Purinergic Receptors in the Rat Lens: Activation of P2X Receptors following Hyperosmotic Stress

Abstract

Purpose.: A range of P2Y and P2X receptors is expressed in the rat lens. Because most P2X receptors are located in the cytoplasm, the authors sought to determine whether P2X receptors are functionally active.

Methods.: Rat lenses were cultured under either isotonic or hypertonic conditions in the presence of P2 receptor agonists or antagonists. Lenses were fixed and cryosectioned, cell membranes were labeled, and confocal microscopy was used to determine whether the different reagents affected cell morphology.

Results.: Application of the P2 receptor inhibitor PPADS to lenses cultured under isotonic conditions induced extracellular space dilations between fiber cells in a distinct zone in the outer cortex. This damage was not caused by the inhibition of P2X₁ or P2Y₁ because more selective antagonists for P2X₁ (MRS2159) and P2Y₁ (MRS2179) either did not cause any damage or induced extracellular dilations located between superficial fiber cells at the lens modiolus, respectively. Although the P2 agonists ATPγS and ADPβS both induced a distinctive disruption to cell morphology in the same localized zone as PPADS, the P2X-specific agonist α,β-methyl-ATP induced no change to cell morphology. However, under hypertonic conditions that cause the insertion of P2X_1,4 into the membranes, α,β-methyl-ATP induced a localized zone of damage that was associated with changes in actin distribution.

Conclusions.: The results show that P2X receptors may play a minimal role in mediating ion fluxes in the rat lens under steady state conditions. In contrast, hypertonic cell shrinkage activates previously inactive P2X receptors in the lens, suggesting P2X receptors may play a role in the lens in response to osmotic stress.

Expression of purinergic (P2) receptors has been reported widely among tissues of the eye, including the retina, ciliary body, cornea, and lens. ¹ P2 receptors use extracellular nucleotides (adenosine 5′-triphosphate [ATP], adenosine 5′-diphosphate [ADP], uridine 5′-triphosphate [UTP], and uridine 5′-diphosphate [UDP]) as signaling molecules to activate members of either the ionotropic P2X (P2X_1–7) or the metabotropic P2Y (P2Y_{1,2,4,6,11–14}) subfamily. ^2,3 P2X receptors form trimeric channels, and their activation by ATP results in rapid influx of Ca²⁺ and Na⁺ ions. ⁴ Members of the P2Y subfamily are activated in an isoform-specific manner by a range of nucleotides, which results in either the downstream activation of phospholipase C and mobilization of Ca²⁺ from the endoplasmic reticulum (P2Y₁, P2Y₂, P2Y₄, P2Y₆, P2Y₁₁) or the modulation of cAMP levels by stimulation (P2Y₁₁) or inhibition (P2Y₁₂, P2Y₁₃, P2Y₁₄) of adenyl cyclase. ²  

In the ocular lens, functional studies have suggested the expression of P2Y-like receptors in sheep lens epithelial cells, ^5,6 sheep lentoids, ⁷ human lens epithelial cells, ^8–10 and rabbit whole lenses. ¹¹ Various roles for these P2Y receptors have been suggested in the lens, including the control of gap junction coupling in the epithelium, ^6,12 regulation of cell division and growth, ¹³ Ca²⁺-dependent modulation of K⁺ channels, ¹⁴ and modulation of Na⁺-K⁺ pump activity. ¹¹ Recently in the rat lens, we have identified at the molecular level the differentiation-dependent expression not only of P2Y isoforms (P2Y_1,2,4,6) ¹⁵ but also of P2X isoforms (P2X_1–4,6,7). ¹⁶ The molecular expression of P2 receptors identified to date in the rat lens is summarized in Table 1. In rat lens epithelial cells, the predominantly cytoplasmic localization of expressed P2 receptors, together with minimal Ca²⁺ mobilization in response to exogenous ATP, suggested P2 receptors are inactive. ^15,16 In cortical fiber cells, P2Y receptors, in particular P2Y₁ and P2Y₂, appeared to be functionally expressed, as indicated by the mobilization of intracellular Ca²⁺ in response to agonist application. ¹⁵ In contrast, P2X receptors appeared to be predominantly located in the cytoplasm of cortical fiber cells, and we have hypothesized that they act as a functionally inactive reserve pool of receptors. ¹⁶ To test this proposal, we subsequently demonstrated that the subcellular localizations of P2X_1,4,6 isoforms could be induced to associate with the plasma membrane of fiber cells in a specific zone of the outer cortex of the lens when rat lenses were under either osmotic or hyperglycemic stress. ¹⁷ Although we have yet to show that this membrane relocation results in functional activation of the P2X receptors, our results suggest that recruitment of P2X receptors from a cytoplasmic pool to the membrane in response to osmotic stress may play a role in regulating lens volume and ultimately the maintenance of lens transparency. 

Table 1.

 

Summary of Purinergic Receptor Expression and Their Predominant Localization in the Rat Lens ^15,16

Isoform	mRNA Expression (Total)	Expression in Equatorial Epithelial Cells	Expression in Cortical Fiber Cells	Expression in Mature Fiber Cells
P2Y1	✓	—	Membranous	—
P2Y2	✓	—	Membranous	—
P2Y4	✓	Cytoplasmic	Cytoplasmic	—
P2Y6	✓	Cytoplasmic	Cytoplasmic	—
P2Y12	Not expressed	N/A	N/A	N/A
P2Y13	Not expressed	N/A	N/A	N/A
P2Y14	Not expressed	N/A	N/A	N/A
P2X1	✓	Cytoplasmic	Cytoplasmic	—
P2X2 *	✓	Membranous	Membranous/cytoplasmic	—
P2X3	✓	Cytoplasmic	Cytoplasmic	Membranous
P2X4	✓	Cytoplasmic	Cytoplasmic	Membranous
P2X5	✓	Cytoplasmic	Not determined	Not determined
P2X6	✓	Cytoplasmic	Cytoplasmic	Membranous
P2X7	✓	Cytoplasmic	Cytoplasmic	—

*  Expression of P2X₂ was associated with the apical-apical interface between equatorial epithelial and superficial cells; in other cortical cells, expression was minimal and predominantly cytoplasmic.

A rodent model of diabetic cataract is characterized by liquefaction of cells located in a distinct zone of the outer cortex that starts as a localized swelling of cells and gradually progresses into cell column disruption and tissue liquefaction. ¹⁸ Our laboratory has been able to mimic the zone-specific damage to cortical fiber cells observed in diabetic cataract by organ culturing lenses under isotonic conditions in the presence of pharmacologic blockers of chloride transport. ^19,20 By using a variety of pharmacologic reagents specific for different transport proteins, in combination with electrophysiological and molecular localization studies, our laboratory has produced a molecular model of steady state volume regulation in the rat lens mediated by localized Cl⁻ ion influx and efflux within the lens cortex. ²¹ The membrane insertion of P2X_1,4,6 in response to osmotic stress and hyperglycemic stress occurs specifically in this influx zone, suggesting that P2X channels may enhance ion influx in this zone in response to changes in cell volume. Furthermore, because P2X channels are permeable to Ca²⁺, their overactivation in this zone could trigger the activation of Ca²⁺-dependent proteases that mediate the localized tissue liquefaction observed in diabetic cataract. ¹⁸  

To further investigate whether P2X receptors are involved in modulating fiber cell volume, we have used in this study our functional imaging approach to investigate the effects of a variety of P2 receptor agonists and antagonists on fiber cell morphology in rat lenses organ cultured under either isotonic or hypertonic conditions. Our results show that modulation of P2Y and P2X receptors has dramatically different effects on fiber cell morphology. P2Y₁ receptors modulate ion uptake associated with fiber cell elongation in peripheral fiber cells outside the influx zone, but P2X receptor agonists have no effect on fiber cell morphology in any region of the lens under isotonic conditions, a result consistent with their predominantly cytoplasmic location under these culture conditions. However, after hypertonic challenge to induce P2X insertion, the subsequent addition of P2X agonists caused a localized zone of cell swelling, specifically in the influx zone, implicating the overactivation of P2X receptors in the etiology of diabetic cataract. 

Materials and Methods

Reagents

The C-terminal specific P2X₁ and P2X₄ affinity-purified polyclonal antibodies and their corresponding control peptides were all purchased from Alomone Laboratories Ltd. (Jerusalem, Israel). The F-actin marker phallotoxin-Alexa Fluor 488, the goat anti-rabbit Alexa Fluor 488 secondary antibody, and the membrane marker wheat germ agglutinin conjugated to tetramethyl rhodamine isothiocyanate (WGA-TRITC) were all purchased from Molecular Probes Inc. (Eugene, OR). The P2 receptor agonists and antagonists adenosine 5′-[γ-thio]triphosphate (ATPγS), adenosine 5′-[β-thio]diphosphate (ADPβS), α,β-methyleneadenosine 5′-triphosphate (α,β-methylATP), pyridoxal-phosphate-6-azophenyl-2′, 4′-disulfonate (PPADS), N⁶-methyl-2′-deoxyadenosine-3′,5′-bisphosphate (MRS2179), and pyridoxal-α⁵-phosphate-6-phenylazo-4′-carboxylic acid (MRS2159) were all purchased as dilithium or trilithium salts from Sigma-Aldrich (St. Louis, MO) and were dissolved in distilled water to a final stock concentration of 20 mM. Phosphate-buffered saline was prepared from PBS tablets (Sigma-Aldrich Inc). Unless otherwise stated, all other chemicals were purchased from Sigma-Aldrich. 

Animals

All studies were approved by the University of Auckland Animal Ethics Committee and were undertaken in adherence to the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. Three to 4-week-old Wistar rats of either sex were killed by CO₂ asphyxiation followed by cervical dislocation. Whole eyes were removed immediately after death, and lenses were extracted through incisions on the posterior sclera. Lenses were immediately processed for incubation experiments or were fixed for immunohistochemistry. 

Lens Organ Culturing

Dissected lenses were immediately placed in a 24-well culture plate containing 2 mL prewarmed Artificial aqueous humor (AAH; 125 mM NaCl, 0.5 mM MgCl₂, 4.5 mM KCl, 10 mM NaHCO₃, 2 mM CaCl₂, 5 mM glucose, 20 mM sucrose, 10 mM HEPES, pH 7.2–7.4, 300 ± 5 mOsm) per well and were incubated at 37°C. After 1 hour, lenses that became cloudy were discarded. Lenses that remained clear after 1 hour were carefully transferred to wells containing 2 mL AAH containing 1% penicillin-streptomycin-neomycin and supplemented with or without pharmacologic modulators of P2 receptors. To induce hypertonic cell shrinkage, lenses were incubated in hypertonic AAH (200 mM NaCl, 0.5 mM MgCl₂, 4.5 mM KCl, 10 mM NaHCO₃, 2 mM CaCl₂, 5 mM glucose, 20 mM sucrose, 10 mM HEPES, pH 7.2∼7.4, 425 ± 5 mOsm) + 1% penicillin-streptomycin-neomycin. To induce hypertonic cell shrinkage and examine the effects of α,β-methylATP, lenses were first cultured in hypertonic AAH for 6 hours before the addition of this agonist. In all in culturing experiments, lenses were incubated at 37°C for a total period of 18 hours before they were fixed for immunohistochemistry and morphologic analysis. 

Immunohistochemistry

Whole lenses were fixed in 0.75% paraformaldehyde in PBS (final pH 7.0–7.5, adjusted by 2 M NaOH) for 24 hours at room temperature, cryoprotected, and cryosectioned according to protocols previously established in our laboratory. ²² Cryosections were washed three times in PBS. For morphologic analysis, cryosections were incubated for 1 hour at room temperature with WGA-TRITC (1:100 in PBS) to label cell membranes. To examine P2X_1,4 receptor localization, cryosections were incubated for 1 hour at room temperature in blocking solution (3% wt/vol bovine serum albumin, 3% vol/vol normal goat serum in PBS) to minimize nonspecific binding of antibodies. Sections were then incubated in primary antibody (P2X_1,4, 1:200) overnight at 4°C. Sections were washed three times in PBS and then were incubated in goat anti-rabbit Alexa 488 secondary antibody (1:400) for 3 hours at room temperature. Control sections were prepared by omitting the primary antibody. Sections were washed three times in PBS followed by 1-hour incubation in WGA-TRITC (1:100 in PBS). To label for F-actin, sections were incubated for 1 hour at room temperature with phallotoxin-Alexa488 (1:50) and WGA-TRITC (1:100) in 1% BSA + PBS. All sections were then extensively washed in PBS and mounted in antifade reagent (AF100; Citifluor Ltd., Leicester, UK). All images were captured using either a confocal laser scanning microscope (TCS2; Leica Microsystems, Wetzlar) or a confocal microscope (FV1000; Olympus Corp., Tokyo, Japan) and sequential scanning to individually acquire signals from WGA-TRITC and Alexa 488. Images were obtained at a resolution of 0.39 to 0.19 μm/pixel for all lower resolution images and at 0.078 μm/pixel for all higher resolution images. Raw images were processed using graphics editing software (CS3; Adobe Photoshop, Mountain View, CA) to combine separate channels and to remove noise (median filter radius = 1 pixel). 

Results

To investigate the role of P2 receptors in the lens and to visualize the spatially distinct activities of these receptors in the lens cortex, we cultured rat lenses for 18 hours in the presence of various pharmacologic P2 receptor antagonists and agonists. Because the interpretation of our results is critically dependent on knowledge of the relative potencies of the different pharmacologic reagents for the different P2 receptor isoforms, this information is briefly reviewed in Table 2. PPADS is a nonselective, but nonuniversal P2 receptor antagonist that exhibits minimal cross-reactivity with other G-protein–coupled receptors at concentrations <100 μM. PPADS has a high potency at P2Y₁, P2X₁, P2X₂, P2X₃, and P2X₅ receptors and a weak effect on P2X₆ and P2X₇ receptors, and it is practically ineffective for P2Y₂, P2Y₄, P2Y₆, and P2X₄ receptors. ^23,24 MRS2179 is a PPADS derivative with particularly high potency for P2Y₁ and a weaker potency for P2X₁ and P2X₃ receptors ^24,25,42 and is, therefore, commonly used to investigate the function of the P2Y₁ receptor isoform. ²³ In contrast, another synthetic PPADS analogue, MRS2159, has a high potency for P2X₁ compared with other P2 receptors, and has no effect on the P2Y₁ receptor isoform. ²⁶ By using these three antagonists, we will be able to distinguish among generic P2 receptor inhibition (PPADS), P2Y₁ inhibition (MRS2179), and P2X₁ inhibition (MRS2159). 

Table 2.

 

Summary of Potency Profiles for Modulators of P2 Receptors^23–41

Reagents	P2Y				P2X
	P2Y1	P2Y2	P2Y4	P2Y6	P2X1	P2X2	P2X3	P2X4	P2X5	P2X6	P2X7
Antagonists
PPADS	+++	−	−	*1	+++	++	+++	−	+++	+	+
MRS2179	+++	−	−	−	++	−	++	−	ND	ND	ND
MRS2159	−	ND	ND	ND	++++	+	+++	ND	ND	ND	ND
Agonists
ATPγS	++	++	++	−	+++	++	++	++	+++	++	−
ADPβS	++	ND*	ND*	+*2	ND*	ND*	ND*	ND*	ND*	ND*	ND*
α,β,-meATP	−	−	−	−	+++	−	++	+	+++*3	+++	−

 The plus signs indicate relative potency of the pharmacologic reagent for a P2 receptor isoform based on pEC50 of recombinant receptors. +, pEC50 between 4.0 and approximately 4.9; ++, pEC50 between 5.0 and approximately 5.9; +++, pEC50 between 6.0 and approximately 6.9; ++++, pEC50 >7.0; −, no measurable activity or pEC50 < 4.0.

*  1, partial (≈69%) inhibition at 100 μM; *2, partial (≈50%) agonist; *3, partial (≈50%) agonist in rat, inactive in other species; ND, not determined; ND*, agonist potency of ADPβS is not determined in most P2 receptor isoforms. However, the physiological analogue ADP is ineffective on these receptors.

In contrast to antagonists, the development of agonists that have isoform specificity and that are metabolically stable has been challenging ²⁴ ; therefore, the commercial availability of P2 receptor agonists is limited. ATPγS is a synthetic ATP analogue that can activate P2Y₁, ²⁷ P2Y₂, ²⁸ and many of the P2X isoforms. ³⁰ ADPβS is a synthetic analogue to ADP that strongly activates P2Y₁ receptors. ²⁷ α,β-methylATP is an ATP analogue that is generally regarded as a P2X receptor agonist. ^30,31 However, within the P2X family, the potency of this agonist varies between isoforms. For example, it potently activates P2X₁, P2X₃, and P2X₆ receptors but is relatively ineffective on P2X₂, P2X₄, and P2X₇ receptors. As a result, α,β,-methylATP is often used to pharmacologically distinguish these two groups of P2X isoforms. ^3,29,30 By using these three agonists, we will be able to distinguish among generic P2 receptor activation (ATPγS), P2Y₁ activation (ADPβS), and P2X activation (α,β,-methylATP). 

Effect of P2 Antagonists on Lens Morphology under Isotonic Conditions

To determine the relative roles of P2 receptors under isotonic conditions, the effects of three P2 receptor antagonists—PPADS, MRS2179, and MRS2159—on cell morphology were first compared (Fig. 1). Equatorial cryosections obtained from lenses cultured in isotonic AAH for 18 hours maintained their regular cellular architecture (Fig. 1A). In contrast, equatorial sections from lenses cultured in the presence of the broad-range P2 receptor inhibitor, PPADS, resulted in a zone of extracellular space dilations approximately 150 μm from the lens capsule (Fig. 1B). The damage to fiber cell morphology induced by PPADS was identical, in terms of form and location, to that induced by the Cl⁻ channel and the NKCC1 inhibitors NPPB and bumetanide, respectively, indicating that the general P2 agonist, PPADS, was blocking ion influx in this region of the lens. ^19,21 To further identify the P2 receptors responsible for blocking ion influx in this zone, lenses were incubated in either MRS2159 (P2X₁ > P2X₃; Fig. 1C) or MRS2179 (P2Y₁ > P2X₁ > P2X₃; Fig. 1D). Equatorial sections from lenses cultured in the presence of MRS2159 showed normal cell morphology that was maintained at even higher concentrations of the drug up to 80 μM (Fig. 1C). The lack of effect of MRS2159 on lens morphology suggests that P2X₁ or P2X₃ does not contribute to lens volume regulation under isotonic conditions, a result consistent with the predominantly cytoplasmic localization of P2X receptors previously observed in the lens cortex. ¹⁶  

Figure 1.

 

Differential effects of P2 receptor antagonists on cell morphology. Equatorial cryosections were obtained from lenses incubated for 18 hours in AAH in the presence or absence of P2 antagonists and were labeled with the membrane marker WGA. (A) Control lenses incubated with AAH showed no signs of morphologic damage. (B) Lenses incubated with 5 μM PPADS exhibited a zone of extracellular space dilation (inset) localized approximately 150 μm from the lens capsule. (C) Lenses incubated with 80 μM P2X₁ inhibitor MRS2159 showed no damage. (D) Lenses incubated with 5 μM P2Y₁ inhibitor MRS2179 resulted in extracellular space dilations between peripheral fiber cells (inset). Representative images from at least six lenses with each pharmacologic reagent.

In contrast, MRS2179 (Fig. 1D) induced extracellular space dilations similar to those seen after the incubation of lenses in PPADS. Despite the similar effects in terms of damage phenotype and dose dependence, the actual location of the extracellular space dilations induced by PPADS and MRS2179 were dramatically different. This is most evident in axial cryosections (Fig. 2). In axial sections from lenses incubated in AAH, we see the ordered arrangement of fiber cells organized in column-like arrays (Figs. 2A, 2D). However, in the presence of PPADS, we can see that damage is induced in the deep cortex that extends as a distinctive band beyond the equator toward both poles (Fig. 2B, solid arrows). On closer examination, it can be seen that the damage is a result of extracellular dilations formed between fiber cells (Fig. 2E). In contrast, the damage induced by MRS2179 was predominantly confined to peripheral fiber cells near the modiolus (Fig. 2C). At higher magnification, it was evident that, like PPADS, MRS2179 induced extracellular dilations between the membranes of adjacent fiber cells (Fig. 2F). Given that MRS2159 (P2X₁ > P2X₃) did not have any effect under isotonic conditions, the only other isoform that could be responsible for the effect of MRS2179 (P2Y₁ > P2X₁ > P2X₃) is P2Y₁. Taken together, these results show that two distinct P2 receptor antagonists are capable of blocking ion uptake in two spatially different regions in the lens cortex but that P2Y₁ receptors are not directly involved in mediating ion uptake blocked by PPADS in the deeper influx zone. 

Figure 2.

 

Spatial locations of morphologic damage induced by P2 antagonists. (A–C) Bright-field overview shots of axial sections taken from lenses incubated with AAH alone (A), AAH + 10 μM PPADS (B), or AAH + 10 μM MRS2179 (C). (D–F) Axial cryosections were labeled with the membrane marker WGA, and high-magnification images were taken by confocal microscopy to reveal cell morphology in the indicated areas (box and inset) after incubation in AAH (D), AHH + 10 μM PPADS (E), and AHH + 10 μM MRS2179 (F). Representative images from at least four lenses in each condition.

Effects of P2 Agonists on Lens Morphology under Isotonic Conditions

To attempt to further identify the P2 receptors responsible for the damage observed in the ion influx zone, lenses were organ cultured under isotonic conditions in the presence of the P2 agonists ATPγS, ADPβS, and α,β-methylATP (Fig. 3). Incubation of lenses with ATPγS (Figs. 3A–D) and ADPβS (Figs. 3E–H) again induced cell damage in a localized ion influx zone found approximately 150 μm from the lens capsule. However, the damage phenotype induced by both ATPγS and ADPβS in this zone was completely different from that induced by PPADS (Fig. 1B) or by NPPB and bumetanide in previous studies. ^19,21 This damage was dose dependent and manifested initially as a loss of the hexagonal cross-sectional profile of cell membranes at low concentrations (80 μM–160 μM; Figs. 3A, 3B, 3E, 3F) and progressed to complete deformation of cellular architecture at higher concentrations (320 μM; Figs. 3C, 3D, 3G, 3H). In contrast, exposure of lenses to α,β-methylATP for 18 hours had no effect on lens morphology, even at higher concentrations (Figs. 3I–L). 

Figure 3.

 

Effect of various P2 receptor agonists on lens morphology. Equatorial cryosections taken from lenses incubated for 18 hours in the presence of different concentrations of P2 receptor agonists ATPγS (A–D), ADPβS (E–H), or α,β,methyl-ATP (I–L). Sections were labeled with the membrane marker WGA and were imaged using confocal microscopy. Agonists concentrations were at 80 μM (A, E, I), 160 μM (B, F, J), and 320 μM (C, G, K). High-power images of cell morphology in the areas indicated by the boxes in (C), (G), and (K) are shown in (D), (H), and (L) for lenses incubated in 320 μM of each agonist. Representative images from at least four lenses at 80 and 160 μM and eight lenses at 320 μM.

This lack of effect of α,β-methylATP on fiber cell morphology is consistent with the predominantly cytoplasmic location of P2X receptors in this area of the lens (Figs. 4A, 4C). Because the subcellular location of selected P2X receptors can be changed from the cytoplasm to the membrane by exposure to osmotic or hyperglycemic stress, ¹⁷ we wanted to determine whether these P2X receptors could now be activated by α,β-methylATP. To achieve this, we chose to use hypertonic stress to induce the recruitment of P2X₁ and P2X₄ receptors to the membrane because this manipulation has minimal impact on overall lens tissue architecture, though a slight compaction of fiber cells located approximately 100 μm from the lens periphery consistent with cell shrinkage was evident (Fig. 4B). After the 6-hour preincubation of lenses in hypertonic AAH to induce P2X₁ and P2X₄ membrane insertion, α,β-methylATP was applied to the lenses for 12 hours to complete the 18-hour culture period. It is now clearly evident that after membrane insertion of P2X₁ and P2X₄ induced by hypertonic stress, α,β-methylATP induces morphologic damage to the cortical influx zone (Fig. 4D). Interestingly, in this α,β-methylATP-induced damage zone, labeling for both P2X₁ and P2X₄ is dramatically reduced, potentially indicating that binding of the antibody to its epitope has been reduced as a downstream consequence of P2X channel activation. 

Figure 4.

 

Effect of α,β-methylATP under isotonic and hypertonic conditions. Cryosections were taken from lenses incubated for 18 hours in isotonic AAH (A), hypertonic AAH (B), isotonic AAH + 320 μM α,β-methylATP (C), and hypertonic AAH + 320 μM α,β-methylATP (D). Insets: magnified images taken from the nucleated zone (A–C) and damage zone (D) were doubled-labeled with either anti-P2X₁ or -P2X₄ antibody (green) and the membrane marker WGA (red).

The rat lens is known to contain Ca²⁺-dependent proteases such as calpain, which, when activated by the elevation of intracellular Ca²⁺ levels, have been shown to mediate the proteolysis of cytoskeletal components such as vimentin, filensin, and F-actin in fiber cells. ^43–45 Given that P2X receptors are permeable to Ca²⁺, it is possible that the damage induced by α,β-methylATP under hypertonic conditions may be the result of Ca²⁺ influx through the recruited P2X₁ and P2X₄ receptors and the subsequent activation of Ca²⁺-dependent proteases. Such activation of Ca²⁺-dependent proteases and degradation of P2X proteins may account for the loss of P2X₁ and P2X₄ labeling observed in Figure 4. To test this proposal, rat lenses were incubated in hypertonic media in the absence (Fig. 5A) and presence (Fig. 5B) of α,β-methylATP and were labeled with phalloidin, a marker for F-actin. Hypertonic stress alone did not induce appreciable change to the expression of actin, which was associated predominantly with the narrow side domains of fiber cell membranes in the outer cortex (Fig. 5A). However, when hypertonic-cultured lenses were incubated in the presence of α,β-methylATP, an abrupt change in the actin labeling pattern was observed that was restricted to the influx zone (Fig. 5B). Under higher magnification, it is apparent that this change in actin labeling is associated with a change from strong actin labeling of the narrow side membranes to a more diffuse cytoplasm labeling pattern (Fig. 5C), a result consistent with actin degradation by localized protease activation. 

Figure 5.

 

Effect of α,βmeATP under hypertonic conditions on the localization of F-actin. Equatorial cryosections were taken from lenses incubated for 18 hours in hypertonic AAH (A) or in hypertonic AAH + 320 μM α,βmeATP (B, C). Sections were labeled with phalloidin-Alexa 488 to label F-actin. Higher magnification image (C) highlights the abrupt change in actin labeling. Arrowhead: initiation of the damage zone located approximately 150 to 200 μm from the lens periphery. Representative images from three lenses treated with α,βmeATP.

Comparison of the damage phenotypes induced by α,β-methylATP under hypertonic conditions and by ATPγS under isotonic conditions tends to indicate that the two ATP agonists have different morphologic effects (Fig. 6). In addition to changes in actin labeling (Fig. 6A), the damage induced by α,β-methylATP under hypertonic conditions included rupturing of cell membranes (Fig. 6B), a result consistent with the activation of cellular proteases. In contrast, morphologic damage induced by ATPγS was not associated with such actin redistribution because the labeling, though distorted, remained associated with the narrow sides of the fiber cells (Fig. 6C) and showed a subtler deformation of fiber cell membranes (Fig. 6D) than cell swelling and membrane rupture. This comparison tends to indicate the effects of α,β-methylATP, and ATPγS (and, by analogy, ADPβS) are mediated by different P2 receptor pathways. 

Figure 6.

 

Effect of α,β-methylATP under hypertonic conditions and ATPγS under isotonic conditions on the fiber cell membrane and the localization of cytoskeletal actin. Equatorial cryosections were taken from lenses incubated for 18 hours in hypertonic AAH + 320 μM α,β-methylATP (A, B) or in isotonic AAH + 320 μM ATPγS (C, D). Sections were double-labeled with F-actin marker phalloidin-Alexa 488 to label F-actin (A, C) and the membrane marker WGA (B, D), and images were taken at the same location. Arrowheads: initiation of this damage zone located approximately 150 to 200 μm from the lens periphery (A, B), showing that F-actin became cytoplasmic a few fiber cell layers before the liquefaction of fiber cells (A, B).

Discussion

In this study, we have investigated whether P2X receptors are active in the lens by monitoring the effects pharmacologic modulators of P2Y and P2X have on fiber cell morphology in organ-cultured rat lenses. Our results indicate roles for P2Y₁ receptors in peripheral fiber cells under isotonic conditions that are consistent with the functional activation of P2Y receptors and subsequent mobilization of intracellular calcium observed in earlier studies in sheep ⁵ and human ^8,10 lens epithelia and rat fiber cells. ¹⁵ In contrast, P2X receptors appear to play only limited roles in the lens under steady state conditions, a result consistent with their predominantly cytoplasmic subcellular location detected previously by immunohistochemistry. ¹⁶ However, in response to hypertonic stress, P2X receptors are recruited to the plasma membrane, ¹⁷ where they were then activated by the addition of the P2X-specific agonist α,β-methylATP (Fig. 4). After hypertonic stress and subsequent P2X_1/4 membrane insertion, α,β-methylATP induced a localized zone of cell swelling and membrane rupture that altered the subcellular distribution of the cytoskeletal protein actin, presumably through Ca²⁺ influx and the subsequent activation of Ca²⁺-dependent proteases (Fig. 6). These observations and the conclusions arising from them are associated with a number of caveats that must first be acknowledged before discussing the potential significance of our results to lens volume regulation and the etiology of diabetic cataract. 

The results of our experiments are critically dependent on the penetration properties, purity, and specificity of the pharmacologic agents used and the effects of agonist application on receptor desensitization and internalization. Since the major effect of all reagents was the induction of a damage zone located approximately 150 μm into the lens, we can safely assume that all pharmacologic reagents diffused into the lens and reached receptors expressed on fiber cells located at least in the cortical region of the lens. With regard to agonist purity, the synthetic analogues of nucleotides used in this study are more stable than naturally occurring nucleotides, but they could have been degraded slowly by ectonucleotidases or they could have contained contaminating nucleotides ²⁴ that might have affected the interpretation of our results. However, the differential effects of the two synthetic ATP analogues, ATPγS and αβmeATP, suggest their degradation or contamination by ATP was unlikely. Finally, because the lens expresses a multitude of P2Y and P2X receptor isoforms, it is difficult to definitively attribute an observed damage phenotype to a particular isoform because of the lack of isoform-specific agonists and antagonists. This is compounded by rapid isoform-specific desensitization and subsequent receptor internalization that can affect P2 receptor expression after agonist application. ^46,47  

Despite the potential caveats associated with the pharmacologic reagents available to us, our results show that the application of general P2 antagonists and agonists all produced damage to fiber cells in a discrete localized zone of the outer cortex. In a series of previous studies, ^19–21 we have shown that a variety of ion channels and transporters mediate ion uptake in this zone and that their inhibition causes ions and water to accumulate in the extracellular space, causing it to swell or dilate. PPADS induced an identical zone of extracellular space dilations, suggesting that it also blocked ion uptake in this area of the lens. Based on an analysis of the expression patterns of the different P2 receptors (Table 1) and the relative potency of PPADS (Table 2), it is most likely that the observed morphologic effects of this general P2 inhibitor can be attributed to either direct blockade of ion uptake mediated by a number of P2X channel isoforms (P2X₁, P2X₃, or P2X₅; P2X₂ is expressed only at the apical-apical interface) or to indirect inhibition of a P2Y₁ signaling pathway that regulates downstream ion uptake mediated by Cl⁻ channels or transporters. The lack of effect of MRS2179 and MRS2159 in the influx zone tends to rule out P2Y₁ and P2X₁ receptors, respectively, as mediators of the extracellular space dilations induced by PPADS in this area of the lens. Therefore, PPADS probably blocks a small subpopulation of P2X₃ and P2X₅ receptors that are membrane resident and normally respond to the constitutive release of ATP from fiber cells into the extracellular space (Suzuki-Kerr H, unpublished data, 2009) to promote ion uptake. The blockade of this basal P2X uptake pathway by PPADS would cause ions to accumulate between fiber cells, inducing osmotic swelling of the normally tight extracellular space. At first glance, this conclusion is somewhat at variance with the absence of fiber cell damage observed after application of the P2X receptor agonist α,β-methylATP under isotonic conditions. However, if the P2X₃ receptor isoform was responsible for this basal P2X-dependent ion uptake pathway, it would offer an explanation for the lack of effect of α,β-methylATP under isotonic conditions. P2X₃ rapidly desensitizes on application of agonists, ⁴ and it is possible that the prolonged application of agonist renders this isoform insensitive to α,β-methylATP, preventing any detectable effects on fiber cell morphology. If this is indeed the case, the differential effects of α,β-methylATP on fiber cell morphology in lenses organ cultured in isotonic and hypertonic conditions could be attributed not only to an increased recruitment of additional P2X receptors to the plasma membrane but also to a change in P2X properties from a P2X₃ receptor that undergoes rapid desensitization to P2X_1,4 or P2X₄ channels that are less prone to desensitization. 

The lack of an effect of α,β-methylATP on fiber cell morphology under isotonic conditions is consistent with the fact that most P2X receptors in the outer cortex are predominantly located in the cytoplasm as a pool of inactive receptors that can be recruited to different membrane domains in an isoform-specific manner under conditions of osmotic and hyperglycemic stress. ¹⁷ In this study we now show that, at least for hypertonic solutions, the P2X_1,4 receptor recruitment to the membrane is functionally active, as indicated by the ability of α,β-methylATP to induce fiber cell swelling. Although the damage induced by α,β-methylATP under hypertonic conditions occurs in the same localized band of fiber cells that exhibit morphologic damage in response to the addition of ATPγS or ADPβS, under isotonic conditions, the damage phenotypes were different, indicating that the different agonists activated different pathways. Indeed the ability of ADPβS to induce damage suggests that under isotonic conditions, P2Y receptors are activated. Considering that P2Y₁ and P2Y₂ are both localized to the plasma membranes of fiber cells throughout the whole of the lens cortex, the application of ATPγS or ADPβS would be expected to stimulate P2Y receptors throughout the whole region, not the just the observed localized zone. It seems likely, therefore, that the localized membrane deformation observed for both ATPγS and ADPβS may reflect a secondary effect of overstimulating P2Y signaling pathways, presumably by mobilization of intracellular Ca²⁺ stores, that result in fiber cells losing their ordered cellular structure specifically in this influx zone. If this deformation of fiber cell membranes was in fact caused by the mobilization of intracellular Ca²⁺, then Ca²⁺ increase does not appear to be sufficient to induce the degradation of cytoskeletal actin by Ca²⁺-dependent proteolysis (Fig. 6). In contrast, the P2X agonist α,β-methylATP caused cell swelling, membrane rupture, and degradation of cytoskeletal proteins in the same zone of the lens after the membrane insertion of P2X₁ and P2X₄ in response to hypertonic challenge. 

The predominant effect of MRS2179 was to induce extracellular space dilations between peripheral fiber cells of the ion efflux zone located near the modiolus, the region in the lens where equatorial epithelial cells initiate the elongation that marks their differentiation into fiber cells. ⁴⁸ In cultured retinal pigment epithelial cells, it has been shown that βFGF, a well known initiator of fiber cell differentiation, ⁴⁹ stimulates the release of ATP, ⁵⁰ raising the possibility that growth factors such as βFGF trigger ATP release and the subsequent activation of P2Y₁ signaling pathways that stimulate the ion influx and volume expansion necessary to drive fiber cell elongation. This observation is peripheral to the present study but may warrant future investigation. 

It is apparent that P2Y receptors are functionally active in the cortex of the rat lens and that, in addition to other roles, they may play a role in regulating ion influx, particularly in elongating fiber cells. P2X receptors appear to play a role in mediating basal ion uptake in the influx zone. Osmotic and hyperglycemic stress can significantly increase this basal level of P2X activity through the recruitment of other P2X receptor isoforms from an inactive cytoplasmic pool, ¹⁷ presumably to regulate the volume of fiber cells and to prevent extracellular space swelling. However, if these recruited P2X receptors are inappropriately activated (as occurred in this study by application of α,β-methylATP after hypertonic stress), they can cause cellular damage that would compromise lens transparency. In this regard, it is interesting to speculate that inappropriate activation of these recruited P2X receptors may be involved in the progression of diabetic cataract because, in a diabetic rat model, lens fiber cell damage starts as a discrete zone of cell swelling in the lens outer cortex located in the same influx zone as identified in this and other studies that, over time, expands into a more extensive region of tissue liquefaction. ¹⁸  

The progression of diabetic cataract is also associated with depolarization of the membrane potential and elevation of intracellular Na⁺ and Ca²⁺ concentrations. ⁵¹ This accumulation of Ca²⁺ in cataractous lenses is thought to arise from increased membrane permeability to cations rather than loss of active transport, ^52,53 suggesting activation of a cation leak pathway in the later stages of cortical cataractogenesis. Our results indicate that P2X receptors could be the cation leak pathway activated in cortical cataract. Because of their permeability properties, activation of P2X receptor channels would cause not only the membrane depolarization associated with cortical cataract but also the Ca²⁺ overload and subsequent proteolytic cytoskeletal degradation. Furthermore, the recruitment of P2X receptors, specifically in the influx zone, offers an explanation for why the membrane damage observed in diabetic cortical cataract is localized to a distinct zone of the outer cortex. Although further studies are required to test this hypothesized involvement of P2X receptors in the formation of diabetic cataract, our results have identified a potential target for the development of novel therapies to slow the progression of diabetic cortical cataract. 

Footnotes

 Supported by the Marsden Fund of New Zealand and the University of Auckland Research Committee. HSK was a holder of a Tertiary Education Commission of New Zealand Bright Futures Top Achiever Doctoral Scholarship. JCL is the holder of a Foundation for Research, Science and Technology Postdoctoral Fellowship.

Footnotes

 Disclosure: H. Suzuki-Kerr, None; J.C. Lim, None; P.J. Donaldson, None