Abstract: We are reporting here, the cloning and characterization of the first cyclin from Leishmania mexicana. We have identified a cyclin-like motif from the L. major genome sequencing project. A cyclin homologue was cloned and sequenced from L. mexicana genome and it showed 96.1% amino acid identity with the putative L. major cyclin. It has also sequence identity to mitotic cyclins from other organisms. Southern analysis showed that it is present as a single copy gene. CYCa has been over-expressed in E. coli as a histidine fusion and western blot has confirmed the immunoreactive property of the recombinant cyclin, which then used to reconstitute active recombinant L. mexicana CRK3. No phosphorylation of histone HI was detected by both wild type and mutated CRK3 on the activation assays suggesting that phosphorylation status and cyclin binding are important for reconstituting protein kinase activity. The results confirm that we have isolated a cyclin molecule from L. mexicana (LmCYCa) which may play an important role in the regulation of the parasite cell cycle.
INTRODUCTION
To date, there are 19 human cyclins; 26 cyclins been identified in the budding yeast; S. cerevisiae and 9 cyclins in the unicellular parasite T. brucei (Doerig et al., 2000; Van Hellemond et al., 2000; Hammarton et al., 2003). Studies on yeast have shown that S. cerevisiae cells are viable in the absence of any two, but not three of CLNs 1-3, suggesting that CLN function is redundant (Reed et al., 1992). Moreover, cyclins C, D and E can functionally substitute for the absence of CLNs 1-3 in yeast when overexpressed (Reed et al., 1992). Further more, the mitotic cyclins, when overexpressed, can substitute for the G1 cyclins in yeast. As a consequence, yeast complementation approach has been established as a tool for identification of novel cyclins from different organisms. Using this approach recently, a novel cyclin has been isolated from the parasite Toxoplasma gondii (Kvaal et al., 2002). The role that some of these cyclins play in the organism is well known, for example the cyclin B/Cdc2 complex acts as the primary promoter of mitosis. On the other hand, the role of others cyclins is not fully understood, even their CDK-partner has yet to be identified. With the advent of the most recent RNA interference technique (RNAi), the role of some trypanosomatid cyclins has been investigated (Hammarton et al., 2003). The activities of the Cdk/cyclin complexes are periodic throughout the cell cycle. This periodicity is achieved by regulation at many levels and is necessary for the proper coordination of cell cycle events (Gould, 2000).
The relative specificity of cyclin localization compared with that of Cdc2 has led to the suggestion that cyclins target Cdks to particular intracellular sites as substrates.
The first putative cyclin to be identified from trypanosome was TbCYC1 (Affranchino et al., 1993). Later work on TbCYC1 (Hua et al., 1997) showed that there is no fluctuation on its expression during the cell cycle. Another work on TbCYC1 (Hammarton et al., 2000) showed its inability to complement a budding yeast G1 cyclin mutant and to bind neither the fission yeast CDK binding protein p13suc1 nor the leishmanial homologue p12cks1. Moreover same workers have shown a 22 amino acid frameshift in the cyclin box of the previously published CYC1. Therefore it is concluded that TbCYC1 is not a mitotic cyclin. Two more cyclins have been isolated from T. brucei (Van Hellemond et al., 2000) CYC2 and CYC3 both are identified by rescue of the Saccharomyces cerevisiae mutant DL1, which is deficient in CLN G1 cyclin function. In addition, the CYC3 is reported with a low level of sequence identity to mitotic B-type cyclins from a variety of organisms; since it contains a destruction box-like motif within its C-terminus. However, the CYC2 has sequence identity to the Neurospora crassa PREG1 and the S. cerevisiae PHO80 cyclin. More over, both CYC2 and CYC3 are present as single copy genes in the T. brucei genome and expressed in all life cycle stages (Van Hellemond et al., 2000). Further more, immuno-precipitation and yeast two-hybrid system have demonstrated the ability of CYC2 to interact with CRK3 from T. brucei (Van Hellemond et al., 2000).
The in vivo overexpression of the trypanosomatid cyclins in the live trypanosome has been investigated. For instance, the expression of the tagged-cyclin in the trypanosome has no effect on either the morphology or the growth as compared to the native over-expressed cyclin (Van Hellemond and Mottram, 2000). The degradation of mitotic B-type cyclins is dependent on the presence of the destruction box (King et al., 1996). The destruction box motif is required for the formation of cyclin-ubiquitin conjugates, which are subsequently targeted to and degraded by the proteasome (King et al., 1996).
The L. major genome sequences are now serving as a basis for examining and comparing specific functional regions of other Leishmania species. Exploring the sequences of the cosmid LM25. Contig98 (LmFlchr25) by blasting the L. major genome with mammalian cyclin sequences, using the VectorNTI software, has revealed the presence of two Open Reading Frames (ORFs). In addition, a blast search of the GenBank/EMBL database with (752 d2) CDS from L. major Contig LM.25 revealed many sequences with significant alignment to all genes that encode G2/mitotic-specific cyclins or G1/S-specific cyclins from different organisms. In the light of these results, we proposed that (752 d2) CDS is a leishmanial cyclin. Therefore, we set our goal to isolate and characterize this cyclin genetically and biochemically from L. mexicana. We also wish for to study the possible in vitro activation of recombinant L. mexicana CRK3 by this cyclin.
MATERIALS and METHODS
Study area: The study was carried out at the Wellcome Centre for Molecular Parasitology, Anderson College and the period of the study was from May 1999 to March 2002.
Cloning of L. mexicana CYCa coding sequence: Genomic DNA was prepared from L. mexicana as previously described (Ali et al., 2003). Two oligonucleotide primers for the PCR reaction were designed. One PCR primer OL813 includes a NdeI restriction site at 5’ end and a priming sequence (752-770 bp) of the LM24. Contig98, which is homologous to the 5’ end of the DNA corresponding to the N-terminus of the putative L. major cyclin motif that serves as the forward primer. The second primer OL814 contains an XhoI restriction site and a priming sequence (1657-1675 bp) of the LM24.Contig98 which is homologous to the 3’ end of the DNA corresponding to the C-terminus of the putative L. major cyclin motif that serves as the reverse primer. Twenty five cycles of PCR were performed with the following parameters: 94°C for 1 min, 55°C for 1 min and 72°C for 2 min. The final extension reaction was increased to 7 min to ensure maximal full-length product. The PCR product of expected size (about 926 bp) was purified and subsequently sub-cloned into pCR-Script Vector (Stratagene) to give plasmid pGL612.
Sequence analysis of LmmCYCa gene: Two recombinant clones, one from each separate PCR reaction, were selected for sequencing. Plasmid isolations from E. coli were performed using Qiaprep columns (Qiagen) and sequencing was done using different primers by the dideoxy chain termination method, using an ABI automatic sequencer. The commercially available T3 and T7 sequencing primers were used together with other sequencing primers based on the LmmCYCa sequence. The obtained sequences were assembled and aligned using the Vector NTI software; the Contig Express and AlignX programs, respectively.
Over-expression of L. mexicana CYCa in E. coli cells: The cloned L. mexicana CYCa fragment was obtained from the construct pGL612. This fragment was then sub-cloned in frame with the histidine-tag in the pET21-a expression vector (Invitrogen) to generate the plasmid pGL630, in which the leishmanial cyclin ORF is tagged by a six-histidine residue from the C-terminus end. This was done in order to over-express the L. mexicana cyclin under the control of the T7 promoter. The resulting pGL630 construct was used to transform the BL-21 (DE3) strain of the E. coli. Expression of the recombinant protein was induced as previously described, with some conditions having been changed. Induction was performed in LB medium containing ampicillin, by adding isopropyl β-D-thiogalactoside (IPTG; Sigma) at a final concentration of 4 mM to culture with an OD600 of 0.6, followed by further 4 h of incubation at 30°C and then overnight at 19°C. The induced samples were analyzed by SDS-polyacrylamide gel electrophoresis and stained with 0.25% Coomassie Blue. Finally, expression of the L. mexicana cyclin in the bacterial cells was checked by western blot analysis using the Mono-clonal anti-(his)4 anti-bodies (Qiagen).
Kinase assay of the L. mexicana CRK3his incubated with the L. mexicana CYCahis: The constructs CRK3his (pGL448), CRK3hisE178 (pGL665) and CYCahis (pGL630) were separately purified. CRK3his was purified by FPLC chromatography (BioCAD). While, CYCahis was partially purified by chromatography over Ni2+-NTA-resin (Qiagen Purification kit). Purified CRK3 was combined with the purified CYChis (equal volumes), alternatively, both lysates were, incubated at 4°C for 4 h, purified on Ni-NTA agarose beads and then the eluates, together with the bound beads were subjected to activation assay. Phosphorylation reactions were performed in a total volume of 20 μL containing KAB. Six different recombinant CRK3his and recombinant CYCahis amounts were used that ranged from 1 to 10 μL for each reaction. Reactions were initiated by the addition of 1 μL (10 μCi) of [γ-32P] ATP diluted in KAB buffer (1M MOPS (pH7.2), 500 mM EGTA, 1 M MgCl2, 500 mM DTT) and incubated at 30°C for 20 min. Reactions were terminated by adding 20 μL Laemmli SDS-PAGE loading buffer (2x) and heating to 100°C for 5 min. Incorporation of [γ-32P] ATP into the substrate histone H1 was then quantified by separation of samples on 12.5% Laemmli gels and exposing dried gels to autorad film.
Southern blot analysis of L. mexicana CYCa: Genomic DNA was prepared from 10×108 promastigotes. Cells were collected by centrifugation and incubated in 4% SDS at 65°C for 30 min in SE buffer (100 mM EDTA pH 8.0, 300 mM NaCl, 10 mM Tris HCl pH 7.5). A volume of 25 μL (20 mg mL-1) proteinase K were added and incubated at 37°C for another 30 min. DNA was extracted using the phenol-chloroform extraction method. Five micrograms of genomic DNA was digested with BamHI, BglII, EcoRI, HindIII, NcoI, SalI and XhoI (New England Biolabs). The digested samples were resolved on a 0.8% TAE agarose gel. After electrophoresis, DNA was denatured in 0.05 N NaOH, neutralized with 0.5 M Tris HCl, transferred onto Hybond-N nylon membrane (Amersham Life Science) by capillary action through a 20 x Standard Saline Citrate (SSC) solution and UV cross-linked using a UV cross-linker (Stratagene).
Before hybridization, nylon membrane was blocked during 4-5 h at 65°C in solution (Church Gilberts) (0.34 mM Na2HPO4, 0.16 mM NaH2PO4.H2O, 7% (w/v) SDS, 1 mM EDTA (pH 8.0). Probe was radiolabeled with the random priming DNA labeling method (Stratagene), using [α-32P]-dCTP (Amersham Pharmacia Biotech). Hybridization was carried out overnight at 65°C, in blocking solution containing 4 μCi mL-1 of denatured probe. After hybridization, membrane was washed twice with 2 x SSC, 0.1% SDS at 65°C and autoradiographed.
RESULTS
Cloning and isolation of a new cyclin gene from L. mexicana: A putative L. major cyclin was identified by blasting the L. major genome sequence with the mammalian cyclin sequences. Consequently, the cyclin homologue was amplified by PCR from the L. mexicana genomic DNA and the cloned PCR was sequenced. The LmmCYCa coding region is 50.7% GC, which is consistent with the average GC content of 56% found in L. mexicana genes (Swindle and Tait, 1996).
High identity was observed with the mitotic cyclins from a variety of organisms. Moreover, a single Open Reading Frame (ORF) was obtained that encoding a predicted protein of 309 amino acid with an estimated molecular mass of 35.28 kDa. The analysis of the novel cyclin protein sequence revealed that it has an isoelectric point (Ip) of 6.28. The amino acid composition is rich in hydrophobic residues that constitute 40%. It is suggested that Leucine might contributed largely to this hydrophobicity of the cyclin protein, since it presents as 11%.
BLAST searches (Altschul et al., 1997) with the L. mexicana CYCa identified cyclin family members budding yeast, fission yeast, T. brucei, T. cruzi and humans. To define conserved features of the so-far identified mitotic cyclins, amino acid sequences from 18 cyclins were aligned and 16 conserved amino acids were identified. LmmCYCa cyclin contains all of the 16 conserved amino acids (not shown) and 56 of the 109 semi-conserved amino acids (shown by dots in Fig. 1. Obviously the Leishmania contains a cyclin box, but it is not conserved enough with other species to be identifiable.
The sequence comparison has shown that the L. mexicana cyclin as well as the putative L. major cyclin has some features not present in other cyclins. For example, there is seven randomly distributed dipeptides of the phenylalanine/leucine residues (FL) in the cyclin-box region (Two in TbCYC6 and one in Hs cyclin A). In addition, this FL dipeptide is not only restricted to the cyclin-box region of TbCYC6, but it also present in the C-terminus region (Fig. 1).
| Fig. 1: | Sequence alignment of L. mexicana CYCa with those of the putative
L. major CYC, T. brucei CYC6, human cyclins A and B. Multiple alignment
of cyclin family sequences, performed by AlignXTM program from the Vector
NTI package software. Shaded black and grey residues represent identities
among these cyclins. Asterisks represent the serine phosphorylation sites
in the human and Xenopus cyclin B1. Dashes represent gaps in sequences.
Hs, Homo sapiens; Tb, Trypanosoma brucei; Lmaj, Leishmania major. GenBankTM
accession numbers as follows: 1 NP001228; 2, XP043879. TbCYC6 (AJ496539.2).
The cyclin-box (CDK-binding domain) is underlined. The FL dipeptides are
shown in red |
Given that both phenylalanine and Leucine are hydrophobic residues, therefore, the cyclin-box is the most hydrophobic part of the cyclin; this may suggest a special requirement for the binding of the PSTAIRE-box of the CRK.
The predicted amino acid sequence shares the highest similarity with the putative L. major cyclin (LmajCYC) with 98.1% similarities and 96.1% identities through the entire coding sequence. A homology has also been shown to the T. brucei CYC6 (Hammerton, unpublished) with 23% identities through the entire coding sequence and 31.7% in the region 47-225 amino acids. The later region is the cyclin-box, the domain that thought to be involved in the CDK binding (Pines, 1995). However, from the sequence of the newly identified leishmanial cyclin, the central region (Under-lined sequence in Fig. 1 seems to contain all parts that are highly conserved in the mitotic cyclins subfamily. In addition, the LmmCYCa has significant similarities with human cyclins A (40.1%) and B1 (35%) in the cyclin-box region. Moreover, LmmCYCa shows 43 of the conserved residues of the mitotic cyclins, most of them in the cyclin-box region and few in the C-terminus regions (Black- highlighted residues in Fig. 1.
Comparison of the LmmCYCa with human cyclins A and B1 together with the T. brucei CYC6 (Fig. 2), have revealed that no N-terminus extension is present in the leishmanial cyclin; this feature is also shared by the T. brucei CYC6. The human cyclins A and B1 also have 50-amino acid insertion near the N-terminus this is absent from the LmmCYCa. Interestingly, all of the phosphorylation sites that shown to be important for the nuclear translocation of the human as well as Xenopus (Li et al., 1997, Winters et al., 2001) cyclin B are present in this insertion. This observation raises the possibility that leishmanial cyclins might have a different mechanism that regulates their spatial as well as temporal existence during the cell cycle.
T. brucei CYC6 is the only trypanosomal cyclin out of the eight identified to date that shows significant homology to LmmCYCa, however, some differences were observed. For example there are two insertions in the cyclin-box domain. One insertion is short, while the second is Q-rich 60-amino acids region that forms of repeats of glutamine residues. In addition, a C-terminus extension is present in T. brucei CYC6 and absent from the L. mexicana CYCa.
The evolutionary rates of the yeast, trypanosome, Leishmania and human cyclin family members are expected to be different. Therefore, when compared with other cyclin superfamily proteins, using the Neighbor Joining method (NJ), LmmCYCa diverges within the mitotic cyclin family class A and B, providing a reasonable degree of confidence for identifying LmmCYCa as a member of the mitotic cyclin family (Fig. 3).
| Fig. 2: | A schematic representation showing the domains characteristic
of the mitotic cyclins of trypanosomatids. The green colored region represents
the cyclin-box. The blue is the N-terminus region, while the red colored
is the C-terminus region. The insertions are illustrated in pink colour |
Interestingly, LmmCYCa apparently diverges before human B cyclins family branching into the B1 and B2 subfamilies.
Genomic organization of L. mexicana CYCa: To determine the copy number of LmmCYCa gene, southern blot analysis of L. mexicana genomic DNA was performed. A single hybridizing fragment under high stringency conditions was detected with digests using the enzymes BglII, EcoRI, HindIII, NcoI and XhoI (which do not cut inside the LmmCYCa coding region). A 6.0 kb BglII fragment, a 1.2 kb EcoRI fragment, a 1.8 kb HindIII, a 4.0 kb NcoI and a 2.7 kb XhoI fragment were hybridized to the labelled probe. However, two hybridized bands, a 3.5 and 1.2 kb fragments of SalI were shown, consistent with the presence of SalI site within the ORF of the LmmCYCa. No further hybridizing bands appeared after longer exposures with high stringently washed membranes. Taken together, this southern blot analysis revealed single bands, suggesting that LmmCYCa is present as a single copy gene in the L. mexicana genome (Fig. 4). This pattern resembles other characterized trypanosome cyclin genes, reinforcing the hypothesis of preferential organization of cyclins in trypanosomatid as single copy genes.
Over-expression and production of recombinant L. mexicana CYCa: Recombinant cyclin for preparative purification was obtained by large-scale cultivation of pGL630-containing cells. As shown in Fig. 5 all samples were resuspended in Laemmli buffer, boiled for 5 min and were then subjected to SDS-PAGE to estimate the size of the recombinant protein in the total cell protein extract. In the induced cells, protein band with a molecular mass of approx. 35 kDa was detected, which was consistent with the predicted size of his-tagged cyclin.
| Fig. 3: | Phylogenetic analysis of L. mexicana cyclin with T.
brucei, yeast and human cyclins using AlignXTM program of
the VectorNTI suite. Species abbreviations are as follows: Hs, Homo sapiens;
Sc, Saccharomyces cerevisiae; Tb, T. brucei; Tz, T. cruzi,
Lmm, Leishmania mexicana; Lmaj, Leishmania major; C.e, Caenorhabditis..
elegans. Accession numbers as follows: H.s cyclin A, (Accession No.
NP001228), H.s cyclin B1, (Accession No. XP043879), H.s cyclin B2, (Accession
No. AAD09309), H.s cyclin D1, (Accession No. XP029777), H.s cyclin D2, (Accession
No. X68452), H.s cyclin E1, (Accession No. BC035498), H.s cyclin H, (Accession
No. AF477979), C.e cyclin A1 (Accession No. P34638), C.e cyclin B, (Accession
No.AAA84394), Mouse cyclin B1, (Accession No. XM193189) CLB1 (Accession
No. M65069), CLN1 (Accession No. M33264), CLN2 (Accession No. M33265), CLN3
(Accession No. S14054), PHO80 (Accession No. X07464), TzCYC4, (Accession
No. AF237588) TzCYC5, (Accession No. AF237589) TzCYC6 (Accession No. AF237587),
TbCYC2 (Accession No. AJ242519), TbCYC3 (Accession No. AJ242520), TbCYC6
(Accession No. AJ496539.2) |
| Fig. 4: | Southern blot analysis of L. mexicana genomic DNA,
probed with the LmmCYCa fragment under high stringency conditions.
5 μg of L. mexicana (MNYC/BZ/62/M379) genomic DNA was digested
with BglII (lane 1), EcoRI (lane 2), HindIII (lane
3), NcoI (lane 4), SalI (lane 5) and XhoI (lane 6),
electrophoresed through a 0.8% agarose gel, blotted to a nylon membrane
and probed with the 936 bp NdeI/XhoI fragment of pGL612 |
| Fig. 5: | SDS-PAGE Localization of the recombinant L. mexicana
CYChis. Cells of the strain BL-21 (DE3) [pET-21a, pGL630], either non-induced
(Lane 1 and 4) or IPTG-induced (Lanes 2, 3, 5 and 6) were extracted and
subjected to SDS-PAGE (12.5%). Pellets preparations were loaded on lanes
1, 2 and 3, while supernatants preparations were loaded on lanes 4, 5 and
6. The molecular mass marker (bio-Rad) was indicated. The proteins in lanes
1-6 were stained with Coomassie Blue. The 35 kDa band of the L. mexicana histidine-tagged CYCa was detected on the supernatant preparation, upon
induction overnight at 19°C with both 1 mM (lanes 2 and 5) and 5 mM
(lanes 3 and 6) IPTG |
| Fig. 6: | Western blot analysis of expression of L. mexicana CYCahis in BL-21 (DE3) bacterial cells. Lysates of induced bacterial cells;
pellet preparations (lanes 2 and 3), supernatant preparations (lanes 5 and
6) and non-induced BL-21 cells (lanes 1 and 4) were resolved by SDS-PAGE
on a 12.5% gel: non-induced cells (lane 1), transformed cells induced at
19°C with 1 mM IPTG (lanes 2 and 5) and transformed cells induced at
19°C with 5 mM IPTG (lanes 3 and 6). All were probed with anti-(His)4
anti-bodies. ECL detection was carried out after incubation with horseradish
peroxidase-linked anti-rabbit IgG |
In order to examine the type and nature of the newly identified L. mexicana cyclin in E. coli cell extract, we used a commercial available anti-(his)4 anti-body that recognize the histidine-tagged protein with high specificity. No anti-body specific for CYC protein was available at that time. Previous studies showed that Trypanosomatid cyclins are present at low levels in any life cycle stage due to their turnover mechanism (Van Hellemond and Mottram, 2000; Van Hellemond et al., 2000). As shown in Fig. 6, the immunoblot using anti-(his)4 anti-bodies detected a single protein band. In addition, the size of the detected signals is consistent with the size of the recombinant LmmCYCahis that was analysed by the SDS-PAGE. However, this signal was not detected in the non-induced cells
Reconstitution of the recombinant L. mexicana CRK3his kinase complex through incubation with recombinant L. mexicana cyclin: CYC-CRK kinase assay: Evidence is accumulating that cyclin-dependent kinase 2 (Cdk2) and its homolog Cdc2, requires cyclin binding and phosphorylation of Thr-160 for activation in vivo (Gu et al., 1992; Brown et al., 1999). However, less information about the mechanism of the activation process is available. Particularly, no information is available for the natural cyclin partner(s) of the L. mexicana CRK3 and of its mechanism of activation. In this study, novel bacterial expression and purification systems for LmmCRK3 (Ali et al., 2003) and LmmCYCa were developed that allow mechanistic studies of the activation process to be performed in the absence of cell extracts.
To elucidate the possibility that L. mexicana CYCa cyclin is an activator of a leishmanial CRK, we decided to test its interaction with the CRK3 protein kinase. From the finding that the L. mexicana CRK3 is active at the G2/M phase of the Leishmania cell cycle (Hassan et al., 2001) together with the presence of significant homology of LmmCYCa to mitotic cyclins, we decided to test the interaction of these two proteins.
| Fig. 7: | Reconstitution of the recombinant L. mexicana CRK3his
kinase activity through incubation with recombinant L. mexicana cyclin:
Ni-NTA purified wild type LmmCRK3his (lane 2), mutated LmmCRK3hisE178 (lane
4), combinations of purified wild type LmmCRK3his and purified LmmCYCahis
(lane 3), combinations of purified mutated LmmCRK3hisE178 and partially
purified LmmCYCahis (lane 5), were assayed for kinase activity. Preparations
were used each in a separate kinase assay containing 5 μL (lanes 2-5)
of the recombinant protein, while 5 μL of Cdc2/cyclin B was used in
the control reaction (lane 1), together with 50 μg histone H1, 10 μCi
[γ-32P] ATP (Amersham). Radiolabeled proteins were analyzed
by SDS-PAGE (12.5% gel) (Top panel) and visualized by autoradiography (Bottom
panel). The gel was exposed to X-ray film at room temperature for 3 days
and quantified on phospho-imager (Typhoon) |
To assay for the kinase activity of the LmmCRK3his, pure and partially pure recombinants LmmCRK3his and LmmCYCahis, respectively were obtained by small-scale purification on Ni-NTA agarose beads and were incubated as shown in Fig. 7, only one signal was detected by p34Cdc2/cyclin B, but none of the other reactions.
DISCUSSION
A novel cyclin from L. mexicana was isolated its gene product is highly homologous to the mitotic cyclins, particularly, A-type ones, from several organisms, mainly in the cyclin-box region. In spite of the high homology to cyclin B-type from yeast and mammals, the absence of the destruction-box is clearly indicated. This observation suggests that LmmCYCa leishmanial cyclin may be directed for degradation by ubiquitination pathway different from that known for others cyclin A/B family. This is in consistence with the fact that trypanosomatids have mechanisms regulating their cell cycle differ from that of other organisms (Doerig et al., 2000; Hammarton et al., 2003; Banerjee et al., 2006)
L. mexicana cyclin is more similar in sequence to the mitotic cyclins A/B and CLBs than to the CLNs. Although of its similarity to mitotic cyclins, L. mexicana cyclin may act elsewhere in the cell cycle. Indeed, mere sequence similarity may not be enough to define the point of action of a cyclin in the cell cycle. This speculation is supported by previously finding (Hunter and Pines, 1991), when it showed that the S. pombe cig1+ gene resembles the mitotic B-type cyclins in sequence but may act in G1. This confirmed later (Martín-Castellanos et al., 2000) as fission yeast cells lacking cig1 and cig2 had a cell cycle distribution similar to that of wild-type cells, with a short G1 and a long G2. The findings of the sequence homology and the phylogenetic branching, suggest that LmmCYCa is probably an ancestral mitotic cyclin family member, related to the mammalian class A or B as shown in (Fig. 3).
The Southern blot analysis showed that LmmCYCa is present as a single coy gene in the L. mexicana genome. There are several lines of evidence that most of the so far identified trypanosome cyclins are present as a single copy. CYC2 and CYC3 are present as single-copy genes in the genome of the African Trypanosome (Van Hellemond et al., 2000), T. brucei CYC6 is also present as single copy (Hammarton et al., 2003), a similar genomic organization to that of the LmmCYCa, which can add more to the degree of similarity between these two cyclins in addition to their sequence homology. CYC6 has shown to have a direct role in the cell cycle of the African trypanosome (Hammarton et al., 2003), a similar role for the LmmCYCa could possibly be present in the parasite Leishmania in the light of structure/function relationship.
Cell cycle research in trypanosomatids has contributed to the successful approach of in vitro activation of CDKs. Transfection of mammalian COS-7 cells with TzCRK1 has proved that a trypanosome CRK protein can bind mammalian cyclins E, D3 and A (Gomez et al., 1998). Yeast two hybrid interaction studies and co-immune-precipitation experiments have demonstrated that CYC2 could bind to CRK3 in T. brucei (Van Hellemond et al., 2000). These findings have proved that similar complexes exist in the unicellular organism that could possibly be responsible for the regulation of the cell cycle in a similar way to multi-cellular organisms. In this study we tested the possibility that the newly identified CYCa is the activating partner for CRK3 i.e., an attempt to in vitro activate the bacterial produced CRK3his with the CYCahis was performed. Although, a very faint signal was observed but it is difficult to declare that the histone H1 was phosphorylated by an active CRK3his /CYCa complex. However, as the phosphorylation status might be responsible for this inactivity, a mutated CRK3his was used in parallel, to check for this possibility. All conditions gave negative results. The cyclin has not been purified optimally and the whole experimental conditions need to be optimized before conclusions are drawn on the possibility of reconstituting CRK3his kinase activity. Our very recent study (Gomes et al., 2010) has shown that CRK3: CYCA complex is active in the absence of T-loop phosphorylation. A recent study in L. donovani (Banerjee et al., 2006) has shown that Leishmania donovani cyclin 1 (LdCyc1) forms a complex with cell cycle kinase subunit CRK3 (LdCRK3) and is possibly involved in S-phase-related activities.
Taken together, the high sequence and structural homology to the mammalian mitotic cyclins, these results raise the possibility that the newly identified leishmanial cyclin has a role in regulating the cell cycle in the Leishmania and itself may be regulated by a mechanism different from that of the unicellular yeast, but more similar to multicellular organisms, possibly via the ubiquitination pathway.
In conclusion, initial experiments have shown no activity for the recombinant histidine-tagged L. mexicana CRK3. This may be due to insufficient trials, thus more optimization is needed in order to draw a conclusion on whether the novel mitotic leishmanial cyclin (LmmCYCa) is the regulatory partner for the G2/M phase kinase (LmmCRK3) or not.
Finally, all these results indicate that L. mexicana might have a class of cyclins that control the activity of the CRK proteins and that a complex mechanism would exist in order to regulate the kinases involved in the cell cycle and the differentiation processes of the parasite.
ACKNOWLEDGMENTS
This study was supported by funds from the UNDP/World Bank/WHO, Special Programme for Research and Training in Tropical Diseases (TDR) (grant ID. 971080).