ARTICLE
https://doi.org/10.1038/s42003-021-02511-5 OPEN
Neutrophils dominate in opsonic phagocytosis of
P. falciparum blood-stage merozoites and protect
against febrile malaria
Asier Garcia-Senosiain 1,2,10, Ikhlaq Hussain Kana 1,2,10, Subhash Singh 3,9✉, Manoj Kumar Das4,
Morten Hanefeld Dziegiel 5,6, Sanne Hertegonne 2,7, Bright Adu 8✉ & Michael Theisen 1,2✉
Antibody-mediated opsonic phagocytosis (OP) of Plasmodium falciparum blood-stage mer-
ozoites has been associated with protection against malaria. However, the precise con-
tribution of different peripheral blood phagocytes in the OP mechanism remains unknown.
Here, we developed an in vitro OP assay using peripheral blood leukocytes that allowed us to
quantify the contribution of each phagocytic cell type in the OP of merozoites. We found that
CD14 + +CD16− monocytes were the dominant phagocytic cells at very low antibody levels
and Fc gamma receptor (FcγR) IIA plays a key role. At higher antibody levels however,
neutrophils were the main phagocytes in the OP of merozoites with FcγRIIIB acting syner-
gistically with FcγRIIA in the process. We found that OP activity by neutrophils was strongly
associated with protection against febrile malaria in longitudinal cohort studies performed in
Ghana and India. Our results demonstrate that peripheral blood neutrophils are the main
phagocytes of P. falciparum blood-stage merozoites.
1 Department for Congenital Disorders, Statens Serum Institut, Copenhagen, Denmark. 2 Centre for Medical Parasitology at Department of International
Health, Immunology and Microbiology, University of Copenhagen and Department of Infectious Diseases, Copenhagen University Hospital, Rigshospitalet,
Copenhagen, Denmark. 3 Indian Institute of Integrative Medicine, Jammu, India. 4 National Institute of Malaria Research, Field Unit, Ranchi, Jharkhand, India.
5 Blood Bank KI 2034, Department of Clinical Immunology, Copenhagen University Hospital, Copenhagen, Denmark. 6 Department of Clinical Medicine,
University of Copenhagen, Copenhagen, Denmark. 7 Ghent University, Ghent, Belgium. 8Noguchi Memorial Institute for Medical Research, College of Health
Sciences, University of Ghana, Accra, Ghana. 9Present address: Consytel Life Sciences Pvt. Ltd., Hyderabad, India. 10These authors contributed equally: Asier
Garcia-Senosiain, Ikhlaq Hussain Kana. ✉email: subhash0974@gmail.com; BAdu@noguchi.ug.edu.gh; mth@ssi.dk
COMMUNICATIONS BIOLOGY |           (2021) 4:984 | https://doi.org/10.1038/s42003-021-02511-5 | www.nature.com/commsbio 1
1234567890():,;
ARTICLE COMMUNICATIONS BIOLOGY | https://doi.org/10.1038/s42003-021-02511-5
B lood-stage Plasmodium spp. infection is responsible for the stained merozoites were quantified by flow cytometry. We foundclinical manifestations of malaria and is characterized by that although both the neutrophils and monocytes were active inintraerythrocytic replication of asexual parasites and the the phagocytosis of merozoites, neutrophils were more active
release of free merozoites into circulation to invade new despite the antibody type (i.e., IP or NIP) used to opsonize
erythrocytes1. Free merozoites are more vulnerable to immune merozoites (Fig. 1b). Importantly, median merozoite-
factors, including antibodies, complement, and leukocytes that phagocytosis by neutrophils in presence of IP was markedly
protect the host (reviewed in2–4). Passive transfer of IgG anti- enhanced compared to NIP (68-fold) and no plasma (unopso-
bodies from individuals with naturally acquired immunity (NAI) nized [UOP]) (95-fold) controls. Moreover, opsonization with IP
against malaria to Plasmodium falciparum-infected patients could led to a substantial shift (Wilcoxon signed-rank test, P= 0.002) in
control parasiteamia and alleviate fever5,6. Malaria-specific IgG the relative distribution in the phagocytic fraction with
antibodies may exert their biological function either by directly neutrophils constituting 94% and monocytes only 6% as
preventing merozoite invasion of erythrocytes7,8 or in co- compared to 70 and 30%, respectively, in the unopsonized
operation with immune effector cells through cell-surface Fc condition (Fig. 1b, lower panel). This indicates that the opsonic
gamma receptors (FcγR)9,10. Engagement of FcγRs triggers phagocytosis of blood-stage merozoites by the neutrophils may be
potent immune cell mechanisms resulting in the release of soluble a potent immune protective mechanism against blood-stage
factors11 and/or phagocytosis12–14 leading to parasite killing. Of malaria infection. Next, to estimate the capacity of each
these mechanisms, the opsonic phagocytosis (OP)12–14 has by far phagocyte to take up UOP or antibody-opsonized merozoites,
been the most consistent correlate of protection against malaria in phagocytosis indices (%) were calculated for both the neutrophils
sero-epidemiological studies. However, all the bioassays used in and monocytes (Fig. 1c). The median percentage of merozoite-
studying these mechanisms have mostly relied on the use of phagocytosis by neutrophils was found to be significantly higher
preselected cell types, such as monocytes11,15 or neutrophils11,16, (1.4-fold) than that of monocytes (Wilcoxon signed-rank test,
or cell lines belonging to definite lineages13,14,16,17, which may P= 0.002). Merozoite-phagocytosis by both the neutrophils and
not clearly depict the peripheral blood environment. In a recent monocytes were abolished by pretreatment of cells with
study, peripheral blood neutrophils in a whole-leukocyte assay cytochalasin D (Cyt D), a reliable inhibitor of phagocytosis
were shown to play a prominent role in the clearance of P. fal- (Fig. 1c). To determine the kinetics of merozoite OP by PBLs, we
ciparum sporozoites demonstrating their role in pre-erythrocytic investigated the time course for the uptake of opsonized
stage malaria immunity18. The role of the different peripheral merozoites by each phagocyte (Fig. 1d). IP-opsonized merozoites
blood phagocytes in opsonic phagocytosis of blood-stage P. fal- were avidly internalized by both the neutrophils and monocytes
ciparum merozoites and their overall contribution to blood-stage with a mean of 85% of neutrophils and 84% of monocytes
malaria immunity is yet to be described using such unbiased phagocytosing merozoites by the first 1 min (Fig. 1d). Merozoite-
in vitro bioassays. Here, using an in vitro phagocytosis assay, phagocytosis could be abolished by Cyt D, demonstrating that
which included all peripheral blood leukocytes (PBL) (in pro- opsonized merozoites become actively ingested shortly after
portion to their normal composition in peripheral blood), we contact with the respective phagocytes (Fig. 1d).
investigated the role of different cell types and their respective Of the monocyte subsets, we found that classical (CD14 + +
FcγRs in IgG-mediated phagocytosis of blood-stage P. falciparum CD16−) monocytes contributed most to the merozoite-
merozoites. Using our assay, we identified neutrophils and phagocytosis in our assay compared to nonclassical (CD14 +
monocytes as the two major phagocytic cell types, and that the CD16 + ) and intermediate (CD14 + +CD16 + ) monocytes
levels of antibodies deposited on the merozoite surface (as indi- (Supplementary Figure 1a and b). Interestingly, the less abundant
cated by IFA titers) determined which cell type played the CD14 + CD16 + and CD14 + +CD16 + subsets were the most
dominant phagocytic role. Higher levels of antibody deposition efficient phagocytes on a per cell basis (Supplementary Figure 1c).
on merozoites (as indicated by higher IFA titers), stimulated Taken together, we conclude here that neutrophils are the main
neutrophils to play a dominant phagocytic role, while with lower phagocytes of antibody-opsonized merozoites in peripheral blood
levels of antibody deposition on merozoites (as indicated by lower while classical monocytes dominate the monocytic contribution
IFA titers) the monocytes played the dominant phagocytic role. due to their relatively higher numbers although not being the
Further, using the phagocytosis assay we tested the association most efficient phagocytes on per cell basis.
between IgG-mediated OP of blood-stage P. falciparum mer-
ozoites and protection against febrile malaria in two different
longitudinal cohort studies performed in Ghana and India. The degree of antibody opsonization influences merozoite-
phagocytosis by neutrophils and monocytes. To further char-
acterize merozoite-phagocytosis by PBLs, we investigated whether
Results the degree of antibody opsonization influenced phagocytosis. The
Neutrophils are the major cell types involved in IgG-mediated merozoites were opsonized with increasing dilutions of immune
phagocytosis of P. falciparum blood-stage merozoites. To plasma (n= 21) and added separately to PBLs from a single
investigate IgG-mediated phagocytosis of P. falciparum blood- donor. There was a clear dose-dependent effect on merozoite-
stage merozoites, PBLs from whole blood samples were used to phagocytosis by neutrophils and monocytes demonstrating that
ensure that the different leukocytes (lymphocytes, monocytes, phagocytosis depends on the amount of antibody deposited on
and granulocytes) were present in their natural physiological the merozoite surface for which we relied on quantification
proportions in the phagocytosis assays. through IFA titers (Fig. 2a). Overall, neutrophils were the main
The Danish donor PBLs used contained approximately 6% phagocytes of antibody-opsonized merozoites, thus confirming
monocytes, 59% neutrophils, 32% lymphocytes, and 3% eosino- the above results. Interestingly, there seems to be a shift in the
phils and basophils (Fig. 1a). For IgG source, we used de- relative contribution of neutrophils and monocytes to the overall
complemented human plasma samples at 1:8000 dilutions from phagocytosis with decreasing amounts of opsonizing antibodies
malaria immune individuals (immune plasma [IP]); or malaria observed at increasing plasma dilution levels (Fig. 2a). At high
nonexposed Danish donors (nonimmune plasma [NIP]) as in our antibody levels neutrophils are the main phagocytes, while the
initial experiments complement did not play a significant role in contribution of monocytes in merozoite-phagocytosis became
the OP of merozoites. PBLs that phagocytosed ethidium bromide- more prominent at lower levels of opsonizing antibodies at higher
2 COMMUNICATIONS BIOLOGY |           (2021)4 :984 | https://doi.org/10.1038/s42003-021-02511-5 | www.nature.com/commsbio
COMMUNICATIONS BIOLOGY | https://doi.org/10.1038/s42003-021-02511-5 ARTICLE
Fig. 1 Malarial antibodies induce merozoite-phagocytosis by neutrophils and monocytes. a Percentage of individual leukocytes in PBLs from Danish
blood donors (n= 10) was enumerated by flow cytometry. Horizontal lines represent median values. b From each donor, 6 × 104 PBLs were incubated with
unopsonized merozoites (UOP), merozoites opsonized with a pool of nonimmune plasma (NIP), or with immune plasma (IP) for 30min at 37 °C for
phagocytosis to occur. Cells were stained with anti-CD14, CD16, CD45, and CD66b antibodies and analyzed by flow cytometry to enumerate ethidium
bromide (EtBr)-positive cells. The mean number (no.) of neutrophils (blue) and monocytes (red), which ingested merozoites are shown in the upper panel,
and the percent contribution of neutrophils and monocytes to overall PBL phagocytosis is shown in lower panels for the given opsonized/unopsonized
condition. c Mean percentage of monocytes or neutrophils in PBL preparations that phagocytosed merozoites in the presence of NIP, IP, or UOP condition.
The percent phagocytosis of IP-opsonized merozoites is also shown for cytochalasin D (Cyt D) treated PLBs. d Kinetics of IP-opsonized merozoite uptake
by monocytes and neutrophils present in Cyt D pretreated or untreated PBLs. Phagocytosis was stopped at indicated time points, and percent phagocytosis
by the neutrophils and monocytes is expressed as the mean of triplicates. Data are represented from one of the two independent assays. Results were
obtained with PBLs from 10 different Danish blood donors (a–c). P values were determined by Wilcoxon signed-rank test b and c. Asterisks represent
P values (**P < 0.01).
COMMUNICATIONS BIOLOGY |           (2021) 4:984 | https://doi.org/10.1038/s42003-021-02511-5 | www.nature.com/commsbio 3
ARTICLE COMMUNICATIONS BIOLOGY | https://doi.org/10.1038/s42003-021-02511-5
Fig. 2 Merozoite-phagocytosis by PBLs depends on the degree of antibody opsonization. Merozoites were opsonized with serial dilutions of individual
plasma (n= 21) from Ghanaian children a or a pool of the same plasma b and c and incubated with PBLs (6 × 104 per well) for 30min at 37 °C for
phagocytosis to occur. Cells were stained with anti-CD14, CD16, CD45, and CD66b antibodies and analyzed by flow cytometry to enumerate EtBr-positive
cells. a Left y-axis shows the median number (no.) of EtBr-positive neutrophils (blue solid line) and monocytes (red solid line) present in PBLs in the
presence of individual plasma samples (n= 21). Right y-axis shows the median percentage of EtBr neutrophils (dotted blue line) and monocytes (dotted
red line) in EtBr-positive PBL fraction in the presence of individual plasma samples (n= 21). b Mean percent of EtBr-positive neutrophils and monocytes in
EtBr-positive PBL fraction in the presence of pooled plasma is plotted against the amount of antibody deposited on the merozoite surface as quantified by
the merozoite-immunofluorescence assay (IFA). c Mean percentage of EtBr-positive neutrophils and monocytes present in PBL preparations either
untreated or pretreated with FcγR blockers in the presence of pooled plasma is plotted against merozoite-IFA. The experiments in b and c were performed
in triplicates.
plasma dilutions (generalized estimating equation, P < 0.0001). being the most efficient phagocytic cell at this low antibody
Next, the dose-titration experiment was repeated with the pool of concentration. Importantly, merozoite-phagocytosis by both of
the immune plasma tested above. The phagocytic activity of PBLs these cell types could be abolished by the blocking of all surface-
was plotted against the amount of malarial IgG deposited on the exposed FcγRs using a combination of anti-CD16, anti-CD32,
merozoite surface as determined by merozoite-IFA (Fig. 2b). As and anti-CD64 antibodies prior to their use in the phagocytosis
expected, the neutrophils were the main phagocyte at high anti- assay, demonstrating that phagocytosis is FcγR-dependent at all
body concentrations (merozoite-IFA titer >104), while monocytes IgG concentrations tested (Fig. 2c).
became the dominant phagocyte at lower antibody concentrations Collectively, these findings suggest that opsonization of
corresponding to <10% saturation of the merozoite surface merozoites leads to prominent phagocytosis by monocytes at
(Fig. 2b). Nonimmune plasma promoted low (<10%) phagocy- low IgG densities, whereas neutrophils are the dominant
tosis activities at all antibody concentrations. We observed that phagocytes at higher IgG densities, as reflected by the
with higher dilutions of plasma samples corresponding to change merozoite-IFA titer.
in IFA titers from 105 to 103 (A.U.), there was a distinct change in
the phagocytic activities of neutrophils and monocytes (Fig. 2b).
While at higher IFA titers >105 neutrophils were observed to be IgG-mediated merozoite-phagocytosis by neutrophils and
the dominant phagocytic cells, with higher plasma dilutions monocytes depend on engagement of distinct FcγRs. To dissect
resulting in IFA titers <104 the phagocytic activity of neutrophils the role of FcγR signaling in merozoite-phagocytosis by blood
decreased linearly to near zero on a per cell basis (Fig. 2c). phagocytes, we investigated PBLs from 19 blood donors. We used
Interestingly, the monocytes which displayed a less dominant anti-CD16 (FcγRIIIB), anti-CD32 (FcγRIIA), and anti-CD64
phagocytic role at higher IFA titers >105, displayed a dominant (FcγRI) antibodies to block individual FcγRs. Median merozoite-
phagocytic role at higher plasma dilutions resulting in IFA titers phagocytosis activity by neutrophils was reduced by 90% as
<104, and further retained robust phagocytic activity on a per cell compared to isotype control when the interaction of the
basis even at a higher dilution of 103 (Fig. 2c). The shift in the merozoite-antibodies with FcγRIIIB was blocked (Friedman test,
relative contribution of phagocytes is explained by monocytes P= 0.0009) (Fig. 3a). Phagocytosis by neutrophils was unaffected
4 COMMUNICATIONS BIOLOGY |           (2021) 4:984 | https://doi.org/10.1038/s42003-021-02511-5 | www.nature.com/commsbio
COMMUNICATIONS BIOLOGY | https://doi.org/10.1038/s42003-021-02511-5 ARTICLE
Fig. 3 Phagocytosis of antibody-opsonized merozoites depends on FcγRIII in neutrophils and FcγRII in monocytes. The PBLs (n= 19) were treated with
10 µg/ml of antibodies against CD16, CD32, CD64, or an isotype control for 30min and then incubated with IP-opsonized merozoites for phagocytosis to
occur. Graphs show the relative phagocytosis (%) of specific FcγR blocker treated neutrophils a and monocytes b using the untreated homologous
phagocytes as reference. Boxes indicate the median and interquartile range. Whiskers mark the minimum and maximum values. P values were determined
by the Friedman test and Dunn’s multiple comparisons test. Asterisks represent P values (*P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001).
by the addition of anti-CD32 or anti-CD64 antibodies suggesting phagocytosis, 22.1%; IQR, 5.4–38%), respectively increased with
that FcγRIIIB is the main receptor for phagocytosis by neu- the age of children, reflective of cumulative exposure to malaria.
trophils (Fig. 3a). Interestingly, simultaneous blocking of both the This was statistically significant for both cell types; however, the
FcγRIIIB and FcγRIIA in neutrophil preparations further reduced stronger association was observed for neutrophils (Odds ratio
the median merozoite-phagocytosis activity by 61% (Wilcoxon [OR]=20.4; 95% CI= 4.3–96.7; P= 0.0001) compared to
signed-rank test, P= 0.0010) suggesting that these receptors monocytes (OR= 6.8; 95% CI= 2.5–18.1; P= 0.0001).
complement each other in IgG-mediated phagocytosis by neu- Next, children were categorized into two equal groups based on
trophils under the conditions tested here. the median phagocytosis value for the respective phagocytes to
The data also demonstrated that blocking antibodies specific examine the risk of symptomatic malaria for each phagocytosis-
for CD32 inhibited median merozoite-phagocytosis by monocytes responder group using the Cox-regression models to calculate
by about 42% compared to isotype control (Friedman test, hazard ratios (HRs). Children in the high phagocytosis group for
P= 0.0015). Blocking antibodies specific for CD16 and CD64 had both neutrophils (unadjusted [uHR] = 0.48; 95% CI= 0.31–0.75;
no such effects (Fig. 3b) suggesting that FcγRIIA but not P= 0.001) and monocytes (uHR= 0.51; 95% CI= 0.33–0.80;
FcγRIIIB and FcγRI is necessary for merozoite-phagocytosis by P= 0.003) had a significantly lower risk of febrile malaria over a
monocytes. 42-week follow-up period compared to those with low-level
Altogether, these data suggest that FcγRIIIB (in synergism with responses (Fig. 4b). The strength of protective association was
FcγRIIA) is a major mediator of merozoite-phagocytosis by similar between neutrophils and monocytes. Adjusting the Cox-
neutrophils while FcγRIIA alone seems to perform this function regression models for age, as it is correlated with parasite
in monocytes. exposure and febrile malaria outcome in this cohort20, marginally
influenced the magnitude of protective associations (Fig. 4b).
Acquisition of IgG capable of mediating opsonic phagocytosis Next, we repeated the phagocytosis assays using highly purified
by PBLs is associated with protection against febrile malaria. neutrophils (>99%; obtained from a Danish blood donor) and
To investigate the association between antibody-mediated mer- merozoites opsonized with the same Ghanaian plasma samples
ozoite-phagocytosis by human PBLs and protection against feb- (n= 140)19 (Supplementary Table 1). Children in the high
rile malaria, we used clinical data and plasma samples (n= 140) purified neutrophil-phagocytosis group had a significantly
from a well-established longitudinal cohort survey (LCS) per- (uHR= 0.43; 95% CI, 0.28–0.68; P < 0.001 and age-adjusted
formed in Ghanaian children19 (Supplementary Table 1). The (aHR)= 0.47; 95% CI, 0.30–0.75; P= 0.001) higher probability of
PBLs used here in phagocytosis assays were obtained from a remaining free of malaria during the study period than those in
Danish blood donor. A wide range of phagocytic responses was the low purified neutrophil-phagocytosis group. Thus, supporting
observed for both neutrophils (median number of phagocytosing the importance of neutrophils in parasite-killing mechanisms.
cells 1154; interquartile range (IQR), 65–2717), and monocytes To further extend these findings and to strengthen the finding
(median number of phagocytosing cells, 578; IQR, 147.3–1004), that neutrophils are important for NAI against malaria, we
and phagocytosis levels with neutrophils was significantly higher comparatively assessed plasma samples (n= 121) in PBL-
compared to monocytes (Wilcoxon matched-pairs signed-rank phagocytosis assay and used clinical data from a LCS conducted
test, P < 0.0001) (Fig. 4a). For most of the plasma samples, neu- in India21,22 (Supplementary Table 1). In general, neutrophils in
trophils were the main phagocyte of merozoites due to their PBL preparations (obtained from a Danish blood donor) showed
greater numbers and apparently stronger phagocytic efficiency higher phagocytosis activity than monocytes (Fig. 4c). Cox
(Fig. 4a). Overall, phagocytic activities for neutrophils (median proportional-hazard models used to determine estimates of
phagocytosis, 24.6%; IQR, 1.6–56.7%) and monocytes (median protection confirmed that subjects in high phagocytosis group
COMMUNICATIONS BIOLOGY |           (2021) 4:984 | https://doi.org/10.1038/s42003-021-02511-5 | www.nature.com/commsbio 5
ARTICLE COMMUNICATIONS BIOLOGY | https://doi.org/10.1038/s42003-021-02511-5
Fig. 4 Neutrophil-phagocytosis of merozoites is dominant compared to monocytes and predicts protection against febrile malaria. Paired aligned dot
plots coupled to box and whiskers plots showing the number of EtBr-positive neutrophils and monocytes present in PBLs from Danish blood donors which
were used as effector cells to test the phagocytosis mediating ability of antibodies present in Ghanaian a and Indian c cohort plasma samples. Both the
Ghanaian children (n= 140; protected= 59, susceptible= 81) b and the Indian participants (n= 121; protected= 71, susceptible= 50) d were categorized
into two equal groups based on the median phagocytosis values obtained with neutrophils or monocytes, and the risk of encountering febrile malaria during
the follow-up periods was calculated with Cox-regression models comparing those with high versus low (reference group) phagocytosis. Values represent
age-adjusted (filled triangles) and unadjusted (filled squares) hazard ratios with 95% confidence intervals. Boxes indicate the median and interquartile
range. Whiskers mark the minimum and maximum values. P values were determined by Wilcoxon signed-rank test a and c. Asterisks represent P values
(****P < 0.0001).
for neutrophils (aHR= 0.35; 95% CI= 0.19–0.66; P= 0.001) but of P. falciparum extracellular merozoites. We have focused on
not monocytes (aHR= 0.71; 95% CI= 0.40–1.25; P= 0.23) had a asexual blood-stage parasites because they are clinically important
significantly lower risk of symptomatic malaria over the follow-up and responsible for all symptoms associated with P. falciparum
period compared to those with low-level responses (Fig. 4d). malaria. Using an in vitro phagocytosis assay, which included
Collectively, these findings reinforce the importance of physiological proportions of all peripheral blood leukocytes from
merozoite-phagocytosis in NAI in both Africa and Asia and healthy donors, we demonstrated that both the human neu-
points to a particularly important and dominant role of trophils and monocytes were highly active in the phagocytosis of
peripheral blood neutrophils. extracellular merozoites. We found that neutrophils were the
dominant phagocyte of merozoites in our in vitro assay.
Relationship between antibody-mediated merozoite-phagocy- In some respects, this is not surprising, as neutrophils are the
tosis by PBLs and antibody-dependent cellular inhibition most abundant (50-70%) leukocytes in human blood-formingthe first line of defence following an infection. They possess a
activity in Ghanaian children. The antibody-dependent cellular
inhibition (ADCI) assay measures the parasite-killing capacity of range of antimicrobial effector functions like respiratoryburst activity, which releases reactive oxygen species (ROS).
antibodies in collaboration with human monocytes. We, there-
fore, compared the phagocytosis activity of PBLs with the ADCI ROS are highly toxic to intraerythrocytic malaria parasitedevelopment24–26 and neutrophil-mediated antibody-depen-
activity previously reported for the same IgG preparations
(n= 85) used here23. There was no correlation between dent respiratory burst (ADRB) is associated with protectionagainst clinical malaria16. However, it is quite surprising that
merozoite-phagocytosis by neutrophils (Spearman r=−0.04;
P= 0.68) or monocytes (Spearman r=−0.06; P= 0.5) and the despite such aforementioned potent antimicrobial properties,the implications of IgG-mediated phagocytosis by neutrophils
ADCI activity suggesting that phagocytosis by both the mono-
cytes and neutrophils are distinct mechanisms and different from in immunity against blood-stage malaria parasites have beenquite understudied27–30.
that of ADCI (Fig. 5). In an attempt to characterize effector functions of specific
phagocytes in blood, several factors were investigated, which may
Discussion potentially influence their capacity for phagocytosis of mer-
Here we have taken an unbiased approach to investigate the role ozoites. Initially, we demonstrated that the degree of antibody
of different blood leukocytes in antibody-mediated phagocytosis opsonization was important for merozoite-phagocytosis by both
6 COMMUNICATIONS BIOLOGY |           (2021) 4:984 | https://doi.org/10.1038/s42003-021-02511-5 | www.nature.com/commsbio
COMMUNICATIONS BIOLOGY | https://doi.org/10.1038/s42003-021-02511-5 ARTICLE
Fig. 5 ADCI activity is not associated with merozoite-phagocytosis by neutrophils or monocytes. Scatterplots showing the relationship between ADCI
activity and merozoite-phagocytosis by neutrophils a and monocytes b in a Ghanaian cohort of children. Here, the data are presented for 85 children
(protected [red dots]= 33, susceptible [blue dots] = 52) for whom both phagocytosis and ADCI activity were available. Spearman’s correlation coefficient
r, with associated P value is shown for both protected (red font) and susceptible (blue font) children in each plot.
the neutrophil and monocyte cell populations. However, contrary bind to FcγRIIA, which has a 10- to 40-fold higher affinity for
to expectations, we observed that monocytes were actively IgG compared to FcγRIIIB34. Thus, favoring FcγR activation on
ingesting merozoites at lower antibody concentrations compared monocytes and phagocytosis. Higher antibody concentrations
to neutrophils suggesting that the threshold of antimerozoite in plasma leads to a proportional change in the amount of IgG
antibodies to mediate OP for peripheral monocytes is lower than deposited on the merozoite surface thereby enhancing
that of peripheral neutrophils. Once the threshold of anti- engagement of FcγRIIIB on neutrophils and subsequently
merozoite antibodies for phagocytosis by neutrophils has been phagocytosis. The very high density of FcγRIIIB (⁓1.4 × 106
reached, these cells became the dominant phagocyte accounting receptors per cell) on neutrophils may further lead to extensive
for 94% of the phagocytic fraction. Thus, emphasizing the receptor clustering at the membrane35 and enhanced activation
importance of neutrophils for the elimination of P. falciparum and merozoite-phagocytosis.
blood-stage parasites. To investigate the role of neutrophils in NAI against clinical
Since IgG antibodies mediate cellular immune effector malaria, we have used a well-characterized LCS to show that the
pathways through binding to FcγRs encoded by different genes neutrophil-phagocytosis activity of IgG is significantly associated
we examined the role of specific FcγRs on merozoite- with reduced risk against febrile malaria in Ghanaian children.
phagocytosis. Human monocytes and neutrophils both Our finding that children whose antibodies showed merozoite-
express FcγRIIA with a subset of monocytes expressing phagocytosis activities by neutrophils above the median had a
FcγRIIIA while FcγRIIIB expression is exclusive to 48% (effect of age accounted) reduced risk of febrile malaria
neutrophils31. We found that anti-CD16 antibodies efficiently compared to children with activities below the median strongly
blocked merozoite-phagocytosis demonstrating that FcγRIIIB is supports the notion that the neutrophils play an important role in
the main receptor for antibody-mediated phagocytosis by malaria immunity. This association was observed in two inde-
neutrophils. Interestingly, anti-CD32 antibodies enhanced the pendent experiments using both the PBLs and highly purified
blocking activity of anti-CD16 antibodies suggesting that (>99%) neutrophils. These findings corroborate a recent study
FcγRIIIB and FcγRIIA act in concert. FcγRIIIB is bound to the that showed that neutrophil opsonic phagocytosis of sporozoites
cell membrane through its glycosylphosphatidylinositol (GPI) is important in malaria immunity and suggest that the role of
anchor and is completely lacking a signal module or signaling neutrophils in controlling parasitemia may transcend different
capacity by itself23. Therefore, FcγRIIIB most likely requires stages of the parasite18. Since the PBL-phagocytosis assay devel-
help from other membrane protein(s) with signaling capacity. oped here measures the phagocytic activity of all leukocytes, our
Some studies have suggested FcγRIIIB may act synergistically analysis also demonstrated that the phagocytosis promoting
with activating receptors such the ITAM-bearing single-chain activity of monocytes was significantly associated with a reduced
FcγRIIA for signaling32, thus providing a plausible explanation risk of febrile malaria in this Ghanaian cohort. While the role of
for the synergy between anti-CD16 and anti-CD32 antibodies monocytes in NAI against malaria is well established12–14,23, we
in merozoite-phagocytosis observed here. A recent report has show here that antibody-mediated phagocytosis of blood-stage P.
further demonstrated that human neutrophils may express low falciparum by neutrophils is also involved in NAI. Due to their
levels of FcγRIIIA33. Whether FcγRIIIB and FcγRIIIA might greater numbers, we predict that neutrophils are the main effector
cooperate in merozoite-phagocytosis by neutrophils remains to cells that control blood-stage malaria parasite multiplication
be investigated. Our blocking experiments also demonstrated in vivo a notion, which may also hold true for pre-erythrocytic
that FcγRIIA was the sole receptor for mediating the phago- stage infections18.
cytosis of merozoites by monocytes. We, therefore, hypothesize Since NAI against malaria is most likely a composite of mul-
that differences in FcγR density, as well as specificity and affi- tiple antiparasite protective mechanisms we sought to compare
nity toward IgG subclasses might explain the different thresh- the data obtained here with previous data from the same LCS. We
olds for merozoite-phagocytosis by neutrophils and monocytes, found no correlation between merozoite-phagocytosis by either
respectively. At low antibody concentrations (merozoite-IFA monocytes or neutrophils and ADCI confirming our past
titer <104 in our PBL assay), opsonized merozoites will initially observation that there was no significant relationship between
COMMUNICATIONS BIOLOGY |           (2021) 4:984 | https://doi.org/10.1038/s42003-021-02511-5 | www.nature.com/commsbio 7
ARTICLE COMMUNICATIONS BIOLOGY | https://doi.org/10.1038/s42003-021-02511-5
phagocytosis and ADCI20. Collectively these data suggest Leukocyte purification. Peripheral blood leukocytes (PBLs) were isolated from
that phagocytosis and ADCI are distinct parasite-killing whole blood samples from Danish donors by centrifugation at 800 g for 25 min.
mechanisms. The layer containing leukocytes was harvested and the contaminating red blood
cells were lysed by mixing with 10 parts of lysis buffer (155 nM NH Cl, 10 mM
While this study has focused on FcγR-mediated phagocytosis, 4KHCO3, and 0.1 mM EDTA; pH 7.4) and incubating for 10 minutes. After washing
yet other studies have suggested that clearance of merozoites may once with PBS, leukocytes were resuspended in cell medium (RPMI-1640 supple-
also involve complement36,37. The current study suggested that mented with 10% fetal bovine serum, 25 mM HEPES, 4 mM L-glutamine, and
complement did not enhance antibody-mediated phagocytosis of 25 µg/ml gentamicin). Cells were counted using a hemocytometer and diluted with
cell medium to a concentration of 6 × 105 cells/ml.
merozoites. We did not further investigate the potential role of For neutrophil purification, samples from healthy donors were carefully layered
complement-receptor(s) in merozoite clearance since this was on top of an isotonic Percoll solution with a density of 1.077 g/ml and centrifuged
outside the scope of the present study. at 800 g for 25 min. The leukocytes were resolved into two distinct bands. The
This study has identified both the monocytes and neutrophils upper layer containing monocytes and the lower one containing neutrophils. The
as two major peripheral blood leukocytes, with different phy- contaminating red blood cells were lysed as described above. Neutrophils werefurther purified using the EasySep Human Neutrophil Isolation Kit (StemCell
siological properties, contributing to IgG-mediated phagocy- Technologies) following the manufacturer’s instructions. Purified cells were
tosis of malaria parasites, which play role in immunity against resuspended in cell medium, counted, and diluted with cell medium to a
malaria. Altogether, our results provide an important glimpse concentration of 6 × 105 cells/ml.
into the various facets of multiple IgG-mediated cellular
responses contributing to the phenomenon of antibody- Parasite culture and merozoite isolation. P. falciparum strain NF54 was cultured
mediated NAI to malaria. The neutrophils constitute the at 4% hematocrit in parasite growth medium (RPMI-1640 supplemented with25 mM HEPES, 4 mM L-glutamine, 5 g/l AlbuMAX, 0.02 g/l hypoxanthine, and
largest myeloid cell subset, and are considered short-lived 25 µg/ml gentamicin). The culture was synchronized by treating with 5% sorbitol
frontline responders against pathogens; however, recent evi- for 10 min. Mature trophozoites/early schizonts were purified with a magnetic
dence suggests they have the capacity to prime adaptive separation unit and cultured further. The merozoites were obtained by filtering
immune responses during phagocytosis38. The monocytes on mature schizonts using a 1.2 µm filter. After removing hemozoin, merozoites were
stained with 10 µg/ml of ethidium bromide for 30 min and washed twice with
the other hand appear to have comparatively lower phagocytic fluorescence-activated cell sorting (FACS) buffer (PBS with 0.5% BSA+ 2 mM
activity than neutrophils but are significantly longer-lived effi- EDTA). Merozoites were opsonized with plasma for 30 min before addition to the
cient antigen-presenting cells. Thus, OP of blood-stage mer- phagocytosis assays. A 1:8,000 dilution of the plasma was used throughout unless
ozoites by both cell types seem to complement each other in the stated otherwise. This dilution corresponds to a merozoite-IFA titer of 40,000, and
it also corresponds to the higher antibody concentration at which neutrophils were
development of protective immunity against malaria, and our found to be the dominant phagocyte. We also used further serial dilutions of
results show that their respective IgG-mediated phagocytic plasma with a lower merozoite-IFA titer of 10,000 at which monocytes were found
activities are associated with protection against febrile malaria to be the major phagocytic cell.
in field settings.
In summary, our study shows that neutrophils contribute Phagocytosis assay. Leukocytes were transferred to 96-well U-bottom plates
4
significantly to IgG-mediated phagocytosis of blood-stage P. fal- containing 6 × 10 cells in 100 µl of cell medium per well. Opsonized, ethidium
ciparum merozoites and protection against febrile malaria. These bromide-stained merozoites (4–6 × 10
5 per well) were washed twice with FACS
buffer, resuspended in cell medium, and 100 µl were added to each leukocyte
findings further elucidate neutrophil antibody-mediated immune containing well. After an incubation of 30 min (unless stated otherwise) at 37 °C
mechanisms against blood-stage malaria parasites. and 5 % CO2, plates were centrifuged in a prechilled centrifuge and washed
twice with ice-cold FACS buffer to stop phagocytosis. Cells were resuspended in
200 µl of cold FACS buffer and incubated for 1 hour at 4 °C with 1:1600 FITC
antihuman CD14 (clone TuK4; Thermo Fisher Scientific MA1-82074), 1:400
Methods BV786 antihuman CD16 (clone 3G8; BD Biosciences 563690), 1:800 APC
Ethics statement. Ethical approval for Danish blood donor samples was given by antihuman CD45 (clone HI30; BD Biosciences 555485), and 1:800 BV421
the Scientific Ethics Committee of Copenhagen and Frederiksberg, Denmark. antihuman CD66b (clone G10F5; BD Biosciences 562940) antibodies. After
Samples from anonymous Danish blood donors (18–60 years of age) obtained for washing thrice with FACS buffer, sample fluorescence was quantified with a
control purposes at Copenhagen University Hospital were used. These individuals CytoFLEX S (Beckman Coulter Life Sciences). Phagocytosis was determined by
are resident of central Copenhagen and provided written consent to have a small measuring the ethidium bromide fluorescence using the 610/20 nm detector.
portion of their blood stored anonymously and used for research purposes. All data Data analysis were performed with Kaluza Analysis Software version 2.1
were analyzed anonymously. (Beckman Coulter Life Sciences). The gating strategy used is shown in Sup-
The Ghanaian LCS was approved by the Institutional Review Board of Noguchi plementary Figure 2.
Memorial Institute for Medical Research of the University of Ghana, Accra, Ghana,
and the Indian LCS was approved by the Institutional Ethics Committee of the
National Institute of Malaria Research, Indian Council of Medical Research, New Blocking of Fcγ receptors. Previous to the incubation with opsonized merozoites,
Delhi, India. Written informed consent was given by the parents and guardians of cells were incubated for 30 min at 37 °C in cell medium with 10 µg/ml of antihu-
children before they were enrolled in the studies. man CD16 (clone 3G8; BD Biosciences 555404), antihuman CD32 (clone FLI8.26;
BD Biosciences 555447), antihuman CD64 (clone 10.1; BD Biosciences 555525), or
an isotype control to block Fcγ receptors 3, 2, and 1, respectively. To block all
surface-exposed FcγRs a combination of anti-CD16, anti-CD32, and anti-CD64
Longitudinal cohort studies. The Indian LCS study was conducted in Dumargarhi antibodies were used. After the incubation period, merozoites were added and the
in the state of Jharkhand22. Briefly, out of 945 individuals (aged 1–82 years) that experiment continued as described above. Phagocytosis values for treated cells are
were enrolled in the study, 386 were sampled at cross-sectional survey-1 and were presented relative to the value of untreated cells incubated with the same immune
followed up actively and passively for malaria case detection for 13-months (April plasma.
2015 to April 2016). In the current study, we focused on 121 participants (aged
3–60 years) who were definitively exposed to P. falciparummalaria (Supplementary
Table 1). Merozoite flow cytometry-based immunofluorescence assay. The amount of
The African study was conducted in Asutsuare, Damgbe, West District, in IgG deposited on the merozoite surface was determined by flow cytometry-based
Ghana19. In total, 798 children under 12 years were enrolled in May 2008. Venous immunofluorescence assay (merozoite-IFA) as described in detail
20. Median
blood was obtained at enrollment and children were followed up actively and fluorescence intensity values are reported.
passively for malaria detection for 42-weeks. Here, we have used available samples
from 140 children (aged 1–12 years) (Supplementary Table 1). At both sites, febrile Statistics and Reproducibility. The Wilcoxon signed-rank test was used to
malaria was defined as any P. falciparum parasitemia confirmed by microscopy of evaluate differences between two groups of paired observations. The Friedman test
stained thick and thin blood smears plus reported fever or axillary temperature with Dunn’s multiple comparisons test was applied to estimate differences between
≥37.5 °C at the time of the visit. Individuals who suffered at least one case of febrile three or more paired groups. Generalized estimating equations were utilized to
malaria during follow-up were considered susceptible (India, 50 [41.3%]; Ghana, 81 account for paired observations in generalized linear models. Association between
[57.9%]), while those who did not experience any episodes of febrile malaria time-to-first febrile malaria episode and the categorized levels of opsonic phago-
despite having parasites at baseline were considered protected (India, 71 [58.7%]; cytosis by neutrophils or monocytes was analyzed by age-adjusted Cox-regression
Ghana, 59 [42.1%]). models. The correlations between opsonic phagocytosis and ADCI activity were
8 COMMUNICATIONS BIOLOGY |          ( 2021) 4:984 | https://doi.org/10.1038/s42003-021-02511-5 | www.nature.com/commsbio
COMMUNICATIONS BIOLOGY | https://doi.org/10.1038/s42003-021-02511-5 ARTICLE
assessed by the Spearman correlation coefficient (r). All statistical analyses were 21. Garcia-Senosiain A. et al. Peripheral merozoite surface proteins are targets of
performed as two-sided tests. P values of less than 0.05 were considered significant. naturally acquired immunity against malaria both in India and in Ghana.
Statistical analysis was performed using Prism 9 (GraphPad Software, Inc.) and R Infect. Immun. 88, e00778–e00719 (2020).
version 3.6.2 with the package geepack39. 22. Das, M. K. et al. Malaria epidemiology in an area of stable transmission in
tribal population of Jharkhand, India. Malar. J. 16, 181 (2017).
Reporting Summary. Further information on research design is available in the Nature 23. Tiendrebeogo, R. W. et al. Antibody-dependent cellular inhibition is
Research Reporting Summary linked to this article. associated with reduced risk against febrile malaria in a longitudinal cohort
study involving ghanaian children. Open Forum Infect. Dis. 2, ofv044 (2015).
24. Allison, A. C. & Eugui, E. M. The role of cell-mediated immune responses in
Data availability resistance to malaria, with special reference to oxidant stress. Annu Rev.
The data that support the findings of this study are available from the corresponding Immunol. 1, 361–392 (1983).
author upon reasonable request and pending agreement from relevant ethics committees 25. Clark, I. A. & Hunt, N. H. Evidence for reactive oxygen intermediates causing
for clinical data. hemolysis and parasite death in malaria. Infect. Immun. 39, 1–6 (1983).
26. Nnalue, N. A. & Friedman, M. J. Evidence for a neutrophil-mediated
Received: 25 May 2021; Accepted: 27 July 2021; protective response in malaria. Parasite Immunol. 10, 47–58 (1988).
27. Boross, P. et al. Destructive arthritis in the absence of both FcgammaRI and
FcgammaRIII. J. Immunol. 180, 5083–5091 (2008).
28. Knackstedt S. L. et al. Neutrophil extracellular traps drive inflammatory
pathogenesis in malaria. Sci. Immunol. 4, eaaw0336 (2019).
29. Aitken, E. H., Alemu, A. & Rogerson, S. J. Neutrophils and malaria. Front.
Immunol. 9, 3005 (2018).
References 30. Amulic, B., Moxon, C. A. & Cunnington, A. J. A more granular view of
1. Mackintosh, C. L., Beeson, J. G. & Marsh, K. Clinical features and neutrophils in malaria. Trends Parasitol. 36, 501–503 (2020).
pathogenesis of severe malaria. Trends Parasitol. 20, 597–603 (2004). 31. Boross P., van de Poel K., Van de Winkel J. G. J. & Leusen J. H. W.
2. Beeson, J. G. et al. Merozoite surface proteins in red blood cell invasion, Encyclopedia of Life Sciences (ELS) (John Wiley & Sons, Ltd 2008).
immunity and vaccines against malaria. FEMS Microbiol. Rev. 40, 343–372 32. Wirthmueller, U., Kurosaki, T., Murakami, M. S. & Ravetch, J. V. Signal
(2016). transduction by Fc gamma RIII (CD16) is mediated through the gamma
3. Hill, D. L., Schofield, L. & Wilson, D. W. IgG opsonization of merozoites: chain. J. Exp. Med. 175, 1381–1390 (1992).
multiple immune mechanisms for malaria vaccine development. Int J. 33. Golay, J. et al. Human neutrophils express low levels of FcgammaRIIIA, which
Parasitol. 47, 585–595 (2017). plays a role in PMN activation. Blood. 133, 1395–1405 (2019).
4. Healer J., Chiu C. Y. & Hansen D. S. Mechanisms of naturally acquired 34. Bruhns, P. et al. Specificity and affinity of human Fcgamma receptors and
immunity to P. falciparum and approaches to identify merozoite antigen their polymorphic variants for human IgG subclasses. Blood. 113, 3716–3725
targets. Parasitology. 145, 1–9 (2017). (2009).
5. McGregor, I. A., Carrington, S. P. & Cohen, S. Treatment of East African P. 35. Goodridge, H. S., Underhill, D. M. & Touret, N. Mechanisms of Fc receptor
falciparum malaria with West African gammaglobulin. Trans. R. Soc. Trop. and dectin-1 activation for phagocytosis. Traffic. 13, 1062–1071 (2012).
Med. Hyg. 57, 170–175 (1963). 36. Reiling, L. et al. Targets of complement-fixing antibodies in protective
6. Cohen, S., McGregor, A. & Carrington, S. Gamma globulin and aquired immunity against malaria in children. Nat. Commun. 10, 610 (2019).
immunity to human malaria. Nature. 192, 733–737 (1961). 37. Boyle, M. J. et al. Human antibodies fix complement to inhibit Plasmodium
7. Hodder, A. N., Crewther, P. E. & Anders, R. F. Specificity of the protective falciparum invasion of erythrocytes and are associated with protection against
antibody response to apical membrane antigen 1. InfectImmun. 69, 3286–3294 malaria. Immunity. 42, 580–590 (2015).
(2001). 38. Meinderts, S. M. et al. Neutrophils acquire antigen-presenting cell features after
8. Holder, A. A. The carboxy-terminus of merozoite surface protein 1: structure, phagocytosis of IgG-opsonized erythrocytes. Blood Adv. 3, 1761–1773 (2019).
specific antibodies and immunity to malaria. Parasitology. 136, 1445–1456 39. Halekoh, U., Højsgaard, S. & Yan, J. The R package geepack for generalized
(2009). estimating equations. J. Stat. Softw. 15, 1–11 (2006).
9. Khusmith, S. & Druilhe, P. Cooperation between antibodies and monocytes
that inhibit in vitro proliferation of Plasmodium falciparum. Infect. Immun.
41, 219–223 (1983). Acknowledgements
10. Pleass, R. J. & Woof, J. M. Fc receptors and immunity to parasites. Trends We thank the study participants, their parents, and guardians for volunteering and
Parasitol. 17, 545–551 (2001). making this work possible. This study was supported by the Danish Ministry of Foreign
11. Bouharoun-Tayoun, H., Attanath, P., Sabchareon, A., Chongsuphajaisiddhi, Affairs (DFC file no. 14-P01-GHA) and the Danish Council for Strategic Research (grant
T. & Druilhe, P. Antibodies that protect humans against Plasmodium number 13127).
falciparum blood stages do not on their own inhibit parasite growth and
invasion in vitro, but act in cooperation with monocytes. J. Exp. Med. 172,
1633–1641 (1990). Author contributions
12. Hill, D. L. et al. Opsonising antibodies to P. falciparum merozoites associated A.G.S. and M.T. designed experiments. A.G.S., I.H.K., and S.H. performed experiments.
with immunity to clinical malaria. PLoS ONE. 8, e74627 (2013). S.S., M.K.D., M.H.D., and B.A. provided samples and clinical data. S.S. and M.T. pre-
13. Osier, F. H. et al. Opsonic phagocytosis of Plasmodium falciparum merozoites: pared the first draft. All authors contributed to discussing the data and proofreading the
mechanism in human immunity and a correlate of protection against malaria. manuscript.
BMC Med. 12, 108 (2014).
14. Kana, I. H. et al. Breadth of functional antibodies is associated with Competing interests
Plasmodium falciparum merozoite phagocytosis and protection against febrile The authors declare no competing interests.
malaria. J. Infect. Dis. 220, 275–284 (2019).
15. Tiendrebeogo, R. W. et al. High-throughput tri-colour flow cytometry
technique to assess Plasmodium falciparum parasitaemia in bioassays. Malar. Additional information
J. 13, 412 (2014). Supplementary information The online version contains supplementary material
16. Joos, C. et al. Clinical protection from falciparum malaria correlates with available at https://doi.org/10.1038/s42003-021-02511-5.
neutrophil respiratory bursts induced by merozoites opsonized with human
serum antibodies. PLoS One 5, e9871 (2010). Correspondence and requests for materials should be addressed to S.S., B.A. or M.T.
17. Hill, D. L. et al. Efficient measurement of opsonising antibodies to
Plasmodium falciparum merozoites. PLoS One 7, e51692 (2012). Peer review information Communications Biology thanks Ana Rodrigues and the other,
18. Feng, G. et al. Mechanisms and targets of Fcgamma-receptor mediated anonymous, reviewer(s) for their contribution to the peer review of this work. Primary
immunity to malaria sporozoites. Nat. Commun. 12, 1742 (2021). Handling Editors: Jesmond Dalli and Karli Montague-Cardoso. Peer reviewer reports are
19. Adu, B. et al. Fc gamma receptor IIIB (FcgammaRIIIB) polymorphisms are available.
associated with clinical malaria in Ghanaian children. PLoS ONE. 7, e46197
(2012). Reprints and permission information is available at http://www.nature.com/reprints
20. Kana, I. H. et al. Naturally acquired antibodies target the glutamate-rich
protein on intact merozoites and predict protection against febrile malaria. J. Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in
Infect. Dis. 215, 623–630 (2017). published maps and institutional affiliations.
COMMUNICATIONS BIOLOGY |           (2021) 4:984 | https://doi.org/10.1038/s42003-021-02511-5 | www.nature.com/commsbio 9
ARTICLE COMMUNICATIONS BIOLOGY | https://doi.org/10.1038/s42003-021-02511-5
Open Access This article is licensed under a Creative Commons
Attribution 4.0 International License, which permits use, sharing,
adaptation, distribution and reproduction in any medium or format, as long as you give
appropriate credit to the original author(s) and the source, provide a link to the Creative
Commons license, and indicate if changes were made. The images or other third party
material in this article are included in the article’s Creative Commons license, unless
indicated otherwise in a credit line to the material. If material is not included in the
article’s Creative Commons license and your intended use is not permitted by statutory
regulation or exceeds the permitted use, you will need to obtain permission directly from
the copyright holder. To view a copy of this license, visit http://creativecommons.org/
licenses/by/4.0/.
© The Author(s) 2021
10 COMMUNICATIONS BIOLOGY |           (2021) 4:984 | https://doi.org/10.1038/s42003-021-02511-5 | www.nature.com/commsbio