Purified box C/D snoRNPs are able to reproduce site-specific 2'-O-methylation of target RNA in vitro

Small nucleolar RNAs (snoRNAs) are associated in ribonucleoprotein particles localized to the nucleolus (snoRNPs). Most of the members of the box C/D family function in directing site-specific 2'-O-methylation of substrate RNAs. Although the selection of the target nucleotide requires the antisense element and the conserved box D or D' of the snoRNA, the methyltransferase activity is supposed to reside in one of the protein components. Through protein tagging of a snoRNP-specific factor, we purified to homogeneity box C/D snoRNPs from the yeast Saccharomyces cerevisiae. Mass spectrometric analysis demonstrated the presence of Nop1p, Nop58p, Nop56p, and Snu13p as integral components of the particle. We show that purified snoRNPs are able to reproduce the site-specific methylation pattern on target RNA and that the predicted S-adenosyl-L-methionine-binding region of Nop1p is responsible for the catalytic activity.     

Domains of human U4atac snRNA required for U12-dependent splicing in vivo

U4atac snRNA forms a base-paired complex with U6atac snRNA. Both snRNAs are required for the splicing of the minor U12-dependent class of eukaryotic nuclear introns. We have developed a new genetic suppression assay to investigate the in vivo roles of several regions of U4atac snRNA in U12-dependent splicing. We show that both the stem I and stem II regions, which have been proposed to pair with U6atac snRNA, are required for in vivo splicing. Splicing activity also requires U4atac sequences in the 5' stem-loop element that bind a 15.5 kDa protein that also binds to a similar region of U4 snRNA. In contrast, mutations in the region immediately following the stem I interaction region, as well as a deletion of the distal portion of the 3' stem-loop element, were active for splicing. Complete deletion of the 3' stem-loop element abolished in vivo splicing function as did a mutation of the Sm protein binding site. These results show that the in vivo sequence requirements of U4atac snRNA are similar to those described previously for U4 snRNA using in vitro assays and provide experimental support for models of the U4atac/U6atac snRNA interaction.     

Tumour class prediction and discovery by microarray-based DNA methylation analysis

Aberrant DNA methylation of CpG sites is among the earliest and most frequent alterations in cancer. Several studies suggest that aberrant methylation occurs in a tumour type-specific manner. However, large-scale analysis of candidate genes has so far been hampered by the lack of high throughput assays for methylation detection. We have developed the first microarray-based technique which allows genome-wide assessment of selected CpG dinucleotides as well as quantification of methylation at each site. Several hundred CpG sites were screened in 76 samples from four different human tumour types and corresponding healthy controls. Discriminative CpG dinucleotides were identified for different tissue type distinctions and used to predict the tumour class of as yet unknown samples with high accuracy using machine learning techniques. Some CpG dinucleotides correlate with progression to malignancy, whereas others are methylated in a tissue-specific manner independent of malignancy. Our results demonstrate that genome-wide analysis of methylation patterns combined with supervised and unsupervised machine learning techniques constitute a powerful novel tool to classify human cancers.     

Does delivery volume of family physicians predict maternal and newborn outcome?

Background

The number of births attended by individual family physicians who practice intrapartum care varies. We wanted to determine if the practice–volume relations that have been shown in other fields of medical practice also exist in maternity care practice by family doctors.

Methods

For the period April 1997 to August 1998, we analyzed all singleton births at a major maternity teaching hospital for which the family physician was the responsible physician. Physicians were grouped into 3 categories on the basis of the number of births they attended each year: fewer than 12, 12 to 24, and 25 or more. Physicians with a low volume of deliveries (72 physicians, 549 births), those with a medium volume of deliveries (34 physicians, 871 births) and those with a high volume of deliveries (46 physicians, 3024 births) were compared in terms of maternal and newborn outcomes. The main outcome measures were maternal morbidity, 5-minute Apgar score and admission of the baby to the neonatal intensive care unit or special care unit. Secondary outcomes were obstetric procedures and consultation patterns.

Results

There was no difference among the 3 volume cohorts in terms of rates of maternal complications of delivery, 5-minute Apgar scores of less than 7 or admissions to the neonatal intensive care unit or the special care unit, either before or after adjustment for parity, pregnancy-induced hypertension, diabetes, ethnicity, lone parent status, maternal age, gestational age, newborn birth weight and newborn head circumference at birth. High- and medium-volume family physicians consulted with obstetricians less often than low-volume family physicians (adjusted odds ratio [OR] 0.586 [95% confidence interval, CI, 0.479–0.718] and 0.739 [95% CI 0.583–0.935] respectively). High- and medium-volume family physicians transferred the delivery to an obstetrician less often than low-volume family physicians (adjusted OR 0.668 [95% CI 0.542–0.823] and 0.776 [95% CI 0.607–0.992] respectively). Inductions were performed by medium-volume family physicians more often than by low-volume family physicians (adjusted OR 1.437 [95% CI 1.036–1.992].

Interpretation

Family physicians'' delivery volumes were not associated with adverse outcomes for mothers or newborns. Low-volume family physicians referred patients and transferred deliveries to obstetricians more frequently than high- or medium-volume family physicians. Further research is needed to validate these findings in smaller facilities, both urban and rural.More than 20 years ago, Luft and associates¹ conducted one of the earliest volume–outcome studies. Since then, many studies addressing the relation between volume of procedures and patient outcomes have been published.^2,3 In some of these studies, either the hospital size or the physician procedural volume was used as a surrogate for physician expertise. Among studies analyzing hospital volumes and outcomes, better outcomes have been associated with higher patient volumes in some instances^4,5,6,7 but not others.^3,8,9 Some studies of individual provider volume have shown a positive relation between volume and outcomes,^10,11 whereas others have shown no relation or inconsistent results.^3,12 Finally, a few studies analyzing both hospital volume and provider volume have reported a positive volume–outcome relation.^13,14Criticism levelled at the methods used in volume–outcome studies have addressed the lack of adjustment for case mix, different cutoff points for volume categories and retrospective design.³ Other factors that have an effect on patient outcomes but that have not been included in previous volume analyses include health maintenance organization status, physician certification and years since graduation, and patient socioeconomic status, age and ethnicity. Furthermore, most of the studies on volume have covered surgical or oncology specialities.The few studies that have been done on volume and outcome in maternity care have shown variable effects. Rural health care is often associated with lower volumes of obstetric procedures. However, no differences in maternal or newborn outcomes have been shown in some comparisons of births in urban and rural locations.^15,16,17,18 Other studies have shown poorer maternal and newborn outcomes in low-volume hospitals, neonatal intensive care units (NICUs) and rural locations.^19,20,21,22 Conversely, higher volume (hospitals with more than 1000 deliveries per year) has been associated with more maternal lacerations or complications.²³When the health care provider has been the unit of analysis, a relation between volume and maternal or newborn outcome has been demonstrated in at least one study²⁴ but not in others.^25,26 Low volume has been defined as 20 to 24 deliveries per year.^24,26 Hass and colleagues²⁴ reported an adjusted odds ratio (OR) of 1.4 for low birth weight for infants delivered by low-volume non-board-certified physicians relative to high-volume non-board-certified physicians; the adjusted OR was 1.56 for low-volume board-certified physicians relative to high-volume board-certified physicians (98.7% of whom were obstetricians).Possible explanations for the differences among studies include differences in health care delivery systems, insurance coverage, experience and training of providers, maternal risk factors, triage or transfer of high-risk cases, choice of outcome measures, and changes over time in access to care, quality assurance and standard of living. Relations have been reported between maternal or newborn outcomes and smoking, maternal history of low birth weight (for previous pregnancies), pregnancy–induced hypertension, diabetes, prepregnancy weight, gestational weight gain, maternal height and age, multiple gestation, previous vaginal birth after cesarean section, history of previous delivery problems, parity, large-for-date fetus, ethnicity and fetal sex.^25,27,28,29 Few studies of the relation between volume of births and obstetric outcome have been able to control for these potentially confounding variables and adjust for maternal risk factors.Our database of detailed accounts of births in one hospital setting allowed us to examine this issue more rigorously. We posed 2 research questions: Is there a relation between the volume of deliveries attended by individual family physicians and maternal and newborn outcomes? If there are differences in outcomes, are they related to different physician practice styles and consultation patterns?     

Instability of a premutation-sized CGG repeat in FMR1 YAC transgenic mice

Fragile X syndrome results from the massive expansion of a CGG repeat in the 5' untranslated region of the gene FMR1. Data suggest that the hyperexpansion properties of FMR1 CGG repeats may depend on flanking cis-acting elements. We have therefore used homologous recombination in yeast to introduce an in situ CGG expansion corresponding to a premutation-sized allele into a human YAC carrying the FMR1 locus. Several transgenic lines were generated that carried repeats of varying lengths and amounts of flanking sequence. Length-dependent instability in the form of small expansions and contractions was observed in both male and female transmissions over five generations. No parent-of-origin effect or somatic instability was observed. Alterations in tract length were found to occur exclusively in the 3' uninterrupted CGG tract. Large expansion events indicative of a transition from a premutation to a full mutation were not observed. Overall, our results indicate both similarities and differences between the behavior of a premutation-sized repeat in mouse and that in human.     

Specificity of elongation factor EF-TU for hydrophobic peptides

The elongation factor EF-Tu carries aminoacyl-tRNAs to the A-site of the ribosome during the elongation process of protein biosynthesis. We, and others, have recently reported that the Escherichia coli EF-Tu interacts with unfolded and denatured proteins and behaves like a chaperone in protein folding and protection against protein thermal denaturation. In this study, we have identified EF-Tu binding sites in protein substrates by screening cellulose-bound peptides scanning the sequences of several proteins. The binding motifs recognized by EF-Tu in protein substrates are also recognized by the chaperone DnaK and mainly consist of hydrophobic clusters. EF-Tu interacts as efficiently as DnaK with the membrane spanning sequence of the membrane protein phospholemman and with the signal sequence of alkaline phosphatase. It interacts less efficiently with several other hydrophobic clusters of lysozyme and alkaline phosphatase, which are also DnaK substrates and fails to bind to several DnaK binding sites. Our results suggest that EF-Tu, like DnaK, interacts albeit more weakly with the hydrophobic regions of substrate protein and are consistent with the hypothesis that it possesses chaperone properties.     

Structure of FAD-bound L-aspartate oxidase: insight into substrate specificity and catalysis

L-Aspartate oxidase (Laspo) catalyzes the conversion of L-Asp to iminoaspartate, the first step in the de novo biosynthesis of NAD(+). This bacterial pathway represents a potential drug target since it is absent in mammals. The Laspo R386L mutant was crystallized in the FAD-bound catalytically competent form and its three-dimensional structure determined at 2.5 A resolution in both the native state and in complex with succinate. Comparison of the R386L holoprotein with the wild-type apoenzyme [Mattevi, A., Tedeschi, G., Bacchella, L., Coda, A., Negri, A., and Ronchi, S. (1999) Structure 7, 745-756] reveals that cofactor incorporation leads to the ordering of two polypeptide segments (residues 44-53 and 104-141) and to a 27 degree rotation of the capping domain. This motion results in the formation of the active site cavity, located at the interface between the capping domain and the FAD-binding domain. The structure of the succinate complex indicates that the cavity surface is decorated by two clusters of H-bond donors that anchor the ligand carboxylates. Moreover, Glu121, which is strictly conserved among Laspo sequences, is positioned to interact with the L-Asp alpha-amino group. The architecture of the active site of the Laspo holoenzyme is remarkably similar to that of respiratory fumarate reductases, providing strong evidence for a common mechanism of catalysis in Laspo and flavoproteins of the succinate dehydrogenase/fumarate reductase family. This implies that Laspo is mechanistically distinct from other flavin-dependent amino acid oxidases, such as the prototypical D-amino acid oxidase.