39th Meeting of the Polish Biochemical Society Gdañsk 16–20 September 2003

SESSION 10

Structure and function of RNA

Organized by Z. Szweykowska-Kuliñska

234

Session 10. Structure and function of RNA

2003

407

Lecture

Diseases caused by mitochondrial RNA mutations Ewa Bartnik Zak³ad Genetyki, Uniwersytet Warszawski, ul. Pawiñskiego 5A, Warszawa

The human mitochondrial genome is a small circular DNA molecule encoding the synthesis of 13 proteins, 22 tRNAs and 2 rRNAs. Perturbations in mitochondrial function due to mutations in genes encoded in both mitochondrial and nuclear DNA can lead to a number of diseases, including blindness, deafness, neuromuscular disorders and diabetes. Changes in mi-

tochondrial function have been associated with aging, and somatic mutations in mitochondrial DNA can occur during carcinogenesis. Numerous mutations have been found in genes encoding tRNA and rRNA. I will present data on these mutations, including our own results on analyzing mutations involved in mitochondrial myopaties and mutations found in tumours.

408

Lecture

Ribozymes of the hepatitis delta virus: new findings on catalytic properties of ribonucleic acids Jerzy Ciesio³ka Instytut Chemii Bioorganicznej, Polska Akademia Nauk, ul. Noskowskiego 12/14, 61-704 Poznañ

Although the delta ribozymes have been studied for fifteen years the most important information concerning their structure and mechanism of catalysis were obtained very recently. The determination of the crystal structure of the genomic ribozyme is, undoubtedly, one of the most important achievements of ‘ribozymology’ in recent years. Details of the X-ray structure have greatly stimulated further studies on the folding of these ribozymes into functionally active molecules as well as on the mechanism of RNA catalysis. The ability of the delta ribozymes to carry out general acid-base catalysis by nucleotide side chains has been assumed in two proposed mechanisms of self-cleavage. Subsequently, a similar strategy has been suggested for the ribosome and peptide bond formation. General acid-base catalysis by nucleotide side chains increases

the catalytic repertoire of RNA and suggests that similar strategies could operate in other RNA-catalyzed reactions. Recently, we have used the in vitro selection approach to select for antigenomic ribozyme variants that are active in the presence of Mg(II). The following ribozyme regions were randomized: L3, J1/4 and J4/2, and after 10 cycles of selection-amplification, cloning and sequencing of the enriched ribozyme pool, several functional ribozymes were identified. Characterization of their catalytic activity, also in the presence of other than Mg(II) divalent metal ions, revealed that some nucleotide substitutions changed ionic preferences of selected variants. Interestingly, in some variants the cytosine residue proposed to play a crucial role in the cleavage mechanism of the wild type ribozyme was mutated.

409

Lecture

New mRNA 5’-cap analogues as useful tools in translation studies in eukaryota 1

1

1

1

1

Edward Dar¿ynkiewicz , Jacek Jemielity , Janusz Stêpiñski , Joanna ¯uberek , Anna NiedŸwiecka , Magdalena 1 1 1 2 Lewdorowicz , Dorota Haber , Ryszard Stolarski , Robert Rhoads 1 — Zak³ad Biofizyki IFD, Uniwersytet Warszawski, ul. ¯wirki i Wigury 93, 02-089 Warszawa, 2 — Department of Biochemistry and Molecular Biology, Louisiana State University Health Sciences Center, Shreveport, USA

Synthetic analogues of 5’-terminal cap of eukaryotic mRNA and snRNA have played an important role in elucidating such physiological processes as mRNA translation, pre-mRNA splicing, intracellular transport of mRNA and snRNAs, and mRNA turnover. Particularly

useful are RNAs capped with synthetic analogues, which are produced by in vitro transcription of a DNA template using a bacteriophage RNA polymerase in the presence of ribonucleoside triphosphates and a cap dinucleotide such as m7Gp3G. Unfortunately, due to the presence of

2003

39th Meeting of the Polish Biochemical Society

a 3’-OH on both the 7-methylguanosine and guanosine moieties, up to half of the mRNAs contain caps incorporated in the reverse orientation [Pasquinelli et al. (1995) RNA’ 1: 957–967]. Previously we designed and synthesized two “anti-reverse” cap analogues (ARCAs), m73’dGp3G and m27,3’OGp3G, that cannot be incorporated in the reverse orientation because of lack of free 3’-OH in 7-methylguanosine moiety [Stepinski et al. (2001) RNA, 7: 1486–1495]. In the present study we have synthesized several new cap analogues modified in the C2’ and C3’ positions of 7-methylguanosine and in the number of phosphate residues, m27,2’OGp3G, m72’dGp3G, m72’dGp4G, m27,2’OGp4G, m27,3’OGp4G,

235

m7Gp5G, and m27,3’OGp5G. These were analysed for conformation in solution, binding affinity to eIF4E, inhibition of in vitro translation, and the ability to stimulate cap-dependent translation in vitro when incorporated into mRNA. The results indicate that modifications at C2’, like those at C3’, prevent reverse incorporation, that tetra- and pentaphosphate cap analogues bind eIF4E and inhibit translation more strongly than their triphosphate counterparts, and that tetraphosphate (but not pentaphosphate) ARCAs promote cap-dependent translation more effectively than any cap analogue tested to date. Supported by the State Committee for Scientific Research: PBZ-KBN-059/T09/10.

410

Lecture

Splicing of U12-dependent introns in plants Artur Jarmo³owski Department of Gene Expression, Adam Mickiewicz University, ul. Miedzychodzka 5, 60-371 Poznañ

While several animal U12-dependent introns have been analyzed thoroughly, no significant study of U12 splicing in plants has been made. We scanned the Arabidopsis genome for U12 introns. Our computational analysis revealed that only 0.06% of all Arabidopsis introns have features characteristic of U12 intervening sequences. The U12 introns were distributed among 69 genes, all contain one U12-dependent intron except for two genes that have two U12 introns. There was no clear preference for the position of U12-type intervening sequences within the genes analyzed. Three U12-dependent introns were subjected to detailed splicing analysis in planta and in tobacco protoplasts. The introns were derived from three different Arabidopsis genes, which encode nuclear cap-binding protein (AtCBP20), glutathione synthetase (Atgsh2) and LUMINIDEPENDENS (AtLD). The introns studied were spliced efficiently in their normal context

within plant tissues. However, in tobacco protoplasts, excision from zein pre-mRNA was generally poor. There were notable differences in efficiency of the U12 introns splicing. While AtCBP20 and Atgsh2 U12 type introns were spliced only at 10% and 5% respectively, AtLD intron was spliced with 50% efficiency. Deletion of natural exonic fragments from the LD construct decreased splicing of the U12 intron to 9%, suggesting the presence of a putative exonic enhancer within the exons. We identified by deletion analysis a 40-nucleotide fragment in the upstream exon that stimulates splicing of the U12 intron more than 5 times. Splicing of U12 dependent introns in protoplasts was not enhanced by over expression of known plant SR proteins and U-rich RNA binding protein UBP1. Our experiments suggest that RBP45, a member of the plant hnRNP-like protein family, can be involved in regulation of U12-dependent splicing in plants.

411

Lecture

Pathways for nuclear mRNA degradation in yeast 1

2

2

2

Joanna Kufel , Laura Milligan , Claire Torchet , Cecile Bousquet-Antonelli , David Tollervey

2

1 — Zak³ad Genetyki, Uniwersytet Warszawski, ul. Pawinskiego 5a, 02-106 Warszawa, 2 — ICMB, University of Edinburgh, EH9 3JR Edinburgh, United Kingdom

Control of mRNA stability is a key step in the regulation of gene expression. During the degradation of cytoplasmic mRNA, deadenylation-dependent removal of the 5’-cap by the decapping complex Dcp1p/Dcp2p is activated by the cytoplasmic Lsm1-7p complex. Pathways that degrade mRNA precursors in the yeast nucleus have also been identified and different

pathways appear to utilize the same RNA degradation machinery. Unspliced pre-mRNAs are rapidly degraded predominantly 3’-5’ by the nuclear exosome, whereas 5’-3’ degradation by the exonuclease Rat1p plays a minor role. The nuclear exosome also functions in the degradation of pre-mRNAs that have failed to undergo correct 3’-end formation. However,

236

Session 10. Structure and function of RNA

in this pathway there is no evidence for significant 5’-3’ degradation. 5’-3’ degradation of nuclear pre-mRNAs also requires their decapping prior to exonuclease digestion. Cells lacking nuclear Lsm2-8 proteins or the decapping factor Dcp1p stabilize pre-mRNA degradation intermediates that would normally be subject to 5’-3’ degradation. These intermediates remain 5’-capped, indicating that the nuclear Lsm2-8p complex normally targets RNA substrates for decapping. Mature HS, GAL and MET mRNAs that are trapped in the nucleus due to a block in mRNA export are also stabilized and remain 5’-capped in lsm or dcp1 mutants. These nuclear-restricted mRNAs remain poly-

2003

adenylated until their degradation, indicating that nuclear mRNA degradation does not involve incremental deadenylation, which is a key feature of cytoplasmic turnover. Nuclear Lsm8p can be UV cross-linked to nuclear-restricted poly(A)+ RNA. This indicates that the Lsm2-8p complex interacts directly with mRNA substrates to activate decapping prior to 5’-3’ degradation. Nuclear-restricted HS and MET mRNAs are also degraded 3’-5’ by Rrp6p, an exonuclease that is specifically associated with the nuclear form of the exosome complex but not by the core exosome. These data show that degradation of different nuclear substrates has distinct features suggesting that they are targeted by related but distinct turnover pathways.

412

Lecture

New types of RNA within the cell Maciej Szymañski, Marek ¯ywicki, Miros³awa Barciszewska, Jan Barciszewski Instytut Chemii Bioorganicznej, Polska Akademia Nauk, ul. Noskowskiego 12/14, 61-704 Poznañ

Recent analyses of the human genome revealed that only about 1.5% is transcribed and translated into protein. Current data indicate that a considerable fraction of genomic DNA transcripts is responsible for regulation of gene expression on both transcriptional and posttranscriptional level. Interestingly both types of mechanisms involve non-protein coding RNAs. Noncoding RNAs (ncRNAs) are stable and functional RNA molecules without coding capacity. They can be broadly divided into two major group: housekeeping ncRNAs and regulatory ncRNAs. The RNAs belonging to the first group are necessary for primary cell functions. The second group can be divided into several subclasses depending on the mechanisms of action. Regulatory ncRNAs are able to affect expression of protein-coding genes on both transcriptional and posttranscriptional levels. These RNAs act by multiple mechanisms such as RNA-RNA base pairing, RNA-protein interactions and intrinsic RNA activity and regulate diverse cellular functions, including chro-

matin structure modification, RNA processing, mRNA stability and translation. Their involvement in some key genetic processes, like genomic imprinting or X chromosome inactivation in higher eukaryotes reveals their importance in cell metabolism. Noncoding RNAs have been also implicated in development and their defects can lead to various genetic disorders such as Prader-Willi syndrome, Beckwith-Wiedemann syndrome or cancerogenesis. The fact, that over 98% of human genome does not encode proteins stimulates efforts for ncRNAs gene finding. In contrast to protein coding genes, extremely variable structural properties of this type of molecules in addition to lack of open reading frames makes them very hard targets for computational based genome search. Till now there is lack of efficient method of ncRNA genes prediction. The important role of noncoding transcripts is however evident from the analysis of the mouse transcriptome, where approximately one third of total cDNA clones does not show protein-coding potential.

413

Oral Presentation

Analysis of putative human RNA methyltransferase that introduces methyl group on the first anticodon position in pre-tRNA-Leu(CAA). 1

2

3

Bartosz Brzezicha , Marcin Schmidt , Izabela Maka³owska , Zofia Szweykowska-Kuliñska

1

1 — Zak³ad Ekspresji Genów, Instytut Biologii Molekularnej i Biotechnologii, Uniwersytet im. Adama Mickiewicza, ul. Miêdzychodzka 5, 60-371 Poznañ, 2 — Katedra Biotechnologii i Mikrobiologii ¯ywnoœci, Akademia Rolnicza im. Augusta Cieszkowskiego, ul. Wojska Polskiego 48, 60-627 Poznañ, 3 — Computational Genomics Center, The Huck Institute for Life Sciences, Pennsylvania State University, 506 Wartik Lab, University Park, 16802 Pennsylvania, USA

Yeast m5C34-tRNA-Leu-methyltransferase Trm4 methylates first anticodon position in pre-tRNA-Leu(CAA) in intron-dependent manner. It is known that the absence of the enzyme is not lethal for the cell but causes disturbances in

mRNA decoding. Human genes coding for tRNA-Leu(CAA) contain 22–25 nt long introns. We have shown, that in the nuclear HeLa cell extract human pre-tRNA-Leu(CAA) is, as in the case of yeast, methylated in the intron-dependent manner.

2003

39th Meeting of the Polish Biochemical Society

Full sequence of the methyltransferase gene in yeast has been established. Searching for orthologous sequences in human GenBank databases revealed the presence of ESTs with high orthology to yeast enzyme sequence. We aligned two putative mRNAs that differ from each other by the presence of longer N-terminus in the protein after translation. Both have the same locus on fifth human chromosome: 5p15.32. Both cDNAs have been amplified using Human HeLa Marathon-Ready cDNA (Clontech) as a template. After sequencing we observed point mutations that were probably incorporated by Taq Polymerase during PCR reaction. In longer cDNA some mutations was responsible for changing encoded amino-acids. Mutations have been reversed by site-directed mutagenesis. Afterward, both cDNAs have been cloned into prokaryotic expression vector pET-28a(+) and the

237

shorter one has been also cloned into yeast expression vector p416 GALS. Overexpression of shorter protein in E. coli is efficient, contrary to the longer one. Enzymatic tests of shorter protein expressed in E. coli resulted in the lack of the methyltransferase activity. Expected activity has not been also detected in yeast extract after transformation of yeast line in which the endogenous methyltransferase gene has been deleted. Analogous experiments will be performed for testing activity of longer protein. We have established localization of both proteins in HeLa cells. The proteins have been expressed as a fusion to GFP. We observed localization using confocal microscopy. Both proteins localize in nucleus with clearly visible nucleolar localization. This observation is in accordance to previous data from which we know that posttranscriptional modifications of tRNA and rRNA take place mainly in the nucleolus.

414

Oral Presentation

Analysis of the posttranscriptional regulation of DNA polymerase beta expression in the rat cells Ewa Grzybowska, Anna Wilczyñska, El¿bieta Sarnowska, Janusz Siedlecki Zak³ad Biologii Molekularnej, Centrum Onkologii — Instytut, ul. Roentgena 5, 02-781 Warszawa

DNA polymerase beta plays a key role in base excision repair (BER). Its overexpression leads to a mutator phenotype, characteristic in certain cancers and its knockout causes apoptotic death of neurons during embryonic development in mice. Two DNA polymerase beta transcripts are expressed in rat cells and they differ (due to alternative polyadenylation) only in the length of their 3’ untranslated regions (3’UTRs). The transcript containing the shorter 3’UTR is expressed in most tissues, with significantly higher levels in testis, while the one with the longer 3’UTR is expressed mostly in brain and lungs. Both 3’UTRs contain specific motifs — putative protein binding sites — that could be responsible for DNA polymerase beta posttranscriptional regulation and tissue-specific expression of the two transcripts. Putative regulatory sequences of the DNA polymerase beta have been isolated and cloned in order to ana-

lyze their ability to bind specific proteins in the yeast three-hybrid system and EMSA. EMSA performed with protein fractions from rat cells, confirmed specific protein binding to the chosen mRNA regions. Three-hybrid analysis of one of these regions has revealed the interaction with HAX-1 protein, a potent anti-apoptotic factor. The effect of both 3’UTR sequences on transcript stability and translatability has also been analyzed using both, luciferase and beta-globin reporter systems. We have established that the presence of short 3’UTR causes an increase in luciferase expression, in contrast to the long 3’UTR, which exerts opposite effect. These results show that 3’UTRs play an important role in regulation of DNA polymerase beta expression and further identification of specifically binding proteins will help to elucidate the mechanisms behind it.

415

Oral Presentation

Efficiency of RNA polyadenylation by poly(A) polymerase in Escherichia coli depends on bacterial growth rate Jacek Jasiecki, Grzegorz Wêgrzyn Katedra Biologii Molekularnej, Uniwersytet Gdañski, ul. K³adki 24, 80-822 Gdañsk

RNA polyadenylation occurs not only in eukaryotes but also in bacteria. In prokaryotes, polyadenylated RNA molecules are usually degraded more efficiently

than non-modified transcripts. Here, we demonstrate that two transcripts, which were shown previously to be substrates for poly(A) polymerase I (PAP I), Escherichia

coli lpp mRNA and bacteriophage lambda oop RNA, are polyadenylated more efficiently in slowly growing cells than in rapidly growing bacteria. Intracellular levels of PAP I varied in inverse proportion to bacterial growth rate. Moreover, transcription from a promoter for the pcnB gene (encoding for PAP I) was shown to be more efficient under conditions of low bacterial growth rates.

We conclude that efficiency of RNA polyadenylation in E. coli is higher in slowly growing bacteria due to more efficient expression of the pcnB gene. This may allow regulation of stability of certain transcripts (those subjected to PAP I-dependent polyadenylation) in response to various growth conditions.

416

Poster

Synthesis of nucleosides tri- and tetraphosphates resistant to hydrolysis Jacek Jemielity, Marcin Ka³ek, Janusz Stêpiñski, Edward Dar¿ynkiewicz Zak³ad Biofizyki IFD, Uniwersytet Warszawski, ul. ¯wirki i Wigury 93, 02-089 Warszawa

The series of nucleosides tri- and tetraphosphates containing methylene group instead of terminal pyrophosphoric bond has been synthesized. Such modified compounds are resistant to either chemical and enzymatic hydrolysis [1, 2]. These derivatives may be applied in studies of decapping process and other biochemical researches. The method used to insert the methylene bridge to the phosphate chain consists in coupling an imidazole derivative NpnIm with bisphosphonate[3]. The elaborated synthetic method enables connecting the substrates with high yields and does not require a usage of any protective groups, what makes it very useful in synthesis of unstable compounds like for instance modified derivatives of 7-methylguanosisne. Preparation of analogue contain-

ing two CH2 groups is far more complicated. Currently, we are carrying out a synthesis of GpCH2pCH2pG. Crucial step in route to this compound is a condensation of protected guanosine with bis(O-benzyl phosphonomethyl)phosphonic acid benzyl ester, via Mitsunobu reaction[4]. References: 1. Thevenod F, Anderie I, Schulz I (1994) J Biol Chem, 269: 24410. 2. Tobin T, Akera T, Lee CY, Brody TM (1974) Biochem Biophys Acta, 345: 102. 3. Jemielity J et al, RNA — in press. 4. Klein E, Mons S, Valleix A, Mioskowski C, Lebeau L (2002) J Org Chem, 67146. Supported by the State Committee for Scientific Research: PBZ-KBN-059/T09/10.

417

Poster

HIV RT-mediated site-specific nonhomologous RNA recombination Anna Kurzynska-Kokorniak, Paulina Jackowiak, Marek Figlerowicz Instytut Chemii Bioorganicznej, Polska Akademia Nauk, ul. Noskowskiego 12/14, 61-704 Poznañ

Our preliminary studies conducted in vivo on brome mosaic virus (BMV) and in vitro on human immunodeficiency virus (HIV) showed that local hybridization between RNA molecules (local RNA-RNA duplex formation) can induce site-specific nonhomologous recombination if a double-stranded region is accompanied by short homologous sequences. A specific localization of the latter causes the local RNA-RNA duplex to exist in two different conformations. We hypothesized that the observed conformational dynamism of recombining molecules is responsible for the precise transfer of the viral polymerase from one template (called RNA donor) to the other (called RNA acceptor). To verify the above hypothesis we have created an in vitro system enabling effective studies on the mechanism of nonhomologous site-specific RNA recombina-

tion. Using this system we have tested the recombination activity of HIV-1 reverse transcriptase (HIV-1 RT ? p51/p66 heterodimer) and of the p66/p66 homodimer. To establish the role of RNA structure in the studied process, we have constructed several RNA donor and RNA acceptor molecules sharing short homologous sequences of various length (3, 5 and 10 nucleotides). We have also tried to determine the enzymes’ ability to switch from the 5’ end of the donor template to the 3’ end of the acceptor template. Moreover, we have tested HIV-1 RT and p66/p66 homodimer recombination activities in the system that offers optimal conditions for site-specific template switching by reverse transcriptase. Our results indicate that neither local hybridization between template molecules nor the presence of short

2003

39th Meeting of the Polish Biochemical Society

homologous sequences, can alone induce site-specific template switching by HIV-1 RT or p66/p66 homodimer. Only the simultaneous occurrence and the

239

proper localization of both motifs lead to recombinant formation.

418

Poster

Appearance of novel PSTVd variants upon coinfection with two non-infectious PSTVd clones Wojciech Podstolski, Anna Góra-Sochacka, W³odzimierz Zagórski-Ostoja Instytut Biochemii i Biofizyki, Polska Akademia Nauk, ul. Pawiñskiego 5A, 02-106 Warszawa

In the course of studies on the genetic stability of the infectious PSTVd variants I2 and I4, among a diverse population of progeny molecules, two non-infectious clones, I2-50 and I4-37, were found. The I2-50 and I4-37 variants are considered non-infectious on the basis of lack of visible symptoms on Lycopersicon esculentum ‘Rutgers’ test plants and negative dot blot hybridization results. The presence of such mutants among the population of viable variants remains a puzzling observation. To investigate the biological properties and possible interactions between these molecules, two types of experiments were carried out. Double inoculation experiments were performed in which monomeric full-length cDNA copies of the I2-50 and I4-37 variants were used. None of the twenty plants inoculated with either variant (I2-50 or I4-37) used alone as controls, or a mixture of the variants showed visible symptoms. However, dot blot hybridization analyses revealed that 7 out of the 30 doubly (I2-50 and I4-37) inoculated plants were PSTVd positive. The PSTVd progeny isolated from three selected plants was analyzed revealing the presence of 10 different PSTVd variants. Eight of them have not been previously described.

The same experiment was conducted, but using monomeric (+) RNA transcripts. Contrary to the results obtained with the cDNA copies, all inoculations yielded positive signals (in 3/15 plants for I2-50, 2/15 for I4-37 and 10/30 for the mixture of transcripts). This result shows that transcripts of apparently non-infectious cDNA can induce infection. Analysis of PSTVd clones isolated from three selected plants revealed that all the progeny of I2-50 and the mixture of both transcripts reversed to the infectious I2 parental type. In contrast in the progeny of I4-37, 8 sequences were detected, 6 of them were not previously described. It must be stressed that in neither of these two types of experiments, were I2-50 nor I4-37 sequences detected. Analysis and comparison of the sequences identified in both experiments with sequences retrieved from the NCBI database made it possible to identify mutations responsible for the apparent lack of infectivity and draw general conclusions on nucleotide preferences at certain positions in the PSTVd genome. Our experiments show that mixed infection with two apparently non-infectious PSTVd variants is followed by the appearance of progeny recombinants with restored regions allowing systemic infection.

419

Poster

AtCBP20 — the cap-binding protein from Arabidopsis thaliana has an nuclear localization signal Artur Rogowski, Maciej Kmieciak, Artur Jarmo³owski Department of Gene Expression, Adam Mickiewicz University, ul. Miedzychodzka 5, 60-371 Poznañ

All transcripts made by eukaryotic RNA polymerase II are co-transcriptionally modified at their 5’ ends by the addition of a cap structure. It has been shown that the cap structure is involved in several different aspects of RNA metabolism, such as pre-mRNA splicing, RNA export from the nucleus to the cytoplasm and polyadenylation. These functions are mediated by a nuclear complex of cap-binding proteins called CBC. It consists of two cap-binding proteins: CBP20 and

CBP80. We have recently identified and characterized a similar cap-binding complex from A. thaliana. AtCBP20 has a predicted molecular mass of 30 kDa. This protein contains a highly conserved canonical RNA-binding domain and a plant specific long C-terminal tail. The C-terminal part of AtCBP20 is rich in glycine and arginine residues. It has been shown that CBC from HeLa cells is transported into the nucleus due to nuclear localization signal (NLS) located in the

240

Session 10. Structure and function of RNA

2003

ments. Mutations in this motifs have revealed that they function in nuclear transport independently. Our results indicate that this NLS is located in the plant specific C-terminal tail of AtCBP20. This data suggest that in plants CBP20 is responsible for transport of the cap-binding complex from the cytoplasm to the nucleus.

N-terminal domain of CBP80. Human CBP20 does not contain any NLS. Using transient expression in transfected tobacco protoplasts and fluorescent microscopy, we have shown that CBP20 from A. thaliana has its own nuclear localization signal. We have recently shown that this signal consist of two separated ele-

420

Poster

Studies of SIAH1, potential mRNA target for NANOS1-PUMILIO2 protein complex in the human germ cells 1

2

3

1

3

Anna Spik , S³awomir Oczkowski , Jacek B³a¿ewicz , Maciej Kotecki , Piotr Formanowicz , Jadwiga 1 Jaruzelska 1 — Polska Akademia Nauk, Instytut Genetyki Cz³owieka, ul. Strzeszyñska 32, 60-479 Poznañ, 2 — Instytut Informatyki, Politechnika Poznañska, ul. Piotrowo 3A, Poznañ, 3 — Polska Akademia Nauk, Instytut Chemii Bioorganicznej, ul. Noskowskiego 12, Poznañ

Nanos and Pumilio in Drosophila. We found that one of these candidates is SIAH1. This gene is homologous to Sina, the fertility gene of Drosophila. Indeed, our experiments demonstrated that SIAH1 is expressed in human testes. The bend-shift approach has shown that NANOS1 and PUMILIO2 bind 3’UTR of the SIAH1 mRNA. In the next step, the conserved 3’UTR motifs of SIAH1 were mutated by introducing nucleotide substitutions using site directed mutagenesis. The binding efficiency of NANOS1-PUMILIO2 complex to 3’UTR mutated variants was substantially diminished. Although further tests are needed we believe that SIAH1 is strong candidate for the NANOS1-PUMILIO2 complex mRNA target in the human germ cells.

The Nanos-Pumilio complex is Drosophila translational repressor in morphogenesis, and in development of the germ cells. We have recently identified homologous protein complex (NANOS1-PUMILIO2) in the male human germ-line. The high conservation of the RNA-binding domains of these proteins suggests that they may function as translational regulators also in humans. Therefore, identification of target mRNAs seems crucial to understand the role of NANOS1 and PUMILIO2 in development of the human germ-line. Electronic searches of the GeneBank using software we have generated enabled us to retrieve candidate human mRNA targets of the NANOS1-PUMILIO2 complex. These candidate mRNAs contain 3’UTR motifs identical to motifs of the known mRNA targets of

421

Poster

Study of cleavage of the phosphodiester bond of mRNA 5’-cap analogs promoted by copper ion and copper complexes 1

2

1

2

1

Alicja Stachelska , Satu Mikkola , Zbigniew Wieczorek , Sanna Anderson , Monika Pietrzak , Satu Valakoski

2

1 — Department of Physics and Biophysics, University of Warmia and Mazury in Olsztyn, ul. Oczapowskiego 4, 10-719 Olsztyn, 2 — Department of Chemistry, University of Turku, FIN-20014 Turku, Finland

The aim of this studies was to investigate the hydrolysis of mRNA compound ApppA and 5’-cap analogs: 2+ m7GpppG and m7GppppG in the presence of Cu ions 2+ 2+ and of Cu TerPy and Cu BiPy complexes to find efficient catalysts for cleavage its phosphodiester bound. Comparison of rate constants of hydrolysis of ApppA, 2+ m7GpppG in the presence and absence of Cu ion shows its significant role in the cleavage of the phospodiester bond of studied compounds. The results show that hydroxoligand of metal ion catalyst is involved in this reaction. Comparison between the hydro-

lysis of ApppA, m7GpppG and m7GppppG in the pres2+ ence of Cu complexes show that the complex of 2+ 2+ Cu TerPy is more effective catalyst than Cu BiPy for hydrolysis of the cap-analogs at pH 8.0. The hydrolysis 2+ of ApppA in presence Cu TerPy proceeds about 2.7 2+ times faster than in the presence of Cu BiPy. However, m7GpppG hydrolysis in the presence of 2+ Cu TerPy is about 1.33 times faster than that in the 2+ presence of Cu BiPy. The hydrolysis of m7GppppG is about one order of magnitude faster in the presence of 2+ 2+ Cu TerPy than in the presence of Cu BiPy.

2003

39th Meeting of the Polish Biochemical Society

The results support suggestion that a hydroxoligand of a phosphate-bound metal ion acts as a nucleophile in 2+ the hydrolysis reaction [2]. The role of the Cu complex catalysts is significant, but is more complicated 2+ than for Cu ion promoted hydrolysis [1]. The acidic aquoligand may be needed for the phospodiester bond

241

cleavage, more than one process. In some cases sufficient catalysis can be induced be one metal ion, but sometimes two separate catalysts are required. References: 1. Anderson S, LDhde M, Mikkola S, Morris G, Stachelska A, Valakoski S, Williams NH (2002) Collect Symp, Ser 5. 2. Valakoski S, Heiskanen S, Anderson S, LDhde M, Mikkola S (2002) J Chem Soc, Perkin Trans, 2604–610.

422

Poster

Homologous RNA recombination as a factor stabilizing and destabilizing a viral genome 1

1

1

Anna Urbanowicz , Magdalena Alejska , Marek Figlerowicz , Jozef Bujarski

2

1 — Instytut Chemii Bioorganicznej, Polska Akademia Nauk, ul. Noskowskiego 12/14, 61-704 Poznañ, 2 — Plant Molecular Biology Center and Biology Department, Northern Illinois University, DeKalb, USA

Enormous genetic variability is one of the most unusual features of RNA viruses. There are two main sources of the genetic polymorphism of RNA genomes: error prone replication and RNA recombination. Error prone replication introduces into a viral genome a wide spectrum of point mutations. They can arise so frequently because RNA polymerases lack proofreading activity. The estimated average mutation –4 rate amounts to 10 per nucleotide per replication cycle. RNA recombination is responsible for more profound changes within the viral genome like sequence deletion, sequence insertion or the exchange of the entire coding domain. In the course of our studies we are trying to establish what the real role of homologous recombination in the virus life cycle is and this way to learn more about the mechanisms generating new viral strains or species. The object of our study is brome mosaic virus (BMV) — a model (+)RNA virus of plants. Its genome consists of three RNA segments, RNA 1, 2 and 3. To determine the frequency of homologous cross-

overs between the same genomic BMV molecules, we induced mixed infections in plants, using pairs of BMV genomic RNAs bearing marker mutations in coding and noncoding regions. We observed that marker mutations in 3’ and 5’ noncoding regions of BMV RNAs were very efficiently removed by imprecise replication. An analysis of the distribution of the stable marker mutations in the nascent virus revealed that homologous recombination events occur very often during virus genome replication. The frequency of recombination in the coding regions ranged from 10% to 25%. We also identified a recombination hot spot in the intergenic region of RNA3, where the frequency of crossovers reached 71%. That region might serve as a place for the exchanging of whole genes. In general our results suggest that the majority of nascent viruses are recombinants. This indicates that homologous RNA recombination is one of the main factors responsible for both the stabilization and destabilization of the viral genome.

423

Poster

A procedure to generate in vitro RNA transcripts with homogenous 3’ ends Agnieszka Wich³acz, Jerzy Ciesio³ka Instytut Chemii Bioorganicznej, Polska Akademia Nauk, ul. Noskowskiego 12/14, 61-704 Poznañ

It is known that during in vitro run–off transcription RNA polymerases often generate heterogeneous 3’ ends. In particular, T7 RNA polymerase attaches additional nucleotides to the nascent RNA transcript and at the same time may show premature termination close to its 3’ end. This results in decreased number of the full-length, correctly transcribed products. However, in many applications such as NMR, crystallization studies

and ligation reactions homogeneous transcripts are indispensable. One approach for the preparation of RNA transcripts with homogeneous 3’ terminus is the use of ribozymes which are attached to their 3’ ends. These constructs can cleave-off themselves acting in cis. On the contrary to hammerhead or hairpin ribozymes, delta ribozymes do not have particular requirements on the sequence

242

Session 10. Structure and function of RNA

upstream of the cleavage site and then can be applied to any RNA. In this communication we have shown that the antigenomic delta ribozyme could also be applied in trans to correctly process the 3’ ends of RNA transcripts. Two different transcripts were extended with the sequence of only 7 nucleotides complementary to

2003

the ribozyme recognition site. Following the cleavage reaction with the ribozyme acting as ‘molecular scissors’ homogeneity of transcripts was proven by TLC analysis of their 3’ terminal nucleotides. The proposed procedure is useful for virtually any RNA molecule that we desire to be homogeneous at its 3’ end.