As shown in several studies, RNA helicases are involved in a wide variety of processes, some of them being essential for survival, as demonstrated for the yeast putative RNA helicases, where their knockouts were lethal . These results are essential for the correct annotation of the Giardia genome, since many of the helicases identified in this study were automatically annotated either as helicases without indicating any further information and others just as hypothetical proteins (http://www.giardiadb.org).
The genome of a number of organisms contains a large number of putative helicases  and, as we found in this work, the relationship between the number of DEAD-box and DExH-box RNA helicases is conserved in Giardia as it is has been reported for other organisms (Table 1). Although Giardia is considered as an early-branching eukaryote and has a smaller and more compact genome , our findings regarding the type and number of RNA helicases in Giardia highlight the importance of these molecules in the biology of eukaryotic cells.
Since only a few DExD/H-box RNA helicases have been characterized biochemically, most of the reports assigning a putative function are based on the presence of the conserved and characteristic motifs that can define a putative RNA helicase and its family. Here we used the presence of those motifs for classification performing an in silico approach and then by manual identification of each motif. Then we confirmed and refined each motif at each position. Our results were in agreement with the phylogenetic tree obtained, because SF2 helicases were grouped specifically according to their sequence conservation as well as with the conservation of their motifs.
The particular finding within the Giardia Ski2 family regarding the internal duplication of the ORF GL50803_87022, having two helicases and Sec63 domains, probably indicates that the origin of this protein was by a fusion event of two ancestral prokaryotic genes, as proposed for the RNA helicases from Entamoeba histolytica EhDExH1 and EhDExH10  and other homologous proteins from phylogenetically distant species. Unfortunately, the significance of this duplication found only in two early-branching parasitic intestinal protozoa is still unknown.
The DEAD-box protein family is present in many organisms, being the major RNA family of helicases, which seem to be involved in many, if not all, steps of RNA metabolism . Although some DEAD-box helicases are closely related and have been described as paralogs , the comparison among amino acid sequences of all full-length sequences showed no paralogous DEAD-box helicases in Giardia because these proteins only share 14–29% identity and 24–43% similarity.
Regarding Giardia differentiation into cysts, it is known that encystation comprises the formation of a resistant cyst wall that allows the parasite to survive under hostile external environmental conditions and guarantees the transmission of the infection to susceptible hosts . Several encystation-specific genes have been identified and characterized during the last decade, and have shown to be up-regulated with similar kinetics during encystation, suggesting that their regulation is at the transcriptional level . Several reports also described putative transcription factors that regulate the expression of encystation-specific genes [71–74]. It was assumed that the encystation process is controlled at multiple levels (basic transcription, enhancement or de-repression) . Moreover, it was hypothesized that epigenetic chromatin modifications via histone acetylation/deacetylation may participate in modulation of stage differentiation in this parasite . In higher organisms, different RNA helicases have been described to interact with histone deacetylases (HDACs), such as the known transcriptional regulator DP103 (Ddx20, Gemin3), which was found to immunoprecipitate with histone deacetylases HDAC2 and HDAC5, suggesting a role in transcription repression through HDACs recruitment . In addition, the role of the RNA helicases p68 (Ddx5) and p72 (Ddx17) as transcription repressors when interacting with HDAC1 , HDAC2 and HDAC3 has been reported . Our findings regarding the levels of induction of the RNA helicase genes by qPCR were diverse, ranging from a smooth 2-4-fold induction in some DEAD-box genes to a high (20-31 times) relative expression in other genes. Two genes, DEAD-box GL50803_13791 and DEAH-box GL50803_13200, presented a marked induction of 554 and 228 times, respectively, under the encystation conditions. Notably, the up-regulation of the encystation-specific gene coding for CWP2 increased up to 2,187 times compared to its expression in trophozoites.
In Giardia, the RNAi machinery controlling antigenic variation has been found to involve a Dicer enzyme with unique characteristics when compared to Dicer enzymes from higher eukaryotes. Giardia Dicer lacks the DExD/H helicase domain as well as double-stranded RNA binding motifs present in other Dicer homologs. Because we are only starting to understand the different roles of RNA helicases in RNAi, there are still many unresolved questions. Since different RNA helicases might operate at different steps in the RNAi pathway or might play different roles, the presence of thirty two putative DExD/H-box helicases in the Giardia genome and their differential patterns of expression during antigenic variation support their importance for RNAi. It would be relevant to determine the role of particular Giardia RNA helicases for different subsets of miRNA or siRNAs. However, it is already clear that the presence of a RNA helicase activity (unwinding or as adaptor proteins) is necessary for the correct functioning at different steps of the RNAi pathway. As it was proposed in several reports, there are a number of potential roles for RNA helicases in RNAi . Our findings in the qPCR experiments during antigenic variation suggest that RNA helicases may participate in RNAi. This could be the case of the G. lamblia putative DEAD-box helicase GL50803_15048, which was found to present high homology with the DmBel helicase and also with the DEAD-box RNA helicases p68 and p72. Taking into account that some studies pointed out extensive overlapping and interplay among small RNA directed silencing machineries  and different RNA helicases operate either at different steps or playing different roles in the RNAi pathway, the involvement of this G. lamblia RNA helicase (GL50803_15048) in post-transcriptional gene silencing deserve further analysis.
Although we did not find a putative helicase in Giardia with high similarity to the HCD of higher eukaryotes Dicer enzyme, it has been proposed that Dicer helicase domain is required for siRNA, but not miRNA, processing . Point mutations within the helicase domain or Dicer lacking a functional HCD showed that pre-miRNA processing does not require helicase participation, but that it is necessary for long dsRNA (siRNA processing) .
In Giardia, we have demonstrated that purified RdRP generates high-molecular-weight VSP RNAs in vitro only when more than one VSP transcript is present in the reaction mixture  and proposed a mechanism where variations in either the general or local concentrations of different VSP transcripts may determine which transcript will circumvent the silencing system, as was suggested to occur in higher eukaryotes . In addition, it has been proposed by others groups the presence in Giardia of a miRNA biogenesis pathway reminiscent of the canonical miRNA biogenesis pathway found in higher organisms [25, 80], and they have identified conserved putative microRNA target site of several variant surface protein (VSP) mRNAs. Here Giardia Dicer apparently would assume the functions of both a Drosha and a Dicer, although no RNA-binding protein DAWDLE (DDL) homolog has yet been identified in this parasite. Furthermore Giardia Dicer must shuttle between the cytoplasm and the nucleus to process pri- and pre-miRNAs, although we determined its cellular localization by expressing a hemagglutinin-tagged version of the protein. Similar to that observed in other cells, Giardia Dicer localizes to the cytoplasm .
On one hand, the lack of the RNA helicase domain in Giardia Dicer is in agreement with the occurrence of a miRNA pathway. But, on the other hand, it was also proposed that a deletion or mutation of the helicase domain of human Dicer leads to a more active enzyme in vitro for cleavage of a perfectly matched 37-nt linear duplex RNA , allowing the enzyme to rapidly reinitiate cleavage on the long substrates. This last being the case for the generation of perfectly matched VSP dsRNAs by RdRP and subsequent degradation . It is also important to highlight that the lack of the helicase domain was proposed to increase the effectiveness of long hairpins for intracellular applications in which multiple siRNAs are desired, as could be the case for VSP mRNA degradation. Interestingly, gDicer without the RNA helicase domain can complement the absence of the entire Dicer in S. pombe. The lack of the RNA helicase domain in Giardia Dicer or, in other words, the inclusion of the RNA helicase domain in Dicer enzymes of higher eukaryotes, raises new questions about the function of this domain in Dicer activity and regulation.