Characterization of transgenic mouse lines expressing Cre recombinase in the retina

Abstract

The mammalian retina consists of five major classes of neuronal cells, as well as glial cells, and it contains more than 50 cell types. The ability to manipulate gene expression in specific cell type(s) in the retina is important for understanding the molecular mechanisms of retinal function and diseases. The Cre/LoxP recombination system has become a powerful tool, allowing gene deletion, over-expression, and ectopic expression in vivo in a cell- and tissue-specific fashion. The key to this tool is the availability of Cre mouse lines with cell- or tissue-type specific expression of Cre recombinase. To date, a large number of Cre-transgenic mouse lines have been generated to target Cre recombinase expression to specific neuronal and glial cell populations in the central nervous system; however, information about the expression patterns of Cre recombinase lines in the retina is largely lacking. In this study, we examined and characterized the expression patterns of Cre recombinase in the retinas of 15 Cre-transgenic mouse lines. Significant Cre-induced recombination or expression of Cre recombinase was observed in the majority of these lines. In particular, we found several Cre lines in which the Cre-induced recombination was found to target exclusively or predominantly a single type or class of retinal cells, including bistratified retinal ganglion cells, starburst amacrine cells, rod bipolar cells, and Müller glial cells. In other lines, the Cre-induced recombination was found in several retinal cell types. These Cre lines provide a valuable resource for retinal research.

Figure 1

Grik4-cre mouse line expresses Cre recombinase in a population of ganglion cells. (A–C) The expression of EYFP viewed in the GCL/IPL in a retinal whole-mount at low (A) and high (B) magnification. (C) EYFP-labeled axons viewed in the GCL/NFL. (D) Vertical section shows the stratification of the dendrites of an EYFP-positive ganglion cell (left panel) in comparison with ChAT-labeled bands (middle panel). The merged image is shown in the right panel in (D). The arrows in (C) and in the left panel in (D) indicate the axon of the ganglion cell. Scale bars: 50 μm in A, 25 μm in B–D. INL: inner nuclear layer; IPL: inner plexiform layer; GCL: ganglion cell layer; NFL: nerve fiber layer.

Figure 2

Expression of Cre recombinase in starburst amacrine cells in the ChAT-cre/Jax mouse line. The retina was photographed in a whole-mount in the GCL (A), the INL (B, the same region as in A), and a vertical section (C). The retina was double-labeled for EYFP (left panels) and ChAT (middle panels). Superimpositions are shown in the right panels. Scale bars: 100 μm in A and B, 25 μm in C. OPL, outer plexiform layer; INL: inner nuclear layer; IPL: inner plexiform layer; GCL: ganglion cell layer.

Figure 3

Expression of Cre recombinase in the ChAT-cre/Gsat mouse line. (A–B) A whole-mount retina was double-labeled for EYFP (left panels) and ChAT (middle panels). Merged images (right panels) revealed no co-localization in either the GCL (A) or the INL (B). (C) In a vertical section, EYFP-positive cells (left panel) did not exhibit ChAT immunofluorescence (middle panel). The merged image demonstrates the absence of co-localization (right panel). Scale bars: 100 μm in A and B, 25 μm in C. OPL: outer plexiform layer; INL: inner nuclear layer; IPL: inner plexiform layer; GCL: ganglion cell layer.

Figure 4

The expression of Cre recombinase in bipolar cells. (A–B) The cell somas of bipolar cells were viewed in retinal whole-mounts of PCP2-cre/GFP (A) and Chx10-cre/GFP (B) mice in the INL (left panels), and their axon terminals, in the IPL (right panels). For the PCP2-cre/GFP line, some weakly labeled cells located in the GCL were encountered (arrow). (C–D) Retinal vertical sections of PCP2-cre/GFP (C) and Chx10-cre/GFP (D) mice were double-labeled for GFP (left panels) and PKC (middle panels), with merged images shown in the right panels. (E) A retinal section from a Chx10-cre/GFP mouse was stained for GFP (left panel) and Cre (middle panel), and co-localization is shown in a merged image (right panel). Scale bars: 100 μm in A and B, 25 μm in C–E. OPL: outer plexiform layer; INL: inner nuclear layer; IPL: inner plexiform layer; GCL: ganglion cell layer; bv: blood vessel.

Figure 5

Mouse lines with predominant expression of Cre recombinase in Müller glial cells. Whole-mount retinas of Foxg1-cre (upper panels), NSE-cre (middle panels) and Eno2-cre (lower panels) lines were analyzed in the same regions in the GCL (A, F and K) and the INL (B, G and L). Retinal vertical sections were immunostained for EYFP (C, H and M) and GS (D, I and N), and superimposed images were created (E, J and O). In addition to Müller cells, some weakly labeled neurons were encountered in the NSE-cre and Eno2-cre lines (arrows). Scale bars: 50 μm for whole-mounts (A and B, F and G, K and L), 25 μm for retinal sections (C–E, H–J, M–O). INL: inner nuclear layer; IPL: inner plexiform layer; GCL: ganglion cell layer.

Figure 6

Mouse lines with expression of Cre recombinase in multiple retinal cell types. (A–D) In the Thy1-cre mouse line, strong EYFP expression was detected in sparse patches of endfeet (A) and somas (B) of Müller cells, as well as some weakly labeled neurons. (C–D) Vertical sections showed the region with patches of EYFP-labeled Müller cells (C) and the region where only EYFP-labeled neurons were observed (D). (E–G) In the Wnt1-cre mouse line, high EYFP expression was detected in Müller cells in the GCL (E) and the INL (F). Blood vessels also appeared to be labeled (E). In the GCL and INL, some weakly labeled somas were encountered. A characteristic view of a vertical section in which Müller cells and weakly labeled bipolar cells and amacrine cells can be seen (G). (H–J) In the Mnx1-cre mouse line, strong EYFP fluorescence was observed in Müller cells in the GCL (H) and in the INL (I) in a retinal whole-mount. The weak EYFP expression was detected in bipolar cells and amacrine cells in vertical sections (J). (K) In the CamK2a-cre mouse line, weak EYFP fluorescence was detected sparsely in different types of retinal neurons but not in Müller cells. Scale bars: 50 μm in A, B, E, F, H and I, 25 μm in C, D, G, J and K. OPL: outer plexiform layer; INL: inner nuclear layer; IPL: inner plexiform layer; GCL: ganglion cell layer.