ProDCoNN: Protein design using a convolutional neural network

Designing protein sequences that fold to a given three-dimensional (3D) structure has long been a challenging problem in computational structural biology with significant theoretical and practical implications. In this study, we first formulated this problem as predicting the residue type given the 3D structural environment around the C _α atom of a residue, which is repeated for each residue of a protein. We designed a nine-layer 3D deep convolutional neural network (CNN) that takes as input a gridded box with the atomic coordinates and types around a residue. Several CNN layers were designed to capture structure information at different scales, such as bond lengths, bond angles, torsion angles, and secondary structures. Trained on a very large number of protein structures, the method, called ProDCoNN (protein design with CNN), achieved state-of-the-art performance when tested on large numbers of test proteins and benchmark datasets.     

大豆转录因子GmERF5的克隆、表达及功能分析

ERF转录因子是植物中特有的转录因子家族之一, 在植物响应生物和非生物胁迫过程中发挥重要的调控作用。通过对大豆(Glycine max)吉林32未成熟胚的表达谱分析, 利用RT-PCR技术从大豆中克隆了1个新的ERF转录因子GmERF5。GmERF5具有237个氨基酸残基, 分子量为26.09 kDa, 等电点为6.85, 其开放阅读框长714 bp。该转录因子蛋白与Gh-ERF2蛋白的同源性最高, 它们同属ERF亚家族的第IV亚类。实时荧光定量PCR分析表明, 该蛋白基因在大豆的根中表达量最高, 且受干旱、高盐、低温及乙烯、脱落酸和茉莉酸甲酯的诱导上调表达。亚细胞定位实验结果表明, GmERF5蛋白定位于细胞核中。转录激活能力分析结果显示, GmERF5可以激活报告基因的表达, 为转录激活子。综合以上结果, 认为GmERF5可能作为转录调控因子参与大豆生物和非生物胁迫的应答。     

Production of a mutant of large-scale loach Paramisgurnus dabryanus with skin pigmentation loss by genome editing with CRISPR/Cas9 system

CRISPR/Cas9 system has been developed as a highly efficient genome editing technology to specifically induce mutations in a few aquaculture species. In this study, we described induction of targeted gene (namely tyrosinase, tyr) mutations in large-scale loach Paramisgurnus dabryanus, an important aquaculture fish species and a potential model organism for studies of intestinal air-breathing function, using the CRISPR/Cas9 system. Tyr gene in large-scale loach was firstly cloned and then its expressions were investigated. Two guide RNAs (gRNAs) were designed and separately transformed with Cas9 in the loach. 89.4% and 96.1% of injected loach juveniles respectively displayed a graded loss of pigmentation for the two gRNAs, in other words, for target 1 and target 2. We classified the injected loach juveniles into five groups according to their skin color phenotypes, including four albino groups and one wild-type-like group. And one of them was clear albino group, which was of high ornamental and commercial value. More than 50 clones for each albino transformant with a visible phenotype in each target were randomly selected and sequenced. Results obtained here showed that along with the increase of pigmentation, wild-type alleles appeared in the injected loach juveniles more often and insertion/deletion alleles less frequently. This study demonstrated that CRISPR/Cas9 system could be practically performed to modify large-scale loach tyr to produce an albino mutant of high ornamental and commercial value, and for the first time showed successful use of the CRISPR/Cas9 system for genome editing in a Cobitidae species.

     

Introgression lines of Solanum sitiens,a wild nightshade of the Atacama Desert,in the genome of cultivated tomato

The wild tomato relative Solanum sitiens is a xerophyte endemic to the Atacama Desert of Chile and a potential source of genes for tolerance to drought, salinity and low‐temperature stresses. However, until recently, strong breeding barriers prevented its hybridization and introgression with cultivated tomato, Solanum lycopersicum L. We overcame these barriers using embryo rescue, bridging lines and allopolyploid hybrids, and synthesized a library of introgression lines (ILs) that captures the genome of S. sitiens in the background of cultivated tomato. The IL library consists of 56 overlapping introgressions that together represent about 93% of the S. sitiens genome: 65% in homozygous and 28% in heterozygous (segregating) ILs. The breakpoints of each segment and the gaps in genome coverage were mapped by single nucleotide polymorphism (SNP) genotyping using the SolCAP SNP array. Marker‐assisted selection was used to backcross selected introgressions into tomato, to recover a uniform genetic background, to isolate recombinant sub‐lines with shorter introgressions and to select homozygous genotypes. Each IL contains a single S. sitiens chromosome segment, defined by markers, in the genetic background of cv. NC 84173, a fresh market inbred line. Large differences were observed between the lines for both qualitative and quantitative morphological traits, suggesting that the ILs contain highly divergent allelic variation. Several loci contributing to unilateral incompatibility or hybrid necrosis were mapped with the lines. This IL population will facilitate studies of the S. sitiens genome and expands the range of genetic variation available for tomato breeding and research.     

Molecular Typing of Pathogenic Leptospira Serogroup Icterohaemorrhagiae Strains Circulating in China during the Past 50 Years

Background

Leptospirosis is one of the most important neglected tropical infectious diseases worldwide. Icterohaemorrhagiae has been throughout recent history, and still is, the predominant serogroup of this pathogen in China. However, very little in detail is known about the serovars or genotypes of this serogroup.

Methodology/Principal Findings

In this study, 120 epidemic strains from five geographically diverse regions in China collected over a 50 year period (1958~2008), and 8 international reference strains characterized by 16S rRNA sequencing and MLST analysis. 115, 11 and 2 strains were identified as L. interrogans, L. borgpetersenii, and L. kirschneri, respectively. 17 different STs were identified including 69 ST1 strains, 18 ST17, 18 ST128, 9 ST143 and 2 ST209. The remaining 12 strains belonged to 12 different STs. eBURST analysis demonstrated that, among the clonal complexes isolated (CCs), CC1 accounted for 73.3% (88/120) strains representing three STs: ST1, ST128 and ST98. ST1 was the most likely ancestral strain of this CC, followed by singleton CC17 (17/120) and CC143 (11/120). Further analysis of adding 116 serogroup Icterohaemorrhagiae strains in the MLST database and studies previously described using global eBURST analysis and MST dendrogram revealed relatively similar ST clustering patterns with five main CCs and 8 singletons among these 244 strains. CC17 was found to be the most prevalent clone of pathogenic Leptospira circulating worldwide. This is the first time, to our knowledge, that ST1 and ST17 strains were distributed among 4 distinct serovars, indicating a highly complicated relationship between serovars and STs.

Conclusions/Significance

Our studies demonstrated a high level of genetic diversity in the serogroup Icterohaemorrhagiae strains. Distinct from ST17 or ST37 circulating elsewhere, ST1 included in CC1, has over the past 50 years or so, proven to be the most prevalent ST of pathogenic leptospires isolated in China. Moreover, the complicated relationship between STs and serovars indicates an urgent need to develop an improved scheme for Leptospira serotyping.     

Cloning, expression, purification, and activity assay of proteins related to D-lactic acid formation in Lactobacillus rhamnosus

Two proteins that might be responsible for D-lactic acid (D-LA) formation were screened from the genome database of Lactobacillus rhamnosus GG. The coding genes of the two proteins in L. rhamnosus CASL, ldhD1 and ldhD2, were cloned and expressed in Escherichia coli Rosetta with an inducible expression vector pETDuet™-1 (Novagen, Darmstadt, Germany), respectively. The two purified proteins, LdhD-1 and LdhD-2, migrated as a single protein band separately, both corresponding to an apparent molecular mass between 35 kDa and 45 kDa on sodium dodecyl sulfate polyacrylamide gel electrophoresis. The specific activities of LdhD-1 and LdhD-2 catalyzing pyruvate to LA were 0.02 U/mg and 0.21 U/mg, respectively. The configuration of LA converted from pyruvate was determined using high-performance liquid chromatography equipped with a chiral column. Only D-LA was detected when LdhD-1 and LdhD-2 were tested. In summary, the two proteins cloned and expressed in this study were most probably responsible for D-LA formation during fermentation of L. rhamnosus CASL.