Functional characterization of heat shock proteins 40 from plasmodium falciparum

Abstract

Molecular chaperones are ubiquitous proteins that assist in the folding and unfolding or assembly and disassembly of other proteins or protein complexes. One of the first experiments that led to the discovery of these proteins was when Drosophila that were subjected to heat stress exhibited puffing patterns at specific chromosomal loci. These loci later turned out to encode for specific genes whose transcription was up-regulated upon exposure to high temperatures, and these proteins were christened as “Heat shock proteins” (Hsps). Subsequent studies demonstrated that Hsps participated in or “chaperoned” the proper folding of nascent polypeptide chains. Since then, our awareness of the diverse roles played by this group of proteins has steadily increased, and today their functions have been implicated in processes ranging from protein folding, assembly, and regulation of signal transduction to differentiation and development in higher eukaryotes. Recently, molecular chaperones have also been implicated in regulating evolution. Molecular chaperones have been classified into different classes based on their molecular sizes—Hsp90, Hsp70, Hsp60, Hsp40, and the small Hsps. Each class has a conserved domain architecture that allows them to perform specific functions. Hsp90 is crucial for the growth and development of several cell types, and Hsp90 knockouts in yeast and mice have shown to be lethal. Hsp70s are an important group of molecular chaperones that participate in a diverse array of cellular functions such as protein folding, protein degradation, apoptosis, and protein trafficking, to name a few. Hsp70 is up-regulated in cells upon exposure to stressors such as high temperature, toxic chemicals, and nutrient deprivation. Hsp60 is typically organellar in eukaryotic cells and is present as GroEL chaperonin in bacteria. Hsp60 is thought to be important in organellar transport of proteins. Hsp40s are a group of Hsp70 co-chaperones that regulate the ATPase activity of Hsp70 and confer substrate specificity to this chaperone. Molecular chaperones have been shown to be closely linked to tumorigenesis. Specifically, Hsp90 is implicated in this process since it chaperones several cell signaling proteins, including many transcription factors and protein kinases, which are mutated in cancer cells. Therefore, inhibitors of Hsp90 have been shown to be especially effective as anti-cancer agents. Molecular chaperones have gained importance in the recent past as drug targets in infectious diseases. Plasmodium falciparum, the protozoan parasite that causes malaria, has been shown to rely heavily upon its repertoire of molecular chaperones for growth and development in its life cycle within human erythrocytes. Hsp90 inhibitors have been shown to not only decrease parasitemia but also to inhibit the progression of the parasite from the early ring stage to trophozoite stage. Recently, Hsp70 inhibitors have also been shown to abrogate parasite growth in cultures. Almost 2% of the parasite genome encodes for molecular chaperones. Chaperones belonging to the Hsp90, Hsp70, and Hsp60 classes in the parasite have been explored in earlier studies. However, the Hsp40 class of chaperones is still underexplored in the parasite. The Hsp40 class of co-chaperones are known to modulate Hsp70 activity in most systems by recruiting them into various pathways and up-regulating its ATPase activity. All members of the Hsp40 co-chaperone family share a highly conserved 70 amino acids long “J-domain”, which in turn contains a signature tri-peptide motif (H-P-D) through which they interact with the N-terminal domain of Hsp70. We have first identified all the Hsp40 family Hsp70 partners. The two Hsp40 selected for further investigations were PfJ4 and KAHsp40. We generated rabbit antisera against them and investigated the expression, localization, and interactions of the endogenously expressed PfHsp40s in the blood stages of P. falciparum. The main findings from our study are summarized below. PfJ4 - an intra-parasitic Hsp40 PfJ4 (PFL0565w) is a type II Hsp40 (lacking the Cys-rich Zn finger motif) that contains an N-terminal J-domain. The J-domain of PfJ4 contains the conserved “H-P-D” motif. Antiserum raised against purified recombinant full-length PfJ4 recognized a single protein of the expected molecular weight (28 kDa) and pI (9.49) in total parasite lysates. The antiserum also pulled down PfJ4 from parasite lysates in immunoprecipitations (IPs). Indirect immunofluorescence analysis (IFA) indicated that PfJ4 synthesis began in the ring stages, and it was found to be present in all the asexual stages of the parasite. As expected, it was found to be intra-parasitic throughout its life cycle in the human blood. The staining pattern of PfJ4 within the parasite was diffuse, indicating a cytosolic localization. We found that PfJ4 was an abundant protein in the trophozoite stages with a concentration of almost 0.5 µg/100 µg of total parasite protein. PfJ4 was also inducible by almost 2-fold when the parasite was subjected to heat shock at 41°C for 2 hours. PfJ4 was therefore an abundant, heat-inducible Hsp40 with a bona fide J-domain capable of interaction with an Hsp70 partner. PfHsp70-I (PF08_0054) is the best-studied Hsp70 in the parasite and also an abundant cytosolic intra-parasitic protein. PfHsp70-I corresponds to almost 1-2% of the total parasite protein at the trophozoite stage. To investigate whether PfJ4 could be a possible co-chaperone of PfHsp70-I, we first carried out size exclusion chromatography with total parasite lysates to examine the molecular complexes in which these chaperones were present within the parasite. We found that PfJ4 was present in a single large complex having a molecular weight of about 669 kDa, whereas PfHsp70-I formed several complexes within the parasite. However, PfHsp70-I was found to peak in two complexes having approximate molecular weights of 440 kDa and 150 kDa. At the same time, a fraction of the PfHsp70-I signal was found to co-elute with the PfJ4 signal at 669 kDa, indicating that a fraction of PfHsp70-I may be associated with PfJ4. To examine the PfJ4 and PfHsp70-I interaction further, we carried out a combination of co-immunoprecipitation (co-IP), denaturing IP, and reciprocal IP assays. We found that the PfJ4 co-IP pulled down four interactors, of which two had molecular weights of approximately 70 kDa and 90 kDa. Denaturing IP of PfHsp70-I revealed that the 70 kDa interactor of PfJ4 could be PfHsp70-I. Reciprocal co-IP carried out with a-PfJ4 and a-PfHsp70-I antisera revealed that while PfHsp70-I was pulled down with PfJ4, PfJ4 was also pulled down with PfHsp70-I, indicating the presence of PfJ4 and PfHsp70-I in a common complex. The identity of PfHsp70-I was confirmed by simultaneous two-dimensional gel electrophoresis of PfJ4 and PfHsp70-I co-IPs, wherein the molecular weight and pI of the 70 and 90 kDa interactors of PfJ4 were found to be identical to that of PfHsp70-I and PfHsp90 in the PfHsp70-I co-IP. Therefore, we concluded that PfJ4 interacted with PfHsp70-I and possibly PfHsp90 within the parasite. PfJ4-PfHsp70-I is the first Hsp40-Hsp70 complex to be reported in the parasite. PfJ4 and PfHsp70-I, both are present within the parasite cytosol, are abundant chaperones, and are heat-inducible, indicating that this chaperone pair performs constitutive roles but may have special significance during heat stress in the parasite. Further, a phylogram drawn with PfHsp40 and yeast Hsp40 clustered PfJ4 along with Djpl, a cytosolic type II Hsp40 in yeast, which has been shown to be involved in peroxisomal protein import. We therefore think that PfJ4-PfHsp70-I might be associated with some aspect of protein trafficking within the parasite. KAHsp40 - An Exported PfHsp40 One of the major events that occurs during the life cycle of P. falciparum is the development of "knobs"—erythrocyte membrane complexes that contain parasite-encoded proteins as well as host-derived proteins. Knobs mediate cytoadherence of infected erythrocytes with endothelial cells of blood vessels and uninfected erythrocytes. Knobs are responsible for most of the complications associated with P. falciparum malaria. Although some understanding exists about proteins that constitute knobs, the mechanism of knob assembly and trafficking of knob components from the parasite is ill-understood. Recent advances in the area of protein trafficking in the parasite have implicated parasite-encoded Hsp40s in the process of export and trafficking of virulence factors such as knob components in the erythrocyte cytosol. Of the 44 Hsp40s identified in P. falciparum, 17 contain a "Plasmodium export element" or PEXEL motif required for export into the erythrocyte. Of these 17 PfHsp40s, KAHsp40 was found to be part of a chromosomal cluster along with two genes known to be essential for knob formation, namely, Knob-associated histidine-rich protein (KAHRP) and P. falciparum erythrocyte membrane protein 3 (PfEMP3). Genes present in chromosomal clusters in P. falciparum are located on neighboring chromosomal loci whose transcripts peak within the same time window during the asexual stages of the parasite. These genes are thought to be, but are not necessarily, functionally associated. Since KAHsp40 formed a chromosomal cluster with KAHRP and PfEMP3, we thought it possible that they could also be functionally associated and therefore pursued investigations on KAHsp40. We first raised antisera against a C-terminal peptide of KAHsp40, which we used to specifically recognize and pull down the protein from total parasite lysate. This antiserum recognized a specific protein of about 48 kDa in western blots and denaturing immunoprecipitations (IPs), reaffirming its specificity towards KAHsp40. This antiserum was used for all the experiments carried out with KAHsp40. We first examined KAHsp40 localization by cell fractionation and immunofluorescence (IFA). To fractionate the infected erythrocyte, we used saponin, a detergent that allows separation of the parasite compartment from the erythrocyte cytosol. Denaturing IPs from both fractions indicated the presence of KAHsp40 in both the parasite compartment and the erythrocyte cytosol. Further, IFA of infected erythrocytes in ring, trophozoite, and schizont stages indicated that PFB0090c was present in the ring stages but was exported into the erythrocyte cytosol only in trophozoite stages. Denaturing IPs performed at different time points during the asexual stages of the parasite confirmed the stage-specific nature of KAHsp40 export. Pulse labeling of parasites followed by denaturing IP indicated that the export of KAHsp40 in the trophozoite stage was rapid, occurring within 15 minutes of its synthesis. As mentioned earlier, KAHsp40 may be functionally associated with knob proteins KAHRP and PfEMP3. In the infected erythrocyte, KAHRP has been shown to be essential for knob formation, and deletion of the KAHRP gene leads to a knobless phenotype. KAHRP is a knob-marker in the infected erythrocyte plasma membrane. We carried out double immunolocalization by IFA for KAHsp40 and KAHRP and found that the two proteins co-localized on the infected erythrocyte plasma membrane in structures reminiscent of knobs, suggesting functional association. Further reciprocal co-IP of KAHsp40 and KAHRP revealed that the two proteins also physically associate with each other in the erythrocyte. To further explore the association of KAHsp40 with knobs, we carried out double immunolocalization by IFA for KAHsp40 and PfEMP1, which is the major cytoadherent ligand that is present in knobs. Interestingly, KAHsp40 and PfEMP1 also co-localized with each other in discrete structures on the erythrocyte plasma membrane. The above results suggested that KAHsp40 was a knob-associated Hsp40. The J-domain of KAHsp40 is located 17 amino acids following the PEXEL sequence and contains the "H-P-D" motif, making it capable of interaction with an Hsp70 partner. However, none of the parasite-encoded Hsp70s are known to be exported into the erythrocyte cytosol. We therefore explored the possibility that KAHsp40 interacted with the host Hsp70 in the infected erythrocyte. Indeed, co-IPs carried out using antibodies against host Hsp70 followed by two-dimensional gel electrophoresis pulled down KAHsp40 from the infected erythrocyte cytosol, indicating interaction between the parasite Hsp40 and host Hsp70. All studies discussed above, as well as other studies that implicate molecular chaperones and other factors in the life cycle and virulence of the malaria parasite, have been carried out with disease models of malaria, such as the laboratory strain 3D7 and P. berghei, the mouse model of malaria. In order to examine the relevance of these proteins during the real disease scenario, we need to examine their expression and functions in parasites as present in malaria patients. Towards this, we have carried out proteomic analyses of clinically isolated malaria parasites, as described below. Clinical Proteomics of Plasmodium falciparum Most of our understanding of infectious diseases, including malaria, comes from experiments carried out with laboratory models of the disease. Although laboratory models have progressed far towards simulating conditions identical to the real disease scenario, they still provide us with a limited understanding of the disease. Essentially, laboratory models lack the all-important interaction between the parasite and the host that exists during host-infection of the parasite. Therefore, it is essential that pathogen biology is understood in its true context, in addition to the invaluable information obtained from laboratory models. Indeed, several aspects of Plasmodium biology have been shown to be severely affected by their immediate environment, which is drastically different under culture conditions and within the host. Although analysis of clinical P. falciparum has been attempted before, they are limited to either transcriptome analyses or analysis of clinically derived parasites that are adapted to culture conditions. There is increasing awareness that most diseases arise due to perturbations of cellular protein networks. In other words, most diseases are proteomics diseases. Prior to our study, proteomics analysis of clinical malaria parasites derived from peripheral circulation of patients and not adapted to culture conditions did not exist. This task is fraught with several challenges, such as low parasitemia of infected patient samples, the presence of only ring stages in peripheral circulation during P. falciparum infection, and contamination of parasite proteins with host proteins, due to which less abundant parasite proteins are not detected. We have, for the first time, extracted the proteome of clinical P. falciparum from patient-derived infected blood and identified about 88 proteins from the same. This list of parasite proteins, although limited, reflects the most abundant proteins present in clinical malaria parasites and contains several proteins uniquely expressed by clinical parasites in the ring stages. We have also detected some proteins for which protein expression evidence did not exist, thereby implicating these proteins in pathogenesis-related functions. We have also identified several well-known as well as potential drug targets in malaria. All the information obtained from this study has now been made available to the malaria research community on the latest release of the Plasmodium database, PlasmoDB.