- Email: [email protected]
Connexin α1 and Cell Proliferation in the Developing Chick Retina
Experimental Neurology 156, 326–332 (1999) Article ID exnr.1999.7027, available online at http://www.idealibrary.com on
Connexin ␣1 and Cell Proliferation in the Developing Chick Retina David L. Becker and Peter Mobbs* Department of Anatomy and Developmental Biology and *Department of Physiology, University College London, Gower Street, London WC1E 6BT, United Kingdom Received October 15, 1998; accepted November 2, 1998
During the formation of the eye, high levels of connexin ␣1 (connexin 43) are expressed within the tissues of the cornea, lens, and neural retina. In order to determine whether connexin ␣1 plays a role in the regulation of cell proliferation we have used a novel antisense technique to reduce its expression early in development (embryonic days 2–4). Application of Pluronic gel, containing antisense oligodeoxynucleotides (ODNs) to connexin ␣1, to one eye of early chick embryos results in a rapid and significant reduction of ␣1 protein which lasts for 24–48 h. Embryos grown for 48 h, after ODN application to one eye, showed a marked reduction in the diameter of the treated, compared to that of the contralateral untreated, eye. Sections cut from the treated eyes showed that the retina was also reduced in size. TUNEL labeling and staining with propidium iodide showed that apoptosis within the retinae of both treated and untreated eyes was rare and thus that the reduction in the area of the retina brought about by antisense ODNs directed at connexin ␣1 was unlikely to be the result of increased cell death. However, the number of mitotic figures in the ventricular zone of the antisense-treated retinae revealed by propidium iodide staining was significantly reduced (P F 0.0001) to 53 ⴞ 3.5% (n ⴝ 5) of that in the contralateral untreated control eyes. Embryos in which one eye was sham operated, treated with pluronic gel, or treated with sense ODN showed no significant changes in eye size or in the number of mitotic figures within the neural retina. These results point to a role for connexin ␣1-mediated gap-junctional communication in controlling the early wave of neurogenesis in the chick retina. r 1999 Academic Press Key Words: antisense; Pluronic gel; connexin 43; neurogenesis; eye; development.
INTRODUCTION
The adult retina originates from a single layer of neuroepithelial cells. Precisely how it attains its final organization from these simple origins is not fully understood but the process clearly requires precise control of both cell type and cell number. During the 0014-4886/99 $30.00 Copyright r 1999 by Academic Press All rights of reproduction in any form reserved.
genesis of the retina each stem cell within the ventricular zone undergoes asymmetric division to give rise to another stem cell, which continues in the cell cycle, and a non-stem-cell daughter which extends a process toward the vitread surface of the retina and migrates to its final location (36). During the proliferative process many more cells than required are generated and later in development about half are eliminated in waves of programmed cell death (28, 30–32, 35). Gap-junctional communication has been implicated in the control of cell migration, differentiation, synapse formation, and correlation of the spontaneous electrical activity necessary for the refinement of synaptic connections (for reviews see 5, 6, 9, 13, 17–21). Electron microscopic studies have shown that there are high levels of gap junctions expressed in the developing retina during the periods in which these developmental events occur (monkey, 34; chick, 12; Xenopus, 10). However, a precise role for this form of intercellular communication in these processes has yet to be established. Recently gap-junctional communication has been implicated in the control of cell proliferation in the eye (16), brain, and other tissues (for reviews see 13, 22, 29, 39). Connexin ␣1 appears to be a predominant connexin during embryonic development, being expressed at high levels in many relatively undifferentiated tissues before being replaced with other connexins as the tissue matures and differentiates (1–3, 8, 14, 38). In order to investigate whether connexin ␣1 expression is involved in the regulation of cell proliferation in the eye and retina at these early times we have employed a novel antisense technique which uses Pluronic gel to deliver connexin ␣1-specific antisense oligodeoxynucleotides (ODNs) to the eye to reduce its expression (3). We show that reducing connexin ␣1 expression by these means results in a reduction in the eye size and a decrease in the number of dividing cells in the ventricular zone of the retina. MATERIALS AND METHODS
A detailed description of the design and selection of ODNs to connexin ␣1, tests for their specificity and penetration into tissue, and the measures used to
326
A ROLE FOR CONNEXIN ␣1 IN RETINAL NEUROGENESIS
establish the absence of toxic effects at the concentration employed here (for criteria see 37), appears in Becker et al. (3, 4). In brief, fertilized White Leghorn eggs were incubated at 38°C and staged according to Hamburger and Hamilton (15). Eggs were windowed and the vitelline and amniotic membranes over the area to be treated were opened using fine forceps (33). To aid in staging of the embryos ink was injected under the amniotic sac. A solution of 30% Pluronic F-127 gel (BASF Corp., Ludwigshafen, Germany) in phosphatebuffered saline (5 µL) was used to deliver unmodified connexin ␣1-specific antisense ODNs to one eye of developing chick embryos at stages 11–13 (embryonic day 2—E2). Pluronic gel is liquid at low temperatures, 0–4°C, but sets when dropped onto the embryo at physiological temperature. The rapid setting of the gel as it warms necessitates swift application and use of instruments that are kept on ice in order to prolong the working period. The gel stays in place for at least 12 h and acts as both a mild surfactant and a reservoir for the ODNs which are slowly released. Following application of the gel eggs were sealed with tape and replaced in the incubator for 48 h. Addition of an FITC tag to some ODNs was used to demonstrate their entry into treated tissue. Immunostaining of whole-mount embryos with connexinspecific antibodies, combined with confocal microscopy, was used to monitor the time course of connexin ␣1 knockdown and recovery and to check that the expression of other connexins was not affected. Control groups consisted of embryos in which one eye was shamoperated and 5 µL of phosphate-buffered saline (PBS), Pluronic gel alone, or Pluronic gel plus sense ␣1 ODN was applied. Connexin expression, as assessed by immunocytochemistry, was unaffected in all of the control groups. The ODNs used were selected on the basis that they were specific for connexin ␣1 and did not form homodimers, stem loops, or hairpins. ODNs were applied at 0.5–1.0 µM final concentration following dosedependent analysis during preliminary experiments covering a range of concentrations from 0.05 to 50 µM. Toxicity effects only became apparent with ODN concentrations greater than 10 µM. For histology embryos were gently removed from the egg, rinsed in PBS, and fixed in 4% paraformaldehyde for 30 min. Eye diameter was measured using a Leica MZ8 dissection microscope equipped with a video attachment. Following this the heads were removed, cut into left and right halves, and either sectioned using a Vibroslicer (Campden Instruments, London, UK) or processed as whole mounts. For vibroslicing the tissues were mounted in 5% agar (BDH, Poole, UK) and serial 100-µm-thick sections cut along the nasotemporal axis. Sections passing through the center of the retina were selected for further processing. The tissue slices were rinsed in PBS, permeabilized (0.1 M lysine, 0.1% Triton
327
X-100 in PBS) overnight, and then stained for 30 min in propidium iodide (1 µg/ml in PBS) before washing and mounting in Citifluor (Canterbury, London, UK) on cavity slides. TUNEL labeling was carried out on whole mounts and Vibratome sections using a Boerhinger Manheim (Lewes, UK) kit. Stained tissues were examined on a Leica TCS4D or TCS-SP laser scanning confocal microscope (Milton Keynes, UK) using excitation and emission wavelengths of 568 and 617 nm, respectively. Vibratome sections were examined at low power to check for folding of the neural retina. A series of overlapping single images were then taken, using a ⫻40 objective, around the circumference of the retina. At this power mitotic figures were readily visible within the layer of proliferating cells in the ventricular zone. These images were stored and used to construct montages from which the total number of mitotic cells and the length of the retina within the section could be determined. The number of mitotic figures per 100 µm was then calculated. Animals were selected at random for such analyses and counts were carried out blind. All figures are given as the mean ⫾ standard error. Eye size and the counts of mitotic figures are also given as ratios (treated/ untreated eye) to eliminate the variations between individuals resulting from slight differences in the stages of the embryos employed. Eye size data were compared using the Kruskal–Wallis and Dunn’s multiple comparison tests. Differences in the number of mitotic cells in the treated and untreated eyes of the same embryo were tested using the paired t test. Differences in the number of mitoses between the untreated and the treated eyes of embryos in which sense and antisense were used were established using Bonferroni’s multiple comparison test. RESULTS
We have previously demonstrated the time course and specificity of the knockdown of connexin ␣1 protein (see Fig. 2 in Refs. 3 and 4). Immunostaining of whole-mount treated embryos shows that the expression of connexin ␣1 protein is dramatically reduced in treated tissue but the expression patterns of connexins 1 and 2 are unaffected. Eye size was measured and the results were expressed as the ratio of the diameter of the treated eye to that of the control eye in each embryo of the antisensetreated group and those of the sham-operated, Pluronic gel-treated, and Pluronic gel with sense ODN-treated control groups. Connexin ␣1-specific antisense ODN application reduced the diameter of the treated eyes (Figs. 1A and 3) by 14 ⫾ 4% (n ⫽ 9). The ratios of the diameters of the control eyes treated with sense, Pluronic gel, or PBS were 1.02 ⫾ 0.01 (n ⫽ 11), 1.03 ⫾ 0.02 (n ⫽ 9), and 1.0 ⫾ 0.004 (n ⫽ 10), respectively. The
328
A ROLE FOR CONNEXIN ␣1 IN RETINAL NEUROGENESIS
329
FIG. 2. Montages of confocal microscope images of propidium iodide-stained retinae showing mitotic figures in the ventricular zone some of which are arrowed. The retina treated with antisense to knockdown connexin ␣1 shows a reduced number of mitotic figures compared to the control retina of that embryo. Bars 50 µm.
FIG. 1. A chick embryo which has had connexin ␣1-antisense ODNs in Pluronic gel applied to its right eye 48 h prior to fixation. The right eye is markedly smaller than the left, control eye (A). Low-power confocal microscope images of Vibratome sections through left and right eyes (B and C, respectively) of the embryo shown in A reveal that the smaller eye contains a small, unfolded, and otherwise apparently normal retina. Mitotic figures stained with propidium iodide are visible within the ventricular zone of the retina. High-power confocal microscope images showing dividing cells in the ventricular zone of control and antisense-treated retinae (D and E). Bars: (A) 2.5 mm, (B and C) 500 µm, (D and E) 10 µm.
330
BECKER AND MOBBS
difference between the antisense and the sense-treated groups was significant at the 1% level and between the antisense and the sham-operated and Pluronic acid controls at the 5% level (Kruskal–Wallis/Dunn’s multiple comparison test). In order to investigate whether the small, antisensetreated eyes contained a normal-sized neural retina within an abnormally small eye capsule as has been reported previously for small eye defects (7), nine heads were taken at random from the sense- and antisensetreated groups and sectioned. These showed the area of the retina to be reduced in proportion to the rest of the eye, to be unfolded, and to be of roughly normal thickness (Figs. 1B and 1C). In order to determine whether the reduction in eye size brought about by antisense connexin ␣1 was caused by induction of apoptotic cell death we carried out TUNEL staining of both whole-mount eyes and Vibratome slices. However, this staining revealed so few TUNEL-positive cells (four apoptotic cells in a total of four retinae) as to preclude any statistical analysis. Apoptotic cells appeared with a similarly extremely low incidence in slices stained with propidium iodide. These results suggest that there is little or no elevation of apoptotic cell death associated with knockdown of connexin ␣1 protein and that apoptosis is a rare event at this time unless the time taken to clear dead cells from the retina is very short. In order to determine the effect of connexin ␣1 knockdown on cell proliferation in the ventricular zone of the retina we counted the number of cells with mitotic figures in Vibratome sections from the treated and control eyes of the antisense- and sense-treated groups. A total of nine pairs of eyes were selected at random from antisense- (n ⫽ 5) and sense- (n ⫽ 4)
FIG. 4. Treatment with antisense ODNs directed at connexin ␣1 results in a significant reduction in the number of mitotic figures in the ventricular zone of these retina compared to the contralateral, untreated eyes of the same embryo (P ⬍ 0.0005—paired t test) and the eyes in sense-treated control animals (P ⬍ 0.0001—Student’s unpaired t test).
treated groups. For each eye the number of mitotic figures (Figs. 1D and 1E) in the ventricular zone of the retina was imaged and counted within a single sagittal section through the midline of the eye capsule. An overlapping series of images taken of the sections was montaged and the length of the retina measured (Fig. 2). There was a large decrease in the number of mitotic figures in the connexin ␣1-treated eyes compared to the contralateral, control eyes (1.34 ⫾ 0.1 and 2.55 ⫾ 0.12 mitoses per 100 µM, respectively, n ⫽ 5) (Fig. 4). This difference was significant at 0.05% level (paired t test). In contrast the number of mitotic figures in the sensetreated embryos was roughly equal in both the treated and the control eyes (2.06 ⫾ 0.15, 1.98 ⫾ 0.12, n ⫽ 4). The reduction in the number of mitotic figures in the ␣1 ODN-treated eye was also significant compared to the number of mitotic cells seen in both the experimental and the control eyes of the sense-treated animals (P ⫽ 0.01 and 0.05, respectively, Bonferroni’s multiple comparison test). Allowing for minor variations in the stage of the embryos by comparing the ratio of the number of mitotic figures in the untreated and treated eyes of antisense (0.53 ⫾ 0.04, n ⫽ 5) with that of the sense-treated control embryos (1.04 ⫾ 0.03, n ⫽ 4) gives results significant at the 0.01% level (unpaired Student’s t test). DISCUSSION
FIG. 3. Histogram showing the ratio of treated and untreated eye diameters in embryos in which one eye has been treated with antisense or sense ODNs in Pluronic gel or with Pluronic gel or PBS alone. Antisense treatment results in the treated eye being smaller than that of the untreated contralateral eye. The symbols show the ratios of the diameters of the treated to those of the untreated eyes of the individual embryos in each group.
The normal organogenesis of the eye requires the smooth orchestration of a cascade of gene expression. The expression patterns of many of these genes are interrelated and often have complex feedback loops that act on one another. In the developing retina the expression of connexins many be related to, and affected by, several other diverse genes, including growth factors cell adhesion molecules (11, 22–24, 26, 27), patterning genes (3), and retinoids (14), all of which
A ROLE FOR CONNEXIN ␣1 IN RETINAL NEUROGENESIS
contribute to the coordination of retinal development. Here we have shown that knockdown of connexin ␣1 with an antisense construct in the developing chick eye, at E2–4, a time when connexin ␣1 is normally expressed at high levels and cell proliferation is high, reduces eye size. The growth of the eye as a whole is affected, with the retina, sclera, and lens remaining in normal proportion with one another. A possible explanation for the small size of the antisense-treated eyes is that the reduction in connexin ␣1 expression leads to apoptotic cell death throughout the eye. However, both propidium iodide and TUNEL labeling show apoptotic cells to be virtually absent in the retinae of the experimental and control groups. It is possible that apoptosis occurred at early times after the application of the antisense ODN and that the dead cells were phagocytosed and cleared from the retina within the 48 h before we examined the eyes. However, given the large and significant reduction in the number of mitotic cells in the retinae of eyes treated with antisense ODNs directed at connexin ␣1, a more likely explanation appears to be that reduced connexin ␣1 expression leads to smaller retinae through a reduction in the rate of cell division. The reduction in cell proliferation seen when connexin ␣1 protein is reduced suggests a requirement for good gap-junctional communication via this connexin for cell division to occur at normal rates in the formation of the early eye cup, unlike the situation in the subependymal and rostral migratory stream of the mouse, in which decreased ␣ expression is associated with increased proliferation (25). The mechanism by which connexin ␣1 influences cell proliferation in the eye and retina is not clear. ACKNOWLEDGMENTS We thank Jeremy Cook and Colin Green for their comments on the manuscript. D.L.B. is a Royal Society University Research Fellow and thanks them for their support. P.M. is supported by the Wellcome trust.
REFERENCES 1.
Becker, D. L., C. David-LeClerc, and A. E. Warner 1992. The relationship of gap junctions and compaction in the preimplantation mouse embryo. Development Suppl.: 113–118. 2. Becker, D. L., W. H. Evans, C. R. Green, and A. E. Warner. 1995. Functional analysis of amino acid sequences in connexin 43 involved in intercellular communication through gap junctions. J. Cell Sci. 108: 1455–1467. 3. Becker, D. L., I. McGonnel, H. Makarenkova, K. Patel, C. Tickle, J. Lorimer, and C. R. Green. 1999. Roles for alpha 1 connexin in morphogenesis of chick embryos using a novel antisense approach. Dev. Genet., in press. 4. Becker, D. L., J. S. Lin, and C. R. Green. 1999. Pluronic gel as a means of antisense delivery. In R. Leslie, Ed., Antisense Technology in the CNS. Oxford Univ. Press, London, in press.
331
5. Bruzzone, R., and C. Ressot. 1997. Connexins, gap junctions and cell–cell signalling in the nervous system. Eur. J. Neurosci. 9: 1–6. 6. Cook, J. E., and D. L. Becker. 1995. Gap junctions in the retina. Microsc. Res. Tech. 31: 408–419. 7. Coulombre, A. J. 1956. The role of intraocular pressure in the development of the chick eye. J. Exp. Zool. 133: 211–225. 8. Dermeitzel, R., O. Traub, T. K. Hwang, E. Beyer, M. V. L. Bennet, and D. C. Spray. 1989. Differential expression of three gap junction proteins in developing and mature brain tissues. Proc. Natl. Acad. Sci. USA 86: 10148–10152. 9. Dermietzel, R., and D. C. Spray. 1993. Gap junctions in the brain: Where, what type, how many and why? Trends Neurosci. 16: 186–192. 10. Dixon, J. S., and J. R. Cronley-Dillon. 1972. The fine structure of the developing retina in Xenopus laevis. J. Embryol. Exp. Morphol. 28: 659–666. 11. Doble, B. W., and E. Kardami. 1995. Basic fibroblast growth factor stimulates connexin 43 expression and intercellular communication of cardiac fibroblasts. Mol. Cell Biochem. 143: 81–87. 12. Fujisawa, H., H. Morioka, K. Watanabe, and H. Nakamura. 1976. A decay of gap junctions in association with cell differentiation of neural retina in chick embryonic development. J. Cell Sci. 22: 585–596. 13. Fulton, B. P. 1995. Gap junctions in the developing nervous system. Perspect. Dev. Neurobiol. 2: 327–334. 14. Green, C. R., L. Bowles, A. Crawley, and C. Tickle. 1994. Expression of the connexin 43 gap junctional protein in tissues at the tip of the chick limb bud is related to epithelial– mesenchymal interactions that mediate morphogenesis. Dev. Biol. 161: 12–21. 15. Hamburger, V., and H. Hamilton. 1951. A series of normal stages in the development of the chick embryo. J. Morphol. 88: 49–92. 16. Illia, M., and G. Jeffery. 1996. Delayed neurogenesis in the albino retina: Evidence of a role for melanin in regulating the pace of cell generation. Dev. Brain Res. 95: 176–183. 17. Kandler, K. 1997. Coordination of neuronal activity by gap junctions in the developing neocortex. Semin. Dev. Biol. 8: 43–51. 18. Kandler, K., and L. C. Katz. 1995. Neuronal coupling and uncoupling in the developing nervous system. Curr. Opin. Neurobiol. 5: 98–105. 19. Kandler, K., and L. C. Katz. 1995. Coordination of neuronal activity in developing visual cortex by gap junction-mediated biochemical communication. J. Neurosci. 18(4): 1419–1427. 20. Kandler, K., and L. C. Katz. 1998. Relationship between dye coupling and spontaneous activity in developing ferret visual cortex. Dev. Neurosci. 20: 59–64. 21. Kumar, N. M., and N. B. Gilula. 1996. The gap junction communication channel. Cell 84: 381–388. 22. Loewenstein, W. R., and B. Rose. 1992. The cell–cell channel in control of growth. Semin. Cell. Biol. 3: 59–79. 23. Makarenkova, H., D. L. Becker, C. Tickle, and A. E. Warner. 1997. Fibroblast growth factor 4 directs gap junction expression in the mesenchyme of the vertebrate limb bud. J. Cell Biol. 138: 1–13. 24. Meyer, R. A., D. W. Laird, J-P. Revel, and R. G. Johnson. 1992. Inhibition of gap junction and adherins junction assembly by connexin and A-CAM antibodies. J. Cell Biol. 119: 179–189. 25. Miragall, F., P. Albiez, H. Bartels, U. de-Vries, and R. Dermietzel. 1997. Expression of the gap junction protein connexin43 in the subependymal layer and rostral migratory stream of the
332
BECKER AND MOBBS mouse: Evidence for an inverse correlation between intensity of connexin43 expression and cell proliferation activity. Cell Tissue Res. 287: 243–253.
26.
27.
28. 29.
30.
31.
Musil, L. S., B. A. Cunningham, G. M. Edelman, and D. A. Goodenough. 1990. Differential phosphorylation of the gap junction protein connexin 43 in junctional communicationcompetent and -deficient cell lines. J. Cell Biol. 111: 2077–2088. Nadarajah, B., H. Makarenkova, D. L. Becker, W. H. Evans, and J. G. Parnavelas. 1998. Basic FGF increases communication between cells of the developing neocortex. J. Neurosci. 18: 7881–7890. Oppenheim, R. W. 1991. Cell death during the development of the nervous system. Annu. Rev. Neurosci. 14: 453–501. Pitts, J. D., M. E. Finbow, and E. Kam. 1988. Junctional communication and cellular differentiation. Br. J. Cancer. 58: 52–57. Prada, C., J. Puga, L. Perez-Mendez, R. Lopez, and G. Ramirez. 1991. Spatial and temporal patterns of neurogenesis in the chick retina. Eur. J. Neurosci. 3: 559–569. Rager, G., and U. Rager. 1978. Systems matching by degeneration. I. A quantitative electron microscopic study of the generation and degeneration of retinal ganglion cells in the chicken. Exp. Brain Res. 33: 65–78.
32. Snow, R. L., and J. A. Robson. 1994. Ganglion cell neurogenesis, migration and early differentiation in the chick retina. Neuroscience 58: 399–409. 33. Tickle, C. 1993. Chick limb buds. Chap. 13, pages 119–126 in C. D. Stern and P. W. H. Holland, Eds., Essential Developmental biology: A Practical Approach. Oxford Univ. Press, London. 34. Townes-Anderson, E., and G. Raviola. 1981. The formation and distribution of intercellular junctions in the rhesus monkey optic cup: The early development of the cilio-iridic and sensory retinas. Dev. Biol. 85: 209–232. ¨ tting, and S. Thanos. 1990. Target dependence 35. Vanselow, J., D. U of chick ganglion cells during embryogenesis: Cell survival and dendritic development. J. Comp. Neurol. 295: 235–247. 36. Waid, D. K., and S. McLoon. 1995. Immediate differentiation of ganglion cells following mitosis in the developing retina. Neuron 14: 117–124. 37. Wagner, R. W. 1994. Gene inhibition using antisense oligodeoxynucleotides. Nature 372: 333–335. 38. Weins, D., L. Jensen, J. Jasper, and J. Becker. 1995. Developmental expression of connexins in the chick embryo myocardium and other tissues. Anat. Rec. 241: 541–553. 39. Yamasaki, H. 1996. Role of disrupted gap junctional intercellular communication in detection and characterization of carcinogens. Mutat. Res. 365: 91–105.
Connexin ␣1 and Cell Proliferation in the Developing Chick Retina David L. Becker and Peter Mobbs* Department of Anatomy and Developmental Biology and *Department of Physiology, University College London, Gower Street, London WC1E 6BT, United Kingdom Received October 15, 1998; accepted November 2, 1998
During the formation of the eye, high levels of connexin ␣1 (connexin 43) are expressed within the tissues of the cornea, lens, and neural retina. In order to determine whether connexin ␣1 plays a role in the regulation of cell proliferation we have used a novel antisense technique to reduce its expression early in development (embryonic days 2–4). Application of Pluronic gel, containing antisense oligodeoxynucleotides (ODNs) to connexin ␣1, to one eye of early chick embryos results in a rapid and significant reduction of ␣1 protein which lasts for 24–48 h. Embryos grown for 48 h, after ODN application to one eye, showed a marked reduction in the diameter of the treated, compared to that of the contralateral untreated, eye. Sections cut from the treated eyes showed that the retina was also reduced in size. TUNEL labeling and staining with propidium iodide showed that apoptosis within the retinae of both treated and untreated eyes was rare and thus that the reduction in the area of the retina brought about by antisense ODNs directed at connexin ␣1 was unlikely to be the result of increased cell death. However, the number of mitotic figures in the ventricular zone of the antisense-treated retinae revealed by propidium iodide staining was significantly reduced (P F 0.0001) to 53 ⴞ 3.5% (n ⴝ 5) of that in the contralateral untreated control eyes. Embryos in which one eye was sham operated, treated with pluronic gel, or treated with sense ODN showed no significant changes in eye size or in the number of mitotic figures within the neural retina. These results point to a role for connexin ␣1-mediated gap-junctional communication in controlling the early wave of neurogenesis in the chick retina. r 1999 Academic Press Key Words: antisense; Pluronic gel; connexin 43; neurogenesis; eye; development.
INTRODUCTION
The adult retina originates from a single layer of neuroepithelial cells. Precisely how it attains its final organization from these simple origins is not fully understood but the process clearly requires precise control of both cell type and cell number. During the 0014-4886/99 $30.00 Copyright r 1999 by Academic Press All rights of reproduction in any form reserved.
genesis of the retina each stem cell within the ventricular zone undergoes asymmetric division to give rise to another stem cell, which continues in the cell cycle, and a non-stem-cell daughter which extends a process toward the vitread surface of the retina and migrates to its final location (36). During the proliferative process many more cells than required are generated and later in development about half are eliminated in waves of programmed cell death (28, 30–32, 35). Gap-junctional communication has been implicated in the control of cell migration, differentiation, synapse formation, and correlation of the spontaneous electrical activity necessary for the refinement of synaptic connections (for reviews see 5, 6, 9, 13, 17–21). Electron microscopic studies have shown that there are high levels of gap junctions expressed in the developing retina during the periods in which these developmental events occur (monkey, 34; chick, 12; Xenopus, 10). However, a precise role for this form of intercellular communication in these processes has yet to be established. Recently gap-junctional communication has been implicated in the control of cell proliferation in the eye (16), brain, and other tissues (for reviews see 13, 22, 29, 39). Connexin ␣1 appears to be a predominant connexin during embryonic development, being expressed at high levels in many relatively undifferentiated tissues before being replaced with other connexins as the tissue matures and differentiates (1–3, 8, 14, 38). In order to investigate whether connexin ␣1 expression is involved in the regulation of cell proliferation in the eye and retina at these early times we have employed a novel antisense technique which uses Pluronic gel to deliver connexin ␣1-specific antisense oligodeoxynucleotides (ODNs) to the eye to reduce its expression (3). We show that reducing connexin ␣1 expression by these means results in a reduction in the eye size and a decrease in the number of dividing cells in the ventricular zone of the retina. MATERIALS AND METHODS
A detailed description of the design and selection of ODNs to connexin ␣1, tests for their specificity and penetration into tissue, and the measures used to
326
A ROLE FOR CONNEXIN ␣1 IN RETINAL NEUROGENESIS
establish the absence of toxic effects at the concentration employed here (for criteria see 37), appears in Becker et al. (3, 4). In brief, fertilized White Leghorn eggs were incubated at 38°C and staged according to Hamburger and Hamilton (15). Eggs were windowed and the vitelline and amniotic membranes over the area to be treated were opened using fine forceps (33). To aid in staging of the embryos ink was injected under the amniotic sac. A solution of 30% Pluronic F-127 gel (BASF Corp., Ludwigshafen, Germany) in phosphatebuffered saline (5 µL) was used to deliver unmodified connexin ␣1-specific antisense ODNs to one eye of developing chick embryos at stages 11–13 (embryonic day 2—E2). Pluronic gel is liquid at low temperatures, 0–4°C, but sets when dropped onto the embryo at physiological temperature. The rapid setting of the gel as it warms necessitates swift application and use of instruments that are kept on ice in order to prolong the working period. The gel stays in place for at least 12 h and acts as both a mild surfactant and a reservoir for the ODNs which are slowly released. Following application of the gel eggs were sealed with tape and replaced in the incubator for 48 h. Addition of an FITC tag to some ODNs was used to demonstrate their entry into treated tissue. Immunostaining of whole-mount embryos with connexinspecific antibodies, combined with confocal microscopy, was used to monitor the time course of connexin ␣1 knockdown and recovery and to check that the expression of other connexins was not affected. Control groups consisted of embryos in which one eye was shamoperated and 5 µL of phosphate-buffered saline (PBS), Pluronic gel alone, or Pluronic gel plus sense ␣1 ODN was applied. Connexin expression, as assessed by immunocytochemistry, was unaffected in all of the control groups. The ODNs used were selected on the basis that they were specific for connexin ␣1 and did not form homodimers, stem loops, or hairpins. ODNs were applied at 0.5–1.0 µM final concentration following dosedependent analysis during preliminary experiments covering a range of concentrations from 0.05 to 50 µM. Toxicity effects only became apparent with ODN concentrations greater than 10 µM. For histology embryos were gently removed from the egg, rinsed in PBS, and fixed in 4% paraformaldehyde for 30 min. Eye diameter was measured using a Leica MZ8 dissection microscope equipped with a video attachment. Following this the heads were removed, cut into left and right halves, and either sectioned using a Vibroslicer (Campden Instruments, London, UK) or processed as whole mounts. For vibroslicing the tissues were mounted in 5% agar (BDH, Poole, UK) and serial 100-µm-thick sections cut along the nasotemporal axis. Sections passing through the center of the retina were selected for further processing. The tissue slices were rinsed in PBS, permeabilized (0.1 M lysine, 0.1% Triton
327
X-100 in PBS) overnight, and then stained for 30 min in propidium iodide (1 µg/ml in PBS) before washing and mounting in Citifluor (Canterbury, London, UK) on cavity slides. TUNEL labeling was carried out on whole mounts and Vibratome sections using a Boerhinger Manheim (Lewes, UK) kit. Stained tissues were examined on a Leica TCS4D or TCS-SP laser scanning confocal microscope (Milton Keynes, UK) using excitation and emission wavelengths of 568 and 617 nm, respectively. Vibratome sections were examined at low power to check for folding of the neural retina. A series of overlapping single images were then taken, using a ⫻40 objective, around the circumference of the retina. At this power mitotic figures were readily visible within the layer of proliferating cells in the ventricular zone. These images were stored and used to construct montages from which the total number of mitotic cells and the length of the retina within the section could be determined. The number of mitotic figures per 100 µm was then calculated. Animals were selected at random for such analyses and counts were carried out blind. All figures are given as the mean ⫾ standard error. Eye size and the counts of mitotic figures are also given as ratios (treated/ untreated eye) to eliminate the variations between individuals resulting from slight differences in the stages of the embryos employed. Eye size data were compared using the Kruskal–Wallis and Dunn’s multiple comparison tests. Differences in the number of mitotic cells in the treated and untreated eyes of the same embryo were tested using the paired t test. Differences in the number of mitoses between the untreated and the treated eyes of embryos in which sense and antisense were used were established using Bonferroni’s multiple comparison test. RESULTS
We have previously demonstrated the time course and specificity of the knockdown of connexin ␣1 protein (see Fig. 2 in Refs. 3 and 4). Immunostaining of whole-mount treated embryos shows that the expression of connexin ␣1 protein is dramatically reduced in treated tissue but the expression patterns of connexins 1 and 2 are unaffected. Eye size was measured and the results were expressed as the ratio of the diameter of the treated eye to that of the control eye in each embryo of the antisensetreated group and those of the sham-operated, Pluronic gel-treated, and Pluronic gel with sense ODN-treated control groups. Connexin ␣1-specific antisense ODN application reduced the diameter of the treated eyes (Figs. 1A and 3) by 14 ⫾ 4% (n ⫽ 9). The ratios of the diameters of the control eyes treated with sense, Pluronic gel, or PBS were 1.02 ⫾ 0.01 (n ⫽ 11), 1.03 ⫾ 0.02 (n ⫽ 9), and 1.0 ⫾ 0.004 (n ⫽ 10), respectively. The
328
A ROLE FOR CONNEXIN ␣1 IN RETINAL NEUROGENESIS
329
FIG. 2. Montages of confocal microscope images of propidium iodide-stained retinae showing mitotic figures in the ventricular zone some of which are arrowed. The retina treated with antisense to knockdown connexin ␣1 shows a reduced number of mitotic figures compared to the control retina of that embryo. Bars 50 µm.
FIG. 1. A chick embryo which has had connexin ␣1-antisense ODNs in Pluronic gel applied to its right eye 48 h prior to fixation. The right eye is markedly smaller than the left, control eye (A). Low-power confocal microscope images of Vibratome sections through left and right eyes (B and C, respectively) of the embryo shown in A reveal that the smaller eye contains a small, unfolded, and otherwise apparently normal retina. Mitotic figures stained with propidium iodide are visible within the ventricular zone of the retina. High-power confocal microscope images showing dividing cells in the ventricular zone of control and antisense-treated retinae (D and E). Bars: (A) 2.5 mm, (B and C) 500 µm, (D and E) 10 µm.
330
BECKER AND MOBBS
difference between the antisense and the sense-treated groups was significant at the 1% level and between the antisense and the sham-operated and Pluronic acid controls at the 5% level (Kruskal–Wallis/Dunn’s multiple comparison test). In order to investigate whether the small, antisensetreated eyes contained a normal-sized neural retina within an abnormally small eye capsule as has been reported previously for small eye defects (7), nine heads were taken at random from the sense- and antisensetreated groups and sectioned. These showed the area of the retina to be reduced in proportion to the rest of the eye, to be unfolded, and to be of roughly normal thickness (Figs. 1B and 1C). In order to determine whether the reduction in eye size brought about by antisense connexin ␣1 was caused by induction of apoptotic cell death we carried out TUNEL staining of both whole-mount eyes and Vibratome slices. However, this staining revealed so few TUNEL-positive cells (four apoptotic cells in a total of four retinae) as to preclude any statistical analysis. Apoptotic cells appeared with a similarly extremely low incidence in slices stained with propidium iodide. These results suggest that there is little or no elevation of apoptotic cell death associated with knockdown of connexin ␣1 protein and that apoptosis is a rare event at this time unless the time taken to clear dead cells from the retina is very short. In order to determine the effect of connexin ␣1 knockdown on cell proliferation in the ventricular zone of the retina we counted the number of cells with mitotic figures in Vibratome sections from the treated and control eyes of the antisense- and sense-treated groups. A total of nine pairs of eyes were selected at random from antisense- (n ⫽ 5) and sense- (n ⫽ 4)
FIG. 4. Treatment with antisense ODNs directed at connexin ␣1 results in a significant reduction in the number of mitotic figures in the ventricular zone of these retina compared to the contralateral, untreated eyes of the same embryo (P ⬍ 0.0005—paired t test) and the eyes in sense-treated control animals (P ⬍ 0.0001—Student’s unpaired t test).
treated groups. For each eye the number of mitotic figures (Figs. 1D and 1E) in the ventricular zone of the retina was imaged and counted within a single sagittal section through the midline of the eye capsule. An overlapping series of images taken of the sections was montaged and the length of the retina measured (Fig. 2). There was a large decrease in the number of mitotic figures in the connexin ␣1-treated eyes compared to the contralateral, control eyes (1.34 ⫾ 0.1 and 2.55 ⫾ 0.12 mitoses per 100 µM, respectively, n ⫽ 5) (Fig. 4). This difference was significant at 0.05% level (paired t test). In contrast the number of mitotic figures in the sensetreated embryos was roughly equal in both the treated and the control eyes (2.06 ⫾ 0.15, 1.98 ⫾ 0.12, n ⫽ 4). The reduction in the number of mitotic figures in the ␣1 ODN-treated eye was also significant compared to the number of mitotic cells seen in both the experimental and the control eyes of the sense-treated animals (P ⫽ 0.01 and 0.05, respectively, Bonferroni’s multiple comparison test). Allowing for minor variations in the stage of the embryos by comparing the ratio of the number of mitotic figures in the untreated and treated eyes of antisense (0.53 ⫾ 0.04, n ⫽ 5) with that of the sense-treated control embryos (1.04 ⫾ 0.03, n ⫽ 4) gives results significant at the 0.01% level (unpaired Student’s t test). DISCUSSION
FIG. 3. Histogram showing the ratio of treated and untreated eye diameters in embryos in which one eye has been treated with antisense or sense ODNs in Pluronic gel or with Pluronic gel or PBS alone. Antisense treatment results in the treated eye being smaller than that of the untreated contralateral eye. The symbols show the ratios of the diameters of the treated to those of the untreated eyes of the individual embryos in each group.
The normal organogenesis of the eye requires the smooth orchestration of a cascade of gene expression. The expression patterns of many of these genes are interrelated and often have complex feedback loops that act on one another. In the developing retina the expression of connexins many be related to, and affected by, several other diverse genes, including growth factors cell adhesion molecules (11, 22–24, 26, 27), patterning genes (3), and retinoids (14), all of which
A ROLE FOR CONNEXIN ␣1 IN RETINAL NEUROGENESIS
contribute to the coordination of retinal development. Here we have shown that knockdown of connexin ␣1 with an antisense construct in the developing chick eye, at E2–4, a time when connexin ␣1 is normally expressed at high levels and cell proliferation is high, reduces eye size. The growth of the eye as a whole is affected, with the retina, sclera, and lens remaining in normal proportion with one another. A possible explanation for the small size of the antisense-treated eyes is that the reduction in connexin ␣1 expression leads to apoptotic cell death throughout the eye. However, both propidium iodide and TUNEL labeling show apoptotic cells to be virtually absent in the retinae of the experimental and control groups. It is possible that apoptosis occurred at early times after the application of the antisense ODN and that the dead cells were phagocytosed and cleared from the retina within the 48 h before we examined the eyes. However, given the large and significant reduction in the number of mitotic cells in the retinae of eyes treated with antisense ODNs directed at connexin ␣1, a more likely explanation appears to be that reduced connexin ␣1 expression leads to smaller retinae through a reduction in the rate of cell division. The reduction in cell proliferation seen when connexin ␣1 protein is reduced suggests a requirement for good gap-junctional communication via this connexin for cell division to occur at normal rates in the formation of the early eye cup, unlike the situation in the subependymal and rostral migratory stream of the mouse, in which decreased ␣ expression is associated with increased proliferation (25). The mechanism by which connexin ␣1 influences cell proliferation in the eye and retina is not clear. ACKNOWLEDGMENTS We thank Jeremy Cook and Colin Green for their comments on the manuscript. D.L.B. is a Royal Society University Research Fellow and thanks them for their support. P.M. is supported by the Wellcome trust.
REFERENCES 1.
Becker, D. L., C. David-LeClerc, and A. E. Warner 1992. The relationship of gap junctions and compaction in the preimplantation mouse embryo. Development Suppl.: 113–118. 2. Becker, D. L., W. H. Evans, C. R. Green, and A. E. Warner. 1995. Functional analysis of amino acid sequences in connexin 43 involved in intercellular communication through gap junctions. J. Cell Sci. 108: 1455–1467. 3. Becker, D. L., I. McGonnel, H. Makarenkova, K. Patel, C. Tickle, J. Lorimer, and C. R. Green. 1999. Roles for alpha 1 connexin in morphogenesis of chick embryos using a novel antisense approach. Dev. Genet., in press. 4. Becker, D. L., J. S. Lin, and C. R. Green. 1999. Pluronic gel as a means of antisense delivery. In R. Leslie, Ed., Antisense Technology in the CNS. Oxford Univ. Press, London, in press.
331
5. Bruzzone, R., and C. Ressot. 1997. Connexins, gap junctions and cell–cell signalling in the nervous system. Eur. J. Neurosci. 9: 1–6. 6. Cook, J. E., and D. L. Becker. 1995. Gap junctions in the retina. Microsc. Res. Tech. 31: 408–419. 7. Coulombre, A. J. 1956. The role of intraocular pressure in the development of the chick eye. J. Exp. Zool. 133: 211–225. 8. Dermeitzel, R., O. Traub, T. K. Hwang, E. Beyer, M. V. L. Bennet, and D. C. Spray. 1989. Differential expression of three gap junction proteins in developing and mature brain tissues. Proc. Natl. Acad. Sci. USA 86: 10148–10152. 9. Dermietzel, R., and D. C. Spray. 1993. Gap junctions in the brain: Where, what type, how many and why? Trends Neurosci. 16: 186–192. 10. Dixon, J. S., and J. R. Cronley-Dillon. 1972. The fine structure of the developing retina in Xenopus laevis. J. Embryol. Exp. Morphol. 28: 659–666. 11. Doble, B. W., and E. Kardami. 1995. Basic fibroblast growth factor stimulates connexin 43 expression and intercellular communication of cardiac fibroblasts. Mol. Cell Biochem. 143: 81–87. 12. Fujisawa, H., H. Morioka, K. Watanabe, and H. Nakamura. 1976. A decay of gap junctions in association with cell differentiation of neural retina in chick embryonic development. J. Cell Sci. 22: 585–596. 13. Fulton, B. P. 1995. Gap junctions in the developing nervous system. Perspect. Dev. Neurobiol. 2: 327–334. 14. Green, C. R., L. Bowles, A. Crawley, and C. Tickle. 1994. Expression of the connexin 43 gap junctional protein in tissues at the tip of the chick limb bud is related to epithelial– mesenchymal interactions that mediate morphogenesis. Dev. Biol. 161: 12–21. 15. Hamburger, V., and H. Hamilton. 1951. A series of normal stages in the development of the chick embryo. J. Morphol. 88: 49–92. 16. Illia, M., and G. Jeffery. 1996. Delayed neurogenesis in the albino retina: Evidence of a role for melanin in regulating the pace of cell generation. Dev. Brain Res. 95: 176–183. 17. Kandler, K. 1997. Coordination of neuronal activity by gap junctions in the developing neocortex. Semin. Dev. Biol. 8: 43–51. 18. Kandler, K., and L. C. Katz. 1995. Neuronal coupling and uncoupling in the developing nervous system. Curr. Opin. Neurobiol. 5: 98–105. 19. Kandler, K., and L. C. Katz. 1995. Coordination of neuronal activity in developing visual cortex by gap junction-mediated biochemical communication. J. Neurosci. 18(4): 1419–1427. 20. Kandler, K., and L. C. Katz. 1998. Relationship between dye coupling and spontaneous activity in developing ferret visual cortex. Dev. Neurosci. 20: 59–64. 21. Kumar, N. M., and N. B. Gilula. 1996. The gap junction communication channel. Cell 84: 381–388. 22. Loewenstein, W. R., and B. Rose. 1992. The cell–cell channel in control of growth. Semin. Cell. Biol. 3: 59–79. 23. Makarenkova, H., D. L. Becker, C. Tickle, and A. E. Warner. 1997. Fibroblast growth factor 4 directs gap junction expression in the mesenchyme of the vertebrate limb bud. J. Cell Biol. 138: 1–13. 24. Meyer, R. A., D. W. Laird, J-P. Revel, and R. G. Johnson. 1992. Inhibition of gap junction and adherins junction assembly by connexin and A-CAM antibodies. J. Cell Biol. 119: 179–189. 25. Miragall, F., P. Albiez, H. Bartels, U. de-Vries, and R. Dermietzel. 1997. Expression of the gap junction protein connexin43 in the subependymal layer and rostral migratory stream of the
332
BECKER AND MOBBS mouse: Evidence for an inverse correlation between intensity of connexin43 expression and cell proliferation activity. Cell Tissue Res. 287: 243–253.
26.
27.
28. 29.
30.
31.
Musil, L. S., B. A. Cunningham, G. M. Edelman, and D. A. Goodenough. 1990. Differential phosphorylation of the gap junction protein connexin 43 in junctional communicationcompetent and -deficient cell lines. J. Cell Biol. 111: 2077–2088. Nadarajah, B., H. Makarenkova, D. L. Becker, W. H. Evans, and J. G. Parnavelas. 1998. Basic FGF increases communication between cells of the developing neocortex. J. Neurosci. 18: 7881–7890. Oppenheim, R. W. 1991. Cell death during the development of the nervous system. Annu. Rev. Neurosci. 14: 453–501. Pitts, J. D., M. E. Finbow, and E. Kam. 1988. Junctional communication and cellular differentiation. Br. J. Cancer. 58: 52–57. Prada, C., J. Puga, L. Perez-Mendez, R. Lopez, and G. Ramirez. 1991. Spatial and temporal patterns of neurogenesis in the chick retina. Eur. J. Neurosci. 3: 559–569. Rager, G., and U. Rager. 1978. Systems matching by degeneration. I. A quantitative electron microscopic study of the generation and degeneration of retinal ganglion cells in the chicken. Exp. Brain Res. 33: 65–78.
32. Snow, R. L., and J. A. Robson. 1994. Ganglion cell neurogenesis, migration and early differentiation in the chick retina. Neuroscience 58: 399–409. 33. Tickle, C. 1993. Chick limb buds. Chap. 13, pages 119–126 in C. D. Stern and P. W. H. Holland, Eds., Essential Developmental biology: A Practical Approach. Oxford Univ. Press, London. 34. Townes-Anderson, E., and G. Raviola. 1981. The formation and distribution of intercellular junctions in the rhesus monkey optic cup: The early development of the cilio-iridic and sensory retinas. Dev. Biol. 85: 209–232. ¨ tting, and S. Thanos. 1990. Target dependence 35. Vanselow, J., D. U of chick ganglion cells during embryogenesis: Cell survival and dendritic development. J. Comp. Neurol. 295: 235–247. 36. Waid, D. K., and S. McLoon. 1995. Immediate differentiation of ganglion cells following mitosis in the developing retina. Neuron 14: 117–124. 37. Wagner, R. W. 1994. Gene inhibition using antisense oligodeoxynucleotides. Nature 372: 333–335. 38. Weins, D., L. Jensen, J. Jasper, and J. Becker. 1995. Developmental expression of connexins in the chick embryo myocardium and other tissues. Anat. Rec. 241: 541–553. 39. Yamasaki, H. 1996. Role of disrupted gap junctional intercellular communication in detection and characterization of carcinogens. Mutat. Res. 365: 91–105.