| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH |
|
|
|
|||||||
|
|||||||||
Electronic Letters to:
|
|
Electronic letters published:
![[Read eLetter]](/icons/misc/sectionDOWN.gif){kind=icon}
Usher 1 Protein Complexes in the Retina
Myosin VIIa in Photoreceptor Cell Synapses May Contribute to an Usher 1 Protein Complex in the Retin
|
|
|||
|
David S. Williams , Daniel Gibbs
Send letter to journal:
dswilliams{at}ucsd.edu David S. Williams, et al.
|
Reiners et al.1 immunolocalized the Usher 1 proteins, harmonin, cadherin23, and myosin VII in mouse retinas. They reported colocalization of all three proteins in the photoreceptor synapse, and, consistent with in vitro binding studies,2 concluded that the proteins form a photoreceptor synaptic complex. We commend the authors for their thoroughness in studying the localization of harmonin. However, their localization of myosin VIIa and cadherin23 is less rigorous, and their colocalization of at least the myosin VIIa in the photoreceptor synapse appears to be an artifact. The myosin VIIa antibody used by Reiners et al.1 was from "pAb 2.2," which was generated in our laboratory. Although some lots of pAb2.2 antiserum appear specific for myosin VIIa on western blots, all require purification for labeling sections3 a point made clearer now with the availability of a null mutant as a negative control. In our hands, antiserum labels thick cryosections of normal mouse retina in a manner similar to that shown by Reiners et al.1 (their Fig. 8A). Label is evident in the RPE, photoreceptor inner segments and the OPL (Fig. 1A). The same labeling of the inner segment and OPL is also present in retinas from Myo7a4626SB/4626SB shaker-1 mice (Fig. 1B). These mice possess a nonsense Gln720Stop mutation4 and appear to be null for even part of Myo7a.5 Therefore, labeling of their retinas by pAb2.2 (which was generated against a fusion protein including amino acids 941-1071 of Myo7a3) is unlikely to represent labeling of myosin VIIa. Labeling by affinity-purified pAb2.2 is detected only in the apical RPE by standard fluorescence microscopy of thick sections6 (Fig. 1C). Higher resolution images also enable detection of label in the connecting cilium with affinity-purified pAb2.2, but labeling of the photoreceptor inner segment or synapse is not evident at any level.3 Affinity-purified pAb2.2 does not label Myo7a4626SB/4626SB retinas at all (Fig. 1D).
Phenotypic analysis of shaker-1 retinas supports a distribution of myosin VIIa that is limited to the apical RPE and photoreceptor cilium. Mutant phenotypes have been identified in both these regions,5,7,8 but the photoreceptor inner segment and synapse appear normal. Reiners et al.1 argued that the defective electroretinogram (ERG) response of shaker-1 retinas supports a dysfunction of the photoreceptor synapse. However, both the a- and b-waves of the ERG have similarly-reduced amplitudes,9 indicating that, while the photoreceptor response is smaller, transmission of that response to other cells is unaffected. This ERG defect contrasts with that of mice with known synaptic defects; such mice have a reduced ERG b:a-wave ratio.10 Hence, evidence indicates that retinal myosin VIIa is indeed restricted to the apical RPE and photoreceptor cilium, as described earlier by others.3,6 Reiners et al.1 showed that harmonin is found in neither of these two locations, so their conclusion that myosin VIIa and harmonin may interact in the retina seems unlikely. Daniel Gibbs and David S. Williams Departments of Pharmacology and Neurosciences, UCSD School of Medicine, CA References 1. Reiners J, Reidel B, El-Amraoui A, et al. Differential distribution of harmonin isoforms and their possible role in Usher-1 protein complexes in mammalian photoreceptor cells. Invest Ophthalmol Vis Sci. 2003;44:5006-5015. |
|||
|
|
|||
|
Uwe Wolfrum , Jan Reiners
Send letter to journal:
wolfrum{at}uni-mainz.de Uwe Wolfrum, et al.
|
In the letter "Usher 1 Protein Complexes in the Retina" by Gibbs and Williams, the authors doubt the localization of myosin VIIa in the synaptic terminal of photoreceptor cells, which we have recently demonstrated.1 The authors justify their doubts by the demonstration of the nonspecificity of the myosin VIIa antiserum used in our study. Indeed, we applied in our study the pAb 2.2 myosin VIIa antibody, which was generated during collaborative efforts between our laboratory and the Williams lab.2,3 The doubts expressed in the letter prompted us to prove our results in Myo7a+/4626SB and the myosin VIIa null mutant Myo7a4626SB/4626SB shaker-1 mice using different antibodies against myosin VIIa including pAb 2.2. In our studies, the pAb 2.2 serum also decorated the photoreceptor synapses in the outer plexiform layer of Myo7a4626SB/4626SB shaker-1 retinas which confirm the results presented by Gibbs and Williams (data not shown). However, in our hands indirect immunofluorescence studies performed with either the affinity purified antibody pAb 2.1 or pAb 2.2 revealed staining in the synaptic region of photoreceptor cells of wild type animals and Myo7a+/4626SB (Figs. 1A, 1C). In contrast, the immunofluorescence in the outer plexiform layer in the cryosections through Myo7a4626SB/4626SB shaker-1 retinas was drastically reduced (Figs. 1B, 1D). Furthermore, immunoelectron microscopy with the affinity purified antibody pAb 2.1 provides further evidence that myosin VIIa is present at the synapse of photoreceptor cells (Figs. 1F, 1G). To occlude antigen masking or destruction during the preparation for immunocytochemical staining, we determined the synaptic localization of myosin VIIa in Western blot analysis of crude synaptosome preparations of rat retinas.4 This biochemical data confirmed our immunocytochemical results regarding the synaptic localization of myosin VIIa (Fig. 1E). Analysis of synaptosome fractions from brain tissue further specified myosin VIIa as a common synaptic protein (data not shown). Since the pAb 2.1/2.2 showed also minor epifluorescence in cryosections through retinas of the myosin VIIa null mutant, we decided to apply a different myosin VIIa antibody to them. This polyclonal antibody was generated against the myosin VIIa-tail and has been characterized previously.5 Indirect immunofluorescence of this anti-myosin VIIa antibody confirmed the localization of myosin VIIa in photoreceptor synapses of wild type and Myo7a+/4626SB mice, in the absence of any synaptic staining in the retina of Myo7a4626SB/4626SB shaker-1 mice (Fig. 2). The differences between the results presented by Gibbs and Williams and our findings are probably due to diverse preparation procedures. Although we do not know exactly how Gibbs and Williams and Hasson et al.6 have processed mouse retinas, the specimens were most probably chemically fixed before cryosectioning and immunocytochemical labeling. It has been previously reported that epitope masking occurs in chemically fixed retinal cryosections.7 To avoid such a phenomenon, we routinely use, if possible, unfixed mouse retinas in immunocytochemical analyses, of which as far as we know no epitope masking is possible. Previous investigations revealed that not only are other Usher 1 proteins, e.g., harmonin, localized at the photoreceptor synapse, but also other proteins identified as myosin VIIa interacting proteins, in particular vezatin, protein kinase A and MyRIP (myosin VIIa- and Rab-interacting protein), were shown as synaptic proteins (U. Wolfrum, unpublished data and other8,9,10). The functions of these proteins are commonly associated with the interaction of myosin VIIa at specific cellular domains. Furthermore, in the model system of Drosophila, myosin VIIa serves as a molecular motor at the neuro-muscular synaptic junctions (Marella SS, Steinert J, Cheney C, Schuster CM, manuscript in preparation, 2004) indicating that myosin VIIa is a common protein at synapses across species. In conclusion, several lines of evidence indicate that the localization of myosin VIIa at the ribbon synapses of mammalian photoreceptor cells is not artifactual. Hence, myosin VIIa and other molecules including Usher proteins may assemble into a supramolecular protein complex integrated by the scaffold protein harmonin (USH1C).1 Jan Reiners and Uwe Wolfrum Institute of Zoology, Johannes Gutenberg University, Mainz, Germany
Figure 1. Retinal localization of myosin VIIa by affinity purified antibodies pAb2.1 and pAb2.2. (A-D) Indirect immunofluorescence with affinity purified antibodies pAb2.1 (A, B) and pAb2.2 (C, D) in retinal cryosections from Myo7a+/4626SB (A, C) and Myo7a4626SB/4626SB (B, D). Both myosin VIIa antibodies stain synapses in the outer plexiform layer of the Myo7a+/4626SB retina. This staining is drastically reduced in Myo7a4626SB/4626SB mice. Methods: Animal use conformed to the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. Goat anti-rabbit Alexa 488 conjugated secondary antibodies (green) were used. Nuclei are counterstained with DAPI (blue). Sections were analyzed with a conventional Leica DMRP fluorescence microscope and images were obtained with a Hamamatsu Orca ER CCD Camera set up. Identical camera settings were used for the micrographs compared (exposure time 120 ms). Figures were identically processed with Adobe Photoshop and tables were composed in CoralDraw 11. (E) Western blot analysis of crude synaptosome preparation from wild type mouse retinas with a mixture of affinity purified myosin VIIa antibodies pAb2.1 and 2.2. A high molecular band at about 240 kDa was detected which was previously specified as myosin VIIa.2,3 (F, G) Silver enhanced immunogold labeling of myosin VIIa the photoreceptor synapse of wild type (F) and Myo7a4626SB/4626SB (G) mice. Immunoelectron microscopy was performed as previously described.11 The ultrathin sections were analyzed in a Tecnai 12 Biotwin transmission electron microscope. Bars: (A-D) 5 µm; (F) 0.15 µm; (G) 0.2 µm.
Figure 2. Indirect immunofluorescence with polyclonal myosin VIIa-tail antibody in retinal cryosections from Myo7a+/4626SB (B, D) and Myo7a4626SB/4626SB (C, E) mice. (A) Diagram of a rod photoreceptor cell: a light sensitive outer segment (OS) is linked via a non-motile connecting cilium (CC) with an inner segment (IS). The synaptic terminal (S) connects the photoreceptor cell with bipolar and horizontal cells. N: nucleus. Intense immunofluorescence labeling of myosin VIIa was present in the synapses in the outer plexiform layer of the Myo7a+/4626SB retina (B). Higher magnification image of the outer plexiform layer further visualized specific synaptic pattering (D). Synaptic staining was absent in the retina from Myo7a4626SB/4626SB mice (C, E). Methods, see legend of figure 1. Bars: (B-E) 5 µm. References 1. Reiners J, Reidel B, El Amraoui A, et al. Differential distribution of harmonin isoforms and their possible role in Usher-1 protein complexes in mammalian photoreceptor cells. Invest Ophthalmol Vis Sci. 2003;44:5006-5015. |
|||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH |
eLetters: 
