Sequence variation of PfEMP1-DBLα in association with rosette formation in Plasmodium falciparum isolates causing severe and uncomplicated malaria

Background Rosetting and cytoadherence of Plasmodium falciparum- infected red blood cells have been associated with severity of malaria. ICAM-1 and CD36 are the main host cell receptors, while PfEMP1-DBLα is a major parasite ligand, which can contribute to rosette formation. This study is aimed at demonstrating whether the highly polymorphic PfEMP1-DBLα sequences occurring among Thai isolates causing severe and uncomplicated malaria are associated with their ability to form rosettes and reflected the clinical outcome of the patients. Methods Two hundred and ninety five PfEMP1-DBLα sequences from Thai clinical isolates causing severe and uncomplicated malaria were evaluated by sequencing and direct comparison using the specific text string analysis functions in Microsoft Excel and Perl. The relationships between the PfEMP1-DBLα sequences were also analysed by network analysis. The binding abilities of parasitized red blood cells (PRBCs) to CD36, wild type ICAM-1, ICAM-1 Kilifi and ICAM-1 S22/A under static condition were included. Results Two hundred and eighty one non-identical amino acid sequences were identified (< 95% sequence identity). When the distributions of semi-conserved features (PoLV1–4 and sequence group) within the rosetting domain PfEMP1-DBLα were observed, close similarity was found between isolates from the two disease groups. The sequence group 1 representing uncomplicated malaria was significantly different from the sequence group 3 representing the majority of severe malaria ( p = 0.027). By using a simple non-phylogenetic approach to visualize the sharing of polymorphic blocks (position specific polymorphic block, PSPB) and cys/PoLV among DBLα sequences, the sequence group 1 was split from the other five sequence groups. The isolates belonging to sequence group 5 gave the highest mean rosetting rate (21.31%). However, within sequence group 2 and group 6, the isolates causing severe malaria had significantly higher rosetting rate than those causing uncomplicated malaria ( p = 0.014, p = 0.007, respectively). Conclusion This is the first report of PfEMP1-DBLα analysis in clinical Thai isolates using semi-conserved features (cys/PoLV and PSPBs). The cys/PoLV group 5 gave the highest rosetting rate. PfEMP1-DBLα domains in Thai isolates are highly diverse, however, clinical isolates from severe and uncomplicated malaria shared common sequences.