Our current research interests are in the areas of post-transcriptional RNA modification in Archaea and Eukaryotes, and apoptosis and cell cycle progression in human cells. RNA modifications can regulate gene expression, which is essential for the control of cellular metabolism, growth and differentiation. Archaea often have eukaryote-like processes, but at a basic level. Therefore, they can serve as much simpler model systems to gain insights into complex eukaryotic events. Some archaeal proteins involved in RNA processing acquire additional roles in the biology of eukaryotes.

Previously, we mainly worked with archaeal organisms. We sequenced a nearly complete set of tRNAs from Haloferax volcanii (Halobacterium volcanii) and the 16S rRNA gene of this organism, characterized several modified nucleotides present in archaeal tRNAs, reported the presence of introns in several archaeal tRNA genes and characterized the splicing of these introns.

RNA-guided RNA modifications in Archaea

The tRNAs and rRNAs of all organisms contain many different types of nucleotide modifications. In eukaryotes, most pseudouridines (Ψ) and 2'-O-methylated residues of rRNA are produced by snoRNPs (small nucleolar ribonucleoproteins). The RNA components of the snoRNPs function as guides to select the sites of target rRNA modification and protein components catalyze the modification. Guide RNAs for the 2'-O-methylations and pseudouridylations are referred as box C/D and box H/ACA RNAs, respectively. Homologs of both box C/D and box H/ACA snoRNAs and their associated proteins have been reported in Archaea. These RNAs are known as sRNAs (small nucleolar RNA-like RNAs) in Archaea, because the Archaea being prokaryotes, lack nuclei and associated nucleoli. To characterize these modification reactions, we have developed in vitro systems using recombinant archaeal proteins and in vitro transcribed guide and target RNAs.

Using our in vitro box C/D RNA system, we showed that 2'-O-methylations at positions 34 and 39 of the pre-tRNATrp of H. volcanii occur through guide-target pairing in trans, although both guide and targets are present in cis in the pre-tRNA. Using the same system, we also showed that methylation of the two nucleotides, guided by the two sites of a dual-guide box C/D RNA present in the intron of the pre-tRNA occur sequentially. Using similar systems, we are also studying 2'-O-methylation reactions produced by single-guide box C/D RNAs of H. volcanii and differences between the modifications by the single- and dual-guide RNAs.

We have identified a double stem-loop box H/ACA RNA of H. volcanii. One of its loops produces one pseudouridine (at position 2605) of the 23S rRNA and the other loop produces two pseudouridines (at positions 1940 and 1942) in the 23S rRNA. Using our in vitro system, we are studying structure/function relationship of the two stem-loops of this H/ACA RNA. The functioning of the two stem-loops are significantly different.

We are also studying RNA-guided 2'-O-methylation and pseudouridylation by in vivo methods. For this, we have been using strains of H. volcanii where the gene for either a guide RNA or a protein of the guide RNP is deleted from the genome. Corresponding RNA modifications are absent in these strains. These modifications can be recovered by expressing a plasmid-borne copy of the gene deleted from the genome. We use these systems to characterize the structure/function relationships of the RNAs and proteins needed for the reactions inside the cells. Recently we identified several residues and motifs in an archaeal Cbf5, the pseudouridine synthase and in an archaeal Fib, the 2'-O-methyltransferase of the sRNPs, that are required for their in vivo activity.

Pseudouridines in the common (TΨC) arm of archaeal and human tRNAs

Pseudouridine (Ψ) is almost universally present at position 55 of tRNAs in all three domains (Archaea, Bacteria and Eukaryotes) of life. Bacterial TruB protein and its homolog Pus4 in yeast convert U at this position to Ψ. This reaction is catalyzed in Archaea by the Pus10 protein, which is not a member of TruB/Pus4 family of pseudouridine synthases. Pus10 homologs are found in most Archaea and eukaryotes, but not in Bacteria and yeast. This coincides with the presence of Ψ54 in most tRNAs of Archaea and a few tRNAs of higher eukaryotes and its absence in Bacteria and yeast. Instead, most tRNAs of Bacteria and eukaryotes contain ribothymidine at position 54. We have shown that archaeal Pus10 proteins can produce tRNA Ψ54 in addition to its tRNA Ψ55 synthase activity. Recently we also showed that a tRNA methyltransferase TrmY is responsible for converting this Ψ54 into hypermodified m1Ψ54. We identified certain domains/motifs of archaeal Pus10 that serve as crucial determinants for its Ψ54 activity.

The homology of human Pus10 with archaeal Pus10 suggests that the former may also have a tRNA Ψ54 synthase activity. Our current work suggests that human Pus10 does have tRNA pseudouridine synthase activity, but it is restricted to only some specific tRNAs. We are now determining how only certain human tRNAs are selected for Pus10-mediated pseudouridylation.

Role of human PUS10 in apoptosis and cell cycle progression

The human PUS10 gene has been suggested to be involved in TRAIL-induced apoptosis. Apoptosis or programmed cell death is a genetically determined mode of cell death that relies on energy-dependent cascade of caspase-mediated molecular events. It is important for natural cell turnover during development and aging, and in the elimination of virus-infected and damaged cells. It is of clinical relevance for cancer treatment by irradiation or drugs. There are two major pathways of apoptosis: extrinsic (death receptor) and intrinsic (mitochondrial). Binding of a ligand, such as TRAIL, to its specific death receptors on the cell surface initiates the extrinsic pathway, mainly by activation of caspase-8, whereas "stress signals'' or internal factors initiate the intrinsic pathway through activation of pro-apoptotic proteins, e.g., BAK and BAX. There is also "cross-talk" between the two pathways via activation of BID.

Recombinant TRAIL can induce apoptosis in several types of cancer cells, but leaves normal cells unaffected. Therefore, it is being developed as a drug for cancer treatment. However, certain cancer cells are either resistant to TRAIL or develop resistance during treatment. This is mainly due to mutations of various factors involved in apoptosis. We analyzed the role of PUS10 in TRAIL-induced apoptosis. PUS10 is mainly present in the nucleus. Early during apoptosis, PUS10 translocates to mitochondria via CRM1-mediated export with the concurrent release of cytochrome c and SMAC from mitochondria. Caspase-3 is required for PUS10 translocation, which reciprocally amplifies the activity of caspase-3 through the intrinsic/mitochondrial pathway. This suggests that in addition to cytoplasmic factors, nuclear factors also play a direct role in the major apoptosis pathways. However, p53 is not involved in TRAIL-induced PUS10 movement. The caspase-3-mediated movement of PUS10 and the release of mitochondrial contents enhancing caspase-3 activity creates a feedback amplification loop for caspase-3 action. Therefore, any defect in the movement or interactions of PUS10 would reduce the TRAIL-sensitivity of tumor cells.

During our apoptosis-related work, we created some PUS10 knockdown cell lines. Cell doubling times and levels of several proteins involved in cell cycle progression of these knockdown cells differ from the parent cells. We are now exploring the relationship between expression of PUS10 and cell cycle progression.

Selected publication ( Complete list )