Carleton JB, Lovell PV, McHugh A, Marzulla T, Horback KL, Mello CV (2014) An Optimized Protocol for High-Throughput In Situ Hybridization of Zebra Finch Brain. Cold Spring Harbor Protoc. doi: 10.1101/pdb.prot084582. [Epub ahead of print]
Animal and tissue preparation
All birds used in the preparation of ZEBrA are adult male zebra finches (Taeniopygia guttata) obtained from our own breeding colony or purchased from local breeders. Since our main goal is to study brain gene expression, and not gene regulation by singing behavior, hearing, or behavioral state, all birds are first isolated overnight (12:12-hour light-dark cycle) in custom-built acoustic isolation chambers to reduce non-specific auditory stimulation. On the following morning (~9:00 AM) birds are monitored briefly (~15 min) to confirm they are not singing, and then sacrificed by decapitation. For brains used in full Atlas Series, birds are monitored for at least 1 hour prior to sacrifice. Brains are quickly dissected out, split in the midline, and hemispheres are placed in plastic molds, covered in OCT compound and frozen in a dry ice/isopropanol bath; both sides are used in ZEBrA. Brains are then sectioned on a cryostat (10 µm), and two consecutive sections mounted per slide, to allow for replication of hybridization in adjacent sections.
Identification of Zebra Finch Orthologs and Gene Model Curation
For all of the genes presented in ZEBrA (see About), we systematically verify gene identity by using a combination of BLAT alignments followed by synteny analysis (verification of flanking genes) in available avian genomes (zebra finch, chicken), well assembled/curated mammalian genomes (human, mouse), and to clarify orthology, other vertebrate genomes (lizard, frog, zebra fish). When searching for genes that are of physiological relevance or are neuroanatomical markers of the mammalian brain, we use the predicted sequences of human Ensembl gene models that corresponded to specific genes and gene families. When necessary (e.g. poor sequence similarity), we use corresponding Ensembl gene model predictions from other species more closely related to zebra finches. For clones identified through array screenings, we start by BLAT aligning the corresponding EST sequences and then search for alignments to annotated loci. Gene verification then proceeds as described above. As a result of these efforts, we have curated a large collection of Ensembl models, and have identified several subsets where the model is either incorrectly annotated, or orthology cannot be unambiguously established; such corrections and revisions are indicated by an asterisk appended to the Ensembl gene model ID. We have also uncovered loci that currently do not have any predicted models; in these cases we provide the chromosome location.
Unless otherwise indicated, all probes are derived from the ESTIMA zebra finch brain cDNA collection. For each gene presented we have selected a single cDNA clone that is partially overlapping and/or contiguous with the predicted Ensembl gene model based on a series of EST reads. In a small number of cases, we have selected clones that are non-overlapping, but found to align within ~500bp of the 3’-end of a gene model. Given the close proximity of these reads to the gene model, evidence of polyadenylation (i.e. poly-A tail), and correct orientation to the +/- DNA strand, we are reasonably confident that these cDNAs correspond to the 3’-end of the gene. These clones are indicated by an asterisk appended to the gene name (e.g. DNM3*). In cases where we were able to connect the clone to the corresponding model in chicken (via cross-BLAT alignment), we remove the asterisk. In rare cases the only available clone is non-overlapping and separated from the 3’-end of the gene by one gap (or in rare instances a few gaps) in the genomic sequence. Since the size of these gaps is not known, we are only able to tentatively assign them a gene name. Such genes are indicated by the presence of two asterisks appended to the gene name (e.g. KCNS3**). ZEBrA also includes some clones that are not associated with a model in the genome (e.g. novel genes or known orthologs not predicted by Ensembl). Where possible we select clones consisting primarily of non-coding or 3’-untranslated (3-UTR) EST sequence, since these regions are unlikely to contain conserved domains present in related members of gene families, but several clones also contain stretches of coding sequence. To confirm clone specificity, we perform BLAT alignments of selected ESTs to the zebra finch genome. The majority of ESTs align unambiguously to a single locus. In a small number of cases an EST may align to secondary loci, but the secondary alignment scores are typically much lower, indicating a high likelihood of probe specificity. Finally, in some cases where no ESTIMA clones are available for a given gene, we have generated template by RT-PCR cloning.
Probe preparation and in Situ Hybridization
We perform non-radioactive in situ hybridization with DIG-labeled antisense riboprobes utilizing an optimized protocol, as detailed in:
1. Carleton et al. (2014) An Optimized Protocol for High-Throughput In Situ Hybridization of Zebra Finch Brain. Cold Spring Harbor Protoc. doi: 10.1101/pdb.prot084582. [Epub ahead of print]
2. Lovell et al. (2008) Birdsong ‘‘Transcriptomics’’: Neurochemical Specializations of the Oscine Song System. PLoS ONE 3(10): e3440
3. Velho et al. (2005) Co-induction of activity-dependent genes in songbirds. Eur J Neurosci 22(7): 1667-78.
4. Mello and Clayton (1994) Song-induced ZENK gene expression in auditory pathways of songbird brain and its relation to the song control system. J Neurosci 14: 6652–6666.
For every probe we first perform an initial test hybridization utilizing our standard protocol and, if necessary, another 1-2 rounds of optimization. These initial hybridizations allow us to determine best hybridization conditions and an initial assessment of brain distribution patterns. Genes with highly differential expression patterns are then hybridized and presented as an Atlas Series (4-9 levels, ~0.2 - 4.0 mm from midline), while genes with non-differential patterns are hybridized and presented as a Subset Series (2-3 levels, ~1.0 and 2.4 mm from midline), focusing on documenting expression in song nuclei. In both cases above we keep track of the hemisphere used, but we do not systematically compare signal across hemispheres. Other genes that exhibit detectable but non-differential expression patterns, but also show high background and/or very low signal are included in ZEBrA as Provisional Series. Finally, genes that show no detectable signal, even after one or a few attempts at lower hybridization stringency are presented without images as Undetectable. Since the various steps above are performed in different brains, we typically examine brain expression in 2-4 different birds for each gene included in ZEBrA.
Completed slides are imaged with a digital scanner (Olympus Nanozoomer) with help from Vadim Pinskiy and Alex Tolpygo in the Mitra lab (at Cold Spring Harbor Laboratory). The Mitra lab then converts each image from its native Nanozoomer NDPI file into compressed JPG. At OHSU, we use Photoshop (Adobe) to adjust the contrast, brightness, and color balance of each image, as well as to correct any artifacts introduced during slide processing (e.g. scratches, chromagen precipitate). Additional color-balancing is also performed across each image series to ensure that background levels are similar across brain sections. Each image is then rotated to approximately match a level from a series of line drawings derived from a histological atlas of the zebra finch brain developed by Drs. Harvey Karten and Partha Mitra. Finally, images are Zoomified, and made available at ZEBrA.