Hsp60 is targeted to a cryptic mitochondrion-derived organelle ("crypton") in the microaerophilic protozoan parasite Entamoeba histolytica

Abstract

Entamoeba histolytica is a microaerophilic protozoan parasite in which neither mitochondria nor mitochondrion-derived organelles have been previously observed. Recently, a segment of an E. histolytica gene was identified that encoded a protein similar to the mitochondrial 60-kDa heat shock protein (Hsp60 or chaperonin 60), which refolds nuclear-encoded proteins after passage through organellar membranes. The possible function and localization of the amebic Hsp60 were explored here. Like Hsp60 of mitochondria, amebic Hsp60 RNA and protein were both strongly induced by incubating parasites at 42 degreesC. 5' and 3' rapid amplifications of cDNA ends were used to obtain the entire E. histolytica hsp60 coding region, which predicted a 536-amino-acid Hsp60. The E. histolytica hsp60 gene protected from heat shock Escherichia coli groEL mutants, demonstrating the chaperonin function of the amebic Hsp60. The E. histolytica Hsp60, which lacked characteristic carboxy-terminal Gly-Met repeats, had a 21-amino-acid amino-terminal, organelle-targeting presequence that was cleaved in vivo. This presequence was necessary to target Hsp60 to one (and occasionally two or three) short, cylindrical organelle(s). In contrast, amebic alcohol dehydrogenase 1 and ferredoxin, which are bacteria-like enzymes, were diffusely distributed throughout the cytosol. We suggest that the Hsp60-associated, mitochondrion-derived organelle identified here be named "crypton," as its structure was previously hidden and its function is still cryptic.

FIG. 1

Cartoon of RT-PCR products and constructs for testing function and localization of amebic Hsp60. (A) Amebic hsp60 RT-PCR products. Primers are defined in Materials and Methods. (B) Construct used to test amebic Hsp60 in groEL bacteria. GroEL protein sequences are in black. (C) pJST4-Hst60 plasmid used for stable transfection of amebae. The coding regions of the amebic hsp60 gene (striped box) and bacterial neomycin phosphotransferase (black box) were expressed under amebic actin 1 gene promoters. (D) hsp60-myc, trunc-hsp60-myc, and ferredoxin-myc constructs.

FIG. 2

Results of RT-PCR with primers to the amebic Hsp60 by using total RNA from amebae cultured in the absence of heat shock (37°C) and after heat shock (1 h at 42°C). An ethidium-stained agarose gel of RT-PCR products was electronically reversed for ease of reproduction. Control RT-PCR with primers to the amebapore was performed with the same RNA targets. Lanes 3 and 6 contain negative controls (N) with no target RNA. These RT-PCR results are qualitative rather than quantitative.

FIG. 3

High-power view of two-dimensional protein gels of amebae incubated at 37°C and after heat shock (42°C). Arrows mark putative Hsp60s, which were in the same location as recombinant Hsp60 expressed in amebae transfected with the pJST4-Hsp60 plasmid (data not shown). Note that the putative Hsp60 is not the only protein which changes in abundance with heat shock.

FIG. 4

Plate showing groEL mutants incubated at 37°C (permissive temperature) or 44°C (nonpermissive temperature). Each groEL mutant was streaked as a cross, which was marked with a pen. A nontransformed groEL mutant was maintained on plates containing tetracycline (left), while groEL mutants transformed with the pSE380 vectors were maintained on plates containing tetracycline and ampicillin (right). A groEL mutant complemented with the amebic Hsp60-E. coli groEL fusion gene (Eh hsp60) grew at both temperatures. In contrast, the control groEL mutant [(−) vector] failed to grow at 44°C. Similarly, a groEL mutant which was transformed with the pSE380 vector without an insert [(−) insert] failed to grow at 44°C.

FIG. 5

Alignment of the amino and carboxyl termini of the predicted E. histolytica Hsp60 in single-letter code with GroEL proteins of E. coli (X07850) (18) and Clostridium thermocellum (Z68137) and Hsp60s of Dictyostalium discoideum (U72247), T. vaginalis (1, U26966 [5] and 2, U57000 [39]), Plasmodium falciparum (U38963) (50), L. major (U59320) (37), Trypanosoma cruzi (X67473) (49), Euglena gracilis (U49053) (56), Saccharomyces cerevisiae (M33301) (23), Schizosaccharomyces pombe (D50609) (57), Histoplasma capsulatum (L11390), Caenorhabditis elegans (L36035), Drosophila melanogaster (X99341), Homo sapiens (M22382) (45), Arabidopsis thaliana (Z11547), and Cucurbita sp. (pumpkin, X70868) (51). Periods indicate amino acids identical to those of the amebic Hsp60, dashes indicate gaps, and asterisks indicate stop codons. Vertical boxes indicate location of primers used previously to identify a segment of the amebic Hsp60 (7). Horizontal boxes indicate proven organelle-targeting presequences of amebic, trichomonad, Histoplasma, human, and pumpkin Hsp60. The T. vaginalis Hsp60 sequence is a composite of two published sequences, each of which is incomplete, so the total length could not be determined.

FIG. 6

Alignment of the amino-terminal organelle-targeting presequence of E. histolytica Hsp60 with those of T. vaginalis hydrogenosomal proteins (3, 20). Dashes indicate amino acids identical to those of the amebic Hsp60. To identify the amebic presequence, an Hsp60 spot was excised from a two-dimensional gel of amebic proteins and sequenced by Edman degradation.

FIG. 7

Fluorescence confocal micrographs of E. histolytica trophozoites localizing an Hsp60-associated organelle and contrasting it with the amebic cytosol or pinosomes. Antibodies to Hsp60 (A) identified a short, cylindrical organelle in a nontransfected parasite. Antibodies to a myc identified two similar organelles in transfected amebae overexpressing Hsp60 with a C-terminal myc tag (B). However, most parasites labeled with anti-Hsp60 or anti-myc antibodies contained a single organelle. In contrast, anti-myc antibodies had a cytosolic distribution when parasites were transfected with a truncated hsp60 gene encoding an Hsp60 lacking Ser and positively charged residues in its presequence (C). Antibodies to ADH1 in nontransfected parasites (D) gave a cytosolic distribution, with some labeling of the nucleus. Similarly, anti-myc antibodies had a cytosolic, if somewhat granular, distribution when parasites were transfected with a ferredoxin gene (E) encoding a peptide with a myc tag at its carboxy terminus. Pinocytosed FITC-dextran by nontransformed amebae fills hundreds of vesicles (F), most of which were small.