Supervised cluster analysis for microarray data based on multivariate Gaussian mixture

MOTIVATION: Grouping genes having similar expression patterns is called gene clustering, which has been proved to be a useful tool for extracting underlying biological information of gene expression data. Many clustering procedures have shown success in microarray gene clustering; most of them belong to the family of heuristic clustering algorithms. Model-based algorithms are alternative clustering algorithms, which are based on the assumption that the whole set of microarray data is a finite mixture of a certain type of distributions with different parameters. Application of the model-based algorithms to unsupervised clustering has been reported. Here, for the first time, we demonstrated the use of the model-based algorithm in supervised clustering of microarray data. RESULTS: We applied the proposed methods to real gene expression data and simulated data. We showed that the supervised model-based algorithm is superior over the unsupervised method and the support vector machines (SVM) method. AVAILABILITY: The program written in the SAS language implementing methods I-III in this report is available upon request. The software of SVMs is available in the website http://svm.sdsc.edu/cgi-bin/nph-SVMsubmit.cgi     

Multipoint genetic mapping of quantitative trait loci with dominant markers in outbred populations

We present a multipoint algorithm for mapping quantitative trait loci (QTLs) using dominant markers. The algorithm is designed for outbred populations and is particularly suited for large families. The algorithm works with either codominant or dominant markers, either of which may be interspersed within the same linkage map. Concurrently, the algorithm also partitions dominance variance at the QTL. Computer simulations show that with large families, QTL mapping with dominant markers can be almost as powerful as with bi-allelic, codominant markers. Yet despite this, other situations show a large standard deviation in the estimate of the QTL position, thus making QTL mapping with dominant markers in outbred populations a useful detection tool, albeit limited in its resolution. This revised version was published online in July 2006 with corrections to the Cover Date.     

Quantitative trait loci responsible for variation in sexually dimorphic traits in Drosophila melanogaster

To understand the mechanisms of morphological evolution and species divergence, it is essential to elucidate the genetic basis of variation in natural populations. Sexually dimorphic characters, which evolve rapidly both within and among species, present attractive models for addressing these questions. In this report, we map quantitative trait loci (QTL) responsible for variation in sexually dimorphic traits (abdominal pigmentation and the number of ventral abdominal bristles and sex comb teeth) in a natural population of Drosophila melanogaster. To capture the pattern of genetic variation present in the wild, a panel of recombinant inbred lines was created from two heterozygous flies taken directly from nature. High-resolution mapping was made possible by cytological markers at the average density of one per 2 cM. We have used a new Bayesian algorithm that allows QTL mapping based on all markers simultaneously. With this approach, we were able to detect small-effect QTL that were not evident in single-marker analyses. Our results show that at least for some sexually dimorphic traits, a small number of QTL account for the majority of genetic variation. The three strongest QTL account for >60% of variation in the number of ventral abdominal bristles. Strikingly, a single QTL accounts for almost 60% of variation in female abdominal pigmentation. This QTL maps to the chromosomal region that Robertson et al. have found to affect female abdominal pigmentation in other populations of D. melanogaster. Using quantitative complementation tests, we demonstrate that this QTL is allelic to the bric a brac gene, whose expression has previously been shown to correlate with interspecific differences in pigmentation. Multiple bab alleles that confer distinct phenotypes appear to segregate in natural populations at appreciable frequencies, suggesting that intraspecific and interspecific variation in abdominal pigmentation may share a similar genetic basis.     

Theoretical basis of the Beavis effect

The core of statistical inference is based on both hypothesis testing and estimation. The use of inferential statistics for QTL identification thus includes estimation of genetic effects and statistical tests. Typically, QTL are reported only when the test statistics reach a predetermined critical value. Therefore, the estimated effects of detected QTL are actually sampled from a truncated distribution. As a result, the expectations of detected QTL effects are biased upward. In a simulation study, William D. Beavis showed that the average estimates of phenotypic variances associated with correctly identified QTL were greatly overestimated if only 100 progeny were evaluated, slightly overestimated if 500 progeny were evaluated, and fairly close to the actual magnitude when 1000 progeny were evaluated. This phenomenon has subsequently been called the Beavis effect. Understanding the theoretical basis of the Beavis effect will help interpret QTL mapping results and improve success of marker-assisted selection. This study provides a statistical explanation for the Beavis effect. The theoretical prediction agrees well with the observations reported in Beavis's original simulation study. Application of the theory to meta-analysis of QTL mapping is discussed.     

Chromosomal regions harboring genes for the work to femur failure in mice

The work to failure is defined as the maximum energy bone can absorb before breaking, and therefore is a direct test of the risk of fracture. To determine the genetic loci influencing work to failure, we have performed a high density genome-wide scan in 633 (MRL × SJL) F₂ female mice. Five loci (P <0.005) with significant effects on work to failure were found on chromosomes 2, 7, 8, 9, and X, which collectively explained around 20% variance of work to femur failure in F₂ mice. Of those, only the QTL on chromosome 9 was concordant with bone mineral density (BMD) QTLs. Eight significant interactions (P <0.01) between marker loci were identified, which accounted for an equivalent amount of F₂ variance (23%) to combined single QTL effects. Our results demonstrate that most of the genetic loci regulating work to failure are different from those for BMD in the 7-week-old female mice. If this is also true in humans, this finding will challenge the predictive value of BMD for the risk of fracture. Electronic Publication     

Mechanisms for 2-methoxyestradiol-induced apoptosis of prostate cancer cells

Prostate and breast carcinomas are sex hormone-related carcinomas, which are known to be associated with an over-expression of the proto-oncogene Bcl-2. Here, we report that 2-methoxyestradiol (2-ME), an endogenous metabolite of estrogen that does not bind to nuclear estrogen receptors, effectively induces apoptosis in Bcl-2-expressing human prostate and breast carcinoma cells in vitro and in a rat prostate tumor model in vivo. In several cell lines derived from prostate, breast, liver and colorectal carcinomas, 2-ME treatment led to an activation of c-Jun N-terminal kinase (JNK) and phosphorylation of Bcl-2, which preceded the induction of apoptosis. In summary, our data suggest that 2-ME induces apoptosis in epithelial carcinomas by causing phosphorylation of JNK, which appears to be correlated with phosphorylation of Bcl-2.     

CRISPR/Cas-based screening of a gene activation library in Saccharomyces cerevisiae identifies a crucial role of OLE1 in thermotolerance

CRISPR/Cas-based (clustered regularly interspaced short palindromic repeats/CRISPR-associated) screening has been proved to be an efficient method to study functional genomics from yeast to human. In this study, we report the development of a focused CRISPR/Cas-based gene activation library in Saccharomyces cerevisiae and its application in gene identification based on functional screening towards improved thermotolerance. The gene activation library was subjected to screening at 42°C, and the same library cultured at 30°C was set as a control group. After five successive subcultures, five clones were randomly picked from the libraries cultured at 30 and 42°C, respectively. The five clones selected at 30°C contain the specificity sequences of five different single guide RNAs, whereas all the five clones selected at 42°C contain the specificity sequence of one sgRNA that targets the promoter region of OLE1. A crucial role of OLE1 in thermotolerance was identified: the overexpression of OLE1 increased fatty acid unsaturation, and thereby helped counter lipid peroxidation caused by heat stress, rendering the yeast thermotolerant. This study described the application of CRISPR/Cas-based gene activation screening with an example of thermotolerant yeast screening, demonstrating that this method can be used to identify functional genes in yeast.     

High concentration magnesium inhibits extracellular matrix calcification and protects articular cartilage via Erk/autophagy pathway

The significant cytopathological changes of osteoarthritis are chondrocyte hypertrophy, proteoglycan loss, extracellular matrix (ECM) calcification, and terminally, the replacement of cartilage by bone. Meanwhile, magnesium ion (Mg²⁺), as the second most abundant divalent cation in the human body, has been proved to inhibit the ECM calcification of hBMSCs (human bone marrow stromal cells), hVSMCs (Human vascular smooth muscle cells), and TDSCs (tendon-derived stem cells) in vitro studies. The ATDC5 cell line, which holds chondrocyte characteristics, was used in this study as an in vitro subject. We found that Mg²⁺ can efficiently suppress the ECM calcification and downregulate both hypertrophy and matrix metalloproteinase-related genes. Meanwhile, Mg²⁺ inhibits the formation of autophagy by inhibiting Erk phosphorylation signaling and lowers the expression of LC3, and eventually effectively reduces the formation of ECM calcification in vitro. In this study, we also used destabilization of the medial meniscus (DMM)-induced osteoarthritis (OA) animal model to further confirm the protective effect of Mg²⁺ on articular cartilage. Compared with the control group (saline-injected), continuous intra-articular magnesium chloride (MgCl₂) injection can significantly alleviate the severity of cartilage calcification in OA animal model. Immunofluorescence staining also revealed that saline-injected DMM group had a higher positive rate of LC3 expression in cartilage chondrocytes, compared with MgCl₂-injected DMM group. In general, Mg²⁺ can significantly downregulate the hypertrophic gene Runx2, MMP13, and Col10α1, upregulate the chondrogenic genes Sox9 and Col1α1, inhibit the Erk phosphorylation signaling, reduce the expression of autophagy protein LC3, and effectively inhibit the ECM calcification of ATDC5. In vivo study also proved that intra-articular injection of Mg²⁺ protected knee cartilage by inhibiting the autophagy formation.     

Bayesian shrinkage analysis of quantitative trait Loci for dynamic traits

Many quantitative traits are measured repeatedly during the life of an organism. Such traits are called dynamic traits. The pattern of the changes of a dynamic trait is called the growth trajectory. Studying the growth trajectory may enhance our understanding of the genetic architecture of the growth trajectory. Recently, we developed an interval-mapping procedure to map QTL for dynamic traits under the maximum-likelihood framework. We fit the growth trajectory by Legendre polynomials. The method intended to map one QTL at a time and the entire QTL analysis involved scanning the entire genome by fitting multiple single-QTL models. In this study, we propose a Bayesian shrinkage analysis for estimating and mapping multiple QTL in a single model. The method is a combination between the shrinkage mapping for individual quantitative traits and the Legendre polynomial analysis for dynamic traits. The multiple-QTL model is implemented in two ways: (1) a fixed-interval approach where a QTL is placed in each marker interval and (2) a moving-interval approach where the position of a QTL can be searched in a range that covers many marker intervals. Simulation study shows that the Bayesian shrinkage method generates much better signals for QTL than the interval-mapping approach. We propose several alternative methods to present the results of the Bayesian shrinkage analysis. In particular, we found that the Wald test-statistic profile can serve as a mechanism to test the significance of a putative QTL.     

Disruption of transforming growth factor-beta signaling by curcumin induces gene expression of peroxisome proliferator-activated receptor-gamma in rat hepatic stellate cells

Activation of hepatic stellate cells (HSC), the major effectors of hepatic fibrogenesis, is coupled with sequential alterations in gene expression, including an increase in receptors for transforming growth factor-beta (TGF-beta) and a dramatic reduction in the peroxisome proliferator-activated receptor-gamma (PPAR-gamma). The relationship between them remains obscure. We previously demonstrated that curcumin induced gene expression of PPAR-gamma in activated HSC, leading to reducing cell proliferation, inducing apoptosis and suppressing expression of extracellular matrix genes. The underlying molecular mechanisms are largely unknown. We recently observed that stimulation of PPAR-gamma activation suppressed gene expression of TGF-beta receptors in activated HSC, leading to the interruption of TGF-beta signaling. This observation supported our assumption of an antagonistic relationship between PPAR-gamma activation and TGF-beta signaling in HSC. In this study, we further hypothesize that TGF-beta signaling might negatively regulate gene expression of PPAR-gamma in activated HSC. The present report demonstrates that exogenous TGF-beta1 inhibits gene expression of PPAR-gamma in activated HSC, which is eliminated by the pretreatment with curcumin likely by interrupting TGF-beta signaling. Transfection assays further indicate that blocking TGF-beta signaling by dominant negative type II TGF-beta receptor increases the promoter activity of PPAR-gamma gene. Promoter deletion assays, site-directed mutageneses, and gel shift assays localize two Smad binding elements (SBEs) in the PPAR-gamma gene promoter, acting as curcumin response elements and negatively regulating the promoter activity in passaged HSC. The Smad3/4 protein complex specifically binds to the SBEs. Overexpression of Smad4 dose dependently eliminates the inhibitory effects of curcumin on the PPAR-gamma gene promoter and TGF-beta signaling. Taken together, these results demonstrate that the interruption of TGF-beta signaling by curcumin induces gene expression of PPAR-gamma in activated HSC in vitro. Our studies provide novel insights into the molecular mechanisms of curcumin in the induction of PPAR-gamma gene expression and in the inhibition of HSC activation.