Search

Aims/hypothesis: High-mobility group box 1 (HMGB1), an evolutionarily conserved chromosomal protein, was rediscovered to be a ‘danger signal’ (alarmin) that alerts the immune system once released extracellularly. Therefore, it has been recognised contributing to the pathogenesis of autoimmune diabetes, but its exact impact on the initiation and progression of type 1 diabetes, as well as the related molecular mechanisms, are yet to be fully characterised. Methods: In the current report, we employed NOD mice as a model to dissect the impact of blocking HMGB1 on the prevention, treatment and reversal of type 1 diabetes. To study the mechanism involved, we extensively examined the characteristics of regulatory T cells (Tregs) and their related signalling pathways upon HMGB1 stimulation. Furthermore, we investigated the relevance of our data to human autoimmune diabetes. Results: Neutralising HMGB1 both delayed diabetes onset and, of particular relevance, reversed diabetes in 13 out of 20 new-onset diabetic NOD mice. Consistently, blockade of HMGB1 prevented islet isografts from autoimmune attack in diabetic NOD mice. Using transgenic reporter mice that carry a Foxp3 lineage reporter construct, we found that administration of HMGB1 impairs Treg stability and function. Mechanistic studies revealed that HMGB1 activates receptor for AGE (RAGE) and toll-like receptor (TLR)4 to enhance phosphatidylinositol 3-kinase (PI3K)–Akt–mechanistic target of rapamycin (mTOR) signalling, thereby impairing Treg stability and functionality. Indeed, high circulating levels of HMGB1 in human participants with type 1 diabetes contribute to Treg instability, suggesting that blockade of HMGB1 could be an effective therapy against type 1 diabetes in clinical settings. Conclusions/interpretation: The present data support the possibility that HMGB1 could be a viable therapeutic target to prevent the initiation, progression and recurrence of autoimmunity in the setting of type 1 diabetes., SCOPUS: ar.j, info:eu-repo/semantics/published

The currently available therapies for Alzheimer’s disease (AD) and related forms of dementia are limited by modest efficacy, adverse side effects, and the fact that they do not prevent the relentless progression of the illness. The purpose of the studies described here was to investigate the neuroprotective effects of the nicotine metabolite cotinine as well as a small series of cotinine and nicotine analogs (including stereoisomers) and to compare their effects to the four clinically prescribed AD therapies.

MicroRNAs (miRNAs), a group of small non-coding RNAs, can regulate gene expression by triggering translation repressing and/or RNA degradation. Recent studies proved that they were deeply involved in tumorigenesis. Here we show that, compared with the two normal prostate cell lines, eighteen miRNAs in the array were over-expressed and 18 showed lower levels of expression in at least 2 of 3 cancer cell lines using miRNA microarray. MiR-99a, miR-335, miR-375 and miR-625 were the most significantly over-expressed miRNAs, whereas miR-155, miR-205, miR-224, miR-422a, miR-422b, miR-452, and miR-452* (* refer to miR-452-3p) were among the most down-regulated miRNAs. Expressions of several of the most significantly differential expression miRNAs were confirmed by Northern blot analysis and real-time RT-PCR. In this study, the expression of miR-10a, miR-99a and miR-125b were also up-regulated in prostate cancer tissues compared with their own normal tissues. For miR-99a,its expression was up-modulated in prostate cancer in comparison with all three normal tissues. Whereas miR-27aand miR-205have lower level expression in cancer tissues than that in normal tissues. Some miRNAs are aberrantly expressed in human prostate cancer suggesting their involvement in the development and progression of this malignancies. Further studies of these deregulation miRNAs will help clarify their role in tumorigenesis and detect their potential clinical usefulness for early diagnosis, prognosis and therapy of prostate cancer.

Purpose: We recently showed that protein expression profiling of serum using surface-enhanced laser desorption/ionization time-of-flight mass spectrometry (SELDI-TOF MS) has potential as a diagnostic approach for detection of prostate cancer. As a parallel effort, we have been pursuing the identification of the protein(s) comprising the individual discriminatory “peaks” and evaluating their utility as potential biomarkers for prostate disease. Experimental Design: We employed liquid chromatography, gel electrophoresis and tandem mass spectroscopy to isolate and identify a protein that correlates with observed SELDI-TOF MS mass/charge (m/z) values. Immunodepletion, immunoassay, and Western analysis were used to verify that the identified protein generated the observed SELDI peak. Subsequent immunohistochemistry was used to examine the expression of the proteins in prostate tumors. Results: An 8,946 m/z SELDI-TOF MS peak was found to retain discriminatory value in each of two separate data sets with an increased expression in the diseased state. Sequence identification by liquid chromatography-MS/MS and subsequent immunoassays verified that an isoform of apolipoprotein A-II (ApoA-II) is the observed 8,946 m/z SELDI peak. Immunohistochemistry revealed that ApoA-II is overexpressed in prostate tumors. SELDI-based immunoassay revealed that an 8.9-kDa isoform of ApoA-II is specifically overexpressed in serum from individuals with prostate cancer. ApoA-II was also overexpressed in the serum of individuals with prostate cancer who have normal prostate-specific antigen (0-4.0 ng/mL). Conclusions: We have identified an isoform of ApoA-II giving rise to an 8.9K m/z SELDI “peak” that is specifically overexpressed in prostate disease. The ability of ApoA-II to detect disease in patients with normal prostate-specific antigen suggests potential utility of the marker in identifying indolent disease.

Summary. Training data in a supervised learning problem consist of the class label and its potential predictors for a set of observations. Constructing effective classifiers from training data is the goal of supervised learning. In biomedical sciences and other scientific applications, class labels may be subject to errors. We consider a setting where there are two classes but observations with labels corresponding to one of the classes may in fact be mislabeled. The application concerns the use of protein mass-spectrometry data to discriminate between serum samples from cancer and noncancer patients. The patients in the training set are classified on the basis of tissue biopsy. Although biopsy is 100% specific in the sense that a tissue that shows itself to have malignant cells is certainly cancer, it is less than 100% sensitive. Reference gold standards that are subject to this special type of misclassification due to imperfect diagnosis certainty arise in many fields. We consider the development of a supervised learning algorithm under these conditions and refer to it as partially supervised learning. Boosting is a supervised learning algorithm geared toward high-dimensional predictor data, such as those generated in protein mass-spectrometry. We propose a modification of the boosting algorithm for partially supervised learning. The proposal is to view the true class membership of the samples that are labeled with the error-prone class label as missing data, and apply an algorithm related to the EM algorithm for minimization of a loss function. To assess the usefulness of the proposed method, we artificially mislabeled a subset of samples and applied the original and EM-modified boosting (EM-Boost) algorithms for comparison. Notable improvements in misclassification rates are observed with EM-Boost.

Purpose: New and more consistent biomarkers of head and neck squamous cell carcinoma (HNSCC) are needed to improve early detection of disease and to monitor successful patient management. The purpose of this study was to determine whether a new proteomic technology could correctly identify protein expression profiles for cancer in patient serum samples. Experimental Design: Surface-enhanced laser desorption/ionization-time of flight-mass spectrometry ProteinChip system was used to screen for differentially expressed proteins in serum from 99 patients with HNSCC and 102 normal controls. Protein peak clustering and classification analyses of the surface-enhanced laser desorption/ionization spectral data were performed using the Biomarker Wizard and Biomarker Patterns software (version 3.0), respectively (Ciphergen Biosystems, Fremont, CA). Results: Several proteins, with masses ranging from 2,778 to 20,800 Da, were differentially expressed between HNSCC and the healthy controls. The serum protein expression profiles were used to develop and train a classification and regression tree algorithm, which reliably achieved a sensitivity of 83.3% and a specificity of 100% in discriminating HNSCC from normal controls. Conclusions: We propose that this technique has potential for the development of a screening test for the detection of HNSCC.

We present a method of data reduction using a wavelet transform in discriminant analysis when the number of variables is much greater than the number of observations. The method is illustrated with a prostate cancer study, where the sample size is 248, and the number of variables is 48,538 (generated using the ProteinChip technology). Using a discrete wavelet transform, the 48,538 data points are represented by 1271 wavelet coefficients. Information criteria identified 11 of the 1271 wavelet coefficients with the highest discriminatory power. The linear classifier with the 11 wavelet coefficients detected prostate cancer in a separate test set with a sensitivity of 97% and specificity of 100%.

Discovery of “signature” protein profiles that distinguish disease states (eg, malignant, benign, and normal) is a key step towards translating recent advancements in proteomic technologies into clinical utilities. Protein data generated from mass spectrometers are, however, large in size and have complex features due to complexities in both biological specimens and interfering biochemical/physical processes of the measurement procedure. Making sense out of such high-dimensional complex data is challenging and necessitates the use of a systematic data analytic strategy. We propose here a data processing strategy for two major issues in the analysis of such mass-spectrometry-generated proteomic data: (1) separation of protein “signals” from background “noise” in protein intensity measurements and (2) calibration of protein mass/charge measurements across samples. We illustrate the two issues and the utility of the proposed strategy using data from a prostate cancer biomarker discovery project as an example.

Background: The low specificity of the prostate-specific antigen (PSA) test makes it a poor biomarker for early detection of prostate cancer (PCA). Because single biomarkers most likely will not be found that are expressed by all genetic forms of PCA, we evaluated and developed a proteomic approach for the simultaneous detection and analysis of multiple proteins for the differentiation of PCA from noncancer patients. Methods: Serum samples from 386 men [197 with PCA, 92 with benign prostatic hyperplasia (BPH), and 96 healthy individuals], randomly divided into training (n = 326) and test (n = 60) sets, were analyzed by surface-enhanced laser desorption/ionization (SELDI) mass spectrometry. The 124 peaks detected by computer analyses were analyzed in the training set by a boosting tree algorithm to develop a classifier for separating PCA from the noncancer groups. The classifier was then challenged with the test set (30 PCA samples, 15 BPH samples, 15 samples from healthy men) to determine the validity and accuracy of the classification system. Results: Two classifiers were developed. The AdaBoost classifier completely separated the PCA from the noncancer samples, achieving 100% sensitivity and specificity. The second classifier, the Boosted Decision Stump Feature Selection classifier, was easier to interpret and used only 21 (compared with 74) peaks and a combination of 21 (vs 500) base classifiers to achieve a sensitivity and specificity of 97% for the test set. Conclusions: The high sensitivity and specificity achieved in this study provides support of the potential for SELDI, coupled with a bioinformatics learning algorithm, to improve the early detection/diagnosis of PCA.

Organophosphate (OP)-based pesticides have been used extensively for decades, and as a result, they have become almost ubiquitous in our environment. There is clinical and animal evidence to suggest that chronic exposures to OPs can lead to cognitive dysfunction and other neurological abnormalities, although the mechanism for these effects is unknown. We previously reported that repeated, subthreshold exposures (defined as doses not associated with signs of acute toxicity) to the commonly used OP chlorpyrifos (CPF) resulted in protracted impairments in the performance of attention and memory-related tasks in rodents as well as deficits in axonal transport ex vivo (in the sciatic nerve). Here, we investigated the effects of CPF and its active metabolite CPF oxon (CPO) on the dynamics and movement of mitochondria in rat primary cortical neurons using time-lapse imaging techniques. Exposure to CPF (1.0–20.0 μM) or CPO (5.0 nM–20.0 μM) for 1 or 24 h resulted in a concentration-dependent increase in mitochondrial length, a decrease in mitochondrial number (indicative of increased fusion events), and a decrease in their movement in axons. The changes occurred at concentrations of CPF and CPO that did not inhibit acetylcholinesterase activity (the commonly cited mechanism of acute OP toxicity), and they were not blocked by cholinergic receptor antagonists. Furthermore, the changes did not seem to be associated with direct (OP-related) effects on mitochondrial viability or function (i.e., mitochondrial membrane potential or ATP production). The results suggest that an underlying mechanism of organophosphate-based deficits in cognitive function might involve alterations in mitochondrial dynamics and/or their transport in axons.

Background Our previous work indicated that novel analogs of choline have cytoprotective effects in vitro that might be useful in neurodegenerative conditions such as Alzheimer's disease (AD). Furthermore, two lead compounds (JWB1-84-1 and JAY2-22-33) from a library of more than 50 improved cognitive performances in a transgenic mouse model of AD. The purpose of these experiments was to more specifically investigate the neuroprotective capabilities of these lead compounds both in vitro and in vivo. Results We used N2a cells which express a Swedish mutation in the amyloid precursor protein and presenilin 1 genes to investigate the effect of JWB1-84-1 and JAY2-22-33 on β-amyloid (Aβ) levels and found that both compounds significantly reduced Aβ levels. JWB1-84-1 and JAY2-22-33 also protected rat primary cortical neurons from Aβ toxicity. Subsequently, we utilized the nematode Caenorhabditis elegans (C. elegans) as an in vivo model organism to identify potential molecular targets of these compounds. In the C. elegans model of Aβ toxicity, human Aβ is expressed intracellularly in the body wall muscle. The expression and subsequent aggregation of Aβ in the muscle leads to progressive paralysis. Conclusion We found that JAY2-22-33 (but not JWB1-84-1) significantly reduced Aβ toxicity by delaying paralysis and this protective effect required both the insulin signaling pathway and nicotinic acetylcholine receptors (nAChRs).

Objective To evaluate head and neck squamous cell carcinomas (HNSCCs) for differences in protein expression between oral cavity, oropharynx, larynx, and hypopharynx subsites. Design Retrospective proteomic analysis using tissue microarray (TMA) and 2-dimensional difference gel electrophoresis (2D-DIGE). For the TMA, automated quantitative protein expression analysis was used to interrogate levels of 4 cell-cycle regulatory proteins chosen for their known roles in cancer (cyclin D1, p53, Rb, and p14). For the 2D-DIGE, lesional and normal adjacent tissues were enriched by laser capture microdissection. Total protein was extracted, analyzed by 2D-DIGE with saturation dye labeling, and evaluated for relative abundance levels of individual protein spots. Setting Two tertiary-care academic medical centers. Patients Seventy-one patients with HNSCC for TMA, and 14 patients with HNSCC with frozen tumor and normal tissue for 2D-DIGE. Results The automated quantitative analysis of protein expression analysis revealed no difference between subsite for cyclin D1, p53, Rb, or p14 expression. The 2D-DIGE study was based on 28 gels (14 cancer gels and 14 adjacent normal gels), and 732 spots were identified as matching across more than 90% of gels. Significance was evaluated based on false discovery rate (FDR) estimated from permuted data sets. There were no significant differences in protein expression between subsites (FDR greater than or equal to 30% in all instances). Conclusions Observed differences in outcomes between HNSCCs from different subsites may not reflect differences in tumor biologic characteristics between subsites. Rather, it is possible that observed clinical heterogeneity among HNSCCs may be based on other factors, such as viral vs chemical carcinogenesis.

Colon cancer is one of the most common cancers in the world. Overexpression of MDR1 mRNA and P-gp is associated with the classic multidrug resistance of colon cancer cells. In our previous study, we reported on the transient specific reversal of MDR1/P-gp-dependent multi-drug resistance by RNA interference (RNAi) in colon cancer cells. In this study, RNAi targeting the MDR1 gene stably reversed MDR1/P-gp-dependent multidrug resistance in colon cancer cells. The plasmid vectors pSilencer-#4029, encoding #4029 MDR1 siRNA, and pSilencer-#4123, encoding #4123 MDR1 siRNA, were constructed and then transfected into COLO 320DM, a colon cancer multidrug-resistant cell line. Clone cells were screened by G418 and identified by RT-PCR and Western blot analysis. Cellular viability was measured using the MTT assay. Cell cycle analysis and intracellular adriamycin accumulation were assessed by flow cytometry. MDR1 mRNA and P-gp expression in positive clone cells, those stably transfected with MDR1 siRNA, was inhibited. The IC50 values of the antitumor drugs were significantly decreased in the positive clones compared to the COLO 320DM parent cell line, and the PI/AI values of positive clones treated with antitumor drugs were significantly decreased compared to the parent cells. In addition, the intracellular adriamycin accumulation of positive clones treated with adriamycin was significantly increased compared with COLO 320DM. This demonstrates that the stable transfection of plasmid vectors encoding MDR1 siRNA can stably reverse the MDR1/P-gp-dependent multidrug resistance of colon cancer cells.

Objective To determine the sensitivity and specificity of surface-enhanced laser desorption/ionization time-of-flight mass spectrometry (SELDI-TOF-MS) for papillary thyroid carcinoma (PTC) detection. Design The SELDI-TOF-MS protein profiles of patients with PTC, patients with benign nodular disease (BND), and healthy controls were analyzed to determine the sensitivity and specificity of SELDI-TOF-MS assay for PTC detection. Data analysis was performed to process the spectral data and classify the disease status of the patients. Setting Academic tertiary care hospital. Patients Serum samples were collected prospectively from 7 patients with PTC, 8 patients with BND, and 7 healthy control volunteers. Intervention All patients diagnosed as having PTC or BND underwent thyroidectomy from October 21, 2004, to January 31, 2006. Main Outcome Measures Twenty-two serum samples were analyzed. Results Most protein peaks resolved by the SELDI-TOF-MS assay were in the range of 1 to 20 kDa. Classification tree analysis based on peak expression distinguished patients with PTC from those with BND with 85.7% sensitivity and 100% specificity. Serum samples from patients with PTC differed most significantly from those of patients with BND by the underexpression of a protein peak at 11 101 Da. Conclusions This pilot study demonstrates that proteomic analysis of serum protein profiles distinguishes patients with PTC from patients with BND with a high degree of sensitivity and specificity. Further investigation into the clinical utility of this technology in PTC biomarker detection and surveillance is warranted.

Objective: To determine the sensitivity and specificity of surface-enhanced laser desorption and ionization timeof-flight mass spectrometry (SELDI-TOF-MS) assay for head and neck squamous cell carcinoma (HNSCC) disease surveillance. Design: The SELDI-TOF-MS serum protein profiles of patients with HNSCC were analyzed to determine the sensitivity and specificity of the SELDI assay for HNSCC detection following definitive treatment. Setting: Academic research. Patients: Thirty-two patients with previously untreated HNSCC. Intervention: Serum samples were collected prospectively at 3-month intervals following treatment during a 24-month follow-up period. Main Outcome Measures: Ninety-three serum samples were analyzed. Results: The SELDI-TOF-MS identified protein peaks in the range of 0 to 100 kDa. Classification tree analysis based onpeakexpressiondistinguishedpretreatmentfrom6-month posttreatmentsampleswith75.0%sensitivityand87.5%specificity. Samples collected at 3 months following treatment did not significantly differ from pretreatment samples. Serum samples from patients who were disease free at 6 months or longer following treatment differed from matched pretreatmentsamplesbytheoverexpressionofaproteinpeakat6495 Da, while serum samples from patients with recurrence differed from matched pretreatment samples by the underexpression of a protein peak at 4493 Da. Conclusions: Proteomic analysis of serum protein profiles distinguishes pretreatment and posttreatment samples from patients with HNSCC with a high degree of sensitivity and specificity. After 6 months, serum protein profiles seem to have distinct differences in peak expression based on disease status. Further investigation of the clinical usefulness of this technology in HNSCC detection and surveillance is warranted.

Motivation: Due to advances in experimental technologies, such as microarray, mass spectrometry and nuclear magnetic resonance, it is feasible to obtain large-scale data sets, in which measurements for a large number of features can be simultaneously collected. However, the sample sizes of these data sets are usually small due to their relatively high costs, which leads to the issue of concordance among different data sets collected for the same study: features should have consistent behavior in different data sets. There is a lack of rigorous statistical methods for evaluating this concordance or discordance.Methods: Based on a three-component normal-mixture model, we propose two likelihood ratio tests for evaluating the concordance and discordance between two large-scale data sets with two sample groups. The parameter estimation is achieved through the expectation-maximization (E-M) algorithm. A normal-distribution-quantile-based method is used for data transformation.Results: To evaluate the proposed tests, we conducted some simulation studies, which suggested their satisfactory performances. As applications, the proposed tests were applied to three SELDI-MS data sets with replicates. One data set has replicates from different platforms and the other two have replicates from the same platform. We found that data generated by SELDI-MS showed satisfactory concordance between replicates from the same platform but unsatisfactory concordance between replicates from different platforms.Availability: The R codes are freely available at http://home.gwu.edu/~ylai/research/ConcordanceContact: ylai@gwu.edu

Proteomic profiling of serum is an emerging technique to identify new biomarkers indicative of disease severity and progression. The objective of our study was to assess the use of surface-enhanced laser desorption/ionization time-of-flight mass spectrometry (SELDI-TOF MS) to identify multiple serum protein biomarkers for detection of liver disease progression to hepatocellular carcinoma (HCC). A cohort of 170 serum samples obtained from subjects in the United States with no liver disease (n = 39), liver diseases not associated with cirrhosis (n = 36), cirrhosis (n = 38), or HCC (n = 57) were applied to metal affinity protein chips for protein profiling by SELDI-TOF MS. Across the four test groups, 38 differentially expressed proteins were used to generate multiple decision classification trees to distinguish the known disease states. Analysis of a subset of samples with only hepatitis C virus (HCV)-related disease was emphasized. The serum protein profiles of control patients were readily distinguished from each HCV-associated disease state. Two-way comparisons of chronic hepatitis C, HCV cirrhosis, or HCV-HCC versus healthy had a sensitivity/specificity range of 74% to 95%. For distinguishing chronic HCV from HCV-HCC, a sensitivity of 61% and a specificity of 76% were obtained. However, when the values of known serum markers alpha fetoprotein, des-gamma carboxyprothrombin, and GP73 were combined with the SELDI peak values, the sensitivity and specifity improved to 75% and 92%, respectively. In conclusion, SELDI-TOF MS serum profiling is able to distinguish HCC from liver disease before cirrhosis as well as cirrhosis, especially in patients with HCV infection compared with other etiologies.

Background: Protein expression profiling for differences indicative of early cancer has promise for improving diagnostics. This report describes the first stage of a National Cancer Institute/Early Detection Research Network-sponsored multiinstitutional evaluation and validation of this approach for detection of prostate cancer. Methods: Two sequential experimental phases were conducted to establish interlaboratory calibration and standardization of the surface-enhanced laser desorption (SELDI) instrumental and assay platform output. We first established whether the output from multiple calibrated Protein Biosystem II SELDI-ionization time-of-flight mass spectrometry (TOF-MS) instruments demonstrated acceptable interlaboratory reproducibility. This was determined by measuring mass accuracy, resolution, signal-to-noise ratio, and normalized intensity of three m/z “peaks” present in a standard pooled serum sample. We next evaluated the ability of the calibrated and standardized instrumentation to accurately differentiate between selected cases of prostate cancer and control by use of an algorithm developed from data derived from a single site 2 years earlier. Results: When the described standard operating procedures were established at all laboratory sites, the across-laboratory measurements revealed a CV for mass accuracy of 0.1%, signal-to-noise ratio of ∼40%, and normalized intensity of 15–36% for the three pooled serum peaks. This was comparable to the intralaboratory measurements of the same peaks. The instrument systems were then challenged with sera from a selected group of 14 cases and 14 controls. The classification agreement between each site and the established decision algorithm were examined by use of both raw peak intensity boosting and ranked peak intensity boosting. All six sites achieved perfect blinded classification for all samples when boosted alignment of raw intensities was used. Four of six sites achieved perfect blinded classification with ranked intensities, with one site passing the criteria of 26 of 28 correct and one site failing with 19 of 28 correct. Conclusions: These results demonstrate that “between-laboratory” reproducibility of SELDI-TOF-MS serum profiling approaches that of “within-laboratory” reproducibility as determined by measuring discrete m/z peaks over time and across laboratories.

Multiple studies have reported that analysis of serum and other bodily fluids using surface enhanced laser desorption/ionization time of flight mass spectroscopy (SELDI-TOF-MS) can identify a “fingerprint” or “signature” of spectral peaks that can separate patients with a specific disease from normal control patients. Ultimately, classification by SELDI-TOF-MS relies on spectral differences in position and amplitude of resolved peaks. Since the reproducibility of quantitation, resolution and mass accuracy of the SELDI-TOF-MS, or any high throughput mass spectrometric technique, has never been determined this method has come under some skepticism as to its clinical usefulness. This manuscript describes a detailed design of a three-phase study to validate the clinical usefulness of SELDI-TOF-MS in the identification of patients with prostatic adenocarcinoma (PCA). At the end of this validation study, the usefulness of the general SELDI-TOF-MS approach to identifying patients with PCA will be demonstrated and how it compares with PCA diagnosis by measuring prostate specific antigen.

With recent advances in mass spectrometry techniques, it is now possible to investigate proteins over a wide range of molecular weights in small biological specimens. This advance has generated data-analytic challenges in proteomics, similar to those created by microarray technologies in genetics, namely, discovery of 'signature' protein profiles specific to each pathologic state (e.g. normal vs. cancer) or differential profiles between experimental conditions (e.g. treated by a drug of interest vs. untreated) from high-dimensional data. We propose a data-analytic strategy for discovering protein biomarkers based on such high-dimensional mass spectrometry data. A real biomarker-discovery project on prostate cancer is taken as a concrete example throughout the paper: the project aims to identify proteins in serum that distinguish cancer, benign hyperplasia, and normal states of prostate using the Surface Enhanced Laser Desorption/Ionization (SELDI) technology, a recently developed mass spectrometry technique. Our data-analytic strategy takes properties of the SELDI mass spectrometer into account: the SELDI output of a specimen contains about 48,000 (x, y) points where x is the protein mass divided by the number of charges introduced by ionization and y is the protein intensity of the corresponding mass per charge value, x, in that specimen. Given high coefficients of variation and other characteristics of protein intensity measures (y values), we reduce the measures of protein intensities to a set of binary variables that indicate peaks in the y-axis direction in the nearest neighborhoods of each mass per charge point in the x-axis direction. We then account for a shifting (measurement error) problem of the x-axis in SELDI output. After this pre-analysis processing of data, we combine the binary predictors to generate classification rules for cancer, benign hyperplasia, and normal states of prostate. Our approach is to apply the boosting algorithm to select binary predictors and construct a summary classifier. We empirically evaluate sensitivity and specificity of the resulting summary classifiers with a test dataset that is independent from the training dataset used to construct the summary classifiers. The proposed method performed nearly perfectly in distinguishing cancer and benign hyperplasia from normal. In the classification of cancer vs. benign hyperplasia, however, an appreciable proportion of the benign specimens were classified incorrectly as cancer. We discuss practical issues associated with our proposed approach to the analysis of SELDI output and its application in cancer biomarker discovery.

The low specificity of the prostate-specific antigen (PSA) test makes it a poor biomarker for early detection of prostate cancer (PCA). Because single biomarkers most likely will not be found that are expressed by all genetic forms of PCA, we evaluated and developed a proteomic approach for the simultaneous detection and analysis of multiple proteins for the differentiation of PCA from noncancer patients.Serum samples from 386 men [197 with PCA, 92 with benign prostatic hyperplasia (BPH), and 96 healthy individuals], randomly divided into training (n = 326) and test (n = 60) sets, were analyzed by surface-enhanced laser desorption/ionization (SELDI) mass spectrometry. The 124 peaks detected by computer analyses were analyzed in the training set by a boosting tree algorithm to develop a classifier for separating PCA from the noncancer groups. The classifier was then challenged with the test set (30 PCA samples, 15 BPH samples, 15 samples from healthy men) to determine the validity and accuracy of the classification system.Two classifiers were developed. The AdaBoost classifier completely separated the PCA from the noncancer samples, achieving 100% sensitivity and specificity. The second classifier, the Boosted Decision Stump Feature Selection classifier, was easier to interpret and used only 21 (compared with 74) peaks and a combination of 21 (vs 500) base classifiers to achieve a sensitivity and specificity of 97% for the test set.The high sensitivity and specificity achieved in this study provides support of the potential for SELDI, coupled with a bioinformatics learning algorithm, to improve the early detection/diagnosis of PCA.

The objective of this study was to discover protein biomarkers that differentiate malignant from nonmalignant cell populations, especially early protein alterations that signal the initiation of a developing cancer. We hypothesized that Surface Enhanced Laser Desorption/Ionization-time of flight-mass spectrometry-assisted protein profiling could detect these protein alterations.Epithelial cell populations [benign prostatic hyperplasia (BPH), prostate intraepithelial neoplasia (PIN), and prostate cancer (PCA)] were procured from nine prostatectomy specimens using laser capture microdissection. Surface Enhanced Laser Desorption/Ionization-time of flight-mass spectrometry analysis was performed on cell lysates, and the relative intensity levels of each protein or peptide in the mass spectra was calculated and compared for each cell type.Several small molecular mass peptides or proteins (3000-5000 Da) were found in greater abundance in PIN and PCA cell lysates. Another peak, with an average mass of 5666 Da, was observed to be up-regulated in 86% of the BPH cell lysates. Higher levels of this same peak were found in only 22% of the PIN lysates and none of the PCA lysates. Expression differences were also found for intracellular levels of prostate-specific antigen, which were reduced in PIN and PCA cells when compared with matched normals. Although no single protein alteration was observed in all PIN/PCA samples, combining two or more of the markers was effective in distinguishing the benign cell types (normal/BPH) from diseased cell types (PIN/PCA). Logistic regression analysis using seven differentially expressed proteins resulted in a predictive equation that correctly distinguished the diseased lysates with a sensitivity and specificity of 93.3 and 93.8%, respectively.We have shown that the protein profiles from prostate cells with different disease states have discriminating differences. These differentially regulated proteins are potential markers for early detection and/or risk factors for development of prostate cancer. Studies are under way to identify these protein/peptides, with the goal of developing a diagnostic test for the early detection of prostate cancer.

The prostate-specific antigen test has been a major factor in increasing awareness and better patient management of prostate cancer (PCA), but its lack of specificity limits its use in diagnosis and makes for poor early detection of PCA. The objective of our studies is to identify better biomarkers for early detection of PCA using protein profiling technologies that can simultaneously resolve and analyze multiple proteins. Evaluating multiple proteins will be essential to establishing signature proteomic patterns that distinguish cancer from noncancer as well as identify all genetic subtypes of the cancer and their biological activity. In this study, we used a protein biochip surface enhanced laser desorption/ionization mass spectrometry approach coupled with an artificial intelligence learning algorithm to differentiate PCA from noncancer cohorts. Surface enhanced laser desorption/ionization mass spectrometry protein profiles of serum from 167 PCA patients, 77 patients with benign prostate hyperplasia, and 82 age-matched unaffected healthy men were used to train and develop a decision tree classification algorithm that used a nine-protein mass pattern that correctly classified 96% of the samples. A blinded test set, separated from the training set by a stratified random sampling before the analysis, was used to determine the sensitivity and specificity of the classification system. A sensitivity of 83%, a specificity of 97%, and a positive predictive value of 96% for the study population and 91% for the general population were obtained when comparing the PCA versus noncancer (benign prostate hyperplasia/healthy men) groups. This high-throughput proteomic classification system will provide a highly accurate and innovative approach for the early detection/diagnosis of PCA.

Improving early detection, diagnosis, treatment monitoring and prognosis of cancer will require rapid and high throughput detection, identification, and measurement of multiple biomarkers. In this study, we demonstrate the versatility of the innovative SELDI ProteinChip(R) MS technology for the rapid, reproducible and simultaneous identification of four well-characterized prostate cancer-associated (PCA) biomarkers, prostate specific antigen (free and complexed forms), prostate specific peptide, prostate acid phophatase and prostate specific membrane antigen in cell lysates, serum and seminal plasma. Proteins corresponding to the mass of these biomarkers could readily be captured and detected using either chemically defined or antibody coated ProteinChip(R) arrays. Several (yet to be identified) proteins were found upregulated in cell lysates of pure populations of PCA cells procured by laser capture microdissection (LCM) when compared with mass spectra of normal cell lysates. Coupling LCM with SELDI provides tremendous opportunities to discover and identify the signature proteins associated with each stage of tumor development. Collectively, these observations demonstrate the potential of SELDI for the discovery and simultaneous detection of and clinical assay development for PCA biomarkers in complex biological mixtures.

No control cell line was available for previous RNA interference studies on reversal of multidrug resistance (MDR) in colon cancer cells. Here, human COLO 320DM, with HT-29 as the control, colon cancer cell lines were used to investigate the reversal of MDR1/P-gp-dependent MDR by siRNA (#4123 and #4029 MDR1 siRNAs) targeting to MDR1 mRNA. Both siRNAs inhibited expression of MDR1 and P-gp in COLO 320DM. The minimum inhibition concentrations were 5 nmol/l of #4123 and 25 nmol/l of #4029. #4123 MDR1 siRNA took effect in 4, 5 and 6 days at doses of 5, 25 and 100 nmol/l, respectively. Increased cytotoxicity of the antitumor drugs adriamycin and vincristine with increased intracellular adriamycin accumulation accompanied inhibition of MDR1 mRNA and P-gp expression. No such effects were found in the HT-29 control. MDR1 siRNAs specifically reversed the MDR of colon cancer cells demonstrating a possible new approach for treating MDR1/P-gp-dependent multidrug resistance.

Head and neck squamous cell carcinoma (HNSCC) accounts for more than 37,000 new cases in the US each year. More than half of the patients present in an advanced stage, which confers a poor prognosis. However if detected early this cancer maybe highly curable. Currently there is no biomarker in clinical practice for early detection or prognosis prediction. Many studies are attempting to discover potential biomarkers in tissue, saliva and serum. Promoter methylation suppresses gene expression in cancer and several candidates have been demonstrated to be cancer specific biomarkers. Increased mitochondrial content is another feature that has been demonstrated related to cancer. Our goal is to validate potential biomarkers previously described by our group and others to be linked to HNSCC, in serum samples from a cohort of patients with longitudinal follow up and a subset of normal subjects’ samples. We assessed mitochondrial DNA content (Cox I and Cox II) and methylation status of EDNRB and KIF1a by quantitative PCR in an evaluation set of samples consisting of 75 pre and post treatment serum samples from HNSCC patients and from 11 healthy cancer-free subjects. In tumors, the observed methylation frequencies were 17%, 22% for EDNRB and KIF1a, respectively. There was no observed methylation for EDNRB and KIF1a in normal samples. Increased mitochondrial DNA content was observed in 54% and 59% of the HNSCC samples, by COX1 and COX2, respectively. And no increased content was noted in the normal set. Combining all the markers and analyzing as a panel (at least one positive), we obtained a sensitivity of 83% with specificity of 100%. When we compared post treatment samples to pre treatment samples, we noticed a decreased mitochondrial DNA content in patients without recurrence, and an increase in patients that recurred, however, due to the small number of samples, these associations needs to be confirmed. Our findings indicate that a serum based panel that includes mitochondrial DNA content analysis and cancer specific methylation markers may serve as a high quality detection tool for HNSCC, as well as recurrence prediction. Validation of these preliminary data is ongoing in a larger independent cohort with comprehensive clinical and follow up data. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 102nd Annual Meeting of the American Association for Cancer Research; 2011 Apr 2-6; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2011;71(8 Suppl):Abstract nr 3170. doi:10.1158/1538-7445.AM2011-3170

Objectives To analyze surface-enhanced laser desorption and ionization time-of-flight mass spectrometry (SELDI-TOF-MS) protein profiles of patients with head and neck squamous cell carcinoma (HNSCC) and healthy controls and to determine the sensitivity and specificity of SELDI assay for HNSCC detection before and after treatment. Design Proteomic analysis and comparison of serum samples. Setting Tertiary care academic medical center. Subjects Seventy-eight patients with HNSCC and 68 healthy controls. Main Outcome Measures Serum samples were prospectively collected from 78 patients with HNSCC and 68 healthy control volunteers. SELDI-TOF-MS was performed on serum samples to identify protein peaks in the range of 0 to 100 kDa. Classification analysis of the spectral data was performed and used to classify the disease status of the patients. Results The SELDI-TOF-MS assay generated serum protein profiles ranging from 0 to 100 kDa. After background subtraction, mass calibration, and normalization, 545 protein peaks were identified. Classification tree analysis based on peak expression correctly classified patients with HNSCC with 82% sensitivity and 76% specificity. Subgroup analysis correctly classified 83% of oral cavity tumors, 81% of oropharyngeal tumors, and 88% of laryngeal tumors. Pretreatment and posttreatment samples were available from 12 patients, and the posttreatment samples were correctly classified in 86% of the patients at 3 months and 75% of the patients at 6 months. Conclusions Proteomic SELDI-TOF-MS analysis of serum protein profiles distinguishes patients with HNSCC from controls with a high degree of sensitivity and specificity. Further investigation into the clinical utility of this technology in HNSCC detection and surveillance is warranted.

Background: New and more consistent biomarkers of head and neck squamous cell carcinoma (HNSCC) are needed to improve early detection of disease and to monitor successful patient management. Objective: To determine if a new proteomic technology can correctly identify protein expression profiles for cancer in patient serum samples as well as detect the presence of a known tumor marker. Design: Direct proteomic analysis and comparison. Methods: The surface-enhanced laser desorption/ ionization time of flight mass spectrometry (SELDI-TOF) ProteinChip system was used to screen for differentially expressed proteins in serum samples from 99 patients with HNSCC, 25 “healthy” smokers, and 102 healthy (normal) controls. Protein peak clustering and classification analyses of the SELDI spectral data were performed. Results: Several proteins, with masses ranging from 2778 to 20 800 Da, were differentially expressed between patients with HNSCC and the normal controls. The serum protein expression profiles were used to develop a classification tree algorithm, which achieved a sensitivity of 83.3% and a specificity of 90% in discriminating HNSCC from normal and healthy smoker controls. The positive and negative predictive values were 80% and 92%, respectively. A peak with an average mass of 10 068 Da was detected in sera from HNSCC patients and identified as the known biomarker metallopanstimulin-1 (MPS-1), based on mass. Peak relative intensity of the 10 068-Da protein correlated consistently with MPS-1 levels detected by radioimmunoassay in serum samples of HNSCC patients and controls. The 10068-Da peak was provisionally identified as MPS-1 by SELDI immunoassay. Conclusion: We propose that this technique may allow for the development of a reliable screening test for the early detection and diagnosis of HNSCC, as well as the potential identification of tumor biomarkers. Arch Otolaryngol Head Neck Surg. 2004;130:98-104

Background Recent technological advances in mass spectrometry pose challenges in computational mathematics and statistics to process the mass spectral data into predictive models with clinical and biological significance. We discuss several classification-based approaches to finding protein biomarker candidates using protein profiles obtained via mass spectrometry, and we assess their statistical significance. Our overall goal is to implicate peaks that have a high likelihood of being biologically linked to a given disease state, and thus to narrow the search for biomarker candidates. Results Thorough cross-validation studies and randomization tests are performed on a prostate cancer dataset with over 300 patients, obtained at the Eastern Virginia Medical School using SELDI-TOF mass spectrometry. We obtain average classification accuracies of 87% on a four-group classification problem using a two-stage linear SVM-based procedure and just 13 peaks, with other methods performing comparably. Conclusions Modern feature selection and classification methods are powerful techniques for both the identification of biomarker candidates and the related problem of building predictive models from protein mass spectrometric profiles. Cross-validation and randomization are essential tools that must be performed carefully in order not to bias the results unfairly. However, only a biological validation and identification of the underlying proteins will ultimately confirm the actual value and power of any computational predictions.