Disease gene candidates revealed by expression profiling of retinal ganglion cell development

Abstract

To what extent do postmitotic neurons regulate gene expression during development or after injury? We took advantage of our ability to highly purify retinal ganglion cells (RGCs) to profile their pattern of gene expression at 13 ages from embryonic day 17 through postnatal day 21. We found that a large proportion of RGC genes are regulated dramatically throughout their postmitotic development, although the genes regulated through development in vivo generally are not regulated similarly by RGCs allowed to age in vitro. Interestingly, we found that genes regulated by developing RGCs are not generally correlated with genes regulated in RGCs stimulated to regenerate their axons. We unexpectedly found three genes associated with glaucoma, optineurin, cochlin, and CYP1B1 (cytochrome P450, family 1, subfamily B, polypeptide 1), previously thought to be primarily expressed in the trabecular meshwork, which are highly expressed by RGCs and regulated through their development. We also identified several other RGC genes that are encoded by loci linked to glaucoma. The expression of glaucoma-linked genes by RGCs suggests that, at least in some cases, RGCs may be directly involved in glaucoma pathogenesis rather than indirectly involved in response to increased intraocular pressure. Consistent with this hypothesis, we found that CYP1B1 overexpression potentiates RGC survival.

Figure 1.

Gene expression of RGCs through development. A, Reproducibility of samples. Two biological replicate samples of E17 RGCs are plotted against each other, with 2-, 3-, 5-, and 10-fold difference bars included. B, Number of genes expressed in RGCs, as defined by present calls in Microarray Suite 5.0 (Affymetrix), plotted as a histogram. C, Of the probes present in at least 10% of RGC samples through development, the number of probe sets that change through RGC development from E17 to P21, plotted as a histogram. D, P21 RGC versus adult whole retinal gene expression, plotting only the genes from U34A GeneChips found present in the RGC samples. Tenfold difference lines are plotted.

Figure 2.

Validation of microarray data. A, Comparison of fold changes (FC) in gene expression during RGC development by GeneChip (average of at least 3 samples) and real-time RT-PCR (average of 5 samples; 1 representative of 3 separate data replications are shown). B, Lysyl oxidase protein is upregulated in rat retinal cryosections in P8 RGCs compared with E20 retinal sections in the ganglion cell layer (GCL) and the adjacent inner plexiform layer in which RGC dendrites are located (top row). A similar upregulation is seen from E20 to P8 RGC cultures (bottom row). Scale bar: top row, 450 μm; bottom row, 50 μm.

Figure 3.

Non-negative matrix factorization. Datasets were weaned to include only genes present in at least 10% of RGC samples and changing at least threefold through development, analyzed by NMF, and internally sorted to amplify boundary edges. The axes demonstrate the ages of the RGC samples analyzed.

Figure 4.

Hierarchical cluster analysis limited to four clusters revealed one cluster of genes that decreased perinatally, two clusters of genes that increased perinatally (only 1 is shown here), and a cluster of genes that spiked at birth and then returned to lower levels after the first postnatal week. Each graph shows the normalized expression levels through development of the highest fold-changing individual genes. Probes from these clusters are listed in supplemental Table 3 (available at www.jneurosci.org as supplemental material).

Figure 5.

Genes expressed by RGCs that are regulated both during development and in regeneration. A, Venn diagram demonstrating distribution of genes analyzed. B, Pie chart demonstrating four classes of genes regulated up or down during development and up or down during regeneration.

Figure 6.

Expression of cochlin, CYP1B1 (CYP), and optineurin (OPTN) mRNA and protein in RGCs in vitro and in vivo. A, Expression levels of the three mRNAs through RGC development from E17 to P21 from GeneChip data. B, Immunoreactivity in the embryonic (E20), postnatal (P8), and early adult (P21) retina. Bright, punctuate staining in the ganglion cell layer (up in each panel) are particularly observed for CYP1B1 at E20 and for optineurin at P21. Cochlin also demonstrates an increase in expression level in the ganglion cell layer from E20 to P21, mimicking the mRNA data. C, Western blot analysis demonstrates cochlin, CYP1B1, and optineurin expression in lysates (15 μg of protein per lane) from purified E20 and P8 RGCs. High CYP1B1 protein level is detected embryonically, whereas increased optineurin and cochlin protein levels are higher postnatally. Blotting against actin (bottom row) demonstrates similar protein loading. D, Immunoreactivity in purified E20 and P8 RGCs after 24 h in culture. Mixed nuclear and cytoplasmic stainings were observed for cochlin (green) and optineurin (red). Perinuclear localization of endogenous CYP1B1 (red) is prominent at both ages. Scale bars, 40 μm.

Figure 7.

Effect of glaucoma-associated genes on RGC survival in vitro. A, B, Embryonic or postnatal RGCs as marked were transfected with plasmid controls (Ctrl) or cDNAs for CYP1B1 (CYP), optineurin (OPTN), and cochlin (COCH; performed in separate experiments and thus separated with double hash marks). RGC survival was measured 3 d later and normalized to control. Statistical analysis using SE of proportions yielded significant Z scores for CYP1B1 only (*). Representative experiments are shown. C, Embryonic RGCs transfected with an siRNA construct directed against CYP1B1 (si-CYP) demonstrated lower survival compared with control cultures.