The aim of this work was to identify various genes involved in ribosomal biogenesis pathway of malaria parasite. Various proteins and RNA molecules like RNA helicases, U3 snoRNA, RNA MRP have been identified and reported earlier in Pf [36
]. The structures and sequences of 5' ETS of the different stage specific pre-rRNA are also known from a previous study, but no information is available for 3' ETS[18
]. The present study highlights the differences in the pathway, where gene for a particular function (either protein or RNA) is missing in the Pf genome. Absence of any factor would indicate deviation from the known pathway and would call for identification of alternative mechanisms present in malaria parasites.
In spite of the differences in the gene structure and transcription pattern of rRNA genes in Plasmodium, the protein components of ribosome do not show much difference. We could identify homologues for each of the proteins present in large and small subunit of the ribosome, except RPL29. The sequence identity with yeast and human homologues for most of the ribosomal proteins were higher than 60%, which indicate conservation of protein structure and function.
Most of the trans-acting factors responsible for ribosomal biogenesis are present in the parasite genome. Proteins present in the various complexes like box C/D snoRNP, box H/ACA snoRNP, U3 snoRNP and exosome have their homologues in the parasite genome. We were unable to find a homologue for the endoribonuclease Rnt1p, responsible for the cleavage in 3' ETS of pre-rRNA. This step of ribosomal biogenesis in malaria parasite seems to differ from yeast and the protein responsible for this function in Pf may be a drug target. Further study is required to identify the sequences of 3' ETS of pre-rRNA and various proteins involved in the alternative pathway for its processing.
Chakrabarti et al
have reported the existence of 2'-O-methylation and snoRNAs in malaria parasites[37
]. They have identified non coding RNAs using GC content, RNA folding potential and sequence conservations. We have predicted box C/D snoRNA genes of P. falciparum
using a different computational method based on hidden markov model (HMM), which has enabled us to identify even those snoRNA which are present in only one species. Identification of species specific snoRNA helps to understand the mechanism of snoRNA evolution. Out of the 18 snoRNA genes in this study, 16 were reported by Chakrabarti et al
. Additionally we have identified some new snoRNA genes not listed in the previous report. These are: PFS16, which guides methylation on a site identical to that for PFS15; and PFS14, whose size differs amongst species due to a small AA rich insertion in the gene. Additionally, we also report the presence of two identical copies of PKS11 in P. knowlesi
genome. Identification of gene duplication in the genome shed new light on the mechanism of evolution of snoRNAs in Plasmodium
We have confirmed the expression of snoRNA genes using northern hybridization and reverse transcriptase PCR assays. We have also identified the orthologs of these genes in other Plasmodium
species. A comparative study of these snoRNAs has revealed features unique to malaria parasite. Most of the snoRNAs in vertebrates are localized to introns of protein coding genes [15
]. In yeast, most snoRNAs are transcribed from their independent promoters [39
], barring a few intron-encoded genes. In the case of Plasmodium
, we found a mixture of localization patterns observed in yeast, vertebrates and plants (Fig ). Table summarizes the comparative study of these snoRNAs, all of which have similar target sites in Pf, yeast, humans and Arabidopsis. All the 18 human snoRNAs are located in introns as compared to only nine in case of Pf and four for yeast. Eight of the snoRNAs in yeast are monocistronic and five exist as polycistronic genes. In the case of Arabidopsis, all these are present in cluster and are transcribed as polycistronic genes, whereas, in Pf two gene clusters were observed, one in the case of PFS11 and PFS12 and another in the case of PFS15 and PFS16 (Fig ). In both vertebrates and yeast, one intron harbours single snoRNA gene but in plants, there are reports of clustered snoRNA genes present in a single intron [35
]. Plasmodium falciparum
has a cluster of two snoRNAs, viz PFS15 and PFS16, which are present in the same intron of PF 14_0027 (Fig ).
Figure 6 Genomic organization of snoRNA. Graphical representations of localization pattern of the identified 18 snoRNAs in P. falciparum and the snoRNAs having similar methylation sites in human, Arabidopsis and yeast. In Arabidopsis, localization pattern of 16 (more ...)
Localization of homologous snoRNA genes
Trypanosomatids are unicellular parasitic protozoa which are the causative agents of several infamous parasitic diseases, such as African trypanosomiasis, caused by Trypanosoma brucei
; Chagas' disease, caused by Trypanosoma cruzi
; and leishmaniasis, caused by Leishmania
species and have a dual host like malaria parasites. Most snoRNAs in these organisms are clustered in reiterated repeats that carry a mixed population of C/D and H/ACA-like RNAs[41
]. Prediction of the modifications guided by these RNAs and using partial mapping data, 84 2'-O
-methyls (Nms) and 32 pseudouridines were identified on rRNAs, suggesting a high occurrence of Nms as compared to pseudouridines on rRNA. Occurrence of a mixed population of box C/D and H/ACA snoRNAs and a higher number of Nms than psuedouridylation is in line with findings in Plasmodium
as reported by Chakrabarti et al
The functional and evolutionary significance of UTRs of eukaryotic transcripts remains unclear. There are reports of introns in the UTRs of RNA transcripts but their functional significance is unknown[42
]. In this work, we show that snoRNA PFS9 is contained within the 3' UTR of the RNA transcript of ribosomal protein L7a using RT-PCR. Identification of polyA site of the mRNA using 3' RACE would be another method to prove this. The homolog of PFS9, PVS9 in P. vivax
is also located in the 3' UTR of the ribosomal protein L7a gene. Localization of snoRNAs in 3' UTR is a novel organization not reported in any other organism till date.
PFS10 is located in an intron of a ribosomal protein PF14_0230 but its homologues in P. chabaudi
, P. berghei
and P. yoelii
are present downstream of the last exon. The three snoRNAs PFS5, PFS6 and PFS7 are present in introns of Enp1 gene of Pf but are localized differently in different species (Fig and ). From these observations, it seems likely that human and malaria parasite snoRNAs prefer to localize in introns rather than in intergenic regions. Harbouring snoRNA genes in introns and UTRs of a constitutively expressed gene may be a more efficient and coordinated method for transcription. In case of P. vivax
, the mRNA of Enp1 gene does not code for the full protein. It is probable that the mRNA harbouring snoRNA genes has lost its translational capability, which is reported in many cases [43
An interesting observation shows that PFS1 and PFS2 are evolutionarily linked and may have evolved from a common ancestor. Both PFS1 and PFS2 have an identical antisense region. PFS1 has two regions complementary to rRNA, one for 18S Tm1370 and another for 28S Gm3308, whereas PFS2 has for 18S A1129 and 28S Am3307. Regions for 28S – 3307 and 3308 are similar, except for deletion of a cytidine from PFS1 antisense region, leading to change in target site.
Studies on mammalian snoRNA genes have revealed that they are a new class of mobile genetic elements [26
]. It has been proposed that retroposition followed by genetic drift is a mechanism that can increase snoRNA diversity during vertebrate evolution to eventually acquire new RNA-modification functions. In this study, our results imply that this mechanism may hold true in Plasmodium
also. In the first case, two identical copies of a homologue of PFS11 are present in the P. knowlesi
genome, whereas only a single copy is present in other species. Sequence alignment of these two loci with flanking regions suggests a 'copy-paste mechanism', as observed in case of retrotransposons (Fig ). We propose that snoRNA duplication may be due to their behaviour as snoRTs because of two reasons 1) both the copies are 100% identical with few bases of overhangs and one of them had a poly A at the 5' end of the sequence (present on PKN.002755), whereas the other copy doesn't have it; presence of polyA tail is an important feature of transposons that traverse using RNA intermediates 2) Only P. knowlesi
has two copies whereas other species of Plasmodium
have only one copy. The parental copy may be the snoRNA present on PKN.000135 which may have been lost during evolution in other species. Since this is the only identified cases of snoRT in Plasmodium
, our hypothesis cannot be conclusively verified. Similar studies on H/ACA snoRNAs may help to draw a final conclusion. In another case, PFS15 is conserved in Plasmodium
species but its paralog – PFS16, is absent from another species, indicative of gene duplication.