Artigo Acesso aberto Revisado por pares

Aquaporins 6-12 in the human eye

2012; Wiley; Volume: 91; Issue: 6 Linguagem: Inglês

10.1111/j.1755-3768.2012.02547.x

ISSN

1755-3768

Autores

Thuy Linh Tran, Toke Bek, Lars Holm, Morten la Cour, Søren Nielsen, Jan Ulrik Prause, Aleksandra Rojek, Steffen Hamann, Steffen Heegaard,

Tópico(s)

Neonatal Respiratory Health Research

Resumo

Purpose: Aquaporins (AQPs) are widely expressed and have diverse distribution patterns in the eye. AQPs 0–5 have been localized at the cellular level in human eyes. We investigated the presence of the more recently discovered AQPs 6–12 in the human eye. Methods: RT-PCR was performed on fresh tissue from two human eyes divided into the cornea, corneal limbus, ciliary body and iris, lens, choroid, optic nerve, retina and sclera. Each structure was examined to detect the mRNA of AQPs 6–12. Twenty-one human eyes were examined using immunohistochemical and immunofluorescence techniques to determine the topographical localization of AQPs 6–12. Results: mRNA transcripts of AQP7, AQP9 and AQP11 were found in the ciliary body, corneo-limbal tissue, optic nerve, retina and sclera. AQP9 and AQP11 mRNA was also detected in the choroid. No mRNA of AQP6, AQP8, AQP10 or AQP12 was detected. Anti-AQP7 immunolabelling was detected in the corneal epithelium, corneal endothelium, trabecular meshwork endothelium, ciliary epithelia, lens epithelium, the inner and outer limiting membrane of the retina, the retinal pigment epithelium and the capillary endothelium of all parts of the eye. AQP9 immunolabelling was detected in the nonpigmented ciliary epithelium and retinal ganglion cells. AQP11 immunolabelling was detected in the corneo-limbal epithelium, nonpigmented ciliary epithelium and inner limiting membrane of the retina. Conclusion: Selective expression of AQP7, AQP9 and AQP11 was found within various structures of the human eye. The detection of these aquaporins in the eye implies a role that may be related not only to water transport but also to the transport of glycerol, lactate and ammonia, with importance for metabolism, especially in the retina. The eye is mainly comprised of water, and thus, correct functioning is strongly dependent on water secretion and absorption mechanisms – in other words, water homoeostasis (Hamann 2002). The transport of water across the corneal epithelium and endothelium maintains corneal transparency (Hamann 2002). Rapid changes in the water content of the iris stroma facilitate changes in shape during pupil constriction and dilatation (Hamann 2002). In the retina, water is transported transcellularly across the retinal pigment epithelium into the choroid, preventing subretinal oedema and retinal detachment (Hamann et al. 1998; Hamann 2002). The intraocular pressure (IOP) is maintained by the aqueous humour (Gabelt & Kaufman 2003). Aqueous humour is secreted by the pigmented and nonpigmented epithelia of the ciliary body in a concerted action involving active membrane proteins and passive ion and water channels – the aquaporins (Gabelt & Kaufman 2003). Aquaporins are selective channels in the cell membrane and mainly facilitate the transport of water in response to osmotic gradients (Verkman et al. 2008). In humans, the aquaporin family contains 13 isoforms divided into (1) classic aquaporins, comprising AQP0, AQP1, AQP2, AQP4 and AQP5, which transport water molecules only; (2) aquaglyceroporins, comprising AQP3, AQP7, AQP9 and AQP10, which besides water also allow the passage of small molecules like ammonia, glycerol and lactate; and (3) unorthodox aquaporins, comprising AQP6, AQP8, AQP11 and AQP12, whose transport profiles have not yet been fully elucidated (Rojek et al. 2008). Studies using knockout mice have demonstrated that the transport of water and small solutes by aquaporins is important for physiological functions in multiple organs, for example, urine concentration, exocrine gland secretion, brain hydration, neural signal transduction and metabolism, fat metabolism and skin hydration (Hara-Chikuma & Verkman 2008; Rojek et al. 2008). New findings suggest that AQP4 also plays a crucial role in regulating the proliferation, migration and differentiation of neuronal stem cells (Kong et al. 2008). AQP4 also modulates neural signal transduction and neuronal excitability by regulating the osmotic and ionic gradients around the neurons (Kong et al. 2008; Takagi et al. 2009). We have previously mapped the distribution of AQPs 1–5 in human and rat eyes (Hamann et al. 1998). Studies have shown that AQP0 contributes to lens transparency, AQP1 is involved in the secretion and drainage of aqueous humour, AQP3 and AQP5 have corneal and conjunctival barrier functions, and AQP4 plays an important role in retinal water homoeostasis (Hamann et al. 1998; Hamann 2002; Wax et al. 2009). The recent discovery of new isoforms (AQPs 6–12) means that further mapping is now required, as the location and functional characteristics of aquaporins may indicate their specific roles in the normal and diseased eye. Fifteen paraffin-embedded human eyes and two recently enucleated nonfixed human eyes were examined. All eyes had been removed because of uveal melanoma. However, only areas without any macro- and microscopic signs of tumour invasion were examined. Furthermore, six normal paraffin-embedded human eyes, removed because of orbital tumour, were examined. The two nonfixed eyes were classified as T3aN0M0 without retinal detachment. Of the paraffin-embedded eyes, six eyes were classified as T2aN0M0, four eyes had T2bN0M0 uveal melanoma and fives eyes had T3aN0M0 uveal melanoma. Seven of the eyes had minor retinal detachment and two had focal retinal invasion. The study followed the tenets of the Declaration of Helsinki. 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). Eye specimens were obtained with informed consent from patients undergoing enucleation of the eye. Fresh tissue from two eyes was divided into the cornea, corneal limbus, ciliary body and iris, lens, choroid, optic nerve, retina and sclera, and mRNA was extracted from each of these. The retina was detached from the choroid by injection of physiological saline solution by which the RPE cells were separated from the neuroretina. Thus, the RPE cells were included in the choroid sample. RNA from human kidney was extracted as a positive control. Tissue from the two nonfixed eyes, containing all of the above-mentioned structures except the lens and optic nerve, was also processed for cryostat sections. Specimens for cryostat section were fixed by immersion in 4% paraformaldehyde in 0.1 m sodium cacodylate buffer, pH 7.2 for 4 h at 4°C, followed by a brief rinse in 5% PBS buffer, pH 7.4 and cryoprotected in 30% sucrose PBS, pH 7.4 at 4°C overnight. The tissue was embedded in Tissue-Tek (Sakura Finetek, Torrance, CA, USA) and snap frozen using liquid nitrogen. Tissue blocks were stored at −80°C until sections were cut. Cryosections (12 μm) were stained for immunofluorescence analysis. Eyes for immunohistochemical staining were fixed by immersion in 4% paraformaldehyde in 0.1 m sodium cacodylate buffer, pH 7.2 for 4 h at 4°C, dehydrated in ethanol followed by xylene and finally embedded in paraffin. Staining of 4-μm sections was carried out using either indirect immunofluorescence or indirect immunoperoxidase. Total RNA was extracted from fresh tissues using Trizol (Invitrogen, Taastrup, Denmark) in combination with RNeasy Mini Kits (Qiagen, Hilden, Germany) according to the manufacturers' instructions. After DNAse treatment (RNase-Free DNase; Qiagen), the RNA was reverse-transcribed using 2 U/μl reverse transcriptase (Superscript II; Invitrogen) in the presence of oligo dT primers. Primers were designed to specifically amplify AQPs 6–12 (Table 1) and were designed to span over an intron. The correctly sized amplification fragment was produced in PCRs with human genomic DNA as template. PCR (35 cycles) was performed (HotStarTaq Master Mix; Qiagen) using 1 pmol of each primer: hot-start at 95°C for 15 min, denaturation at 95°C for 30 s, annealing at 58°C for 30 s and elongation at 72°C for 1 min. For AQP8 and AQP12, the annealing temperature was 60°C. Negative controls were performed without reverse transcriptase or cDNA. PCR for β-actin was performed to assess the quality of cDNA. Human kidney cDNA was used as a positive control, confirming the expression of AQP6, AQP7 and AQP11. PCR products were separated by 1.5% agarose gel electrophoresis, and the products were photographed under ultraviolet illumination. Paraffin sections from all 21 eyes were deparaffinized and rehydrated. For immunoperoxidase labelling, endogenous peroxidase was blocked by 0.5% H2O2 in absolute methanol for 30 min at room temperature. To reveal antigens, sections were submerged in 1 mm Tris solution (pH 9.0) supplemented with 0.5 mm EGTA and heated in a microwave oven at 650 W for 6 min (98°C) and then at 350 W for 10 min (95°C). After this treatment, sections were left to cool for 30 min in the buffer. Non-specific binding of IgG was prevented by incubating the sections in 50 mm NH4Cl for 30 min, followed by blocking in PBS supplemented with 1% BSA, 0.05% saponin and 0.2% gelatin. Sections were incubated overnight at 4°C with primary antibodies diluted in phosphate-buffered saline (PBS) supplemented with 0.1% BSA and 0.3% Triton-X-100. The sections were rinsed with PBS containing 0.1% bovine serum albumin (BSA) and then incubated with horseradish peroxidase-conjugated (HRP) secondary antibody (P448, 1:200; Dako, Glostrup, Denmark). Labelling was visualized by 0.05% 3,3-diaminobenzidine tetrahydrochloride (DAB). Sections were then counterstained with Mayer's modified haematoxylin and mounted with coverslips. Additional tyramide amplification was employed to enhance the signal from AQP9 as this protein occurs at low levels. After overnight incubation with anti-AQP9 antibody as described above, the sections were incubated with HRP-conjugated secondary antibody (donkey anti-rabbit; Thermo Scientific, Rockford, IL, USA). After washing in PBS containing 0.1% BSA, the sections were incubated with TSA Plus Biotin Amplification solution (Perkin Elmer, Waltham, MA, USA) for 10 min. Antibody binding was visualized with FITC-conjugated streptavidin. Fluorescence techniques were employed for double-labelling of the tissues. After simultaneous overnight incubation with primary antibody as described above plus anti-glial fibrillary acidic protein (anti-GFAP) antibody, the sections were incubated with fluorescent secondary antibody (donkey anti-rabbit antibodies AlexaFluor 488, donkey anti-goat AlexaFluor 546; Invitrogen). After washing (PBS for 3× 10 min), the sections were mounted with coverslips. AQP 11 immunolabelling was performed on cryosections. Cryosections were left at room temperature for 30 min, and antigens were retrieved through graded alcohol (99%, 96% and 70%). Sections were blocked with PBS supplemented with 1% BSA, 0.05% saponin and 0.2% gelatin. The subsequent steps were as described for immunofluorescence staining. Affinity-purified polyclonal antibodies to human AQP7 were raised in rabbits against a peptide corresponding to the human C-terminal amino acids: 321–341 (CANRSSVHPAPPLESMALEHF). Affinity-purified polyclonal antibodies to AQP11 were raised in rabbits against a peptide corresponding to the rat N-terminal amino acids: 1–14 (MSALLGCPEVQDTC). This peptide has a homology of 67% between humans and rats. Both antibodies were affinity purified. Rabbit anti-human AQP9 antibody was purchased from Abcam (Cambridge, UK; ab84828). Goat anti-GFAP antibody was purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA; sc-6170). The specificity of the anti-AQP7 and anti-AQP9 antibodies was verified by staining control human kidney and liver, respectively. The reactions lacking the primary antibody produced no signal. Furthermore, immunolabelling with preimmunized serum was also performed. Immunoblotting for AQP7, AQP9 and AQP11 did not detect any signal in the membrane samples from the eye, and control samples suggested that the antibodies were unsuitable for utilization in immunoblotting. RT-PCR detected AQP7, AQP9 and AQP11 in the two human eyes. AQP7 mRNA was found in the ciliary body, corneo-limbal tissue, optic nerve, retina and sclera (Fig. 1). A weak transcript signal of AQP7 was seen in the choroid and cornea. The primer pair for AQP7 amplified two bands of 189 and 557 bp. Subsequent sequencing of the gel extraction product showed that the bands corresponded to two different splice variants of AQP7 (Ensembl sequences ENST00000379507 and ENST00000379506). The 189-bp band was the stronger of the two in all eye samples in which AQP7 was detected as well as the control sample (human kidney). AQP9 mRNA was detected in the choroid, ciliary body, corneo-limbal tissue, retina and sclera. AQP11 mRNA was detected in the choroid, ciliary body, cornea, corneo-limbal tissue, optic nerve, retina and sclera. RNA could not be retrieved from the lens of the two eyes, and the lens was subsequently excluded from further analysis using RT-PCR. Transcripts of AQP6, AQP8, AQP10 and AQP12 were not detected in the examined tissues. The control β-actin was detected in the cornea, corneal limbus, ciliary body, choroid, optic nerve, retina and sclera. However, amplification in the sclera was less efficient. Aquaporin (AQP) mRNA transcript expression in normal human eyes. AQP mRNA expression examined in the choroid (CH), ciliary body and iris (CB), cornea (CO), limbus (LIM), optic nerve (ON), retina (RE) and sclera (SC). AQP7 mRNA (189 and 557 bp) in CH, CB, LIM, ON, RE and SC of the eye. AQP9 mRNA (196 bp) in CH, CB, LIM, RE and SC. AQP11 mRNA (121 bp) in all the above structures of the eye. Human kidney cDNA (K) and β-actin (318 bp) served as positive controls, whereas genomic DNA (d) served as the negative control. RT(−) = reverse transcriptase negative samples; RT(+) = reverse transcriptase positive samples. Immunolabelling of normal eyes and eyes with uveal melanoma was identical in histological normal areas. Positive labelling with anti-AQP7 was detected in the cytoplasm of the basal cells in the anterior corneal epithelium and conjunctival epithelium in the limbal region (Fig. 2A,B). AQP7 labelling was strongest in the basal cells of the conjunctival epithelium, with lower staining intensity in the outer cell layers. The basal cells of the epithelium in the limbal area also showed positive labelling with anti-AQP11 (Fig. 2C). The corneal endothelium showed positive AQP7 labelling of the cytoplasm (Fig. 2D). Immunolocalization of AQP7 and AQP11 in the cornea and corneo-limbal region. Bar = 25 μm. (A) Positive labelling with anti-AQP7 antibody in the basal cells of the anterior corneal epithelium. The basal cells show abundant staining throughout the cornea. (B) In the limbal area and the zone of transition to conjunctiva, AQP7 labelling is most strongly expressed in the basal cells of the epithelium (arrow) with decreasing staining intensity in the outer cell layers. The endothelial cells of the capillaries show strong labelling (arrowheads). (C) Immunofluorescence labelling of AQP11 in the basal cell layer of the epithelium in the limbal area of the cornea (green; arrow). (D) The corneal endothelium demonstrates strong binding of anti-AQP7 in the cytoplasm. The endothelial cells of the trabecular meshwork and Schlemm's canal showed positive labelling for AQP7 (Fig. 3A). Immunolocalization of AQP7 in the trabecular meshwork and lens. Bar = 25 μm. (A) Immunolabelling of AQP7 of the trabecular meshwork endothelium (arrow) and Schlemm's canal (arrowhead). (B) The anterior lens epithelium is positive for AQP7 with strong labelling at the apical membrane. Immunohistochemistry showed positive AQP7 labelling of the anterior lens epithelium with marked labelling of the apical membrane (Fig. 3B). The lens epithelium cytoplasm showed unspecific labelling. The pars plicata of the ciliary body revealed strong AQP7 immunolabelling of the basolateral membranes and in the cytoplasm of the nonpigmented ciliary epithelium (Fig. 4A). Anti-AQP7 labelling was also seen in the basolateral membranes of the pigmented epithelial cells (Fig. 4A). However, no labelling was detected in the cytoplasm of the pigmented ciliary epithelium (Fig. 4A). The intense cytoplasmic labelling in the nonpigmented epithelium masked visualization of potential labelling in the juxtaposed apical cell membranes of the two epithelial layers. The ciliary stroma was unlabelled, but the endothelium of capillaries and postcapillary venules in the stroma exhibited strong AQP7 immunolabelling (Fig. 4A). The pigmented epithelium of the iris was negative for AQP7 immunolabelling, and there was a clear change in immunoreactivity at the base of the iris. Immunolocalization of AQP7, AQP9 and AQP11 in the iris and ciliary body. Bar = 25 μm. (A) Strong AQP7 labelling in the basolateral membrane and cytoplasm of the nonpigmented ciliary epithelium (arrow) and basolateral membrane of the pigmented ciliary epithelium (arrowhead). Positive AQP7 labelling of capillary endothelium (double arrow). (B) The cytoplasm of the nonpigmented ciliary epithelium show labelling with anti-AQP9 antibody (arrow). (C) Labelling of AQP11 in the basal membrane of the nonpigmented ciliary epithelium (arrow). (D) Weak labelling of AQP9 in the smooth muscle fibres of the iris sphincter muscle (arrow) and iris dilator muscle (arrowhead). AQP9 was detected in the cytoplasm of the nonpigmented ciliary epithelium (Fig. 4B). Furthermore, the ciliary and iridal muscles showed weak labelling of AQP9 (Fig. 4D). AQP11 was detected in the basal membrane of the nonpigmented ciliary epithelium (Fig. 4C). The endfeet of Müller cells at the inner limiting membrane (ILM) showed positive labelling for AQP7 and AQP11 (Fig. 5A,C). AQP7 was also detected in the outer limiting membrane (OLM; Fig. 5A). The cytoplasm of the retinal pigment epithelium showed positive labelling for AQP7 (Fig. 5A). The retinal ganglion cells showed positive labelling for AQP9 in the cytoplasm (Fig. 5B). AQP9 labelling was also detected in the astrocyte processes ensheathing the retinal capillaries and in the inner segments of the photoreceptors (Fig. 5B). Immunolocalization of AQP7, AQP9 and AQP11 in the retina. Bar = 25 μm. (A) Positive labelling of AQP7 in the outer limiting membrane (arrowheads) of the retina and endothelium of small retinal capillaries (arrows). The endfeet of Müller cells shows positive labelling at the inner limiting membrane (double arrows). The retinal pigment epithelium (RPE) is marked (star). Insert: Positive labelling of AQP7 in the RPE cell cytoplasm (arrow). The drusen basal to the RPE are negative (asterisk). The RPE above the drusen expresses AQP7. (B) Positive labelling of AQP9 in the cytoplasm of retinal ganglion cells (arrow). AQP9 labelling of astrocytes processes around a retinal capillary (double arrow, upper insert). AQP9 labelling of the inner segments of the photoreceptors (insert) in both rods (arrow, insert) and cones (arrowhead, lower insert). (C) The endfeet of Müller cells at the inner limiting membrane shows positive labelling (green) for AQP11 (arrow). GFAP is positively labelled in both astrocyte and Muller cell processes (red). Astrocyte processes ensheathing the retinal capillaries are positively labelled for GFAP (glial fibrillary acidic protein) (red; double arrow). Part of the Müller cell endfeet shows co-labelling of both antibodies (yellow; arrowhead). The capillary endothelium was positive for AQP7 in all vessels. This study was performed to investigate the presence of AQPs with particular interest in human eyes. The expression and localization of AQPs 6–12 in human eyes were investigated, and the presence of AQP7, AQP9 and AQP11 was confirmed by RT-PCR analysis and immunohistochemical staining. Unfortunately, Western blot could not be performed, which could have supported the findings further. Eyes with uveal melanoma were used because the availability of normal and healthy eyes is extremely limited. Cornea donor eyes normally show marked autolytic changes with significant degradation of RNA even after a short-time post-mortem (Jonas et al. 1992). Thus, the use of cornea donor eyes for RT-PCR analysis would result in loss of important data. We compared eyes with uveal melanoma and eyes enucleated because of orbital tumour and observed no differences in AQP immunoreactivity in normal histological areas. Rather than using animal eyes, where both eyes and control tissue are easily accessible, we performed our study using human tissue. Studies using animal eyes also have the advantage of animal knock out models. However, results obtained from animal studies cannot be transferred directly to humans in general, therefore, studies on human tissue are necessary. In this study, transcripts of AQP6, AQP8, AQP10 and AQP12 were not detected. However, other studies have reported the presence of AQP6, AQP8 and AQP10 transcripts in rat retinas and AQP6 immunolabelling of the outer plexiform layer (Tenckhoff et al. 2005; Iandiev et al. 2011). The disparity of species might explain the different results. Two splice variants of AQP7 mRNA were detected in both eye tissue and kidney control tissue corresponding to Ensembl sequences ENST00000379507 and ENST00000379506. However, the latter transcript was longer than the sequence registered in the database. The anti-AQP7 antibody only recognizes the first splice variant of AQP7 (ENST00000379507) because the two transcripts have different C-terminals. The protein translation and cellular localization of the alternate AQP7 splice variant were therefore impossible to verify. AQP7 and AQP11 were expressed in a pattern similar to that of AQP3 in the corneo-limbal region (Hara-Chikuma & Verkman 2008). Previous studies have shown that AQP3 participates in cell migration and proliferation in the wound healing process because of its water- and glycerol-transporting properties (Levin & Verkman 2006; Hara-Chikuma & Verkman 2008). AQP7 and AQP11 might act in concert with AQP3 to mediate corneal wound healing. AQP7, AQP9 and AQP11 were found in the ciliary epithelia, suggesting a role in aqueous humour secretion. AQP1 and AQP4 have been shown to be important in aqueous humour secretion and are responsible for most of the passive transport of water (Gabelt & Kaufman 2003). Absence of AQP labelling in the iris pigment epithelium further supports a transporting role for the AQPs in the ciliary epithelia, as the iris pigment epithelium does not have a transport function. However, AQP7, AQP9 and AQP11 possibly only contribute to water secretion in the formation of aqueous humour to a minor extent, as the water permeability of AQP7, AQP9 and AQP11 is much lower than that of AQP1 and AQP4 (Yang & Verkman 1997; Tsukaguchi et al. 1999; Patil et al. 2001; Sohara et al. 2005; Yakata et al. 2011). AQP9 was found in the cytoplasm of ganglion cells in the retina, confirming previous studies (Dibas et al. 2007; Naka et al. 2010). However, other studies have also found AQP9 in the catecholaminergic amacrine cells (Iandiev et al. 2006). Labelling of AQP9 was also detected in astrocytes processes ensheathing the retinal capillaries and in the cytoplasm of the inner segments of the photoreceptors. AQP9 has formerly been found in the brain, primarily in the astrocytes, and has been suggested to provide neurons with lactate and glycerol for energy metabolism (Bringmann et al. 2006; Yang et al. 2011). The presence of AQP9 in both the brain and retina is not surprising given their close connection. Accordingly, AQP9 may have a similar function in the retina and brain and may facilitate the uptake of lactate or glycerol into the retinal ganglion cells and photoreceptors (Naka et al. 2010). AQP7 and AQP11 were localized to the endfeet of Müller cells at the ILM. AQP4 and the inwardly rectifying potassium channel, Kir4.1, are co-localized in the ILM and probably maintain the spatial buffering of K+ into the vitreous body following synaptic transmission (Goodyear et al. 2009). AQP4 facilitates the accompanying osmotic water transport in response to K+-flux through Kir4.1 (Goodyear et al. 2009). The concurrent localization of AQP7 and AQP11 at the ILM suggests that these aquaporins also participate in water transport into the vitreous body. Interestingly, Kir4.1 has also been localized to the OLM and partially transports K+ back into the subretinal space (Nagelhus et al. 1999). The resulting osmotic gradient across the OLM is presumably equilibrated by water transport through ion channels and cotransporters (Goodyear et al. 2009), a process in which AQP7 may be involved. APQ7 labelling was also present in the cytoplasm of the retinal pigment epithelium (RPE). Among many important functions, RPE cells support photoreceptor cell survival, form the outer blood–retinal barrier and maintain transepithelial ion- and fluid transport between the retina and choroid (Kaufman & Alm 2003). The abundant cotransport proteins in the apical membrane of the RPE are probably responsible for the majority of water transport (Hamann 2002), but none of the known aquaporins has yet been localized to the RPE cell membrane with certainty. Specifically, in a previous study using immunoblotting, immunocytochemistry and immunoelectron microscopy in intact human and rat ocular tissue (Hamann et al. 2010), the highly water-permeable AQP1 was found to be absent from the RPE, although a study on cultured RPE cells suggested the opposite (Stamer et al. 2003). However, it is unlikely that AQP7 contributes to the water transport through the RPE to any significant extent, as immunolabelling showed a cytoplasmatic localization of AQP7. The demonstration of AQP7, AQP9 and AQP11 in various structures of the human eye is interesting, as these aquaporins are capable of transporting molecules other than water. They have been localized to specific and important regulatory sites within the human eye where the transport of both water and other solutes with metabolic properties occurs. For instance, AQP7 was also found in the corneal endothelium, trabecular meshwork endothelium and lens epithelium. Besides water transport, AQP7 may also contribute to the transport of nutrients. Furthermore, AQP7 and AQP9 may transport solutes that are useful in energy metabolism in the retina. For this reason, the role of aquaporins in tissues like the retina and the optic nerve probably exceeds that of just fluid regulation. This study was supported by grants from the Danish Eye Research Foundation, the Synoptik Foundation, Aase and Ejnar Danielsen Foundation and Civilingeniør Lars Andersen Legat.

Referência(s)
Altmetric
PlumX