The Use of High-Throughput DNA Sequencing in the Investigation of Antigenic Variation: Application to Neisseria Species

Antigenic variation occurs in a broad range of species. This process resembles gene conversion in that variant DNA is unidirectionally transferred from partial gene copies (or silent loci) into an expression locus. Previous studies of antigenic variation have involved the amplification and sequencing of individual genes from hundreds of colonies. Using the pilE gene from Neisseria gonorrhoeae we have demonstrated that it is possible to use PCR amplification, followed by high-throughput DNA sequencing and a novel assembly process, to detect individual antigenic variation events. The ability to detect these events was much greater than has previously been possible. In N. gonorrhoeae most silent loci contain multiple partial gene copies. Here we show that there is a bias towards using the copy at the 3′ end of the silent loci (copy 1) as the donor sequence. The pilE gene of N. gonorrhoeae and some strains of Neisseria meningitidis encode class I pilin, but strains of N. meningitidis from clonal complexes 8 and 11 encode a class II pilin. We have confirmed that the class II pili of meningococcal strain FAM18 (clonal complex 11) are non-variable, and this is also true for the class II pili of strain NMB from clonal complex 8. In addition when a gene encoding class I pilin was moved into the meningococcal strain NMB background there was no evidence of antigenic variation. Finally we investigated several members of the opa gene family of N. gonorrhoeae, where it has been suggested that limited variation occurs. Variation was detected in the opaK gene that is located close to pilE, but not at the opaJ gene located elsewhere on the genome. The approach described here promises to dramatically improve studies of the extent and nature of antigenic variation systems in a variety of species.     

The NOD2 p.Leu1007fsX1008 Mutation (rs2066847) Is a Stronger Predictor of the Clinical Course of Crohn's Disease than the FOXO3A Intron Variant rs12212067

Background

Very recently, a sub-analysis of genome-wide association scans revealed that the non-coding single nucleotide polymorphism (SNP) rs12212067 in the FOXO3A gene is associated with a milder course of Crohn''s disease (CD) (Cell 2013;155:57–69). The aim of our study was to evaluate the clinical value of the SNP rs12212067 in predicting the severity of CD by correlating CD patient genotype status with the most relevant complications of CD such as stenoses, fistulas, and CD-related surgery.

Methodology/Principal Findings

We genotyped 550 CD patients for rs12212067 (FOXO3A) and the three common CD-associated NOD2 mutations rs2066844, rs2066847, and rs2066847 and performed genotype-phenotype analyses.

Results

No significant phenotypic differences were found between the wild-type genotype TT of the FOXO3A SNP rs12212067 and the minor genotypes TG and GG independently from NOD2 variants. The allele frequency of the minor G allele was 12.7%. Age at diagnosis, disease duration, body mass index, surgery rate, stenoses, fistula, need for immunosuppressive therapy, and disease course were not significantly different. In contrast, the NOD2 mutant p.Leu1007fsX1008 (rs2066847) was highly associated with penetrating CD (p = 0.01), the development of fistulas (p = 0.01) and stenoses (p = 0.01), and ileal disease localization (p = 0.03). Importantly, the NOD2 SNP rs2066847 was a strong separator between an aggressive and a mild course of CD (p = 2.99×10⁻⁵), while the FOXO3A SNP rs12212067 did not separate between mild and aggressive CD behavior in our cohort (p = 0.35). 96.2% of the homozygous NOD2 p.Leu1007fsX1008 carriers had an aggressive disease behavior compared to 69.3% of the patients with the NOD2 wild-type genotype (p = 0.007).

Conclusion/Significance

In clinical practice, the NOD2 variant p.Leu1007fsX1008 (rs2066847), in particular in homozygous form, is a much stronger marker for a severe clinical phenotype than the FOXO3A rs12212067 SNP for a mild disease course on an individual patient level despite its important impact on the inflammatory response of monocytes.     

CRF-based models of protein surfaces improve protein-protein interaction site predictions

Background

The identification of protein-protein interaction sites is a computationally challenging task and important for understanding the biology of protein complexes. There is a rich literature in this field. A broad class of approaches assign to each candidate residue a real-valued score that measures how likely it is that the residue belongs to the interface. The prediction is obtained by thresholding this score.Some probabilistic models classify the residues on the basis of the posterior probabilities. In this paper, we introduce pairwise conditional random fields (pCRFs) in which edges are not restricted to the backbone as in the case of linear-chain CRFs utilized by Li et al. (2007). In fact, any 3D-neighborhood relation can be modeled. On grounds of a generalized Viterbi inference algorithm and a piecewise training process for pCRFs, we demonstrate how to utilize pCRFs to enhance a given residue-wise score-based protein-protein interface predictor on the surface of the protein under study. The features of the pCRF are solely based on the interface predictions scores of the predictor the performance of which shall be improved.

Results

We performed three sets of experiments with synthetic scores assigned to the surface residues of proteins taken from the data set PlaneDimers compiled by Zellner et al. (2011), from the list published by Keskin et al. (2004) and from the very recent data set due to Cukuroglu et al. (2014). That way we demonstrated that our pCRF-based enhancer is effective given the interface residue score distribution and the non-interface residue score are unimodal.Moreover, the pCRF-based enhancer is also successfully applicable, if the distributions are only unimodal over a certain sub-domain. The improvement is then restricted to that domain. Thus we were able to improve the prediction of the PresCont server devised by Zellner et al. (2011) on PlaneDimers.

Conclusions

Our results strongly suggest that pCRFs form a methodological framework to improve residue-wise score-based protein-protein interface predictors given the scores are appropriately distributed. A prototypical implementation of our method is accessible at http://ppicrf.informatik.uni-goettingen.de/index.html.     

Improved MPI collectives for MPI processes in shared address spaces

As the number of cores per node keeps increasing, it becomes increasingly important for MPI to leverage shared memory for intranode communication. This paper investigates the design and optimization of MPI collectives for clusters of NUMA nodes. We develop performance models for collective communication using shared memory and we demonstrate several algorithms for various collectives. Experiments are conducted on both Xeon X5650 and Opteron 6100 InfiniBand clusters. The measurements agree with the model and indicate that different algorithms dominate for short vectors and long vectors. We compare our shared-memory allreduce with several MPI implementations—Open MPI, MPICH2, and MVAPICH2—that utilize system shared memory to facilitate interprocess communication. On a 16-node Xeon cluster and 8-node Opteron cluster, our implementation achieves on geometric average 2.3X and 2.1X speedup over the best MPI implementation, respectively. Our techniques enable an efficient implementation of collective operations on future multi- and manycore systems.     

Determination of phosphatidylethanol in blood from alcoholic males using high-performance liquid chromatography and evaporative light scattering or electrospray mass spectrometric detection

The `pathologic' phospholipid, phosphatidylethanol (PEth), formed only in the presence of ethanol, was determined in extracts of human blood using high-performance liquid chromatography with evaporative light scattering detection (ELSD) or electrospray (ES) mass spectrometry. Separation was performed using a diol column and a normal-phase binary gradient system. Decreasing concentrations of PEth (15 to 1 nmol/ml blood) could be detected by ELSD in three male alcoholics, up to 3 weeks after the beginning of an alcohol-free period. Using ES, levels down to 100 pmol/ml blood was detected. The molecular species of PEth were similar to those of phosphatidylcholine found in the same blood sample. The method provides a rapid quantitative and qualitative determination of PEth in blood. The limits of detection were 200 pmol (≈125 ng) using ELSD and 140 fmol (≈100 pg) using ES, total amounts injected.     

Quantitating intraparticle O2 gradients in solid supported enzyme immobilizates: Experimental determination of their role in limiting the catalytic effectiveness of immobilized glucose oxidase

Enzymatic O₂‐dependent oxidations are receiving increased attention for use in fine chemicals synthesis. Solid supported oxidation catalysts often show poor efficiency due to pronounced O₂ diffusion restriction. Internal O₂ supply therefore constitutes a key parameter for optimizing the enzyme immobilization. We herein describe an optical sensing method for quantitation of space‐averaged intraparticle O₂ concentrations in porous Sepabeads carriers. The method applies phosphorescence lifetime measurements on Sepabeads labeled with an O₂ sensitive indicator dye. Using glucose oxidase immobilized at different loadings (0.005–12 mg/g) on labeled Sepabeads, we analyzed in real time during the enzymatic reaction the formation of O₂ concentration differences between bulk liquid and the intraparticle environment. We show that the O₂ gradient at apparent steady state increased with increasing enzyme loading, so that O₂ eventually became totally depleted from inside the highly loaded carriers. We also show that the residual intraparticle O₂ concentration was correlated with the catalytic effectiveness factor (η) of the enzyme immobilizate used, thus providing a direct measure of the magnitude of O₂ diffusion limitation. Once corrected for diffusional effect, η was no longer dependent on enzyme loading and its constant value now described the intrinsic activity of immobilized glucose oxidase. Three common procedures of enzyme immobilization, involving adsorption, cross‐linking, and covalent attachment, are shown to differ widely concerning the obtained intrinsic activity. Therefore, intraparticle O₂ concentration data enable distinction between diffusional restriction and activity loss as the two principal factors limiting the effectiveness of immobilized O₂ dependent enzymes, and thus they inform rational design of an optimally active oxidation biocatalyst on solid support. Biotechnol. Bioeng. 2013; 110: 2086–2095. © 2013 Wiley Periodicals, Inc.