Multi-parametric cytometry from a complex cellular sample: Improvements and limits of manual versus computational-based interactive analyses

The wide possibilities opened by the developments of multi-parametric cytometry are limited by the inadequacy of the classical methods of analysis to the multi-dimensional characteristics of the data. While new computational tools seemed ideally adapted and were applied successfully, their adoption is still low among the flow cytometrists. In the purpose to integrate unsupervised computational tools for the management of multi-stained samples, we investigated their advantages and limits by comparison to manual gating on a typical sample analyzed in immunomonitoring routine. A single tube of PBMC, containing 11 populations characterized by different sizes and stained with 9 fluorescent markers, was used. We investigated the impact of the strategy choice on manual gating variability, an undocumented pitfall of the analysis process, and we identified rules to optimize it. While assessing automatic gating as an alternate, we introduced the Multi-Experiment Viewer software (MeV) and validated it for merging clusters and annotating interactively populations. This procedure allowed the finding of both targeted and unexpected populations. However, the careful examination of computed clusters in standard dot plots revealed some heterogeneity, often below 10%, that was overcome by increasing the number of clusters to be computed. MeV facilitated the identification of populations by displaying both the MFI and the marker signature of the dataset simultaneously. The procedure described here appears fully adapted to manage homogeneously high number of multi-stained samples and allows improving multi-parametric analyses in a way close to the classic approach. © 2016 International Society for Advancement of Cytometry.

Postoperative serum proteomic profiles may predict metastatic relapse in high-risk primary breast cancer patients receiving adjuvant chemotherapy

A total of 30-50% of early breast cancer (EBC) patients considered as high risk using standard prognostic factors develop metastatic recurrence despite standard adjuvant systemic treatment. A means to better predict clinical outcome is needed to optimize and individualize therapeutic decisions. To identify a protein signature correlating with metastatic relapse, we performed surface-enhanced laser desorption/ionization-time of flight mass spectrometry profiling of early postoperative serum from 81 high-risk EBC patients. Denatured and fractionated serum samples were incubated with IMAC30 and CM10 ProteinChip arrays. Several protein peaks were differentially expressed according to clinical outcome. By combining partial least squares and logistic regression methods, we built a multiprotein model that correctly predicted outcome in 83% of patients. The 5-year metastasis-free survival in 'good prognosis' and 'poor prognosis' patients as defined using the multiprotein index were strikingly different (83 and 22%, respectively; P<0.0001, log-rank test). In a multivariate Cox regression including conventional pathological factors and multiprotein index, the latter retained the strongest independent prognostic significance for metastatic relapse. Major components of the multiprotein index included haptoglobin, C3a complement fraction, transferrin, apolipoprotein C1 and apolipoprotein A1. Therefore, postoperative serum protein pattern may have an important prognostic value in high-risk EBC.

Feature extraction and signal processing for nylon DNA microarrays

Background

High-density DNA microarrays require automatic feature extraction methodologies and softwares. These can be a potential source of non-reproducibility of gene expression measurements. Variation in feature location or in signal integration methodology may be a significant contribution to the observed variance in gene expression levels.

Results

We explore sources of variability in feature extraction from DNA microarrays on Nylon membrane with radioactive detection. We introduce a mathematical model of the signal emission and derive methods for correcting biases such as overshining, saturation or variation in probe amount. We also provide a quality metric which can be used qualitatively to flag weak or untrusted signals or quantitatively to modulate the weight of each experiment or gene in higher level analyses (clustering or discriminant analysis).

Conclusions

Our novel feature extraction methodology, based on a mathematical model of the radioactive emission, reduces variability due to saturation, neighbourhood effects and variable probe amount. Furthermore, we provide a fully automatic feature extraction software, BZScan, which implements the algorithms described in this paper.

Vanin genes are clustered (human 6q22-24 and mouse 10A2B1) and encode isoforms of pantetheinase ectoenzymes

The mouse Vanin-1 molecule plays a role in thymic reconstitution following damage by irradiation. We recently demonstrated that it is a membrane pantetheinase (EC 3.56.1.-). This molecule is the prototypic member of a larger Vanin family encoded by at least two mouse (Vanin-1 and Vanin-3) and three human (VNN1, VNN2, VNN3) orthologous genes. We now report (1) the structural characterization of the human and mouse Vanin genes and their organization in clusters on the 6q22-24 and 10A2B1 chromosomes, respectively; (2) identification of the human VNN3 gene and the demonstration that the mouse Vanin-3 molecule is secreted by cells, and (3) that the Vanin genes encode different isoforms of the mammalian pantetheinase activity. Thus, the Vanin family represents a novel class of secreted or membrane-associated ectoenzymes. We discuss here their possible role in processes pertaining to tissue repair in the context of oxidative stress.

[DNA arrays: technological aspects and applications]

The Human Genome Project has allowed considerable progress in the construction of physical and genetic maps and the identification of genes involved in human sicknesses. The accelerated accumulation of biological information and knowledge is due in large part to the sequencing projects of other organisms, which in fact paved the way for the Human Genome Project. In parallel, recently developed techniques which take advantage of genomic sequences allow large scale molecular analyses resulting in the functional annotation of many of the proteins represented by these genes. This is the goal of functional genomics. These progresses are at the origin of the present revolution in biomedical research. DNA microarrays are playing a dominant role compared to the other developing technologies since they are relatively easy to make and use and are applicable to numerous scientific inquiries. They allow the simultaneous analysis of several thousands of genes in biological samples from sick or healthy tissues, at the genome or transcriptome level. The data obtained is expected to result in major advances in the health sciences. In addition to an improved understanding of the complex molecular interaction networks of healthy cells and tissues, a more precise genetic characterization of the molecular mechanisms involved in pathology should result in the identification of new therapeutic targets and the development of new medicines. The genetic profiles thus obtained should also permit the definition of new pathologic subclasses not recognizable by traditional clinical factors, as well as new markers for susceptibility to certain illnesses, and new prognostic markers or methods of predicting responses to treatment. In this article, we present the different approaches and potential applications of DNA microarray technology, in particular as applied to cancer research.

[Molecular typing of breast cancer: transcriptomics and DNA microarrays]

Breast cancer is the most frequent and deadly cancer of women. Its great heterogeneity makes prognosis and response to current treatments highly variable and difficult to predict. Mammary oncogenesis remains poorly understood. These issues should benefit from recent development of techniques capable of large-scale molecular analyses. The use of cDNA array techniques allows for the simultaneous analysis of the mRNA expression levels of thousands of genes in mammary tumor cell lines and breast tumors. Expression profiles will help classify tumors and provide new prognostic tools and potential therapeutic targets. They will also boost our knowledge of the molecular events responsible for the development and progression of this cancer.

Gene expression profiling of primary breast carcinomas using arrays of candidate genes

Breast cancer is characterized by an important histoclinical heterogeneity that currently hampers the selection of the most appropriate treatment for each case. This problem could be solved by the identification of new parameters that better predict the natural history of the disease and its sensitivity to treatment. A large-scale molecular characterization of breast cancer could help in this context. Using cDNA arrays, we studied the quantitative mRNA expression levels of 176 candidate genes in 34 primary breast carcinomas along three directions: comparison of tumor samples, correlations of molecular data with conventional histoclinical prognostic features and gene correlations. The study evidenced extensive heterogeneity of breast tumors at the transcriptional level. A hierarchical clustering algorithm identified two molecularly distinct subgroups of tumors characterized by a different clinical outcome after chemotherapy. This outcome could not have been predicted by the commonly used histoclinical parameters. No correlation was found with the age of patients, tumor size, histological type and grade. However, expression of genes was differential in tumors with lymph node metastasis and according to the estrogen receptor status; ERBB2 expression was strongly correlated with the lymph node status (P < 0.0001) and that of GATA3 with the presence of estrogen receptors (P < 0.001). Thus, our results identified new ways to group tumors according to outcome and new potential targets of carcinogenesis. They show that the systematic use of cDNA array testing holds great promise to improve the classification of breast cancer in terms of prognosis and chemosensitivity and to provide new potential therapeutic targets.

Differential gene expression in CD3epsilon- and RAG1-deficient thymuses: definition of a set of genes potentially involved in thymocyte maturation

A set of 3000 mouse thymus cDNAs was analyzed by extensive measurement of expression using complex-probe hybridization of DNA arrays ("quantitative differential screening"). The complex probes were initially prepared using total thymus RNA isolated from C57BL/6 wild-type (WT), CD3epsilon- and RAG1-deficient mice. Over 100 clones displaying over- or under-expression by at least a factor of two between WT and knockout (KO) thymuses were further analyzed by measuring hybridization signatures with probes from a wide range of KO thymuses, cell types, organs, and embryonic thymuses. A restricted set of clones was selected by virtue of their expression spectra (modulation in KO thymuses and thymocytes, lymphoid cell specificity, and differential expression during embryonic thymus development), sequenced at one extremity, and compared to sequences in databases. Clones corresponding to previously identified genes (e.g., Tcrbeta, Tcf1 or CD25) showed expression patterns that were consistent with existing data. Ten distinct clones corresponding to new genes were subjected to further study: Northern blot hybridization, in situ hybridization on thymus sections, and partial or complete mRNA sequence determination. Among these genes, we report a new serine peptidase highly expressed in cortical epithelial cells that we have named thymus-specific serine peptidase (TSSP), and an acidic protein expressed in thymocytes and of unknown function that we have named thymus-expressed acidic protein (TEAP). This approach identifies new molecules likely to be involved in thymocyte differentiation and function.

An ESTs description of the new Vanin gene family conserved from fly to human

Circulation and tissue colonization are essential properties of lymphoid cells and involve major families of adhesion molecules (e.g. , integrin, selectin, mucin-like, and molecules from the immunoglobulin superfamily). The mouse Vanin-1 molecule was recently identified and found to be involved in the colonization of the thymus by hematopoietic precursor cells. Here we show based on computational analysis of EST sequence database resources that Vanin-1 belongs to a new family of related molecules present from drosophila to human. This family includes the amidase enzyme Biotinidase, and a central protein domain is shared between Vanin and Nitrilase families, suggesting that Vanin molecules might bear an enzymatic activity. Five of these molecules were new uncharacterized cDNA sequences only described as ESTs. The three human Vanin genes map to the same region of Chromosome 6q. The detailed results are consultable at the VANIN web page (http://tagc. univ-mrs.fr/pub/vanin/).

Sensitivity issues in DNA array-based expression measurements and performance of nylon microarrays for small samples

DNA or oligonucleotide arrays are widely used for large-scale expression measurements, using various implementations: macroarrays in which DNA is spotted onto nylon membranes of relatively large dimensions (with radioactive detection) on the one hand; microarrays on glass slides and oligonucleotide chips, both used with fluorescent probes, on the other hand. Nylon micro-arrays with colourimetric detection have also been described recently. The small physical dimensions of miniaturized systems allow small hybridization volumes (2-100 microl) and provide high probe concentrations, in contrast to macroarrays. We show, however, that actual sensitivity (defined as the amount of sample necessary for detection of a given mRNA species) is in fact similar for all these systems and that this is mostly due to the very different amounts of target material present on the respective arrays. We then demonstrate that the combination of nylon microarrays with(33)P-labelled radioactive probes provides 100-fold better sensitivity, making it possible to perform expression profiling experiments using submicrogram amounts of unamplified total RNA from small biological samples. This has important implications in basic and clinical research and makes this alternative approach particularly suitable for groups operating in an academic context.

Expression profiling: DNA arrays in many guises

DNA arrays have become the preferred method for large-scale expression measurement. Such data are needed in view of the large amounts of sequence data available: expression levels in a number of different tissues or situations provide a first step toward functional characterisation of new entities revealed by DNA sequencing. Although the basic principle of measurement is in all cases based on hybridisation of a mixed probe derived from tissue RNA to large sets of DNA fragments representing many genes, a number of different forms of implementation of this principle are at hand. They are briefly described and compared, emphasizing the important issue of sensitivity and sample requirements and the accessibility of the methods to academic scientists. When these factors are taken into account, it appears that, contrary to a largely prevalent impression, the "best" approach is not necessarily always provided by the widely advertised glass microarrays or oligonucleotide chips.

Expression scanning of an array of growth control genes in human tumor cell lines

Analysis of gene expression on a medium- or large-scale is an increasingly recognized method for functional and clinical investigations based on the now extensive catalog of known or partially sequenced genes. The accessibility of this approach can be enhanced by using readily available technology (macroarrays on Nylon, radioactive detection) and the IMAGE resource to assemble sets of targets. We have set up such a medium-scale, flexible system and validated it by the study of quantitative expression levels for 120 genes in six cell lines, including three mammary carcinoma cell lines. A number of important parameters are identified as necessary for the assembly of a valid set and the obtention of good-quality quantitative data. The extensive data assembled in this survey identified potential targets of carcinogenesis, for example the CRABP2 and GATA3 transcription factor genes. We also demonstrate the feasibility of this procedure for relatively small tumor samples, without recourse to probe amplification methods.

Engagement of natural cytotoxicity programs regulates AP-1 expression in the NKL human NK cell line

NK cell cytotoxicity is a fast and efficient mechanism of target cell lysis. Using transcription analysis, such as multiplex messenger assays, we show here that natural cytotoxicity exerted by the human NKL cell line correlates with mRNA accumulation of very early activator protein (AP)-1 transcription factor genes such as JunB, FosB and c-Fos. In addition, DNA-binding activities of Jun-Fos heterodimers were observed by electrophoretic mobility shift assays during the course of natural cytotoxicity. Interaction between immunoglobulin-like transcript-2/leukocyte Ig-like receptor 1 on NKL cells and HLA-B27 on target cells leads to an impairment of NKL natural cytotoxicity, which correlates with an absence of JunB, FosB, and c-Fos transcription, as well as an absence of their DNA-binding activity. Our studies thus indicate that, despite the rapidity of NK cell-mediated lysis, AP-1 transcription factor is activated during the early stage of NK cell cytolytic programs and that engagement of NK cell inhibitory receptors for MHC class I molecules impairs the very early activation of AP-1.

'LABNOTE', a laboratory notebook system designed for academic genomics groups

We have developed a relational laboratory database system, adapted to the daily book-keeping needs of laboratories that must keep track of information acquired on hundreds or thousands of clones in an effective and user-friendly fashion. Data, whether final or related to experiments in progress, can be accessed in many different ways, e.g. by clone name, by gene, by experiment or through DNA sequence. Updating, import and export of results is made easier by specially developed tools. This system, in network version, serves several groups in our Institute and (over the Internet) elsewhere, and is instrumental in collaborative studies based on expression profiling. It can be used in many similar situations involving progressiveaccumulation of information on sets of clones or related objects.

Differential mRNA expression in untreated and TNF-alpha elicited murine dendritic cells precursors

We have compared the pattern of gene expression in long term cultured precursor dendritic cells (DC), either untreated (immature) or cultured for two days in the presence of recombinant murine (rm)-TNF alpha (mature). The hybridization signature of complex cDNA probes prepared from total RNA extracted from immature and mature DC were analyzed using a mouse thymic cDNA library, gridded on high density filters. For each clone spotted on the filters, we have measured using an imaging plate device the hybridization signals of the complex probe obtained from immature or mature DC. Comparative analysis of these values allowed us to identify differentially expressed gene products. Our goal is to identify a new set of genes induced or repressed during DC maturation elicited by rmTNF alpha treatment.

Multiplex messenger assay: simultaneous, quantitative measurement of expression of many genes in the context of T cell activation

The hybridization signature approach, using colony filters and labeled complex probes, can provide high throughput measurement of gene activity. We describe here the implementation of this method to follow the expression levels of 47 genes in resting and activated T cells, as well as in epithelial cells. Using 4-fold spotting of colonies, imaging plate detection and various correction and normalization procedures, the technique is sensitive enough to quantify expression levels for sequences present at 0.005% abundance in the probe. Comparison with Northern blotting shows good consistency between the two methods. Upon activation of a T cell clone by an anti-CD3 antibody variations ranging from 2- to 20-fold are measured, some of which had not been reported previously. This 'multiplex messenger assay' method, performed using available commercial apparatus, can be used in many cases where simultaneous assessment of mRNA levels for many genes is of interest.

From hybridization image to numerical values: a practical, high throughput quantification system for high density filter hybridizations

Hybridization to sets of bacterial colonies or PCR products arrayed on high density filters is used in a number of experimental schemes. In many cases it is desirable to collect quantitative information ('hybridization signatures') rather than indications on 'positive' and 'negative' colonies. We present a practical system, based on an imaging plate analyser and a customized version of commercial software, that makes such quantification feasible, and define its performance in terms of reproducibility and linearity. The system is far superior to methods based on autoradiography and should be useful in many projects that involve the increasingly popular high density filter format.

Differential gene expression in the murine thymus assayed by quantitative hybridization of arrayed cDNA clones

High-throughput measurement of hybridization signatures obtained using complex probes prepared from poly(A)+ RNA and high-density cDNA colony filters is described. The performance of the system, elimination of artifacts, and verification of the validity of the data are discussed. cDNAs corresponding to sequences present at levels of approximately 0.01% in the complex probe can be detected. Good correlation is observed between expression profiles determined by this method and by Northern blotting. The method is applied to a preliminary investigation of differential expression in three cell types present in the murine thymus.