3-Deazaadenosine analogues of p5'A2'p5'A2'p5'A: synthesis, stereochemistry, and the roles of adenine ring nitrogen-3 in the interaction with RNase L

Sequence-specific 3-deazaadenosine (c(3)A)-substituted analogues of trimeric 2',5'-oligoadenylate, p5'A2'p5'A2'p5'A, were synthesized and evaluated for their ability to activate human RNase L (EC 3.1.2.6) aiming at the elucidation of the nitrogen-3 role in this biochemical process. Substitution of either 5'-terminal or 2'-terminal adenosine with c(3)A afforded the respective analogues p5'(c(3)A)2'p5'A2'p5'A and p5'A2'p5'A2'p5'(c(3)A) that were as effective as the natural tetramer itself as activators of RNase L (EC(50)=1nM). In contrast, p5'A2'p5'(c(3)A)2'p5'A showed diminished RNase L activation ability (EC(50)=10nM). The extensive conformational analysis of the c(3)A-substituted core trimers versus the parent natural core trimer by the (1)H and (13)C NMR, and CD spectroscopy displayed close stereochemical similarity between the natural core trimer and (c(3)A)2'p5'A2'p5'A and A2'p5'A2'p5'(c(3)A) analogues, thereby strong evidences for the syn base orientation about the glycosyl bond of the c(3)A residue of the latter were found. On the contrary, an analogue A2'p5'(c(3)A)2'p5'A displayed rather essential deviations from the spatial arrangement of the parent natural core trimer.     

Kinetics of the processing of the precursor to 4.5 S RNA, a naturally occurring substrate for RNase P from Escherichia coli.

A study was made of the cleavage by M1 RNA and RNase P of a non-tRNA precursor that can serve as a substrate for RNase P from Escherichia coli, namely, the precursor to 4.5 S RNA (p4.5S). The overall efficiency of cleavage of p4.5S by RNase P is similar to that of wild-type tRNA precursors. However, unlike the reaction with wild-type tRNA precursors, the reaction catalyzed by the holoenzyme with p4.5S as substrate has a much lower Km value than that catalyzed by M1 RNA with the same substrate, indicating that the protein subunit plays a crucial role in the recognition of p4.5S. A model hairpin substrate, based on the sequence of p4.5S, is cleaved with greater efficiency than the parent molecule. The 3'-terminal CCC sequence of p4.5 S may be as important for cleavage of this substrate as the 3'-terminal CCA sequence is for cleavage of tRNA precursors.     

Structural changes of human RNase L upon homodimerization investigated by Raman spectroscopy

RNase L, a key enzyme in the host defense system, is activated by the binding of 2'-5'-linked oligoadenylates (2-5A) to the N-terminal ankyrin repeat domain, which causes the inactive monomer to form a catalytically active homodimer. We focused on the structural changes of human RNase L as a result of interactions with four different activators: natural 2-5 pA(4) and three tetramers with 3'-end AMP units replaced with ribo-, arabino- and xylo-configured phosphonate analogs of AMP (pA(3)X). The extent of the RNase L dimerization and its cleavage activity upon binding of all these activators were similar. A drop-coating deposition Raman (DCDR) spectroscopy possessed uniform spectral changes upon binding of all of the tetramers, which verified the same binding mechanism. The estimated secondary structural composition of monomeric RNase L is 44% α-helix, 28% β-sheet, 17% β-turns and 11% of unordered structures, whereas dimerization causes a slight decrease in α-helix and increase in β-sheet (ca. 2%) content. The dimerization affects at least three Tyr, five Phe and two Trp residues. The α-β structural switch may fix domain positions in the hinge region (residues ca. 336-363) during homodimer formation.     

Mechanistic studies on depurination and apurinic site chain breakage in oligodeoxyribonucleotides.

Products formed from defined oligodeoxyribonucleotide tetramers (oligonucleotides) by depurination at pH 5.0 and 90 degrees C followed by chain breakage at the resulting apurinic sites (AP sites) were assigned by reversed phase HPLC. Through kinetic analysis, rate constants of depurination and subsequent chain breakage reactions were measured. Depurination of the oligonucleotides with purine bases locating at the terminal positions was several times faster than those with purines at the internal ones. The pKa values for the N7 of the G residues and the activation energies of the depurination were essentially independent of the position of the bases. The frequency factor was found to be responsible for the observed difference of the depurination rates. In contrast, the chain breakage by beta-elimination was several times faster for the AP sites formed at the internal positions than those at the 5'-terminal positions. It is suggested that an electron withdrawing phosphate group attached to the 5'-side of an AP site facilitates the chain cleavage.     

Purine 8-bromination modulates the ribonuclease L binding and activation abilities of 2',5'-oligoadenylates. Possible influence of glycosyl torsion angle

Analogs of the 2',5'-linked adenylate trimer diphosphate (pp5'A2'p5'A2'p5'A or 2-5A) containing 8-bromoadenosine in the first, second, third, first and third, or second and third nucleotide positions (from the 5' terminus) were synthesized and found to vary dramatically in their ability to bind to and activate the RNase L of mouse L cells. Whenever the 8-bromoadenosine residue was substituted for adenosine in the first or 5'-terminal residue, there resulted a marked decrease in ability to bind to the 2-5A-dependent endonuclease. A similar result was obtained when the second adenosine nucleotide was replaced by 8-bromoadenosine. To the contrary, all analogs that bore an 8-bromoadenosine (br8A) in the third or 2'-terminal position were bound about as well as parent 2-5A to RNase L. Additionally, the 5'-diphosphate pp5'A2'p5'A2'p5' (br8A) was 10 times more effective than 2-5A as an inhibitor of translation. An increase in stability could not explain this significantly enhanced ability since the 2'-terminally brominated analog showed a similar half-life to 2-5A itself. Finally of particular interest was the analog monophosphate p5'A2'p5'(br8A)2'p5'(br8A) which possessed nearly 10% of the translational inhibitory activity of 2-5A triphosphate itself. These results suggest that changes in the base-sugar torsion angles of 2-5A may modulate both binding to and activation of mouse L cell RNase L.     

Phosphorothioate analogues of 2',5'-oligoadenylate. Enzymatically synthesized 2',5'-phosphorothioate dimer and trimer: unequivocal structural assignment and activation of 2',5'-oligoadenylate-dependent endoribonuclease

In continued studies to elucidate the requirements for binding to and activation of the 2',5'-oligoadenylate-dependent endoribonuclease (RNase L), chirality has been introduced into the 2',5'-oligoadenylate (2-5A, p3An) molecule to give the Rp configuration in the 2',5'-internucleotide backbone and the Sp configuration in the alpha-phosphorus of the pyrophosphoryl moiety of the 5'-terminus. This was accomplished by the enzymatic conversion of (Sp)-ATP alpha S to the 2',5'-phosphorothioate dimer and trimer by the 2-5A synthetase from lysed rabbit reticulocytes. The most striking finding reported here is the ability of the 2',5'-phosphorothioate dimer 5'-triphosphate (i.e., p3A2 alpha S) to bind to and activate RNase L. p3A2 alpha S displaces the p3A4[32P]pCp probe from RNase L with an IC50 of 5 X 10(-7) M, compared to an IC50 of 5 X 10(-9) M for authentic p3A3. Further, p3A2 alpha S activates RNase L to hydrolyze poly(U)-3'-[32P]pCp (20% at 2 X 10(-7) M), whereas authentic p3A2 is unable to activate the enzyme. Similarly, the enzymatically synthesized p3A2 alpha S at 10(-6) M activated RNase L to degrade 18S and 28S rRNA, whereas authentic p3A2 was devoid of activity. p3A3 alpha S was as active as authentic p3A3 in the core--cellulose and rRNA cleavage assays. The absolute structural and configurational assignment of the enzymatically synthesized p3A2 alpha S and p3A3 alpha S was accomplished by high-performance liquid chromatography, charge separation, enzymatic hydrolyses, and comparison to fully characterized chemically synthesized (Rp)- and (Sp)-2', 5'-phosphorothioate dimer and trimer cores.(ABSTRACT TRUNCATED AT 250 WORDS)     

Probing the activation site of ribonuclease L with new N6-substituted 2',5'-adenylate trimers

2-5A trimer [5'-monophosphoryladenylyl(2'-5')adenylyl(2'-5')adenosine] activates RNase L. While the 5'-terminal and 2'-terminal adenosine N(6)-amino groups play a key role in binding to and activation of RNase L, the exocyclic amino function of the second adenylate (from the 5'-terminus) plays a relatively minor role in 2-5A's biological activity. To probe the available space proximal to the amino function of the central adenylate of 2-5A trimer during binding to RNase L, a variety of substituents were placed at that position. To accomplish this, the convertible building block 5'-O-dimethoxytrityl-3'-O-(tert-butyldimethylsilyl)-6-(2,4-dinitrophenyl)thioinosine 2'-(2-cyanoethylN,N-diisopropylphosphoramidite) was prepared as a synthon to introduce 6-(2,4-dinitrophenyl)thioinosine into the middle position of the 2-5A trimer during automated synthesis. Post-synthetic treatment with aqueous amines transformed the (2,4-dinitrophenyl)thioinosine into N(6)-substituted adenosines. Assays of these modified trimers for their ability to bind and activate RNase L showed that activation activity could be retained, albeit with some sacrifice compared to unmodified p5'A2'p5'A2'p5'A. Thus, the spatial domain about this N(6)-amino function could be available for modifications to enhance the biological potency of 2-5A analogues and to ligate 2-5A to targeting vehicles such as antisense molecules.     

Alternative function of a protein kinase homology domain in 2', 5'-oligoadenylate dependent RNase L.

RNase L is the 2',5'-oligoadenylate (2-5A)-dependent endoribonuclease that functions in interferon action and apoptosis. One of the intriguing, albeit unexplained, features of RNase L is its significant homology to protein kinases. Despite the homology, however, no protein kinase activity was detected during activation and RNA cleavage reactions with human RNase L. Similarly, the kinase plus ribonuclease domains of RNase L produced no detectable protein kinase activity in contrast to the phosphorylation obtained with homologous domains of the related kinase and endoribonuclease, yeast IRE1p. In addition, neither ATP nor pA(2'p5'A)3was hydrolyzed by RNase L. To further investigate the function of the kinase homology in RNase L, the conserved lysine at residue 392 in protein kinase-like domain II was replaced with an arginine residue. The resulting mutant, RNase LK392R, showed >100-fold decreases in 2-5A-dependent ribonuclease activity without reducing 2-5A- or RNA-binding activities. The greatly reduced activity of RNase LK392Rwas correlated to a defect in the ability of RNase L to dimerize. These results demonstrate a critical role for lysine 392 in the activation and dimerization of RNase L, thus suggesting that these two activities are intimately linked.     

RNase L, a crucial mediator of innate immunity and other cell functions

The 2-5A/RNase L pathway is one of the first cellular defences against viruses. RNase L is an unusual endoribonuclease which activity is strictly regulated by its binding to a small oligonucleotide, 2-5A. 2-5A itself is very unusual, consisting of a series of 5'- triphosphorylated oligoadenylates with 2'-5' bonds. But RNase L activity is not limited to viral RNA cleavage. RNase L plays a central role in innate immunity, apoptosis, cell growth and differentiation by regulating cellular RNA stability and expression. Default in its activity leads to increased susceptibility to virus infections and to tumor development. RNase L gene has been identified as HPC1 (Hereditary Prostate Cancer 1) gene. Study of RNase L variant R462Q in etiology of prostate cancer has led to the identification of the novel human retrovirus closely related to xenotropic murine leukemia viruses (MuLVs) and named XMRV.     

Only one 3'-hydroxyl group of ppp5' A2'p5'A2'p5' A (2-5A) is required for activation of the 2-5A-dependent endonuclease

To investigate the relative importance of each of the ribose 3'-hydroxyl groups of 2-5A (ppp5' A2'p5'A2'-p5' A) in determining binding to and activation of the 2-5A-dependent endonuclease (RNase L), the 3'-hydroxyl functionality of each adenosine moiety of 2-5A trimer triphosphate was sequentially replaced by hydrogen. The analog in which the 5'-terminal adenosine was replaced by 3'-deoxyadenosine (viz. ppp5'(3'dA)-2'p5' A2'p5' A) was bound to RNase L as well as 2-5A itself and was only 3 times less potent than 2-5A as an activator of RNase L. On the other hand, when the second adenosine unit was replaced by 3'-deoxyadenosine (viz. ppp5' A2'p5'(3'dA)2'p5' A), binding to RNase L was decreased by a factor of eight relative to 2-5A trimer and, even more dramatically, there was a 500-1000-fold drop in ability to activate the 2-5A-dependent endonuclease. Finally, when the 3'-hydroxyl substituent was converted to hydrogen in the 2'-terminal residue of 2-5A, a significant increase in both binding and activation ability occurred. We conclude that only the 3'-hydroxyl group of the second (from the terminus) nucleotide residue of 2-5A is needed for effective activation of RNase L.     

Evidence that discontinuous DNA replication in Escherichia coli is primed by approximately 10 to 12 residues of RNA starting with a purine

Intact primer RNA for discontinuous DNA replication of Escherichia coli has been detected by specific labeling in vitro of its 5'-terminal tri- (or di-) phosphate group with vaccinia guanylyltransferase and [alpha-32P]GTP. A mutant defective either in RNase H or in both RNase H and DNA polymerase I accumulated about 10 or 30 times more intact primer RNA, respectively, than wild-type cells. The primers started with purine in an A to G ratio of 5 and the most abundant 5'-terminal dinucleotide sequence was (p)ppA-Pu. The chain length of the intact primer RNA was approximately 10 to 12 nucleotide residues. The structural properties of the E. coli primer RNa resemble those of the eukaryotic primer RNA.     

2- and 8-azido photoaffinity probes. 1. Enzymatic synthesis, characterization, and biological properties of 2- and 8-azido photoprobes of 2-5A and photolabeling of 2-5A binding proteins

The 2- and 8-azido trimer 5'-triphosphate photoprobes of 2-5A have been enzymatically synthesized from [gamma-32P]2-azidoATP and [alpha-32P]8-azidoATP by 2-5A synthetase from rabbit reticulocyte lysates. Identification and structural determination of the 2- and 8-azido adenylate trimer 5'-triphosphates were accomplished by enzymatic hydrolyses with T2 RNase, snake venom phosphodiesterase, and bacterial alkaline phosphatase. Hydrolysis products were identified by HPLC and PEI-cellulose TLC analyses. The 8-azido photoprobe of 2-5A displaces p3A4[32P]pCp from RNase L with affinity equivalent to p3A3 (IC50 = 2 X 10(-9) M in radiobinding assays). The 8-azido photoprobe also activates RNase L to hydrolyze poly(U) [32P]pCp 50% at 7 X 10(-9) M in core-cellulose assays. The 2- and 8-azido photoprobes and authentic p3A3 activate RNase L to cleave 28S and 18S rRNA to specific cleavage products at 10(-9) M in rRNA cleavage assays. The nucleotide binding site(s) of RNase L and/or other 2-5A binding proteins in extracts of interferon-treated L929 cells were investigated by photoaffinity labeling. Dramatically different photolabeling patterns were observed with the 2- and 8-azido photoprobes. The [gamma-32P]2-azido adenylate trimer 5'-triphosphate photolabels only one polypeptide with a molecular weight of 185,000 as determined by SDS gel electrophoresis, whereas the [alpha-32P]8-azido adenylate trimer 5'-triphosphate covalently photolabels six polypeptides with molecular weights of 46,000, 63,000, 80,000, 89,000, 109,000, and 158,000. Evidence that the photolabeling by 2- and 8-azido 2-5A photoprobes was highly specific for the p3A3 allosteric binding site was obtained as follows.(ABSTRACT TRUNCATED AT 250 WORDS)     

Cytotoxic mechanisms of inhibitor of RNA synthesis, 1-(3-C-ethynyl-beta-D-ribo-pentofuranosyl)cytosine (ECyd, TAS-106).

We investigated the molecular mechanisms of cell death induced by 1-(3-C-ethynyl-beta-D-ribo-pentofuranosyl)cytosine (ECyd, TAS-106: Figure 1), a potent inhibitor of RNA synthesis, using mouse mammary tumor FM3A cells and human fibrosarcoma HT1080 cells. ECyd induced the characteristics of apoptosis on these cells, such as morphological changes, DNA fragmentations and caspase-3-like protease activation. General caspases inhibitor (Z-Asp-CH2-DCB) inhibited cell death. Interestingly, we also found that ECyd induced rRNA fragmentation with the size of 3.2, 2.8 and 1.5 kb, and which might be caused by inhibition of RNA synthesis. rRNA fragmentation was mainly occurred in D8 domain of 28 S rRNA, and the end of 5'-terminal sequence of 1.5 kb fragment was C3220pC3221p or C3221pG3222p, that was identical to the recognition sequence of RNase L. Furthermore, the fragmentation patterns of rRNA digested with RNase L resembled that of ECyd treated cells in shape. These results indicate that antitumor mechanisms of ECyd are involved in activation of RNase L. rRNA fragmentation may be one of the death events as a result of inhibition of RNA synthesis and play an important role in the antitumor activity of ECyd.     

Synthesis of 2-5As possessing base-modified adenosines and their activities to human recombinant RNase L.

The unique 2',5'-oligoadenylate (2-5A) acts as a potent inhibitor of translation in vertebrate cells through the activation of a constituent latent 2-5A-dependent endoribonuclease (RNase L). This 2-5A system plays a major role in the interferon natural defense mechanism against viral infection. We report the syntheses of base-modified adenosine-substituted 2-5A derivatives, their interaction with recombinant human RNase L and their biological stability.     

Regulation of mitochondrial mRNA stability by RNase L is translation-dependent and controls IFNalpha-induced apoptosis

Interferons (IFNs) inhibit the growth of many different cell types by altering the expression of specific genes. IFNs activities are partly mediated by the 2'-5' oligoadenylates-RNase L RNA decay pathway. RNase L is an endoribonuclease requiring activation by 2'-5' oligoadenylates to cleave single-stranded RNA. Here, we present evidence that degradation of mitochondrial mRNA by RNase L leads to cytochrome c release and caspase 3 activation during IFNalpha-induced apoptosis. We identify and characterize the mitochondrial translation initiation factor (IF2mt) as a new partner of RNase L. Moreover, we show that specific inhibition of mitochondrial translation with chloramphenicol inhibits the IFNalpha-induced degradation of mitochondrial mRNA by RNase L. Finally, we demonstrate that overexpression of IF2mt in human H9 cells stabilizes mitochondrial mRNA, inhibits apoptosis induced by IFNalpha and partially reverses IFNalpha-cell growth inhibition. On the basis of our results, we propose a model describing how RNase L regulates mitochondrial mRNA stability through its interaction with IF2mt.     

Purine 8-substitution modulates the ribonuclease L binding and activation abilities of 2',5'-oligoadenylates

Analogues of the 2',5'-linked adenylate trimers monophosphate (p5'A2'p5'A2'p5'A) containing 8-hydroxypropyladenosine, 8-bromoadenosine, and 8-hydroxyadenosine in the first, second, and third nucleotide positions were tested for their ability to bind to and activate RNase L of mouse L cells. p5'AHPr2'p5'AHPr2'p5'AHPr (pAHPr3) (1b) and p5'ABr2'p5'ABr2'p5'ABr (pABr3) (1d) were markedly decreased in ability to bind to the 2-5A dependent endonuclease. On the other hand, analogue of the 2',5'-linked adenylate trimer monophosphate substituted by 8-hydroxyadenosine in the first, second, and third nucleotide position was bound about as well as parent 2-5A [pppA(2'p5'A)2] (p3A3) (1e) to RNase L. Additionally, p5'AOH2'p5'AOH2'p5'AOH (pAOH3) (1c) was as active as parent 2-5A in the rRNA cleavage assay, while pAHPr3 (1b) and pABr3 (1d) were devoid of activity. The 8-substituted analogues of 2-5A were more resistant to the degradation by the (2',5') phosphodiesterase. Finally of particular interest was monophosphate, pAOH3 (1c) which possessed nearly 100% of the translation inhibitory activity of 2-5A triphosphate itself. These results suggest that changes in the base-sugar torsion angles of 2-5A may modulate both binding to and activation of mouse L cell RNase L.