Analysis of the Plasmodium and Anopheles transcriptomes during oocyst differentiation

Abstract

Understanding the life cycle of the malaria parasite in its mosquito vector is essential for developing new strategies to combat this disease. Subtractive hybridization cDNA libraries were constructed that are enriched for Plasmodium berghei and Anopheles stephensi genes expressed during oocyst differentiation on the midgut. Sequencing of 1485 random clones led to the identification of 1137 unique expressed sequence tags. Of the 608 expressed sequence tags with data base hits, 320 (53%) had significant matches to the non-redundant protein data base, whereas 288 (47%) with matches only to genomic data bases represent novel Plasmodium and Anopheles genes. Transcription of six novel parasite genes and two previously identified asexual stage genes was up-regulated during oocyst differentiation. In addition, the mRNA for an Anopheles fibrinogen domain gene was induced on day 2 after an infectious blood meal, at the time of ookinete to oocyst differentiation. The subcellular distribution of MAEBL, a sporozoite surface protein, is developmentally regulated from presumed storage organelles in day 15 oocysts to uniform distribution on the surface in day 22 oocysts. This redistribution may reflect a sporozoite maturation program in preparation for salivary gland invasion. Furthermore, apical membrane antigen 1, another parasite surface molecule, is translationally regulated late in sporozoite development, suggesting a role during infection of the vertebrate host. The present results and those of an accompanying report (Abraham, E. G., Islam, S., Srinivasan, P., Ghosh, A. K., Valenzuela, J., Ribeiro, J. M., Kafatos, F. C., Dimopoulos, G., & Jacobs-Lorena, M. (2003) J. Biol. Chem. 279, 5573-5580) provide the foundation for studies seeking to understand at the molecular level Plasmodium development and its interactions with the mosquito.

Fig. 1. Construction of subtraction libraries

For each library a mixture of equal amounts of cDNA from each of the samples listed in the middle set of three boxes was subtracted from a pool of equal amounts of cDNA from samples in the Infected midgut box at the top. Infected blood, cDNA from blood-stage parasites obtained from infected mice; Uninfected and infected midgut, cDNA prepared from mosquito midguts dissected at the indicated times (hours or days) after a non-infected or infected blood meal, respectively. Unfed, cDNA from midguts of mosquitoes that did not have a blood meal.

Fig. 2. Functional classification of ESTs based on BLASTX and BLASTN similarities

A total of 363 unique sequences from the mid-oocyst library and 245 unique sequences from the late-oocyst library had data base hits (Table I) and were grouped based on their predicted biological function. A, mid-oocyst library. B, late-oocyst library. Genomic Survey Sequence (GSS) is a sequence that has similarity with genomic DNA.

Fig. 3. Analysis of the AsFBN1

A, sequence comparison of the A. stephensi (As) and A. gambiae (Ag) fibrinogen domain protein genes with horseshoe crab tachylectins 5A and 5B. Amino acid residues that are conserved between all the four proteins are shaded light, and residues that conserved within the mosquito proteins are shaded dark. Arrow, putative signal peptide cleavage site; Dotted line, fibrinogen-like domain; *, conserved cysteine residues. AgFBN24, A. gambiae fibrinogen domain protein 24; TL5A, tachylectin 5A; TL5B, tachylectin 5B. B, AsFBN1 is induced by Plasmodium in the mosquito midgut. Northern blot analysis shows that the gene is induced at 48 h after an infected but not after a non-infected, blood meal. The numbers refer to the time in hours after a blood meal at which the midguts were dissected. 0h, sugar-fed mosquitoes; UInf, uninfected blood meal; Inf, infected blood meal. C, Northern blots such as the one shown in B were quantified using phosphorimaging. The signals were normalized to a mosquito loading control (mitochondrial rRNA; Ref. 29) and are plotted relative to the value of sugar-fed controls. Gray and black bars represent expression profiles from two independent sets of RNA samples.

Fig. 4. Temporal expression patterns of selected ESTs from the subtraction libraries

A, reverse transcription-PCR products of samples indicated above each lane were amplified for 25 cycles with primers for genes specified to the left of each panel, fractionated by gel electrophoresis, and transferred onto nylon membranes. Radioactive probes generated by PCR amplification of the corresponding cloned EST were hybridized to the membranes. Plasmodium ribosomal protein gene (PbRP; accession number BF295783) was used as a loading control, and the circumsporozoite protein gene (accession number M14135) was used as a positive control. Similar profiles were obtained in at least four experiments with at least two independently isolated RNA samples. M, clones from the mid-oocyst library; L, clones from the late-oocyst library; bl, RNA from blood stage parasites; 24 h N-in, RNA from guts dissected 24 h after a non-infected blood meal; 24 h Inf, RNA from guts dissected 24 h after an infected blood meal; D4, D9, and D15, RNA from guts dissected at the indicated number of days after an infected blood meal; Unknown, EST having homology to the Plasmodium genomic data base that does not have a predicted open reading frame; Hypothetical, EST having homology to the Plasmodium genomic data base that has a predicted open reading frame. B, linearity of signal response. Increasing amounts of template (relative amounts indicated at the top of each lane) were amplified for 25 cycles and analyzed as in panel A. Note that the strength of the signal is proportional to the amount of RNA template.

Fig. 5. Differential localization of P. berghei MAEBL in oocyst and salivary gland sporozoites

Anti-MAEBL and anti-CS antibodies were used to localize the corresponding proteins on sporozoites. The source of sporozoites was as follows: MG D15, midguts on day 15 after infection; MG D22, midguts on day 22 after infection; SG D25, salivary glands on day 25 after infection. The sporozoites were counterstained with 4,6-diamidino-2-phenylindole (DAPI) to show the location of the nucleus. The MG D15 pattern was consistently observed in three independent preparations. No reproducible differences of protein distribution between day 22 MG and day 25 SG sporozoites were observed.

Fig. 6. Regulation of AMA1 expression in sporozoites

A, anti-AMA1 and anti-CS antibodies were used to localize the corresponding proteins on sporozoites. The source of sporozoites was as follows: MG D15, midguts on day 15 after infection; MG D22, midguts on day 22 after infection; SG D25, salivary glands on day 25 after infection. Sporozoites with no AMA1 expression were stained with 4,6-diamidino-2-phenylindole, which labels the nucleus, to show the presence of the parasite. B, cross-sectional view of salivary gland sporozoites labeled with anti-AMA1 antibody. Immuno-electron microscopy shows the surface localization of AMA1 on salivary gland-invaded sporozoites. No staining was observed with midgut sporozoites (data not shown).