human erythrocytes: chemistry, polymorphism, and their developmental change Semin Hematol, 18(1), 39 – 62. Hakomori, Sen-itiroh. (1999). Antigen structure and genetic basis of histo-blood groups A, B and O: their changes associated with human cancer. Biochimica et Biophysica Acta (BBA)-General Subjects, 1473(1), 247-266. Harmening, D., F., Glenda, & T., Barbara. (2012). The ABO Blood Group System. In D. Harmening (Ed.), Mordern blood banking and transfusion practices (6th ed., pp. 119- 146). Philadelphia: F. A. Davis Company. 88 University of Ghana http://ugspace.ug.edu.gh Hassinen, A., Rivinoja, A., Kauppila, A., & Kellokumpu, S. (2010). Golgi Nglycosyl- transferases form both homo- and heterodimeric enzyme complexes in live cells. Journal of Biological Chemistry, 285(23), 17771-17777. Hosoi, E., Hirose, M., & Hamano, S. (2003). Expression levels of H-type alpha (1,2)- fucosyltransferase gene and histo-blood group ABO gene corresponding to hematopoietic cell differentiation. Transfusion, 43(1), 65-71. Hosseini-Maaf, B., Hellberg, A., Rodrigues, M. J., Chester, M. A., & Olsson, M. L. (2003). ABO exon and intron analysis in individuals with the Aweak B phenotype reveals a novel O1v-A2 hybrid allele that causes four missense mutations in the A transferase. BMC Genetics, 4(1), 17. Hrubisko, M., Calkovska, Z., Mergancova, O., & Gallova, K. . (1966). Beobachtungen Ober Varianten des Blutgruppensystems ABO. I Studie der Variante Am. Blut, 13(4), 137- 142. Hult, A. K., Yazer, M. H., & Jørgensen, R. (2010). Weak A phenotypes associated with novel ABO alleles carrying the A2-related 1061C deletion and various missense substitutions. Transfusion, 50(7), 1471 – 1486. Hult, Annika. (2013). Studies of the ABO and FORS Histo-Blood Group Systems: Focus on Flow Cytometric and Genetic Analysis. Lund University, Faculty of Medicine Doctoral Dissertation Series, 2013(63). Hurst, F. P., Sajjad, I., & Elster, E. A. (2010). Transplantation of A2 kidneys into B and O recipients leads to reduction in waiting time: USRDS experience. Transplantation 89(11), 1396–1402. Irshaid, N. M., Chester, M. A., & Olsson, M. L. (1999). Allele-related variation in minisatellite repeats involved in the transcription of the blood group ABO gene. Transfus Medicine, 9(3), 219–226. Issitt, P. D., & Anstee, D. J. (1998). Applied blood group serology (4th ed.). Durham NC: Montgomery Scientific Publications. Iyiola, O. A., Igunnugbemi, O. O., & Bello, O. G. (2012). Gene frequency of ABO and RH(D) blood group allels in Lagos, South-West Nigeria. The Egyptian Journal of Medical Human Genetics, 12(2), 147–153. Kabat, E. A. (1956). Blood Group Substances; Their Chemistry and Immunochemistry. New York: Academic Press. Kannagi, R., Levery, S. B., & Hakomori, S. (1984 ). Blood group H antigen with globo-series structure; Isolation and characterization from human blood group O erythrocytes. FEBS Letters, 175(2), 397 – 401. Kassahun, T., Yohannes, P., & Mebeaselassie, A. (2015). Frequency distribution of ABO and Rh (D) blood. The Egyptian Journal of Medical Human Genetics. 16(1), 71-76. Kelly, Robert J., Rouquier, S., Giorgi, D., Lennon, Gregory G., & Lowe, John B. (1995). Sequence and Expression of a Candidate for the Human Secretor Blood Group α (1, 2) Fucosyltransferase Gene (FUT2) Homozygosity For An Enzyme-Inactivating Nonsense Mutation Commonly Correlates with the non-Secretor Phenotype. Journal of Biological Chemistry, 270(9), 4640-4649. Khan, M. N., Khaliq, I., Bakhsh, A., Akhtar, M. S., & Amin ud Din, M. (2009). Distribution of ABO and Rh D blood groups in the population of Poonch District, Azad Jammu and Kashmir. Eastern Mediterranean health journal, 15, 717-721. doi: 10.26719/2009.15.3.717 Klein, H. G., & Anstee, D. J. (2006). ABO, Lewis and P groups and Ii antigens. In H. Klein & D. Anstee (Eds.), Mollison's Blood Transfusion in Clinical Medicine (11th ed., pp. 114-150). Oxford: Blackwell Publishing. 89 University of Ghana http://ugspace.ug.edu.gh Knowles, R. W., Bai, Y., Daniels, G. L., & Watkins, W. M. (1982). Monoclonal anti‐Type 2 H: an antibody detecting a precursor of the A and B blood group antigens. International Journal of Immunogenetics, 9(2), 69-76. Kobata, Akira. (1972). Isolation of oligosaccharides from human milk. Methods in Enzymology, 28, 262-271. Koda, Y., Soejima, M., Wang, B., & Kimura, H. (1997). Structure and Expression of the Gene Encoding Secretor‐Type Galactoside 2‐α‐l‐fucosyltransferase (FUT2). European journal of biochemistry, 246(3), 750-755. Kominato, Y., Hata, Y., & Takizawa, H. (1999). Expression of human histo - blood group ABO genes is dependent upon DNA methylation of the promoter region. Journal of Biological Chemistry, 274(52), 37240–37250. Kominato, Y., Hata, Y., & Takizawa, H. (2002 ). Alternative promoter identifi ed between a hypermethylated upstream region of repetitive elements and a CpG island in human ABO histo -blood group genes. Journal of Biological Chemistry, 277(40), 37936 – 37948. Koscielak, J., Lenkiewicz, B., Zielenski, J., & Seyfried, H. (1986). Weak A phenotypes possibly caused by mutation. Vox sanguinis, 50(3), 187-190. Kretchy, J. P., Doku, G. N., Annor, R. A., Addy, B. S., & Asante, R. K. (2017). Distribution of ABO blood group/Rhesus factor in the Eastern Region of Ghana towards effective blood bank inventory. Scholars Journal of Applied Medical Sciences, 5(3), 821-826. Kudo, T., Iwasaki, H., & Nishihara, S. (1996). Molecular genetic analysis of the human Lewis histo-group system. II. Secretor gene inactivation by a novel single missense mutation A385T in Japanese nonsecretor individuals. Journal of Biological Chemistry, 271(16), 9830–9837. Landsteiner, K., & Levine, P. (1926). On the cold agglutinins in human serum. Journal of Immunology, 12(6), 441–460. Landsteiner, K., & Witt, D. H. (1926 ). Observations on the human blood groups. Irregular reactions. Isoagglutinins in sera of group IV. The factor A1. Journal of Immunology, 11(3), 221 – 247. Landsteiner, Karl. (1900). Zur Kenntnis der antifermentativen, lytischen und agglutinierenden Wirkungen des Blutserums und der Lymphe. Zentralbl. Bakteriol, 27, 357-362. Lanset, S., Ropartz, C., Rousseau, P. Y., Guerbet, Y., & Salmon, C. (1966). Une famille comportant les phenotypes Bombay: OABh et OBh. Transfusion, 9(3), 255-258. Larson, G., & Samuelsson, B. E. (1980). Blood-group type glycosphingolipids of human cord blood erythrocytes. Journal of Biochemistry, 88(3), 647-657. Lattes, L., & Cavazzuti, A. (1924 ). Sur l’existence d’un troisième élément d’isoagglutination. Journal of Immunology, 9(5), 407 – 425. Lau, P., Sererat, S., Beaty, J., Oilschlager, R., & Kini, J. (1990). Group A variants defined with a monoclonal anti-A reagent. Transfusion 30(2), 142-145. Le Pendu, J., Cartron, J. P., Lemieux, R. U., & Oriol, R. (1985). The presence of at least two different H-blood-group-related j3-D-Gal a.-2-L-fucosyltransferases in human serum and the genetics of blood group H substances. American Journal of Human Genetics, 37(4), 749-760. Le Pendu, J., Lambert, F., Gerard, G., Vitrac, D., Moilicone, R., & Oriol, R. (1986). On the specificity of human anti-H antibodies. Vox sanguinis, 50(4), 223-226. Le Pendu, J., Lambert, F., & Samuelsson, B. (1986). Monoclonal antibodies specific for Type 3 and Type 4 chain-based blood group determinants: relationship to the A1 and A2 subgroups. Glycoconjugate Journal, 3(3), 255 – 271. 90 University of Ghana http://ugspace.ug.edu.gh Le Pendu, J., Lemieux, R. U., Lambert, F., Dalix, A. M., & Oriol, R. (1982). Distribution of H type 1 and H type 2 antigenic determinants in human sera and saliva. American Journal of Human Genetics, 34(3), 402-415. Lemieux, R. U. (1978). Human blood groups and carbohydrate chemistry. Chemical Society Review, 7(4), 423-452. Lemieux, R. U., & Driguez, H. (1975). The chemical synthesis of 2-0-(alpha-L- fucopyranosyl)-3-O-(alpha-D-galactopyranosyl)o-galactose. Terminal structure of the blood-group B antigenic determinant. Journal of the American Chemical Society, 97(14), 4069-4071. Livingstone, Frank B., Gershowitz, H., Neel, James V., Zuelzer, Wolf W., & Solomon, Marvin D. (1960). The distribution of several blood group genes in Liberia, the Ivory Coast and Upper Volta. American journal of physical anthropology, 18(3), 161-178. Lowe, John B. (1995). Biochemistry and Biosynthesis of ABH and Lewis Antigens; Characterization of Blood Group-Specific Glycosyltransferases. In C. JEAN-PIERRE & ROUGER (Eds.), Blood Cell Biochemistry (Vol. VI, pp. 75-109). New York: Springer Science+Business Media. Lundblad, A. (1978 ). Oligosaccharides from human urine. Methods Enzymol, 50, 226 – 235. Mackie, R. I., Sghir, A., & Gaskins, H. R. (1999). Developmental microbial ecology of the neonatal gastrointestinal tract. The American Journal of Clinical Nutrition, 65(5), 1035S–1045S. Mak, K. H., Lubenko, A., Greenwell, P., Voak, D., Yan, K. F., & Poole, J. (1996). Serologic characteristics of H-deficient phenotypes among Chinese in Hong Kong. Transfusion, 36(11-12), 994-999. Makela, O., Ruoslathi, E., & Ehnholm, C. (1969). Subtypes of human ABO blood-groups and subtype-specific antibodies. The Journal of Immunology, 102(3), 763-771. Marcus, D. M., & Cass, L. E. (1969 ). Glycosphingolipids with Lewis blood group activity: uptake by human erythrocytes. Science, 164(3879), 553–555. Marr, A. M. S., Donald, A. S. R., & Morgan, W. T. J. (1968). Two new oligosaccharides obtained from an Lea - active glycoprotein. Biochemical Journal, 110(4), 789–791. May, R. M., & du Toit, E. D. (1989). Blood group gene frequencies of four population groups in the western Cape. South African Medical Journal, 12(76), 647-650. Mifsud, N. A., Watt, J. M., & Condon, J. A. (2000). A novel cis-AB variant allele arising from a nucleotide substitution A796C in the B transferase gene. Transfusion, 40(10), 1276–1277. Monteiro, F., Tavares, G., Ferreira, M., Amorim, A., Bastos, P., Rocha, C., . . . Cunha- Ribeiro, L. M. (2011). Technologies involved in molecular blood group genotyping. International Society of Blood Transfusion, 6(1), 1–6. Morel, P. A., Watkins, W. M., & Greenwell, P. (1984). Genotype A1B expressed as phenotype A2B in a Black population. Transfusion, 24, 426. Morgan, W., & van Heyningen, R. (1944). The occurrence of A, B and O blood group substances in pseudo-mucinous ovarian cyst fluids. Brit. J. exp. Path, 25(5). Munro, J. R., & Schachter, H. (1973). The presence of two GDP-L-fucose:glycoprotein fucosyltransferases in human serum. Archives of Biochemistry and Biophysics, 156(2), 534-542. Murray, N. A., & Roberts, I. A. G. (2007). Hemolytic disease of the newborn. Archives of Disease in childhood-fetal and neonatal edition, 92(2), F83-F88. Musa, A. U., Ndakotsu, M. A., Abdul-Aziz, H., Kilishi, A., & Aliyu, I. (2015). Distribution of ABO and Rhesus blood group systems among blood donors in Sokoto North- western Nigeria. Journal of Applied Hematology, 6(3), 136-138. 91 University of Ghana http://ugspace.ug.edu.gh Naing, L., Winn, T. , & Rusli, B. N. (2006). Practical Issues in Calculating the Sample Size for Prevalence Studies. Archives of Orofacial Sciences, 1, 9-14. Ndoula, S. T., Noubiap, J. J., Nansseu, J. R., & Wonkam, A. (2014). Phenotypic and allelic distribution of the ABO and rhesus (D) blood groups in the Cameroonian population. International Journal of Immunogenetics, 41(3), 206-210. Nijenhuis, L. E., & Bratlie, K. (1962). ABO antibodies in twins. Vox Sanguinis, 7(2), 236– 238. O ’ Keefe, D. S., & Dobrovic, A. (1996). Decreased stability of the O allele mRNA transcript of the ABO gene. Blood, 87(7), 3061–3068. Ogasawara, K., Tsuneyama, H., & Uchikawa, M. (2010). A new cisAB allele with 796A>G (Met266Val) mutation may generate from B allele. Vox Sanguinis, 99, 334. Ogasawara, K., Yabe, R., Uchikawa, M., Bannai, M., Nakata, K., Takenaka, M., . . . Tokunaga, K. (1998). Different alleles cause an imbalance in A2 and A2B phenotypes of the ABO blood group. Vox sanguinis, 74(4), 242-247. Oguchi, Y., Kawaguchi, T., Suzuta, T., & Osawa, T. (1978). The nature of human blood group A3 erythrocytes. Vox sanguinis, 34(1), 32 – 39. Olsson, M. L., & Chester, M. A. (1996a). Frequent occurrence of a variant O1 gene at the blood group ABO locus. Vox sanguinis, 70(1), 26–30. Olsson, M. L., & Chester, M. A. (1996b). Polymorphisms at the ABO locus in subgroup A individuals. Transfusion, 36(4), 309-313. Oriol, R., Daniiovs, J., & Hawkins, B. R. (1981). A new genetic model proposing that the Se gene is a structural gene closely linked to the H gene. American Journal of Human Genetics, 33(3), 421-431. Oriol, R., Le Pendu, J., & Mollicone, R. (1986). Genetics of ABO, H, Lewis. X and related antigens. Vox Sanguinis, 51(3), 161-171. Oriol, R., Mollicone, R., & Cailleau, A. (1999). Divergent evolution of fucosyltransferase genes from vertebrates, invertebrates, and bacteria. syltransferase genes from vertebrates, invertebrates and bacteria. Glycobiology, 9(4), 323-334. Oriol, Rafael. (1995). ABO, Hh, Lewis, and secretion. Serology, genetics and tissue distribution. In J.-P. Cartron & P. Rouger (Eds.), Blood Cell Biochemistry (Vol. 6, pp. 37–66). New York: Plenum Press. Owen, R. D. (1954). Heterogeneity of antibodies to the human blood groups in normal and immune sera. Journal of Immunology, 73(1), 29 – 39. Pacuska, T., & Koscieak, J. (1974). a1-2 Fucosyltransferase of human bone marrow. FEBS Letters, 41(2), 348-351. Patenaude, S. I., Seta, N. O. L., & Borisov, S. N. (2002). The structural specificity in human ABO(H) blood group biosynthesis. Nature Structural & Molecular Biology, 9(9), 685-690. Pawlak, Z., & Lopez, M. (1979). Développement des antigènes ABH et Ii chez les enfants de O à 16 ans. Review Fr Transfus Immunohématol, 3(1979), 253–263. Pramanik, T., & Pramanik, S. (2000). The distribution of ABO and Rh blood groups in Nepalese students: a report. Eastern mediterranian Health Journal, 6(1), 156-158. Putkonen, T. (1930). Ü ber die gruppenspezifi schen Eigenschaften Verschiedener Korperfl ussigkeiten. Acta Soc Med Fenn Duodecim Ser A, 14, 113–153. Race, C., & Watkins, W. M. (1972). The enzymic products of the human A and B blood group genes in the serum of 'Bombay' Oh donors. FEBS Letters, 27(1), 125-130. Race, C., Ziderman, D., & Watkins, W. M. (1968). An a-D-galactosyltransferase associated with the blood-group B character. Biochemical Journal, 107(5), 733-735. Race, R., & Sanger, R. (1975). Blood groups in man (6th ed.): Blackwell Scientific publications Oxford. 92 University of Ghana http://ugspace.ug.edu.gh Ravn, V., & Dabelsteen, E. (2000). Tissue distribution of histo‐blood group antigens. Apmis, 108(1), 1-28. Reed, T. E. (1964). The frequency and nature of blood group A3. Transfusion, 4(6), 457 – 460. Regan, Fiona. (2016). Blood cell antigens and antibodies: erythrocytes, platelets and granulocytes. In B. Bain, I. Bates, M. Laffan & M. Lewis (Eds.), Dacie and Lewis Practical Haematology (11 ed., pp. 482-513): Elsevier Health Sciences. Rege, V. P., Painter, T. J., Watkins, W. M., & Morgan, W. T. J. (1964). Isolation of serologically active fucose - containing oligosaccharides from human blood - group H substance. Nature 203(4943), 203. Reid, Marion E., Lomas-Francis, C., & Olsson, Martin L. (2012). The blood group antigen factsbook (3rd ed.). Waltham: Academic press. Rochant, H., Tonthat, H., Henri, A., Titeux, M., & Dreyfus, B. (1976). Abnormal distribution of erythrocytes A1 antigens in preleukemia as demonstrated by an immunofluorescence technique. Nouv Rev Fr Hematol Blood Cells, 17(1-2), 237 – 255. Saitou, N., & Yamamoto, F. (1997). Evolution of primate ABO blood group genes and their homologous genes. Molecular Biology and Evolution 14(4), 399–411. Sakamoto, J., Yin, B. W. T., & Lloyd, K. O. (1984 ). Analysis of the expression of H, Lewis, X, Y and precursor blood group determinants in saliva and red cells using a panel of mouse monoclonal antibodies Molecular Immunology, 21(11), 1093 – 1098. Salmon, C., & Cartron, J. P. (1977). Interactions in AB heterozygotes. CRC Handbook Series In Clinical Laboratory Science, 131-138. Sambo, N., Yerima, A. M., Amaza, D. S., Bukar, A. A., & Gali, R. M. (2016). Association of ABO (H) Secretor Status with Diabetes Mellitus. IOSR Journal of Dental and Medical Sciences 15(6), 1-5. Sano, R., Nakajima, T., & Takahashi, K. (2012). Expression of ABO blood-group genes is dependent upon an erythroid cell-specific regulatory element that is deleted in persons with the Bm phenotype. Blood, 119(22), 5301–5310. Sarnesto, A., Kohlin, T., & Hingsgaul, O. (1992). Purification of the secretor-type beta- galactoside alpha 1-2-fucosyltransferase from human serum. Journal of Biological Chemistry, 267(4), 2737-2744. Schachter, H., Michaels, M. A., Crookston, M. C., Tilley, C. A., & Crookston, J. H. (1971). A quantitative difference in the activity of blood group A-specific N- acetylgalactosaminyltransferase in serum from A1 and A2 human subjects Biochemical and Biophysical Research Communications, 45(4), 1011 – 1018. Schachter, H., Michaels, M. A., Tilley, C. A., Crookston, M. C., & Crookston, J. H. (1973). Qualitative differences in the N-acetyl-d-galactosaminyltransferases produced by humanA1 andA2 genes Proc Natl Acad Sci USA, 70(1), 220 – 224. Schenkel-Brunner, H., & Tuppy, H. (1973). Enzymatic conversion of human blood-group-O erythrocytes into A2 and A1 cells by α-N-acetyl-d-galactosaminyl transferases of blood-group-A individuals European journal of biochemistry, 34(1), 125–128 Schenkel-Brunner, Helmut. (1982). Studies on blood - groups A1 and A2: Further Evidence for the Predominant Influence of Quantitative Differences in the Number of A Antigenic Sites Present on A1 and A2 Erythrocytes. European journal of biochemistry, 122(3), 511 – 514. Schenkel-Brunner, Helmut. (2000). Human Blood Groups (2nd ed.). Wien: Springer-Verlag. Schenkel‐Brunner, H., Chester, Alan M., & Watkins, W. M. (1972). α‐l‐FucosyItransferases in Human Serum from Donors of Different ABO, Secretor and Lewis Blood‐Group Phenotypes. European journal of biochemistry, 30(2), 269-277. 93 University of Ghana http://ugspace.ug.edu.gh Schiff, F., & Sasaki, H. (1932). Der Ausscheidungstypus, ein auf serologischem Wege nachweisbares Mendelndes Merkmal. Klin Wochenschr, 11(34), 1426–1429. Shastry, S., & Bhat, S. (2010). Imbalance in A2 and A2B phenotype frequency of ABO group in South India. Blood Transfusion, 8(4), 267. Shechter, Y., Etzioni, A., Levene, C., & Greenwell, P. (1995). A Bombay individual lacking Hand Le antigens but expressing normal levels of a-2- and a-4-fucosyltransferases. Transfusion 35(9), 773-776. Smith, S., Okai, I., Abaidoo, C. S., & Acheampong, E. (2018). Association of ABO Blood Group and Body Mass Index: A Cross-Sectional Study from a Ghanaian Population. Journal of Nutrition and Metabolism, 2018, 6. doi: 10.1155/2018/8050152 Sokol, R. J., Booker, D. J., Stamps, R., & Windle, J. A. (1995). Autoimmune haemolysis and red cell autoantibodies with ABO blood group specifi city. Haematologia, 26(3), 121 – 129. Springer, G. F., Horton, Richard E., & Forbes, M. (1959). Origin of anti-human blood group B agglutinins in white Leghorn chicks. Journal of Experimental Medicine, 110(2), 221-244. Springer, Georg F. (1971). Blood-group and Forssman antigenic determinant shared beween microbes and mammalian cells. Progress in Allergy, 15, 9–77. Ssebati, E. (1973). A study of the African H-deficient A2 and A2B phenotypes. Vox sanguinis, 24(2), 165-170. Storry, J., & Olsson, M. . (2009). The ABO blood group system revisited: a review and update. Immunohematology, 25(2), 48-59. Sturgeon, P., McQuiston, D., & Van Camp, S. (1973). Quantitative Studies on Salivary Blood Group Substances: II. Normal Values 1. Vox sanguinis, 24(2), 114-125. Sujata, G., AkankSha, A., Richa, B., & Niraj, kumar N. (2017). A1 and A2 Sub-Types of Blood Group‘A’: A Reflction of their Prevalence in North Karnataka Region. Journal of Clinical and Diagnostic Research, 12(5), 40. Svensson, L., Rydberg, L., de Mattos, L. C., & Henry, S. M. (2009). Blood group A1 and A2 revisited an immunochemical analysis. Vox sanguinis, 96(1), 56–61. Szulman, Aron E. (1965). The ABH antigens in human tissues and secretions during embryonal development. Journal of Histochemistry & Cytochemistry, 13(8), 752-754. Szulman, Aron E. (1971). The histological distribution of the blood group substances in man as disclosed by immunofluorescence: IV. The ABH antigens in embryos at the fifth week post fertilization. Human pathology, 2(4), 575-585. Taniuchi, F., Higai, K., Tanaka, T., Azuma, Y., & Matsumoto, K. (2013). Transcriptional regulation of fucosyltransferase 1 gene expression in colon cancer cells. The Scientific World Journal, 2013. doi: 10.1155/2013/105464 Tesfaye, K., Petros, Y., & Andargie, M. (2015). Frequency distribution of ABO and Rh (D) blood group alleles in Silte Zone, Ethiopia. Egyptian Journal of Medical Human Genetics, 16(1), 71-76. Thuresson, B., Chester, M. A., Storry, J. R., & Olsson, M. L. (2008). ABO transcript levels in peripheral blood and erythropoietic culture show different allele-related patterns independent of the CBF/NF-Y enhancer motif and multiple novel allele-specific variations in the 5′- and 3′-noncoding regions. Transfusion, 3(48), 493 – 504. Tsai, L‐C., Kao, L‐G., Chang, J‐G., Lee, H‐H., Linacre, A., & Lee, James C‐I. (2000). Rapid identification of the ABO genotypes by their single‐stand conformation polymorphism. Electrophoresis: An International Journal, 21(3), 537-540. Twu, Y. C., Hsieh, C. Y., & Tu, L. C. (2006). Expression of the histo-blood group B gene predominates in AB-genotype cells. Transfusion, 11(46), 1988 – 1996. 94 University of Ghana http://ugspace.ug.edu.gh Voak, D., Lodge, T., & Reed, J. (1970). A possible explanation for the expression of A2B phenotypes observed in some populations. Vox sanguinis, 18(5), 471–474. von Dungern, E., & Hirschfeld, L. (1910). Ueber Vererbung gruppenspezifischer Strukturen des Blutes II. Z ImmunForsch, 6, 284–292. von Dungern, E., & Hirszfeld, L. (1911). Ü ber gruppenspezifi scheStrukturen des Blutes III. Zeitschrift für Immunologische Forschung, 8, 526 – 562. Wagner, F. F., & Flegel, W. A. (1997). Polymorphism of the h allele and the population frequency of sporadic nonfunctional alleles. Transfusion, 37(3), 284-290. Watkins, W. M. (1978). Blood group gene specified glycosyltransferases in rare ABO groups and in leukaemia. Revue francaise de transfusion et immuno-hematologie, 21(1), 201– 228. Watkins, W. M. (1980). Biochemistry and genetics of the ABO, Lewis and P blood group systems. Advances in Human Genetics, 10, 1–136. Watkins, W. M., Greenwell, P., Yates, A. D., & Johnson, P. H. (1988). Regulation of expression of carbohydrate blood group antigens. Biochimie, 70(11), 1597 – 1611. Watkins, W. M., & Morgan, W. (1957). Specific inhibition studies relating to the Lewis blood-group system. Nature, 180(4594), 1038 – 1040. Watkins, W. M., & Morgan, W. (1959). Possible genetical pathways for the biosynthesis of blood group mucopolysaccharides. Vox sanguinis, 4(2), 97-119. Weiner, W., Lewis, H. B. M., Moores, P., Sanger, R., & Race, R. R. (1957). A gene, y, modifying the ABO group antigen A. Vox sanguinis, 2(1), 25-37. West, L. J., & Platt, J. L. (2010). And justice for all: consideration of ABO compatibility in allocation of hearts for infant transplantation. Circulation, 121(17), 1884–1886. Westhoff, Connie M., & Reid, Marion E. (2007). ABO and related antigens and antibodies. In C. D. Hillyer, L. E. Silberstein, P. M. Ness, K. C. Anderson & J. D. Roback (Eds.), Blood Banking and Transfusion Medicine (pp. 69-79). Philadelphia: Elsevier. Wiener, A. S. (1951). Origin of naturally occurring hemagglutinins and hemo lysins; a review. Journal of Immunology, 66(2), 287-295. Wilczynska, Z., Miller-Podraza, H., & Koscielak, J. (1980). The contribution of different glycoconjugates to the total ABH blood group activity of human erythrocytes. FEBS Letters, 112(2), 277–279. Xiang, D, Liu, X , & Guo, ZH. (2006). Studies of ABO subgroups in the Chinese Han population in Shanghai area. Chin J Blood Transfusion (in Chinese), 1(19), 25-26. Yamaguchi, H., Okubo, Y., & Tanaka, M. (1970). A rare blood Bx analagous to Ax in a Japanese family. Proceedings of the Japan Academy, 46(5), 446 – 449. Yamakami, K. (1926). The individuality of semen, with reference to its property of inhibiting specifically isohemoagglutination. The Journal of Immunology, 12(3), 185-189. Yamamoto, F. (2004). Review: ABO blood group system—ABH oligosaccharide antigens, anti-A and anti-B, A and B glycosyltransferases and ABO genes. Immunohematology, 20(1), 3-22. Yamamoto, F., Cid, E., Yamamoto, M., Saitou, N., Bertranpetit, J., & Blancher, A. (2014). An integrative evolution theory of histo-blood group ABO and related genes. Scientific Reports, 4, 6601. Yamamoto, F., Clausen, H., & White, T. (1990). Molecular genetic basis of the histo-blood group ABO system. Nature, 345(6272), 229-233. Yamamoto, F., McNeill, P. D., & Hakomori, S. (1992). Human histo - blood group A2 transferase coded by A2 allele, one of the A subtypes, is characterized by a single base deletion in the coding sequence, which results in an additional domain at the carboxyl terminal. Biochemical and Biophysical Research Communications, 187(1), 366 – 374. 95 University of Ghana http://ugspace.ug.edu.gh Yamamoto, F., McNeill, P. D., & Yamamoto, M. (1993). Molecular genetic analysis of the ABO blood group system: 4. Another type of O allele. Vox Sanguinis, 64(3), 175– 178. Yamamoto, M., Yamamoto, F., Luong, T., Williams, T., Kominato, Y., & Yamamoto, F. (2003). Expression profiling of 68 glycosyltransferase genes in 27 dif ferent tissues by the systematic mult iplex reverse transcription-polymerase chain reaction method revealed clustering of sexually-related tissues in hierarchiacal clustering algorithm analysis. Electrophoresis 24(14), 2295–2307. Yazawa, S., & Furukawa, K. (1980). a-L-Fucosyltransferases related to biosynthesis of blood group substances in human saliva. Journal of Immunogenetics 7(2), 137-148. Yazer, M. H., & Palcic, M. M. (2005). The importance of disordered loops in ABO glycosyltransferases. Transfusion Medicine Review 19(3), 210 – 216. Yoshida, A. (1983). The existence of atypical blood group Galactosyltransferase which causes an expression A2 character in A1B red blood cells. Am journal of human Genetics. 35(6), 1117-11125. Yoshida, A., Dave, V., & Hamilton, H. B. (1988). Imbalance of blood group A subtypes and the existence of superactive B gene in Japanese in Hiroshima and Nagasaki. American Journal of Human Genetics, 43(4), 422-428. Yu, L-C., Chang, C-Y., Twu, Y-C., & Lin, M. (2000). Human histo-blood group ABO glycosyltransferase genes: Different enhancer structures with different transcriptional activities. Biochemical and Biophysical Research Communications, 273(2), 459–466. 96 University of Ghana http://ugspace.ug.edu.gh APPENDICES APPENDIX 1: Ethical approval from National Blood Service, Ghana 97 University of Ghana http://ugspace.ug.edu.gh APPENDIX 2: Ethical approval from College of Health Sciences, University of Ghana 98 University of Ghana http://ugspace.ug.edu.gh APPENDIX 3: participant informed consent form PREVALENCE OF BLOOD GROUP A2 AMONG GROUP A AND AB DONORS WHO VISIT THE SOUTHERN AREA BLOOD CENTRE I, Evans Owusu Agyapong from University of Ghana, am investigating the “PREVALENCE OF BLOOD GROUP A2 AMONG GROUP A AND AB DONORS WHO VISIT THE SOUTHERN AREA BLOOD CENTRE”. Blood is grouped into four ABO groups (A, B, AB and O) based on the types of antigens present on the red cells. Blood groups are very important in transfusion and transplantation as they help in determining the right match for recipients. They are also important in the prevention of Haemolytic Disease of the Newborn (HDN). This study will help us to know the number of blood group A individuals who belong to subgroup A2. The study will further help to determine the number of A2 individuals who have developed Anti-A1 antibody in their serum which can cause haemolytic transfusion reaction. BENEFITS Knowing the frequency of A2 and anti-A1will inform policy and decision making including the possible incorporation of blood group A2 screening in routine blood grouping tests which will help in reducing the risk of blood transfusion reactions and HDN. RISKS There is no risks associated with participation in this study since only your already donated blood will be used. I invite you to participate in this study by allowing us to use 2mls of your donated blood (if group A or AB) to screen for antigen A2 and subsequently, anti-A1. Please note that taking part in this study is entirely voluntary and strictly confidential. You may refuse to take part or withdraw from the study without any objection or penalty. I assure you that your blood sample will be used only for the purpose of this research. Contacts: 99 University of Ghana http://ugspace.ug.edu.gh You can contact the investigator, Evans Owusu Agyapong, for any clarification on the study on 0247566099 I (participant’s name) ……………………………………………………….... have understood the nature and the purpose of the study and have agreed to participate in the study. ………………………………… ……………………… …………………… Signature/Thumb print Date Telephone (Your signature or thumb print above indicate your willingness to participate in this study) Witnessed by (name).......................... Signature/Thumb print............................ Date................ ……………………………………………… ................................... ................. Full name of investigating team member Signature Date 100 University of Ghana http://ugspace.ug.edu.gh APPENDIX 4: Data collection form COLLEGE OF HEALTH SCIENCES, UNIVERSITY OF GHANA BLOOD GROUP A2 PROJECT DATA COLLECTION FORM FORM NO……........ 1.1 Participant’s ID PATIDNM 1.2 Date of visit: DATEVIST 1.3 Participant’s age: PATAGE 1.4 Participant’s gender: PATGEN 1.5 Participant’s nationality PATNAT 1.6 Participant’s residence PATRSD 1.7 First time Donor? YES NO PATAGE 1.8.0 Retained Donor? YES NO PATGEN 1.8.1 If retained donor, number of donations done: 0-5 6-10 >10 DONTYP 1.9 Status 1. Consented 2. Refused 3. Withdrawn STATUS 2.0 SAMPLE COLLECTION 2.1 Was sample collected? 1. Yes 2. No SMPLCLN 2.2 If No, reason 1. Refusal 2. NA 3. Other, specify SMPNCD 101 University of Ghana http://ugspace.ug.edu.gh APPENDIX 5: Washing of red blood cells 1ml of the cells was transferred into a 10ml capacity test tube (12x75mm) which was topped up with saline (0.9%) and centrifuged at 3500rpm for two minutes. The supernatant was decanted and the process was repeated three times after which the last clear supernatant was completely aspirated using a clean pipette. 102 University of Ghana http://ugspace.ug.edu.gh APPENDIX 6: Preparation of 3% red cell suspension 0.2mls of the previously washed packed red blood cells was dispensed into a 12x75mm test tube containing 2.8mls of normal saline (0.9%). The tube was capped and its contents mixed thoroughly by gently inverting it continuously several times. 103 University of Ghana http://ugspace.ug.edu.gh APPENDIX 7: REAGENTS USED A1 Reagent red cells (Lot No: R0123291, NHS Blood and Transplant, Liverpool, UK) A2 Reagent red cell (Lot No: R0223291, NHS Blood and Transplant, Liverpool, UK) B Reagent red cell (Lot No: R0333291, NHS Blood and Transplant, Liverpool, UK) Anti-A (Lot No: A1116-24A, Fortress diagnostics Ltd, UK) Anti-B (Lot No: E61040-B, Fortress diagnostics Ltd, UK) Anti-D (lot No. D116-16B, Fortress diagnostics Ltd, UK) Anti-A1 lectin (Lot No: 116117-B4, Lorne laboratories Ltd, UK) 104