Previous Article | Next Article ![]()
Molecular and Cellular Biology, August 2006, p. 6016-6023, Vol. 26, No. 16
0270-7306/06/$08.00+0 doi:10.1128/MCB.02347-05
Copyright © 2006, American Society for Microbiology. All Rights Reserved.
J. David Barrass,
Alastair P. Droop,
Christophe Dez, and
Jean D. Beggs*
Wellcome Trust Centre for Cell Biology, University of Edinburgh, King's Buildings, Mayfield Road, Edinburgh EH9 3JR, United Kingdom
Received 8 December 2005/ Returned for modification 30 January 2006/ Accepted 30 May 2006
|
|
|---|
|
|
|---|
Upon completion of the splicing reaction, the spliced exon RNA (mRNA) is released and the postsplicing ribonucleoprotein complex dissociates in an active process that involves two members of the ATP-dependent DExH box RNA helicase family, Prp22 and Prp43 (3, 26, 32). Prp22 is needed for release of the spliced exons (32, 43), while Prp43 is required for disassembly of the spliceosome and release of the excised intron in its branched, lariat form (26). The U4 snRNP reassociates with the U6 snRNP (31, 41) to form the U4/U6 di-snRNP that will then join the U5 snRNP to form the U4/U6.U5 tri-snRNP once more.
Although many protein components of the splicing machinery have been identified through their physical or genetic interactions with known splicing factors, the functions of a large number of these remain to be determined. For example, the name SPP382 (S. Pandit and B. C. Rymond, unpublished results; http://db.yeastgenome.org/cgi-bin/locus.pl?locus=spp382) was given to Saccharomyces cerevisiae open reading frame YLR424W to indicate its ability to suppress the prp38-1 mutation that causes a defect in spliceosome maturation (45); however, the function of the Spp382 protein is unknown. Spp382 has been reported to associate directly or indirectly with many protein components of the splicing machinery, including Prp43 and components of the U5 snRNP (1, 7, 13, 16), and recently the name Ntr1 (nineteen complex related) was proposed because of its interaction with the NTC (38). In addition, Spp382/Ntr1 interacts with Ykr022c/Ntr2 (16, 17, 38, 39).
We demonstrate here that Ntr1 and Ntr2 are spliceosome associated and coprecipitate mainly excised intron from an in vitro splicing reaction mixture. Under nonsplicing conditions, Ntr1 and Ntr2 coprecipitate the U2, U5, and U6 snRNAs. Northern and microarray analyses show increased levels of pre-mRNAs in cells depleted of Ntr1, and in vitro, Ntr1-depleted extract has low splicing activity and accumulates excised intron in a postsplicing complex that includes Prp8 and Cef1. Glycerol gradient analysis of Ntr1-depleted extract suggests a defect in recycling snRNPs, and the U5 snRNP protein Prp8 is shown by immunoprecipitation to be associated with eightfold more U2 snRNA and with U6 snRNA that is not complexed with U4 snRNA, supporting its accumulation in the U2/U5/U6 postsplicing complex. The intron release factor Prp43 cofractionates with Ntr1 and Ntr2, and in the absence of Ntr1, Prp43 is unable to associate with spliceosomes to release the excised intron. These results complement and extend the recently reported results of Tsai et al. (38), demonstrating that Ntr1 and Ntr2 function in the Prp43-mediated release of the excised intron. We propose that Ntr1 acts as a spliceosome receptor or RNA-targeting factor to promote interaction of Prp43 with the excised intron, possibly assisted by the Ntr2 protein.
|
|
|---|
|
View this table: [in a new window] |
TABLE 1. Plasmids and yeast strains used in this study
|
|
View this table: [in a new window] |
TABLE 2. Deoxyoligonucleotides used in this study
|
-32P]ATP to produce a uniformly labeled splicing substrate (29). Glycerol gradient analysis. For glycerol gradient analysis (essentially as described in reference 4), 80 µl of splicing extract (without added ATP or pre-mRNA) was diluted to 200 µl with 120 µl of GG buffer (20 mM HEPES [pH 7.0], 100 mM KCl, 0.2 mM EDTA) and layered onto an 11-ml 10 to 30% glycerol gradient containing GG buffer. After centrifugation at 37,000 rpm for 17 h in an SW40 Ti rotor (Beckman) at 4°C, 400-µl fractions were collected and stored at 70°C. Alternate fractions were analyzed by Western or Northern blotting.
Immunoprecipitation. Immunoprecipitations were performed as described previously (37), with IPP150 buffer (6 mM HEPES [pH 7.9], 150 mM NaCl, 5 mM MgCl2, 0.1% [vol/vol] Nonidet P-40) and with immunoglobulin G (IgG) agarose (Sigma)-, antihemagglutinin (anti-HA) agarose (mouse anti-HA [F-7]; Santa Cruz Biotechnology), anti-His agarose (rabbit anti-His [H-15]; Santa Cruz Biotechnology)-, or protein A-Sepharose-bound antibodies, i.e., anti-Prp8 rabbit polyclonal antibodies (raised against a peptide corresponding to amino acids 2 to 34 of Prp8; this lab), anti-HA antibodies (Santa Cruz Biotechnology), affinity-purified anti-Prp43 antibodies raised against a peptide corresponding to amino acids 714 to 727 or obtained as a gift from Y. Henry (20). For immunoprecipitation of spliceosomes, splicing reaction mixtures (10 µl for splicing activity or 50 µl for immunoprecipitation) were incubated at 23°C for 25 min and the reaction products were fractionated on a 7% (wt/vol) polyacrylamide-8 M urea gel and visualized by autoradiography of the dried gel. For immunoprecipitation of snRNPs (see Fig. 1A and 3C), splicing extracts were used but without addition of ATP or pre-mRNA.
![]() View larger version (80K): [in a new window] |
FIG. 1. Ntr1 and Ntr2 associate with snRNPs and excised intron RNA. (A) Coprecipitation of snRNAs with Ntr1 and Ntr2. Splicing extracts (50 µl) with nontagged proteins (BMA38a; lane 4), Ntr2-TAP (KL2T; lane 5), or Ntr1-TAP (KL4T; lane 6) were mixed with an equal volume of precipitation buffer containing IgG-agarose beads and incubated at 4°C for 2 h. The beads were washed in buffer containing 150 mM NaCl and deproteinized, and the RNAs were precipitated, fractionated in an 8% polyacrylamide-8 M urea gel, blotted, and probed for U1, U2, U4, U5, and U6 snRNAs. Total RNA extracted from 12.5 µl of each extract is shown in lanes 1 to 3. (B) Coprecipitation of splicing complexes by Ntr1 and Ntr2. Splicing extracts derived from cultures with nontagged proteins (BMA38a; lanes 1 and 6), Prp8-TAP (RG8T; lanes 2 and 7), Ntr1-TAP (KL4T; lanes 3 and 8), Ntr2-TAP (KL2T; lanes 4 and 9), and Ntr2-TAP with HA-Ntr1 depleted (KL4G2T cells grown in glucose for 10 h; lanes 5 and 10) were incubated under splicing conditions in vitro (60-µl total volume) with 32P-labeled ACT1 RNA as the substrate. The reactions were stopped after 25 min, and 10-µl volumes were removed as splicing controls (not shown). The remaining 50-µl volumes were mixed with IgG-agarose beads and incubated at 4°C for 2 h. Beads were washed in buffer containing IPP150 and deproteinized, and the RNAs were precipitated. The unbound samples were also deproteinized, and 20% aliquots were analyzed (lanes 1 to 5) alongside the precipitates (lanes 6 to 10) by fractionation in a 7% polyacrylamide-8 M urea gel and autoradiography. The various RNA species are indicated diagrammatically to the right of panel B, with rectangles representing exons, a thin line representing the intron, and a lariat loop representing the branched form of the intron following the first catalytic step of splicing. IP, immunoprecipitation.
|
![]() View larger version (63K): [in a new window] |
FIG. 3. Depletion of Ntr1 leads to a redistribution of snRNP complexes. To examine the profile of snRNP distribution, non-Ntr1-depleted (A) and Ntr1-depleted (B) cell extracts (no added ATP or pre-mRNA) were centrifuged through 10 to 30% glycerol gradients and RNAs were extracted from alternate fractions and analyzed by Northern blotting with hybridization with probes for U1, U2, U4, U5, and U6 snRNAs. The sedimentation of snRNP complexes is indicated. In addition to U4/U6.U5 tri-snRNPs and U5 snRNPs, fractions 1 to 7 may contain some endogenous spliceosome and postsplicing complexes. (C) Coprecipitation of snRNAs by anti-Prp8 antibodies in extracts from Ntr1-depleted (lanes 4 to 6) or nondepleted (lanes 1 to 3) cells. RNAs were analyzed by fractionation in a 6% polyacrylamide-8 M urea gel and Northern blotting. Preimmune (PI) serum was used as a background control. (D) Same as panel C, but the precipitated snRNAs were analyzed by nondenaturing gel electrophoresis as described previously (41). IP, immunoprecipitation.
|
Northern analysis of RNAs. RNAs were deproteinized by sodium dodecyl sulfate-proteinase K treatment, followed by phenol-chloroform-isoamyl alcohol extraction and fractionated in a 6% (wt/vol) polyacrylamide-8 M urea gel. The sn(o)RNAs were detected by Northern analysis with end-labeled deoxyoligonucleotides complementary to U1, U2, U4, U5, and U6 or U3 (Table 2). The amounts of radioactivity in RNA bands were quantified with a Storm 860 PhosphorImager and ImageQuant software (Molecular Dynamics). The amounts in precipitates were normalized against the input amounts.
|
|
|---|
Depletion of Ntr1. The NTR1 gene is essential for cell viability. Therefore, to investigate the requirement for Ntr1 protein for splicing, yeast strain KL4G, in which chromosomal NTR1 is transcribed under the control of the glucose-repressible PGAL1 promoter, was grown in galactose medium and then shifted to glucose medium. As a control, a second culture was treated identically but was transferred back to galactose medium. Approximately 2 h after the metabolic shift, the growth of the culture slowed considerably and then the cells continued to grow slowly (approximately 4-h doubling time; data not shown). Northern analysis of extracted RNA showed a barely detectable amount of unspliced pre-U3 RNA 2 h after the shift, with the amount increasing for 6 to 10 h postshift (Fig. 2A, lanes 6 to 10). Microarray analysis showed a mild splicing defect for some transcripts of intron-containing genes 2 h after the shift and a substantial genomewide accumulation of unspliced pre-mRNAs by 10 h postshift (data not shown). Extract from KL4G cells grown for 10 h in glucose contained a barely detectable amount of Ntr1 (Fig. 2B, lane 2) and, when incubated under in vitro splicing conditions with an ACT1 substrate RNA, produced only a small amount of spliced exons but accumulated a substantial amount of excised intron (Fig. 2C, lane 6) compared to nondepleted extract (Fig. 2C, lane 3). Incubation of the reaction mixtures with anti-Prp8 (lanes 8 and 9) or anti-Cef1 (lanes 10 and 11) antibodies showed that in Ntr1-depleted extract the excised intron is associated with both Prp8 and Cef1, supporting its accumulation in a postsplicing complex. When KL4G2T (same as KL4G but with TAP-tagged Ntr2) cells were treated similarly by growth in glucose for 10 h to deplete Ntr1, TAP-Ntr2 coprecipitated a substantial amount of excised intron from a splicing reaction mixture (Fig. 1B, lane 10). Therefore, Ntr2 is able to associate with the excised intron complex despite the substantial depletion of Ntr1. The difference in the amounts of the spliced exon and excised intron species in the Ntr1-depleted reaction mixture is most likely due to inhibition of splicing as a consequence of the retention of essential splicing factors within a postsplicing complex, in addition to which the excised intron RNA in the complex is protected against degradation (26). This indicates that Ntr1 is required for efficient pre-mRNA splicing and for metabolism of excised intron and suggests that Ntr1 might play a role in recycling the postspliceosome components for new rounds of splicing.
![]() View larger version (64K): [in a new window] |
FIG. 2. Effects of Ntr1 depletion on splicing in vitro. (A) Northern analysis of U3 snoRNA. BMA38 (lanes 1 to 5) or KL4G (PGAL1-3HA-NTR1; lanes 6 to 10) cells were grown in YPGR (galactose medium) to an OD600 of 0.5 and then transferred to YPD (glucose medium; repressing conditions). RNA was extracted after 0, 2, 4, 6, or 10 h following the shift, subjected to denaturing polyacrylamide gel electrophoresis, blotted, and probed for U3A snoRNA. (B and C) KL4G cells were grown in YPGR to an OD600 of 0.5 and then transferred to YPD and grown for 10 h before harvesting to produce a splicing extract. A nondepleted control culture of KL4G was produced by similar treatment but with a return to YPGR instead of YPD. (B) Western blot assay showing the relative amounts of HA-tagged Ntr1 in KL4G cells at 0 h (lane 1) and at 10 h (lane 2) after the shift to YPD. Nop1 was probed as a loading control. (C) In vitro splicing assays with extracts derived from Ntr1-depleted (lanes 4 to 6) and nondepleted (lanes 1 to 3) cells. Aliquots (5 µl) were withdrawn and halted after incubation for 0, 10, and 25 min at 23°C. The RNAs were analyzed by fractionation in a 7% polyacrylamide-8 M urea gel and autoradiography. Splicing reaction mixtures with Ntr1-depleted (lanes 9 and 11) and nondepleted (lanes 7, 8, and 10) extracts were incubated with preimmune (PI) serum (lane 7), anti-Prp8 antibodies (lane 8 and 9), or anti-Cef1 antibodies (lanes 10 and 11), and precipitated RNAs were extracted and analyzed as described above. The various RNA species are indicated diagrammatically on the right as described in the legend to Fig. 1. IP, immunoprecipitation.
|
Next, the association of snRNAs with Prp8 in Ntr1-depleted and nondepleted extracts was examined by immunoprecipitation. As expected, anti-Prp8 antibodies efficiently pulled down U5, U4, and U6 snRNAs from nondepleted extract (Fig. 3C, lane 3) because of its presence in U5 snRNP and U4/U6.U5 tri-snRNP (15). With Ntr1-depleted extract, immunoprecipitation of Prp8 pulled down an eightfold increased amount of U2 (compared to nondepleted extract and normalized against the total amounts in each extract) but only half the amount of U4 (Fig. 3C, lane 6) and an equivalent amount of U6. The increased association of Prp8 with a U2 snRNA-containing complex and decreased association with U4 snRNP in Ntr1-depleted extract supports the accumulation of Prp8 in the U2/U5/U6 excised intron-containing complex. These precipitates were also analyzed by nondenaturing gel electrophoresis to distinguish free U6 snRNA from a U4/U6 heterodimer that is derived from U4/U6.U5 tri-snRNPs following phenol extraction (nota bene, Prp8 is not a component of U4/U6 di-snRNPs or free U6 snRNPs). As shown in Fig. 3D, the Prp8 precipitate from Ntr1-depleted extract contained less U4/U6 heterodimer and more free U6 snRNA (presumably derived from the U2/U5/U6 complex) than the Prp8 precipitate from nondepleted extract. This supports the conclusion from the glycerol gradients that Ntr1 depletion results in a defect in recycling of snRNPs and indicates that Prp8 accumulates in a U2/U5/U6 snRNP complex.
Ntr1, Ntr2, and Prp43 cofractionate in at least two distinct complexes. Ntr2, Prp8, and the excised intron release factor Prp43 are among many proteins that have been found in affinity-selected complexes along with Ntr1 (13, 16). To investigate this further, extract from KL4G2T cells containing HA-Ntr1 and Ntr2-TAP was fractionated in a glycerol gradient and alternate fractions were analyzed by precipitation of Ntr2-TAP with IgG-agarose and detection of precipitated proteins by Western blotting. Interestingly, HA-Ntr1 was associated with Ntr2-TAP in most of the gradient fractions (Fig. 4A, fractions 1 to 23), whereas Prp8 was associated with Ntr2-TAP only in the higher-density fractions (fractions 3 to 7) that contain high-molecular-weight complexes. Similarly, glycerol gradient fractionation of extract from KL4G cells (HA-tagged Ntr1 and nontagged Ntr2), followed by precipitation with anti-HA antibodies, showed coprecipitation of Prp8 and of Prp43 with HA-Ntr1 in the same high-density fractions (3 to 7) (Fig. 4B). Therefore, as Ntr2 pulls down Ntr1 and Prp8 and Ntr1 pulls down Prp8 and Prp43 in the same high-density fractions, it is most likely that these four proteins are present in the same high-molecular-weight complex. In addition, it is notable that the bulk of Ntr1-associated Ntr2 and Prp43 occurs in lower-density fractions (Fig. 4A and B, fractions 13 to 21). This likely represents an Ntr1/Ntr2/Prp43 complex that is not snRNP associated. Also, Ntr2-TAP coprecipitated both HA-Ntr1 and Prp8 in extract from galactose-grown (nondepleted) KL4G2T (PGAL-HA-Ntr1 and Ntr2-TAP) cells and coprecipitated Prp8 in extract from glucose-grown KL4G2T cells, despite the depletion of Ntr1 (Fig. 4C, lanes 5 and 6). Thus, although the level of Prp8 that coprecipitated with Ntr2 was lower in Ntr1-depleted extract, the association of Ntr2 and Prp8, whether direct or indirect, is not absolutely Ntr1 dependent, as is the case for its association with excised intron (Fig. 1B).
![]() View larger version (51K): [in a new window] |
FIG. 4. Associations among Ntr1, Ntr2, Prp43, and Prp8. (A and B) Glycerol gradient fractionation (without added ATP or pre-mRNA). (A) Extract from galactose-grown KL4G2T cells containing HA-Ntr1 and Ntr2-TAP was fractionated in a 10 to 30% glycerol gradient. Alternate gradient fractions were immunoprecipitated with IgG-agarose, and the precipitated proteins were analyzed by Western blotting with anti-HA and anti-Prp8 (8.6) antibodies. (B) KL4G cell extract containing HA-Ntr1 was fractionated in a 10 to 30% glycerol gradient. Alternate gradient fractions were immunoprecipitated with anti-HA agarose, and the precipitated proteins were analyzed by Western blotting with affinity-purified anti-Prp43 antibodies and with anti-Prp8 antibodies. (C) Extracts (without added ATP or pre-mRNA) from galactose-grown (lanes 2 and 5) or glucose-grown (lanes 3 and 6) KL4G2T cells or from BMA38a (untagged control) cells (lanes 1 and 4) were incubated with IgG-agarose to precipitate Ntr2-TAP, and the precipitated proteins were analyzed by Western blotting with anti-HA and anti-Prp8 antibodies. Note that the Ntr2-TAP protein in whole extract is very poorly recognized by IgG on the blot. IP, immunoprecipitation.
|
![]() View larger version (36K): [in a new window] |
FIG. 5. Immunoprecipitation (IP) of splicing reaction mixtures shows that Ntr1 is required for association of Prp43 with the excised intron. Ntr1-depleted (lanes 2, 3, 5, and 6) and nondepleted (lanes 1 and 4) extracts were prepared from KL4G cells grown as described in the legend to Fig. 2. Splicing reaction mixtures (50 µl) were assembled with 32P-labeled ACT1 substrate RNA and 0.9 µg of purified recombinant Prp43T123A. For reconstitution of the Ntr1-depleted extract (lanes 3 and 6), the splicing reaction mixture was mixed with 3HA-Ntr1 affinity selected from galactose-grown KL4G cell extract (overproducing 3HA-Ntr1). Following incubation for 25 min at room temperature, 90% of each splicing reaction mixture was mixed with 20 µl of affinity-purified anti-Prp43 antibodies immobilized on Sepharose CL45 beads in IPP150. After mixing at 4°C for 1 h, the beads were washed and the RNA was extracted. Precipitated RNA (lanes 4 to 6) and RNA extracted from 10% aliquots of the total reaction mixtures (lanes 1 to 3) were separated on a 7% polyacrylamide-8 M urea gel. For each reconstitution reaction mixture, 10 µl of extract from YPGR-grown KL4G cells was incubated with anti-HA-agarose beads for 1 h at 4°C and the beads were washed five times with IPP150 buffer and once in 0.6 M potassium phosphate buffer (pH 7) and then incubated with 0.5 µg of HA peptide (Sigma) for 8 h at 4°C to elute the HA-Ntr1 from the beads. Recombinant His10-tagged Prp43T123A protein was produced in Escherichia coli BL21-CodonPlus(DE3)RIL (Stratagene) and purified essentially as described by Martin et al. (26). The various RNA species are indicated diagrammatically on the right as described in the legend to Fig. 1.
|
|
|
|---|
Taken together, these data provide compelling evidence that Ntr1 and Ntr2 are present in the late-stage excised intron-containing spliceosome complex and that Ntr1 is required for the interaction of Prp43 with the excised intron prior to its release and recycling of the splicing factors. While the manuscript was under review, Tsai et al. (38) published data showing that metabolic depletion of Ntr2 resulted in accumulation of pre-mRNA and excised lariat-intron in vivo and that a dimeric complex of Ntr1 and Ntr2 is required for dissociation of spliceosomes and release of lariat-intron in vitro. The data presented here complement and extend that study, showing that Ntr1 is required both in vivo and in vitro for efficient splicing and for Prp43 to function in release of the excised intron and recycling of snRNPs. Tsai et al. (38) found that immunodepletion of Ntr1 quantitatively removed Ntr2 from cell extract and vice versa, indicating that these two proteins form a stable complex. We have shown that, following metabolic depletion of Ntr1, Ntr2 is still present in a complex with Prp8 (Fig. 4C) and with excised intron (Fig. 1B), indicating that Ntr2 can associate with late splicing complexes independently of Ntr1 but is not by itself sufficient to activate the intron release function of Prp43.
Ntr1 was originally named Spp382 after the role of this protein in suppression of the prp38-1 temperature-sensitive growth defect. Prp38 is a tri-snRNP-specific protein that was shown to be dispensable for spliceosome assembly but required for spliceosome maturation (6, 45). The depletion of Prp38 caused a spliceosome maturation defect with accumulation of U4/U6, and the temperature-sensitive prp38-1 mutant strain accumulated unspliced pre-mRNA at the nonpermissive temperature (45). The suppression of prp38-1 by Spp382/Ntr1 may suggest an additional role for this protein at an earlier stage, during spliceosome maturation. Alternatively, ntr1 mutations may alleviate the prp38-1 defect indirectly, for example, by moderating the interaction of Ntr1 with a common interacting factor or altering the rate of spliceosome assembly by affecting the availability of splicing factors.
Near its N terminus, Ntr1 contains a G-patch region (glycine-rich sequence found in nucleic binding proteins [2], amino acids 61 to 106) that is the most highly conserved region of this protein. The mammalian G-patch protein called TFIP11 (tuftelin-interacting protein) is a possible ortholog of Ntr1. Human TFIP11 has been found in affinity-purified spliceosomes (18, 46), and mouse TFIP11 was reported to affect the alternative splicing of an adenovirus E1A reporter transcript in a transfection assay (44).
Interestingly, yeast Spp2, another G-patch protein and spliceosome component, interacts with the ATP-dependent DExH-box splicing factor Prp2 (33). As this interaction involves the G-patch sequence in Spp2 and is required for the recruitment of Prp2 to the spliceosome prior to the first catalytic step of splicing, it was proposed that Spp2 may be an accessory factor that confers spliceosome specificity on Prp2 (33). Thus, Ntr1 appears to be an analogous accessory factor for Prp43, targeting it to the excised intron complex, in addition to which it might conceivably be a cofactor for Prp43 helicase activity. Tsai et al. (38) showed that the G-patch region of Ntr1 interacted with Prp43 in a two-hybrid assay. Thus, Ntr1 might promote the interaction of Prp43 with the excised intron through an RNA binding activity of the G-patch region. The role of Ntr2 in these proposed functions remains to be determined.
Intriguingly, Prp43 was recently shown to function also in ribosome synthesis and to associate both with ribosome precursor complexes and with mature ribosomes (12, 20, 21). It is not known how Prp43 is targeted to these distinct cellular processes or whether this dual function allows Prp43 to coordinate the regulation of ribosome biogenesis with pre-mRNA splicing, a process that is vital for ribosomal protein synthesis in budding yeast as many ribosomal protein genes contain an intron. It will be interesting to determine whether Ntr1 acts as an accessory factor for both of these processes or interacts with only a fraction of Prp43, conferring splicing specificity and possibly regulating this in response to different metabolic requirements of the cell.
This work was funded by studentships from The Darwin Trust of Edinburgh to K.L.B. and T.A. and by Wellcome Trust grant 067311 to J.D.B. G.E.-G. is supported by NIH grant R15 GM072622. C.D. was supported by an EMBO fellowship. J.D.B. is a Royal Society Darwin Trust Professor.
Permanent address: W. M. Keck Science Center of the Claremont Colleges, Claremont, CA 91711. ![]()
Present address: Department of Biology (Area 7), University of York, York YO10 5YW, United Kingdom. ![]()
|
|
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»