Conceived and designed the experiments: SC DG IW CKB. Performed the experiments: SC DG IW. Analyzed the data: SC DG IW CKB. Contributed reagents/materials/analysis tools: SC IW. Wrote the paper: SC DG IW CKB.
The authors have declared that no competing interests exist.
In response to light, most retinal neurons exhibit gradual changes in membrane potential. Therefore K+ channels that mediate threshold currents are well-suited for the fine-tuning of signal transduction. In the present study we demonstrate the expression of the different Kv11 (ether-à-go-go related gene; erg) channel subunits in the human and mouse retina by RT PCR and quantitative PCR, respectively. Immunofluorescence analysis with cryosections of mouse retinae revealed the following local distribution of the three Kv11 subunits: Kv11.1 (m-erg1) displayed the most abundant expression with the strongest immunoreactivity in rod bipolar cells. In addition, immunoreactivity was found in the inner part of the outer plexiform layer (OPL), in the inner plexiform layer (IPL) and in the inner segments of photoreceptors. Immunoreactivity for Kv11.2 (m-erg2) was observed in the outer part of the OPL and throughout the IPL. Double-labeling for vGluT1 or synaptophysin indicated a mainly presynaptic localization of Kv11.2. While no significant staining for Kv11.3 (m-erg3) was detected in the neuronal retina, strong Kv11.3 immunoreactivity was present in the apical membrane of the retinal pigment epithelium. The different expression levels were confirmed by real-time PCR showing almost equal levels of Kv11.1 and Kv11.2, while Kv11.3 mRNA expression was significantly lower. The two main splice variants of Kv11.1, isoforms a and b were detected in comparable levels suggesting a possible formation of cGMP/cGK-sensitive Kv11.1 channels in photoreceptors and rod bipolar cells. Taken together, the immunohistological results revealed different expression patterns of the three Kv11 channels in the mouse retina supposing distinct physiological roles.
The vertebrate retina is a neuronal network which consists of six major cell types. The incoming light is detected by photoreceptors and the information is subsequently passed through bipolar cells to ganglion cells which form the optic nerve. The processing of the incoming signals already occurs in the retina via feedback mechanisms and lateral connections between retinal neurons
Kv11 (ether à-go-go related gene; erg) K+ channels belong to the EAG family of voltage-gated K+ channels. Most members of this family (Kv10 or eag, Kv11 or erg, and Kv12 or elk channels) share an activation threshold at relatively negative potentials
In the present study we unveiled the expression patterns of all three Kv11 subunits in the mouse retina. Our results point to different physiological roles for the different subunits.
All animal experiments were approved by the local authorities of the City of Hannover (Niedersächsisches Landesamt für Verbraucherschutz und Lebensmittelsicherheit) and were conducted in accordance with the German law on the protection of experimental animals and the European Communities Council Directive of November 24 1986 (86/609/EEC).
For the use of human material, tenets of the Declaration of Helsinki were followed, informed consent was obtained, and Institutional Human Experimentation Committee approval was granted for the studies.
We investigated the mRNA expression of all three Kv11 channel subunits in mouse and human retinas by RT PCR. To yield mRNA from mouse retinas, C57Bl/6J mice were sacrificed by cervical dislocation and the retinas were isolated immediately from the enucleated eye cups and put into lysis buffer from RNeasy Mini Kit for RNA extraction (Qiagen). Human retinal mRNA was taken from human donor eyes within 24 hours after death. The RNA was reverse transcribed and used for RT PCR reactions as described elsewhere
gene | organism (Acc. No.) | Sequence | first base |
Kv11.1 (erg1) | human (NM_000238) | fd |
1616 |
rev |
2168 | ||
Kv11.2 (erg2) | human (NM_173092) | fd |
2309 |
rev |
2795 | ||
Kv11.3 (erg3) | human (NM_033272) | fd |
2243 |
rev |
2799 | ||
Kv11.1 (m-erg1a) | mouse (NM_013569) | fd |
746 |
rev |
981 | ||
Kv11.1 (m-erg1b) | mouse (AF012869) | fd |
258 |
rev |
467 | ||
Kv11.2 (m-erg2) | mouse (NM_001037712) | fd |
2225 |
rev |
2418 | ||
Kv11.3 (m-erg3) | mouse (NM_133207) | fd |
3092 |
rev |
3255 | ||
Ribosomal protein L32 | mouse (NM_172086) | fd |
300 |
rev |
407 |
For comparative quantification of the expression of the three Kv11 channel subunits we extracted RNAs from retinas of C57Bl/6J mice (2 male, 2 female, 1 month old) using the RNeasy Mini Kit (Qiagen). The mRNAs were reverse transcribed and the real-time experiments performed as described earlier
To access the retinal distribution patterns of the Kv11 channel subunits, we performed immunohistological experiments with eyes from C57Bl/6J mice. For these studies 1-month old animals (n = 10) of both sexes were used. Prior to enucleation, the mice were anesthetized by intraperitoneal injection of 16% sodium pentobarbital solution (Narcoren®, Merial, Hallbergmoos, Germany, 5 µl/g body weight) and transcardially perfused with 4% formaldehyde in 0.1 M sodium cacodylate buffer, pH 7.3. After enucleation the eyes were post-fixed in 4% paraformaldehyde for 30 minutes and cryoprotected in graded sucrose solutions (10%, 20% and 30%) in 0.1 M cacodylate buffer, pH 7.3. The 14 µm cryosections were permeabilized in blocking buffer (PBS with 0.2% Triton X-100, 0.02% sodium azide, 5% donkey serum). The primary antibodies listed in
antigen | antiserum | source | Dilution |
erg1 | rabbit anti-erg1 | Chemicon AB5930 | 1∶1000 |
erg2 | rabbit anti-erg2 | this report | 1∶1000 |
erg3 | rabbit anti-erg3 | this report | 1∶500 |
PKCα | mouse anti-PKCα | Santa Cruz Biotechnology sc-8393 | 1∶250 |
calbindin | mouse anti-calbindin | Abcam ab9481 | 1∶1000 |
vGluT1 | guinea pig anti-vGluT1 | Synaptic Systems 135304 | 1∶5000 |
ChAT | goat anti-choline acetyltransferase | Chemicon (Millipore) AB144P | 1∶100 |
synaptophysin | goat anti-synaptophysin | Santa Cruz Biotechnology sc-7568 | 1∶50 |
GAD2/GAD65 | guinea pig anti-GAD2/GAD65 | Synaptic Systems 198104 | 1∶1000 |
ezrin | mouse anti-ezrin | Abcam ab4069 | 1∶250 |
The immunoreactivities were visualized by appropriate secondary antibodies: donkey anti-mouse conjugated to DyLight 488, donkey anti-rabbit DyLight 649,donkey anti-goat DyLight 549 or donkey anti-guinea pig DyLight 488 (Jackson ImmunoResearch). The cell nuclei were counter stained by 10 minutes incubation with bis-benzimide solution (Hoechst 33258, Sigma). Confocal imaging was performed with a TCS SP2 AOBS scan head and an inverted Leica DM IRB.
For the generation of anti-Kv11.2 and anti-Kv11.3 antibodies, fusion proteins of glutathione-S-transferase and rat Kv11.2 (amino acids 722–950; Acc No. NP_446389) or rat Kv11.3 (amino acids 976–1090; Acc No. NP_571987) were prepared by subcloning the corresponding cDNA fragments into the pGEX vector (Pharmacia) and expression in E. coli. The fusion proteins were purified with glutathione sepharose 4B beads (Pharmacia) and rabbits were immunized with the fusion proteins.
Prior to the immunohistological experiments, suitability of the used Kv11 antibodies was tested by immunocytochemistry on HEK293 cells transiently transfected with rat Kv11.1, Kv11.2 or Kv11.3 cDNA. We used rat cDNA because the antibodies were also raised against parts of the rat Kv11 subunits. To allow a direct comparison of subcellular Kv11 subunit location with antibody staining we transfected Kv11 channel subunits with a C-terminal EGFP tag subcloned from the pEGFP-N vector (Clontech). Goat anti-rabbit antibodies conjugated to Alexa Fluor 546 (Molecular Probes; 1∶4000) were used to visualize Kv11 channel immunoreactivity. The three Kv11 antibodies were found to label specifically transfected cells and the Kv11 immunoreactivity exhibited identical subcellular distribution compared to the EGFP signal of the tagged Kv11 channels (
Previous studies demonstrated the expression of the three Kv11 channel subunits only in rat retinas
RT PCR experiments showing the expression of transcripts of all three Kv11 channel subunits in the mouse (A) and human (B) retina; * indicates the 600 (A) or 500 bp (B) fragments of the molecular weight marker (100 bp ladder). In A the expression of the most abundant splice-variants of Kv11.1, isoforms a and b, in the mouse retina is also shown. (C) Real-time PCR experiments for the three Kv11 channel genes revealed comparable levels of expression for Kv11.1 and Kv11.2, while Kv11.3 is expressed at a significantly lower level in the mouse retina. (D) Using primers specific for the two most abundant splice-variants of Kv11.1 we detected the two splice-variants in a ratio of 1.8∶1. The expression levels of erg genes were normalized to the expression level of the ribosomal protein L32. Each experiment was carried out with four retinal mRNA preparations.
The distribution of Kv11 channel subunits in the mouse retina was investigated using the specific Kv11.1, Kv11.2 and Kv11.3 antibodies in immunofluorescence experiments on eye cup slices. These experiments demonstrated strong expression of both, Kv11.1 and Kv11.2 subunits, in the neuronal retina while there was only a very small fraction of Kv11.3-positive cells in the neuronal retina (
Confocal fluorescence micrographs of mouse retinas labeled with antibodies against Kv11.1, Kv11.2 and Kv11.3 and counter stained for cell nuclei. (A) The Kv11.1 immunoreactivity was observed in IS, the OPL, the INL and the IPL. (B) Immunoreactivity for Kv11.2 was detected in the OPL and in the IPL. (C) Immunostaining of retina with the Kv11.3-specific antibody revealed no significant signals in the neuronal retina, but in the retinal pigment epithelium. GCL: ganglion cell layer; IPL: inner plexiform layer; INL: inner nuclear layer; OPL: outer plexiform layer; ONL: outer nuclear layer; IS: photoreceptor inner segments; OS: photoreceptor outer segments; RPE: retinal pigment epithelium.
Immunoreactivity for Kv11.2 was present in the outer part of the OPL and throughout the IPL. While Kv11.1 immunoreactivity was found in both, neuronal somata and cell processes, the expression of Kv11.2 subunits was restricted to processes of different retinal neurons. Thus, Kv11.1 and Kv11.2 subunits displayed a different expression pattern within the IPL and the OPL. Nevertheless, a partial overlap in the IPL could point to the presence of heteromeric Kv11 channels, since different Kv11 subunits can assemble to form heteromeric channels
Kv11.3 immunoreactivity was not found at a significant level in the neuronal retina, but we detected strong expression of Kv11.3 subunits in the retinal pigment epithelium.
Double-labeling experiments were performed in the attempt to further circumscribe the Kv11 immunoreactivity to certain neurons or cell layers. The only study on the localization of Kv11 channel proteins in the retina so far reported an expression of Kv11.1 subunits in somata and primary dendrites of horizontal cells
(A) PKCα-specific antibodies were used for double-labeling experiments with Kv11.1. An almost perfect co-localization of both proteins was detected in rod bipolar cells. The box highlights the section shown in the bottom row in higher magnification illustrating the co-localization of Kv11.1 and PKCα. (B) Calbindin-specific antibodies were used for double-labeling experiments with Kv11.1. No significant co-localization of proteins was found in the mouse retina (the small boxes highlight cell bodies of horizontal cells immunolabeled by calbindin). The large box highlights the section shown in the bottom row in higher magnification illustrating the distinct expression of Kv11.1 and calbindin in the mouse retina. (C) Double-labeling experiments with Kv11.1 and vGluT1-specific antibodies. Some co-localization was detected in the IPL, especially in the axonal terminals of bipolar cells. In the OPL no co-localization could be detected. The box highlights the section shown in the bottom row in higher magnification illustrating the distinct expression of Kv11.1 and vGluT1 in the OPL of the mouse retina. (D) Double-labeling experiments with Kv11.1 and ChAT-specific antibodies. The merged data show no co-localization of both proteins. The box highlights the section shown in the bottom row in higher magnification illustrating the distinct expression of Kv11.1 and ChAT. On the right of each panel a part of the bright field picture is shown. RPE: retinal pigment epithelium; OS: outer segments; IS: inner segments; ONL: outer nuclear layer; OPL: outer plexiform layer; INL: inner nuclear layer; IPL: inner plexiform layer; GCL: ganglion cell layer. Scale bars: 20 µm. The results of counter staining of cell nuclei (blue) is included in the merged pictures shown on the right.
In contrast to this postsynaptic distribution of Kv11.1 subunits in the OPL, Kv11.2 seemed to be expressed presynaptically in this layer (see
(A) The presynaptic marker of glutamatergic synapses vGluT1 was used for double-labeling experiments with Kv11.2. The merged data show a high degree of co-localization of both proteins. The box highlights the section shown in the bottom row in higher magnification illustrating the co-expression of Kv11.2 and vGluT1 in the IPL and OPL. (B) Double-labeling for the unspecific presynaptic marker synaptophysin and for Kv11.2. The merged data show a high degree of overlap in the OPL. In the IPL, Kv11.2 immunoreactivity is present in a subset of synaptophysin-positive presynaptic structures. The box highlights the section shown in the bottom row in higher magnification. (C) Double-labeling for the GABAergic marker GAD2/GAD65 and for Kv11.2. The merged data show no co-localization of both proteins. The box highlights the section shown in the bottom row in higher magnification illustrating the distinct expression of Kv11.2 and GAD2/GAD65. (D) A PKCα-specific antibody was used for double-labeling experiments with the Kv11.2-specific antibody. Some co-localization was detected in the IPL, especially in the axonal terminals of bipolar cells. In the OPL no co-localization could be detected. The box highlights the section shown in the bottom row in higher magnification illustrating the distinct expression pattern of Kv11.2 and PKCα in the OPL. On the right of each panel a part of the bright field picture is shown. RPE: retinal pigment epithelium; OS: outer segments; IS: inner segments; ONL: outer nuclear layer; OPL: outer plexiform layer; INL: inner nuclear layer; IPL: inner plexiform layer; GCL: ganglion cell layer. Scale bars: 20 µm. The results of counter staining of cell nuclei (blue) is included in the merged pictures shown on the right.
We were not able to perform direct double-labeling experiments with Kv11.1 and Kv11.2 specific antibodies as both antibodies were raised in rabbits. But we had a strong overlap of Kv11.1 and PKCα on the one hand (
Although immunostaining with Kv11.3-specific antibody revealed no immunoreactivity in the neuronal retina, there was a strong signal in the apical membrane of the retinal pigment epithelium (RPE). Double-labeling experiments with an antibody against ezrin, a marker of the apical membrane of the RPE confirmed the expression of Kv11.3 subunits in the apical membrane of the RPE (
Ezrin, a marker of the apical membrane of the RPE was used in double-labeling experiments with Kv11.3. The merged data show an almost perfect co-localization of both proteins. Arrows point to the additional expression of Kv11.3 subunits in retinal and choroidal blood vessels. The box highlights the section shown in the bottom row in higher magnification illustrating the coexpression of K11.3 and ezrin in the apical membrane of the RPE. A part of the corresponding bright field picture is shown on the right. RPE: retinal pigment epithelium; OS: outer segments; IS: inner segments; ONL: outer nuclear layer; OPL: outer plexiform layer; INL: inner nuclear layer; IPL: inner plexiform layer; GCL: ganglion cell layer. Scale bars: 20 µm. The results of counter staining of cell nuclei (blue) is included in the merged pictures shown on the right.
The present PCR data show for the first time the mRNA expression of all three Kv11 channel subunits in the human retina and we confirmed and further analyzed their expression in the mouse eye. To unveil the distinct cellular expression patterns of the three Kv11 channel subunits we performed immunohistological studies. While Kv11.1 subunits displayed a relatively wide-spread distribution in different retinal layers, Kv11.2 subunit expression was detected predominantly presynaptically in the glutamatergic photoreceptors and bipolar cells, and Kv11.3 immunoreactivity was detected in the apical membrane of the retina supporting cell layer of the RPE and in choroidal and retinal blood vessels. The RPE plays a major role in spatial buffering of ions in the subretinal space
Interestingly, Kv11.1 displays a distribution in the mouse retina comparable to that of Kv10.1 (eag1)
Other modulations of retinal Kv11 channels which might be involved in adaption could include metabotropic GluR1-mediated Kv11 current inhibition as recently shown in mitral cells
Kv11.2 channels are expressed presynaptically in glutamatergic synapses of the retina, suggesting their contribution to the synaptic release of glutamate in photoreceptors and bipolar cells. The axons of OFF bipolar cells terminate in the outer half of the IPL, whereas those of ON and rod bipolar cells terminate in the inner half of the IPL
Taken together, our immunohistological data revealed very different expression patterns for the three Kv11 channel subunits in the retina. These differences favor distinct functional roles for the channels in the retina. Kv11.1 channels, possibly heteromers formed by 1a and 1b subunits, may be involved in the setting of the excitability of rod bipolar cells and photoreceptors. In contrast, Kv11.2 channels with their presynaptic expression could be involved in the fine tuning of the presynaptic membrane potential thereby adjusting the neurotransmitter release in these cells. In sharp contrast to the expression of Kv11.1 and Kv11.2 in the neuronal retina, Kv11.3 immunoreactivity was found to be expressed in the RPE where the channels could contribute to the spatial buffering of K+ ions.
The Kv11 antibodies used for the immunostaining of retina slices were tested for their suitability using heterologously expressed Kv11 channels. HEK cells were transiently transfected with rat cDNA for Kv11.1, Kv11.2 and Kv11.3 tagged with EGFP. Bright field pictures are shown in the left panels. Kv11 channel expression directly indicated by the EGFP fluorescence (middle panels) coincided with Kv11 channel immunoreactivity visualized by the red fluorescent secondary antibody (right panels). Although both, the Kv11.1 antibody and the Kv11.2 antibody showed some cross-reactivity with Kv11.2 and with Kv11.1 channels, respectively, this cross-reactivity was obviously too weak for immunohistology because both channels showed completely different expression patterns (e.g. in the OPL. Compare
(TIF)
We would like to thank A. Hasse, T. Kock and N. Stamm for their valuable help in testing the suitability of the used Kv11 antibodies and performing PCR experiments (T.K.).