Lack of CaBP1/Caldendrin or CaBP2 Leads to Altered Ganglion Cell Responses

Targeting of the Cabp2 gene. A, Scheme of the mouse Cabp2 gene with its exons. Arrows above the scheme indicate primers used to clone by PCR Cabp2 genomic fragments. The PGK-DTA and HSV-TK cassettes were included in the targeting vector for negative selection in transfected ES cells. In the targeting vector, the neo cassette (positive selection) replaces exon 1 and exon 2A of the Cabp2 gene. The targeting vector is constructed by using a ∼5-kb DNA fragment as long arm that extends upstream of the initial ATG and covers the CaBP2 promoter. tdTomato was cloned and fused on the initiation codon of CaBP2. The short arm is a ∼2.0-kb genomic fragment encompassing exon 2B to intron 5 of the Cabp2 gene. Arrows (FH1064, FH1139, and Fh1140) below the scheme indicate primers used to select targeted Cabp2 allele. The location of MfeI restriction site as well as the probe (probe SB) used for analysis of the targeted allele using Southern blot are also indicated. B, Southern blot analysis of ES cell clones. MfeI-digested genomic DNA isolated from wild-type B6/Blu ES cells (WT) or targeted clones 1, 37, and 40 was analyzed by Southern blot with an external probe as shown in A and shows a fragment of 13.4 kb for the wild-type Cabp2 allele and a fragment of 9.1 kb for the targeted Cabp2 allele. C, PCR-based analysis of mice for Cabp2 targeting. A 2.3- and 2.75-kb PCR product is amplified with primers FH1140 and FH1139 as shown in A for the WT Cabp2 allele and with primers FH1064 and FH1139 for the targeted allele, respectively.

Analysis of the expression of CaBP1/CD or CaBP2 in WT, Cabp1^–/–, or Cabp2^–/– mice. A, PCR analysis of CaBP1 and caldendrin transcripts. PCR amplification of both S-CaBP1/L-CaBP1 variants or caldendrin using RNA from Cabp1^+/+ or Cabp1^–/– mouse retinas. PCR amplification in the absence of DNA (–) was used as a negative control, and amplification of GAPDH was the positive control. B, Analysis of CaBP1/CD proteins using immunohistochemistry. Immunolocalization of CaBP1/CD in the retinas of 2-month-old Cabp1^+/+ and Cabp1^–/– mice using rat anti-CaBP1/CD antibodies. Nuclei are labeled with Hoechst. Lack of CaBP1/CD immunoreactivity confirms targeting of the Cabp1 gene. Scale bar: 20 μm. C, PCR analysis of CaBP2 transcripts. PCR amplification of both S-CaBP2/L-CaBP2 variants from Cabp2^+/+ or Cabp2^–/– mouse retina. Controls as described in A. D, Analysis of CaBP2 proteins using immunohistochemistry. Immunolocalization of CaBP2 in the retina of 2-month-old Cabp2^+/+, Cabp2^–/–, Cabp1^–/–, and Cabp1^–/–/Cabp2^–/– mice using rat anti-CaBP2 antibodies. Nuclei are labeled with Hoechst. Cross-reactivity of anti-CaBP2 antibody is revealed by labeling of Cabp2^–/– mouse retina with this antibody and results in staining signals in the inner retina of Cabp2^–/– retina. The absence of staining in Cabp1^–/–/Cabp2^–/– double-KO mice demonstrates that the cross-reactivity of anti-CaBP2 antibodies is against the CaBP1/CD proteins. Specific labeling of CaBP2 in the inner retina is revealed by labeling of Cabp1^–/– mouse retina with anti-CaBP2 antibodies. Scale bar: 20 μm.

Analysis of the CaBP1/CD-expressing cells in Cabp1^+/+ and Cabp1^–/– mice. A, CaBP1/CD localizes in cells expressing NK3R in the inner retina. Analysis of Cabp1^+/+ mouse retina labeled with anti-NK3R (green) and rat anti-CaBP1/CD (red) using confocal microscopy. All cells labeled with anti-NK3R are also stained with anti-CaBP1/CD. Arrows in the right panels point to some of the yellow colocalization signals. INL, inner nuclear layer; GCL, ganglion cell layer. B, CaBP1/CD localizes in many cells expressing calretinin in the inner retina of Cabp1^+/+ mouse retina. Legend as in A with anti-calretinin (green). C, CaBP1/CD colocalizes with synaptotagmin 2 in the outer IPL. Legend as in A with anti-Syt2 (green). D–F, Morphology of CaBP1/CD-deficient cells (NK3R stained type 1 and 2 bipolar cells, calretinin-stained amacrine cells, and Syt2-stained type 2 bipolar cells) is normal in Cabp1^–/– mouse retina. Scale bars: 10 μm.

Ribbon synapse morphology in the OPL and IPL of Cabp1^+/+ and Cabp1^–/– mice. A, Confocal images showing en face views of the axon terminals of Syt2-labeled type 2 cone OFF bipolar cells (selected in IPL outer sublamina) or NK3R-labeled type 1 and 2 cone OFF bipolar cells in Cabp1^+/+ and Cabp1^–/– retina whole mounts. Synaptic ribbons were visualized with anti-Ctbp2 (green). The axons and ribbons appear normal in both Cabp1^+/+ and Cabp1^–/– retinas. Scale bar: 2 μm. B, Representative electron micrographs of cone OFF bipolar cells in mouse retina cross-sections through the IPL within 5 μm from the inner nuclear layer. Normal ribbons and tethered vesicles are observed both in Cabp1^+/+ and Cabp1^–/– retinas. Scale bar: 200 nm. C, En face views of the dendrites of type 2 cone OFF and type 6 cone ON bipolar cells in the OPL of mouse retina whole mounts labeled with anti-Syt2 (red) and anti-Ctbp2 (green). Scale bar: 2 μm. D, Representative electron micrographs of mouse retina cross-sections through the OPL. Scale bar: 200 nm.

Immunolocalization of CaBP2 in mouse retina. Confocal images of Cabp1^–/– (A–F) or Gus-GFP (G–I) mouse retinas double labeled for CaBP2 and CaBP5 (A–C), calsenilin (D–F), and GFP (G–I). Sublaminae 1–5 are indicated in C. Scale bar: 10 μm.

Analysis of the CaBP2-expressing cells in Cabp2^+/+ and Cabp2^–/– mice. A, CaBP2 localizes in some cells expressing NK3R in the inner retina. Analysis of Cabp2^+/+ mouse retina labeled with anti-NK3R (green) and rat anti-CaBP2 (red) using confocal microscopy. Some cells labeled with anti-NK3R are also stained with anti-CaBP2. Arrows in the right panels point to some of the yellow colocalization signals. INL, inner nuclear layer; GCL, ganglion cell layer. B, CaBP2 localizes in some cells expressing synaptotagmin 2 in the inner retina of Cabp2^+/+ mice. Legend as in A with anti-Syt2 (green). The area shown at higher magnification in the bottom panels is indicated by the white box. C, CaBP2-labeled cells terminate their axons in sublamina 1 and sublamina 3/4. Legend as in A with anti-calretinin (green). Sublaminae 1–5 are indicated in the right panel. D–F, The overall morphology of the retina and CaBP2-deficient cells (NK3R- and synaptotagmin-labeled bipolar cells) is normal in Cabp2^–/– mouse retinas.

Ribbon synapse morphology in the OPL and IPL of Cabp2^+/+ and Cabp2^–/– mice. A, Confocal images of en face views of the axon terminals of Syt2-labeled type 6 cone ON bipolar cells (selected in IPL inner lamina) or NK3R-labeled cone OFF bipolar cells in Cabp2^+/+ and Cabp2^–/– retina whole mounts. Synaptic ribbons were visualized with anti-Ctbp2 (green). The axons and ribbons of Cabp2^+/+ and Cabp2^–/– retinas appear similar. Scale bar: 2 μm. B, Representative electron micrographs of cone OFF bipolar cells in mouse retina cross-sections through the IPL within 2 μm from the inner nuclear layer. Scale bar: 200 nm. C, En face views of the dendrites of type 2 cone OFF and type 6 cone ON bipolar cells in the OPL of mouse retina whole mounts labeled with anti-Syt2 (red) or anti-Ctbp2 (green). Scale bar: 2 μm. D, Representative electron micrographs of cone terminals in mouse retina cross-sections through the OPL. Scale bar: 200 nm.

ON alpha ganglion cell responses in Cabp2^+/+ and Cabp2^–/– mice. A, Maximum-projection image of an exemplar ON alpha ganglion cell filled with dye postrecording. B, Exemplar spike raster from an ON alpha ganglion cell in response to 100% contrast step. C, Average excitatory synaptic currents in response to 10-ms flash of 100% contrast. D, Bar graph comparing the peak amplitude of synaptic response shown in C across all cells between control and Cabp2^–/–. E, Linear-nonlinear (LN) model: a time-varying random stimulus and the resulting ganglion cell excitatory synaptic inputs were used to derive the linear filter and static nonlinearity that relate the stimulus to the response (top). F, Average normalized linear filters for the stimuli. G, Quantification of the time to peak, i.e., latency of the peaks in linear filters. H, Nonlinearities of ganglion cells for the noise stimuli. I, Quantification of the response range of the measured response across all cells.

OFF alpha ganglion cell responses in Cabp1^+/+ and Cabp1^–/– mice. A, B (i), Maximum projection image of an exemplar OFF sustained and OFF transient alpha ganglion cell filled with dye postrecording. A, B (ii), Exemplar spike raster in response to 100% contrast step. A, B (iii), Average excitatory synaptic currents in response to a 10-ms flash of 100% contrast from control and Cabp1^–/– retinas. A, B (iv), Quantification of the peak amplitude of synaptic response shown in iii across all cells between control and Cabp1^–/– retinas. A, B (v), Linear-nonlinear (LN) model: average normalized linear filters for the stimuli. A, B (vi), Quantification of the time to peak, i.e., latency of the peaks in linear filters. A, B (vii), Average nonlinearities of ganglion cells for the noise stimuli. A, B (viii), Quantification of the response range of the measured response across all cells between control and Cabp1^–/– retinas.

Spike responses of ON and OFF alpha ganglion cells in control, Cabp1^–/–, and Cabp2^–/– mice. A, Average peristimulus histogram of a spike response to a 100% contrast step in ON alpha ganglion cells in Cabp2^+/+ and Cabp2^–/– retinas. B, Quantification of the peak spike rate shown in A. C, Average peristimulus histogram of a spike response to a 100% contrast step in OFF transient ganglion cells in Cabp1^+/+ and Cabp1^–/– retinas. D, Quantification of the peak spike rate shown in C.