Investigation of host red blood cell receptors essential for rosetting in severe Plasmodium falciparum malaria

Plasmodium falciparum malaria kills almost half a million people every year, many of whom are children living in Africa. Rosetting is a pathological phenomenon which is associated with all types of severe malaria and occurs when two or more uninfected erythrocytes adhere to an erythrocyte infected with the mature form of the P. falciparum parasite. It is thought that these rosettes may cause obstruction of the microvasculature leading to the serious complications seen in severe malaria. Understanding the molecular mechanisms of rosetting could therefore lead to the development of new adjuvant therapies for severe disease. The overall aims of this work were to reassess the evidence for previously described host erythrocyte rosetting receptors, explore new methods of investigating rosetting mechanisms, including generating knockdown/out erythrocytes from CD34+ haematopoietic stem cells and immortalised erythroid precursors, and to identify novel rosetting receptors. This thesis begins by reassessing the evidence that the glycosaminoglycans, heparan sulfate (HS) and chondroitin sulfate (CS) are involved in rosetting. Contrary to previously published work, results from experiments using carefully validated enzymes to cleave HS or CS did not support the hypothesis that HS or CS are important host receptors for rosetting across the six P. falciparum strains tested. In addition, I found no evidence to suggest that HS or CS are actually present on mature erythrocytes, though HS was detected on early bone marrow-derived erythrocyte precursors. Secondly, I investigated the use of induced pluripotent stem cells (iPSC) and cells cultured from adult bone marrow CD34+ stem cells (cRBC) as a tool to produce knockdown erythrocytes using RNA inference techniques. The cRBC appeared, both morphologically and by receptor profiling with flow cytometry, to be a good approximation of reticulocytes. However, unexpectedly, the cRBC derived erythrocytes were only able to form rosettes with two of the four parasite lines tested. Further study into the subtle differences in receptor expression levels between cRBC and peripheral erythrocytes suggested that Band 3 could be a potential novel rosetting receptor. This hypothesis was supported by the results of rosette disruption experiments which showed that antibodies to the Wrightb antigen, carried on Band 3, were capable of significantly disrupting rosettes of mature erythrocytes across all parasites strains tested. Finally, I used a new, immortalised erythroid precursor line, the “EJ” cells, developed by the Duraisingh laboratory at Harvard university, to further investigate the role of Band 3 and the Wrightb antigen in rosetting. Band 3, glycophorin A (GYPA) and CR1 knockout EJ cells, created using CRISPR/Cas9 technology, were tested for rosetting ability with six parasite lines. As the Wrightb antigen requires both Band 3 and GYPA to properly form, both these knockout EJ cells also lacked Wrightb. Compared to wildtype EJ cells, GYPA and Band 3 knockout EJ cells had reduced rosetting. However, the rosetting rates were similar between the two knockout lines, suggesting that neither the presence of GYPA or Band 3 alone can rescue the poor rosetting phenotype in the absence of Wrightb. In summary, I have found that while there is little evidence to support the involvement of HS or CS in rosetting, the Wrightb antigen carried on Band 3 may be an important, strain-transcending rosetting receptor and could represent a useful therapeutic target to reduce rosetting. In addition, I have developed new techniques for investigating rosetting receptors using a novel erythroid precursor line. The EJ cells also have great potential for the development of a rosetting screen to identify other candidates to help reduce the mortality and morbidity of severe malaria in the future