- Email: [email protected]
Early born lineage of retinal neurons express class III β-tubulin isotype
BR A IN RE S E A RCH 1 1 76 ( 20 0 7 ) 1 1 –1 7
a v a i l a b l e a t w w w. s c i e n c e d i r e c t . c o m
w w w. e l s e v i e r. c o m / l o c a t e / b r a i n r e s
Research Report
Early born lineage of retinal neurons express class III β-tubulin isotype Rajesh K. Sharma⁎, Peter A. Netland Department of Ophthalmology, University of Tennessee Health Science Center, Memphis, TN 38105, USA
A R T I C LE I N FO
AB S T R A C T
Article history:
Aim: Class III β-tubulin, a constituent of neuronal microtubules, has been frequently used as a
Accepted 12 July 2007
marker for the neuronal lineage in developmental biology. In retina, it is often used as a marker
Available online 22 August 2007
for ganglion cells. We investigated the developmental expression of this protein in retina and identified the cell types expressing it to gain a better understanding of whether preferred
Keywords:
expression of this isotype in certain retinal neurons plays a cell specific role, or whether it is
Ganglion cell
only a part of an intrinsic developmental program. Methods: Immunohistochemistry was done
Cytoskeleton
using an antibody against class III β-tubulin and other retinal cell specific markers in adult
RPE
retinae of mice. Rabbit and human retinae were used to investigate if there are any species-
Cell lineage
specific differences. Results: Class III β-tubulin was found in ganglion cells, certain amacrine cells, some horizontal cell processes and cone photoreceptors. Class III β-tubulin was already
TUJ1
expressed in the earliest developmental stage studied (Embryonic day 14) in developing nerve fiber layer but became distinct at the day of birth when immunoreactive cells were located in the ganglion cell layer (ganglion and displaced amacrine cells), proximal parts of neuroblastic/ inner nuclear layer (amacrine cells) and distal part of neuroblastic/outer nuclear layer (photoreceptors). In one animal, class III β-tubulin containing bodies were found in the retinal pigment epithelium cells. Conclusions: Class III β-tubulin is not solely expressed by ganglion cells and, therefore, cannot be used as an exclusive marker for these cells. Results show that the expression of class III β-tubulin was not related to cell morphology or cell function, but rather to the cell lineage (early born retinal neurons) suggesting that the expression of class III β-tubulin in certain cell types may be due to the cell specific developmental program. © 2007 Elsevier B.V. All rights reserved.
1.
Introduction
Eukaryotic cytoskeleton is made up of three distinct components: microtubules, intermediate filaments and microfilaments. Microtubules are vital for cellular functions. They form tracks for trafficking membrane-bound organelles and are essential for the neurite growth during development and regeneration. Although the neuronal and non-neuronal micro-
tubules are composed of the same basic constituents, neurons contain different isotypes of tubulins, post-translational modifications and microtubule-associated proteins (Heidemann, 1996). Multiple genes exist for both α- and β-tubulin. Most αand β-tubulin isotypes are universally expressed; however, some are preferentially expressed in certain tissue types. Most notably, class III and IVa β-tubulins are neuron-specific. The reason for preferred expression of various isotypes is not clear.
⁎ Corresponding author. E-mail address: [email protected] (R.K. Sharma). 0006-8993/$ – see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.brainres.2007.07.090
12
BR A IN RE S EA RCH 1 1 76 ( 20 0 7 ) 1 1 –17
The class III β-tubulin isotype is one of the six tubulin isotypes expressed in mammalian tissue (Sullivan and Cleveland, 1986). Its expression is associated with early neural differentiation (Easter et al., 1993; Lee et al., 1990; Moody et al., 1989). Retinal ganglion cells, that are involved in glaucoma, have been identified by class III β-tubulin in vitro (Fournier and McKerracher, 1997; Hu et al., 2006). Whether
this isotype is also expressed by other retinal neurons is not well studied. In this study, we have investigated the developmental expression of class III β-tubulin in mouse retina and compared its expression in mice, rabbit, and human retinae. Our results provide a clue as to whether cell-specific expression of class III β-tubulin is because of its suitability with neuronal
BR A IN RE S E A RCH 1 1 76 ( 20 0 7 ) 1 1 –1 7
environment, or because genes of certain isotypes are a part of cell-specific developmental programs.
2.
Observations and results
2.1.
Immunoreactivity in developing mice retina
2.1.1.
E 14 retinae
Faint immunoreactivity was seen in what appeared to be the developing nerve fiber layer (Fig. 1A).
2.1.2.
PN 0 retinae
In PN 0 retina, a thin immunoreactive inner plexiform layer (IPL) was noted. There were immunoreactive cells on both sides of the IPL (Fig. 1B). Immunoreactive cells could also be seen on the distal part of the neuroblastic cell mass. These cells projected proximally to various depths of the neuroblastic cell mass with a center-to-peripheral gradient. Projections from the cells in the central part of the retina were larger than in peripheral parts. No outer plexiform layer (OPL) was formed at this stage.
2.1.3.
PN 3 retinae
Immunoreactive cells in the distal part of the retina became more distinct and so did immunoreactivity in cells adjacent to IPL (Fig. 1C).
2.1.4.
PN 7 retina
The immunoreactive IPL was thicker and showed three illdefined sublaminae (Fig. 1D). Immunoreactive cells were seen on both the proximal and the distal sides of the IPL. Certain cells in the GCL were large and more intensely stained. NFL was intensely immunoreactive. Immunoreactive cells in the proximal parts of the INL were located toward the IPL. Immunoreactive cells were also seen in the ONL. These cells were less differentiated in the peripheral parts than in the central. In the peripheral parts, these cells projected toward, but not all the cells reached, the OPL. However, in the central part, projections of the cells could be seen reaching the OPL.
2.1.5.
2.1.6.
13
Adult retinae
Immunoreactivity in the adult retina was intense (Fig. 1F). Immunoreactive photoreceptor cells could be seen in the ONL. The cell bodies of these cells were located the distal parts of the ONL and their projections reached up to the OPL (Fig. 2B). The inner segments of these photoreceptors were also immunoreactive (Fig. 2A and B). Cells in the INL were mostly located in the proximal parts and morphologically were of two subtypes; large and small. There were clusters of large cells in the extreme peripheral part of the retina (Fig. 2C). Large cells were sparse and widely spaced out. These cells also had long neurites. Smaller cells were more numerous and had smaller neurite projections. Immunoreactive cells in the GCL were also distinctly of two morphological types; small and relatively more intensely stained large cells. However, larger cell types were more numerous in the GCL than in the INL. Immunoreactivity was intense in the NFL. As expected, calbindin immunoreactivity was seen in horizontal cells located at the distal most part of the INL including their projections in the OPL, and in certain amacrine cells in the INL and the GCL (Fig. 3C and D). Three distinct immunoreactive plexuses were seen in the IPL (Fig. 3C). Some of the TUJ-1 immunoreactive cells in the INL colocalized with calbindin; however, there were numerous calbindin immunoreactive cells that did not colocalize with TUJ1 and vice versa (Fig. 3C and D). Perikarya of the cells proximal to OPL did not colocalize, but some weak colocalization could be seen in their processes. In the GCL, cells that were weakly immunoreactive to TUJ-1 colocalized with calbindin. Large cells intensely stained with TUJ1 in the GCL projected to the nerve fiber layer. These cells were morphologically judged to be the ganglion cells. These cells did not colocalize with calbindin (Fig. 3D). PNA labels cone photoreceptors, and as expected staining was seen in certain photoreceptor cells including their terminals in the OPL (Fig. 3E and F). In mice retinae, it was more prominent in inner segments and OPL terminals (Fig. 3E). PNA staining colocalized with the TUJ-1 immunoreactivity both in rabbits and in mice (Fig. 3E and F).
2.2.
Immunoreactivity in adult rabbit and human retina
2.2.1.
Rabbit retina
PN 14 retinae
Immunoreactivity approached adult levels (Fig. 1E). At this stage, photoreceptor outer segments were distinct. Immunoreactivity was observed in certain photoreceptor cells in the ONL and their inner segments. In addition, staining was present in OPL, IPL and cells adjacent to it.
TUJ1 immunoreactivity in the rabbit retina was identical to that of mice retinae (Fig. 2E, F and G). Immunoreactivity was seen in certain photoreceptor cells (cones) including their cell bodies, neurite and the inner segments. Immunoreactive cells were also
Fig. 1 – Immunoreactivity in developing mice retina: (A) Shows faint immunoreactivity in the NFL of an E 14 retina. (B) PN 0 Retina showing immunoreactivity in differentiating cells on both sides of the IPL (brackets; vertical arrows), distal neuroblastic cell (NBC) mass projecting proximally (inset; horizontal arrow), as well as in the IPL and the NFL. (C) Immunoreactivity in the NBC in a PN 3 retina (horizontal arrows). (D) At PN 7 retina the immunoreactivity in the IPL is divided in three ill-defined sublaminae (black arrowheads). There are reactive cells on both sides of the IPL (brackets; vertical arrows) including certain large cells in the GCL (thick vertical arrows). Immunoreactive cells in the ONL are still predominantly in the central part (horizontal arrows) and projected toward the OPL (small arrowheads; inset). (E) PN 14 and (F) adult retinae showing immunoreactive photoreceptors(horizontal arrows; also see Fig. 2B), and their inner segments (oblique arrows; also see Fig. 2A and B). Their projections reach the OPL (small arrowheads). Immunoreactive cells in the INL are of two morphological types; large and small (small and large vertical arrows pointing downwards). There were immunoreactive cells in GCL (also see Fig. 2A) and the reactivity in IPL showed three distinct sublaminae (black arrowheads). Horizontal arrows = cells in the ONL, oblique arrows = photoreceptor inner segments, small and large vertical arrows = morphologically distinct cells on either sides of IPL, small arrow heads = OPL and large arrowheads = plexuses in IPL, scale bar = 40 μm.
14 BR A IN RE S EA RCH 1 1 76 ( 20 0 7 ) 1 1 –17
Fig. 2 – Immunoreactivity in adult retina of different species. There are no significant differences in TUJ1 immunoreactivity in mouse (A, B, C), rabbit (E, F, G), and human retinae (I, J, K). Immunoreactivity is seen in certain photoreceptor cells (horizontal arrows; cones) including their inner segments (oblique arrows), cells in the proximal part of the INL (downwardly pointing vertical arrows; amacrine cells), cells in the GCL (upwardly pointing vertical arrows; ganglion cells and certain displaced amacrine cells) and the NFL. Presumed ganglion cells are large and more intensely stained (large upwardly pointing vertical arrows). Extreme peripheral parts of the retina showed clusters of large cell types (arrows; C). Controls showed no immunoreactivity (D, H and L). Horizontal arrows = cells in the ONL, oblique arrows = photoreceptor inner segments, small and large vertical arrows = morphologically distinct cells on either sides of IPL, small arrow heads = OPL and large arrowheads = plexuses in IPL, scale bar = 40 μm.
BR A IN RE S E A RCH 1 1 76 ( 20 0 7 ) 1 1 –1 7
present in the proximal part of the INL (amacrine cells) and the GCL. However, as compared to the mice retinae, the two distinct morphological types (large and small) were less distinct. NFL was intensely immunoreactive. IPL had immunoreactive neurites that formed three ill-defined layers. Interestingly, in some rabbit retinal sections, immunoreactive bodies were seen in the RPE. Because RPE (especially from old animals) is autofluorescent, this possibility was ruled out as the staining was not visible through nonspecific filters. Some, but not all, immunoreactive bodies weakly colocalized with PNA (Fig. 3A and B).
2.2.2.
Human retina
Basically, the pattern of immunoreactivity in human retina resembled mice and rabbit (Fig. 2I, J and K). Immunoreactive photoreceptor cells were present in the ONL. Immunoreactive cells were also present in proximal INL (amacrines) and the GCL. Nerve fiber layer was immunoreactive (Fig. 2I and K).
3.
15
et al., 1997; Ranganathan et al., 1996). This raises an interesting possibility that composition of microtubules and polymorphism in related genes could possibly play a role in the pathogenesis of glaucoma. Another interesting finding was localization of class III β-tubulin immunoreactive bodies in the RPE cells. We also observed that some of these intracellular bodies were also labeled by PNA. The importance of these findings is not clear. It is possible that this was unique to this animal or a result of processing of the eye. Interestingly, fovea is cone rich, and cones have class III β-tubulin; however, an excessive amount of this protein can be cytotoxic (Hari et al., 2003). These findings require further investigation, specifically, if drusen found in association with AMD also expresses class III β-tubulin.
4.
Experimental procedure
4.1.
Animals and tissue preparation
Discussion
Our results show that class III β-tubulin is expressed by cells other than RGCs in the retina; therefore, this labeling alone cannot be used as an unequivocal label of ganglion cells. Based on morphological and immunocytochemical criteria, class III βtubulin was found in cone photoreceptors, certain amacrine cells, processes of certain horizontal cells as well as the ganglion cells. Retinal neurons are born in two phases. Ganglion cells, cones, horizontal cells and certain amacrine cells are born in early phase, whereas rods and bipolar cells in late phase (Harman et al., 1992; Sharma and Ehinger, 2003; Sharma et al., 2003; Sidman, 1961). Our results show that the cells expressing class III β-tubulin were morphologically and functionally diverse, but they all belonged to the early born cell lineage including the calbindin expressing amacrine cells. The expression of class III βtubulin coincides with the beginning of differentiation in these neurons (Sharma and Ehinger, 1997). Interestingly, there are different types of amacrine cells. Calbindin expressing amacrine cells are among those born early during development (Sharma et al., 2003). This suggests that expression of class III β-tubulin is related to cell-specific developmental programs rather than morphology or function. There was no significant speciesspecific difference in the adult retina of the 3 species studied. We did not study the expression profile during development in rabbit or human. However, in adult retina the classes of neurons expressing class III β-tubulin were comparable. The ability of cells to survive treatment in certain cancers is associated with altered expression (Kavallaris et al., 1997) or mutation (Gonzalez-Garay et al., 1999) in α- and β-tubulins, suggesting that composition of cytoarchitecture can have a significant effect on the ability of the cells to survive or undergo apoptosis. A class of anticancer drugs that include paclitaxel (Taxol) and estramustine stabilize spindle microtubules through microtubule polymerization and block mitosis (Jordan et al., 1993). There are differences in the sensitivity of microtubules composed of class III β-tubulin and those assembled from unfractionated tubulin to these drugs (Lu and Luduena, 1993). Class III β-tubulin transcript expression has been shown to alter in tumors showing resistance to antimicrotubule/microtubule-polymerizing drugs (Kavallaris
Embryonic day (E) 14 (n = 2) eyes were enucleated from embryos obtained by cesarean section after sacrificing the pregnant mice. Eyes were also obtained from postnatal day (PN) 0 (n = 3), 3 (n = 3), 7 (n = 3) and 14 (n = 3), and adult C57/black6 mice. The mice were anesthetized with CO2 prior to decapitation. Eyes were enucleated and fixed as described below. Adult rabbit eyes were also used (n = 3). Animals were treated according to the ARVO resolution on animal experimentation. Human eyes were obtained from Mid-South Eye Bank. The donor was a 57year-old adult male who had no history of any ocular disorder. All the eyes were briefly fixed in 4% paraformaldehyde in phosphate-buffered saline (PBS; 0.1 M phosphate, 0.85% NaCl; pH 7.4). The anterior segment of the eyes was removed and the posterior segment was fixed in the same fixative for 24 h. The tissue was rinsed in PBS and then transferred sequentially to PBS containing 5%, 10% and 20% sucrose. The eyes were stored in PBS with 20% sucrose until sectioning. Twelve-micron thick sections were cut on a cryostat and mounted on glass slides coated with gelatin. They were subsequently stored at − 70 °C until used. Sagittal sections were cut in the embryonic and postnatal retinae, and the sections passing close to the optic nerve were used for immunohistochemistry.
4.2.
Immunohistochemistry
Frozen sections were thawed at room temperature and then rinsed with PBS containing 1% bovine serum albumin (BSA; Sigma Chemical Co.). The sections were incubated with the primary antibodies in a humidified chamber overnight. The following primary antibodies were used; TUJ1 monoclonal antibody against class III β-tubulin (1: 500; CRP Inc. Berkeley CA) and calbindin (1: 500; Sigma®) labeling horizontal cells and certain amacrine cells. In addition, cone photoreceptors were also labeled with Peanut agglutinin (PNA) which binds to glycoconjugates associated with cone cell membranes (Blanks and Johnson, 1984). Optimum working concentration and incubation time for the antibodies was determined earlier in pilot experiments. After the incubation, the slides were rinsed with PBS and incubated for 1 h in appropriate secondary antibody conjugated with Cy3 (Jackson Laboratories PA), rinsed
16
BR A IN RE S EA RCH 1 1 76 ( 20 0 7 ) 1 1 –17
in PBS and mounted with Vectashield® (Vector Laboratories Burlingame CA). At least four slides from each specimen were stained, and in each experiment, controls were obtained by omitting the primary antibody. For double labeling, TUJ1 and calbindin antibodies (raised in animal other than mice) were mixed in dilutions previously determined by pilot experiments. Cy3-conjugated anti-mouse and Cy2-conjugated second sec-
ondary antibody were also mixed and used as secondary antibody. For peanut agglutinin (PNA) histochemistry, PNA lectin (FITC-conjugated peanut lectin 0.2 mg/mL Arachis hypogaea; Sigma, St. Louis, MO) was dissolved in the blocking solution at 1:100 dilution and added to the PF fixed retinal sections, incubated for 45 min, washed and mounted in Vectashield®. For double labeling experiments, PNA was
BR A IN RE S E A RCH 1 1 76 ( 20 0 7 ) 1 1 –1 7
added after the last wash in immunohistochemistry. Tissue was viewed on a Nikon photomicroscope fitted with a digital camera.
Acknowledgments The authors thank Professor Dianna A. Johnson for support and Marina A. Kedrov for technical help with the experiments. This work was supported by a generous grant by Gail and Richard Siegal, The Hyde Foundation, NEI grant EY-13080 and an unrestricted grant from Research to Prevent Blindness.
Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.brainres.2007.07.090. REFERENCES
Blanks, J.C., Johnson, L.V., 1984. Specific binding of peanut lectin to a class of retinal photoreceptor cells. A species comparison. Invest. Ophthalmol. Vis. Sci. 25, 546–557. Easter Jr., S.S., Ross, L.S., Frankfurter, A., 1993. Initial tract formation in the mouse brain. J. Neurosci. 13, 285–299. Fournier, A.E., McKerracher, L., 1997. Expression of specific tubulin isotypes increases during regeneration of injured CNS neurons, but not after the application of brain-derived neurotrophic factor (BDNF). J. Neurosci. 17, 4623–4632. Gonzalez-Garay, M.L., Chang, L., Blade, K., Menick, D.R., Cabral, F., 1999. A beta-tubulin leucine cluster involved in microtubule assembly and paclitaxel resistance. J. Biol. Chem. 274, 23875–23882. Hari, M., Yang, H., Zeng, C., Canizales, M., Cabral, F., 2003. Expression of class III beta-tubulin reduces microtubule assembly and confers resistance to paclitaxel. Cell Motil. Cytoskelet. 56, 45–56. Harman, A.M., Sanderson, K.J., Beazley, L.D., 1992. Biphasic retinal neurogenesis in the brush-tailed possum, Trichosurus vulpecula: further evidence for the mechanisms involved in formation of ganglion cell density gradients. J. Comp. Neurol. 325, 595–606.
17
Heidemann, S.R., 1996. Cytoplasmic mechanisms of axonal and dendritic growth in neurons. Int. Rev. Cytol. 165, 235–296. Hu, Y., Cui, Q., Harvey, A.R., 2007. Interactive effects of C3, cyclic AMP and ciliary neurotrophic factor on adult retinal ganglion cell survival and axonal regeneration. Mol. Cell Neurosci. 146 (3), 986–999. Jordan, M.A., Toso, R.J., Thrower, D., Wilson, L., 1993. Mechanism of mitotic block and inhibition of cell proliferation by taxol at low concentrations. Proc. Natl. Acad. Sci. U. S. A. 90, 9552–9556. Kavallaris, M., Kuo, D.Y., Burkhart, C.A., Regl, D.L., Norris, M.D., Haber, M., Horwitz, S.B., 1997. Taxol-resistant epithelial ovarian tumors are associated with altered expression of specific beta-tubulin isotypes. J. Clin. Invest. 100, 1282–1293. Lee, M.K., Tuttle, J.B., Rebhun, L.I., Cleveland, D.W., Frankfurter, A., 1990. The expression and posttranslational modification of a neuron-specific beta-tubulin isotype during chick embryogenesis. Cell Motil. Cytoskelet. 17, 118–132. Lu, Q., Luduena, R.F., 1993. Removal of beta III isotype enhances taxol induced microtubule assembly. Cell Struct. Funct. 18, 173–182. Moody, S.A., Quigg, M.S., Frankfurter, A., 1989. Development of the peripheral trigeminal system in the chick revealed by an isotype-specific anti-beta-tubulin monoclonal antibody. J. Comp. Neurol. 279, 567–580. Ranganathan, S., Dexter, D.W., Benetatos, C.A., Chapman, A.E., Tew, K.D., Hudes, G.R., 1996. Increase of beta(III)- and beta(IVa)-tubulin isotopes in human prostate carcinoma cells as a result of estramustine resistance. Cancer Res. 56, 2584–2589. Sharma, R.K., Ehinger, B., 1997. Mitosis in developing rabbit retina: an immunohistochemical study. Exp. Eye Res. 64, 97–106. Sharma, R.K., Ehinger, B., 2003. Development and structure of the retina. In: Kaufman, P.L., Alm, A. (Eds.), Adler's Physiology of the Eye: Clinical Application. Mosby, St. Louis, pp. 319–347. Sharma, R.K., O'Leary, T.E., Fields, C.M., Johnson, D.A., 2003. Development of the outer retina in the mouse. Brain Res. Dev. Brain Res. 145, 93–105. Sidman, R.L., 1961. Histiogenesis of mouse retina studied with thymidine-H3. In: Smelser, G.K. (Ed.), The Structure of the Eye. Academic Press Inc. Ltd, London, pp. 487–506. Sullivan, K.F., Cleveland, D.W., 1986. Identification of conserved isotype-defining variable region sequences for four vertebrate beta tubulin polypeptide classes. Proc. Natl. Acad. Sci. U. S. A. 83, 4327–4331.
Fig. 3 – Colocalization of class III β-tubulin with calbindin and PNA. (A and B) Immunoreactive bodies in RPE layer with variable staining intensities (vertical arrows in A and B). Some, but not all, immunoreactive bodies weakly colocalized with PNA. (C, D) Colocalization of TUJ1 (red) and calbindin (green) immunoreactivity in mouse (C) and rabbit (D) retinae. In both species, as expected calbindin immunoreactivity is seen in horizontal cells, including their projections to the OPL (small arrowheads). Immunoreactivity is also seen in a subpopulation of amacrine cells in the INL and displaced amacrine cells in the GCL (green and yellow labeling in GCL). Three immunoreactive plexuses are seen in the IPL (black arrowheads). TUJ1 immunoreactivity colocalized with some, but not all amacrine cells (green [thick vertical arrow] and red [small vertical arrow] labeling). In the GCL many cells that are weakly immunoreactive to TUJ1 colocalized with calbindin (yellow labeling in GCL). Perikarya of calbindin immunoreactive horizontal cells did not colocalize with TUJ1 (green labeling; C and D), but some processes weakly colocalized. (E, F) Show colocalization of TUJ1 (red) and PNA (green) in mouse (E) and rabbit (F) retinae. PNA labeling in cone photoreceptors colocalized with TUJ1 immunoreactivity in both rabbit and mouse retina including the inner segments (oblique arrows), and their terminals in the OPL. Horizontal arrows = cells in the ONL, oblique arrows = photoreceptor inner segments, small and large vertical arrows = morphologically distinct cells on either sides of IPL, small arrow heads = OPL and large arrowheads = plexuses in IPL, scale bar = 40 μm.
a v a i l a b l e a t w w w. s c i e n c e d i r e c t . c o m
w w w. e l s e v i e r. c o m / l o c a t e / b r a i n r e s
Research Report
Early born lineage of retinal neurons express class III β-tubulin isotype Rajesh K. Sharma⁎, Peter A. Netland Department of Ophthalmology, University of Tennessee Health Science Center, Memphis, TN 38105, USA
A R T I C LE I N FO
AB S T R A C T
Article history:
Aim: Class III β-tubulin, a constituent of neuronal microtubules, has been frequently used as a
Accepted 12 July 2007
marker for the neuronal lineage in developmental biology. In retina, it is often used as a marker
Available online 22 August 2007
for ganglion cells. We investigated the developmental expression of this protein in retina and identified the cell types expressing it to gain a better understanding of whether preferred
Keywords:
expression of this isotype in certain retinal neurons plays a cell specific role, or whether it is
Ganglion cell
only a part of an intrinsic developmental program. Methods: Immunohistochemistry was done
Cytoskeleton
using an antibody against class III β-tubulin and other retinal cell specific markers in adult
RPE
retinae of mice. Rabbit and human retinae were used to investigate if there are any species-
Cell lineage
specific differences. Results: Class III β-tubulin was found in ganglion cells, certain amacrine cells, some horizontal cell processes and cone photoreceptors. Class III β-tubulin was already
TUJ1
expressed in the earliest developmental stage studied (Embryonic day 14) in developing nerve fiber layer but became distinct at the day of birth when immunoreactive cells were located in the ganglion cell layer (ganglion and displaced amacrine cells), proximal parts of neuroblastic/ inner nuclear layer (amacrine cells) and distal part of neuroblastic/outer nuclear layer (photoreceptors). In one animal, class III β-tubulin containing bodies were found in the retinal pigment epithelium cells. Conclusions: Class III β-tubulin is not solely expressed by ganglion cells and, therefore, cannot be used as an exclusive marker for these cells. Results show that the expression of class III β-tubulin was not related to cell morphology or cell function, but rather to the cell lineage (early born retinal neurons) suggesting that the expression of class III β-tubulin in certain cell types may be due to the cell specific developmental program. © 2007 Elsevier B.V. All rights reserved.
1.
Introduction
Eukaryotic cytoskeleton is made up of three distinct components: microtubules, intermediate filaments and microfilaments. Microtubules are vital for cellular functions. They form tracks for trafficking membrane-bound organelles and are essential for the neurite growth during development and regeneration. Although the neuronal and non-neuronal micro-
tubules are composed of the same basic constituents, neurons contain different isotypes of tubulins, post-translational modifications and microtubule-associated proteins (Heidemann, 1996). Multiple genes exist for both α- and β-tubulin. Most αand β-tubulin isotypes are universally expressed; however, some are preferentially expressed in certain tissue types. Most notably, class III and IVa β-tubulins are neuron-specific. The reason for preferred expression of various isotypes is not clear.
⁎ Corresponding author. E-mail address: [email protected] (R.K. Sharma). 0006-8993/$ – see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.brainres.2007.07.090
12
BR A IN RE S EA RCH 1 1 76 ( 20 0 7 ) 1 1 –17
The class III β-tubulin isotype is one of the six tubulin isotypes expressed in mammalian tissue (Sullivan and Cleveland, 1986). Its expression is associated with early neural differentiation (Easter et al., 1993; Lee et al., 1990; Moody et al., 1989). Retinal ganglion cells, that are involved in glaucoma, have been identified by class III β-tubulin in vitro (Fournier and McKerracher, 1997; Hu et al., 2006). Whether
this isotype is also expressed by other retinal neurons is not well studied. In this study, we have investigated the developmental expression of class III β-tubulin in mouse retina and compared its expression in mice, rabbit, and human retinae. Our results provide a clue as to whether cell-specific expression of class III β-tubulin is because of its suitability with neuronal
BR A IN RE S E A RCH 1 1 76 ( 20 0 7 ) 1 1 –1 7
environment, or because genes of certain isotypes are a part of cell-specific developmental programs.
2.
Observations and results
2.1.
Immunoreactivity in developing mice retina
2.1.1.
E 14 retinae
Faint immunoreactivity was seen in what appeared to be the developing nerve fiber layer (Fig. 1A).
2.1.2.
PN 0 retinae
In PN 0 retina, a thin immunoreactive inner plexiform layer (IPL) was noted. There were immunoreactive cells on both sides of the IPL (Fig. 1B). Immunoreactive cells could also be seen on the distal part of the neuroblastic cell mass. These cells projected proximally to various depths of the neuroblastic cell mass with a center-to-peripheral gradient. Projections from the cells in the central part of the retina were larger than in peripheral parts. No outer plexiform layer (OPL) was formed at this stage.
2.1.3.
PN 3 retinae
Immunoreactive cells in the distal part of the retina became more distinct and so did immunoreactivity in cells adjacent to IPL (Fig. 1C).
2.1.4.
PN 7 retina
The immunoreactive IPL was thicker and showed three illdefined sublaminae (Fig. 1D). Immunoreactive cells were seen on both the proximal and the distal sides of the IPL. Certain cells in the GCL were large and more intensely stained. NFL was intensely immunoreactive. Immunoreactive cells in the proximal parts of the INL were located toward the IPL. Immunoreactive cells were also seen in the ONL. These cells were less differentiated in the peripheral parts than in the central. In the peripheral parts, these cells projected toward, but not all the cells reached, the OPL. However, in the central part, projections of the cells could be seen reaching the OPL.
2.1.5.
2.1.6.
13
Adult retinae
Immunoreactivity in the adult retina was intense (Fig. 1F). Immunoreactive photoreceptor cells could be seen in the ONL. The cell bodies of these cells were located the distal parts of the ONL and their projections reached up to the OPL (Fig. 2B). The inner segments of these photoreceptors were also immunoreactive (Fig. 2A and B). Cells in the INL were mostly located in the proximal parts and morphologically were of two subtypes; large and small. There were clusters of large cells in the extreme peripheral part of the retina (Fig. 2C). Large cells were sparse and widely spaced out. These cells also had long neurites. Smaller cells were more numerous and had smaller neurite projections. Immunoreactive cells in the GCL were also distinctly of two morphological types; small and relatively more intensely stained large cells. However, larger cell types were more numerous in the GCL than in the INL. Immunoreactivity was intense in the NFL. As expected, calbindin immunoreactivity was seen in horizontal cells located at the distal most part of the INL including their projections in the OPL, and in certain amacrine cells in the INL and the GCL (Fig. 3C and D). Three distinct immunoreactive plexuses were seen in the IPL (Fig. 3C). Some of the TUJ-1 immunoreactive cells in the INL colocalized with calbindin; however, there were numerous calbindin immunoreactive cells that did not colocalize with TUJ1 and vice versa (Fig. 3C and D). Perikarya of the cells proximal to OPL did not colocalize, but some weak colocalization could be seen in their processes. In the GCL, cells that were weakly immunoreactive to TUJ-1 colocalized with calbindin. Large cells intensely stained with TUJ1 in the GCL projected to the nerve fiber layer. These cells were morphologically judged to be the ganglion cells. These cells did not colocalize with calbindin (Fig. 3D). PNA labels cone photoreceptors, and as expected staining was seen in certain photoreceptor cells including their terminals in the OPL (Fig. 3E and F). In mice retinae, it was more prominent in inner segments and OPL terminals (Fig. 3E). PNA staining colocalized with the TUJ-1 immunoreactivity both in rabbits and in mice (Fig. 3E and F).
2.2.
Immunoreactivity in adult rabbit and human retina
2.2.1.
Rabbit retina
PN 14 retinae
Immunoreactivity approached adult levels (Fig. 1E). At this stage, photoreceptor outer segments were distinct. Immunoreactivity was observed in certain photoreceptor cells in the ONL and their inner segments. In addition, staining was present in OPL, IPL and cells adjacent to it.
TUJ1 immunoreactivity in the rabbit retina was identical to that of mice retinae (Fig. 2E, F and G). Immunoreactivity was seen in certain photoreceptor cells (cones) including their cell bodies, neurite and the inner segments. Immunoreactive cells were also
Fig. 1 – Immunoreactivity in developing mice retina: (A) Shows faint immunoreactivity in the NFL of an E 14 retina. (B) PN 0 Retina showing immunoreactivity in differentiating cells on both sides of the IPL (brackets; vertical arrows), distal neuroblastic cell (NBC) mass projecting proximally (inset; horizontal arrow), as well as in the IPL and the NFL. (C) Immunoreactivity in the NBC in a PN 3 retina (horizontal arrows). (D) At PN 7 retina the immunoreactivity in the IPL is divided in three ill-defined sublaminae (black arrowheads). There are reactive cells on both sides of the IPL (brackets; vertical arrows) including certain large cells in the GCL (thick vertical arrows). Immunoreactive cells in the ONL are still predominantly in the central part (horizontal arrows) and projected toward the OPL (small arrowheads; inset). (E) PN 14 and (F) adult retinae showing immunoreactive photoreceptors(horizontal arrows; also see Fig. 2B), and their inner segments (oblique arrows; also see Fig. 2A and B). Their projections reach the OPL (small arrowheads). Immunoreactive cells in the INL are of two morphological types; large and small (small and large vertical arrows pointing downwards). There were immunoreactive cells in GCL (also see Fig. 2A) and the reactivity in IPL showed three distinct sublaminae (black arrowheads). Horizontal arrows = cells in the ONL, oblique arrows = photoreceptor inner segments, small and large vertical arrows = morphologically distinct cells on either sides of IPL, small arrow heads = OPL and large arrowheads = plexuses in IPL, scale bar = 40 μm.
14 BR A IN RE S EA RCH 1 1 76 ( 20 0 7 ) 1 1 –17
Fig. 2 – Immunoreactivity in adult retina of different species. There are no significant differences in TUJ1 immunoreactivity in mouse (A, B, C), rabbit (E, F, G), and human retinae (I, J, K). Immunoreactivity is seen in certain photoreceptor cells (horizontal arrows; cones) including their inner segments (oblique arrows), cells in the proximal part of the INL (downwardly pointing vertical arrows; amacrine cells), cells in the GCL (upwardly pointing vertical arrows; ganglion cells and certain displaced amacrine cells) and the NFL. Presumed ganglion cells are large and more intensely stained (large upwardly pointing vertical arrows). Extreme peripheral parts of the retina showed clusters of large cell types (arrows; C). Controls showed no immunoreactivity (D, H and L). Horizontal arrows = cells in the ONL, oblique arrows = photoreceptor inner segments, small and large vertical arrows = morphologically distinct cells on either sides of IPL, small arrow heads = OPL and large arrowheads = plexuses in IPL, scale bar = 40 μm.
BR A IN RE S E A RCH 1 1 76 ( 20 0 7 ) 1 1 –1 7
present in the proximal part of the INL (amacrine cells) and the GCL. However, as compared to the mice retinae, the two distinct morphological types (large and small) were less distinct. NFL was intensely immunoreactive. IPL had immunoreactive neurites that formed three ill-defined layers. Interestingly, in some rabbit retinal sections, immunoreactive bodies were seen in the RPE. Because RPE (especially from old animals) is autofluorescent, this possibility was ruled out as the staining was not visible through nonspecific filters. Some, but not all, immunoreactive bodies weakly colocalized with PNA (Fig. 3A and B).
2.2.2.
Human retina
Basically, the pattern of immunoreactivity in human retina resembled mice and rabbit (Fig. 2I, J and K). Immunoreactive photoreceptor cells were present in the ONL. Immunoreactive cells were also present in proximal INL (amacrines) and the GCL. Nerve fiber layer was immunoreactive (Fig. 2I and K).
3.
15
et al., 1997; Ranganathan et al., 1996). This raises an interesting possibility that composition of microtubules and polymorphism in related genes could possibly play a role in the pathogenesis of glaucoma. Another interesting finding was localization of class III β-tubulin immunoreactive bodies in the RPE cells. We also observed that some of these intracellular bodies were also labeled by PNA. The importance of these findings is not clear. It is possible that this was unique to this animal or a result of processing of the eye. Interestingly, fovea is cone rich, and cones have class III β-tubulin; however, an excessive amount of this protein can be cytotoxic (Hari et al., 2003). These findings require further investigation, specifically, if drusen found in association with AMD also expresses class III β-tubulin.
4.
Experimental procedure
4.1.
Animals and tissue preparation
Discussion
Our results show that class III β-tubulin is expressed by cells other than RGCs in the retina; therefore, this labeling alone cannot be used as an unequivocal label of ganglion cells. Based on morphological and immunocytochemical criteria, class III βtubulin was found in cone photoreceptors, certain amacrine cells, processes of certain horizontal cells as well as the ganglion cells. Retinal neurons are born in two phases. Ganglion cells, cones, horizontal cells and certain amacrine cells are born in early phase, whereas rods and bipolar cells in late phase (Harman et al., 1992; Sharma and Ehinger, 2003; Sharma et al., 2003; Sidman, 1961). Our results show that the cells expressing class III β-tubulin were morphologically and functionally diverse, but they all belonged to the early born cell lineage including the calbindin expressing amacrine cells. The expression of class III βtubulin coincides with the beginning of differentiation in these neurons (Sharma and Ehinger, 1997). Interestingly, there are different types of amacrine cells. Calbindin expressing amacrine cells are among those born early during development (Sharma et al., 2003). This suggests that expression of class III β-tubulin is related to cell-specific developmental programs rather than morphology or function. There was no significant speciesspecific difference in the adult retina of the 3 species studied. We did not study the expression profile during development in rabbit or human. However, in adult retina the classes of neurons expressing class III β-tubulin were comparable. The ability of cells to survive treatment in certain cancers is associated with altered expression (Kavallaris et al., 1997) or mutation (Gonzalez-Garay et al., 1999) in α- and β-tubulins, suggesting that composition of cytoarchitecture can have a significant effect on the ability of the cells to survive or undergo apoptosis. A class of anticancer drugs that include paclitaxel (Taxol) and estramustine stabilize spindle microtubules through microtubule polymerization and block mitosis (Jordan et al., 1993). There are differences in the sensitivity of microtubules composed of class III β-tubulin and those assembled from unfractionated tubulin to these drugs (Lu and Luduena, 1993). Class III β-tubulin transcript expression has been shown to alter in tumors showing resistance to antimicrotubule/microtubule-polymerizing drugs (Kavallaris
Embryonic day (E) 14 (n = 2) eyes were enucleated from embryos obtained by cesarean section after sacrificing the pregnant mice. Eyes were also obtained from postnatal day (PN) 0 (n = 3), 3 (n = 3), 7 (n = 3) and 14 (n = 3), and adult C57/black6 mice. The mice were anesthetized with CO2 prior to decapitation. Eyes were enucleated and fixed as described below. Adult rabbit eyes were also used (n = 3). Animals were treated according to the ARVO resolution on animal experimentation. Human eyes were obtained from Mid-South Eye Bank. The donor was a 57year-old adult male who had no history of any ocular disorder. All the eyes were briefly fixed in 4% paraformaldehyde in phosphate-buffered saline (PBS; 0.1 M phosphate, 0.85% NaCl; pH 7.4). The anterior segment of the eyes was removed and the posterior segment was fixed in the same fixative for 24 h. The tissue was rinsed in PBS and then transferred sequentially to PBS containing 5%, 10% and 20% sucrose. The eyes were stored in PBS with 20% sucrose until sectioning. Twelve-micron thick sections were cut on a cryostat and mounted on glass slides coated with gelatin. They were subsequently stored at − 70 °C until used. Sagittal sections were cut in the embryonic and postnatal retinae, and the sections passing close to the optic nerve were used for immunohistochemistry.
4.2.
Immunohistochemistry
Frozen sections were thawed at room temperature and then rinsed with PBS containing 1% bovine serum albumin (BSA; Sigma Chemical Co.). The sections were incubated with the primary antibodies in a humidified chamber overnight. The following primary antibodies were used; TUJ1 monoclonal antibody against class III β-tubulin (1: 500; CRP Inc. Berkeley CA) and calbindin (1: 500; Sigma®) labeling horizontal cells and certain amacrine cells. In addition, cone photoreceptors were also labeled with Peanut agglutinin (PNA) which binds to glycoconjugates associated with cone cell membranes (Blanks and Johnson, 1984). Optimum working concentration and incubation time for the antibodies was determined earlier in pilot experiments. After the incubation, the slides were rinsed with PBS and incubated for 1 h in appropriate secondary antibody conjugated with Cy3 (Jackson Laboratories PA), rinsed
16
BR A IN RE S EA RCH 1 1 76 ( 20 0 7 ) 1 1 –17
in PBS and mounted with Vectashield® (Vector Laboratories Burlingame CA). At least four slides from each specimen were stained, and in each experiment, controls were obtained by omitting the primary antibody. For double labeling, TUJ1 and calbindin antibodies (raised in animal other than mice) were mixed in dilutions previously determined by pilot experiments. Cy3-conjugated anti-mouse and Cy2-conjugated second sec-
ondary antibody were also mixed and used as secondary antibody. For peanut agglutinin (PNA) histochemistry, PNA lectin (FITC-conjugated peanut lectin 0.2 mg/mL Arachis hypogaea; Sigma, St. Louis, MO) was dissolved in the blocking solution at 1:100 dilution and added to the PF fixed retinal sections, incubated for 45 min, washed and mounted in Vectashield®. For double labeling experiments, PNA was
BR A IN RE S E A RCH 1 1 76 ( 20 0 7 ) 1 1 –1 7
added after the last wash in immunohistochemistry. Tissue was viewed on a Nikon photomicroscope fitted with a digital camera.
Acknowledgments The authors thank Professor Dianna A. Johnson for support and Marina A. Kedrov for technical help with the experiments. This work was supported by a generous grant by Gail and Richard Siegal, The Hyde Foundation, NEI grant EY-13080 and an unrestricted grant from Research to Prevent Blindness.
Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.brainres.2007.07.090. REFERENCES
Blanks, J.C., Johnson, L.V., 1984. Specific binding of peanut lectin to a class of retinal photoreceptor cells. A species comparison. Invest. Ophthalmol. Vis. Sci. 25, 546–557. Easter Jr., S.S., Ross, L.S., Frankfurter, A., 1993. Initial tract formation in the mouse brain. J. Neurosci. 13, 285–299. Fournier, A.E., McKerracher, L., 1997. Expression of specific tubulin isotypes increases during regeneration of injured CNS neurons, but not after the application of brain-derived neurotrophic factor (BDNF). J. Neurosci. 17, 4623–4632. Gonzalez-Garay, M.L., Chang, L., Blade, K., Menick, D.R., Cabral, F., 1999. A beta-tubulin leucine cluster involved in microtubule assembly and paclitaxel resistance. J. Biol. Chem. 274, 23875–23882. Hari, M., Yang, H., Zeng, C., Canizales, M., Cabral, F., 2003. Expression of class III beta-tubulin reduces microtubule assembly and confers resistance to paclitaxel. Cell Motil. Cytoskelet. 56, 45–56. Harman, A.M., Sanderson, K.J., Beazley, L.D., 1992. Biphasic retinal neurogenesis in the brush-tailed possum, Trichosurus vulpecula: further evidence for the mechanisms involved in formation of ganglion cell density gradients. J. Comp. Neurol. 325, 595–606.
17
Heidemann, S.R., 1996. Cytoplasmic mechanisms of axonal and dendritic growth in neurons. Int. Rev. Cytol. 165, 235–296. Hu, Y., Cui, Q., Harvey, A.R., 2007. Interactive effects of C3, cyclic AMP and ciliary neurotrophic factor on adult retinal ganglion cell survival and axonal regeneration. Mol. Cell Neurosci. 146 (3), 986–999. Jordan, M.A., Toso, R.J., Thrower, D., Wilson, L., 1993. Mechanism of mitotic block and inhibition of cell proliferation by taxol at low concentrations. Proc. Natl. Acad. Sci. U. S. A. 90, 9552–9556. Kavallaris, M., Kuo, D.Y., Burkhart, C.A., Regl, D.L., Norris, M.D., Haber, M., Horwitz, S.B., 1997. Taxol-resistant epithelial ovarian tumors are associated with altered expression of specific beta-tubulin isotypes. J. Clin. Invest. 100, 1282–1293. Lee, M.K., Tuttle, J.B., Rebhun, L.I., Cleveland, D.W., Frankfurter, A., 1990. The expression and posttranslational modification of a neuron-specific beta-tubulin isotype during chick embryogenesis. Cell Motil. Cytoskelet. 17, 118–132. Lu, Q., Luduena, R.F., 1993. Removal of beta III isotype enhances taxol induced microtubule assembly. Cell Struct. Funct. 18, 173–182. Moody, S.A., Quigg, M.S., Frankfurter, A., 1989. Development of the peripheral trigeminal system in the chick revealed by an isotype-specific anti-beta-tubulin monoclonal antibody. J. Comp. Neurol. 279, 567–580. Ranganathan, S., Dexter, D.W., Benetatos, C.A., Chapman, A.E., Tew, K.D., Hudes, G.R., 1996. Increase of beta(III)- and beta(IVa)-tubulin isotopes in human prostate carcinoma cells as a result of estramustine resistance. Cancer Res. 56, 2584–2589. Sharma, R.K., Ehinger, B., 1997. Mitosis in developing rabbit retina: an immunohistochemical study. Exp. Eye Res. 64, 97–106. Sharma, R.K., Ehinger, B., 2003. Development and structure of the retina. In: Kaufman, P.L., Alm, A. (Eds.), Adler's Physiology of the Eye: Clinical Application. Mosby, St. Louis, pp. 319–347. Sharma, R.K., O'Leary, T.E., Fields, C.M., Johnson, D.A., 2003. Development of the outer retina in the mouse. Brain Res. Dev. Brain Res. 145, 93–105. Sidman, R.L., 1961. Histiogenesis of mouse retina studied with thymidine-H3. In: Smelser, G.K. (Ed.), The Structure of the Eye. Academic Press Inc. Ltd, London, pp. 487–506. Sullivan, K.F., Cleveland, D.W., 1986. Identification of conserved isotype-defining variable region sequences for four vertebrate beta tubulin polypeptide classes. Proc. Natl. Acad. Sci. U. S. A. 83, 4327–4331.
Fig. 3 – Colocalization of class III β-tubulin with calbindin and PNA. (A and B) Immunoreactive bodies in RPE layer with variable staining intensities (vertical arrows in A and B). Some, but not all, immunoreactive bodies weakly colocalized with PNA. (C, D) Colocalization of TUJ1 (red) and calbindin (green) immunoreactivity in mouse (C) and rabbit (D) retinae. In both species, as expected calbindin immunoreactivity is seen in horizontal cells, including their projections to the OPL (small arrowheads). Immunoreactivity is also seen in a subpopulation of amacrine cells in the INL and displaced amacrine cells in the GCL (green and yellow labeling in GCL). Three immunoreactive plexuses are seen in the IPL (black arrowheads). TUJ1 immunoreactivity colocalized with some, but not all amacrine cells (green [thick vertical arrow] and red [small vertical arrow] labeling). In the GCL many cells that are weakly immunoreactive to TUJ1 colocalized with calbindin (yellow labeling in GCL). Perikarya of calbindin immunoreactive horizontal cells did not colocalize with TUJ1 (green labeling; C and D), but some processes weakly colocalized. (E, F) Show colocalization of TUJ1 (red) and PNA (green) in mouse (E) and rabbit (F) retinae. PNA labeling in cone photoreceptors colocalized with TUJ1 immunoreactivity in both rabbit and mouse retina including the inner segments (oblique arrows), and their terminals in the OPL. Horizontal arrows = cells in the ONL, oblique arrows = photoreceptor inner segments, small and large vertical arrows = morphologically distinct cells on either sides of IPL, small arrow heads = OPL and large arrowheads = plexuses in IPL, scale bar = 40 μm.