• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of pnasPNASInfo for AuthorsSubscriptionsAboutThis Article
Proc Natl Acad Sci U S A. Mar 15, 2005; 102(11): 4146–4151.
Published online Mar 7, 2005. doi:  10.1073/pnas.0407918102
PMCID: PMC554800

A spatially localized rhomboid protease cleaves cell surface adhesins essential for invasion by Toxoplasma


Apicomplexan parasites cause serious human and animal diseases, the treatment of which requires identification of new therapeutic targets. Host-cell invasion culminates in the essential cleavage of parasite adhesins, and although the cleavage site for several adhesins maps within their transmembrane domains, the protease responsible for this processing has not been discovered. We have identified, cloned, and characterized the five nonmitochondrial rhomboid intramembrane proteases encoded in the recently completed genome of Toxoplasma gondii. Four T. gondii rhomboids (TgROMs) were active proteases with similar substrate specificity. TgROM1, TgROM4, and TgROM5 were expressed in the tachyzoite stage responsible for the disease, whereas TgROM2 and TgROM3 were expressed in the oocyst stage involved in transmission. Although both TgROM5 and TgROM4 localized to the cell surface in tachyzoites, TgROM5 was primarily at the posterior of the parasite, whereas adhesins were sequestered in internal micronemes. Upon microneme secretion, as occurs during invasion, the MIC2 adhesin was secreted to the apical end and translocated to the posterior, the site of cleavage, where it colocalized only with TgROM5. Moreover, only TgROM5 was able to cleave MIC adhesins in a cell-based assay, indicating that it likely provides the key protease activity necessary for invasion. T. gondii rhomboids have clear homologues in other apicomplexans including malaria; thus, our findings provide a model for studying invasion by this deadly pathogen and offer a target for therapeutic intervention.

Keywords: microneme, microneme protein protease 1, regulated intramembrane proteolysis, MIC2, protease

Toxoplasma gondii is a member of the phylum Apicomplexa, which contains obligate intracellular parasites including Plasmodium, the agent of malaria, and Cryptosporidium, an agent of diarrheal disease. During its complex life cycle, T. gondii alternates between three different invasive stages: sporozoites, bradyzoites, and tachyzoites. The first two of these stages are involved in transmission: sporozoites are contained within oocysts that are shed into the environment by cats, whereas bradyzoites are persistent tissue forms that are responsible for chronic infection. Tachyzoites are the replicating form responsible for dissemination within the host during acute infection and, thus, are most often associated with symptoms of toxoplasmosis.

Apicomplexans contain a set of apically localized organelles whose sequential secretion is required for invasion of host cells (1). Polarized attachment of T. gondii to the surface of host cells is mediated by the secretion of adhesins stored in organelles called micronemes (2). Microneme proteins containing a transmembrane domain (TMD), namely MIC2, MIC6, MIC8, and MIC12, form heterologous–homologous complexes that expose a variety of adhesive domains at the surface of the parasite (36). After binding host-cell receptors, adhesins are rapidly redistributed toward the posterior through an actin-myosin-dependent process and, ultimately, released into the medium by proteolysis (2, 7).

Cleavage of MIC2 is required for efficient invasion to occur; in its absence, parasites remain attached to cells in a nonpolarized manner and fail to enter (8). However, the parasite protease responsible for C-terminal processing of adhesins including MIC2, called microneme protein protease 1, has not been identified. Recently, the cleavage of MIC2 and MIC6 has been found to occur within the first few residues of their TMDs (9, 10), providing an important clue regarding the identity of microneme protein protease 1.

Rhomboids are serine proteases that are unique in being able to cleave within the first few residues of their substrate TMDs to release domains to the outside of the cell, unlike other examples of intramembranous processing. These proteases typically have seven TMDs and are proposed to work by forming a catalytic triad within the membrane bilayer involving an asparagine, a histidine, and a serine residue contributed by different TMDs (11). Rhomboid-1 from Drosophila melanogaster (Rho-1) was the first member of this widespread protease family to be identified (11) and is responsible for cleaving Spitz, the primary ligand of the EGF receptor in Drosophila. Spitz is synthesized as an inert transmembrane precursor; its cleavage by Rho-1 leads to secretion of the active EGF receptor ligand.

Importantly, Spitz contains a motif within its TMD that is necessary and sufficient for cleavage by many rhomboid proteases (12). A similar motif exists around the cleavage site of MIC adhesins, and this motif is necessary and sufficient for their cleavage (9, 12). Taken together, these observations suggest that a T. gondii rhomboid protease may be the microneme protein protease 1 activity for invasion of host cells. To address this hypothesis directly, we have cloned rhomboid proteases expressed by T. gondii and investigated their role in the cleavage of adhesins.

Materials and Methods

Identification and Cloning of TgROMs. Bacterial, fly, mouse, human, and plant rhomboids were used to search the EST (www.cbil.upenn.edu/apidots) database and the complete draft sequence of the genome (http://ToxoDB.org) of T. gondii by using tblastx. Five putative nonmitochondrial T. gondii rhomboid proteases were identified. TgROM1 (scaffold no. TGG_995345, chromosome VIII), TgROM2 (scaffold no. TGG_995366, chromosome VI), TgROM3 (scaffold no. TGG_995283, chromosome V), TgROM4 (scaffold no. TGG_995368, chromosome VIII), and TgROM5 (scaffold no. TGG_994723, chromosome Ia). Complete ORFs for each gene were amplified from full-length cDNAs generated by using a SMART cDNA library construction kit (Becton Dickinson) from the RH strain, the TgRH* tachyzoite cDNA or VEG sporozoites cDNA, by using the high fidelity polymerase Klentaq LA. The sequences of completed cDNAs were confirmed by cycle sequencing with bigdye 3.1. Cloning and sequencing primers are available on request.

Phylogenetic Analysis. The protein sequences for 24 rhomboids representing the major branches of rhomboids were aligned by using clustalx 1.81 (13) with a gap opening of 10, a gap extension of 0.1, and using protein weight matrix Gonnet 250. Regions of the alignment, where <75% of taxa contained data, were excluded from the phylogenetic analysis. Dendrograms were generated in paup* 4.0 (14) by neighbor-joining and parsimony methods and bootstrapped for 1,000 replicates. Identical trees were obtained from both analyses.

Total RNA Extraction and RT-PCR. Total RNA from the oocysts was extracted, ethanol precipitated twice, and resuspended in RNase-free water (15). Total RNA from Me49 tachyzoites and bradyzoites were obtained after incubation in TRIzol for 5 min at room temperature, extraction with 20% chloroform, and precipitation with 50% isopropanol. RNAs were resuspended in water to a final concentration of 1 mg/ml. mRNAs of genes of interest were reverse transcribed with SuperScript II, and the cDNAs obtained were then amplified for 30 cycles by using Klentaq LA polymerase.

Growth of Host Cells and Parasites. T. gondii tachyzoites of the RH hxgprt (obtained from David Roos, University of Pennsylvania, Philadelphia), the Me49, and the VEG strains were maintained by growth in monolayers of human foreskin fibroblasts as described in ref. 8. Bradyzoites of the Me49 strain were obtained by in vitro differentiation by cultivation in RPMI medium 1640/1% FBS/50 mM Hepes, pH 8.1 at 37°C for 5–7 days in presence of air (16). Fully developed cysts (50–60% of parasites) were harvested by scraping the monolayer and the bradyzoites released from the cysts by digestion in 170 mM NaCl-pepsin (0.1 mg·ml–1)/60 mM HCl for 1 min at 37°C and neutralization with 94 mM Na2CO3. Oocysts of the vascular endothelial growth strain (partially sporulated for 48 h and representing a mixture of VEG oocysts at different stages of sporulation) were obtained as described in ref. 17.

Expression of Tagged TgROMs in T. gondii. Full-length ORFs and an N-terminal hemagglutinin (HA) 9 tag were inserted into the plasmid pGRA1/LacZ (18) replacing the LacZ gene. To express TgROMs under their endogenous promoters, ≈1 kb genomic regions upstream of the start codon were used to replace the GRA1 promoter in the above vector. The promoters and tagged genes were finally cloned into the pHLEM vector and transfected into the RH hxgprt strain with a plasmid harboring the selectable marker HXGPRT (19). Transfected parasites were passaged under drug selection and analyzed between passages 6 and 11. At this point, only a portion of parasites expressed the tagged protein, ranging from 10% to 30% of the total cells.

Indirect Immunofluorescence Microscopy. TgROMs were detected in transiently or semistably transfected parasites grown in human foreskin fibroblast cells. In some cases, extracellular parasites were incubated with 1% ethanol for 1 min at 37°C to induce the secretion of microneme proteins. Cells were fixed in 3.7% paraformaldehyde for 30 min at 4°C, blocked in PBS/Ca2+ and 10% FBS and permeabilized by using 0.05% saponin. Cells were incubated with rabbit anti-HA9 (Zymed), anti-SAG1 (monoclonal DG52) and/or anti-MIC2 (monoclonal 6D10) (20) antibodies and were examined by using a Zeiss Axioplan microscope equipped with phase contrast and epifluorescence optics.

Rhomboid Cleavage Assay. COS cell transfection experiments were performed as described in refs. 11, 12, and 21 by using FuGENE 6 and 250 ng of GFP-Spitz (11) and Star (11), GFP-Spi-type α TGF (TGFα) chimeras (12), or GFP-MIC2, 100 ng of Rho-1 (11), or 100 or 250 ng of TgROMs. TgROMs were tagged with a triple HA tag immediately after the initiating methionine. Medium was removed 24 h after transfection, and cells were incubated in fresh serum-free DMEM for 24 h, except for inhibitor experiments in which media was removed after 2 h and concentrated. Cleavage of Spitz and MIC2 were tested in the presence of Batimastat and BB1101 (20 μM), respectively, which are potent metalloprotease inhibitors. Proteins in supernatants or in cells were detected by Western blot analysis.

Electron Microscopy. Parasites were fixed in 4% paraformaldehyde/0.5% glutaraldehyde in 100 mM Pipes for 1 h at 4°C. Samples were then embedded in 10% gelatin and infiltrated overnight with 2.3 M sucrose/20% polyvinyl pyrrolidone in Pipes at 4°C. Samples were frozen in liquid nitrogen and sectioned with a cryoultramicrotome. Sections were probed with rabbit anti-HA (1513) followed by a 18-nm colloidal gold-conjugated anti-rabbit antibody, stained with uranyl acetate/methyl cellulose, and analyzed by transmission electron microscopy. Parallel controls omitting the primary antibody were consistently negative at the concentration of colloidal gold conjugated secondary antibodies used in these studies.


Identification of Rhomboid-Like Genes in T. gondii. Five putative rhomboids, TgROM1 (AY587210; 249 aa), TgROM2 (AY704176; 283 aa), TgROM3 (AY587209; 264 aa), TgROM4 (AY704175, 634 aa) and TgROM5 (AY587208; 585 aa), were identified by searching the EST database and in the recently completed genome of T. gondii (Fig. 1A and Table 1, which is published as supporting information on the PNAS web site) (http://toxodb.org/ToxoDB.shtml). Their predicted translation products contained the conserved residues belonging to the catalytic triad, an invariant upstream glycine predicted to form the oxyanion-binding hole, and seven TMDs similar to other eukaryotic rhomboids (Fig. 1A; see also Fig. 8, which is published as supporting information on the PNAS web site). TgROMs 4 and 5 were unusual in having long hydrophilic C termini and large loops connecting several TMDs. An additional putative rhomboid, TgROM6 (TgTwinScan_4036), was also encoded in the genome sequence (D. Soldati, personal communication), but it contained a predicted mitochondrial targeting sequence and 6 TMDs, typical of the PARL protein family of mitochondrial rhomboids that are involved in mitochondrial membrane remodeling (not cell surface shedding) (22).

Fig. 1.
T. gondii contains five nonmitochondrial rhomboids, all of which are closely related to the Rho members of the family. (A) Schematic representation of the predicted T. gondii rhomboids. The seven TMDs (gray bars) are numbered. The three conserved catalytic ...

To better depict the relationship between nonmitochondrial T. gondii rhomboids and other members of this class of proteases, we performed a phylogenetic analysis of a select set of rhomboids from flies, mice, humans, plants, bacteria, and parasites. Parasite rhomboids formed three distinct clades that were distantly related to rhomboids in other taxa (Fig. 1B). Notably, several of the T. gondii rhomboids occur in closely related pairs (TgROM1 with TgROM2 and TgROM4 with TgROM5) and, thus, appear to be paralogs, likely derived by gene duplication events. Each of the main groups of T. gondii rhomboids has a single ortholog present in Plasmodium falciparum and Plasmodium yoellii (Fig. 1). In addition, there are at least four distinct rhomboids in Plasmodium that have no apparent homologs in T. gondii (data not shown). Related genes are also found in the genomes of other apicomplexans, including Cryptosporidium, Eimeria, and Neospora (data not shown).

TgROMs Are Active Proteases. To examine the capacity of putative TgROMs to act as proteases, we tested their ability to cleave GFP-Spitz expressed in COS cells, which is an efficient surrogate substrate for many rhomboids from diverse organisms (23). The cleaved form of GFP-Spitz was detected in the media of cells transfected with TgROM1, TgROM3, and TgROM5 (Fig. 2A), whereas cleavage was only apparent in cell lysates with TgROM2 (Fig. 2B). No activity was detected with TgROM4. Increasing the amounts of transfected DNA encoding each active rhomboid increased the amount of cleaved Spitz in cells and media, except for TgROM5 (because high levels of TgROM5 resulted in cytotoxicity). Cleavage of Spitz depends on the active-site serine and its mutation to alanine in TgROM5 completely abolished cleavage (Fig. 2C). All five TgROMs were expressed well in transfected mammalian cells (Fig. 2A), and their relative proteolytic activities were not due to trivial differences in relative expression levels.

Fig. 2.
Activity analysis of TgROMs. (A) Plasmids encoding GFP-Spitz and Star, the transport factor for Spitz (26), were cotransfected into COS cells alone (–), with 100 ng or 250 ng of a plasmid encoding one of the TgROMs, or 100 ng of Drosophila Rho-1 ...

TgROMs Share Substrate Specificity with Drosophila Rho-1. Rhomboids from different organisms are highly specific proteases that rely on helix-destabilizing residues within the top of their substrate TMD for cleavage (12). Accordingly, introduction of mutations into the first seven residues of the Spitz TMD reduced cleavage by Rho-1 but completely abrogated cleavage by all TgROMs, indicating that substrate specificity relies on the same residues within the Spitz TMD (Fig. 3A).

Fig. 3.
Substrate specificity of TgROM proteases. (A) Substrates in schematic form are depicted above each panel (membrane is indicated by two horizontal lines, with extracellular up). Mutating the first seven residues of the Spitz TMD strongly reduced its cleavage ...

Many rhomboid proteases rely, in particular, on a GA residue pair within the 7-aa substrate motif for cleavage (12). Similarly, we compared the ability of TgROMs to cleave two substrates that differed only in the presence or absence of this GA motif (Fig. 3B); inclusion of the GA facilitated cleavage only by TgROM5. This comparison served as an independent and more precise way of assessing substrate specificity and directly confirmed that TgROM5 conforms to the same specificity determinant previously characterized for a large class of rhomboid proteases (12).

Expression Patterns of TgROMs. To begin addressing the biological roles of these rhomboid proteases, we examined their expression patterns by RT-PCR of RNA from the three invasion stages of the parasite: tachyzoites, bradyzoites, and sporozoites (Fig. 4). TgROM1, TgROM4, and TgROM5 were primarily expressed in tachyzoites, the proliferative stage of the life cycle (but were also detected weakly in sporozoites). Conversely, TgROM2 and TgROM3 were expressed primarily in sporozoites. TgROM1 and TgROM4 were also expressed in bradyzoites, although this result may reflect, in part, their expression in tachyzoites because the in vitro tachyzoite-to-bradyzoite conversion was <100%. The dynamic expression of TgROMs in different life cycle stages is consistent with their regulation being primarily transcriptional, like other members of this family, and suggests roles in a variety of biological processes, including invasion.

Fig. 4.
Stage-specific expression of TgROMs. Schematic representation of the three main invasive stages of the T. gondii life cycle. RT-PCR analysis of TgROM expression in invasive stages of the parasite showed that TgROM1, TgROM4, and TgROM5 were expressed primarily ...

Intracellular Localization of TgROMs. Two of the five TgROMs were unlikely to be responsible for cleavage of adhesins during tachyzoite invasion: TgROMs 2 and 3 were not expressed or minimally expressed in tachyzoites. Thus, we analyzed the subcellular localizations of the remaining TgROMs, TgROM1, TgROM4, and TgROM5. Parasite rhomboids were tagged with HA9 at their N termini (because C-terminal tags abolished proteolytic activity; data not shown) and localized by immunofluoresence in transiently transfected parasites.

Expression from their endogenous promoters revealed that TgROM1 colocalized with MIC2 in micronemes in intracellular and extracellular tachyzoites (Fig. 5A). TgROM4 was uniformly distributed at the surface of intracellular and extracellular parasites (Fig. 5B). TgROM5 was primarily on the surface at the posterior end of intracellular parasites, in an opposite pattern from MIC2, which is confined to apical secretory organelles (Fig. 5C). Similarly, in extracellular parasites, TgROM5 was distributed along the length of the cell but was primarily at the back of the parasite and only rarely located at the front. We also examined the distribution of TgROM5 in parasites after induction of microneme secretion by treatment with ethanol as described in refs. 7 and 8 (Fig. 5C Middle and Bottom). Under these conditions, TgROM5 was often concentrated at the extreme back end in a prominent dot that colocalized with MIC2 (arrow in Fig. 5C).

Fig. 5.
Subcellular localization of HA9-tagged TgROM1, TgROM4, and TgROM5 in tachyzoites by using immunofluorescence and electron microscopy. (A) TgROM1 (αHA9) was localized at the apical part of intracellular (Upper) and extracellular parasites (Lower ...

Whereas TgROM4 was evenly distributed along the parasite surface, the distribution of TgROM5 was more heterogeneous. To further evaluate the distribution of TgROM5, we examined its localization by cryoimmunoEM. Immunogold labeling revealed that TgROM5 was distributed in clusters and that it was exposed at the surface of the parasite (Fig. 5D).

Activity of TgROMs on Microneme Adhesins. TgROM1 colocalized with MIC2 in micronemes while TgROMs 4 and 5 were at the cell surface; however, cleavage of MIC protein TMDs does not occur in micronemes, but only once they reach the cell surface. Because the ectodomains of MIC adhesins are very large (Fig. 6A) and could prevent interaction with rhomboid proteases, as is common with other intramembrane proteases, we examined the capacity of TgROMs to process MIC TMDs directly within a native adhesin. Full-length MIC2 was tagged at its N terminus with GFP, and its ability to be cleaved by TgROMs was investigated in the heterologous assay.

Fig. 6.
Cleavage of MIC adhesins by TgROMs. (A) Schematic comparison of full-length MIC2 with the MIC6 and MIC12 synthetic (SYN) substrates used for activity analysis. (B) Full-length TgMIC2 (containing its integrin-like and six thrombospondin domains) was GFP ...

TgROM5 proved to be the only parasite protease capable of processing this substrate efficiently, as detected by the release of GFP-MIC2 into the medium (Fig. 6B). Note that TgROM1 did not have the capacity to cleave MIC2, providing a possible explanation of how they are able to colocalize in micronemes, yet, no cleavage occurs there. The activity of TgROM5 was completely abolished by mutating the catalytic serine to alanine (GSSG→GSAG) (Fig. 6B), establishing that the release depends on its proteolytic activity. The presence of the MIC2-associated protein, which forms a complex with the MIC2 ectodomain, had no effect on the cleavage of MIC2 by TgROM5 (data not shown). Isocoumarins are the only serine protease inhibitors that are known to inhibit all rhomboids tested (24), and MIC2 cleavage by TgROM5 was inhibited by 3,4-dichloroisocomarin (Fig. 6B). Previous studies have indicated that treatment of intact parasites with 3,4-dichloroisocomarin reduces the release of MIC2 into the supernatant, although this effect may also be due in part to decreased secretion of MIC2 from the micronemes (7).

TgROM5 was also efficiently able to cleave the TMDs from MIC6 and MIC12 when they were contained within a chimeric substrate (Fig. 6 C and D). These observations suggest that TgROM5 is the protease primarily responsible for cleaving MIC2, and also indicate that TgROM5 processes multiple microneme adhesins.


Our biochemical and cell biological analyses of T. gondii rhomboid proteases suggest that rhomboids fulfill several distinct developmental functions during the parasite lifecycle. Although further investigation is required to decipher these roles, the analyses presented here allow us to reach several conclusions about their function during invasion.

TgROM5 was the only parasite rhomboid protease to be highly active against the full-length MIC2 adhesin and the TMDs of MIC6 and MIC12. TgROM5 could thus be involved in processing most MIC TMDs during invasion, especially because other MICs, including MIC7, 8, and 9, also share a similar TMD and are likely to serve as good substrates for TgROM5. Consistent with this role, TgROM5 was strongly expressed in tachyzoites, but was also expressed in sporozoites and at low levels in bradyzoites, and was clustered in patches at the surface of tachyzoites. A similar patchy distribution has previously been reported for cell surface-exposed MIC2 (25), and these two proteins were observed to colocalize at the posterior end of the parasite after secretion. Although TgROM5 may also exist as an isoform with a longer N terminus (AY634626), this form is also active (S.U., unpublished observations), and the construct we used to localize TgROM5 is sufficiently long to encode either variant, although it remains unclear which form predominates in the parasite.

Somewhat surprisingly, although TgROM4 is most similar to TgROM5, it was not active against substrates tested here. TgROM4 was uniformly distributed on the cell surface, consistent with a role in processing a cell surface substrate. However, based on this uniform distribution, it would be expected that if TgROM4 were active on MICs, that this result would be a premature release of the adhesins. In contrast, TgROM5 provides an attractive candidate for microneme protein protease 1 because its activity would be regulated by spatial distribution at the back of the organism. Whether TgROM4 and TgROM5 fulfill unique or overlapping functions will require further analysis, including removing their respective activities in vivo by targeted gene disruption.

Taken together, these data suggest a model for MIC adhesin cleavage during invasion (Fig. 7). Adhesins are secreted from micronemes to the apical surface of the parasite at the start of invasion. MIC adhesins bind host-cell receptors, and these complexes are translocated from the anterior to the posterior end of the parasite. TgROM5 is concentrated at the posterior surface, where it cleaves the MIC adhesins upon their translocation, thus releasing the adhesin complexes and allowing parasite entry. Remarkably, this process is strikingly reminiscent of EGF ligand cleavage in Drosophila; EGF ligands are sequestered in the ER, whereas rhomboid proteases are localized in the Golgi apparatus (26). The transport factor Star is required to translocate EGF ligands to the Golgi apparatus, allowing them to be cleaved efficiently by rhomboids and released as active ligands for signaling.

Fig. 7.
Model of MIC adhesin cleavage during parasite invasion. TgROM5 (green) is on the parasite posterior surface and in patches more laterally, whereas MIC adhesins (red) are in micronemes. The sequence of events from left to right is as follows: unattached ...

Cleavage of the transmembrane microneme proteins occurs through a mechanism that is conserved within apicomplexans. PbTRAP, the MIC2 homolog in P. berghei, was correctly processed when expressed in T. gondii (9). The genome of P. falciparum contains eight rhomboid-like genes, three of which are clear homologs of the TgROMs (see Fig. 1), whereas several others are predicted to be mitochondrial. Intriguingly, parasite rhomboids form distinct clades that are not closely similar to any of the previously characterized groupings (27). This divergence may reflect the ancient origin of the phylum Apicomplexa (28, 29) along with the relative absence of molecular sequence data from closely related taxa (i.e., dinoflagellates and ciliates).

There is a serious need for identifying new targets for combating apicomplexan infections, and proteases have proven to be very attractive targets for therapeutic intervention in many other diseases. Based on our findings here, we predict that the Plasmodium rhomboid homologous to TgROM5 (i.e., PfROM4, NP_703414.1) is the primary rhomboid protease involved in P. falciparum thrombospondin-related anonymous protein (PfTRAP) processing during invasion. PfTRAP is essential for sporozoite entry into the salivary glands and invasion of hepatocytes during infection in the vertebrate host (3032). Thus, our work provides a role for rhomboid proteases in processing adhesins important for host-pathogen recognition and suggests a therapeutic target for interrupting infection by sporozoites. Additionally, it is possible that rhomboids play key developmental roles at other stages in the life cycle, and interruption of their functions might offer additional means of interrupting infection by this important group of parasites.

T. gondii rhomboid proteases differed in their expression patterns, subcellular localization, and activities and are thus likely to be involved in multiple functions during the complex life cycle of the parasite. TgROM1 was the most widely expressed rhomboid and, yet, its location within micronemes would appear to preclude its role in processing of MIC protein TMDs. TgROM2 and TgROM3 were largely or exclusively confined to the sporozoite stage. Although the function of these TgROMs remains uncertain, sporozoites undergo a complex program of cell development after meiosis in the oocyst (3335), thus these proteases may play roles in cell signaling or development.

Supplementary Material

Supporting Information:


We thank Wandy Beatty for performing the cryoimmunoEM, Michael White and Jay Radke (Montana State University, Bozeman) for the generous gift of total RNA from sporozoites, Kelinag Tang for construction of full-length cDNA, Julie Suetterlin for expert technical assistance, Roy Black (Immunex) for providing BB1101, and Mike Wolfe for generously providing space and resources to S.U. S.U. was supported by the International Human Frontier Science Program. F.B. was supported in part by National Institutes of Health Grant AI 34036. L.D.S. is a recipient of a Scholar Award in Molecular Parasitology from the Burroughs Wellcome Fund.


Author contributions: F.B., L.D.S., and S.U. designed research; F.B., T.J.J., and S.U. performed research; F.B. and S.U. contributed new reagents/analytic tools; F.B., L.D.S., and S.U. analyzed data; and F.B., L.D.S., and S.U. wrote the paper.

This paper was submitted directly (Track II) to the PNAS office.

Abbreviations: HA, hemagglutinin; TgROM, Toxoplasma gondii rhomboid; TMD, transmembrane domain; TGFα, type α TGF.

Data deposition: The sequence reported in this paper has been deposited in the GenBank database (accession no. AY587208–AY587210, AY704175, and AY704176).


1. Carruthers, V. B. & Sibley, L. D. (1997) Eur. J. Cell Biol. 73, 114–123. [PubMed]
2. Carruthers, V. B., Giddings, O. K. & Sibley, L. D. (1999) Cell. Microbiol. 1, 225–236. [PubMed]
3. Jewett, T. J. & Sibley, L. D. (2004) J. Biol. Chem. 275, 9362–9369. [PubMed]
4. Reiss, M., Viebig, N., Brecht, S., Fourmaux, M., Soete, M., Di Cristina, M., Dubremetz, J. F. & Soldati, D. (2001) J. Cell Biol. 152, 563–578. [PMC free article] [PubMed]
5. Cérede, O., Dubremetz, J. F., Bout, D. & Lebrun, M. (2002) EMBO J. 21, 2526–2536. [PMC free article] [PubMed]
6. Meissner, M., Reiss, M., Viebig, N., Carruthers, V. B., Toursel, C., Tomavo, S., Ajioka, J. W. & Soldati, D. (2001) J. Cell Sci. 115, 563–574. [PubMed]
7. Carruthers, V. B., Sherman, G. D. & Sibley, L. D. (2000) J. Biol. Chem. 275, 14346–14353. [PubMed]
8. Brossier, F., Jewett, T. J., Lovett, J. L. & Sibley, L. D. (2003) J. Biol. Chem. 278, 6229–6234. [PubMed]
9. Opitz, C., Di Cristina, M., Reiss, M., Ruppert, T., Crisanti, A. & Soldati, D. (2002) EMBO J. 21, 1577–1585. [PMC free article] [PubMed]
10. Zhou, X. W., Blackman, M. J., Howell, S. A. & Carruthers, V. B. (2004) Mol. Cell Proteomics 3, 565–576. [PubMed]
11. Urban, S., Lee, J. R. & Freeman, M. (2001) Cell 107, 173–182. [PubMed]
12. Urban, S. & Freeman, M. (2003) Mol. Cell 11, 1425–1434. [PubMed]
13. Higgins, D. G., Thompson, J. D. & Gibson, T. J. (1996) Methods Enzymol. 266, 382–402. [PubMed]
14. Swofford, D. L. (1998) PAUP* Phylogenetic Analysis Using Parsimony and Other Methods (Sinauer, Sunderland, MA), Version 4.0.
15. White, M. W. & Radke, J. R. (1997) Methods 13, 158–170. [PubMed]
16. Soête, M., Camus, D. & Dubremetz, J. F. (1994) Exp. Parasitol. 78, 361–370. [PubMed]
17. Radke, J. R., Gubbels, M. J., Jerome, M. E., Radke, J. B., Striepen, B. & White, M. W. (2004) Mol. Microbiol. 52, 93–105. [PubMed]
18. Howe, D. K., Mercier, C., Messina, M. & Sibley, L. D. (1997) Mol. Biochemical Parasitol. 86, 29–36. [PubMed]
19. Donald, R. G. K. & Roos, D. S. (1998) Mol. Biochem. Parasitol. 91, 295–305. [PubMed]
20. Wan, K. L., Carruthers, V. B., Sibley, L. D. & Ajioka, J. W. (1997) Mol. Biochem. Parasitol. 84, 203–214. [PubMed]
21. Urban, S., Lee, J. R. & Freeman, M. (2002) EMBO J. 21, 4277–4286. [PMC free article] [PubMed]
22. McQuibban, G. A., Saurya, S. & Freeman, M. (2003) Nature 423, 537–541. [PubMed]
23. Urban, S., Schlieper, D. & Freeman, M. (2002) Curr. Biol. 12, 1507–1512. [PubMed]
24. Urban, S. & Wolfe, M. S. (2005) Proc. Natl. Acad. Sci. USA, 102, 1883–1888. [PMC free article] [PubMed]
25. Jewett, T. J. & Sibley, L. D. (2003) Mol. Cell 11, 885–894. [PubMed]
26. Lee, J. R., Urban, S., Garvey, C. F. & Freeman, M. (2001) Cell 107, 161–171. [PubMed]
27. Koonin, E. V., Makarova, K. S., Rogozin, I. B., Davidovic, L., Letellier, M. C. & Pelligrini, L. (2003) Genome Biol. 4, R19. [PMC free article] [PubMed]
28. Escalante, A. A. & Ayala, F. J. (1995) Proc. Natl. Acad. Sci. USA 92, 5793. [PMC free article] [PubMed]
29. Gajadhar, A. A., Marquardt, W. C., Hall, R., Gunderson, J., Ariztia-Carmona, E. V. & Sogin, M. L. (1991) Mol. Biochem. Parasitol. 45, 147–154. [PubMed]
30. Wengelnik, K., Spaccapelo, R., Naitza, S., Robson, K. J. H., Janse, C. J., Bistoni, F., Waters, A. P. & Crisanti, A. (1999) EMBO J. 18, 5195–5204. [PMC free article] [PubMed]
31. Matuschewski, K., Nunes, A. C., Nussenzweig, V. & Ménard, R. (2002) EMBO J. 21, 1597–1606. [PMC free article] [PubMed]
32. Sultan, A. A., Thathy, V., Frevert, U., Robson, K. J. H., Crisanti, A., Nussenzweig, V., Nussenzweig, R. S. & Menard, R. (1997) Cell 90, 511–522. [PubMed]
33. Cornelissen, A. W. & Overdulve, J. P. (1985) Parasitology 90, 35–44. [PubMed]
34. Dubey, J. P., Speer, C. A., Shen, S. K., Kwok, O. C. H. & Blixt, J. A. (1997) J. Parasitol. 83, 870–882. [PubMed]
35. Ferguson, D. J. P., Birch-Andersen, A., Siim, J. C. & Hutchison, W. M. (1979) Acta Pathol. Microbiol. Scand. 87, 183–190. [PubMed]

Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...