An autoimmune attack on uninfected red blood cells likely contributes to anaemia – a shortage of red blood cells – in people with malaria, according to a new study published in eLife.
Anaemia is a common and sometimes deadly complication of malaria infections.
While the immune system must destroy red blood cells infected with the malaria parasite to clear the infection, the study suggests the infection can trigger an ongoing autoimmune attack on uninfected red blood cells, ultimately causing anaemia.
Previous studies in mice with malaria showed that autoimmune antibodies attach to a molecule called phosphatidylserine (PS) on uninfected cells, marking them for destruction by the immune system.
These autoimmune antibodies were also found in patients with malaria-induced anaemia, suggesting they may be the cause of red blood cell loss.
However, it has been difficult to study these immune processes in patients living in areas where malaria is endemic, who may experience repeated infections.
To address this, lead author Dr. Juan Rivera-Correa, a former postdoctoral fellow in the Department of Microbiology at NYU Grossman School of Medicine in New York City, US, and his colleagues decided to study the immune response in people from Germany exposed to malaria for the first time while travelling in Africa.
“Studying the immune cells that produce these autoantibodies during new malaria infections in European travelers represented a unique opportunity,” Dr. Rivera-Correa explains.
In their experiments, the team identified the production of an unusual type of immune B-cell—FcRL5+T-bet+ B-cells—that increases anti-PS antibody production associated with the development of anaemia in these patients.
These immune cells also developed and produced anti-PS antibodies in blood drawn from uninfected people that was then exposed to broken remnants of malaria-infected red blood cells in the laboratory.
“Our results provide the first mechanistic evidence of autoimmune-mediated anaemia in malaria patients and highlight these atypical immune cells as major promoters of this complication,” Rivera-Correa says.
Ana Rodriguez, Professor of Microbiology at NYU Grossman School of Medicine, adds: “There is a great need for novel targeted treatments for malaria-induced anaemia, which is common and can be fatal for many malaria patients.
The unique phenotype and specificity of these immune B-cells could allow them to be used as a biomarker for anaemia or as a target for new therapies.”
Malaria is an intravascular parasitic disease caused by infection with Plasmodium. Plasmodium falciparum is the most prevalent malaria species in Africa. In 2017 there were 219 million cases and 435,000 global malaria deaths with 61% of deaths occurring in children less than five years of age1.
Two common complications of severe malaria (SM) are severe malarial anemia (SMA) and cerebral malaria (CM).
SMA is life-threatening and accounts for over half of childhood malaria deaths in Africa, and occurs through multiple mechanisms, including: direct destruction of infected RBCs, clearance of infected and uninfected RBCs, and insufficient bone marrow production (reviewed in2). Death is preventable in SMA if children receive an appropriate transfusion3,4.
CM is characterized by an unrousable coma with no other identifiable cause, and typically has a mortality rate of 18–21%5. Reported mortality in African children with SM was 10% in the AQUAMAT study and ranged from 4–15% across 11 sites in 9 different African countries4.
Elevations in autoantibodies targeting self antigens, such as DNA, are well recognized in autoimmune disorders like systemic lupus erythematosus (SLE) and are used for diagnosis, prognosis, and understanding disease pathogenesis6,7.
Autoantibodies have been associated with complications in a number of infections8,9, and autoimmunity during and after infection is frequently reported in diseases associated with systemic inflammatory responses— including malaria10,11.
While autoantibodies in malaria were initially attributed to non-specific polyclonal immune activation12, more recent studies demonstrate direct parasite induced secretion of autoantibodies in the host13.
Malaria shares clinical features with autoimmune disorders, including autoimmune anemia. Using a mouse model, it was observed that anti-phopsphatidylserine (PS) antibodies promote anemia during malaria through the binding to PS exposed in uninfected erythrocytes which facilitates their clearance14.
It was also observed that levels of anti-PS antibodies are correlated with malarial anemia in patients14,15.
Although autoantibodies have been reported in experimental models of malaria and patient populations16–18, few studies have investigated the relationship between autoantibodies and SM complications. Acute kidney injury (AKI) is a common complication in pediatric SM associated with significant morbidity and mortality4,19,20, but the pathogenesis is not well understood.
Given reports of elevated autoantibodies in SM and clinical similarities between malaria and autoimmune disorders, we sought to evaluate the relationship between anti-PS and anti-DNA antibodies in children with SM.
We hypothesized children with SM would have elevated autoantibodies associated with complications on admission including anemia and AKI. In this prospective cohort study, we measured admission anti-PS and anti-DNA autoantibodies in Ugandan children with SM and evaluated whether autoantibodies were associated with disease severity and outcomes over two years follow-up.
Discussion
In the present study we evaluate the relationship between autoimmune antibodies, SM complications, and in-hospital and post-discharge mortality in a cohort of Ugandan children with SM and CC. Since autoimmune anti-PS antibodies promote malaria-induced anemia in a mouse model and are correlated with malarial anemia in adults14,15, we hypothesized that these antibodies may also play a role in children with malarial anemia and that other autoimmune antibodies, such as anti-DNA, may contribute to malaria pathogenesis. In this work, we demonstrate that in children with SM, anti-PS antibodies are strongly associated with anemia on admission, post-discharge mortality, and hospital readmissions, and that anti-DNA antibodies are associated with AKI.
Together, the study findings suggest that in children with SM, anti-PS antibodies may play a role in the development of severe anemia and post-discharge morbidity and mortality, and that immune mediated pathways may contribute to malaria-related AKI.
Previous studies have shown that anti-PS autoantibodies contribute to anemia in experimental murine models of malaria by binding to uninfected RBC and promoting their premature clearance from the circulation14.
Analysis of plasma samples from adults with P. falciparum infections have documented an association between anti-PS antibodies andmalarial anemia14,15. This study extends these findings by showing anti-PS and anti-DNA autoantibodies are inversely associated with hemoglobin in children with SM during the acute phase of illness.
Further, we show that children positive for anti-PS antibodies on admission are at risk for multiple hospital readmissions during follow-up. These data are consistent with previous reports describing an association of anti-RBC autoantibodies and anemia in P. vivax25,26, although in these studies only antibodies recognizing proteins, not lipids or nucleic acids (such as PS and DNA), were analyzed.
In mouse models of malaria, infected, but also uninfected RBC, expose PS in their surface14,27, which is recognized by anti-PS antibodies, contributing to anemia14. Our knowledge of this mechanism in human patients is more limited, but an inverse correlation of anti-PS antibodies and hemoglobin levels was described in malarial anemia14,15.
Since in vitro treatment of RBC with artesunate increases PS translocation to the RBC surface28, it is possible anti-PS antibodies binding to PS on the surface of RBC may be an additional mechanism contributing to anemia following artesunate treatment29.
Additional studies are needed to determine: (i) how levels of anti-PS autoantibodies change over infection; (ii) whether anti-PS antibodies may contribute to recurrent anemia in children with repeated malaria infections; and (iii) whether anti-PS autoantibodies accelerate the removal of RBC following parenteral treatment with artesunate.
We observed a correlation between lactate dehydrogenase (LDH) and autoantibody levels, suggesting RBC lysis may be mediated by autoantibodies. Alternatively, it is possible hemolysis enhances autoantibody production through immune activation and exposure to self-antigens. Additional studies are needed to delineate whether RBC hemolysis is mediated by complement fixation on RBC-bound anti-PS antibodies, or whether hemolysis triggers the generation of autoantibodies.
Anti-DNA antibodies also correlated with anemia in children with SM. This association may be explained by the observations that DNA is abundant in the circulation of malaria patients30 and that cell-free DNA binds to uninfected red blood cells31.
RBC-bound DNA could be targeted by infection-induced anti-DNA antibodies promoting their clearance and enhancing anemia. Cell-free DNA in malaria patients declines upon resolution32, which could explain the lack of correlation of anti-DNA antibodies with hospital readmission.
In this study anti-DNA antibodies were strongly associated with the presence and severity of AKI. Previous studies in children with SM suggest the etiology of AKI at presentation is largely related to reduced renal perfusion19 as a result of hypovolemia and impaired microvascular perfusion from parasitized RBC cytoadhesion to the endothelium.
We observed an inverse relationship between eGFR and anti-DNA and anti-PS antibodies during SM, raising questions on whether reduced autoantibody clearance in AKI leads to increases in circulating autoantibodies.
The MHC class 1-like Fc receptor (FcRn) is one receptor important in maintaining serum levels of albumin and IgG, contributing to the long half-life of IgG. However, podocytes and proximal tubule epithelium in the kidney express FcRn that have been shown to promote IgG clearance from the glomerular basement membrane33.
Additionally, other receptors involved in antibody removal such as complement receptor 1 (CR1/CD35) are expressed on podocytes34 and down regulated in other cell types during malaria35.
While it is possible that reduced GFR may lead to altered renal handling of IgG and affect levels of circulating autoantibodies, we did not observe any relationship with total IgG against malaria antigens AMA-1, MSP-1, CSP, or GLURP and any measures of renal function (AKI, creatinine, BUN, or eGFR). These data suggest changes in IgG metabolism are not responsible for the increase in autoantibodies.
Anti-DNA antibodies are associated with glomerulonephritis and AKI in SLE through a variety of mechanisms including: (a) DNA immune complex (IC) entrapment in glomerular basement membranes, (b) non-DNA antigen crossreactivity, (c) binding to apoptoic cellular debris in the glomerulus and acting as “a planted antigen” and/or (d) anti-DNA binding to kidney cells including podocytes, endothelial cells and/or proximal tubule cells leading to renal injury36. P. malariae can lead to a nephrotic syndrome associated with a membrano-proliferative glomerulonephritis with deposits of IgM, IgG and C3 in mesangial and subendothelial tissue37.
However, the pathogenesis of P. falciparum appears to be different and the role of auto-antibodies and IC in kidney pathogenesis during P. falciparum remains unclear. Even if it is well-documented that there are increases in circulating DNA and anti-DNA that form IC in the circulation32, evidence of IC deposition in kidneys of patients with P. falciparum is inconsistent38–40.
The relationship between anti-DNA autoantibodies and glomerulonephritis has not been systematically investigated using immunofluorescent labeling of immunoglobulins and complement to detect and characterize glomerulonephritis in renal biopsies of surviving patients.
Together, these results indicate that an immune mediated pathway, possibly an immune-mediated glomerulonephritis, similar to what is observed with SLE nephritis, could contribute to P. falciparum malaria-associated AKI41. We further speculate that anti-DNA autoantibodies may lead to proximal tubular cell injury and acute tubular necrosis in the absence of glomerulonephritis.
This study has several strengths. We were able to measure autoantibodies in a well characterized population of Ugandan children with SM and compare autoantibodies with malaria-specific antibodies.
The finding of a strong association between anti-DNA autoantibodies and AKI (defined using established KDIGO criteria and using CC to define baseline creatinine), that was not present for anti-PS antibodies or malaria-specific antibodies, suggests a mechanistic link between anti-DNA antibodies and AKI. Further, the prospective design enabled us to evaluate the relationship between autoantibodies, post-discharge mortality, and risk of subsequent hospitalization.
This study has limitations. We included children with two manifestations of severe malaria and therefore cannot comment on the generalizability of these findings to other manifestations of severe or uncomplicated malaria.
As we did not collect longitudinal samples, we cannot draw conclusions on how autoantibody levels change following treatment and whether there are clinical consequences associated with post-treatment autoantibody levels. Apart from the assessment of hemoglobin in the urine, no additional urinalysis was conducted and urine was not stored. Quantification of blood and protein in the urine and renal biopsy findings will be needed to better characterize AKI in SM.
In summary, we report elevations in anti-DNA and anti-PS autoantibodies in SM. Consistent with another study in malarial anemia14,15, we show a relationship between autoantibodies and anemia, this time in children with acute severe malaria in an endemic area. We also show that children testing positive for anti-PS antibodies were more likely to be readmitted to the hospital and readmitted multiple times compared to children negative for anti-PS antibodies. We also report a relationship between anti-DNA antibodies and the presence and severity of AKI. Additional studies are needed to evaluate the impact of malaria treatment on autoantibody levels, and to assess the relationship between post-treatment autoantibodies and long-term complications of severe malaria including recurrent hospitalizations for severe anemia, death, and chronic kidney disease.
More information: Juan Rivera-Correa et al, Atypical memory B-cells are associated with Plasmodium falciparum anemia through anti-phosphatidylserine antibodies, eLife (2019). DOI: 10.7554/eLife.48309
Journal information: eLife
Provided by eLife
