CR1 Knops blood group alleles are not associated with severe malaria in the Gambia

The Knops blood group antigen erythrocyte polymorphisms have been associated with reduced falciparum malaria-based in vitro rosette formation (putative malaria virulence factor). Having previously identified single-nucleotide polymorphisms (SNPs) in the human complement receptor 1 (CR1/CD35) gene underlying the Knops antithetical antigens Sl1/Sl2 and McC(a)/McC(b), we have now performed genotype comparisons to test associations between these two molecular variants and severe malaria in West African children living in the Gambia. While SNPs associated with Sl:2 and McC(b+) were equally distributed among malaria-infected children with severe malaria and control children not infected with malaria parasites, high allele frequencies for Sl 2 (0.800, 1,365/1,706) and McC(b) (0.385, 658/1706) were observed. Further, when compared to the Sl 1/McC(a) allele observed in all populations, the African Sl 2/McC(b) allele appears to have evolved as a result of positive selection (modified Nei-Gojobori test Ka-Ks/s.e.=1.77, P-value <0.05). Given the role of CR1 in host defense, our findings suggest that Sl 2 and McC(b) have arisen to confer a selective advantage against infectious disease that, in view of these case-control study data, was not solely Plasmodium falciparum malaria. Factors underlying the lack of association between Sl 2 and McC(b) with severe malaria may involve variation in CR1 expression levels.     

Lymphocyte subpopulations in the neonate: high percentage of ANAE+ cells with low avidity for sheep erythrocytes

In the present study we have stained for alfa-naphthyl acetate esterase (ANAE) cord blood lymphocytes (CBL) as well as cord blood E rosetting cells. The results obtained showed that the percentage of E rosettes (E+) is lower in cord blood than in adult peripheral blood when rosetting is carried out with untreated sheep red blood cells (SRBC), while there is no significant difference if SRBC are previously treated with 2-aminoethylisothiouronium bromide (AET). ANAE-positive cells (A+) were higher in cord than in adult blood. ANAE staining of E+ cells showed that CBL include a high percentage of A+ cells with low avidity for SRBC which could represent immature lymphocytes related to the T-cell lineage.     

A complement receptor-1 polymorphism with high frequency in malaria endemic regions of Asia but not Africa

Complement receptor-1 (CR1) is a ligand for rosette formation, a phenomenon associated with cerebral malaria (CM). Binding is dependent on erythrocyte CR1 copy number. In Caucasians, low CR1 expressors have two linked mutations. We determined the Q981H and HindIII RFLP distribution in differing population groups to ascertain a possible role in adaptive evolution. We examined 194 Caucasians, 180 Choctaw Indians, 93 Chinese-Taiwanese, 304 Cambodians, 89 Papua New Guineans (PNG) and 366 Africans. PCR/RFLP used HindIII for CR1 expression and BstNI for the Q981H mutation. DNA sequencing and pyrosequencing were performed to resolve inconclusive results. Gene frequencies for the L allele were 0.15 in Africans, 0.16 in Choctaws, 0.18 in Caucasians, 0.29 in Chinese-Taiwanese, 0.47 in Cambodians and 0.58 in PNG. Allelic frequency for 981H were 0.07 in Africans, 0.15 in Caucasians, 0.18 in Choctaws, 0.29 in Chinese-Taiwanese, 0.47 in Cambodians and 0.54 in PNG. The Q981H polymorphism correlates with the HindIII RFLP in most groups except West Africans and appears to be part of a low CR1 expression haplotype. The gene frequency for the haplotype is highest in the malaria-endemic areas of Asia, suggesting that this haplotype may have evolved because it protects from rosetting and CM.     

Erythrocyte CR1 expression level does not correlate with a HindIII restriction fragment length polymorphism in Africans; implications for studies on malaria susceptibility

Complement receptor 1 (CR1) expression level on erythrocytes is genetically determined, and in Caucasian populations is linked to high (H) and low (L) expression alleles identified by a HindIII restriction fragment length polymorphism (RFLP). Erythrocyte CR1 may be an important factor in determining malaria susceptibility, as low expression of CR1 reduces the rosetting of uninfected erythrocytes with Plasmodium falciparum-infected cells, a process that contributes to malaria pathogenesis. Prior to studying CR1 expression and malaria susceptibility, we have investigated whether the quantity of erythrocyte CR1 correlates with the H and L alleles in an African population. Mean erythrocyte CR1 in 149 Malian adults was 415 molecules per cell, which is comparable to Caucasian populations; however, there was no relationship between erythrocyte CR1 level and genotype for the HindIII RFLP (mean CR1 per erythrocyte HH = 414, HL = 419 and LL = 403, P > 0.1, Student's t-test). The conclusions of a previous study of erythrocyte CR1 expression level and malaria susceptibility in West Africa that was based on HindIII RFLP genotyping may therefore need to be re-evaluated.     

An in vivo and in vitro model of Plasmodium falciparum rosetting and autoagglutination mediated by varO, a group A var gene encoding a frequent serotype

In the Saimiri sciureus monkey, erythrocytes infected with the varO antigenic variant of the Plasmodium falciparum Palo Alto 89F5 clone bind uninfected red blood cells (rosetting), form autoagglutinates, and have a high multiplication rate, three phenotypic characteristics that are associated with severe malaria in human patients. We report here that varO parasites express a var gene having the characteristics of group A var genes, and we show that the varO Duffy binding-like 1α₁ (DBL1α₁) domain is implicated in the rosetting of both S. sciureus and human erythrocytes. The soluble varO N-terminal sequence (NTS)-DBL1α₁ recombinant domain, produced in a baculovirus-insect cell system, induced high titers of antibodies that reacted with varO-infected red blood cells and disrupted varO rosettes. varO parasites were culture adapted in vitro using human erythrocytes. They formed rosettes and autoagglutinates, and they had the same surface serotype and expressed the same varO gene as the monkey-propagated parasites. To develop an in vitro model with highly homogeneous varO parasites, rosette purification was combined with positive selection by panning with a varO NTS-DBL1α₁-specific mouse monoclonal antibody. The single-variant, clonal parasites were used to analyze seroprevalence for varO at the village level in a setting where malaria is holoendemic (Dielmo, Senegal). We found 93.6% (95% confidence interval, 89.7 to 96.4%) seroprevalence for varO surface-reacting antibodies and 86.7% (95% confidence interval, 82.8 to 91.6%) seroprevalence for the recombinant NTS-DBL1α₁ domain, and virtually all permanent residents had seroconverted by the age of 5 years. These data imply that the varO model is a relevant in vivo and in vitro model for rosetting and autoagglutination that can be used for rational development of vaccine candidates and therapeutic strategies aimed at preventing malaria pathology.     

Blood group A antigen is a coreceptor in Plasmodium falciparum rosetting

The malaria parasite Plasmodium falciparum utilizes molecules present on the surface of uninfected red blood cells (RBC) for rosette formation, and a dependency on ABO antigens has been previously shown. In this study, the antirosetting effect of immune sera was related to the blood group of the infected human host. Sera from malaria-immune blood group A (or B) individuals were less prone to disrupt rosettes from clinical isolates of blood group A (or B) patients than to disrupt rosettes from isolates of blood group O patients. All fresh clinical isolates and laboratory strains exhibited distinct ABO blood group preferences, indicating that utilization of blood group antigens is a general feature of P. falciparum rosetting. Soluble A antigen strongly inhibited rosette formation when the parasite was cultivated in A RBC, while inhibition by glycosaminoglycans decreased. Furthermore, a soluble A antigen conjugate bound to the cell surface of parasitized RBC. Selective enzymatic digestion of blood group A antigen from the uninfected RBC surfaces totally abolished the preference of the parasite to form rosettes with these RBC, but rosettes could still form. Altogether, present data suggest an important role for A and B antigens as coreceptors in P. falciparum rosetting.     

Plasmodium falciparum rosetting is associated with malaria severity in Kenya.

Rosette formation in 154 fresh Plasmodium falciparum isolates from Kenyan children with mild (n = 54), moderate (n = 64), or severe (n = 36) malaria was studied to determine whether the ability to form rosettes in vitro is correlated with malaria severity. There was a wide distribution of rosette frequencies within each clinical category; however, a clear trend towards higher rosette frequency with increasing severity of disease was seen, with the median rosette frequency of the mild-malaria group (1%; range, 0 to 82%) being significantly lower than those of the moderate-malaria group (5%; range, 0 to 45%; Mann-Whitney U test, P < 0.02) and the severe-malaria group (7%; range, 0 to 97%; Mann-Whitney U test, P < 0.003). Within the severe-malaria category there was no difference in rosetting among isolates from cerebral malaria patients or those with other forms of severe malaria. We also examined the ABO blood groups of the patients from whom isolates were obtained and found that isolates from group O patients (median rosette frequency, 2%; range 0 to 45%) rosetted less well than those from group A (median, 7%; range 0 to 82%; Mann-Whitney U test, P < 0.01) or group AB (median, 11%; range 0 to 94%; Mann-Whitney U test, P < 0.03). We therefore confirm that rosetting is associated with severe malaria and provide further evidence that rosetting is influenced by ABO blood group type. Whether rosetting itself plays a direct role in the pathogenesis of severe malaria or is a marker for some other causal factor remains unknown.     

Rosetting in Plasmodium falciparum: A cytoadherence phenotype with multiple actors

The capacity of Plasmodium falciparum-infected red blood cells to bind uninfected red blood cells (“rosetting”) has been associated with high parasite density in numerous geographic areas and with severe malaria in African children. We summarize here the associations that have emerged from field studies and describe the various experimental models of rosetting that have been developed. A variety of erythrocyte receptors, several serum factors and a number of rosette-mediating PfEMP1 adhesins have been identified. Several var genes code for rosette-forming PfEMP1 adhesins in each P. falciparum genome, so that each clonal line has the capacity to generate distinct types of rosettes. To clarify their respective role in malaria pathogenesis, each of the multiple ligand/receptor interactions should be further studied for fine specificity, binding affinity and the impact of the large population polymorphism of the parasite variant repertoires should be assessed. Interestingly, some major human erythrocyte surface polymorphisms have been identified as affecting rosette formation, consistent with a role for rosetting in life-threatening falciparum malaria.     

Expression of complement receptors type 1 (CR1) and type 3 (CR3) on circulating granulocytes in experimentally provoked asthma

Neutrophils demonstrate increased complement receptor activity, measured by rosetting of C3b-coated erythrocytes, after asthma that was provoked experimentally. However, it is not clear whether the increased rosetting is due simply to increase in receptor numbers or whether other factors, such as cell adhesiveness, are involved. We have therefore enumerated granulocyte complement receptors, after asthma provoked experimentally, with monoclonal antibodies against the receptors and flow cytometry. There was a maximal 28.2 +/- 7.5% and 33.4 +/- 9.5% (mean +/- SEM; n = 15) increase in granulocyte CR1 and CR3, respectively, at 3 hours after asthma induced by antigen. There was a maximal 32.0 +/- 7.3% (mean +/- SEM; n = 7) increase in granulocyte CR1, but no change in granulocyte CR3, at 1 hour after exercise-induced asthma. No significant changes in granulocyte CR1 or CR3 were observed up to 6 hours after methacholine challenge, or after exercise in subjects who did not develop exercise-induced asthma. There was a maximal 33 +/- 9% (mean +/- SEM; n = 8) increase in granulocyte CR1 at 30 minutes, but no increase in granulocyte CR3, after histamine challenge of subjects with asthma. Incubation of whole blood with histamine in vitro did not lead to any enhancement in expression of granulocyte CR1. This suggests that antigen- and exercise-induced release of histamine may augment granulocyte CR1 expression through an indirect mechanism. These data indicate that there is increase in the numerical expression of CR1 on granulocytes, after asthma provoked experimentally, which is accompanied by increases in granulocyte CR3 after bronchoprovocation with antigen, but not histamine or exercise.     

Hyperexpression of ICAM-1 and CD36 in placentas infected with Plasmodium falciparum: a possible role of these molecules in sequestration of infected red blood cells in placentas

AIMS: During pregnancy, Plasmodium falciparum malaria is frequent and associated with maternofetal complications. This could be the consequence of sequestration by several adhesion molecules of parasite-infected red blood cells in syncytiotrophoblast. To investigate the expression of ICAM-1 and CD36, two of the adhesion molecules for Plasmodium falciparum, an immunohistochemical study was carried out in malaria-infected placentas. METHODS AND RESULTS: Thirty-five infected and 35 noninfected samples were chosen randomly. According to the histological classification of Bulmer, the infected placentas were separated in three groups: active, active chronic and past-chronic infection. CD36 was localized in the cytoplasm of stromal cells of terminal villi of infected or noninfected placentas, but not in syncytiotrophoblast. ICAM-1 was detected in the cytoplasm of stromal and endothelial villous cells in both infected and noninfected placentas and in syncytiotrophoblast of eight infected placentas showing more frequently active than active chronic or past-chronic infection (P < 0.001). The percentage of cells immunostained for CD36 or ICAM-1 was evaluated in the terminal villi. The proportion of villous cells, with ICAM-1 and CD36 immunostaining, was significantly higher in infected vs. noninfected placentas (P < 0.0001) and CD36 was detected more in acute inflammatory vs. past-chronic inflammatory placentas (P < 0.05). CONCLUSIONS: The higher expression of ICAM-1 in infected placentas and its localization in syncytiotrophoblast particularly during acute infection, suggest ICAM-1 can act directly in the sequestration of parasite-infected red blood cells (IRBCs). On the other hand, the expression of CD36 is influenced by the presence of IRBCs without being directly implicated in sequestration of IRBCs. The hyperexpression of these two molecules could explain the high frequency of malaria during pregnancy.     

Fresh isolates from children with severe Plasmodium falciparum malaria bind to multiple receptors

The sequestration of Plasmodium falciparum-infected erythrocytes (pRBC) away from the peripheral circulation is a property of all field isolates. Here we have examined the pRBC of 111 fresh clinical isolates from children with malaria for a number of adhesive features in order to study their possible coexpression and association with severity of disease. A large number of adhesion assays were performed studying rosetting, giant rosetting, and binding to CD36, intercellular adhesion molecule 1, platelet endothelial cell adhesion molecule 1, thrombospondin, heparin, blood group A, and immunoglobulins. Suspension assays were performed at the actual parasitemia of the isolate, while all the static adhesion assays were carried out at an equal adjusted parasitemia. The ability to bind to multiple receptors, as well as the ability to form rosettes and giant rosettes, was found to be more frequent among isolates from children with severe versus mild malaria (P = 0.0015). Rosettes and giant rosettes were more frequent for children with severe malaria, and the cell aggregates were larger and tighter, than for those with mild disease (P = 0.0023). Binding of immunoglobulins (97% of isolates) and of heparin (81% of isolates) to infected erythrocytes was common, and binding to heparin and blood group A was associated with severity of disease (P = 0.011 and P = 0.031, respectively). These results support the idea that isolates that bind to multiple receptors are involved in the causation of severe malaria and that several receptor-ligand interactions work synergistically in bringing about severe disease.     

Rosetting in Plasmodium falciparum: a cytoadherence phenotype with multiple actors.

The capacity of Plasmodium falciparum-infected red blood cells to bind uninfected red blood cells ("rosetting") has been associated with high parasite density in numerous geographic areas and with severe malaria in African children. We summarize here the associations that have emerged from field studies and describe the various experimental models of rosetting that have been developed. A variety of erythrocyte receptors, several serum factors and a number of rosette-mediating PfEMP1 adhesins have been identified. Several var genes code for rosette-forming PfEMP1 adhesins in each P. falciparum genome, so that each clonal line has the capacity to generate distinct types of rosettes. To clarify their respective role in malaria pathogenesis, each of the multiple ligand/receptor interactions should be further studied for fine specificity, binding affinity and the impact of the large population polymorphism of the parasite variant repertoires should be assessed. Interestingly, some major human erythrocyte surface polymorphisms have been identified as affecting rosette formation, consistent with a role for rosetting in life-threatening falciparum malaria.     

Another case of a lymphocytotoxic antibody with blood group A1, Leb and A Led associated specificity

A lymphocytotoxic antibody with blood group A1 Leb, and A (A1 + A2) Led, associated specificity was found together with an anti-HLA-DR2 in the serum of a multiparous woman. The A1 Leb and A Led-antibodies could be absorbed with erythrocytes from persons with blood group A1 or A2, irrespective of their Lewis antigens, even if their lymphocytes reacted negatively in this antibody, leaving anti-HLA-DR2 in the serum. Lymphocytes of blood group A1 were able to absorb the anti-A1 Leb and -A Led, whereas those of A2 could only absorb the anti-A Led. Saliva from persons with blood groups A1 Le (a - b +), A2 Le (a- b +), A1 Le (a - b -), secretor, and A2 Le (a - b -), secretor, inhibited the anti-A1 Leb and -A Led when tested in the serum/saliva ratio 50:1, which was not the case with other ABO-Lewis combinations. The woman who produced this antibody has blood group O Le (a - b +) and secretes H, Lea and Leb substances in about the same amount as do other individuals with blood group O Le (a - b +) used as controls. The anti-A in her serum has equal titers against A1 Le (a + b -) and A1 Le (a - b +) red cells.     

Frequency,risk factors and prognostic value of anemia associated with HIV/AIDS in the adult in Mali

Anaemia is a common complication of the HIV infection. To understand the mechanism of HIV associated anaemia and to suggest a consequent therapeutic approach in adults in Mali, we undertook a prospective case/control study in two services of reference with essentially adults recruitment in Bamako. We studied the frequency, the risk factors and the prognosis value of this complication in 133 patients with HIV infection matched to 133 others non HIV infected. The average age of our patients was 36.08 +/- 8.80 years (age range: 19 to 66 years). The frequency of anaemia was significantly higher in patients with HIV infection compared to the controls (78.9% vs. 51.9%; OR = 2.46; 95% CI [1.56-3.92]). Anaemia was more frequent in women than in men (p = 0.00003). A significant association between anaemia and thrombopenia or lymphopenia was observed only in patients with HIV infection. The severity of anemia was positively associated with the HIV2 infection and the progression of the HIV disease. Mortality was more frequently associated to the anaemia (p < 10(-5)) in patients infected by HIV. These findings suggest that bone marrow depression leading to a decreased red blood cells production is the main mechanism of HIV associated anaemia in adult in Mali. Therefore, without evidence of a best cost-effectiveness ratio of a human recombinant erythropoietin treatment in the context of countries with a low income, the therapy of this haematological complication must be an emergency focusing on red blood cells transfusions.     

Immune mechanisms against canine distemper. I. Identification of K cell against canine distemper virus infected target cells in vitro.

Canine peripheral blood lymphocytes, polymorphonuclear leucocytes (PMN) and monocytes (macrophages) were obtained by various cell separation techniques and were tested for their cytotoxic capacity against antibody-sensitized canine distemper virus (CDV) infected Vero cells by an in vitro chromium release assay. Canine lymphocytes were found to destroy CDV infected target cells effectively, while neither PMN nor monocytes (macrophages) could do so. The active lymphocyte was characterized by various rosetting techniques to be a non-T and a non-B lymphocyte. These cells bear no surface immunoglobulin (SIg-) but possessed both Fc receptors (Fc+) and complement receptors (EAC+) suggesting that these cells are neither classical T nor B cells. The possible roles of this K cell in the resistance against canine distemper are discussed.     

Inhibition by monoclonal anticomplement receptor type 1 on interactions between senescent human red blood cells and monocytic-macrophagic cells

The effect of different murine monoclonal antibodies (Mab) specific for the glycoprotein complement receptor type 1 (CR1), type 2 (CR2), and type 3 (CR3) on the adhesion to and on the phagocytosis of human senescent red blood cells (S-RBC) by monocytes or by monocyte-derived macrophages (M phi) was investigated. Murine Mab anti-CR3 (anti-Leu 15 and OKM1) were found to inhibit, in the same order of magnitude, on one hand, the Fc receptors (FcR)-dependent rosetting and phagocytosis, and, on the other hand, the S-RBC rosetting and phagocytosis by adherent monocytes. Thus, the specific involvement of the CR3 epitopes recognized by Mab anti-Leu 15 or by OKM1 in the interactions between S-RBC and monocyte/macrophage could not be demonstrated. Murine Mab anti-CR1 was found to be a significant inhibitor of binding to and of phagocytosis of S-RBC (but not of young [Y] RBC) by monocytes or M phi, whereas Mab OKM5 carrying the same isotype as Mab anti-CR1, but a different specificity, was devoid of any significant inhibitory effect. Furthermore, Y-RBC or S-RBC opsonized with Mab anti-CR1 did not form FcR-dependent rosettes and were not internalized by monocytes; in addition, preincubation of phagocytes with Mab anti-CR1 did not inhibit FcR-dependent rosetting and phagocytosis. These results suggest that the effect of anti-CR1 is mediated through a specific binding to CR1 and not through an FcR blockade. As the role of specifically bound IgG on phagocytosis of human S-RBC by macrophages has previously been demonstrated by several authors, the present study suggests that monocyte-macrophage complement receptor type 1 may act in synergy with Fc receptors in the recognition of S-RBC by macrophages. It is shown in addition that the tripeptide Arg-Gly-Asp, identical to the region of iC3b recognized by CR3 and by several adhesion-promoting receptors that are structurally similar to CR3, such as fibronectin or vitronectin, is a significant inhibitor of the binding to and the phagocytosis of S-RBC by monocytic-macrophagic cells.     

Variant antigens of Plasmodium falciparum encoded by the var multigenic family are multifunctional macromolecules

Cytoadhesion of parasitized red blood cells (PRBCs) to vascular endothelial cells (sequestration) and binding of unparasitized RBCs to PRBCs (rosetting) are virulence factors of Plasmodium falciparum, the species responsible for lethal human malaria. Variant antigens involved in both phenomena have been identified as products of the multicopy var gene family. In this review, progress in the understanding of molecular mechanisms of sequestration is summarized, in particular, concerning the structure of var gene products related to specificity of binding to endothelial receptors, and the origin of var gene diversity.     

Analysis with monoclonal antibodies of human lymphoid cells forming rosettes with rabbit red blood cells.

Rabbit red blood cells have previously been shown to rosette with a subpopulation of thymocytes and with mitogen activated peripheral lymphocytes but not with unstimulated lymphocytes. Using monoclonal antibodies and double marker assays we studied the phenotype of these cells. In thymus, over 90% of rosetting cells express antigens of immature thymocytes (HTA1, OKT6). A proportion of the rosetting cells shows in addition antigens of mature thymocytes (OKT3, UCHT1). These cells probably correspond to a stage of intrathymic maturation between common and mature thymocytes. Virtually all rosetting cells are T cells and express an antigen related to T cell activation (TAC) when lymphocytes are activated by mitogens like PHA or Con A. Few rosetting cells are Ia positive. Two other antigens (OKT9, OKT10) known to be associated with proliferating and immature cells, are found in variable proportions on rosetting cells. After stimulation with allogeneic lymphocytes, fewer rosettes are detected than after stimulation by mitogens. Cells activated by a soluble antigen (PPD) and forming rosettes with rabbit red blood cells have a helper phenotype (Leu3a positive). Screening of leukaemia cell samples revealed that only cells from patients with T-ALL form rosettes with rabbit red blood cells. Rosette formation is almost totally inhibited by a polyclonal anti-thymocyte serum and two monoclonal antibodies (OKT11A,Lyt3) which have been shown to block rosettes with sheep erythrocytes.     

<Emphasis Type="Italic">Plasmodium falciparum</Emphasis> malaria and ABO blood group: is there any relationship?

There is increasing evidence that Plasmodium falciparum malaria is influenced by ABO blood group but the extent of association between both is yet to be well defined. Studies that investigated association between P. falciparum malaria and ABO blood group were identified using MEDLINE search and were systematically reviewed. There were apparent discrepancies and contradictions in the studies as some reported significant association between both while others observed no significant association. This outcome may reflect the complex interaction between P. falciparum malaria and the host immune responses. However, findings from all studies reviewed suggested that individuals of blood group O are relatively resistant to severe disease caused by P. falciparum infection. It was established that parasitized erythrocytes form rosettes more readily with red blood cells (RBCs) of A, B, or AB groups than with blood group O and this parasite-triggered RBC rosette formation is associated with the severity of clinical disease and with the development of cerebral malaria. Differences in rosetting ability were based on the P. falciparum strain-specific preference of rosetting with non-O blood groups and not only a phenomenon of laboratory-propagated strains, but also exist in wild clinical isolates from all major malarious areas of the world.     

High titres of anti-T antibodies and other haemagglutinins in human malaria.

The prevalence of antibodies against (i) human red blood cells (RBC) of A and B groups, (ii) trypsinized O Rh+ RBC and (iii) neuraminidase treated O Rh+ RBC were investigated both in sera of Africans from a malaria endemic area of Upper Volta and in sera of Europeans with acute malaria from a Paris hospital. An increased frequency of high titres of haemagglutinins was observed against A and B as well as O Rh+ trypsinized human RBC, thus confirming previously published results. In addition, agglutination of neuraminidase treated RBC showed that the titres were increased in about 40% of Africans studied and in about 80% of patients with acute malaria. Using agglutination with a specific anti-T lectin and inhibition with two ligands, it was found that sera of malarious patients contain high titres of antibodies directed against the T antigen of neuraminidase treated RBC. The mechanisms of appearance of high titres of autohaemagglutinins in malaria and their possible interference in the anaemia associated with this disease are discussed.