© 2014 APMIS. Published by John Wiley & Sons Ltd. DOI 10.1111/apm.12225
APMIS 122: 772–780
Altered aquaporin expression in glaucoma eyes THUY LINH TRAN,1 TOKE BEK,2 MORTEN LA COUR,3 SØREN NIELSEN,4 JAN ULRIK PRAUSE,1 STEFFEN HAMANN3 and STEFFEN HEEGAARD1,3 1
Eye Pathology Institute, Department of Neuroscience and Pharmacology, University of Copenhagen, Copenhagen; 2Department of Ophthalmology, Aarhus University Hospital, Aarhus; 3Department of Ophthalmology, Glostrup Hospital, University of Copenhagen, Copenhagen; and 4Department of Biomedicin, Aarhus University, Aarhus, Denmark
Tran TL, Bek T, la Cour M, Nielsen S, Prause JU, Hamann S, Heegaard S. Altered aquaporin expression in glaucoma eyes. APMIS 2014; 122: 772–780. Aquaporins (AQP) are channels in the cell membrane that mainly facilitate a passive transport of water. In the eye, AQPs are expressed in the ciliary body and retina and may contribute to the pathogenesis of glaucoma and optic neuropathy. We investigated the expression of AQP1, AQP3, AQP4, AQP5, AQP7 and AQP9 in human glaucoma eyes compared with normal eyes. Nine glaucoma eyes were examined. Of these, three eyes were diagnosed with primary open angle glaucoma; three eyes had neovascular glaucoma; and three eyes had chronic angle-closure glaucoma. Six eyes with normal intraocular pressure and without glaucoma were used as control. Immunohistochemistry was performed using antibodies against AQP1, AQP3, AQP4, AQP5, AQP7 and AQP9. For each specimen, optical densities of immunoprecipitates were measured using Photoshop and the staining intensities were calculated. Immunostaining showed labelling of AQP7 and AQP9 in the nonpigmented ciliary epithelium and the staining intensities were significantly decreased in glaucoma eyes (p = 0.003; p = 0.018). AQP7 expression in the M€ uller cell endfeet was increased (p = 0.046), and AQP9 labelling of the retinal ganglion cells (RGC) showed decreased intensity (p = 0.037). No difference in AQP1, AQP4 and AQP9 expression was found in the optic nerve fibres. This study is the first investigating AQPs in human glaucoma eyes. We found a reduced expression of AQP9 in the retinal ganglion cells of glaucoma eyes. Glaucoma also induced increased AQP7 expression in the M€ uller cell endfeet. In the ciliary body of glaucoma eyes, the expression of AQP7 and AQP9 was reduced. Therefore, the expression of AQPs seems to play a role in glaucoma. Key words: Aquaporin; human eye; glaucoma; immunohistochemistry. Steffen Heegaard, Eye Pathology Institute, Department of Neuroscience and Pharmacology, Faculty of Health Sciences, University of Copenhagen, Frederik V’s vej 11, 1., 2100 Copenhagen, Denmark. e-mail: [email protected]
Glaucoma is a family of common eye disorders presenting with loss of retinal ganglion cells and axons causing optic neuropathy and loss of visual field (1). Glaucoma is progressive and may cause blindness if left untreated (1). In Denmark, one per cent of the population over the age of 45 years has glaucoma (2). The role of aquaporins (AQPs) in the development of glaucoma has not been established. However, AQPs may play a role in the pathogenesis. Aquaporins are expressed at numerous sites where they could contribute to glaucoma and optic neuropathy; AQP1 and AQP4 are expressed in the ciliary body and are important in the secretion of aqueous humour, while AQP4 is expressed in the retina where it plays a role in retinal homoeostasis Received 19 August 2013. Accepted 31 October 2013
772
(3). Moreover, the location of AQP7 and AQP9 in the human eyes indicates that these AQPs have metabolic functions (4). Aquaporins are selective channels in the cell membrane and mainly facilitate a passive transport of water (3). Studies with knockout mice show that AQPs are associated with the transport of water and smaller molecules, affecting physiological functions in several organs, e.g. urine concentration, exocrine gland secretion, brain oedema, neural signal transduction and metabolism, fat metabolism and skin hydration (5, 6). A subset of AQPs is the aquaglyceroporins that facilitate transport of NH4+, urea, glycerol, lactate and ketones in addition to water (7). Normal neuronal tissue mainly metabolizes glucose. However, lactate, pyruvate, glutamate and ketones can also be metabolized to generate energy (8). New findings suggest that
AQUAPORINS AND GLAUCOMA
AQPs are also involved in regulating proliferation, migration and differentiation of neuronal stem cells (9). AQP4 also modulates neural signal transduction and neuronal excitability by regulating osmotic and ionic gradients around the neurons (9). For that reason, it appears that the significance of AQPs in the retina and optic nerve is more comprehensive than fluid regulation, especially with regard to glaucoma and optic neuropathy. The purpose of this study was to investigate changes in AQP expression in human glaucoma eyes compared with normal control eyes.
with aminoethyl-carbazole. All sections were counterstained with Mayer’s modified haematoxylin. To ensure equal staining intensity, all slides of each antibody were stained simultaneously in one batch. Fluorescence techniques were used to visualize glial fibrillary acidic proteins (GFAP) in M€ uller cells showing gliosis in glaucoma eyes. The initial steps of immunolabelling were as described above. However, the sections were incubated for 1 hr with fluorescent secondary antibody instead (donkey anti-rabbit antibodies AlexaFluor 488 and donkey anti-rabbit AlexaFluor 546; Invitrogen, Taastrup, Denmark).
Primary antibodies
MATERIALS AND METHODS Material Nine paraffin-embedded glaucoma eyes and six eyes with normal intraocular pressure (10–18 mmHg) and without glaucoma were retrieved from the tissue bank of the Eye Pathology Institute, University of Copenhagen. The glaucoma eyes were all clinically diagnosed, and comprised three eyes with primary open angle glaucoma (POAG), three eyes with neovascular glaucoma (NVG) and three eyes with chronic angle-closure glaucoma (CACG). The eyes with POAG all had IOP 40 mmHg). All glaucoma eyes presented with loss of retinal ganglion cells (RGC) and atrophy of the nerve fibre layer. Cupping of the disc was also seen in all specimens. The eyes with POAG and NVG showed the greatest loss of RGCs and thinning of the nerve fibre layer. The six normal eyes served as controls. The collection of specimens of human eyes had been approved by the Danish Committee on Biomedical Research Ethics (H-2-2010-034) and the Danish Data Protection Agency (J.nr. 2010-41-4453).
Immunohistochemistry Immediately after removal, the eyes were fixed by immersion in 4% paraformaldehyde in 0.1 M sodium cacodylate buffer, pH 7.2, for 24–48 h, dehydrated in ethanol followed by xylene, and finally embedded in paraffin. Sections of 4 lm were cut by the same technician and staining of all slides of each antibody was carried out simultaneously in one batch using indirect immunoperoxidase. Sections were deparaffinized through graded alcohols and rinsed in distilled water. To reveal antigens, sections were submerged in 1 mM Tris solution (pH 9.0) supplemented with 0.5 mM EDTA and heated in a microwave oven at 97 °C for 20 min. Section were then left to cool for 30 min in the buffer, rinsed in phosphatebuffered saline (PBS; pH 7,4) and incubated with the primary antibody. The primary antibodies were diluted in 2% bovine serum for 1 h, rinsed in PBS, incubated with biotin-conjugated secondary antibody (Chemmate Detection Kit; Dako AS, Glostrup, Denmark), treated with 1% hydrogen peroxide to remove endogenous peroxidase, incubated with streptavidin peroxidase and developed
© 2014 APMIS. Published by John Wiley & Sons Ltd
The primary antibodies used were anti-AQP1 (10), antiAQP3 (11), anti-AQP4 (AQP-004, Alomone, Jerusalem, Israel), anti-AQP5 (12), anti-AQP7 (4) and anti-AQP9 (ab84828, Abcam, Cambridge, UK). Rabbit anti-glial fibrillary acidic protein was purchased from DAKO (Glostrup, Denmark) (Z0334). The specificity of the antibodies was verified by staining normal human kidney, salivary gland and liver. The reactions lacking the primary antibody produced no signal (negative controls).
Quantification of immunohistochemical staining All sections were examined with a light microscope, Axioplan 2 (Carl Zeiss, Jena, Germany). Digital images were obtained of representative fields (920 objective) with an Axiocam HRC (Carl Zeiss). Microscope settings were kept constant throughout image requisition. Digital images from each specimen were incorporated into a single image using Adobe Photoshop CS5.1. The image was normalized by setting the background as the reference point to zero. The ‘Quick Selection’ tool was used to select structures of positively stained cells and the optical density (OD) was generated using the ‘Histogram’ tool (0–255, where 0 = dark and 255 = white) (Fig. 1). The grey level was used as an arbitrary unit (arb. unit) of immunolabelling intensity and was calculated as the difference between the indicated value and the maximum ‘white’ level of 255 (13). The final value of optical density was an average of 10 independent areas of interest for each antibody. Background measurement was performed to evaluate the influence of nonspecific antibody binding.
Statistical analysis One-way Analysis of Variance (ANOVA) with Bonferroni adjustment was performed to identify differences in AQP staining intensity levels between the controls and the glaucoma subgroups, as the OD scores were normally distributed. All statistical values were judged significant if p < 0.05.
RESULTS Immunohistochemical findings in the anterior part of glaucoma eyes
In the cornea, conjunctiva, lens epithelium and trabecular meshwork of glaucoma eyes, the staining
773
TRAN et al.
A
intensity for all tested AQPs was equal to that of control eyes (Table 1). In the ciliary body, there was AQP1, AQP4, AQP7 and AQP9 labelling in the nonpigmented ciliary epithelium. The staining intensity of AQP7 and AQP9 in the nonpigmented ciliary epithelium was significantly decreased (p = 0.003; p = 0.018) (Table 1, Figs 2 and 3), whereas the AQP1 and AQP4 staining intensities were unchanged (p = 0.563; p = 0.979). Immunohistochemical findings in the retina and optic nerve of glaucoma eyes
B
C
Fig. 1. Using photoshop ‘Quick Selection’ tool positively stained structures were demarcated (green line). Optical density (OD) was measured within the demarcated area. (A) Demarcation of positive AQP7 immunolabelling (green line) in nonpigmented ciliary epithelium in the right side of the photomicrograph. (B) Demarcation of positive AQP7 labelling (green line) in M€ uller cells in the left part of the photomicrograph. (C) Demarcation of postitive AQP9 labelling (green line) in a retinal ganglion cell.
774
During retinal injury in glaucoma, the M€ uller cells respond with reactive gliosis, demonstrated with the GFAP labelling (Fig. 4). The staining intensity of AQP7 in the M€ uller cell processes was significantly increased compared with normal eyes (p = 0.046) (Table 1; Fig. 4). M€ uller cells of glaucoma eyes showed labelling of AQP7 in their processes stretching towards the outer retina into the outer plexiform layer, whereas in control eyes, AQP7-labelled M€ uller cell processes were limited to the RGC layer. AQP7 labelling in the outer limiting membrane of the retina and in the capillary endothelium showed no changes in staining intensity. In the retina, AQP4 labelling in M€ uller cells and astrocytes showed no changes in glaucoma eyes (p = 0.799). Immunolabelling of AQP9 was found in the RGC in both glaucoma and normal eyes (Fig. 5). The number of RGCs was considerably decreased and the staining intensity of each retinal ganglion cell was significantly decreased in glaucoma eyes (p = 0.037). In the optic nerve head (ONH), AQP9 labelling was observed in the astrocyte processes. There were no significant differences between glaucoma and control eyes (p = 0.448) (Table 1, Fig. 6). However, the amount of nerve fibres within the optic nerve was reduced. In the optic nerve cranially to the lamina cribrosa, AQP4 and AQP9 labelling was seen in the astrocytes and no significant changes in staining intensities were observed (Table 1). There were no differences in staining intensity between the glaucoma subgroups for any ocular structure (Table 1). Furthermore, there was no correlation between age and staining intensity for any ocular structure. DISCUSSION This is the first study investigating the changes in AQP expression in different types of glaucoma in human eyes. Using optical density measurements, we found significant changes in the expression of © 2014 APMIS. Published by John Wiley & Sons Ltd
AQUAPORINS AND GLAUCOMA
Table 1. Aquaporin (AQP) expression in various ocular structures. One-way Analysis of Variance (ANOVA) with Bonferroni adjustment was performed to identify differences in AQP staining intensity levels between the controls and the glaucoma subgroups. Staining intensity is given in arbitrary units converted from optical density measurement. Values are given as means SEM Ocular structure Normal Glaucoma p value (n = 6) (n = 9) AQP1 Corneal keratocytes 204.1 3.3 203.8 2.9 0.868 Corneal endothelium 200.9 3.0 204.0 2.0 0.832 Trabecular meshwork 199.0 1.3 198.6 1.6 0.375 Ciliary nonpigmented epithelium 193.2 1.2 186.3 4.9 0.563 Lens epithelium 201.3 1.1 204.4 1.9 0.158 AQP3 Conjunctival/corneal epithelium 103.9 3.6 100.5 2.4 0.900 AQP4 Ciliary nonpigmented epithelium 85.0 5.8 87.2 2.7 0.979 Muller cell/astrocytes 157.0 3.0 150.2 3.5 0.799 Optic nerve 112.8 11.3 100.2 7.2 0.328 AQP5 Conjunctival/corneal epithelium 95.3 3.2 91.7 4.8 0.580 AQP7 Corneal epithelium 100.9 11.2 93.8 3.1 0.729 Corneal endothelium 154.8 2.3 162.6 6.0 0.761 Trabecular meshwork 124.9 4.0 119.6 1.4 0.821 Ciliary nonpigmented epithelium 141.37 3.5 107.0 2.7 0.003* Lens epithelium 153.0 7.5 164.1 5.9 0.189 Muller cell endfeet 120.4 2.8 148.0 2.6 0.046* Capillary endothelium 116.3 6.5 113.3 1.9 0.722 AQP9 Ciliary nonpigmented epithelium 148.7 5.4 117.4 3.4 0.018* Retinal ganglion cells 157.8 4.8 124.0 6.6 0.037* Optic nerve head 106.9 12.1 119.0 7.0 0.448 Optic nerve 90.8 2.8 84.9 12.5 0.673 *p < 0.05 was considered statistically significant.
AQP7 and AQP9 in the ciliary body and retina. AQP9 was significantly decreased in the retinal ganglion cells. However, we found no changes in AQP9 expression in the nerve fibres of the optic nerve head as had been reported previously (14, 15). We also found no changes in the expression of AQP1, AQP3, AQP4 and AQP5. We performed measurements of optical density (OD) to standardize the method for quantifying the immunostaining. Today, human glaucoma eyes are not enucleated and quantitative analyses like real-time PCR and Western blot require fresh human tissue. Therefore, quantitative methods studying human glaucoma eyes are unfortunately often limited to immunohistochemistry applied on formalin-fixed paraffin-embedded eyes. However, immunohistochemistry is associated with some variability, as both the technical processing and the intraobserver variability highly impact the interpretation of the immunolabelling. In this study, all sections of each antibody were stained simultaneously in one bath to ensure that procedure parameters concerning ambient environment (tp. and humidity), buffers and dilution of antibodies were kept constant. Also, according to routine, the eyes were immerged and fixed in paraformaldehyde, supplied by the Eye Pathology Institute, immediately after removal. Further preparation of all eyes was handled by the Eye Pathology Institute and all paraffin-embedded specimens had been stored in © 2014 APMIS. Published by John Wiley & Sons Ltd
the same room. All specimens were cut by the same technician at 4 lm in thickness. To uniformly evaluate the immunolabelling, the initial step of analysis involved a fusion of all photomicrographs of each eye and normalizing the background to zero. An arbitrary value of staining intensity of bound antibody was then obtained by measuring the optical density. The linear relationship between the intensity of immunoperoxidase staining and the amount of antigen was the basis for the use of OD measurements as an estimate of staining variation (16, 17). Therefore, variations in staining intensities were quantifiable by measuring OD and statistical comparisons between different samples were applicable (16, 18). Using OD measurements of the ciliary body, we found a significant decrease in AQP7- and AQP9 expression. Optical density for AQP7 and AQP9 was only measured in the nonpigmented ciliary epithelium, as the pigment granules in the pigmented epithelium interfered with OD measurement. Interestingly, no decrease was found in AQP1- and AQP4 expression. AQP1 is a highly permeable water channel and conducts the main part of water from the nonpigmented epithelium into the aqueous humour (1). Constitutive AQP1- and AQP4 expression is consistent with an unaffected production of aqueous humour in glaucoma eyes, regardless of IOP (1). AQP7 and AQP9 are aquaglyceroporins and have the ability to transport both water and
775
TRAN et al.
A
A
B
B
C C 150 Staining intensity/arb. unit
Staining intensity/arb. unit
150
100
50
100
50
0 Control
POAG
NVG
CACG
Fig. 2. Aquaporins 7 immunolabelling in the anterior segment of normal and glaucoma eyes. (A) AQP7 expression in the ciliary epithelia of normal eyes (brown staining). In the immediate subepithelial stroma, there is a faint stain, which is also seen in (B). Bar = 50 lm. (B) In glaucoma eyes, AQP7 expression is decreased. The subepithelial faintly positive stroma is of the same intensity as in (A). Bar = 50 lm. (C) Staining intensity of AQP7 in the nonpigmented ciliary epithelium in normal eyes (control) and in glaucoma subgroups. Staining intensity denotes the optical density of positively labelled cells. The staining intensity of AQP7 is significantly decreased in the glaucoma groups compared with the control group (p = 0.003). There are no statistical differences among the three glaucomatous subgroups (p = 0.107). Arb. unit: arbitrary unit; POAG: primary open angle glaucoma; NVG: neovascular glaucoma; CACG: chronic angle-closure glaucoma.
776
0 Control
POAG
NVG
CACG
Fig. 3. AQP9 immunolabelling in the anterior segment of normal and glaucoma eyes. (A) In normal eyes, AQP9 is expressed in the ciliary epithelia (brown staining; bar = 50 lm). (B) In glaucoma eyes, AQP9 expression is decreased. Bar = 50 lm. (C) Staining intensity of AQP9 in the nonpigmented ciliary epithelium in control and in glaucoma subgroups. Staining intensity denotes the optical density of positively labelled cells. The staining intensity of AQP9 is significantly decreased in the glaucoma groups compared with the control group (p = 0.018). There are no statistical differences among the three glaucomatous subgroups (p = 0.253). Arb.unit: arbitrary units; POAG: primary open angle glaucoma; NVG: neovascular glaucoma; CACG: chronic angle-closure glaucoma. © 2014 APMIS. Published by John Wiley & Sons Ltd
AQUAPORINS AND GLAUCOMA
other small noncharged molecules such as glycerol, lactate and urea that can be used as substrates in energy production (7). Therefore, decrease in both AQP7- and AQP9 expression in the nonpigmented ciliary epithelium may be coupled with changes in the metabolism of the ciliary body and anterior chamber, which could aggravate the glaucomatous retinal damage. M€ uller cells in a glaucomatous retina display a changed expression of membrane proteins with functional consequence for the overall retinal homoeostasis and RGC survival (19). The increased AQP7 expression as seen in this study may be an A
B
C
D
E
Staining intensity/arb. unit
150
100
50
0 © 2014 APMIS. Published by John Wiley & Sons Ltd
attempt to stimulate glutamate uptake by alkalinization of the intracellular space, as AQP7 may facilitate the NH3-transport needed in the glutamate/glutamine cycle (8, 20). Downregulation of AQP4 expression has been observed in animal studies of optic nerve crush and autoimmune uveitis, where the M€ uller cells respond with reparative activation (21, 22). The unchanged AQP4 expression and increased AQP7 expression may be linked to a different M€ uller cell response characteristic of glaucoma eyes (23). In the retina, we found a significantly decreased AQP9 expression in retinal ganglion cells. AQP9 has been suggested to provide neurons in the brain with lactate and glycerol for energy metabolism (8, 24). AQP9 may have a similar function in retinal ganglion cells and facilitate the uptake of lactate and/or glycerol into the retinal ganglion cells and photoreceptors (8, 14, 25). The decrease in AQP9 expression may result in compromised metabolism of the RGCs due to deprivation of nutrients. Consequently, the RGCs may become fragile and susceptible to apoptosis triggered by new glaucomatous insults such as ischaemia, excitotoxicity, further neurotrophic and nutrient deprivation and oxidative stress (26, 27). Interestingly, the three groups of glaucoma eyes
Fig. 4. Aquaporins 7 immunolabelling in glial cells of the normal and glaucomatous retina. (A) In a normal retina, immunofluorescence labelling of AQP7 (green) is seen in the M€ uller cell endfeet (arrows) at the inner limiting membrane and at the outer limiting membrane (double arrow). Glial fibrillary acidic protein (GFAP) labelling (red) is seen in astrocyte- and M€ uller cell processes (arrowhead). Bar = 50 lm. (B) In the same normal retina, immunoperoxidase labelling of AQP7 is seen in the M€ uller cell endfeet (arrows) and at the outer limiting membrane in a normal retina (double arrow). Bar = 50 lm. (C) Immunofluorescense staining of glaucomatous retinas demonstrate an upregulated AQP7 expression (green) in the M€ uller cells (arrows) and AQP7 labelling is also seen in M€ uller cell processes stretching towards the outer retina into the outer plexiform layer. AQP7 labelling of the OLM (double arrow) is not increased. GFAP (red) is upregulated in the M€ uller cells and is expressed in the processes throughout the retina (arrowheads). Bar = 50 lm. (D) In the same glaucomatous retina, immunoperoxidase staining increased AQP7 expression in M€ uller cells (arrows) and OLM (double arrow) is evident. Bar = 50 lm. (E) Based on immunoperoxidase stained sections, AQP7 staining intensity was measured in the M€ uller cells endfeet in control and in glaucoma subgroups. Staining intensity denotes the optical density of positively labelled cells. The staining intensity of AQP7 is significantly increased in the glaucoma groups compared with the control group (p = 0.046). There were no statistical differences among the three glaucomatous subgroups (p = 0.731). Arb. unit, arbitrary unit; POAG, primary open angle glaucoma; NVG, neovascular glaucoma; CACG, chronic angle-closure glaucoma.
777
TRAN et al.
A
A
B
B
C
C 150
Staining intensity/arb. unit
Staining intensity/arb. unit
150
100
50
0 Control
POAG
NVG
CACG
Fig. 5. Aquaporins 9 immunolabelling in retinal ganglion cells of the normal and glaucomatous retina. (A) AQP9 expression is seen in the retinal ganglion cells in normal eyes (arrow). Bar = 50 lm. (B) In glaucoma eyes, AQP9 expression is significantly decreased in the retinal ganglion cells (arrow). Bar = 50 lm. (C) Staining intensity of AQP9 in the retinal ganglion cells in control and in glaucoma subgroups. Staining intensity denotes the optical density of positively labelled cells. The staining intensity of AQP9 is significantly decreased in the glaucoma groups compared with the control group (p = 0.037). There are no statistical differences among the three glaucomatous subgroups (p = 0.9812). Arb. unit, arbitrary unit; POAG, primary open angle glaucoma; NVG, neovascular glaucoma; CACG, chronic angle-closure glaucoma.
778
100
50
0 Control
POAG
NVG
CACG
Fig. 6. Aquaporins 9 immunolabelling in the optic nerve fibres of the normal and glaucomatous eye. (A) AQP9 expression is seen in and along the optic nerve fibres of normal eyes. Bar = 500 lm. (B) In glaucoma eyes, AQP9 expression in the optic nerve fibres has not changed; however, the amount of optic nerve fibres is markedly reduced. Bar = 500 lm. (C) Staining intensity of AQP9 in the optic nerve head in control and in glaucoma subgroups. Staining intensity denotes the optical density of positively labelled cells. The staining intensity of AQP9 is unchanged in the glaucoma groups compared with the control group (p = 0.448). Arb. unit: arbitrary unit; POAG: primary open angle glaucoma; NVG: neovascular glaucoma; CACG: chronic angle-closure glaucoma. © 2014 APMIS. Published by John Wiley & Sons Ltd
AQUAPORINS AND GLAUCOMA
showed the same staining intensities. Therefore, the decreased AQP9 expression may be a common feature of the RGCs when an initial glaucomatous insult occurs. In contrast, the AQP9 expression was unchanged in the nerve fibres of the optic nerve head in all the glaucoma eyes compared with the control group. Moreover, there were no differences in AQP9 expression among the three glaucoma groups regardless of cause. The unchanged AQP9 expression in the ONH suggests that the glaucomatous degeneration in the ONH is a secondary consequence of the initial RGC death. Studies using animal models of artificially elevated IOP showed a downregulation of AQP9 in the optic nerve head (14, 15). The decrease in AQP9 expression may be a transient effect, as the overall period with elevated IOP was considerably shorter than the lapse of glaucoma of the eyes in this study (14, 15). However, the results in both studies should be interpreted with caution as the sample sizes were relatively small. This study is the first investigating AQPs in human glaucoma eyes. The changed expression of AQP7 and AQP9 in the ciliary body may be associated with metabolic changes in glaucoma eyes. Furthermore, downregulation of AQP9 in RGCs indicates AQP9 as a relevant factor in RGC metabolism.
This study was supported by grants from the Danish Eye Research Foundation, Fight for Sight – Denmark, the Synoptik Foundation, Aase and Ejnar Danielsen Foundation and Civilingeniør Lars Andersen Legat.
REFERENCES 1. Wax M, Clark A, Civan MM. Glaucoma. In: Yanof MDJ, editor. Ophthalmology. St. Louis: Mosby, 2009: 1108–16. 2. Lundberg LU, Thygesen J, Damgaard-Jensen L, Serup L, Kessing SV. Glaukompatienter i behandling hos praktiserende ojenlaeger i Danmark. Estimat af antal patienter samt omfang af synsfeltsdefekter [Glaucoma patients treated by practicing ophthalmologists in Denmark. Estimated number of patients and the extent of visual field defects]. Ugeskr Laeger. 2000;162:3028–33. 3. Verkman AS, Ruiz-Ederra J, Levin MH. Functions of aquaporins in the eye. Prog Retin Eye Res 2008;27:420–33. 4. Tran TL, Bek T, Holm L, la COUR M, Nielsen S, Prause JU, et al. Aquaporins 6–12 in the human eye. Acta Ophthalmol. 2013;91:557–63. 5. Rojek A, Praetorius J, Frokiaer J, Nielsen S, Fenton RA. A current view of the mammalian aquaglyceroporins. Annu Rev Physiol. 2008;70:301–27.
© 2014 APMIS. Published by John Wiley & Sons Ltd
6. Hara-Chikuma M, Verkman AS. Aquaporin-3 facilitates epidermal cell migration and proliferation during wound healing. J Mol Med (Berl). 2008;86:221–31. 7. Tsukaguchi H, Weremowicz S, Morton CC, Hediger MA. Functional and molecular characterization of the human neutral solute channel aquaporin-9. Am J Physiol 1999;277:F685–96. 8. Bringmann A, Pannicke T, Grosche J, Francke M, Wiedemann P, Skatchkov SN, et al. Muller cells in the healthy and diseased retina. Prog Retin Eye Res 2006;25:397–424. 9. Kong H, Fan Y, Xie J, Ding J, Sha L, Shi X, et al. AQP4 knockout impairs proliferation, migration and neuronal differentiation of adult neural stem cells. J Cell Sci. 2008;121:4029–36. 10. Nielsen S, Smith BL, Christensen EI, Knepper MA, Agre P. CHIP28 water channels are localized in constitutively water-permeable segments of the nephron. J Cell Biol. 1993;120:371–83. 11. Kwon TH, Frokiaer J, Fernandez-Llama P, Knepper MA, Nielsen S. Reduced abundance of aquaporins in rats with bilateral ischemia-induced acute renal failure: prevention by alpha-MSH. Am J Physiol. 1999;277: F413–27. 12. Nejsum LN, Kwon TH, Jensen UB, Fumagalli O, Frokiaer J, Krane CM, et al. Functional requirement of aquaporin-5 in plasma membranes of sweat glands. Proc Natl Acad Sci U S A. 2002;99:511–6. 13. Nakamura H, Riley F, Sakai H, Rademaker W, Yue BY, Edward DP. Histopathological and immunohistochemical studies of lenticules after epikeratoplasty for keratoconus. Br J Ophthalmol 2005;89:841–6. 14. Naka M, Kanamori A, Negi A, Nakamura M. Reduced expression of aquaporin-9 in rat optic nerve head and retina following elevated intraocular pressure. Invest Ophthalmol Vis Sci. 2010;51:4618– 26. 15. Mizokami J, Kanamori A, Negi A, Nakamura M. A preliminary study of reduced expression of aquaporin9 in the optic nerve of primate and human eyes with glaucoma. Curr Eye Res 2011;36:1064–7. 16. Nabors LB, Songu-Mize E, Mize RR. Quantitative immunocytochemistry using an image analyzer. II. Concentration standards for transmitter immunocytochemistry. J Neurosci Methods. 1988;26:25–34. 17. Taylor CR, Levenson RM. Quantification of immunohistochemistry–issues concerning methods, utility and semiquantitative assessment II. Histopathology 2006;49:411–24. 18. Lehr HA, Jacobs TW, Yaziji H, Schnitt SJ, Gown AM. Quantitative evaluation of HER-2/neu status in breast cancer by fluorescence in situ hybridization and by immunohistochemistry with image analysis. Am J Clin Pathol 2001;115:814–22. 19. Nork TM, Ghobrial MW, Peyman GA, Tso MO. Massive retinal gliosis. A reactive proliferation of Muller cells. Arch Ophthalmol. 1986;104:1383–9. 20. Bringmann A, Pannicke T, Biedermann B, Francke M, Iandiev I, Grosche J, et al. Role of retinal glial cells in neurotransmitter uptake and metabolism. Neurochem Int 2009;54:143–60. 21. Dibas A, Oku H, Fukuhara M, Kurimoto T, Ikeda T, Patil RV, et al. Changes in ocular aquaporin expression following optic nerve crush. Mol Vis. 2010;16:330–40.
779
TRAN et al.
22. Motulsky E, Koch P, Janssens S, Lienart M, Vanbellinghen AM, Bolaky N, et al. Aquaporin expression in blood-retinal barrier cells during experimental autoimmune uveitis. Mol Vis. 2010;16:602–10. 23. Vidal L, Diaz F, Villena A, Moreno M, Campos JG, Perez de Vargas I. Reaction of Muller cells in an experimental rat model of increased intraocular pressure following timolol, latanoprost and brimonidine. Brain Res Bull 2010;82:18–24. 24. Mylonakou MN, Petersen PH, Rinvik E, Rojek A, Valdimarsdottir E, Zelenin S, et al. Analysis of mice with targeted deletion of AQP9 gene provides conclu-
780
sive evidence for expression of AQP9 in neurons. J Neurosci Res 2009;87:1310–22. 25. Badaut J, Brunet JF, Regli L. Aquaporins in the brain: from aqueduct to ‘multi-duct’. Metab Brain Dis 2007;22:251–63. 26. Miki A, Kanamori A, Negi A, Naka M, Nakamura M. Loss of Aquaporin 9 Expression Adversely Affects the Survival of Retinal Ganglion Cells. Am J Pathol. 2013;182:1727–39. 27. Almasieh M, Wilson AM, Morquette B, Cueva Vargas JL, Di Polo A. The molecular basis of retinal ganglion cell death in glaucoma. Prog Retin Eye Res. 2012;31:152–81.
© 2014 APMIS. Published by John Wiley & Sons Ltd
APMIS 122: 772–780
Altered aquaporin expression in glaucoma eyes THUY LINH TRAN,1 TOKE BEK,2 MORTEN LA COUR,3 SØREN NIELSEN,4 JAN ULRIK PRAUSE,1 STEFFEN HAMANN3 and STEFFEN HEEGAARD1,3 1
Eye Pathology Institute, Department of Neuroscience and Pharmacology, University of Copenhagen, Copenhagen; 2Department of Ophthalmology, Aarhus University Hospital, Aarhus; 3Department of Ophthalmology, Glostrup Hospital, University of Copenhagen, Copenhagen; and 4Department of Biomedicin, Aarhus University, Aarhus, Denmark
Tran TL, Bek T, la Cour M, Nielsen S, Prause JU, Hamann S, Heegaard S. Altered aquaporin expression in glaucoma eyes. APMIS 2014; 122: 772–780. Aquaporins (AQP) are channels in the cell membrane that mainly facilitate a passive transport of water. In the eye, AQPs are expressed in the ciliary body and retina and may contribute to the pathogenesis of glaucoma and optic neuropathy. We investigated the expression of AQP1, AQP3, AQP4, AQP5, AQP7 and AQP9 in human glaucoma eyes compared with normal eyes. Nine glaucoma eyes were examined. Of these, three eyes were diagnosed with primary open angle glaucoma; three eyes had neovascular glaucoma; and three eyes had chronic angle-closure glaucoma. Six eyes with normal intraocular pressure and without glaucoma were used as control. Immunohistochemistry was performed using antibodies against AQP1, AQP3, AQP4, AQP5, AQP7 and AQP9. For each specimen, optical densities of immunoprecipitates were measured using Photoshop and the staining intensities were calculated. Immunostaining showed labelling of AQP7 and AQP9 in the nonpigmented ciliary epithelium and the staining intensities were significantly decreased in glaucoma eyes (p = 0.003; p = 0.018). AQP7 expression in the M€ uller cell endfeet was increased (p = 0.046), and AQP9 labelling of the retinal ganglion cells (RGC) showed decreased intensity (p = 0.037). No difference in AQP1, AQP4 and AQP9 expression was found in the optic nerve fibres. This study is the first investigating AQPs in human glaucoma eyes. We found a reduced expression of AQP9 in the retinal ganglion cells of glaucoma eyes. Glaucoma also induced increased AQP7 expression in the M€ uller cell endfeet. In the ciliary body of glaucoma eyes, the expression of AQP7 and AQP9 was reduced. Therefore, the expression of AQPs seems to play a role in glaucoma. Key words: Aquaporin; human eye; glaucoma; immunohistochemistry. Steffen Heegaard, Eye Pathology Institute, Department of Neuroscience and Pharmacology, Faculty of Health Sciences, University of Copenhagen, Frederik V’s vej 11, 1., 2100 Copenhagen, Denmark. e-mail: [email protected]
Glaucoma is a family of common eye disorders presenting with loss of retinal ganglion cells and axons causing optic neuropathy and loss of visual field (1). Glaucoma is progressive and may cause blindness if left untreated (1). In Denmark, one per cent of the population over the age of 45 years has glaucoma (2). The role of aquaporins (AQPs) in the development of glaucoma has not been established. However, AQPs may play a role in the pathogenesis. Aquaporins are expressed at numerous sites where they could contribute to glaucoma and optic neuropathy; AQP1 and AQP4 are expressed in the ciliary body and are important in the secretion of aqueous humour, while AQP4 is expressed in the retina where it plays a role in retinal homoeostasis Received 19 August 2013. Accepted 31 October 2013
772
(3). Moreover, the location of AQP7 and AQP9 in the human eyes indicates that these AQPs have metabolic functions (4). Aquaporins are selective channels in the cell membrane and mainly facilitate a passive transport of water (3). Studies with knockout mice show that AQPs are associated with the transport of water and smaller molecules, affecting physiological functions in several organs, e.g. urine concentration, exocrine gland secretion, brain oedema, neural signal transduction and metabolism, fat metabolism and skin hydration (5, 6). A subset of AQPs is the aquaglyceroporins that facilitate transport of NH4+, urea, glycerol, lactate and ketones in addition to water (7). Normal neuronal tissue mainly metabolizes glucose. However, lactate, pyruvate, glutamate and ketones can also be metabolized to generate energy (8). New findings suggest that
AQUAPORINS AND GLAUCOMA
AQPs are also involved in regulating proliferation, migration and differentiation of neuronal stem cells (9). AQP4 also modulates neural signal transduction and neuronal excitability by regulating osmotic and ionic gradients around the neurons (9). For that reason, it appears that the significance of AQPs in the retina and optic nerve is more comprehensive than fluid regulation, especially with regard to glaucoma and optic neuropathy. The purpose of this study was to investigate changes in AQP expression in human glaucoma eyes compared with normal control eyes.
with aminoethyl-carbazole. All sections were counterstained with Mayer’s modified haematoxylin. To ensure equal staining intensity, all slides of each antibody were stained simultaneously in one batch. Fluorescence techniques were used to visualize glial fibrillary acidic proteins (GFAP) in M€ uller cells showing gliosis in glaucoma eyes. The initial steps of immunolabelling were as described above. However, the sections were incubated for 1 hr with fluorescent secondary antibody instead (donkey anti-rabbit antibodies AlexaFluor 488 and donkey anti-rabbit AlexaFluor 546; Invitrogen, Taastrup, Denmark).
Primary antibodies
MATERIALS AND METHODS Material Nine paraffin-embedded glaucoma eyes and six eyes with normal intraocular pressure (10–18 mmHg) and without glaucoma were retrieved from the tissue bank of the Eye Pathology Institute, University of Copenhagen. The glaucoma eyes were all clinically diagnosed, and comprised three eyes with primary open angle glaucoma (POAG), three eyes with neovascular glaucoma (NVG) and three eyes with chronic angle-closure glaucoma (CACG). The eyes with POAG all had IOP 40 mmHg). All glaucoma eyes presented with loss of retinal ganglion cells (RGC) and atrophy of the nerve fibre layer. Cupping of the disc was also seen in all specimens. The eyes with POAG and NVG showed the greatest loss of RGCs and thinning of the nerve fibre layer. The six normal eyes served as controls. The collection of specimens of human eyes had been approved by the Danish Committee on Biomedical Research Ethics (H-2-2010-034) and the Danish Data Protection Agency (J.nr. 2010-41-4453).
Immunohistochemistry Immediately after removal, the eyes were fixed by immersion in 4% paraformaldehyde in 0.1 M sodium cacodylate buffer, pH 7.2, for 24–48 h, dehydrated in ethanol followed by xylene, and finally embedded in paraffin. Sections of 4 lm were cut by the same technician and staining of all slides of each antibody was carried out simultaneously in one batch using indirect immunoperoxidase. Sections were deparaffinized through graded alcohols and rinsed in distilled water. To reveal antigens, sections were submerged in 1 mM Tris solution (pH 9.0) supplemented with 0.5 mM EDTA and heated in a microwave oven at 97 °C for 20 min. Section were then left to cool for 30 min in the buffer, rinsed in phosphatebuffered saline (PBS; pH 7,4) and incubated with the primary antibody. The primary antibodies were diluted in 2% bovine serum for 1 h, rinsed in PBS, incubated with biotin-conjugated secondary antibody (Chemmate Detection Kit; Dako AS, Glostrup, Denmark), treated with 1% hydrogen peroxide to remove endogenous peroxidase, incubated with streptavidin peroxidase and developed
© 2014 APMIS. Published by John Wiley & Sons Ltd
The primary antibodies used were anti-AQP1 (10), antiAQP3 (11), anti-AQP4 (AQP-004, Alomone, Jerusalem, Israel), anti-AQP5 (12), anti-AQP7 (4) and anti-AQP9 (ab84828, Abcam, Cambridge, UK). Rabbit anti-glial fibrillary acidic protein was purchased from DAKO (Glostrup, Denmark) (Z0334). The specificity of the antibodies was verified by staining normal human kidney, salivary gland and liver. The reactions lacking the primary antibody produced no signal (negative controls).
Quantification of immunohistochemical staining All sections were examined with a light microscope, Axioplan 2 (Carl Zeiss, Jena, Germany). Digital images were obtained of representative fields (920 objective) with an Axiocam HRC (Carl Zeiss). Microscope settings were kept constant throughout image requisition. Digital images from each specimen were incorporated into a single image using Adobe Photoshop CS5.1. The image was normalized by setting the background as the reference point to zero. The ‘Quick Selection’ tool was used to select structures of positively stained cells and the optical density (OD) was generated using the ‘Histogram’ tool (0–255, where 0 = dark and 255 = white) (Fig. 1). The grey level was used as an arbitrary unit (arb. unit) of immunolabelling intensity and was calculated as the difference between the indicated value and the maximum ‘white’ level of 255 (13). The final value of optical density was an average of 10 independent areas of interest for each antibody. Background measurement was performed to evaluate the influence of nonspecific antibody binding.
Statistical analysis One-way Analysis of Variance (ANOVA) with Bonferroni adjustment was performed to identify differences in AQP staining intensity levels between the controls and the glaucoma subgroups, as the OD scores were normally distributed. All statistical values were judged significant if p < 0.05.
RESULTS Immunohistochemical findings in the anterior part of glaucoma eyes
In the cornea, conjunctiva, lens epithelium and trabecular meshwork of glaucoma eyes, the staining
773
TRAN et al.
A
intensity for all tested AQPs was equal to that of control eyes (Table 1). In the ciliary body, there was AQP1, AQP4, AQP7 and AQP9 labelling in the nonpigmented ciliary epithelium. The staining intensity of AQP7 and AQP9 in the nonpigmented ciliary epithelium was significantly decreased (p = 0.003; p = 0.018) (Table 1, Figs 2 and 3), whereas the AQP1 and AQP4 staining intensities were unchanged (p = 0.563; p = 0.979). Immunohistochemical findings in the retina and optic nerve of glaucoma eyes
B
C
Fig. 1. Using photoshop ‘Quick Selection’ tool positively stained structures were demarcated (green line). Optical density (OD) was measured within the demarcated area. (A) Demarcation of positive AQP7 immunolabelling (green line) in nonpigmented ciliary epithelium in the right side of the photomicrograph. (B) Demarcation of positive AQP7 labelling (green line) in M€ uller cells in the left part of the photomicrograph. (C) Demarcation of postitive AQP9 labelling (green line) in a retinal ganglion cell.
774
During retinal injury in glaucoma, the M€ uller cells respond with reactive gliosis, demonstrated with the GFAP labelling (Fig. 4). The staining intensity of AQP7 in the M€ uller cell processes was significantly increased compared with normal eyes (p = 0.046) (Table 1; Fig. 4). M€ uller cells of glaucoma eyes showed labelling of AQP7 in their processes stretching towards the outer retina into the outer plexiform layer, whereas in control eyes, AQP7-labelled M€ uller cell processes were limited to the RGC layer. AQP7 labelling in the outer limiting membrane of the retina and in the capillary endothelium showed no changes in staining intensity. In the retina, AQP4 labelling in M€ uller cells and astrocytes showed no changes in glaucoma eyes (p = 0.799). Immunolabelling of AQP9 was found in the RGC in both glaucoma and normal eyes (Fig. 5). The number of RGCs was considerably decreased and the staining intensity of each retinal ganglion cell was significantly decreased in glaucoma eyes (p = 0.037). In the optic nerve head (ONH), AQP9 labelling was observed in the astrocyte processes. There were no significant differences between glaucoma and control eyes (p = 0.448) (Table 1, Fig. 6). However, the amount of nerve fibres within the optic nerve was reduced. In the optic nerve cranially to the lamina cribrosa, AQP4 and AQP9 labelling was seen in the astrocytes and no significant changes in staining intensities were observed (Table 1). There were no differences in staining intensity between the glaucoma subgroups for any ocular structure (Table 1). Furthermore, there was no correlation between age and staining intensity for any ocular structure. DISCUSSION This is the first study investigating the changes in AQP expression in different types of glaucoma in human eyes. Using optical density measurements, we found significant changes in the expression of © 2014 APMIS. Published by John Wiley & Sons Ltd
AQUAPORINS AND GLAUCOMA
Table 1. Aquaporin (AQP) expression in various ocular structures. One-way Analysis of Variance (ANOVA) with Bonferroni adjustment was performed to identify differences in AQP staining intensity levels between the controls and the glaucoma subgroups. Staining intensity is given in arbitrary units converted from optical density measurement. Values are given as means SEM Ocular structure Normal Glaucoma p value (n = 6) (n = 9) AQP1 Corneal keratocytes 204.1 3.3 203.8 2.9 0.868 Corneal endothelium 200.9 3.0 204.0 2.0 0.832 Trabecular meshwork 199.0 1.3 198.6 1.6 0.375 Ciliary nonpigmented epithelium 193.2 1.2 186.3 4.9 0.563 Lens epithelium 201.3 1.1 204.4 1.9 0.158 AQP3 Conjunctival/corneal epithelium 103.9 3.6 100.5 2.4 0.900 AQP4 Ciliary nonpigmented epithelium 85.0 5.8 87.2 2.7 0.979 Muller cell/astrocytes 157.0 3.0 150.2 3.5 0.799 Optic nerve 112.8 11.3 100.2 7.2 0.328 AQP5 Conjunctival/corneal epithelium 95.3 3.2 91.7 4.8 0.580 AQP7 Corneal epithelium 100.9 11.2 93.8 3.1 0.729 Corneal endothelium 154.8 2.3 162.6 6.0 0.761 Trabecular meshwork 124.9 4.0 119.6 1.4 0.821 Ciliary nonpigmented epithelium 141.37 3.5 107.0 2.7 0.003* Lens epithelium 153.0 7.5 164.1 5.9 0.189 Muller cell endfeet 120.4 2.8 148.0 2.6 0.046* Capillary endothelium 116.3 6.5 113.3 1.9 0.722 AQP9 Ciliary nonpigmented epithelium 148.7 5.4 117.4 3.4 0.018* Retinal ganglion cells 157.8 4.8 124.0 6.6 0.037* Optic nerve head 106.9 12.1 119.0 7.0 0.448 Optic nerve 90.8 2.8 84.9 12.5 0.673 *p < 0.05 was considered statistically significant.
AQP7 and AQP9 in the ciliary body and retina. AQP9 was significantly decreased in the retinal ganglion cells. However, we found no changes in AQP9 expression in the nerve fibres of the optic nerve head as had been reported previously (14, 15). We also found no changes in the expression of AQP1, AQP3, AQP4 and AQP5. We performed measurements of optical density (OD) to standardize the method for quantifying the immunostaining. Today, human glaucoma eyes are not enucleated and quantitative analyses like real-time PCR and Western blot require fresh human tissue. Therefore, quantitative methods studying human glaucoma eyes are unfortunately often limited to immunohistochemistry applied on formalin-fixed paraffin-embedded eyes. However, immunohistochemistry is associated with some variability, as both the technical processing and the intraobserver variability highly impact the interpretation of the immunolabelling. In this study, all sections of each antibody were stained simultaneously in one bath to ensure that procedure parameters concerning ambient environment (tp. and humidity), buffers and dilution of antibodies were kept constant. Also, according to routine, the eyes were immerged and fixed in paraformaldehyde, supplied by the Eye Pathology Institute, immediately after removal. Further preparation of all eyes was handled by the Eye Pathology Institute and all paraffin-embedded specimens had been stored in © 2014 APMIS. Published by John Wiley & Sons Ltd
the same room. All specimens were cut by the same technician at 4 lm in thickness. To uniformly evaluate the immunolabelling, the initial step of analysis involved a fusion of all photomicrographs of each eye and normalizing the background to zero. An arbitrary value of staining intensity of bound antibody was then obtained by measuring the optical density. The linear relationship between the intensity of immunoperoxidase staining and the amount of antigen was the basis for the use of OD measurements as an estimate of staining variation (16, 17). Therefore, variations in staining intensities were quantifiable by measuring OD and statistical comparisons between different samples were applicable (16, 18). Using OD measurements of the ciliary body, we found a significant decrease in AQP7- and AQP9 expression. Optical density for AQP7 and AQP9 was only measured in the nonpigmented ciliary epithelium, as the pigment granules in the pigmented epithelium interfered with OD measurement. Interestingly, no decrease was found in AQP1- and AQP4 expression. AQP1 is a highly permeable water channel and conducts the main part of water from the nonpigmented epithelium into the aqueous humour (1). Constitutive AQP1- and AQP4 expression is consistent with an unaffected production of aqueous humour in glaucoma eyes, regardless of IOP (1). AQP7 and AQP9 are aquaglyceroporins and have the ability to transport both water and
775
TRAN et al.
A
A
B
B
C C 150 Staining intensity/arb. unit
Staining intensity/arb. unit
150
100
50
100
50
0 Control
POAG
NVG
CACG
Fig. 2. Aquaporins 7 immunolabelling in the anterior segment of normal and glaucoma eyes. (A) AQP7 expression in the ciliary epithelia of normal eyes (brown staining). In the immediate subepithelial stroma, there is a faint stain, which is also seen in (B). Bar = 50 lm. (B) In glaucoma eyes, AQP7 expression is decreased. The subepithelial faintly positive stroma is of the same intensity as in (A). Bar = 50 lm. (C) Staining intensity of AQP7 in the nonpigmented ciliary epithelium in normal eyes (control) and in glaucoma subgroups. Staining intensity denotes the optical density of positively labelled cells. The staining intensity of AQP7 is significantly decreased in the glaucoma groups compared with the control group (p = 0.003). There are no statistical differences among the three glaucomatous subgroups (p = 0.107). Arb. unit: arbitrary unit; POAG: primary open angle glaucoma; NVG: neovascular glaucoma; CACG: chronic angle-closure glaucoma.
776
0 Control
POAG
NVG
CACG
Fig. 3. AQP9 immunolabelling in the anterior segment of normal and glaucoma eyes. (A) In normal eyes, AQP9 is expressed in the ciliary epithelia (brown staining; bar = 50 lm). (B) In glaucoma eyes, AQP9 expression is decreased. Bar = 50 lm. (C) Staining intensity of AQP9 in the nonpigmented ciliary epithelium in control and in glaucoma subgroups. Staining intensity denotes the optical density of positively labelled cells. The staining intensity of AQP9 is significantly decreased in the glaucoma groups compared with the control group (p = 0.018). There are no statistical differences among the three glaucomatous subgroups (p = 0.253). Arb.unit: arbitrary units; POAG: primary open angle glaucoma; NVG: neovascular glaucoma; CACG: chronic angle-closure glaucoma. © 2014 APMIS. Published by John Wiley & Sons Ltd
AQUAPORINS AND GLAUCOMA
other small noncharged molecules such as glycerol, lactate and urea that can be used as substrates in energy production (7). Therefore, decrease in both AQP7- and AQP9 expression in the nonpigmented ciliary epithelium may be coupled with changes in the metabolism of the ciliary body and anterior chamber, which could aggravate the glaucomatous retinal damage. M€ uller cells in a glaucomatous retina display a changed expression of membrane proteins with functional consequence for the overall retinal homoeostasis and RGC survival (19). The increased AQP7 expression as seen in this study may be an A
B
C
D
E
Staining intensity/arb. unit
150
100
50
0 © 2014 APMIS. Published by John Wiley & Sons Ltd
attempt to stimulate glutamate uptake by alkalinization of the intracellular space, as AQP7 may facilitate the NH3-transport needed in the glutamate/glutamine cycle (8, 20). Downregulation of AQP4 expression has been observed in animal studies of optic nerve crush and autoimmune uveitis, where the M€ uller cells respond with reparative activation (21, 22). The unchanged AQP4 expression and increased AQP7 expression may be linked to a different M€ uller cell response characteristic of glaucoma eyes (23). In the retina, we found a significantly decreased AQP9 expression in retinal ganglion cells. AQP9 has been suggested to provide neurons in the brain with lactate and glycerol for energy metabolism (8, 24). AQP9 may have a similar function in retinal ganglion cells and facilitate the uptake of lactate and/or glycerol into the retinal ganglion cells and photoreceptors (8, 14, 25). The decrease in AQP9 expression may result in compromised metabolism of the RGCs due to deprivation of nutrients. Consequently, the RGCs may become fragile and susceptible to apoptosis triggered by new glaucomatous insults such as ischaemia, excitotoxicity, further neurotrophic and nutrient deprivation and oxidative stress (26, 27). Interestingly, the three groups of glaucoma eyes
Fig. 4. Aquaporins 7 immunolabelling in glial cells of the normal and glaucomatous retina. (A) In a normal retina, immunofluorescence labelling of AQP7 (green) is seen in the M€ uller cell endfeet (arrows) at the inner limiting membrane and at the outer limiting membrane (double arrow). Glial fibrillary acidic protein (GFAP) labelling (red) is seen in astrocyte- and M€ uller cell processes (arrowhead). Bar = 50 lm. (B) In the same normal retina, immunoperoxidase labelling of AQP7 is seen in the M€ uller cell endfeet (arrows) and at the outer limiting membrane in a normal retina (double arrow). Bar = 50 lm. (C) Immunofluorescense staining of glaucomatous retinas demonstrate an upregulated AQP7 expression (green) in the M€ uller cells (arrows) and AQP7 labelling is also seen in M€ uller cell processes stretching towards the outer retina into the outer plexiform layer. AQP7 labelling of the OLM (double arrow) is not increased. GFAP (red) is upregulated in the M€ uller cells and is expressed in the processes throughout the retina (arrowheads). Bar = 50 lm. (D) In the same glaucomatous retina, immunoperoxidase staining increased AQP7 expression in M€ uller cells (arrows) and OLM (double arrow) is evident. Bar = 50 lm. (E) Based on immunoperoxidase stained sections, AQP7 staining intensity was measured in the M€ uller cells endfeet in control and in glaucoma subgroups. Staining intensity denotes the optical density of positively labelled cells. The staining intensity of AQP7 is significantly increased in the glaucoma groups compared with the control group (p = 0.046). There were no statistical differences among the three glaucomatous subgroups (p = 0.731). Arb. unit, arbitrary unit; POAG, primary open angle glaucoma; NVG, neovascular glaucoma; CACG, chronic angle-closure glaucoma.
777
TRAN et al.
A
A
B
B
C
C 150
Staining intensity/arb. unit
Staining intensity/arb. unit
150
100
50
0 Control
POAG
NVG
CACG
Fig. 5. Aquaporins 9 immunolabelling in retinal ganglion cells of the normal and glaucomatous retina. (A) AQP9 expression is seen in the retinal ganglion cells in normal eyes (arrow). Bar = 50 lm. (B) In glaucoma eyes, AQP9 expression is significantly decreased in the retinal ganglion cells (arrow). Bar = 50 lm. (C) Staining intensity of AQP9 in the retinal ganglion cells in control and in glaucoma subgroups. Staining intensity denotes the optical density of positively labelled cells. The staining intensity of AQP9 is significantly decreased in the glaucoma groups compared with the control group (p = 0.037). There are no statistical differences among the three glaucomatous subgroups (p = 0.9812). Arb. unit, arbitrary unit; POAG, primary open angle glaucoma; NVG, neovascular glaucoma; CACG, chronic angle-closure glaucoma.
778
100
50
0 Control
POAG
NVG
CACG
Fig. 6. Aquaporins 9 immunolabelling in the optic nerve fibres of the normal and glaucomatous eye. (A) AQP9 expression is seen in and along the optic nerve fibres of normal eyes. Bar = 500 lm. (B) In glaucoma eyes, AQP9 expression in the optic nerve fibres has not changed; however, the amount of optic nerve fibres is markedly reduced. Bar = 500 lm. (C) Staining intensity of AQP9 in the optic nerve head in control and in glaucoma subgroups. Staining intensity denotes the optical density of positively labelled cells. The staining intensity of AQP9 is unchanged in the glaucoma groups compared with the control group (p = 0.448). Arb. unit: arbitrary unit; POAG: primary open angle glaucoma; NVG: neovascular glaucoma; CACG: chronic angle-closure glaucoma. © 2014 APMIS. Published by John Wiley & Sons Ltd
AQUAPORINS AND GLAUCOMA
showed the same staining intensities. Therefore, the decreased AQP9 expression may be a common feature of the RGCs when an initial glaucomatous insult occurs. In contrast, the AQP9 expression was unchanged in the nerve fibres of the optic nerve head in all the glaucoma eyes compared with the control group. Moreover, there were no differences in AQP9 expression among the three glaucoma groups regardless of cause. The unchanged AQP9 expression in the ONH suggests that the glaucomatous degeneration in the ONH is a secondary consequence of the initial RGC death. Studies using animal models of artificially elevated IOP showed a downregulation of AQP9 in the optic nerve head (14, 15). The decrease in AQP9 expression may be a transient effect, as the overall period with elevated IOP was considerably shorter than the lapse of glaucoma of the eyes in this study (14, 15). However, the results in both studies should be interpreted with caution as the sample sizes were relatively small. This study is the first investigating AQPs in human glaucoma eyes. The changed expression of AQP7 and AQP9 in the ciliary body may be associated with metabolic changes in glaucoma eyes. Furthermore, downregulation of AQP9 in RGCs indicates AQP9 as a relevant factor in RGC metabolism.
This study was supported by grants from the Danish Eye Research Foundation, Fight for Sight – Denmark, the Synoptik Foundation, Aase and Ejnar Danielsen Foundation and Civilingeniør Lars Andersen Legat.
REFERENCES 1. Wax M, Clark A, Civan MM. Glaucoma. In: Yanof MDJ, editor. Ophthalmology. St. Louis: Mosby, 2009: 1108–16. 2. Lundberg LU, Thygesen J, Damgaard-Jensen L, Serup L, Kessing SV. Glaukompatienter i behandling hos praktiserende ojenlaeger i Danmark. Estimat af antal patienter samt omfang af synsfeltsdefekter [Glaucoma patients treated by practicing ophthalmologists in Denmark. Estimated number of patients and the extent of visual field defects]. Ugeskr Laeger. 2000;162:3028–33. 3. Verkman AS, Ruiz-Ederra J, Levin MH. Functions of aquaporins in the eye. Prog Retin Eye Res 2008;27:420–33. 4. Tran TL, Bek T, Holm L, la COUR M, Nielsen S, Prause JU, et al. Aquaporins 6–12 in the human eye. Acta Ophthalmol. 2013;91:557–63. 5. Rojek A, Praetorius J, Frokiaer J, Nielsen S, Fenton RA. A current view of the mammalian aquaglyceroporins. Annu Rev Physiol. 2008;70:301–27.
© 2014 APMIS. Published by John Wiley & Sons Ltd
6. Hara-Chikuma M, Verkman AS. Aquaporin-3 facilitates epidermal cell migration and proliferation during wound healing. J Mol Med (Berl). 2008;86:221–31. 7. Tsukaguchi H, Weremowicz S, Morton CC, Hediger MA. Functional and molecular characterization of the human neutral solute channel aquaporin-9. Am J Physiol 1999;277:F685–96. 8. Bringmann A, Pannicke T, Grosche J, Francke M, Wiedemann P, Skatchkov SN, et al. Muller cells in the healthy and diseased retina. Prog Retin Eye Res 2006;25:397–424. 9. Kong H, Fan Y, Xie J, Ding J, Sha L, Shi X, et al. AQP4 knockout impairs proliferation, migration and neuronal differentiation of adult neural stem cells. J Cell Sci. 2008;121:4029–36. 10. Nielsen S, Smith BL, Christensen EI, Knepper MA, Agre P. CHIP28 water channels are localized in constitutively water-permeable segments of the nephron. J Cell Biol. 1993;120:371–83. 11. Kwon TH, Frokiaer J, Fernandez-Llama P, Knepper MA, Nielsen S. Reduced abundance of aquaporins in rats with bilateral ischemia-induced acute renal failure: prevention by alpha-MSH. Am J Physiol. 1999;277: F413–27. 12. Nejsum LN, Kwon TH, Jensen UB, Fumagalli O, Frokiaer J, Krane CM, et al. Functional requirement of aquaporin-5 in plasma membranes of sweat glands. Proc Natl Acad Sci U S A. 2002;99:511–6. 13. Nakamura H, Riley F, Sakai H, Rademaker W, Yue BY, Edward DP. Histopathological and immunohistochemical studies of lenticules after epikeratoplasty for keratoconus. Br J Ophthalmol 2005;89:841–6. 14. Naka M, Kanamori A, Negi A, Nakamura M. Reduced expression of aquaporin-9 in rat optic nerve head and retina following elevated intraocular pressure. Invest Ophthalmol Vis Sci. 2010;51:4618– 26. 15. Mizokami J, Kanamori A, Negi A, Nakamura M. A preliminary study of reduced expression of aquaporin9 in the optic nerve of primate and human eyes with glaucoma. Curr Eye Res 2011;36:1064–7. 16. Nabors LB, Songu-Mize E, Mize RR. Quantitative immunocytochemistry using an image analyzer. II. Concentration standards for transmitter immunocytochemistry. J Neurosci Methods. 1988;26:25–34. 17. Taylor CR, Levenson RM. Quantification of immunohistochemistry–issues concerning methods, utility and semiquantitative assessment II. Histopathology 2006;49:411–24. 18. Lehr HA, Jacobs TW, Yaziji H, Schnitt SJ, Gown AM. Quantitative evaluation of HER-2/neu status in breast cancer by fluorescence in situ hybridization and by immunohistochemistry with image analysis. Am J Clin Pathol 2001;115:814–22. 19. Nork TM, Ghobrial MW, Peyman GA, Tso MO. Massive retinal gliosis. A reactive proliferation of Muller cells. Arch Ophthalmol. 1986;104:1383–9. 20. Bringmann A, Pannicke T, Biedermann B, Francke M, Iandiev I, Grosche J, et al. Role of retinal glial cells in neurotransmitter uptake and metabolism. Neurochem Int 2009;54:143–60. 21. Dibas A, Oku H, Fukuhara M, Kurimoto T, Ikeda T, Patil RV, et al. Changes in ocular aquaporin expression following optic nerve crush. Mol Vis. 2010;16:330–40.
779
TRAN et al.
22. Motulsky E, Koch P, Janssens S, Lienart M, Vanbellinghen AM, Bolaky N, et al. Aquaporin expression in blood-retinal barrier cells during experimental autoimmune uveitis. Mol Vis. 2010;16:602–10. 23. Vidal L, Diaz F, Villena A, Moreno M, Campos JG, Perez de Vargas I. Reaction of Muller cells in an experimental rat model of increased intraocular pressure following timolol, latanoprost and brimonidine. Brain Res Bull 2010;82:18–24. 24. Mylonakou MN, Petersen PH, Rinvik E, Rojek A, Valdimarsdottir E, Zelenin S, et al. Analysis of mice with targeted deletion of AQP9 gene provides conclu-
780
sive evidence for expression of AQP9 in neurons. J Neurosci Res 2009;87:1310–22. 25. Badaut J, Brunet JF, Regli L. Aquaporins in the brain: from aqueduct to ‘multi-duct’. Metab Brain Dis 2007;22:251–63. 26. Miki A, Kanamori A, Negi A, Naka M, Nakamura M. Loss of Aquaporin 9 Expression Adversely Affects the Survival of Retinal Ganglion Cells. Am J Pathol. 2013;182:1727–39. 27. Almasieh M, Wilson AM, Morquette B, Cueva Vargas JL, Di Polo A. The molecular basis of retinal ganglion cell death in glaucoma. Prog Retin Eye Res. 2012;31:152–81.
© 2014 APMIS. Published by John Wiley & Sons Ltd
