Skip Navigation

This Article
Right arrow Abstract Freely available
Right arrow Print PDF (190K) Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (21)
Right arrowRequest Permissions
Right arrow Commercial Re-use Guidelines
for Open Access NAR Content
Google Scholar
Right arrow Articles by Trippe, R.
Right arrow Articles by Benecke, B. J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Trippe, R.
Right arrow Articles by Benecke, B. J.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

Nucleic Acids Research Pages 3119-3126  

A highly specific terminal uridylyl transferase modifies the 3[prime]-end of U6 small nuclear RNA
Introduction
Materials And Methods
   Cellular extracts and fractionation
   Templates
   In vitro transcription
   Northern blot hybridization
   Terminal uridylyl transferase assay
Results
   HeLa cells contain a non-specifc and a U6 RNA-specific terminal uridylyl transferase
   Recognition of cellular and synthetic U6 RNA by the specific TUTase
   U6 RNAs from various eucaryotes, but not fission yeast, are modified by the U6-specific TUTase
   U6-specific TUTase restores the four template-encoded UMP residues
Discussion
Acknowledgements
References

A highly specific terminal uridylyl transferase modifies the 3[prime]-end of U6 small nuclear RNA

A highly specific terminal uridylyl transferase modifies the 3[prime]-end of U6 small nuclear RNA

Ralf Trippe, Björn Sandrock, Bernd-Joachim Benecke*

Department of Biochemistry NC6, Ruhr University, D-44780 Bochum, Germany

Received April 2, 1998; Revised and Accepted May 14, 1998

ABSTRACT

HeLa cell extracts contain significant amounts of terminal uridylyl transferase (TUTase) activity. In a template-independent reaction with labeled UTP, these enzymes are capable of modifying a broad spectrum of cellular RNA molecules in vitro. However, fractionation of cell extracts by gel filtration clearly separated two independent activities. In addition to a non-specific enzyme, an additional terminal uridylyl transferase has been identified that is highly specific for cellular and in vitro synthesized U6 small nuclear RNA (snRNA) molecules. This novel TUTase enzyme was also able to select as an efficient substrate U6 snRNA species from higher eucaryotes. In contrast, no labeling was detectable with purified fission yeast RNA. Using synthetic RNAs containing different amounts of transcribed 3[prime]-end UMP residues, high resolution gel electrophoresis revealed that U6 snRNA species with three terminal U nucleotides served as the optimal substrate for the transferase reaction. The 3[prime]-end modification of the optimal synthetic substrate was identical to that observed with endogenous U6 snRNA isolated from HeLa cells. Therefore, we conclude that the specific addition of UMP residues to 3[prime]-recessed U6 snRNA molecules reflects a recycling process, ensuring the functional regeneration for pre-mRNA splicing of this snRNA.

INTRODUCTION

Post-transcriptional modification and processing of newly synthesized RNA molecules represent fundamental aspects of eucaryotic gene expression. In that context, one of the most intensely studied processes is the splicing of nuclear pre-mRNA during mRNA biogenesis (1,2). The splicing reaction is catalyzed by a large ribonucleoprotein (RNP) complex designated the spliceosome (3). This complex is formed by several ribonucleoprotein particles containing distinct small nuclear RNA (snRNA) molecules and their associated proteins. In addition, several non-snRNP proteins are also required for pre-mRNA splicing (1,4,5).

The snRNPs involved in nuclear pre-mRNA splicing consist of five distinct particles containing the snRNAs U1, U2, U4, U5 and U6 respectively (6,7). These snRNAs participate in specific RNA-RNA interactions required for assembly of the spliceosome and for catalysis of individual steps of the two transesterification reactions (8).

Among the snRNA molecules, U6 snRNA is the most highly conserved snRNA, from yeast to man (9), and therefore is believed to play a central role in the catalytic steps of pre-mRNA splicing. In several aspects, U6 snRNA differs from the other spliceosomal RNAs. The majority of these snRNAs are transcribed by RNA polymerase II (10). In contrast, U6 RNA is transcribed by RNA polymerase III (11-13). Furthermore, U6 snRNA contains a [gamma]-methyltriphosphate at its 5[prime]-end (14), as opposed to the trimethylated guanosine cap structure of the other snRNAs. In addition, the 3[prime]-end of U6 snRNA undergoes extensive processing before being incorporated into the U6 snRNP complex (15,16). Finally, U6 RNA is also unique in that UMP residues are both added to and removed from its 3[prime]-end post-transcriptionally (12,17-19). Due to these modifications, human U6 snRNA has been found to be heterogeneous in size (20). The major form terminates with five UMP residues and an unusual 2[prime],3[prime]-cyclic phosphate (>p; 18), but a substantial fraction (~10%) contains an oligouridylate stretch of variable length (up to 12 residues) terminating with a 3[prime]-hydroxyl group (19). The enzymes responsible for these modifications of the 3[prime]-end of the U6 RNA have not been identified and remain to be characterized. On the other hand, HeLa cells are known to contain a 3[prime]-terminal phosphate cyclase that, in an ATP-dependent reaction, catalyzes conversion of a 3[prime]-phosphate to a 2[prime],3[prime]-cyclic phosphodiester (21). Such a 3[prime]-terminal phosphate cyclase has recently been cloned, but template specificity of this enzyme has not been demonstrated (22). Therefore, it is not known at present whether or not this enzyme is responsible for the corresponding U6 snRNA modification.

It is well documented that cellular extracts contain enzyme activities which, in a template-independent reaction, incorporate labeled UMP into endogenous RNA molecules when using [[alpha]-32P]UTP as substrate. Such enzymes, designated terminal uridylyl transferases (TUTases), have been characterized in a variety of tissues and organisms (23 and references therein). TUTases catalyze the transfer of UMP residues to the 3[prime]-hydroxyl group of a large number of RNA species, including U6 RNA (12). Here we report the identification of a TUTase enzyme that exclusively transfers UMP residues to the 3[prime]-end of U6 snRNA molecules. Furthermore, according to its apparent molecular mass and its highly selective template specificity, this enzyme is clearly distinct from the aforementioned non-specific TUTase activities.

MATERIALS AND METHODS

Cellular extracts and fractionation

Cytoplasmic S100 extracts (15 mg/ml) were prepared from HeLa cells as described in detail previously (24). Extracts were used either for in vitro transcription reactions or for partial purification of TUTase enzymes by phosphocellulose P11 (Whatman) chromatography and gel filtration in Superdex G200 columns (Pharmacia).

Templates

Templates used for in vitro transcription in HeLa cell extracts (Fig. 1) were as follows. An adenoviral VA I RNA gene reporter sequence supplemented with the human 7S K terminator region was fused to pol III gene (genes transcribed by RNA polymerase III) promoters with varying amounts of gene internal sequences. This VA I cassette and the entire 7SK(6)VA construct have been described in detail by Emde et al. (25; see their fig. 3 and templates). The mouse U6 fusion construct was based on the mouse U6 snRNA gene (kindly provided by Dr Ram Reddy, Houston, TX) and fusion of the VA I reporter sequence was to position +26 of that gene.

Templates used for in vitro transcription with purified T7 RNA polymerase (Boehringer) were generated by PCR. Two gene-specific primers which also provided additional external sequences were applied. That specific for the 5[prime]-end of the respective gene provided the entire T7 promoter with the missing number of downstream G residues to create the GGG initiator element optimal for T7 RNA polymerase initiation. Therefore, in the case of the U6 RNA gene, the resulting T7 transcripts differed from cellular U6 RNA by two residues (two additional G nucleotides at the 5[prime]-end). The 3[prime]-end-specific oligonucleotide provided a DraI restriction site required for linearization of the template immediately downstream of the three (RoY5 RNA) or four (U6 RNA) U nucleotides, representing the pol III terminator element. In order to in vitro synthesize U6 snRNA molecules with different amounts of 3[prime]-terminating U residues (U6-3, U6-4 and U6-5), the corresponding primer was changed accordingly. The primers used were: T7-U6 upstream, 5[prime]-TTTAATACGACTCACTATAGGGTGCTCGCTTCGGCA-3[prime]; downstream, 5[prime]-TTTTA(3-5)-TATGGAACGCTTCACGAA-3[prime]; T7-RoY5 upstream, 5[prime]-TTTAATACGACTCACTATAGGGAGTTGGTCCGAGTGTTGT-3[prime]; downstream, 5[prime]-TTTTAAACAGCAAGCTAGTCAAG-3[prime]. The double-stranded PCR fragments were cloned into the EcoRV site of the Bluescript KS+ vector (Stratagene). The cDNA of the RoY5 RNA gene (26) has been cloned by RT-PCR (R.Trippe, unpublished results) and the 7S L/VA I hybrid gene (7S L +145) has been described in detail previously (25).

In vitro transcription

An aliquot of 1 µg plasmid DNA was transcribed in 20 µl S100 extract in the presence of 70 mM KCl, 16 mM HEPES-KOH, pH 7.9, 5 mM MgCl2, 1 mM DTT, 0.1 mM EDTA and 12% (v/v) glycerol. Reactions in a final volume of 50 µl were started by addition of nucleotides (0.5 mM each ATP, GTP and CTP and 5 µCi [[alpha]-32P]UTP) and incubated for 1 h at 30°C. Extraction of transcripts and analysis by electrophoresis in denaturing gels was as described (27).

For in vitro transcription with the bacteriophage T7 RNA polymerase, plasmids were linearized with DraI (3[prime]-end) and fragments with the cloned gene constructs recovered after restriction in the upstream polylinker and agarose gel electrophoresis. Purified DNA fragments (50-100 ng) were transcribed essentially as described by Melton et al. (28).

Northern blot hybridization

RNAs were separated in a 2.5% agarose, 2.2 M formaldehyde gel and transferred to a Hybond N membrane (Amersham) as described previously (29). The blot was UV cross-linked for 45 s and then prehybridized for 2 h in 5× SET, 0.1% SDS at 54°C. A 25mer oligonucleotide complementary to U6 RNA (nt +42 to +66), end-labeled with T4 polynucleotide kinase in the presence of [[gamma]-32P]ATP, was used as probe. Hybridization was for 16 h in the same buffer and at the same temperature as above. Washing of the membrane (2 × 15 min in 2× SET, 0.1% SDS) was at room temperature, followed by autoradiography.

Terminal uridylyl transferase assay

The standard TUTase assay was carried out in a final volume of 50 µl with 1 µg cellular RNA or 50 ng T7 RNA polymerase transcripts. The reaction was performed in the presence of 60 mM KCl, 12 mM HEPES-KOH, pH 7.9, 5 mM MgCl2, 2 mM DTT, 0.1 mM EDTA, 12% (v/v) glycerol and 5 µCi [[alpha]-32P]UTP. Following incubation for 1 h at 30°C, RNA was analyzed as above. Cellular RNA (1 mg/ml) was isolated from S100 extracts by standard techniques (using guanidinium isothiocyanate and phenol extraction), dissolved in RNase-free water and stored in aliquots at -70°C.

RESULTS

HeLa cells contain a non-specifc and a U6 RNA-specific terminal uridylyl transferase

HeLa cell extracts are widely used to study the structure and expression of eucaryotic genes by in vitro transcription reactions. In particular, pol III genes (genes transcribed by RNA polymerase III) are rather efficiently transcribed in such extracts, with the appearance of distinct bands demonstrating that accurate initiation as well as termination reactions occurred. In the experiment shown in Figure 1, constructs of a human 7S K RNA gene (lanes 1-3; 30) or a mouse U6 snRNA gene (lanes 4-6; 31) were transcribed in cytoplasmic S100 extracts with [[alpha]-32P]UTP and in the presence of low concentrations of [alpha]-amanitin, to suppress transcription by RNA polymerase II. In addition to the appearance of specific transcripts, these assays revealed an additional labeled band of unknown origin. This major labeled band of an apparent length of ~105 nt (see arrow) was reproducibly observed with different extracts. Furthermore, the appearance of this additional prominent band resulted from a DNA template-independent reaction. In contrast to the transcripts originating from the human 7S K RNA and mouse U6 RNA templates (lanes 1 and 4), labeling of the additional band (arrow) was insensitive to high concentrations (400 µg/ml) of [alpha]-amanitin (lanes 2 and 5) or inactivation of the DNA template by 2 µg/ml actinomycin D (lanes 3 and 6). Furthermore, the same labeled band was also observed if in vitro transcription was performed with a pUC8 vector DNA as template (lane 7). Together, these results indicate that labeling of the RNA in question was not transcriptional. Consequently, this labeling of pre-existing RNA must have been due to a uridylic acid transferase reaction.


Figure 1. In vitro transcription of pol III gene constructs. The human h7SK(6)VA and mouse mU6(26)VA constructs were transcribed in S100 extract in the presence of 400 µg/ml [alpha]-amanitin (lanes 2 and 5) or 2 µg/ml actinomycin D (lanes 3 and 6), with lanes 1, 4 and 7 representing the controls (c) with low concentrations of [alpha]-amanitin (1 µg/ml), to suppress pol II transcription only. Phenol-extracted RNA was analyzed electrophoretically in denaturating (8 M urea) 6% polyacrylamide gels. The arrow indicates the position of in vitro labeled endogenous U6 snRNA. m, labeled DNA marker fragments. Lane 7 shows the reaction without a functional gene, using a pUC8 vector DNA as template.

The occurrence in mammalian cells of non-specific TUTases is clearly evident if phenol-extracted RNA is incubated with cellular extracts in the presence of labeled UTP (12,15). However, an interesting question remaining was why only one such additional labeled RNA band appeared in the in vitro transcription assays if a non-specific transferase was responsible for this result. One possible explanation was, of course, that within these extracts only one endogenous RNA substrate might have been accessible to the transferase. Alternatively, the enzyme transferring UMP residues to pre-existing RNAs of the extract could be specific for only one low molecular weight RNA species. That is why we started to purify the transferase activity by standard chromatographic procedures. After separating S100 extract proteins on conventional phosphocellulose P11 columns (Fig. 2), we observed a broad peak of TUTase activity eluting between 170 and 450 mM KCl (fractions 6-20). In this case, the TUTase activity was analyzed with cytoplasmic RNA as substrate, previously isolated from S100 extracts. As is evident from Figure 2, these fractions contained a non-specific TUTase activity which labeled a broad spectrum of cellular RNA molecules. According to their apparent size, two RNA bands were tentatively designated as 5S and U6 RNA respectively. However, a very different picture emerged if the active fractions of the phosphocellulose column were analyzed after separation on a Superdex G200 column. As shown in Figure 3A, gel filtration of the extract proteins clearly separated two different uridylyl transferase activities. One was associated with the high molecular weight range of the column (fractions 9-17) and represented the non-specific TUTase activity observed before. Again, this activity efficiently labeled a broad spectrum of cellular RNA molecules, which included U6 snRNA (see arrow). In addition, however, a second TUTase activity was found associated with fractions representing proteins of lower molecular mass (fractions 23-27, corresponding to a size of ~60 kDa). This enzyme was highly specific for a single small RNA species, co-migrating with in vitro transcribed mammalian U6 snRNA (see legend to Fig. 3A). Although the entire heterogeneous population of celluar RNA molecules was again present as substrate, only U6 snRNA was labeled in the standard TUTase reaction. This data points to a second transferase with clearly different specificity.


Figure 2. Chromatographic separation of S100 extract proteins by phosphocellulose P11. Aliquots of 75 mg extract proteins were loaded on a 1.6 × 5 cm phosphocellulose column and eluted with a linear gradient (0.1-1 M KCl). Samples of 30 µl of individual fractions (4.5 ml) were assayed for TUTase activity in a standard reaction mixture containing 1 µg exogenous cellular RNA. Analysis of labeled RNA was as in Figure 1. The positions of 5S rRNA and U6 snRNA are indicated by arrows. ft, flow-through fraction.
Figure 3. Analysis of TUTase activities of protein fractions obtained by gel filtration. (A) Proteins associated with the 170-400 mM KCl fractions of the P11 column were concentrated by ammonium sulfate precipitation and applied to a Superdex G200 column. As in Figure 2, individual fractions were analyzed for TUTase activity using exogenous RNA. The position of labeled U6 snRNA is indicated by an arrow and was verified by electrophoresis in parallel of the labeled band obtained with fraction 25 in comparison with mammalian U6 snRNA transcribed in vitro with S100 extracts (data not shown). l, load fraction. (B) TUTase analysis as in (A), however, here 10 µl peak fraction (15) of the non-specific activity of (A) was assayed in combination with another 10 µl active fraction (21, 23, 25, 27 or 29) which revealed a high specificity for U6 snRNA. 15, activity of 10 µl non-specific peak fraction alone. The position of labeled U6 snRNA is again marked.

However, one could not exclude the possibility that the two enzyme activities were generated artificially, for example by dismantling of an enzyme complex by phosphocellulose chromatography. The resulting subunits, subsequently separated by gel filtration, might then represent only residual activities, yet with apparently different specificity. Therefore, both TUTase activities were assayed in combination. As shown in Figure 3B, addition of the peak fraction of the non-specific activity (fraction 15) to fractions containing the specific enzyme (fractions 21-29) did not influence the underlying pattern of the specific TUTase activity. Neither did a shift in peak activity to a different position occur nor an overall increase from the combination of both enzyme activities result. The absence of any synergistic effect or any kind of cooperativity indicates that the two TUTase activities resolved in Figure 3A do indeed represent two separate enzymes. This conclusion is further supported by the finding that separation of the two enzymes can also be achieved on hydroxyapatite columns (R.Trippe, unpublished results), a much milder ion exchange resin than phosphocellulose P11. Together, these results provide firm evidence that HeLa cell extracts contain at least two distinct TUTase activities which reveal different substrate specificities and which can clearly be separated on the basis of their molecular mass.

Recognition of cellular and synthetic U6 RNA by the specific TUTase

In order to substantiate the described differences in specificity of the two TUTase activities, we analyzed both enzymes with different RNA substrates, obtained by in vitro transcription with bacteriophage T7 RNA polymerase. In contrast to applying a heterogeneous population of cellular RNA molecules, the use of defined RNA substrates avoids problems which might arise due to possible competition effects among individual RNA species. For this analysis, gene sequences coding for human 7S L RNA or RoY5 RNA were used in comparison with the U6 snRNA template. We would like to emphasize that this selection was made arbitrarily, because those clones were available in the laboratory. Figure 4A shows the results of such a transferase assay with defined RNA substrates and either the non-specific (lanes 1-4) or the specific enzyme (lanes 5-8). As with total cellular RNA (lane 1), the non-specific enzyme (fraction 15 of Fig. 3A) labeled all individual RNA substrates used in this analysis, i.e. U6 RNA (lane 2), RoY5 RNA (lane 3) and 7S L RNA (lane 4). It should be noted that suppression of labeling by proteins of the non-specific fraction of in vitro synthesized U6 RNA (lane 2) is reproducibly observed with this fraction, very likely due to a newly described U6-specific 3[prime]-exonuclease activity. The nuclease catalyzes an extremely rapid removal of UMP residues from the 3[prime]-end of U6 snRNA and co-elutes from phosphocellulose columns with the TUTases analyzed here (32). This finding also explains the fairly low rate of U6 RNA labeling observed with the P11 fractions in Figure 2, as compared with the non-specific labeling of other small RNAs with the same fractions. Furthermore, the heterogeneous pattern obtained with the RoY5 RNA substrate (Fig. 4A, lane 3) clearly indicates that the non-specific transferase added more than one nucleotide. In contrast to the non-specific enzyme, the specific TUTase activity (fraction 25 of Fig. 3A) was again highly selective for U6 RNA, irrespective of whether total cellular RNA or in vitro transcribed U6 RNA was used (Fig. 4A, lanes 5 and 6). The weaker labeling of T7 transcribed U6 RNA used here (lane 6), as compared with the cellular U6 RNA (lane 5), is due to the four U residues at its 3[prime]-end (see also Fig. 6A, lane 2). The different length of the T7 transcribed U6 RNA is explained by the two additional G residues at its 5[prime]-end, required for efficient initiation of the bacteriophage RNA polymerase (see templates). In contrast to the U6 RNA substrates, the U6-specific TUTase did not label the RoY5 (lane 7) and 7S L RNAs (lane 8), although these RNAs were accepted as substrates by the non-specific enzyme (lanes 3 and 4). Together, these results confirmed the high specificity for U6 snRNA of the newly identified TUTase.

Figure 4. Analysis of the TUTase substrate specificity. (A) The non-specific (lanes 1-4) and specific (lanes 5-8) TUTase activities as obtained from the Superdex G200 column, fractions 15 and 25 respectively, were analyzed with different in vitro synthesized small RNA species, in comparison with cellular RNA (lanes 1 and 5). Synthetic RNAs obtained by in vitro transcription with T7 RNA polymerase (see templates) were: U6 snRNA (lanes 2 and 6); RoY5 RNA (lanes 3 and 7); human 7S L RNA (lanes 4 and 8). (B) Analysis of non-specific (lanes 1-4) and specific (lanes 5-8) TUTase activities with 5 µCi different labeled [[alpha]-32P]nucleoside triphosphate substrates as indicated. The position of the labeled U6 snRNA is marked by an arrow.

To demonstrate that both the non-specific and specific transferases are indeed selective for uridylic acid residues, standard TUTase reactions with cellular RNA were performed with both enzymes, however using different labeled nucleoside triphosphates as substrate. As is evident from Figure 4B, both enzymes were only active in the presence of UTP (lanes 1 and 5). No labeling of cellular RNA was observed with CTP (lanes 2 and 6), GTP (lanes 3 and 7) or ATP (lanes 4 and 8).

U6 RNAs from various eucaryotes, but not fission yeast, are modified by the U6-specific TUTase

Since the novel TUTase enzyme identified here was found to be highly specific for U6 snRNA, we wanted to know whether a similar kind of selectivity was also detectable among U6 RNA molecules from different species. For this, total cellular RNA was isolated from a variety of eucaryotic organisms ranging from yeast to man. Since we did not know the relative amounts of U6 RNA present among total RNA of these organisms, the abundance of U6 RNA was first analyzed. For this, a 25mer oligonucleotide complementary to the most conserved region of U6 snRNA (nt +42 to +66 in mammals) was used as probe in a northern blot analysis. The autoradiogram in Figure 5A demonstrates that with this probe U6 RNA molecules were clearly detectable among total cellular RNA from man (lanes 1 and 8), mouse (lane 2), frog (lane 3), fly (lane 4), worm (lane 5) and fission yeast (lane 6). Therefore, this analysis allows normalization for the relative concentrations of U6 RNA among these species. Subsequently, a standard TUTase reaction with the specific enzyme was performed, applying exactly the same amounts of total cellular RNA from these organisms as used in Figure 5A. The results shown in Figure 5B demonstrate that the specific enzyme was able to accept U6 RNA from mouse (lane 2), frog (lane 3), fly (lane 4) and worm (lane 5) as substrate. The slightly different position of the TUTase-labeled U6 RNA band of Caenorhabditis elegans (Fig. 5B, lane 5) is in agreement with the smaller size of this RNA (102 versus 106 nt; 33). This difference, however, was not resolved in the agarose gel used in Figure 5A. It should be noted that the efficiency of labeling of U6 RNA showed considerable differences among the species analyzed (compare Fig. 5A and B). In particular, with Drosophila U6 RNA (see lanes 4) we reproducibly observe a significant decrease in TUTase labeling which is not understood at present. Furthermore, in contrast to the other eucaryotes analyzed, no U6-specific labeling was observed with total cellular RNA from Schizosaccharomyces pombe as substrate (lane 6). The mature (spliced) version of S.pombe U6 snRNA has been found to be 99 nt in length (34). However, no TUTase-labeled RNA product of that size was obtained here, although the U6 RNA was clearly present in total celluar RNA from fission yeast (see the northern blot analysis in Fig. 5A). Of course, only U6 RNA molecules with a 3[prime]-OH group represent substrates for the addition of uridylic acid residues by the TUTase. Therefore, one might argue that U6 RNA from S.pombe contains only blocked 3[prime]-ends. However, this is not the case. Lund and Dahlberg (18) have shown that in vivo the ratio between blocked and free 3[prime]-OH groups of U6 RNA in S.pombe is ~1:1. That is why we conclude that the absence of specific TUTase labeling with U6 RNA from S.pombe is very likely due to an altered secondary structure of that RNA as compared with other species. The minus RNA control in lane 7 of Figure 5B is to show that no endogenous RNA substrate is associated with the fractions containing the U6-specific TUTase.


Figure 5. Species specificity of TUTase with cellular RNA from different organisms. (A) Northern blot analysis of cellular RNA. RNA isolated from the organisms indicated (36) was separated in a 2.5% agarose, 2.2 M formaldehyde gel, transferred to a nylon membrane, cross-linked and processed as described (29). Total cellular RNA was from HeLa cells (lanes 1 and 8; 1 µg), mouse FM3A cells (lane 2; 2 µg), Xenopus laevis oocytes (lane 3; 6 µg), Drosophila melanogaster Schneider cells (lane 4; 6 µg), Caenorhabditis elegans (lane 5; 3 µg) and Schizosaccharomyces pombe (lane 6; 10 µg) respectively. Lane 7 represents a minus RNA control (-). (B) The same amount of total cellular RNA as in (A) was analyzed in the U6-specific TUTase reaction.

. High resolution gel electrophoresis of different in vitro synthesized U6 RNA molecules after treatment with or without TUTase. (A) Three different templates encoding human U6 RNAs with three (U6-3), four (U6-4) or five (U6-5) U residues at their 3[prime]-ends respectively were transcribed by T7 RNA polymerase in vitro. Samples of 50 ng unlabeled in vitro transcribed U6 RNA were incubated with the specific TUTase and 5 µCi [[alpha]-32P]UTP in a standard reaction (lanes 1-3). For comparison, the same U6 RNAs, however labeled during T7 transcription and not treated with TUTase, are shown in lanes 4-6. To obtain a less heterogeneous RNA template (as is for example evident from lanes 1-6) for the TUTase assay, unlabeled RNA of each T7 RNA polymerase transcription was first separated in high resolution gels and the major band of each reaction recovered (rec.). Subsequently, 50 ng of this purified material were incubated with specific TUTase in the presence of labeled UTP and the products analyzed as before (lanes 7-9). The asterisk (*) is to define in which reaction the RNAs were labeled. (B) Aliquots of 50 ng unlabeled T7 RNA polymerase transcripts were analyzed in high resolution polyacrylamide gels stained with ethidium bromide, either directly (lanes 1-3) or after recovery (rec.) of the main transcripts excised as described before [lanes 7-9; this numbering was chosen to point out that this material was exactly the same as that used for the TUTase reactions in lanes 7-9 of (A)].

U6-specific TUTase restores the four template-encoded UMP residues

Finally, we wanted to see how the 3[prime]-ends of U6 snRNA molecules were modified by the U6-specific TUTase. For this analysis, three different U6 RNA templates, under control of the T7 promoter, were cloned, which differed by the number of T residues at the 3[prime]-end of the transcribed DNA sequence. Upon linearization of the templates with the DraI restriction enzyme (see Materials and Methods), labeling (*) by in vitro transcription with T7 RNA polymerase (Fig. 6A, lanes 4-6) resulted in a population of slightly heterogeneous RNA products. The major band of each reaction represented U6 RNA molecules with three (U6-3), four (U6-4) or five (U6-5) U residues at their 3[prime]-ends respectively. Since all three constructs contain identical 5[prime]-end sequences, the observed heterogeneity must be present at the 3[prime]-end. When equal amounts of unlabeled molecules of such an in vitro transcription reaction were analyzed in a subsequent TUTase reaction with labeled UTP (lanes 1-3), again a slightly heterogeneous population of labeled products was separated by high resolution gel electrophoresis. It is clearly evident, however, that U6-3 RNA (lane 1) was much more efficiently labeled by the TUTase (*) than U6-4 (lane 2) and U6-5 RNA (lane 3), the latter two representing low efficiency substrates for this TUTase reaction. In order to avoid side effects possibly originating from the application of a heterogeneous population of RNA molecules which might obscure these results, we decided to use a more defined substrate for the TUTase reaction. To achieve this, unlabeled in vitro transcribed RNAs were separated in high resolution gels as before and the single major band of each reaction (shown in Fig. 6B) was recovered from the gel. This material (rec.) was again analyzed in the TUTase assay supplemented with labeled UTP. As shown in Figure 6A, lanes 7-9, the results obtained with these defined substrates confirmed the previous observations. Again, U6-3 RNA was the most effective substrate for the TUTase reaction (lane 7), whereas U6-4 (lane 8) and U6-5 RNA (lane 9) represented poor substrates. The weak residual signals still detected in lanes 8 and 9 very likely originate from minor contamination with U6-3 RNA. It cannot be totally excluded that even the clearly defined single bands (as visible in lanes 7-9 of Fig. 6B) after recovery from tightly restricted areas of such gels contain trace amounts of shorter RNA molecules which differ in size by only 1 or 2 nt. This interpretation is consistent with the finding that after gel purification (lanes 8 and 9) labeling of U6-4 and U6-5 RNA species relative to U6-3 was more significantly reduced as if the material was assayed directly (lanes 2 and 3). Therefore, we conclude that the inefficient labeling of U6-4 and U6-5 RNA molecules observed in Figure 6A, lanes 2 and 3, was indeed due to the microheterogeneity of those in vitro synthesized RNA substrates. Thus, it appears that the physiological function of the U6-specific TUTase activity is restricted to restoring the native 3[prime]-end stem structure, which is completed by the four template U residues present in newly transcribed U6 RNA molecules. Consequently, U6 snRNA molecules containing three (or possibly less) U residues at their 3[prime]-ends constitute the physiological substrate for the TUTase enzyme described here. Furthermore, the differences in length, as observed by a comparison of lanes 1-3 with 4-6 of Figure 6A, indicate that in all these cases the TUTase enzyme added only one UMP residue to the RNA species synthesized in vitro.

Figure 6B is to demonstrate that the U6 RNA transcribed in vitro by T7 RNA polymerase indeed shows only minor heterogeneity. Ethidium bromide staining of the transcripts separated in high resolution polyacrylamide gels revealed only one major RNA band (lanes 1-3). Lanes 7-9 show exactly the same amounts of recovered (rec.) material that were applied to the TUTase reaction in lanes 7-9 of Figure 6A. The slightly reduced amount (~40%) of the recovered U6-5 band resulted in a correspondingly decreased activity in lane 9 of Figure 6A, as compared for example with lane 8.

In order to substantiate our conclusion that the TUTase activity identified here catalyzes the restoration of authentic 3[prime]-ends of U6 snRNA molecules, we compared the reaction products obtained with cellular U6 RNA with those obtained with the U6 RNA synthesized by T7 RNA polymerase in vitro (Fig. 7). Since individual 1 nt steps are not readily visible from an autoradiogram of the high resolution gel (upper panel), the PhosphorImager analysis of this gel is also included (lower panel), allowing tracing of individual bands which differ by only 1 nt. As can be seen from the comparison shown in Figure 7, after the TUTase labeling reaction (TUTase*) the position of cellular RNA differs from that of the in vitro synthesized labeled U6-3 RNA (T7*) by a distance which corresponds to an apparent difference in length of 1.5 nt. In agreement with the results of Figure 6A, this difference in length is increased when compared with in vitro synthesized U6-3 RNA subsequently labeled by TUTase (TUTase*) and analyzed as before. The apparent difference in size between TUTase-labeled cellular RNA and the major band of TUTase-labeled U6-3 RNA now corresponds to 2.5 nt. In this context, one has to keep in mind that the RNA transcribed in vitro differs in length from the cellular U6 snRNA by two additional (5[prime]-GG) nucleotides (compare lanes 5 and 6 of Fig. 4A). Taking into account these two additional G residues of the T7 transcripts, the TUTase reaction products obtained with cellular or synthestic RNA show a net difference in migration corresponding to 0.5 nt. One might argue that in this reaction only the transfer of a phosphoryl group was involved. However, chain length analysis in that gel system (Fig. 6A, lanes 4-6) indicates migration effects related to a length difference of exactly 1 nt. Furthermore, comparison of the main U6-3 bands (Fig. 6A, lanes 1 and 4) clearly shows that the TUTase reaction products were elongated by exactly one UMP residue. Therefore, we believe that the net difference in size of 0.5 nt between cellular and synthetic U6 RNA as a consequence of the TUTase reaction reflects a modification of the cellular RNA which is absent from the synthetic T7 transcripts.

Figure 6. Size comparison of the TUTase reaction products obtained with synthetic or cellular U6 RNA. Aliquots of 50 ng unlabeled U6-3 or U6-4 RNA T7 transcripts and 1 µg cellular RNA were incubated with specific TUTase in the presence of labeled UTP. The products were analyzed by high resolution gel electrophoresis in comparison with synthetic U6-3 and U6-4 RNA, which were labeled during the T7 RNA polymerase reaction without subsequent TUTase treatment. (Upper) Autoradiogram of the gel. (Lower) PhosphorImager spectrum of the same gel used for autoradiography above. This image is also provided to allow a distinction between RNA chains which differ in length by only 1 nt. From this image it appears that the labeled TUTase reaction products obtained with cellular RNA migrate at a position corresponding to a band which is 1.5 and 2.5 nt shorter than the main bands of the labeled T7 transcripts obtained with the U6-3 and U6-4 templates respectively.

In summary we conclude that in vivo the physiological substrate of the TUTase characterized here is represented by recessed U6 snRNA molecules which are modified to restore the four 3[prime]-UMP residues present in newly transcribed U6 RNA.

DISCUSSION

Preferential labeling of U6 snRNA molecules in cellular trans-cription extracts by a TUTase activity has been observed in several laboratories (12,15,17,18). The conclusion that this labeling of U6 RNA is the result of a highly selective reaction was mainly based on the observation that in these extracts only a single major product was obtained, the identity of which was confirmed to be U6 snRNA (12). These investigators also excluded the possibility that preferential labeling of this particular RNA species might be due to an enrichment of U6 RNA in those extracts. With the observation that a specific addition of uridylic acid residues to U6 RNA occurred, the question arose whether or not this might simply be a consequence of availability of cellular RNA molecules for this reaction. One could not exclude the possibility that in cellular extracts only the 3[prime]-end of U6 RNA was freely accessible to the transferase enzyme. That conclusion seemed to be supported by the finding that with (phenol-extracted) naked RNA a broad spectrum of cellular RNA molecules was obtained in such a labeling reaction (12,15). Therefore, at least for the non-specific enzyme, the problem of substrate availability is certainly a crucial point. However, the results presented here (Fig. 3) demonstrate that, in addition to a non-specific enzyme, a specific transferase indeed exists in HeLa cells, capable of adding UMP residues to U6 RNA selectively.

A partially purified TUTase activity considered to be specific for U6 snRNA has been described previously (15). However, for two reasons it is extremely unlikely that the enzyme characterized there is identical to the TUTase activity we have identified here. First, as also noted by others (12,17,18; Fig. 1), Tazi et al. (15) obtained selective labeling of U6 RNA in cellular extracts, but, in agreement with all these investigations, their extracts also, when analyzed with phenol-extracted nucleic acids, were found to label a broad spectrum of different cellular RNA molecules. That finding clearly indicated that a non-specific TUTase was observed. Secondly, to trace their transferase activity during the purification steps, Tazi and co-workers (15) used Saccharomyces cerevisiae 5S RNA as substrate for the TUTase reaction. As is evident from the results shown here (see fractions 23-27 in Fig. 3A), the U6-specific TUTase enzyme does not accept 5S RNA as a substrate or any other cellular RNA species, with the exception of U6 snRNA molecules. Thus, in our hands labeling of 5S RNA was only obtained with the non-specific TUTase enzyme. Therefore, a purification scheme based on analysis of 5S RNA modification must have resulted in enrichment of the non-specific enzyme.

It is quite conceivable that the enzyme characterized here exerts its substrate specificity by recognizing, directly or indirectly (via a putative bridging protein), distinct structural elements of its target RNA. In this context, it is interesting to note that fission yeast U6 RNA was not recognized as substrate by the specific TUTase (Fig. 5). Although most regions of that RNA reveal extended homology with the corresponding molecules of higher eucaryotes (9), fission yeast U6 RNA shows considerable sequence diversity affecting its 3[prime]-end stem-loop structure. With the assumption in mind of recognition by structure, one has to postulate that the status of the substrate RNA allows free accessibility of its 3[prime]-end structure.

A number of results have shown that during its life cycle the spliceosomal U6 snRNA undergoes a series of modifying reactions, primarily affecting its 3[prime]-end. Using anti-La antibodies, a high variability in length of the 3[prime]-terminating oligouridylic stretch of the U6 snRNA was observed (18,20). These U6 RNA molecules are characterized by 3[prime]-hydroxyl termini, thought to be required for incorporation into the U4-U6 snRNP complex (15,19). In addition, the vast majority of cellular U6 snRNA molecules have been found to possess an unsual 2[prime],3[prime]-cyclic phosphate (>p) at their 3[prime]-ends (18). It has been suggested that conversion of a U6 snRNA species with a 3[prime]-hydroxyl end into another type with U>p occurs as a consequence of the participation of U6 RNA in pre-mRNA splicing (15,19). Recent results indicate, however, that formation of a phosphate-blocked 3[prime]-end is not a consequence of the involvement of U6 RNA in the splicing reaction (35).

Furthermore, a U6 snRNA-specific nuclease activity has been isolated from HeLa cell nuclear extracts and has recently been characterized in detail (32). This exonuclease has been shown to remove template and non-template (as defined in 32) nucleotides from the 3[prime]-end of human U6 RNA molecules. As with the specific TUTase described here, that nuclease too has been found to specifically recognize a secondary structure located at the 3[prime]-end of the U6 RNA. It should be mentioned that such a nuclease activity was not associated with the partially purified U6-specific TUTase activity analyzed here.

Taken together, the addition and removal of nucleotides implies a sophisticated recycling process for functional U6 snRNA molecules, assuming that these molecules are used more than once. Newly synthesized U6 RNA recruits specific polypeptides to form the U6 snRNP. This complex associates with the U4 snRNP, resulting in formation of the U4-U6 snRNP complex. Up to this point the 3[prime]-end of U6 RNA is not blocked and terminates with a free hydroxyl group. In a subsequent, as yet unidentified, step, however, U6 snRNA acquires a blocked (>p) 3[prime]-end. In order to recycle the unblocked functional form of the RNA, removal of the 3[prime]-cyclic phosphate may well be accomplished by the above-mentioned U6-specific exonuclease (32). Processing of U6 RNA by this nuclease has been shown to remove all terminating UMP nucleotides, including the four template 3[prime] U residues. In a final step of this cycle, the U6-specific TUTase could act to regenerate the correct 3[prime]-terminus of the functional U6 RNA, with a four nucleotide oligo(U) stretch completing the 3[prime]-end stem-loop structure found within newly transcribed U6 RNA.

At present it is not clear, however, which enzyme activity mediates the described 3[prime]-end elongation of U6 RNA with several additional non-template U residues (20). From our experiments (Fig. 6) it appears that the specific enzyme has a clear preference for restoring only the four template U nucleotides present in newly synthesized U6 RNA. Therefore, one possibility might be that this further elongation with non-template nucleotides (32) is a result of the non-specific TUTase activity. This is consistent with our finding that the non-specific enzyme, at least with RoY5 RNA as substrate, can add several U residues to the 3[prime]-end of RNA molecules (Fig. 4A). However, the functional significance of such an addition of non-template UMP residues to U6 RNA remains to be elucidated.

Experiments are in progress to analyze in more detail the structural requirements for the described modifications of U6 snRNA. Furthermore, the availability of monoclonal antibodies directed against the specific TUTase might help to elucidate the physiological significance of 3[prime]-terminating U residues alternately added to and removed from the U6 snRNA.

ACKNOWLEDGEMENTS

We thank Dr Ram Reddy (Houston, TX) for providing the mouse U6 snRNA clone and Halil-Cem Gürsoy for samples of cellular RNA from lower eucaryotes. The expert technical assistance of Nadine Pieda is gratefully acknowledged. Thanks also are due to P.Cichocki for the synthesis of oligonucleotides and to K.Grabert for the photographs. This work was supported by a grant from the Deutsche Forschungsgemeinschaft (Be 531/12-3) to B.J.B.

REFERENCES

1. Moore,M.J., Query,C.C. and Sharp,P.A. (1993) In Gesteland,R.F. and Atkins,J.F. (eds), The RNA World. Cold Spring Harbor Laboratory Press. Cold Spring Harbor, NY, pp. 303-358.

2. Krämer,A. (1996) Annu. Rev. Biochem., 65, 367-409. MEDLINE Abstract

3. Konarska,M.M. and Sharp,P.A. (1986) Cell, 46, 845-855. MEDLINE Abstract

4. Green,M.A. (1991) Annu. Rev. Cell Biol., 7, 559-599. MEDLINE Abstract

5. Guthrie,C. (1991) Science, 253, 157-163. MEDLINE Abstract

6. Lührmann,R., Kastner,B. and Bach,M. (1990) Biochim. Biophys. Acta, 1087, 265-292. MEDLINE Abstract

7. Baserga,S.J. and Steitz,J.A. (1993) In Gesteland,R.F.and Atkins,J.F. (eds), The RNA World. Cold Spring Harbor Laboratory Press. Cold Spring Harbor, NY, pp. 359-381.

8. Madhani,H.D. and Guthrie,C. (1994) Annu. Rev. Genet., 28, 1-26. MEDLINE Abstract

9. Brow,D.A. and Guthrie,C. (1988) Nature, 334, 213-218. MEDLINE Abstract

10. Dahlberg,J.E. and Lund,E. (1988) In Birnstiel,M.L. (ed.), Structure and Function of Major and Minor snRNPs. Springer Verlag, Heidelberg, Germany, pp. 38-70.

11. Kunkel,G.R., Maser,R.L., Calvet,J.P. and Pederson,T. (1986)Proc. Natl. Acad. Sci. USA, 83, 8575-8579. MEDLINE Abstract

12. Reddy,R., Henning,D., Das,G., Harless,M. and Wright,D. (1987)J. Biol. Chem., 262, 75-81. MEDLINE Abstract

13. Moenne,A., Carnier,S., Anderson,G., Margottin,F., Beggs,J. and Sentenac,A. (1990) EMBO J., 9, 271-277. MEDLINE Abstract

14. Singh,R. and Reddy,R. (1989) Proc. Natl. Acad. Sci. USA, 86, 8280-8283. MEDLINE Abstract

15. Tazi,J., Forne,T., Jeanteur,P., Cathala,G. and Brunel,C. (1993) Mol. Cell. Biol., 13, 1641-1650. MEDLINE Abstract

16. Forne,T., Rossi,F., Labourier,E., Antaoine,E., Cathala,G., Brunel,C. and Tazi,J. (1995) J. Biol. Chem., 270, 16476-16481. MEDLINE Abstract

17. Hirai,H., Lee,D.I., Natori,S. and Sekimizu,K. (1988) J. Biochem. (Tokyo), 104, 991-994. MEDLINE Abstract

18. Lund,E. and Dahlberg,J.E. (1992) Science, 255, 327-330. MEDLINE Abstract

19. Terns,M.P., Lund,E. and Dahlberg,J.E. (1992) Mol. Cell. Biol., 12, 3032-3040. MEDLINE Abstract

20. Rinke,J. and Steitz,J.A. (1985) Nucleic Acids Res., 13, 2617-2639. MEDLINE Abstract

21. Filipowicz,W., Konarska,M., Gross,H.J. and Shatkin,A.J. (1983) Nucleic Acids Res., 11, 1405-1418. MEDLINE Abstract

22. Genschik,P., Billy,E., Swianiewicz,M. and Filipowicz,W. (1997) EMBO J., 16, 2955-2967. MEDLINE Abstract

23. Andrews,N.C. and Baltimore,D. (1986) Proc. Natl. Acad. Sci. USA, 83, 221-225. MEDLINE Abstract

24. Weil,P.A., Segall,J., Harris,B., Sun-Yu,N. and Roeder,R.G. (1979)J. Biol. Chem., 254, 6163-6173. MEDLINE Abstract

25. Emde,G., Frontzek,A. and Benecke,B.J. (1997) RNA, 3, 538-549. MEDLINE Abstract

26. Wolin,S.L. and Steitz,J.A. (1983) Cell, 32, 735-744. MEDLINE Abstract

27. Kleinert,H. and Benecke,B.J. (1988) Nucleic Acids Res., 16, 1319-1331. MEDLINE Abstract

28. Melton,D.A., Krieg,P.A., Rebagliati,M.R., Maniatis,T., Zinn,K. and Green,M.R. (1984) Nucleic Acids Res., 12, 7035-7056. MEDLINE Abstract

29. Sambrook,J., Fritsch,E.F. and Maniatis,T. (1989) Molecular Cloning: A Laboratory Manual, 2nd Edn. Cold Spring Harbor Laboratory Press. Cold Spring Harbor, NY.

30. Krüger,W. and Benecke,B.J. (1987) J. Mol. Biol., 195, 31-41. MEDLINE Abstract

31. Ohshima,Y., Okada,N., Tani,T., Itoh,Y. and Itoh,M. (1981) Nucleic Acids Res., 9, 5145-5158. MEDLINE Abstract

32. Booth,B.L.,Jr and Pugh,B.F. (1997) J. Biol. Chem., 272, 984-991. MEDLINE Abstract

33. Guthrie,C. and Patterson,B. (1988) Annu. Rev. Genet., 22, 387-419. MEDLINE Abstract

34. Tani,T. and Ohshima,Y. (1989) Nature, 337, 87-90. MEDLINE Abstract

35. Gu,J., Shumyatsky,G., Makan,N. and Reddy,R. (1997) J. Biol. Chem., 272, 21989-21993. MEDLINE Abstract

36. Chomczynski,P. and Sacchi,N. (1987) Anal. Biochem., 162, 156-159. MEDLINE Abstract

*To whom correspondence should be addressed. Tel: +49 234 700 4232; Fax: +49 234 709 4244; Email: bernd.benecke@rz.ruhr-uni-bochum.de

This page is run by Oxford University Press, Great Clarendon Street, Oxford OX2 6DP, as part of the OUP Journals
Comments and feedback: www-admin{at}oup.co.uk
Last modification: 16 Jun 1998
Copyright©Oxford University Press, 1998.
Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 Nucleic Acids ResHome page
S. B. Patel and M. Bellini
The assembly of a spliceosomal small nuclear ribonucleoprotein particle
Nucleic Acids Res., November 1, 2008; 36(20): 6482 - 6493.
[Abstract] [Full Text] [PDF]

Home page
 RNAHome page
K. Licht, J. Medenbach, R. Luhrmann, C. Kambach, and A. Bindereif
3'-cyclic phosphorylation of U6 snRNA leads to recruitment of recycling factor p110 through LSm proteins
RNA, August 1, 2008; 14(8): 1532 - 1538.
[Abstract] [Full Text] [PDF]

Home page
 Mol. Cell. Biol.Home page
O. S. Rissland, A. Mikulasova, and C. J. Norbury
Efficient RNA Polyuridylation by Noncanonical Poly(A) Polymerases
Mol. Cell. Biol., May 15, 2007; 27(10): 3612 - 3624.
[Abstract] [Full Text] [PDF]

Home page
 RNAHome page
R. Trippe, E. Guschina, M. Hossbach, H. Urlaub, R. Luhrmann, and B.-J. Benecke
Identification, cloning, and functional analysis of the human U6 snRNA-specific terminal uridylyl transferase
RNA, August 1, 2006; 12(8): 1494 - 1504.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
I. Aphasizheva, R. Aphasizhev, and L. Simpson
RNA-editing Terminal Uridylyl Transferase 1: IDENTIFICATION OF FUNCTIONAL DOMAINS BY MUTATIONAL ANALYSIS
J. Biol. Chem., June 4, 2004; 279(23): 24123 - 24130.
[Abstract] [Full Text] [PDF]

Home page
 RNAHome page
E. DUMPELMANN, H. MITTENDORF, and B.-J. BENECKE
Efficient transcription of the EBER2 gene depends on the structural integrity of the RNA
RNA, April 1, 2003; 9(4): 432 - 442.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
R. Aphasizhev and L. Simpson
Isolation and Characterization of a U-specific 3'-5'-Exonuclease from Mitochondria of Leishmania tarentolae
J. Biol. Chem., June 8, 2001; 276(24): 21280 - 21284.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Print PDF (190K) Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (21)
Right arrowRequest Permissions
Right arrow Commercial Re-use Guidelines
for Open Access NAR Content
Google Scholar
Right arrow Articles by Trippe, R.
Right arrow Articles by Benecke, B. J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Trippe, R.
Right arrow Articles by Benecke, B. J.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?