Use of nucleotide analogs by class I and class II CCA-adding enzymes (tRNA nucleotidyltransferase): deciphering the basis for nucleotide selection

We explored the specificity and nature of the nucleotide-binding pocket of the CCA-adding enzyme (tRNA nucleotidyltransferase) by using CTP and ATP analogs as substrates for a panel of class I and class II enzymes. Overall, class I and class II enzymes displayed remarkably similar substrate requirements, implying that the mechanism of CCA addition is conserved between enzyme classes despite the absence of obvious sequence homology outside the active site signature sequence. CTP substrates are more tolerant of base modifications than ATP substrates, but sugar modifications prevent incorporation of both CTP and ATP analogs by class I and class II enzymes. Use of CTP analogs (zebularine, pseudoisocytidine, 6-azacytidine, but not 6-azauridine) suggests that base modifications generally do not interfere with recognition or incorporation of CTP analogs by either class I or class II enzymes, and that UTP is excluded because N-3 is a positive determinant and/or O-4 is an antideterminant. Use of ATP analogs (N6-methyladenosine, diaminopurine, purine, 2-aminopurine, and 7-deaza-adenosine, but not guanosine, deoxyadenosine, 2'-O-methyladenosine, 2'-deoxy-2'-fluoroadenosine, or inosine) suggests that base modifications generally do not interfere with recognition or incorporation of ATP analogs by either class I or class II enzymes, and that GTP is excluded because N-1 is a positive determinant and/or the 2-amino and 6-keto groups are antideterminants. We also found that the 3'-terminal sequence of the growing tRNA substrate can affect the efficiency or specificity of subsequent nucleotide addition. Our data set should allow rigorous evaluation of structural hypotheses for nucleotide selection based on existing and future crystal structures.     

Poly(A) synthesis by RNA polymerase II from rat liver

Multiple forms of DNA-dependent RNA polymerase were resolved by DEAE-Sephadex chromatography. In addition to RNA polymerases, an active poly(A) polymerase was also fractionated. RNA polymerases were examined for their capacity to synthesize poly(A). None of the freshly prepared enzymes could efficiently make poly(A) in presence or absence of exogenous primers. However, “aging” of polymerase II by simple incubation at 37°C resulted in the loss of RNA polymerizing activity with a corresponding increase in poly(A) synthesizing activity. Transformation of RNA polymerase to poly(A) polymerase resulted in reduced capacity to transcribe native DNA and altered chromatographic behavior. The results suggest that subunits of polymerase II obligatory to DNA-dependent RNA synthesis were degraded by “aging” and that a stable subunit of the RNA polymerase could preferentially make poly(A).     

Unusual synthesis by the Escherichia coli CCA-adding enzyme

The tRNA 3' end contains the conserved CCA sequence at the 74-76 positions. The CCA sequence is synthesized and maintained by the CCA-adding enzymes. The specificity of the Escherichia coli enzyme at each of the 74-76 positions was investigated using synthetic minihelix substrates that contain permuted 3' ends. Results here indicate that the enzyme has the ability to synthesize unusual 3' ends. When incubated with CTP alone, the enzyme catalyzed the addition of C74, C75, C76, and multiple Cs. Although the addition of C74 and C75 was as expected, that of C76 and multiple Cs was not. In particular, the addition of C76 generated CCC, which would have conflicted with the biological role of the enzyme. However, the presence of ATP prevented the synthesis of CCC and completely switched the specificity to CCA. The presence of ATP also had an inhibitory effect on the synthesis of multiple Cs. Thus, the E. coli CCA enzyme can be a poly(C) polymerase but its synthesis of poly(C) is regulated by the presence of ATP. These features led to a model of CCA synthesis that is independent of a nucleic acid template. The synthesis of poly(C) by the CCA-adding enzyme is reminiscent of that of poly(A) by poly(A) polymerase and it provides a functional rationale for the close sequence relationship between these two enzymes in the family of nucleotidyltransferases.     

Crystal structures of an archaeal class I CCA-adding enzyme and its nucleotide complexes

CCA-adding enzymes catalyze the addition of CCA onto the 3' terminus of immature tRNAs without using a nucleic acid template and have been divided into two classes based on their amino acid sequences. We have determined the crystal structures of a class I CCA-adding enzyme from Archeoglobus fulgidus (AfCCA) and its complexes with ATP, CTP, or UTP. Although it and the class II bacterial Bacillus stearothermophilus CCA enzyme (BstCCA) have similar dimensions and domain architectures (head, neck, body, and tail), only the polymerase domain is structurally homologous. Moreover, the relative orientation of the head domain with respect to the body and tail domains, which appear likely to bind tRNA, differs significantly between the two enzyme classes. Unlike the class II BstCCA, this enzyme binds nucleotides nonspecifically in the absence of bound tRNA. The shape and electrostatic charge distribution of the AfCCA enzyme suggests a model for tRNA binding that accounts for the phosphates that are protected from chemical modification by tRNA binding to AfCCA. The structures of the AfCCA enzyme and the eukaryotic poly(A) polymerase are very similar, implying a close evolutionary relationship between them.     

Role of tfdC(I)D(I)E(I)F(I) and tfdD(II)C(II)E(II)F(II) gene modules in catabolism of 3-chlorobenzoate by Ralstonia eutropha JMP134(pJP4)

The enzymes chlorocatechol-1,2-dioxygenase, chloromuconate cycloisomerase, dienelactone hydrolase, and maleylacetate reductase allow Ralstonia eutropha JMP134(pJP4) to degrade chlorocatechols formed during growth in 2,4-dichlorophenoxyacetate or 3-chlorobenzoate (3-CB). There are two gene modules located in plasmid pJP4, tfdC(I)D(I)E(I)F(I) (module I) and tfdD(II)C(II)E(II)F(II) (module II), putatively encoding these enzymes. To assess the role of both tfd modules in the degradation of chloroaromatics, each module was cloned into the medium-copy-number plasmid vector pBBR1MCS-2 under the control of the tfdR regulatory gene. These constructs were introduced into R. eutropha JMP222 (a JMP134 derivative lacking pJP4) and Pseudomonas putida KT2442, two strains able to transform 3-CB into chlorocatechols. Specific activities in cell extracts of chlorocatechol-1,2-dioxygenase (tfdC), chloromuconate cycloisomerase (tfdD), and dienelactone hydrolase (tfdE) were 2 to 50 times higher for microorganisms containing module I compared to those containing module II. In contrast, a significantly (50-fold) higher activity of maleylacetate reductase (tfdF) was observed in cell extracts of microorganisms containing module II compared to module I. The R. eutropha JMP222 derivative containing tfdR-tfdC(I)D(I)E(I)F(I) grew four times faster in liquid cultures with 3-CB as a sole carbon and energy source than in cultures containing tfdR-tfdD(II)C(II)E(II)F(II). In the case of P. putida KT2442, only the derivative containing module I was able to grow in liquid cultures of 3-CB. These results indicate that efficient degradation of 3-CB by R. eutropha JMP134(pJP4) requires the two tfd modules such that TfdCDE is likely supplied primarily by module I, while TfdF is likely supplied by module II.     

Interferon induction by platinum(II)-poly(I).poly(C) complexes

The effect of the interaction between poly(I).poly(C) and cis-dichloro-diammineplatinum(II) (cis-Pt), its trans analogue and chloro-diethylene-triamminoplatinum(II) (dien-Pt) on interferon induction activity was investigated. The covalent monodentate fixation of the three compounds on N7 of inosine has different effects on the structure and thermostability of poly(I). poly(C) which is well reflected by the interferon induction activity of the samples. Thus, the sandwich stabilization by dien-Pt at low binding ratios is manifested by an increased interferon induction and a high resistance towards RNAase degradation. The destabilization of the duplex by cis-Pt decreases interferon induction, accompanied by an increase in RNAase sensitivity of the complexes. In the case of trans-Pt the duplex structure is little perturbed and interferon induction is essentially maintained.     

Dissociation of double-stranded poly(I) . poly(C) by cis-diammine-dichloro-Pt(II).

The covalent binding of cis-Pt(NH3)2Cl2 on the double stranded poly(I) . poly(C) induced an irreversible dissociation of the two strands. This dissociation was evidenced mainly by poly(I)-Agarose affinity chromatography which allowed to recover free strands of cis-Pt(NH3)2Cl2-poly(I) from a cis-Pt(NH3)2Cl2-poly(I) . poly(C) complex, by density equilibrium centrifugation where free poly(C) could be isolated, and by acid titrations of the metal-poly(I) . poly(C) complexes. The separation of the two strands of the polyribonucleotide upon cis-Pt(NH3)2Cl2 fixation was shown not to exceed 90--95%. A dissociation curve of the polynucleotide double helix as a function of the amount of bound cis-Pt(NH3)2Cl2 was determined and was shown to be of a characteristic cooperative effect. The fixation of the paltinum compound to poly(I) . poly(C) seemed also to be cooperative.     

Macrophage activation: synergism between hybridoma MAF and poly(I). Poly(C) delivered by liposomes

High concentrations of a murine T cell hybridoma culture supernatant containing macrophage-activating factor (MAF) rendered resident mouse peritoneal macrophages cytotoxic for P815 mastocytoma cells. The capacity of the hybridoma-derived MAF (MAFH) to induce tumoricidal activity increased 10(3) to 10(4)-fold when the lymphokine was encapsulated into liposomes. Combinations of MAFH and poly(I) X poly(C) acted synergistically to render macrophages potently cytotoxic. Subthreshold (nonactivating) concentrations of free or liposome-encapsulated MAFH increased the potency of free poly(I) X poly(C) and liposome encapsulated poly(I) X poly(C). Either as free agent or encapsulated in liposomes, single-stranded poly(I) or poly(C) did not activate macrophages in the presence or absence of MAFH. Double-stranded poly(I) X poly(C) was thus required for macrophage activation and synergism with MAFH.     

CCA-adding enzymes and poly(A) polymerases are all members of the same nucleotidyltransferase superfamily: characterization of the CCA-adding enzyme from the archaeal hyperthermophile Sulfolobus shibatae.

We describe the purification, cloning, and characterization of the CCA-adding enzyme [ATP(CTP):tRNA nucleotidyl transferase] from the thermophilic archaebacterium, Sulfolobus shibatae. Characterization of an archaeal CCA-adding enzyme provides formal proof that the CCA-adding activity is present in all three contemporary kingdoms. Antibodies raised against recombinant, expressed Sulfolobus CCA-adding enzyme reacted specifically with the 48-kDa protein and fully depleted all CCA-adding activity from S. shibatae crude extract. Thus, the cloned cca gene encodes the only CCA-adding activity in S. shibatae. Remarkably, the archaeal CCA-adding enzyme exhibits no strong homology to either the eubacterial or eukaryotic CCA-adding enzymes. Nonetheless, it does possess the active site signature G[SG][LIVMFY]xR[GQ]x5,6D[LIVM][CLIVMFY]3-5 of the nucleotidyltransferase superfamily identified by Holm and Sander (1995, Trends Biochem Sci 20:345-347) and sequence comparisons show that all known CCA-adding enzymes and poly(A) polymerases are contained within this superfamily. Moreover, we propose that the superfamily can now be divided into two (and possibly three) subfamilies: class I, which contains the archaeal CCA-adding enzyme, eukaryotic poly(A) polymerases, and DNA polymerase beta; class II, which contains eubacterial and eukaryotic CCA-adding enzymes, and eubacterial poly(A) polymerases; and possibly a third class containing eubacterial polynucleotide phosphorylases. One implication of these data is that there may have been intraconversion of CCA-adding and poly(A) polymerase activities early in evolution.     

Helix-with-loops structure of polynucleotide. II. Poly(I+CU)

Copolymers of ribocytidylic acid (C) and ribouridylic acid (U) with C contents of 66_.5 and 43 mole-% have been prepared. Ultraviolet, absorption measurements were made of aqueous solutions of the mixtures of these copolymers (poly CU) and polyriboinosinic acid (poly I). In every set of examinations, a maximum hypochromicity (at 230, 248, or 260 mμ) was observed when the solution contained equal moles of C residue in the copolymer and I residue in the homopolymer. This fact was interpreted as indicating that a helix-with-loops structure, similar to that proposed by Fresco and Alberts⁷, is formed. It was shown that this structure is formed more rapidly and melts more sharply than that observed by Fresco and Alberts. An infrared examination was also made of this “helix-with-loops” structure. It was found that a few absorption bands are assignable to the helix part and a few other bands to the loop part of the structure.     

Interferon induction by platinum(II)-poly(I)·poly(C) complexes

The effect of the interaction between poly(I)·poly(C) and cis-dichloro-diammineplatinum(II) (cis-Pt), its trans analogue and chloro-diethylene-triamminoplatinum(II) (dien-Pt) on interferon induction activity was investigated. The covalent monodentate fixation of the three compounds on N⁷ of inosine has different effects on the structure and thermostability of poly(I)·poly(C) which is well reflected by the interferon induction activity of the samples. Thus, the sandwich stabilization by dien-Pt at low binding ratios is manifested by an increased interferon induction and a high resistance towards RNAase degradation. The destabilization of the duplex by cis-Pt decreases interferon induction, accompanied by an increase in RNAase sensitivity of the complexes. In the case of trans-Pt the duplex structure is little perturbed and interferon induction is essentially maintained.     

Poly(ADP-ribose) synthesis and cell division

A clone of HeLa cells has been selected in the presence of 5-methylnicotinamide. In the presence of 10 mM 5-methylnicotinamide the resistant cells grow at 70% of the rate of the same cell culture without 5-methylnicotinamide. 10 mM 5-methylnicotinamide completely inhibits the growth of normal HeLa cells. Both cell types in the absence of 5-methylnicotinamide have the same generation time. Poly (adenosine diphosphate-ribose) synthesis is less sensitive to 5-methylnicotinamide in nuclei isolated from the resistant cells than in nuclei from sensitive cells.     

Inhibition of several enzymes by gold compounds. II. beta-Glucuronidase, acid phosphatase and L-malate dehydrogenase by sodium thiomalatoraurate (I), sodium thiosulfatoaurate (I) and thioglucosoaurate (I)

Bovine liver beta-D-glucuronide glucuronohydrolase, EC 3.2.1.32), wheat germ acid phosphatase (orthophosphoric monoesterphosphohydrolase, EC 3.1.3.2) and bovine liver L-malate dehydrogenase (L-malate: NAD oxidoreductase, EC 1.1.1.37) were inhibited by a series of gold (I) complexes that have been used as anti-inflammatory drugs. Both sodium thiosulfatoaurate (I) (Na AuTs) and sodium thiomalatoraurate (NaAuTM) effectively inhibited all three enzymes, while thioglucosoaurate (I) (AuTG) only inhibited L-malate dehydrogenase. The equilibrium constants (K1) ranged from nearly 4000 microM for the NaAuTM-beta-glucuronidase interaction to 24 microM for the NaAuTS-beta-glucuronidase interaction. The rate of covalent bond formation (kp) ranged from 0.00032 min-1 for NaAuTM-beta-glucuronidase formation to 1.7 min-1 for AuTG-L-malate dehydrogenase formation. The equilibrium data shows that the gold (I) drugs bind by several orders lower than the gold (III) compounds, suggesting a significantly stronger interaction between the more highly charged gold ion and the enzyme. Yet the rate of covalent bond formation depends as much on the structure of the active site as upon the lability of the gold-ligand bond. It was also observed that the more effective the gold inhibition the more toxic the compound.     

Role of class I and class II antigens in the allogenic stimulation: class I and class II recognition in allogenic stimulation; blocking of MLR by monoclonal antibodies and F(ab')2 fragments

Purified monomorphic monoclonal antibodies against Class I and Class II antigens in the inhibition of in vitro allogenic response were assayed. As expected, anti-Class II antibodies are highly inhibitory when used in concentrations greater than 5 micrograms/ml in MLRI and 50 micrograms/ml in MLRII. Surprisingly, anti-Class I monoclonal antibodies are as effective as anti-Class II in inhibiting primary MLR although they have no effect in MLRII. These results were confirmed by using F(ab')2 fragments. The inhibitory effect of anti-Class I has been shown to occur at the stimulator cell level. It is proposed that the allogenic stimulation is elicited after Class I and Class II recognition although only Class II differences are responsible for the proliferative response.     

Production of alloantisera against class II bovine lymphocyte antigens (BoLA) by cross-immunization between class I matched cattle

This paper describes the production of alloantisera directed against bovine major histocompatibility complex (MHC) (BoLA) class II antigens in animals whose MHC phenotypes had been defined by one dimensional isoelectric focusing. Animals of closely matched BoLA class I types were selected by serology and subsequently typed for class I and class II by 1D-IEF of immunoprecipitated antigens. Those with similar class I type by both methods, but differing at the class II locus, were chosen for reciprocal immunization. Cross-immunization was by two skin implantations 6 weeks apart. The resulting antisera showed low titre after the first immunization and elevated titre 3 weeks after the second immunization. The sera reacted strongly with cells expressing specific BoLA class II antigens. The pattern of reactivity correlated well with IEF class II typing on a panel of animals representing all of the class II IEF types present in the Friesian population.     

Characterization of antisera raised against stickleback (Gasterosteus aculeatus) MHC class I and class II molecules

The three-spined stickleback (Gasterosteus aculeatus) is an important model organism for investigations on the maintenance of polymorphism of the major histocompatibility complex (MHC) of vertebrates. Analysis of functional aspects of MHC diversity in stickleback would benefit from the availability of MHC specific reagents. Here we characterize antisera raised against recombinant fusion proteins of stickleback MHC class I alpha and class II alpha and beta. Western blot analysis using recombinant proteins confirmed the specificity of the antisera. In brain and muscle preparations, neither of the MHC types was detectable. High levels of each MHC receptor type were observed in gills and spleen and lower levels in head kidneys. In histological sections of gills, epithelial cells of primary and secondary lamellae stained positive with MHC class I antiserum, while single, scattered cells stained positive for MHC class II. In sections of spleen and head kidney, considerable numbers of cells positive for either MHC type were detected. Molecular weight shift in SDS-PAGE after deglycosylation of MHC class I alpha and class II beta confirmed the predicted glyco-protein character of the molecules. The majority of MHC II alpha was not glycosylated; only a small fraction of MHC II alpha was susceptible to deglycosylation. This suggests differential expression of the two stickleback MHC II alpha genes (Gaac-DAA, Gaac-DBA) only one of which (Gaac-DBA) has a site for N-linked glycosylation.     

Cellular localization of class I (HLA-A,B, C) and class II (HLA-DR and DQ) MHC antigens on the epithelial cells of normal human jejunum

HLA class I and class II (HLA-DR (human I-E equivalent) and DQ (human I-A equivalent] antigens were localized by immunofluorescence technique on thin frozen sections of normal human jejunum using a panel of monomorphic monoclonal antibodies. HLA class I (A, B and C) and HLA-DR molecules were found in the basolateral membrane of enterocytes; HLA-DR were also detected in a patchy distribution in the apical part of enterocytes; HLA-DQ molecules (the human equivalent of the murine I-A molecular subset) were not detected on normal enterocytes. All three molecules were detected on the membrane of lymphocytes and monocytes present in the lamina propria.     

The synthesis and behaviour of pyrazine mononuclear carbonyl complexes of Rh(I), Ir(I), Ru(II) and Os(II)

The synthesis in high yields and the dissociative behaviour in the solid state and in solution of the mononuclear complexes [cis-M(CO)₂Cl(pyz)] (M=Rh, Ir; PYZ=pyrazine) and [fac-M(CO)₃Cl₂(pyz)] (M=Ru, Os) are reported. The mononuclear complexes of Rh and Ir are relatively labile with respect to pyrazine release. Particularly in the case of rhodium they generate even in the solid state the corresponding dinuclear complexes [cis-Cl(CO)₂M(pyz)cis-M(CO)₂Cl] (M=Rh, Ir). The ¹H NMR spectra of these mononuclear Rh and Ir complexes in CHCl₃ solution show, at 25 and 60 °C, respectively, a fast and reversible dissociation of metal coordinated pyrazine, which is hindered by lowering the temperature. Crystallographic aspects of [cis-Ir(CO)₂Cl(pyz)] have been investigated via single crystal X-ray diffraction. The mononuclear complexes of Ru and Os are more stable. In the solid state they do not rearrange, with release of pyrazine, to generate the related dimeric complexes with pyrazine as bridge. In solution, at room temperature, they do not dissociate quickly, although a mixture of monomeric and dimeric pyrazine complexes (ratio monomer to dimer 9:1 and 15:1 for Ru and Os, respectively) is slowly formed by a process which is reverted by addition of excess pyrazine, as expected for a dissociative equilibrium.     

A model for C74 addition by CCA-adding enzymes: C74 addition, like C75 and A76 addition, does not involve tRNA translocation

The CCA-adding enzyme adds CCA to the 3'-end of tRNA one nucleotide at a time, using CTP and ATP as substrates. We found previously that tRNA does not rotate or translocate on the enzyme during the addition of C75 and A76. We therefore predicted that the growing 3'-end of tRNA must, upon addition of each nucleotide, refold to reposition the new 3'-hydroxyl equivalently relative to the solitary nucleotidyltransferase motif. Cocrystal structures of the class I archaeal Archaeoglobus fulgidus enzyme, poised for addition of C75 and A76, confirmed this prediction. We have also demonstrated that an evolutionarily flexible beta-turn facilitates progressive refolding of the 3'-terminal C74 and C75 residues during C75 and A76 addition. Although useful cocrystals corresponding to C74 addition have not yet been obtained, we now show experimentally that tRNA does not rotate or translocate during C74 addition. We therefore propose, based on the existing A. fulgidus cocrystal structures, that the same flexible beta-turn functions as a wedge between the discriminator base (N73) and the terminal base pair of the acceptor stem, unstacking and repositioning N73 to attack the incoming CTP. Thus a single flexible beta-turn would orchestrate consecutive addition of all three nucleotides without significant movement of the tRNA on the enzyme surface.     

Cellular localization of class I (HLA-A, B, C) and class II (HLA-DR and DQ) MHC antigens on the epithelial cells of normal human jejunum

HLA class I and class II (HLA-DR (human I-E equivalent) and DQ (human I-A equivalent] antigens were localized by immunofluorescence technique on thin frozen sections of normal human jejunum using a panel of monomorphic monoclonal antibodies. HLA class I (A, B and C) and HLA-DR molecules were found in the basolateral membrane of enterocytes; HLA-DR were also detected in a patchy distribution in the apical part of enterocytes; HLA-DQ molecules (the human equivalent of the murine I-A molecular subset) were not detected on normal enterocytes. All three molecules were detected on the membrane of lymphocytes and monocytes present in the lamina propria.