Disruption of Plasmodium berghei merozoite surface protein 7 gene modulates parasite growth in vivo

The merozoite form of the malaria parasite attaches to and invades red blood cells (RBCs). Merozoite surface proteins (MSPs) mediate the first interaction between parasite and erythrocyte and are exposed to humoral antibody; therefore, they are vaccine candidates. Ten MSPs have been identified^1,2 ; of these, MSP1 is the most extensively studied.^3,4 Plasmodium falciparum MSP1 is synthesized as an approximately 200-kDa precursor processed into 4 fragments that remain as a complex on the surface of the merozoite.^5,6  At the time of erythrocyte invasion, the complex is shed as the result of a cleavage within the C-terminal 42-kDa fragment.^7,8  The shed complex contains other polypeptides that are fragments of MSP6 and MSP7.^9-12  The role of these associated proteins is unclear, but it has been suggested that they are involved in the structure, processing, and function of the MSP1 complex.

P falciparum MSP7 is the 351-amino acid precursor to a 22-kDa polypeptide associated with the MSP1 complex.¹²  The msp7 gene is expressed in mature schizonts at the same time as msp1. Additionally, msp7 is part of a gene family containing 3 and 6 adjacent members in Plasmodium yoelii and P falciparum, respectively.^12,13  Little is known about the role or function of the MSP7 family in merozoite invasion of RBCs.

We wanted to study the role of MSP7 in merozoite invasion and MSP1-complex biosynthesis and structure. Here we have identified and disrupted msp7 in Plasmodium berghei by double homologous recombination. This msp7 knockout line (MSP7ko) was used to examine the role of MSP7 in MSP1 synthesis and processing and in parasite biology. Deletion of msp7 resulted in a growth phenotype that reflected an effect on parasite invasion of erythrocytes.

Plasmid pDHΔMSP7 containing a DHFR cassette flanked by 420-bp and 360-bp sequences from the 5′ and 3′ ends of the msp7 coding sequence was used to transfect P berghei ANKA schizonts. The resultant knockout parasites (MSP7ko) were cloned and analyzed by Southern blotting.^14,15 

Total RNA was isolated from purified schizonts, blotted and hybridized with a 1-kb DNA fragment derived from msp7.

Schizonts from wild-type (WT) and MSP7ko lines were biosynthetically labeled by incubation for 2 hours at 37°C in [³⁵S]-methionine- and cysteine-containing medium. Labeled parasites were lysed, and specific proteins were immunoprecipitated.¹⁶ 

For immunofluorescence assay (IFA) analysis, thin smears of schizonts were incubated with rabbit antiserum raised against recombinant P yoelii MSP7 (diluted 1:300) and mAb 25.1 (diluted 1:500) against MSP1. Images were captured using a Leica TCS SP2 microscope and software with a 63×/1.32 oil immersion PH3CS objective (Leica Microsystems, Bucks Milton Keynes, United Kingdom), then processed for presentation with Adobe Photoshop software (Adobe Systems, San Jose, CA).

To ascertain whether the disruption of msp7 had an effect on parasite growth in vivo, challenge inocula (100 and 1000 blood-stage parasites) from WT and MSP7ko P berghei were injected intravenously into groups of Balb/c mice (6-8 weeks of age). Forty-eight hours after injection and daily for another 10 days, a thin smear of blood was prepared and stained with Giemsa reagent. For each sample, at least 2000 RBCs were examined by microscopy, and the percentage of parasitemia was calculated. The ratio of mature RBCs and reticulocytes infected by WT and MSP7ko parasites was also determined during the 6- to 10-day period after infection. Images were captured using a Leica DMR microscope equipped with a 63 × oil immersion objective and Leica IM 1000 software.

The P berghei msp7 gene was identified (72% identity/83% similarity with P yoelii MSP7 at the protein level) and targeted for disruption to examine its function. In the MSP7ko parasite line obtained, the DHFR gene was correctly integrated into and disrupted the msp7 locus (Figure 1A). Northern blot analysis failed to detect any expression of msp7 in the MSP7ko parasite, though the gene was highly expressed in the WT parasite (Figure 1A).

Antibodies to MSP7 precipitated a 42-kDa band from an extract of WT parasites that was absent in the extract of MSP7ko parasites (Figure 1B, lanes 1 and 2). Despite the absence of MSP7, MSP1 was produced in similar amounts in WT and MSP7ko parasites (Figure 1B, lanes 3 and 4). The spectrum of polypeptides precipitated by MSP1-specific polyclonal serum, but not normal serum, which represents processing fragments of MSP1, was indistinguishable for the 2 parasite lines (Figure 1B, lanes 5 and 6).

According to IFA analysis, antibodies raised against the C-terminus of MSP7 colocalized with antibodies to MSP1, suggesting coexpression during schizogony in WT parasites. No reactivity with MSP7ko parasites was observed (Figure 1C), though MSP1 was still located at the schizont periphery.

Northern blotting, immunoprecipitation, and IFA experiments clearly showed that MSP7 was not expressed in the MSP7ko parasite. However, the synthesis and surface expression of MSP1 was unchanged. This suggests that either MSP7 is not essential for the synthesis and processing of MSP1 or that other members of the family may substitute its function. In P falciparum the complex between MSP1 and MSP7 is stable in the presence of the detergent NP-40.¹⁰  Either this is not the case in P berghei or 1 of the 2 MSP7-related proteins (MSRP1 or MSRP2) might form a more stable interaction with MSP1 than MSP7 does. In the yeast 2-hybrid system, the minimal region of P yoelii MSP7 interacting with MSP1 was mapped to residues 113-324,¹³  and both MSRP1 and MSRP2 could interact with MSP1. Although this region of MSP7 is not present in the knockout parasite, the processing of MSP1 was unaffected. The expression of MSRP1 and MSRP2 was largely unaffected in the MSP7ko parasite when compared with the WT parasite (data not shown). This clearly suggests that the 3 genes are independently regulated.

Transgenic parasites were viable, and they invaded and multiplied within erythrocytes. However, the overall growth of these parasites was delayed compared with that of WT parasites in first few days (Figure 2A). This might have been attributed to a reduced ability to invade RBCs. However, by day 9 or 10, after an influx of reticulocytes into the peripheral circulation, the growth rate of the 2 parasites was similar, suggesting that the MSP7ko parasites can invade reticulocytes as efficiently as do WT parasites. This was confirmed by calculating the ratio of the percentage parasitemia in mature RBCs and in reticulocytes for WT and MSP7ko parasites during the course of infection (Figure 2B) and was illustrated by the blood smears (Figure 2C). It is possible, therefore, that the MSP7ko phenotype is an impairment of invasion of the mature RBCs. Infectivity of reticulocytes may be the default pathway because WT and MSP7ko parasites invade reticulocytes equally well, consistent with the previous finding showing preferential invasion of reticulocytes by P berghei.^17,18  This preference for reticulocyte invasion will be investigated more thoroughly, for example, by examining whether there are certain molecules on the reticulocyte surface that facilitate parasite invasion.

It will be of interest in the future to disrupt other members of the msp7 family, either alone or in various combinations, to see whether any of them is essential for an interaction with MSP1 and parasite invasion.

We thank David Bacon and Dale Raine for the illustrations.