HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 277, Issue 22, 19511-19520, May 31, 2002

This Article Abstract Full Text (PDF) All Versions of this Article: 277/22/19511    most recent M200500200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Boucher, N. Articles by Papadopoulou, B. Search for Related Content PubMed PubMed Citation Articles by Boucher, N. Articles by Papadopoulou, B.

A Common Mechanism of Stage-regulated Gene Expression in Leishmania Mediated by a Conserved 3'-Untranslated Region Element^*

Nathalie Boucher^§, Ying Wu, Carole Dumas, Marthe Dubé, Denis Sereno^¶, Marie Breton^§, and Barbara Papadopoulou

From the Centre de Recherche en Infectiologie du Centre de Recherche du Centre Hospitalier de Université Laval and the Département de Biologie Médicale, Faculté de Médecine, Université Laval, Québec G1V 4G2, Canada

Received for publication, January 16, 2002, and in revised form, March 20, 2002

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Developmental regulation of mRNA levels in trypanosomatid protozoa is determined post-transcriptionally and often involves sequences located in the 3'-untranslated regions (3'-UTR) of the mRNAs. We have previously identified a developmentally regulated gene family in Leishmania encoding the amastin surface proteins and showed that stage-specific accumulation of the amastin mRNA is mediated by sequences within the 3'-UTR. Here we identified a 450-nt region within the amastin 3'-UTR that can confer amastigote-specific gene expression by a novel mechanism that increases mRNA translation without an increase in mRNA stability. Remarkably, this 450-nt 3'-UTR element is highly conserved among a large number of Leishmania mRNAs in several Leishmania species. Here we show that several of these mRNAs are differentially expressed in the intracellular amastigote stage of the parasite and that the 450-nt conserved element in their 3'-UTRs is responsible for stage-specific gene regulation. We propose that the 450-nt conserved element, which is unlike any other regulatory element identified thus far, is part of a common mechanism of stage-regulated gene expression in Leishmania that regulates mRNA translation in response to intracellular stresses.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Parasites of the genus Leishmania cause cutaneous, mucocutaneous, and visceral infections affecting ~400,000 people each year of the 397 million that are at risk worldwide (1). During its digenetic life cycle, Leishmania alternates between the alimentary tract of the sand fly vector as an extracellular promastigote and the acidic phagolysosomes of macrophages as an intracellular amastigote. Differentiation of the parasite into the amastigote form is a prerequisite for its intracellular survival. Several environmental factors including acidic pH, elevated temperature, and the harmful phagolysosomal milieu trigger cytodifferentiation accompanied by the differential expression of a variety of genes (2-6). Such stage-specific gene expression is crucial for adaptation because Leishmania differentiates from an extracellular to an intracellular parasite. Gene regulation in Leishmania and related trypanosomatids shares unique features that include polycistronic transcription of large RNA units by an -amanitin-sensitive RNA polymerase II, probably in the absence of promoter elements, and pre-mRNA processing into monocistronic mRNAs through a post-transcriptional control mediated by trans-splicing and polyadenylation (7-9). trans-Splicing and polyadenylation are mechanistically coupled in trypanosomatids and recognize regulatory signals that consist of polypyrimidine-rich sequences (10, 11).

Numerous examples in Leishmania species support the notion that developmental regulation of mRNA levels is determined post-transcriptionally by sequences located in the 3'-untranslated regions (3'-UTR)¹ that usually control mRNA stability (12-17). More recently, a novel mechanism of stage-specific regulation affecting pre-mRNA processing has been reported in Leishmania mexicana (18). The role of 3'-UTRs and/or intercistronic regions in stage-specific gene regulation is further supported by the observation that differential expression of tandemly repeated genes is dependent on sequences present downstream of the different copies, which are often divergent within the same genomic locus (18-21). Although several genes differentially expressed in the intracellular amastigote stage have been reported in Leishmania (14, 17, 21-28), the molecular mechanisms that control developmental regulation in this organism are still not well understood.

We have recently identified a Leishmania gene family encoding the amastin surface proteins that are related to Trypanosoma cruzi amastins (29) and showed that these genes are differentially expressed in the intracellular amastigote stage of the parasite and that the 3'-UTR of the amastin mRNA is required for increased mRNA accumulation in amastigotes (17). We have now delineated a 450-nt region within the last third of the amastin 3'-UTR that confers stage-specific regulation and showed that this sequence is highly conserved among the 3'-UTRs of a large number of Leishmania mRNAs, several of which are known or shown in this study to be developmentally regulated in the mammalian-living form of the parasite. We show here that the 3'-UTR 450-nt conserved sequence can increase expression from a reporter mRNA in a stage-specific manner. Regulation by this 3'-UTR element does not increase mRNA abundance or stability, a major way of generating stage-specific gene expression in Leishmania and other trypanosomatids (12-15, 17, 30-32), but instead it increases protein levels, suggesting its implication in translational control. Our data point to a common mechanism of stage-specific regulation in Leishmania that might be utilized by a number of similarly regulated mRNAs. This is the first example of a common mechanism of stage-specific gene expression in protozoan parasites.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Leishmania Growth and Infections-- Leishmania infantum LEM1317 and Leishmania major Friedlin MHOM/IL/80/FRIEDLIN strains have been described previously (33, 34). Leishmania promastigotes were cultured at pH 7.0 and 25 °C in RPMI 1640 medium supplemented with 10% heat-inactivated fetal bovine serum and 5 mg/ml hemin. Adapted axenic amastigotes of L. infantum were maintained in MAA/20 medium at 37 °C in a 5% CO₂ atmosphere as described previously (35). L. major amastigotes were isolated from footpad lesions of infected BALB/c mice as described (36). In vitro J774 murine macrophage infections were done as indicated previously (17, 37). The luciferase (LUC) activity of the recombinant parasites was determined as reported previously (38). The mean LUC activity was expressed in relative light units, and it was ranged from 400,000 to 700,000 in Leishmania grown as promastigotes, from 10,000 to 120,000 in axenic amastigotes, and from 500 to 12,000 in intramacrophage amastigotes. These marked differences in LUC activity depending on the medium and on the conditions of parasite growth are probably due to the extracellular acidic pH in which lysed amastigote cells are exposed and to the higher protease activity found in the phagolysosomes of the macrophages. It has been indeed reported that firefly luciferases are highly susceptible to acidic pH and to proteolysis (39, 40). The level of infection was also determined by optical microscopy examination following Diff Quick staining of cell preparation.

Nucleic Acids and Protein Manipulations-- Leishmania chromosomes were separated by clamped homogeneous electric field electrophoresis as described (41). Total RNA of Leishmania cells was isolated using the guanidinium isothiocyanate method with TRIzol (Invitrogen). Southern and Northern blot hybridizations were performed following standard procedures (42). Probes used in Fig. 5 were amplified by PCR from L. major Friedlin genomic DNA using the following set of primers: P31_02-5', 5'-ATGCAGCGCAGAATCAGCTCTA-3', and P31_02-3', 5'-AAACCCACTTGCGGGCACGA-3' for the Lmflchr31_02 gene; P32_14-5', 5'-GCTGTTGCGTTAGGTGGTGG-3' and P32_14-3': 5'-CCACCGCTGTGAAAACCAGA-3' for the Lmflchr32_14 gene; and AL117263-P1, 5'-ACGCACCTGCAGGCGGTGTCCC-3' and AL117263-P2, 5'-ACACTGCCCCTTCATCTGCC-3' for the 3-ketoacyl-CoA thiolase gene. To prepare soluble protein lysates, Leishmania cells were harvested by centrifugation, washed with Hepes-NaCl, resuspended in lysis buffer (8 M urea, 4% CHAPS, 40 mM Tris-base), and sonicated three times for 30 s. The proteins were quantified by the Amido Black 10B (Bio-Rad), and ~35 µg of total protein extracts were loaded onto a 10% SDS-PAGE. The gels were transferred on a polyvinylidene difluoride membrane (Immobilon-P, Millipore) and blocked for 16 h with PBST (phosphate-buffered saline, 0.1% Tween 20, plus 1% gelatin solution). The first antibody, a goat anti-luciferase pAp (Promega) was diluted at 1:2000 in PBST with 1% gelatin and incubated 90 min with agitation. Following a few washes with PBST, a donkey anti-goat horseradish peroxidase conjugate antibody (Santa Cruz Biotechnology) diluted at 1:5000 in PBST with 1% gelatin was added and incubated for 60 min with the membrane. After additional washes, the final reaction was done with a Renaissance kit (New Life Science Products). LUC protein levels were estimated by densitometric analysis using a PhosphorImager with the ImageQuant 3.1 software.

Recombinant DNA Constructs and Transfections-- Expression vector pSPYNEOLUC was made as described previously (17). To construct vector pSPYNEOLUC-IR, a 467-bp fragment (IR) containing the last 40 nucleotides of the amastin 3'-UTR with the natural poly(A) site (position 3250) and part of the intercistronic sequence between two amastin copies (see Fig. 1A) was amplified by PCR using Pwo DNA polymerase (Roche Molecular Biochemicals) and primers IR-5' (5'-GCTTGCTTTTTGCTTTCTGTCA-3') and IR-3' (5'-GCGGCTCGCCAGTGTAGCAGA-3') and subcloned into pSPYNEOLUC digested with BamHI and filled in by Klenow fragment. To generate the different LUC-chimeric constructs listed in Fig. 1B, various parts of the 3'-UTR of the L. infantum amastin mRNA amplified by PCR by Pwo DNA polymerase and the following sets of primers (P35, 5'-GAGGAAATGAAGTGAAGGCG-3' and P37, 5'-GAGGAACGGAGACAATAATG-3' for amplifying the full-length 3'-UTR; P35 and P40, 5'-TTCCAGGCCTGCAGCGCACG-3' for the first 415 nt; P35 and P41, 5'-CCTCGTCGTCCCCTCGATCA-3' for the first 1000 nt; P37 and P44, 5'-GTGGCTGTCTAACTACACTT-3' for the last 770 nt; P44 and P45, 5'-CCTTGTCTTTGCTCGTCCATTC-3' for the 347-nt subregion within the 770 nt and primers P42, 5'-TGCGGCACGCACCTACACCA-3' and P43, 5'-TAGCGGCCCGCCTTGTCTTTG-3' for the 184-nt subregion) were subcloned into the BamHI site of pSPYNEOLUC filled in by Klenow. The 467-bp fragment (IR) was introduced into HindIII site of the above LUC-chimeric vectors. To construct vectors pSPYNEOLUC-A2 and pSPYNEOLUC-A2-309 shown in Fig. 6, the full-length 3'-UTR of the Leishmania donovani A2 mRNA (A2-5', 5'-GGCTCGGCGTCCGCTTTCCG-3'; A2-3', 5'-TGCACTTTTCGTTTTTCCCGCA-3') (23) and the 309-nt region within the A2 3'-UTR (A2-309-5', 5'-GCGGATCCCGGAAGCGTGGCGA-3', A2-309-3', 5'-CCGGATCCCACCACGAACAA-3') that is homologous to the 450-nt region of the amastin 3'-UTR (see Fig. 4) were amplified using Pwo polymerase and the primers indicated above and subcloned downstream of the LUC gene into the BamHI site of vector pSPYNEOLUC. Vector pSPYNEOLUC-AC008242 (see Fig. 6) was generated by subcloning the PCR-amplified 394-bp region (Lm-ch27/5', 5'-AAGCGCGACGAGAGCACCCT-3', and Lm-ch27/3', 5'-TCGAACAGGGCCATGCGTAT-3') part of the SW3.1 L. major transcript showing homology to the 450-nt conserved element in the amastin 3'-UTR (see Fig. 4), into vector pSPYNEOLUC digested with BamHI and filled in by Klenow. Finally, vector pSPYNEOLUC-31_02 was constructed by subcloning a PCR-amplified fragment of 433 bp (31_02-5', 5'-ACGCCAACGAGTTCTCCAGA-3', and 31_02-3', 5'-GCACAGCTCACCCCCGCCTC-3') present in the 3'-UTR of the Lmflchr31_02 mRNA and shown 73% identity with the amastin 450-nt region into pSPYNEOLUC as indicated above. The 467-bp IR sequence was then introduced downstream of these LUC chimeras as indicated above. 10-20 µg of purified plasmid DNA (Qiagen) was transfected into L. infantum by electroporation as described previously (43). For the majority of the LUC chimeras corresponding to different parts of the amastin 3'-UTR fused to the LUC coding region, polyadenylation started at the amastin mRNA poly(A) site present in the 467-bp amplified IR sequence (see Fig. 1A), as determined by RT-PCR assays (data not shown).

Reporter Gene mRNA Decay-- RNA turnover was measured upon the addition of 10 µg/ml actinomycin D (Sigma) in both promastigote and axenic amastigote cultures of L. infantum-LUC recombinant transfectants as described previously (33). Total RNA was extracted at various time points (0, 1, 3, and 5 h) following actinomycin D treatment. The RNA samples were subjected to Northern blot analysis using the LUC coding region as a probe. The levels of mRNA were normalized by hybridizing the blots with -tubulin probe.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

One-third of the 3'-UTR of the Leishmania Amastin mRNA Is Involved in Stage-specific Gene Regulation-- We have recently identified a new developmentally regulated gene family in Leishmania, which shares a significant homology to the amastin surface proteins of T. cruzi and showed that stage-specific regulation of the amastin mRNA is mediated by sequences within the 3'-UTR (17). To expand our studies on amastin gene developmental regulation, we made a series of chimeric constructs with regions spanning the 3'-UTR and cloned these sequences downstream of the luciferase reporter gene (LUC) (Fig. 1). To direct accurate 5' and 3' processing of the LUC chimeric transcripts, these cassettes were flanked by an upstream -tubulin intergenic region and by a downstream region of 467 bp including the few last nucleotides of the amastin 3'-UTR with the natural poly(A) site of the amastin transcript followed by the amastin intergenic region (17). In trypanosomatids, polyadenylation is often directed by trans-splicing signals that are located 100-400 nt downstream of the polyadenylation site (10, 11, 44, 45). The LUC chimeras were subcloned into a neomycin phosphotransferase (NEO)-expression vector and introduced by electroporation into L. infantum cells (see "Experimental Procedures"). The LUC activity of the stable transfectants was evaluated in promastigotes, axenic amastigotes, and infected murine macrophages in vitro. We have previously shown (17) and reconfirmed in this study that the full-length 3'-UTR of the amastin mRNA is able to increase LUC activity by 13-17-fold specifically in axenic and intramacrophage amastigotes, respectively (Fig. 1B). Deletion mutagenesis demonstrated that the first 1000 nucleotides of the amastin 3'-UTR are not associated with stage-specific regulation of LUC activity. However, the last 770 nucleotides of the amastin 3'-UTR are shown to induce LUC activity by 13-fold in axenic amastigotes and by 24.5-fold in intramacrophage amastigotes but not in promastigotes (Fig. 1B). Deletions spanning the 770-nt region of the amastin 3'-UTR significantly decreased regulation (Fig. 1B). Indeed, a subregion covering the first 347 nucleotides of the 770-nt region showed a 4-5-fold decrease in LUC activity in comparison with the LUC-770 construct (Fig. 1B). Further deletion within the 347-nt region giving rise to the LUC-184-nt chimeric construct has decreased LUC activity by another 50% (Fig. 1B). Thus, these deletion studies localized the cis-acting regulatory element responsible for stage-specific regulation of the amastin mRNA to a 350-770-nt region at the end of the amastin 3'-UTR.

View larger version (22K): [in this window] [in a new window]   Fig. 1.   Deletion mutagenesis within the 3'-UTR of the Leishmania amastin mRNA to delineate the region that directs stage-specific gene expression. A, schematic representation of the L. infantum amastin mRNA. The different PCR-amplified fragments corresponding to defined regions within the amastin 3'-UTR are indicated. The 467-bp region (IR) harboring the last 40 nucleotides of the amastin 3'-UTR with the poly(A) site (position 3250) followed by the downstream intercistronic region of the amastin gene is also indicated. ORF, open reading frame. B, structure of the different LUC-chimeric vectors transfected into L. infantum. Various regions of the amastin 3'-UTR were inserted between the LUC coding region and the 467-bp region (IR). In these expression vectors, the NEO transcript is processed at the 5'-end by a 92-synthetic polypyrimidine stretch followed by an AG (indicated here by the hatched box) that provides trans-splicing signals (68) and at the 3'-end by sequences within the -tubulin intergenic region (IR) recognized by the polyadenylation machinery, and the LUC chimeric mRNAs are processed by the -tubulin intergenic region at the 5'-end (~200 nt upstream of the LUC start codon) and by the 467-bp region at the 3'-end. This sequence is important for 3' processing of the chimeric transcripts, and as in Leishmania polyadenylation is often directed by trans-splicing signals that are located 100-400 nt downstream of the poly(A) site (10). The effect of 3'-UTR deletions on LUC activity was measured in L. infantum-LUC recombinant promastigotes and amastigotes. Amastigotes were isolated either from axenic cultures or from infected murine macrophages in vitro (see "Experimental Procedures"). The results are presented as the relative luciferase fold increase compared with the control transfectant for each growth condition and are the averages of four separate experiments. The values have been normalized with the copy number of the LUC-expressing vectors present in the different Leishmania transfectants.

The cis-Acting Regulatory Region within the Amastin 3'-UTR Confers Stage-specific Gene Expression by a Mechanism Increasing mRNA Translation-- As previously reported for several amastigote-specific transcripts in Leishmania, mRNA abundance is mainly associated to a mechanism controlling mRNA turnover (13, 17, 22). Such an increase in mRNA stability was previously observed with the full-length 3'-UTR of the amastin mRNA (17). We have now examined the role of the 770-nt regulatory region of the amastin 3'-UTR in inducing LUC mRNA stability in a stage-specific manner. To our surprise, the accumulation and/or the stability of the LUC-770 mRNA did not increase in the amastigote stage compared with promastigotes (Fig. 2), despite the fact that the 770-nt region is capable of inducing LUC activity by 25-fold specifically in amastigotes (Fig. 1B). Similarly, the 347-nt subregion shown to induce LUC activity by 6-fold in amastigotes (Fig. 1B) did not increase LUC mRNA stability or accumulation (Fig. 2). We have observed an overall increase in LUC-770 mRNA stability over that of the LUC-347 mRNA but in both developmental stages of the parasite (Fig. 2A). Moreover, this difference was not translated into a higher LUC activity in promastigotes for the LUC-770 construct when compared with LUC-347 (Fig. 1B). These data suggest that the cis-acting regulatory element residing within the 770-nt region of the amastin 3'-UTR is not involved in mRNA stability or in another mechanism of post-transcriptional regulation controlling mRNA abundance. Thus, the increased stability of LUC mRNA in amastigotes mediated by the full-length amastin 3'-UTR and the differential accumulation of the amastin transcript in amastigotes (Ref. 17 and Fig. 5B) should be controlled by another region within the 3'-UTR probably located upstream of the 770-nt region.

View larger version (43K): [in this window] [in a new window]   Fig. 2.   Effect of the cis-acting regulatory element within the amastin 3'-UTR mediating stage-specific gene expression on mRNA steady-state levels. A, kinetics of LUC mRNA decay in L. infantum recombinant transfectants expressing the LUC-770 or LUC-347 chimeric constructs with the last 770 nucleotides of the amastin 3'-UTR or the 347-nt subregion, respectively, fused downstream of the LUC coding region (see Fig. 1B). 5'-end processing of these transcripts is as indicated in Fig. 1B, and 3'-end processing occurs at the amastin poly(A) site, which is part of the 467-bp region (IR in Fig. 1A). Turnover of LUC mRNA was measured in both promastigotes (P) and axenic amastigotes (A) upon incubation of these cells with 10 µg/ml of actinomycin D for various time points (0, 1, 3, and 5 h) prior to RNA extraction. Northern blots were first hybridized with the LUC probe. RNA loading was monitored by hybridization with the -tubulin probe. RNA turnover assays were repeated three times, and similar results were obtained. B, Northern blot analysis of L. infantum promastigotes and axenic amastigotes expressing the LUC-770 and LUC-347 transcripts indicated in A. As a control, we used the LUC-IR construct with the amastin intergenic region fused downstream of the LUC gene (see Fig. 1B). 3' processing of this hybrid mRNA occurred within the last 100 nt of the IR sequence (data not shown). To follow axenic differentiation of L. infantum, the same RNA blot was hybridized to the amastigote-specific amastin gene probe (bottom panel).

The 770-nt region within the amastin 3'-UTR increases LUC reporter activity in a stage-specific manner (Fig. 1B) but not LUC mRNA abundance (Fig. 2), suggesting that regulation may occur at the level of translation. To test this possibility, we measured LUC protein levels in Leishmania recombinant transfectants grown in both developmental stages by Western blot and PhosphorImager analyses. Although there is significantly less LUC-770 mRNA made in amastigotes compared with the mRNA lacking the regulatory region (LUC-IR) (Fig. 2B), the LUC-770 mRNA is highly translated with a 37-fold increase in LUC protein levels (Fig. 3B). This translational control takes place only in the amastigote form of the parasite, and no regulation by the 770-nt region was seen in promastigotes (Fig. 3, compare lanes LUC-IR and LUC-770). The 347-nt subregion also confers an increase in LUC protein levels in amastigotes, although at lower levels (~5-fold), and finally the 184-nt subregion results only in 2-fold induction (Fig. 3B). These data correlate well with the reporter LUC activity results where a ~25-fold increase is conferred by the 770-nt region of the amastin mRNA specifically in amastigotes (Fig. 1B). Altogether, these data clearly establish the involvement of sequences within the 770-nt region of the amastin 3'-UTR in amastigote-specific gene expression through a mechanism that probably regulates translational efficiency.

View larger version (47K): [in this window] [in a new window]   Fig. 3.   Role of the cis-acting regulatory element within the amastin 3'-UTR mediating stage-specific gene expression on mRNA translation. Total protein extracts of L. infantum LUC-recombinant transfectants extracted from promastigote (A) and axenic amastigote (B) cultures were transferred into nitrocellulose membrane, and Western blots were reacted with an anti-LUC antibody as described under "Experimental Procedures." wt, L. infantum wild-type strain; LUC-IR, L. infantum expressing the LUC mRNA fused to the amastin intergenic region (see Fig. 1B); LUC-770, L. infantum expressing the LUC-770 chimeric mRNA (Fig. 1B); LUC-347, L. infantum expressing the LUC-347 mRNA (Fig. 1B); LUC-184, L. infantum expressing the LUC-184 mRNA (Fig. 1B). The LUC protein recognized by the antibody is indicated by an arrow.

A 450-nt Element within the Regulatory Region of the Amastin 3'-UTR Mediating a Stage-specific Translational Control Is Highly Conserved among Many mRNAs in Leishmania-- A rapid adaptation and reversibility to the various environmental conditions encountered upon parasite differentiation and intracellular survival could be very well achieved by developmentally regulated genes that are subjected to a translational control, as has also been reported in other eukaryotic systems (46). To address the question of whether stage-specific translational control mediated by sequences within the 770-nt region of the amastin 3'-UTR might also be encountered in other developmentally regulated mRNAs in Leishmania, we screened the data bases for sequences homologous to the 770-nt region. Remarkably, in silico studies depicted more than 85 BLAST homologies with significant scores, mainly from the ongoing L. major Friedlin genome sequencing project (www.ebi.ac.uk/parasites/leish.html), that include sequences located in the 3'-end of ~60 predicted protein-coding genes and 25 expressed sequence tags (Table I). All of these sequences share a 68-78% identity to the first 450 nucleotides of the 770-nt region within the amastin 3'-UTR (Fig. 4 and Table I). The results obtained by in silico screening were also experimentally confirmed by Southern blot hybridization using the conserved region of the amastin 3'-UTR as a probe. Indeed, this sequence was found to be present in multiple copies in several L. major and L. infantum chromosomes but interestingly did not hybridize to the nonpathogenic Leishmania tarentolae strain, which does not survive for long periods inside macrophages (47) (data not shown).

View larger version (89K): [in this window] [in a new window]   Fig. 4.   Many amastigote-specific transcripts in Leishmania share a highly homologous 450-nt element in their 3'-UTRs. In silico screening using the 770-nt regulatory region of the amastin 3'-UTR as bait depicted a large number of Leishmania sequences displaying a 68-78% identity with the first 450 nucleotides of the 770-nt region. The homologous sequences within the 3'-UTRs of a selected number of amastigote-specific transcripts were aligned using the GCG Pileup program. These include the known L. donovani amastigote-specific genes HSP100 (26), A2 (13), and 5'A2rel (AC010851) (48), the L. major SW3.1 gene (AC008242) encoding histone H1 (25), and the L. major Lmflchr31_02 and Lmflchr34_00 homologs of the amastin gene family and the 3-ketoacyl-CoA thiolase gene (AL117263) shown here to be differentially expressed in the intracellular amastigote stage of the parasite (see Fig. 5). The conserved region within the 3'-UTRs of the HSP100 and A2 amastigote-specific mRNAs was smaller than the average size of 450 nt found in a large number of Leishmania mRNAs (this figure and Table I). Nucleotide 1 corresponds to the position 2500 in the amastin mRNA, and nucleotide 454 corresponds to the position 2948 (see Fig. 1A and Ref. 17).

Almost half of the identified protein-coding genes that might be regulated via the conserved 450-nt element are not annotated or belong to unclassified hypothetical proteins. However, the remaining proteins can be clustered to several classes that include, among others, putative homologs of the amastin gene family in L. major, histones, chaperones, proteins involved in spliceosome assembly, translation initiation, signaling pathways, energy metabolism, cell adhesion, and ubiquitin-dependent protein degradation (Table I). Interestingly, several known amastigote-specific transcripts like the L. donovani A2 (13) and 5'A2rel (48), the L. donovani HSP100 (26, 49), the L. major SW3.1 gene encoding histone H1 (25), and several 3'-end expressed sequence tags from late stationary and amastigote L. major cDNA libraries (www.ebi.ac.uk/parasites/LGN/amastigoteclones.html) also contain sequences highly homologous to the amastin 450-nt element (Fig. 4 and data not shown). The above amastigote-specific genes have been shown to be regulated post-transcriptionally by sequences located in the 3'-UTR (13, 25, 26).

Link between the Conserved 450-nt 3'-UTR Element and Amastigote-specific Gene Expression in Leishmania-- The identification of a 450-nt 3'-UTR element that is conserved among many mRNAs in Leishmania, some of which are known to be differentially expressed in amastigotes (Table I), points to a common regulatory mechanism that might be utilized by a number of stage-specific mRNAs. To test this hypothesis, we examined whether there is a correlation between the presence of the 450-nt conserved element in the 3'-UTRs of these mRNAs and their amastigote-specific gene expression. To this aim, we randomly selected three L. major protein-coding genes from Table I harboring the 450-nt conserved sequence in their 3'-end and used them as probes in Northern blot hybridization studies. These genes correspond to two homologs of the amastin gene family in L. major Friedlin (Lmflchr31_02 and Lmflchr32_14) and to the 3-ketoacyl-CoA thiolase (AL117263). Interestingly, all of these genes were differentially expressed in the intracellular amastigote stage of the parasite (Fig. 5). Hybridization under stringent conditions with each one of the 450-nt conserved elements located in the 3'-end of the above genes also demonstrated a specific accumulation in the amastigote stage (data not shown), further supporting the link between the presence of the 450-nt element within 3'-UTRs and amastigote-specific gene regulation. These are important observations because they suggest that similarly regulated mRNAs are likely to have similar 3'-UTR signals and, therefore, probably share regulatory machinery.

View larger version (43K): [in this window] [in a new window]   Fig. 5.   Leishmania mRNAs harboring the 3'-UTR 450-nt element are developmentally regulated in the amastigote stage of the parasite. Stage-specific expression of selected mRNAs harboring the 450-nt conserved element in their 3'-UTR. Northern blots of total RNA isolated from L. major Friedlin and L. infantum promastigotes (P) and amastigotes (A) were hybridized with specific intragenic probes for the two L. major amastin homologs 31_02 and 32_14 and the L. major 3'-ketoacyl-CoA thiolase gene (AL117263) (Table I). The amastin amastigote-specific gene was used as a control probe. L. major amastigotes were extracted from lymph nodes in the footpads of infected BALB/c mice. In the case of L. infantum, amastigotes were prepared from an axenic culture (see "Experimental Procedures"). The bottom panel represents an ethidium bromide staining of the RNA samples loaded on agarose gel prior to transfer.

To directly assess the role that the conserved 3'-UTR element of mRNAs known or shown here (Fig. 5B) to be differentially expressed in amastigotes may play in their stage-specific regulation, we tested whether these sequences can direct amastigote-specific expression of a reporter gene. Three amastigote-specific genes were selected for this study. First, the conserved element in the 3'-UTR of the L. donovani A2 amastigote-specific mRNA was capable to increase LUC activity specifically in amastigotes by ~6-fold (Fig. 6). This regulation was abolished when the A2-309-nt region was tested in the antisense orientation with respect to LUC mRNA processing. Under the same growth conditions, the full-length 3'-UTR of the A2 mRNA increased LUC activity by 14-fold (Fig. 6). This difference could possibly be attributed to other sequences within the A2 3'-UTR that may regulate stage-specific expression of the A2 mRNA. Similarly, the conserved element in the L. major SW3.1 mRNA (AC008242) conferred a ~4.5-fold increase in LUC activity exclusively in amastigotes and only when cloned in sense orientation (Fig. 6). Moreover, the conserved element residing in the 3'-UTR of the L. major amastin homolog Lmflchr31_02 was shown to increase LUC activity by at least 10-fold in intramacrophage amastigotes (Fig. 6). No regulation was obtained by these homologous 3'-UTR sequences in the extracellular promastigote stage of the parasite (Fig. 6), suggesting that these sequences are only implicated in amastigote-specific gene regulation.

View larger version (20K): [in this window] [in a new window]   Fig. 6.   The conserved 3'-UTR element in several amastigote-specific transcripts confers stage-specific gene regulation. The conserved region within the 3'-UTRs of A2, (309 nt), SW3.1 (394 nt), and the Lmflchr31_02 amastin homolog (433 nt) mRNAs, which is highly homologous to the amastin 450-nt element were subcloned downstream of the LUC reporter gene, transfected into L. infantum, and their role in amastigote-specific gene regulation was evaluated by measuring LUC activity in the transfectants grown under promastigote and axenic amastigote conditions. LUC assays were done in L. infantum, because it is not possible to axenically grow L. major amastigotes. The full-length A2 3'-UTR and the amastin intergenic sequence (IR; see Fig. 1A) were used as controls. 5'- and 3'-end processing of the LUC chimeric mRNAs is as described for Figs. 1 and 2. The results are presented as the relative luciferase fold increase compared with the control transfectant for each growth condition and are the averages of three separate experiments.

The conserved 450-nt 3'-UTR element is part of mRNAs that accumulate at high levels in amastigotes (Fig. 5), but at least for the amastin mRNA, this sequence is not involved in mRNA stability (Fig. 2). Northern blot hybridization and Western blot analysis indicated that similarly to the amastin 3'-UTR conserved element-mediated regulation (Figs. 2 and 3), accumulation of the LUC-A2/309 and LUC-Lmflchr31_02 chimeric mRNAs was comparable in both promastigote and amastigote stages of the parasite, whereas LUC protein levels were specifically increased in amastigotes (data not shown).

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

A Novel 3'-UTR Element Highly Conserved among Several Developmentally Regulated mRNAs in Leishmania-- We have previously identified a developmentally regulated gene family encoding the amastin surface proteins in Leishmania and showed that stage-specific increase in the half-life of the amastin mRNA is mediated by sequences within the 3'-UTR (17). Here, we delineated a region within the last 770 nucleotides of the amastin 3'-UTR that directs stage-specific expression of a reporter gene through a mechanism that increases mRNA translation without an increase in mRNA abundance or stability (Figs. 1-3). It is remarkable that the first 450 nucleotides within this 770-nt regulatory region are highly conserved among a large number of mRNAs in several Leishmania species, some of which are shown to be differentially expressed in the intracellular amastigote form of the parasite (Table I and Figs. 4 and 5). This 450-nt conserved element is often located within 3'-UTRs at different distances (0.35-2 kb) relative to the stop codon in many protein-coding genes of L. major Friedlin, and its presence is directly correlated to the amastigote-specific expression of these genes (Table I, Figs. 1 and 4-6, and data not shown). Our data suggest that developmentally regulated mRNAs in Leishmania harboring the 450-nt 3'-UTR signal are likely to be controlled by the same regulatory machinery.

Few cis-acting regulatory elements within the 3'-UTRs of developmentally regulated mRNAs thought to be responsible for either mRNA turnover or translation have previously been identified in both Leishmania (16, 50) and Trypanosoma species (30-32, 51-53). However, none of these regulatory elements share any significant homology to the 450-nt element identified in this study, indicating its implication in a novel mechanism of stage-specific gene regulation in these organisms. The conserved 3'-UTR 450-nt element is exclusively associated with amastigote-specific gene regulation, because it was not found in any of the known Leishmania metacyclic-specific or promastigote-specific mRNAs (Table II). Moreover, not all of the known amastigote-specific transcripts contain this 450-nt element (Table II), suggesting that stage-regulated gene expression in Leishmania is a complex process probably involving a variety of mechanisms.

A Common Mechanism of Stage-specific Translational Control via a Conserved 450-nt Element within the 3'-UTRs of Many Differentially Expressed mRNAs in Leishmania-- While using the 770-nt region of the L. infantum amastin 3'-UTR harboring the 450-nt conserved element (Figs. 1-3) or the homologous conserved sequence within the 3'-UTRs of an L. major amastin homolog or other known amastigote-specific mRNAs (Fig. 6 and data not shown), we have systematically observed a stage-specific induction in reporter gene activity because of increased mRNA translation in the absence of increased mRNA stability and/or mRNA abundance. Nevertheless, 3'-UTRs of developmentally regulated mRNAs shown here to contain the 450-nt conserved element have been associated with either stage-specific mRNA stability or mRNA abundance (Refs. 13 and 17 and Fig. 5). We propose that stage-specific regulation of this particular subset of Leishmania mRNAs might be controlled by two distinct mechanisms: one acting via the 450-nt conserved element possibly regulating mRNA translation and the other by an adjacent 3'-UTR region that increases mRNA stability. A bipartite mechanism of stage-specific regulation with distinct 3'-UTR elements involved either in mRNA stability or translation has also been reported for the PARP mRNA in trypanosomes (31, 32) and more recently for the Leishmania amazonensis HSP83 3'-UTR (50).

In this study, we show that the conserved 450-nt element in the 3'-UTRs of several Leishmania amastigote-specific mRNAs is an important component of a regulatory machinery that stimulates stage-specific expression probably at the level of translation (Figs. 1B, 3B, and 6 and data not shown). The main advantages of regulating genes through a translational control mechanism are the speed and the readily reversible nature of the response to altering physiological conditions. This enables cells to modulate translational efficiency, energy metabolism, and membrane composition and to activate signal transduction pathways. Several of the genes with predicted functions illustrated in Table I belong to these categories and could very well be regulated at the level of translation via the conserved 450-nt element. However, it is intriguing that expression of genes encoding such diverse cellular functions could possibly be controlled by common regulatory machinery. The major triggering factors that induce Leishmania amastigote differentiation in our axenic culture system are elevated temperature and acidic pH. We have previously shown that the amastin mRNA is rapidly induced upon growth of the parasites at low pH but not at increased temperature (17). Similar results were also obtained with two of the amastin homologs in L. major Friedlin (data not shown) and with the A2 amastigote-specific transcript (13). There are many examples of genes that are regulated by ambient pH in fungi and in bacteria (54, 55). It is thus possible that acidic pH constitutes an important signal for regulating this subset of transcripts via the conserved 450-nt 3'-UTR element.

Translational controls are critical for a variety of developmental processes in a wide range of organisms such as Caenorhabditis elegans, Drosophila, and Xenopus and surprisingly often require regulatory elements in the 3'-UTR (46, 56, 57). The molecular mechanisms by which 3'-UTR elements mediate translational regulation are not well understood. There are models involving regulation either by a looping between 5'- and 3'-UTR sequences (58, 59) or by deadenylation of the target mRNA (60, 61) or by affecting elongation (62) and/or cytoplasmic mRNA localization (reviewed in Ref. 63). All of these examples, however, involve 3'-UTR elements acting as repressors of translation (reviewed in Refs. 64 and 65). 3'-UTR elements that enhance mRNA translation have been identified to act indirectly through their cytoplasmic localization (63, 66) or by promoting cytoplasmic polyadenylation (67). Most of these elements, with the exception of mRNA localization signals, correspond to short AU-rich sequences ranging in size from 4 to 55 nt. The conserved 450-nt 3'-UTR element identified in this study does not contain any of these motifs, and as suggested from deletion analysis (Fig. 1B), the full-size element is required for efficient regulation. Given the extensive length of this 3'-UTR conserved element and its sequence conservation throughout the entire region (Fig. 4 and data not shown), it is likely that a tertiary structure needs to be adopted for an optimal regulation possibly through a protein-RNA interaction. Preliminary data using the mfold algorithm (Bioinformatics Rensselaer) suggest that the conserved 450-nt element found in the 3'-UTRs of several amastigote-specific mRNAs folds into a similar bipartite Y-shaped stem-loop structure (data not shown).

In the present study, we identified a conserved 450-nt element within the 3'-UTRs of a specific class of developmentally regulated mRNAs in Leishmania that is distinct from any other regulatory element directing developmental gene expression not only in trypanosomatid protozoa but also in other eukaryotic organisms. To our knowledge, this is the first report supporting the notion that similarly regulated mRNAs share a common regulatory mechanism via conserved 3'-UTR signals that confer regulation possibly at the level of translation. Translational regulation is poorly studied in protozoan parasites, and future work should permit appreciation of its role in stage-specific gene expression.

	ACKNOWLEDGEMENTS

We thank Drs. Marc Ouellette and Xiao-Fang Huang for critical reading of this manuscript.

	FOOTNOTES

^* This work was supported by Canadian Institutes of Health Research Grant gr-14500 (to B. P.).The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

These authors contributed equally to this work.

^§ Recipient of a Canadian Institutes of Health Research studentship.

^¶ Present address: Laboratoire de Biologie Parasitaire, Institut de Recherche pour le Développement, Montpellier, France.

Member of a Canadian Institutes of Health Research group on host-pathogen interactions. Fonds de Recherche en Santé de Québec Senior Scholar. Burroughs Wellcome Fund New Investigator in Molecular Parasitology. To whom correspondence should be addressed: Centre de Recherche en Infectiologie, Centre Hospitalier Universitaire de Québec, Pavillon CHUL, 2705 boul. Laurier, Ste-Foy, Québec G1V 4G2, Canada. Tel.: 418-654-2705; Fax: 418-654-2715; E-mail: barbara.papadopoulou@crchul.ulaval.ca.

Published, JBC Papers in Press, March 23, 2002, DOI 10.1074/jbc.M200500200

	ABBREVIATIONS

The abbreviations used are: UTR, untranslated region; nt, nucleotide(s); LUC, luciferase; CHAPS, 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonic acid; PBS, phosphate-buffered saline; IR, intergenic region.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES