Changes in ocular aquaporin expression following optic nerve crush

Abstract

Purpose: Changes in the expression of water channels (aquaporins; AQP) have been reported in several diseases. However, such changes and mechanisms remain to be evaluated for retinal injury after optic nerve crush (ONC). This study was designed to analyze changes in the expression of AQP4 (water selective channel) and AQP9 (water and lactate channel) following ONC in the rat.

Methods: Rat retinal ganglion cells (RGCs) were retrogradely labeled by applying FluoroGold onto the left superior colliculus 1 week before ONC. Retinal injuries were induced by ONC unilaterally. Real-time PCR was used to measure changes in AQP4, AQP9, thy-1, Kir4.1 (K(+) channel), and beta-actin messages. Changes in AQP4, AQP9, Kir4.1, B cell lymphoma-x (bcl-xl), and glial fibrillary acidic protein (GFAP) expression were measured in total retinal extracts using western blotting.

Results: The number of RGCs labeled retrogradely from the superior colliculus was 2,090+/-85 cells/mm(2) in rats without any treatment, which decreased to 1,091+/-78 (47% loss) and 497+/-87 cells/mm(2) (76% loss) on days 7 and 14, respectively. AQP4, Kir4.1, and thy-1 protein levels decreased at days 2, 7, and 14, which paralleled a similar reduction in mRNA levels, with the exception of Kir4.1 mRNA at day 2 showing an apparent upregulation. In contrast, AQP9 mRNA and protein levels showed opposite changes to those observed for the latter targets. Whereas AQP9 mRNA increased at days 2 and 14, protein levels decreased at both time points. AQP9 mRNA decreased at day 7, while protein levels increased. GFAP (a marker of astrogliosis) remained upregulated at days 2, 7, and 14, while bcl-xl (anti-apoptotic) decreased.

Conclusions: The reduced expression of AQP4 and Kir4.1 suggests dysfunctional ion coupling in retina following ONC and likely impaired retinal function. The sustained increase in GFAP indicates astrogliosis, while the decreased bcl-xl protein level suggests a commitment to cellular death, as clearly shown by the reduction in the RGC population and decreased thy-1 expression. Changes in AQP9 expression suggest a contribution of the channel to retinal ganglion cell death and response of distinct amacrine cells known to express AQP9 following traumatic injuries.

Figure 1

Optic nerve crush decreased number of retinal ganglion cells (RGCs), thy-1 protein, and mRNA levels. The mean number of RGCs labeled retrogradely from superior colliculus was 2090±85/mm² in sham rats without any treatment (A), which decreased to 1091±78 (47% loss, B) and 497±87/mm² (76% loss, C) on day 7 and 14, respectively. Thy-1 expressed by retinal ganglion cells is widely used as marker for RGC stress and was shown to be reduced following optic nerve crush. Following retinal injuries with optic nerve crush, retinas were dissected and plasma membrane proteins were isolated or total RNA was isolated and transcribed into cDNA. Thirty microgram protein was loaded into each lane. D: Immunoreactive bands for thy-1 and β-tubulin at 2, 7, and 14 days following crush showing a significant reduction in thy-1 protein levels at 7 and 14 day but not at 2 days (94±18%, 33±6%, 24±7% at 2, 7, and 14 day, respectively, n=7). Densitometric quantification is shown in E. Data are expressed as a ratio of the control value and each measurement represents mean±SEM *Denotes statistical significance of thy-1 protein levels in optic nerve crushed retinas versus sham (p<0.005) as determined by one-way ANOVA and Tukey multiple comparison test. Optic nerve crush also decreased thy-1 transcripts significantly (58±7%, 6±2%, at 2 and 7 day, respectively, % p<0.001 versus sham, n=8), as determined by quantitative real-time PCR (E). Thy1- mRNA at 2-weeks was below detection levels and considered zero. Gene expression data of thy-1 is calculated after normalizing with β-actin. **Denote significant differences compared with sham-retinas at p<0.05. Abbreviations: sham eye (C), and crushed (R).

Figure 2

Optic nerve crush decreased AQP4 protein and mRNA levels. Following retinal injuries with optic nerve crush, retinas were dissected and plasma membrane proteins were isolated or total RNA was isolated and transcribed into cDNA. Thirty microgram protein was loaded into each lane. Immunoreactive bands for AQP4 and β-tubulin 2, 7, and 14 days after optic nerve crush showing a significant reduction in AQP4 protein levels. Quantitative measurement using western blot showed that optic nerve crush decreased AQP4 protein levels by ~40% (60±10, 63±10, 38±10, at 2, 7, and 14 day, respectively, n=7, A). Densitometric quantification is shown in B. Data are expressed as a ratio of the treated to control or sham value and each measurement represents mean±SEM *Denote statistical significance of AQP4 protein levels in optic nerve crushed retinas versus sham (p<0.005) as determined by one-way ANOVA and Tukey multiple comparison test. Optic nerve crush also decreased AQP4 transcripts significantly (61±5%, 60±8%, 58±6, at 2, 7, and 14 day, respectively, n=7), as determined by quantitative real-time PCR (B). Gene expression data of AQP4 is calculated after normalizing with ACTB. **Denote significant differences compared with sham-retinas at p<0.05. Abbreviations: sham eye (C), and crushed (R).

Figure 3

Optic nerve crush decreased Kir4.1 protein and mRNA levels. Following retinal injuries with optic nerve crush, retinas were dissected and plasma membrane proteins were isolated. Total RNA was isolated and transcribed into cDNA. Real-time PCR was performed using specific primers (see Methods). mRNA expression of Kir4.1 was adjusted to the mRNA copies of ACTB (reference gene). Thirty microgram protein was loaded into each lane. Immunoreactive bands for Kir4.1 and β-tubulin 2, 7, and 14 days after optic nerve crush showing a significant reduction in Kir4.1 protein levels. Quantitative measurement using western blot showed that elevation of optic nerve crush decreased Kir4.1 protein levels (34±10, 72±6, 22±9, at 2, 7, and 14 day, respectively, A, n=7). Densitometric quantification is shown in B. Data are expressed as a ratio of the control value and each measurement represents mean±SEM *Denote statistical significance of Kir4.1 protein levels in optic nerve crushed retinas versus sham (p<0.005) as determined by one-way ANOVA and Tukey multiple comparison test. Results indicate that mRNA expression level of Kir4.1 was significantly lower in optic nerve crushed retinas compared to sham at 7 and 14 days (55±7%, 41±9%, at 7, and 14 day, respectively, B, n=6). By contrast, optic nerve crush increased Kir4.1 mRNA at 2 days (325±81%, B, n=6). **Denote significant differences compared with sham-retinas at p<0.05. Abbreviations: sham eye (C), and crushed (R).

Figure 4

Optic nerve crush induced novel changes in AQP9 protein and mRNA levels. AQP9 protein levels showed opposite trend to changes in mRNA where they increased at 7 day (174±27%, p<0.001 versus control, A, n=6) but decreased at 2 and 14 day (38±19%, 50±13%, at 2, and 14 day, respectively, A, n=8). AQP9 shows a doublet and both bands were used in densitometric quantification in B. Data are expressed as a ratio of the control value and each measurement represents mean±SEM *Denote statistical significance of AQP9 protein levels in optic nerve crushed retinas versus sham (p<0.005) as determined by one-way ANOVA and Tukey multiple comparison test. Optic AQP9 transcripts were quantified by quantitative real-time PCR (B). Gene expression data of AQP9 is calculated after normalizing with ACTB. AQP9 mRNA increased at 2 and 14 day (150±30%, 200±30%, at 2, and 14 day, respectively, n=5) but decreased at 7 day (44±13%, n=6, B). **Denote significant differences compared with sham-retinas at p<0.05. Abbreviations: sham eye (C), and crushed (R).

Figure 5

Glial fibrillary acidic protein is upregulated following optic nerve crush, while B cell lymphoma-x (bcl-xl) is downregulated. Following retinal injuries with optic nerve crush, retinas were dissected and cytosolic proteins were isolated. Fifty micrograms protein was loaded into each lane. Glial fibrillary acidic protein (GFAP), a cellular marker for retinal injury, was upregulated (A: 450 ±190%, 262±30%, and 300±34%, days 2, 7, and 14, respectively, p<0.001, n=7). Bcl-xl (anti-apoptotic factor) downregulation is involved in initiating apoptosis and not surprisingly was reduced following optic nerve crush (B: 67±9% and 56±10%, at days 7 and 14, respectively, p<0.005, n=7). However, changes in bcl-xl protein levels at day 2 were not statistically significant.