Poly(A) inclusive RNA isoform sequencing (PAIso-seq) reveals wide-spread non-adenosine residues within RNA poly(A) tails.

Message RNA poly(A) tails are vital for their function and regulation. However, the full-length sequence of mRNA isoforms with their poly(A) tails remains undetermined. Here, we develop a method at single-cell level sensitivity that enables quantification of poly(A) tails along with the full-length cDNA while reading non-adenosine residues within poly(A) tails precisely, which we name poly(A) inclusive RNA isoform sequencing (PAIso-seq). Using this method, we can quantify isoform specific poly(A) tail length. More interestingly, we find that 17% of the mRNAs harbor non-A residues within the body of poly(A) tails in mouse GV oocytes. We show that PAIso-seq is sensitive enough to analyze single GV oocytes. These findings will not only provide an accurate and sensitive tool in studying poly(A) tails, but also open a door for the function and regulation of non-adenosine modifications within the body of poly(A) tails.

Prediction of hepatitis C virus interferon/ribavirin therapy outcome based on viral nucleotide attributes using machine learning algorithms.

BACKGROUND: Hepatitis C virus (HCV) causes chronic hepatitis C in 2-3% of world population and remains one of the health threatening human viruses, worldwide. In the absence of an effective vaccine, therapeutic approach is the only option to combat hepatitis C. Interferon-alpha (IFN-alpha) and ribavirin (RBV) combination alone or in combination with recently introduced new direct-acting antivirals (DAA) is used to treat patients infected with HCV. The present study utilized feature selection methods (Gini Index, Chi Squared and machine learning algorithms) and other bioinformatics tools to identify genetic determinants of therapy outcome within the entire HCV nucleotide sequence. RESULTS: Using combination of several algorithms, the present study performed a comprehensive bioinformatics analysis and identified several nucleotide attributes within the full-length nucleotide sequences of HCV subtypes 1a and 1b that correlated with treatment outcome. Feature selection algorithms identified several nucleotide features (e.g. count of hydrogen and CG). Combination of algorithms utilized the selected nucleotide attributes and predicted HCV subtypes 1a and 1b therapy responders from non-responders with an accuracy of 75.00% and 85.00%, respectively. In addition, therapy responders and relapsers were categorized with an accuracy of 82.50% and 84.17%, respectively. Based on the identified attributes, decision trees were induced to differentiate different therapy response groups. CONCLUSIONS: The present study identified new genetic markers that potentially impact the outcome of hepatitis C treatment. In addition, the results suggest new viral genomic attributes that might influence the outcome of IFN-mediated immune response to HCV infection.

Depletion of dimeric all-alpha dUTPase induces DNA strand breaks and impairs cell cycle progression in Trypanosoma brucei.

The enzyme deoxyuridine 5'-triphosphate nucleotidohydrolase (dUTPase) is responsible for the control of intracellular levels of dUTP thus controlling the incorporation of uracil into DNA during replication. Trypanosomes and certain eubacteria contain a dimeric dUTP-dUDPase belonging to the recently described superfamily of all-alpha NTP pyrophosphatases which bears no resemblance with typical eukaryotic trimeric dUTPases and presents unique properties regarding substrate specificity and product inhibition. While the biological trimeric enzymes have been studied in detail and the human enzyme has been proposed as a promising novel target for anticancer chemotherapeutic strategies, little is known regarding the biological function of dimeric proteins. Here, we show that in Trypanosoma brucei, the dimeric dUTPase is a nuclear enzyme and that down-regulation of activity by RNAi greatly reduces cell proliferation and increases the intracellular levels of dUTP. Defects in growth could be partially reverted by the addition of exogenous thymidine. dUTPase-depleted cells presented hypersensitivity to methotrexate, a drug that increases the intracellular pools of dUTP, and enhanced uracil-DNA glycosylase activity, the first step in base excision repair. The knockdown of activity produces numerous DNA strand breaks and defects in both S and G2/M progression. Multiple parasites with a single enlarged nucleus were visualized together with an enhanced population of anucleated cells. We conclude that dimeric dUTPases are strongly involved in the control of dUTP incorporation and that adequate levels of enzyme are indispensable for efficient cell cycle progression and DNA replication.

Roles for TbDSS-1 in RNA surveillance and decay of maturation by-products from the 12S rRNA locus.

The Trypanosoma brucei exoribonuclease, TbDSS-1, has been implicated in multiple aspects of mitochondrial RNA metabolism. Here, we investigate the role of TbDSS-1 in RNA processing and surveillance by analyzing 12S rRNA processing intermediates in TbDSS-1 RNAi cells. RNA fragments corresponding to leader sequence upstream of 12S rRNA accumulate upon TbDSS-1 depletion. The 5' extremity of 12S rRNA is generated by endonucleolytic cleavage, and TbDSS-1 degrades resulting upstream maturation by-products. RNAs with 5' ends at position -141 and 3' ends adjacent to the mature 5' end of 12S rRNA are common and invariably possess oligo(U) tails. 12S rRNAs with mature 3' ends and unprocessed 5' ends also accumulate in TbDSS-1 depleted cells, suggesting that these RNAs represent dead-end products normally destined for decay by TbDSS-1 in an RNA surveillance pathway. Together, these data indicate dual roles for TbDSS-1 in degradation of 12S rRNA maturation by-products and as part of a mitochondrial RNA surveillance pathway that eliminates stalled 12S processing intermediates. We further provide evidence that TbDSS-1 degrades RNAs originating upstream of the first gene on the minor strand of the mitochondrial maxicircle suggesting that TbDSS-1 also removes non-functional RNAs generated from other regions of the mitochondrial genome.

Small ncRNA transcriptome analysis from kinetoplast mitochondria of Leishmania tarentolae.

Gene expression in mitochondria of kinetoplastid protozoa requires RNA editing, a post-transcriptional process which involves insertion or deletion of uridine residues at specific sites within mitochondrial pre-mRNAs. Sequence specificity of the RNA editing process is mediated by oligo-uridylated small, non-coding RNAs, designated as guide RNAs (gRNAs). In this study, we have analyzed the small ncRNA transcriptome from kinetoplast mitochondria of Leishmania tarentolae by generating specialized cDNA libraries encoding size-selected RNA species. Through this screen, a significant number of novel oligo-uridylated RNA species, which we have termed oU-RNAs, has been identified. Most novel oU-RNAs are present as stable RNA species in mitochondria as assessed by northern blot analysis. Thereby, novel oU-RNAs show similar expression levels and sizes as previously reported for canonical gRNAs. Several oU-RNAs are transcribed from both strands of the maxicircle and minicircles components of the mitochondrial genome, from regions where up till now no transcription has been reported. Two stable oU-RNAs exhibit an anchor sequence in antisense orientation to known gRNAs and thus might regulate editing of respective pre-mRNAs. A number of oU-RNAs map in antisense orientation to non-edited protein-coding genes suggesting that they might function by a different mechanism. In addition, our screen shows that all kinetoplast-derived RNAs are prone to some degree of uridylation.

Terminal dUTP nick end labeling (TUNEL) positive cells in the different regions of the brain in normal aging and Alzheimer patients.

This study investigated terminal dUTP nick-end labeling (TUNEL)-positive cells in the frontal, occipital, and hippocampal cortices of seven normal aging and four Alzheimer's patients. Significant increase in TUNEL-positive cells was observed in the frontal and hippocampal cortices of Alzheimer's patients when compared with controls. In the hippocampal cortex, only area CA4 demonstrated a significant increase of TUNEL-positive cells. Double staining of TUNEL-positive cells for glial fibrillary acidic protein revealed that < 13% of the TUNEL-positive nuclei belonged to astrocytes. The results of this study illustrated a differential pattern of cortical degeneration between normal aging and Alzheimer patients.

An in vivo 31P magnetic resonance spectroscopy study of uridine excess in rats fed orotic acid.

Spatially localized 31P NMR spectroscopy was used to assay in vivo the liver of intact rats fed orotic acid (OA) in a diet which produces hepatic steatosis. Twenty-three sets of multiple volume spectra were obtained from twenty-one 265- to 315-g female rats after 0-9 days of feeding either a 1% OA/64% sucrose diet (12 rats) or a 65% sucrose control diet (9 rats). The intensity of the in vivo diphosphodiester resonance ascribed to UDP-hexos(amin)es increased and the phosphomonoester resonance decreased in intensity prior to fatty infiltration. High resolution NMR spectroscopy of extracts of these livers indicated that the UDP-hexos(amin)e peak included four different UDP-sugars including UDP-N-acetylglucosamine (UDP-glcNAc), and that lower phosphocholine (P-Cho) accounted for the lower phosphomonoester resonance in vivo. Increased UDP-glcNAc is thought to reflect impaired lipoprotein glycosylation as a mechanism for hepatic steatosis in orotic acid feeding. P-Cho deficiency has been shown to be due to an increased rate of phosphatidylcholine synthesis. Low P-Cho concentration has been shown to be associated with lipid accumulation in a choline-deficient diet, but was not previously associated with hepatic steatosis in OA feeding. Changes in phosphorus metabolites were observed 2 days prior to development of fatty liver. HPLC assay of uridine nucleotides showed a good correlation between magnetic resonance spectroscopy and HPLC quantitation. In this study there were two biochemical correlates of impaired hepatic lipid secretion detectable by in vivo assay with 31P NMR spectroscopy. This method has application for noninvasive assays in ornithine transcarbamylase-deficient patients.

A small nucleolar RNA is processed from an intron of the human gene encoding ribosomal protein S3.

A human small nucleolar RNA, identified previously in HeLa cells by anti-fibrillarin autoantibody precipitation and termed RNA X, has been characterized. It comprises two uridine-rich variants (148 and 146 nucleotides), which we refer to as snRNA U15A and U15B. Secondary structure models predict for both variants a U15-specific stem-loop structure, as well as a new structural motif that contains conserved sequences and can also be recognized in the other fibrillarin-associated nucleolar snRNAs, U3, U14, and RNA Y. The single-copy gene for human U15A has been found unexpectedly to reside in intron 1 of the ribosomal protein S3 gene; the U15A sequence appears on the same strand as the S3 mRNA and does not exhibit canonical transcription signals for nuclear RNA polymerases. U15A RNA is processed in vitro from S3 intron 1 transcripts to yield the correct 5' end with a 5'-monophosphate; the in vitro system requires ATP for 3' cleavage, which occurs a few nucleotides downstream of the mature end. The production of a single primary transcript specifying the mRNA for a ribosomal or nucleolar protein and a nucleolar snRNA may constitute a general mechanism for balancing the levels of nucleolar components in vertebrate cells.

The effect of orotic acid on the response of the recently infarcted rat heart to hypothermic cardioplegia.

Patients with a recent myocardial infarction have a higher morbidity and mortality than comparable patients with chronic myocardial ischaemia. We postulated that this might be due to a reduced overall tolerance of the heart to cardioplegic arrest in the presence of a recent infarct. We postulated that orotic acid, a pyrimidine precursor which augments the rate of protein synthesis, might improve the response of the recently infarcted heart to cardioplegic arrest. Myocardial infarction was produced in rats by coronary ligation. The rats were then divided into two groups according to whether they were treated with oral orotic acid (10 mg/kg per day) or untreated. A sham-operated (non-infarcted) group served as normal controls. After 2 days, the hearts (n = 12 per group) underwent 1 h of cardioplegic arrest at 23 degrees C on the isolated working heart apparatus. Before arrest, maximum cardiac function in the untreated infarct group was lower than in the normal group (P less than 0.05), whereas in the treated group, function was similar to the normal group. After arrest there was severe depression of cardiac function in the untreated infarct group: only 57% recovery of the pre-arrest value compared with 86% in the normal group (P less than 0.001). In the orotic acid treated group, recovery (90%) was significantly greater than in the untreated group (P less than 0.001) and equivalent to the normal group. Oxygen utilisation, when corrected for external work, was higher in both infarct groups than in the normal group before and after arrest (P less than 0.05 in both cases). Total uridine nucleotide content of the infarcted and non-infarcted zones of the heart was increased. Treatment with orotic acid produced a further upward trend in uridine nucleotide levels. We conclude that an established, recent infarct reduces the overall tolerance of the heart to hypothermic cardioplegia. Treatment with orotic acid improves the function of the infarcted heart following cardioplegic arrest, and may therefore improve the results of urgent cardiac surgery in patients with myocardial infarction.

Antiproliferative effects of 4',5'-unsaturated adenine nucleosides on leukemia L1210 cells in vitro.

Several 4',5'-unsaturated adenine nucleosides were shown to have antiproliferative activity against L1210 leukemia cells in vitro. The active nucleosides were cytotoxic to the L1210 cells as demonstrated by Trypan Blue uptake. The cytotoxicity was not induced by alterations in the ribonucleoside and deoxyribonucleoside triphosphate levels of the L1210 cells.

Identification of 6-azauridine triphosphate in L1210 cells and its possible relevance to cytotoxicity.

L1210 cells treated with 1 mM 6-azauridine (AzUrd) (concentration causing 50% inhibition of cell growth, 3 microM) continued to divide at a reduced rate for 72 h before stopping. However, a 24-h treatment was lethal to 99% of the cells, as determined by colony formation. To investigate the mechanism for this delayed cytotoxicity, the metabolism of AzUrd was studied. Cells incubated with AzUrd contained a new 254 nm-absorbing component, not found in control cells. It appeared to be 6-azauridine-5'-triphosphate, since it was the only peak in the triphosphate region of the chromatogram which contained 3H after incubation of cells with [3H]AzUrd. Incorporation of [3H]AzUrd into the acid-insoluble fraction (nucleic acids) was also detected. A role for this incorporation in the mechanism of AzUrd cytotoxicity was strongly suggested by the observation that cordycepin (0.01 mM) partially protected cells from the lethality of AzUrd, presumably by preventing its incorporation into RNA. The previously known inhibition of pyrimidine de novo synthesis by AzUrd was confirmed by a decrease in the intracellular contents of UTP and CTP in AzUrd-treated cells. Therefore, we propose that the inhibition of pyrimidine de novo synthesis and the incorporation into nucleic acid(s) may act in concert to produce the cytotoxic effects of AzUrd.

Effect of L-histidinol on the metabolism of 5-fluorouracil in the BALB/c x DBA/8 F1 murine tumor system.

L-Histidinol, a structural analogue of histidine, which transiently inhibits proliferation, can protect cells from the toxic effects of proliferation-dependent chemotherapeutic agents such as 5-fluorouracil (FUra). In the BALB/c x DBA/8 F1 (hereafter called CD8F1) murine tumor system, L-histidinol protected mice from FUra-induced leukopenia, weight loss, and mortality; however, the therapeutic index was not improved since L-histidinol also protected the tumor against the toxic effects of FUra. In order to understand the mechanism of this protection, we examined the effects of L-histidinol on the metabolism of FUra. Results indicate that L-histidinol had no effect on the phosphoribosyl pyrophosphate levels in tumor, the activation of FUra to nucleotides or levels of free 5-fluorodeoxyuridine monophosphate in either tumor or bone marrow. L-Histidinol (7 mg/mouse, every 2 h for 5 doses) reduced RNA and DNA synthesis, as measured by 32P incorporation in vivo, by approximately one-half in tumor, and by 70% in bone marrow. This in turn resulted in reduced incorporation of FUra into RNA in both tumor and bone marrow. At 2 h, 4 h, and 24 h after FUra administration the level of FUra in RNA was 24-37% less in both tumor and bone marrow of mice that received L-histidinol with FUra. Using 32P as a monitor of overall RNA synthesis, the [3H]FUra/32P ratio remained unchanged, suggesting that the reduction of FUra incorporation into RNA was due to decreased RNA synthesis rather than a decrease in the number of FUra molecules per RNA chain. In contrast, L-histidinol had no effect on the in vivo inhibition of thymidylate synthetase by 5-fluorodeoxyuridine monophosphate as measured by the incorporation of [3H]-2'-deoxyuridine into DNA or on the percentages of thymidylate synthetase in the free versus 5-fluorodeoxyuridine monophosphate-bound state. We conclude that L-histidinol reduces FUra toxicity by reducing FUra incorporation into RNA. Since the major mechanism of action in the CD8F1 breast tumor system is the incorporation of FUra into RNA, the reduction in toxicity and antitumor activity observed when L-histidinol is combined with FUra is consistent with the observed reduction in tumor and bone marrow RNA containing incorporated FUra residues.

Pyridyl groups for protection of the imide functions of uridine and guanosine. Exploration of their displacement reactions for site-specific modifications of uracil and guanine bases.

For the protection of the O-4 function of uridine and the O-6 of guanosine, 2-, 3- and 4-hydroxypyridines, 2-pyridinethiol, 6-methyl-2-hydroxy- and 6-methyl-3-hydroxypyridines have been employed. These substituted pyridines gave pyridyl-N-and/or pyridyl-O-substituted derivatives, depending both upon the position of the hydroxyl and methyl groups in the pyridine ring, at the C-4 and the C-6 of the uracil and guanine residues, respectively. These groups were found to be good leaving groups for nucleophilic substitution reactions by amines, thiolates and oximate. If needed, the rate of these substitution reactions could be conveniently increased by almost 1000-fold by conversion of the pyridyl moiety to its methiodide.

Location of oligo(uridylic acid) sequences within messenger ribonucleic acid molecules of HeLa cells.

A significant fraction of the polyadenylated mRNAs of HeLa cells contain an oligo(uridylic acid) [oligo(U)] sequence of 15-30 nucleotides. Several different experimental approaches were used to determine if these oligo(U)'s occupied similar sites within all mRNAs. In one approach, poly(adenylic acid)-containing mRNAs [poly(A+) mRNAs] averaging 2800 nucleotides in length were reduced to an average size of 500 nucleotides by controlled alkaline hydrolysis. Over 20% of the oligo(U)-containing fragments isolated from the hydrolysate retained a poly(A) sequence, showing that oligo(U)'s were not exclusively located near 5' ends of mRNA although 20% were apparently close to 3' ends. To confirm these observations, oligo(U)-containing mRNA [oligo(U+) mRNA] was exposed to the 3'-exonucleolytic activity of polynucleotide phosphorylase to produce fragments containing the 5' regions of mRNA. Each of a set of fragments of decreasing length generated by increased times of exposure of the mRNAs to the enzyme was found to have about the same oligo(U) content, including the shortest that averaged 550 nucleotides. These data not only eliminated an exclusive location for oligo(U) in either 3' or 5' ends of mRNA but also suggested that oligo(U)'s might be close to the 5' ends of some mRNAs. To verify this last observation, periodate-oxidized poly(A+) mRNA was labeled at the 5' caps and at 3'-adenosine residues by sodium [3H]borohydride reduction before it was nicked 3-5 times with alkali to produce 5' and 3' end-labeled pieces that could be separated with oligo(thymidylic acid)-cellulose.(ABSTRACT TRUNCATED AT 250 WORDS)

Pseudouridylic modification of a 6S RNA transcribed in vitro from highly repetitive and transcribable (Hirt) sequences of salmon total DNA.

Salmon total DNA was transcribed in a HeLa cell extract, and it was found that a distinct 6S RNA is transcribed by RNA polymerase III. By analogy with a 6S RNA transcribed from mouse total DNA in vitro, which we demonstrated to be the transcripts from type 2 Alu family (Sakamoto et al. (1984) Mol Gen Genet 194:1-6), the genes for the 6S RNA in the salmon genome are presumed to be reiterated in at least 10(4)-10(5) copies and to be dispersed among the genome. A fingerprint analysis showed that the nucleotide sequences of the genes for the 6S RNA are extensively conserved. Furthermore, we demonstrated that at least two positions in the 6S RNA are modified in vitro to pseudouridylic acid. The pseudouridylic acid residues are located in specific oligonucleotides and each of the modified nucleotides amounts to 0.5 to 1.0 mol per mol of RNA. The relationship between salmon highly repetitive and transcribable (Hirt) sequences and rodent type 2 Alu family, and the significance of the pseudouridylic modification in relation to the functional role of these sequences are discussed.

A comparison of purine and pyrimidine pools in Bloom's syndrome and normal cells.

Nucleotide content of normal and Bloom's syndrome fibroblasts and lymphocytes were examined by reversed phase HPLC. The ATP/ADP ratio in primary cultures of normal human fibroblasts was at least three fold higher than in the primary cultures of Bloom's syndrome fibroblasts. After three months in culture the ratios of ATP/ADP of the Bloom's cells approach those of normal fibroblasts. Individual nucleotide measurements showed that initial differences did not reflect excess ADP, but rather very low levels of ATP in Bloom's syndrome fibroblasts. The amount of ATP increased gradually during culture. However, even after three months in culture, significant differences were noted in ATP levels between Bloom's syndrome fibroblasts and normal fibroblasts. Thus a defect in Bloom's syndrome is correlated with a defect in purine biosynthesis or ATP generation.

Oligo(U) sequences present in sea urchin maternal RNA decrease following fertilization.

Oligo(U) tracts were identified and measured in RNA from sea urchin eggs and embryos using a quantitative assay based on the amount of [3H]poly(A) protected from RNase T2 in duplexes with the oligo(U). The oligo(U) amounted to 0.0035% of egg RNA (0.063 X 10(-12) g/egg) and decreased to 0.0015% (0.027 X 10(-12) g/embryo) by 2 hr after fertilization. The oligo(U) tracts had a maximum size of 15-30 nucleotides and were associated with two size classes of RNA. In eggs about half were in 100 to 200 nucleotide RNA and half in mRNA-sized molecules. After fertilization, the oligo(U) in the population of large-mRNA-sized molecules was greatly reduced.

Reversed-phase ion-pair high-performance liquid chromatographic assay of 5-fluorouracil, 5'-deoxy-5-fluorouridine, their nucleosides, mono-, di-, and triphosphate nucleotides with a mixture of quaternary ammonium ions.

Several quaternary ammonium ions were evaluated in the development of a reversed-phase ion-pair high-performance liquid chromatographic assay for the separation of the known nucleosides and nucleotides of 5-fluorouracil (FU) and its analogue, 5'-deoxy-5-fluorouridine (5'-dFUR). The capacity factors of FU, 5'-dRUR, and their eight anabolites including 5-5-fluorouridjine and its mono-, di-, triphosphate and diphosphoglucose, 5-fluorodeoxyuridine and its mono;- and diphosphate nucleotides, were dependent on the chain length and concentration of the counter ions, pH, type of buffer, as well as the type of bonded stationary phase. Separation of FU, 5'-dFUR, their nucleosides and monophosphate nucleotides was readily achieved using tetrabutylammonium ion alone as the counter ion. However, under these conditions, the di- and triphosphate nucleotides were eluted from the column only after lengthy gradient elution and with poor reproducibility. Optimal conditions for a simultaneous separation of the ten fore-mentioned compounds were achieved using a two-step elution with a mixture of tetraethylammonium (C8) and tetrabutylammonium (C16) ions. The first eluent consisted of C8 and C16 ions in a mixture of acetate--phosphate buffer and methanol, the second eluent contained an additional 30 mM phosphate. FU, 5'-dFUR, their nucleosides and monophosphates and diphosphoglucose were separated by isocratic elution from a microBondapak C18 reversed-phase column using the first eluent; and the di- and triphosphate nucleotides were subsequently eluted, isocratically with the second eluent. This assay does not require gradient elution and can be completed within 50 min with good reproducibility.

5'-Terminal cap structures of oligo(uridylic acid)-containing messenger ribonucleic acid from HeLa cells: comparison with other ribonucleic acid subpopulations.

The 5'-terminal cap structures of 32P-labeled oligo(uridylic acid)-containing messenger ribonucleic acid [oligo(U+)mRNA] isolated from HeLa cell polyadenylated [poly(A+)] mRNA were analyzed and compared to those of the poly(A+) mRNA. A method employing P1 nuclease, alkaline phosphatase, and adsorption to activated charcoal showed that the types of cap core (m7 GpppXm) in oligo(U+) mRNA were essentially identical with those in poly(A+) mRNA. Analysis of RNase T2 digestion products of oligo(U+) mRNA demonstrated the presence of both cap 1 (m7GpppXmpYp) and cap 2 (m7GpppXmpYmpZp) in this subpopulation, confirming its cytoplasmic location. The base compositions of these two types of caps were different from each other and nonrandom but did not differ significantly between oligo(U+) and poly(A+) m RNA. The only observed difference between the mRNA populations was a higher ratio of cap 1 and cap 2 in the former. Possible implications of these findings for the relationship between oligo(U+) mRNA and poly(A+) mRNA are discussed.