Lung eQTLs to help reveal the molecular underpinnings of asthma.

Genome-wide association studies (GWAS) have identified loci reproducibly associated with pulmonary diseases; however, the molecular mechanism underlying these associations are largely unknown. The objectives of this study were to discover genetic variants affecting gene expression in human lung tissue, to refine susceptibility loci for asthma identified in GWAS studies, and to use the genetics of gene expression and network analyses to find key molecular drivers of asthma. We performed a genome-wide search for expression quantitative trait loci (eQTL) in 1,111 human lung samples. The lung eQTL dataset was then used to inform asthma genetic studies reported in the literature. The top ranked lung eQTLs were integrated with the GWAS on asthma reported by the GABRIEL consortium to generate a Bayesian gene expression network for discovery of novel molecular pathways underpinning asthma. We detected 17,178 cis- and 593 trans- lung eQTLs, which can be used to explore the functional consequences of loci associated with lung diseases and traits. Some strong eQTLs are also asthma susceptibility loci. For example, rs3859192 on chr17q21 is robustly associated with the mRNA levels of GSDMA (P = 3.55 × 10(-151)). The genetic-gene expression network identified the SOCS3 pathway as one of the key drivers of asthma. The eQTLs and gene networks identified in this study are powerful tools for elucidating the causal mechanisms underlying pulmonary disease. This data resource offers much-needed support to pinpoint the causal genes and characterize the molecular function of gene variants associated with lung diseases.

Identification of a panel of sensitive and specific DNA methylation markers for lung adenocarcinoma.

BACKGROUND: Lung cancer is the number one cancer killer of both men and women in the United States. Three quarters of lung cancer patients are diagnosed with regionally or distantly disseminated disease; their 5-year survival is only 15%. DNA hypermethylation at promoter CpG islands shows great promise as a cancer-specific marker that would complement visual lung cancer screening tools such as spiral CT, improving early detection. In lung cancer patients, such hypermethylation is detectable in a variety of samples ranging from tumor material to blood and sputum. To date the penetrance of DNA methylation at any single locus has been too low to provide great clinical sensitivity. We used the real-time PCR-based method MethyLight to examine DNA methylation quantitatively at twenty-eight loci in 51 primary human lung adenocarcinomas, 38 adjacent non-tumor lung samples, and 11 lung samples from non-lung cancer patients. RESULTS: We identified thirteen loci showing significant differential DNA methylation levels between tumor and non-tumor lung; eight of these show highly significant hypermethylation in adenocarcinoma: CDH13, CDKN2A EX2, CDX2, HOXA1, OPCML, RASSF1, SFPR1, and TWIST1 (p-value << 0.0001). Using the current tissue collection and 5-fold cross validation, the four most significant loci (CDKN2A EX2, CDX2, HOXA1 and OPCML) individually distinguish lung adenocarcinoma from non-cancer lung with a sensitivity of 67-86% and specificity of 74-82%. DNA methylation of these loci did not differ significantly based on gender, race, age or tumor stage, indicating their wide applicability as potential lung adenocarcinoma markers. We applied random forests to determine a good classifier based on a subset of our loci and determined that combined use of the same four top markers allows identification of lung cancer tissue from non-lung cancer tissue with 94% sensitivity and 90% specificity. CONCLUSION: The identification of eight CpG island loci showing highly significant hypermethylation in lung adenocarcinoma provides strong candidates for evaluation in patient remote media such as plasma and sputum. The four most highly ranked loci, CDKN2A EX2, CDX2, HOXA1 and OPCML, which show significant DNA methylation even in stage IA tumor samples, merit further investigation as some of the most promising lung adenocarcinoma markers identified to date.

DNA methylation profile of 28 potential marker loci in malignant mesothelioma.

Patients with malignant mesothelioma (MM), an aggressive cancer associated with asbestos exposure, usually present clinically with advanced disease and this greatly reduces the likelihood of curative treatment. MM is difficult to diagnose without invasive techniques; the development of non-invasively detectable molecular markers would therefore be highly beneficial. DNA methylation changes in cancer cells provide powerful markers that are potentially detectable non-invasively in DNA shed into bodily fluids. Here we examined the methylation status of 28 loci in 52 MM tumors to investigate their potential as molecular markers for MM. To exclude candidate MM markers that might be positive in biopsies/pleural fluid due to contaminating surrounding non-tumor lung tissue/DNA, we also examined the methylation of these markers in lung samples (age- or environmentally induced hypermethylation is frequently observed in non-cancerous lung). Statistically significantly increased methylation in MM versus non-tumor lung samples was found for estrogen receptor 1 (ESR1; p = 0.0002), solute carrier family 6 member 20 (SLC6A20; p = 0.0022) and spleen tyrosine kinase (SYK; p=0.0003). Examination of associations between methylation levels of the 28 loci and clinical parameters suggest associations of the methylation status of metallothionein genes with gender, histology, asbestos exposure, and lymph node involvement, and the methylation status of leucine zipper tumor suppressor 1 (LZTS1) and SLC6A20 with survival.

Isoaspartylation appears to trigger small cell lung cancer-associated autoimmunity against neuronal protein ELAVL4.

Autoantibodies against SCLC-associated neuronal antigen ELAVL4 (HuD) have been linked to smaller tumors and improved survival, but the antigenic epitope and mechanism of autoimmunity have never been solved. We report that recombinant human ELAVL4 protein incubated under physiological conditions acquires isoaspartylation, a type of immunogenic protein damage. Specifically, the N-terminal region of ELAVL4, previously implicated in SCLC-associated autoimmunity, undergoes isoaspartylation in vitro, is recognized by sera from anti-ELAVL4 positive SCLC patients and is highly immunogenic in subcutaneously injected mice and in vitro stimulated human lymphocytes. Our data suggest that isoaspartylated ELAVL4 is the trigger for the SCLC-associated anti-ELAVL4 autoimmune response.

DNA methylation analysis: a powerful new tool for lung cancer diagnosis.

Carcinoma of the lung is the most common cause of cancer death worldwide. The estimated 5-year survival ranges from 6-16%, depending on the cell type. The best opportunity for improving survival of lung cancer patients is through early detection, when curative surgical resection is possible. Although the subjects at increased risk for developing carcinoma of the lung (long-term smokers) can be identified, only 10-20% of this group will ultimately develop the disease. Screening tests of long-term smokers employed to date (radiography and sputum cytology) have not been successful in reducing lung cancer mortality. The application of molecular markers specific for lung cancer offers new possibilities for early detection. Hypermethylation of CpG islands in the promoter regions of genes is a common phenomenon in lung cancer, as demonstrated by the analysis of the methylation status of over 40 genes from lung cancer tumors, cell lines, patient sputum and/or serum. Determination of the methylation patterns of multiple genes to obtain complex DNA methylation signatures promises to provide a highly sensitive and specific tool for lung cancer diagnosis. When combined with the development of non-invasive methods to detect such signatures, this may provide a viable method to screen subjects at risk for lung cancer.

Distinct DNA methylation profiles in malignant mesothelioma, lung adenocarcinoma, and non-tumor lung.

DNA methylation markers provide a powerful tool to make diagnoses based on genetic material obtained directly from tumors or from "remote" locations such as sputum, pleural fluid, or serum. In particular when limited cell numbers are available, amplifyable DNA markers can provide a very sensitive tool for cancer detection and classification. Malignant mesothelioma (MM), an aggressive cancer strongly associated with asbestos exposure, can be difficult to distinguish from adenocarcinoma of the lung when limited material is available. In an attempt to identify molecular markers for MM and adenocarcinoma, we examined the DNA methylation status of 14 loci. Analysis of methylation levels in 10 MM and 8 adenocarcinoma cell lines showed that methylation of APC was significantly elevated in adenocarcinoma compared to MM cell lines (P=0.0003), while methylation of CDH1 was higher in MM (P<0.02). Subsequent examination of the methylation status of the 14 loci in 6 MM and 7 adenocarcinoma primary tumors, which yielded similar methylation profiles, supported these observations. Comparison of methylation in MM cell lines and tumors versus non-tumor lung tissue indicated that APC exhibits less methylation in MM (P=0.003) while RASSF1, PGR1, ESR1, and CDH1 show more methylation in MM, the latter two showing the most significant difference between the two tissue types (P< or = 0.0001). Comparison of methylation in adenocarcinoma cell lines and tumors versus non-tumor lung tissue showed methylation of ESR1, PGR1 and RASSF1 to be significantly elevated in adenocarcinoma, with RASSF1 being most significant (P=0.0002). Thus, with the examination of 14 loci, we have identified 5 candidates that show potential for distinguishing between MM, adenocarcinoma and/or non-cancer lung. Our observations support the strong potential of methylation markers as tools for accurate diagnosis of neoplasms in and around the lung.

Hierarchical clustering of lung cancer cell lines using DNA methylation markers.

Recent analyses of global and gene-specific methylation patterns in cancer cells have suggested that cancers from different organs demonstrate distinct patterns of CpG island hypermethylation. Although certain CpG islands are frequently methylated in many different kinds of cancer, others are methylated only in specific tumor types. Because distinct patterns of CpG island hypermethylation can be seen in tumors from different organs, it seems likely that histological subtypes of cancer within a given organ may exhibit distinct methylation patterns as well. The goal of our study was to determine whether the patterns of CpG island hypermethylation could be used to distinguish between different histological subtypes of lung cancer. We analyzed the methylation status of 23 loci in 91 lung cancer cell lines using the quantitative real-time PCR method MethyLight. Genes PTGS2 (COX2), CALCA, MTHFR, ESR1, MGMT, MYOD1, and APC showed statistically significant differences in the level of CpG island methylation between small cell lung cancer (SCLC) and non-small cell lung cancer cell lines (NSCLC). Hierarchical clustering using a panel consisting of these seven loci yielded two major groups, one of which contained 78% of the SCLC lines. Within this group, a large cluster consisted almost exclusively of SCLC cell lines. Our results show that DNA methylation patterns differ between NSCLC and SCLC cell lines and suggest that these patterns could be developed into a powerful molecular marker to achieve accurate diagnosis of lung cancer.

Classification of individual lung cancer cell lines based on DNA methylation markers: use of linear discriminant analysis and artificial neural networks.

The classification of small cell lung cancer (SCLC) and non-small cell lung cancer (NSCLC) can pose diagnostic problems due to inter-observer variability and other limitations of histopathology. There is an interest in developing classificatory models of lung neoplasms based on the analysis of multivariate molecular data with statistical methods and/or neural networks. DNA methylation levels at 20 loci were measured in 41 SCLC and 46 NSCLC cell lines with the quantitative real-time PCR method MethyLight. The data were analyzed with artificial neural networks (ANN) and linear discriminant analysis (LDA) to classify the cell lines into SCLC or into NSCLC. Models used either data from all 20 loci, or from five significant DNA methylation loci that were selected by a step-wise back-propagation procedure (PTGS2, CALCA, MTHFR, ESR1, and CDKN2A). The data were sorted randomly by cell line into 10 different data sets, each with training and testing subsets composed of 71 and 16 of the cases, respectively. Ten ANN models were trained using the 10 data sets: five using 20 variables, and five using the five variables selected by step-wise back-propagation. The ANN models with 20 input variables correctly classified 100% of the cell lines, while the models with only five variables correctly classified 87 to 100% of cases. For comparison, 10 different LDA models were trained and tested using the same data sets with either the original data or with logarithmically transformed data. Again, half of the models used all 20 variables while the others used only the five significant variables. LDA models provided correct classifications in 62.5% to 87.5% of cases. The classifications provided by all of the different models were compared with kappa statistics, yielding kappa values ranging from 0.25 to 1.0. We conclude that ANN models based on DNA methylation profiles can objectively classify SCLC and NSCLC cells lines with substantial to perfect concordance, while LDA models based on DNA methylation profiles provide poor to substantial concordance. Our work supports the promise of ANN analysis of DNA methylation data as a powerful approach for the development of automated methods for lung cancer classification.

Molecular signature of smoking in human lung tissues.

Cigarette smoking is the leading risk factor for lung cancer. To identify genes deregulated by smoking and to distinguish gene expression changes that are reversible and persistent following smoking cessation, we carried out genome-wide gene expression profiling on nontumor lung tissue from 853 patients with lung cancer. Gene expression levels were compared between never and current smokers, and time-dependent changes in gene expression were studied in former smokers. A total of 3,223 transcripts were differentially expressed between smoking groups in the discovery set (n = 344, P < 1.29 × 10(-6)). A substantial number of smoking-induced genes also were validated in two replication sets (n = 285 and 224), and a gene expression signature of 599 transcripts consistently segregated never from current smokers across all three sets. The expression of the majority of these genes reverted to never-smoker levels following smoking cessation, although the time course of normalization differed widely among transcripts. Moreover, some genes showed very slow or no reversibility in expression, including SERPIND1, which was found to be the most consistent gene permanently altered by smoking in the three sets. Our findings therefore indicate that smoking deregulates many genes, many of which reverse to normal following smoking cessation. However, a subset of genes remains altered even decades following smoking cessation and may account, at least in part, for the residual risk of lung cancer among former smokers. Cancer Res; 72(15); 3753-63. ©2012 AACR.

Low level anti-Hu reactivity: A risk marker for small cell lung cancer?

BACKGROUND: Previous experimental and laboratory studies have implicated antibodies against Hu proteins (anti-Hu) as a potential marker for small cell lung cancer (SCLC); there are no estimates of the association between anti-Hu and SCLC using a population-based design. METHODS: We used stored plasma specimens to evaluate anti-Hu reactivity in relationship to small cell lung cancer in a population-based case-control study. Using Western Blot analysis, we measured anti-Hu reactivity against recombinant Hu family member, HuD, in plasma samples from 41 SCLC cases and 79 controls individually matched for age, race, sex, and smoking status (never, past, current). We analyzed the association between anti-Hu reactivity and SCLC using conditional logistic regression. RESULTS: Anti-Hu reactivity was associated with SCLC, both before and after adjustment for amount of smoking. We observed a smoking-adjusted odds ratio of 3.2 (95% confidence interval from 0.98 to 13.4) comparing subjects above 1800 units (the lower limit of the second tertile of the distribution among antibody positive controls) to subjects with lower reactivity. We also found suggestive evidence in follow-up of our cases that anti-Hu above 1800 units was related to longer-term survival from SCLC. The present research is the first report of anti-Hu reactivity and SCLC in a population-based study. CONCLUSIONS: Given the suggestive evidence in this study, prospective analyses to examine whether anti-Hu reactivity might predict risk of developing SCLC, or whether anti-Hu reactivity could serve as an early marker for SCLC, may be warranted.

N-terminal propeptide of type III procollagen as a biomarker of anabolic response to recombinant human GH and testosterone.

CONTEXT: Biomarkers that predict musculoskeletal response to anabolic therapies should expedite drug development. During collagen synthesis in soft lean tissue, N-terminal propeptide of type III procollagen (P3NP) is released into circulation. We investigated P3NP as a biomarker of lean body mass (LBM) and muscle strength gains in response to testosterone and GH. DESIGN: Community-dwelling older men received GnRH agonist plus 5 or 10 g testosterone gel plus 0, 3, or 5 microg recombinant human GH daily. P3NP levels were measured at baseline and wk 4, 8, 12, and 16. LBM and appendicular skeletal muscle mass (ASM) were measured by dual-energy x-ray absorptiometry. RESULTS: One hundred twelve men completed treatment; 106 underwent serum P3NP measurements. P3NP levels were higher at wk 4 than baseline (6.61 +/- 2.14 vs. 4.51 +/- 1.05, P < 0.0001) and reached plateau by wk 4 in men receiving testosterone alone. However, wk 8 P3NP levels were higher than wk 4 levels in men receiving testosterone plus recombinant human GH. Increases in P3NP from baseline to wk 4 and 16 were significantly associated with gains in LBM (r = 0.26, P = 0.007; r = 0.53, P < 0.001) and ASM (r = 0.17, P = 0.07; r = 0.40, P < 0.0001). Importantly, for participants receiving only testosterone, P3NP increases at wk 4 and 16 were related to muscle strength gains (r = 0.20, P = 0.056 and r = 0.36, P = 0.04). In stepwise regression, change in P3NP explained 28 and 30% of the change in ASM and LBM, respectively, whereas change in testosterone but not IGF-I and age provided only small improvements in the models. CONCLUSION: Early changes in serum P3NP levels are associated with subsequent changes in LBM and ASM during testosterone and GH administration. Serum P3NP may be a useful early predictive biomarker of anabolic response to GH and testosterone.