The aim of this workpackage is to use recently developed high-throughput technologies to study gene expression and protein products in the developing zebrafish on a massively parallel scale: expression profiling based on microarrays, and proteomics based on 2D gel electrophoresis and mass spectrometry.

2,000 oligonucleotide microarrays are being used to investigate gene expression in at least 100 mutant, knock-out and morpholino-induceded knock-down conditions (from WP1 and WP4, respectively) from at least 10 tissues or regulatory pathways in order to identify downstream candidate genes for mutagenesis and as drug targets. This work is mostly carried out by MPI EB and Hubrecht Laboratory. FACS sorting of cells from zebrafish transgenic for GFP (WP2) will provide target DNA for cell type specific expression profiling. In combination with different allele strengths and morpholino dosages this will eventually allow qualitative and semi-quantitative modelling of regulatory pathways in the developing zebrafish embryo, a potential not present in any other model organism.

Separation of proteins on 2D gels followed by mass spectrometry (MALDI-TOF) allows to
characterize changes to the proteome in specific conditions, with respect to translational
control and phosphorylation. A proteomics pipeline has been set up in order to complement the
analysis of expression patterns and expression profiling, which give insight into transcriptional regulation. However, as proteomics currently requires significantly more effort than expression profiling, it is reserved for specific regulatory pathways identified as especially promising.

As part of their commitment to the ZF-MODELS Integrated Project, the Geisler lab (MPI EB) and the Spaink lab (LEI) perform zebrafish expression profiling on request. This implements the ZF-MODELS contract which specifies that expression profiling is to be performed on at least 100 conditions from at least 10 organ systems of interest to the consortium.

As this work involves significant scientific input from our side and the contract does not envision it as a service, we perform it on a collaborative basis (i.e. generally requiring a coauthorship).

Please send your request either to Robert Geisler or to Herman Spaink, addressing the following questions:

• What conditions are you proposing for expression profiling?
• Why do you expect significant up- or downregulation of gene expression in these conditions?
• Does your experiment require expression profiling of multiple stages/time points/alleles/drug concentrations, and if so, which ones?
• How do you ensure that your experiment is controlled as tightly as possible (e.g. by using sibs from the same egglay as a control)?
• What downstream analysis do you expect to perform on the candidate genes that we will identify?
• Why do you think this experiment will yield significant scientific insight?

We will contact you to indicate acceptance of your request, to give you a time estimate for the microarray analysis, and to resolve any remaining questions.

Please send us your biological material only after acceptance of your request. Material should be sent in TRIzol buffer, on dry ice. We will perform RNA extraction using our standard protocols and perform microarray analysis either on six slides (Compugen set, 3x Cy3/Cy5 double labelling with dye swap) or on four slides (Affymetrix, 2x condition and 2x control).

As a result you will receive lists of genes indicating UniGene ID, expression (fold change) and significance (t-test P-value), sorted by expression, for each of your conditions. In addition we can provide you the raw data files to process with your own software. The results will be exclusively accessible to you for a period of up to one year after delivery in order to allow publication in a journal, unless you request an earlier release yourself. After that time we will place them on the public ZF-Espresso webserver and submit the raw data to the ArrayExpress repository.

Alternatively, researchers may request Compugen slides from the lab of Herman Spaink for their own hybridizations. Published work using microarray slides from the ZF-MODELS consortium must acknowledge the ZF-MODELS project. Furthermore, normalized data from such an experiment must be submitted no later than a month after publication to the ZF-Espresso database, and raw data to the ArrayExpress repository .

As outlined above, after publication of a microarray experiment by a consortium partner normalized data are to be made available through the ZF-Espresso database of MPI EB, and raw data through the ArrayExpress repository. The first datasets were submitted to ArrayExpress in 2005.