- Email: [email protected]
Pharmacological modulation of fluid secretion in the pigmented rabbit conjunctiva
Life .%imca,
PI1 SOO24-3205@9)00638-4
ELSEVIER
Vol. 66. No. 7, Pp. PL LOS-1II,2000 cowligM02OMlElasviaSCiihC. Pri&diniheUSA. ASrightareacmd 00%32OYOO/ssee thmt matter
PfL4RM4COLoGYmm Accelerated Communicatbn
PHARMACOLOGICAL
MODULATION OF FLUID SECRETION IN THE PIGMENTED RABBIT CONJUNCTIVA
Michael H. I. Shiue,’ Ashutosh A. Kulkarni,’ Hovhannes J. Gukasyan,’ Jennifer B. Swisher,’ Kwang-Jin Kim,2 and Vincent H. L. Leelv3
Departments of ‘Pharmaceutical Sciences, ‘Ophthalmology, 2Medicine, 2Physiology and Biophysics, 2Molecular Pharmacology and Toxicology, and 2Biomedical Engineering and 2Will Rogers Institute Pulmonary Research Center, Schools of Pharmacy, Medicine, and Engineering, University of Southern California, Los Angeles, CA 90033 (Submitted July 6,1999; accepted August 16, 1999; received in final form September 22,1999)
Abstract. We determined net fluid secretion rate across the pigmented rabbit conjunctiva in the presence and absence of pharmacological agents known to affect active Cl secretion and Na+ absorption. Fluid flow across a freshly excised pigmented rabbit conjunctiva mounted between two Lucite half chambers was measured by a pair of capacitance probes in an enclosed cabinet maintained at 37 “C and a relative humidity of 70%. Fluid transport was also measured in the presence of compounds known to affect active Cl secretion (CAMP, UTP, and ouabain), Na+ absorption (D-glucose), or under the Cl-free condition on both sides of the tissue. Net fluid secretion rate across the pigmented rabbit conjunctiva in the serosal-to-mucosal direction at baseline was 4.3 + 0.2 pl/hr/cm2 (mean f s.e.m.). Net fluid secretion rate was increased approximately two-fold by mucosally applied 1 mM 8-Br CAMP (8.4 + 0.4 pl/hr/cm2) and 10 pM UTP (9.8 f 0.6 p.l/hr/cm2), but was abolished by either serosally applied 0.5 mM ouabain (0.3 k 0.1 pl/hr/cm2) or under the Cl-free conditions (0.06 f 0.04 pl/hr/cm2). Mucosal addition of 20 mM Dglucose decreased net fluid secretion rate to 1.0 If: 0.5 pl/hr/cm2. In conclusion, the pigmented rabbit conjunctiva appears to secrete fluid secondary to active Cl‘ secretion. This net fluid secretion is subject to modulation by changes in active Cl secretion rate and in mucosal fluid composition such as glucose concentration. 8 2000ElsevierScienceInc. Key Words: conjuuctiva, uTP,capacitanceprobe
net fluid
secretion,
active
~hlori&
secretion,
Na+-coupled
glucose
transport,
CAMP,
Introduction Transtissue fluid flow has been demonstrated in epithelia such as rabbit comeal epithelium (l), bovine retinal pigment epithelium (2), human tracheal gland cells (3), and rat intestinal epithelium (4). Typically, fluid secretion across epithelia is driven by active Cl secretion (1). The pigmented rabbit conjunctiva is predominantly a Cl-secreting tissue (5), subject to modulation by CAMP, Corresponding Author: Vincent H.L. Lee, Ph.D., University of Southern California, School of Pharmacy, Department of Pharmaceutical Sciences, 1985 Zonal Avenue, Los Angeles, CA 90033 USA. (Voice: 323-442-1368; FAX: 323-442-1390; E-mail: [email protected]).
PL-106
ConjunctivalFluid Secretion
Vol. 66, No. 7, 2000
Ca*+, protein kinase C (PKC), and purine and pyrimidine nucleotides (6,7). The short circuit current (1,J was stimulated 133% by 1 mM 8-Br CAMP, 107% by 10 l.tM A23187 (a Ca2+ ionophore), 87% by 1 pM phorbol lZmyristate-13-acetate (PMA) (6), and 180% by 10 pM UTP (7). The only report on conjunctival water permeation was a study done by Candia ef al. (8) investigating the effect of hypotonicity on the mucosal to serosal diffusional 3H-water flux across the excised albino rabbit conjunctiva. Thus, the purpose of the present study was to determine whether the pigmented rabbit conjunctiva indeed is capable of secreting fluid and whether such fluid secretion was coupled with active Cl secretion. 8-Br CAMP and UTP were chosen for study, since they stimulated active Cl- secretion the most in our previous studies (6,7). Materials and Methods Reagents: 8-bromoadenosine-3’,5’-cyclic monophosphate (8-Br CAMP), ouabain, D-glucose, Dmannitol, and uridine 5’-triphosphate (UTP) were purchased from Sigma Chemicals Co. (St. Louis, MO). Animals and tissue preparation: Male, Dutch-belted pigmented rabbits (2.5-3.0 kg) were purchased from Irish Farms (Norco, CA) and were treated in compliance with The Guiding Principles in the Care and Use of Animals (DI-IEW Publication, NM 80-23). The detailed procedure for conjunctival tissue preparation was reported previously (5). In brief, rabbits were euthanized with an overdose of sodium pentobarbital solution (325 mg/kg) via a marginal ear vein. Both eyeballs were excised, and the conjunctival tissues were trimmed for mounting as flat sheets between two Lucite half chambers for water flux measurements. Approximately the same region of the conjunctival tissue was used for each experiment. The mucosal and the serosal sides of the tissue (0.385 cm*) were each bathed in bicarbonated Ringer’s solution (BR). Regular BR contained 111.55 n&l NaCl, 4.82 mM KCI, 0.86 mM NaI-IzPOd, 29.20 mM NaHC03, 1.04 mM CaC12, 0.74 mM MgC12, and 5 mM D-glucose. The buffer pH was adjusted to 7.4 by bubbling with 5% CO2 in air. Cl free condition was established by replacing, on an equimolar basis, NaCI, KCl, CaC12, and MgCl2 with sodium isethionate, potassium isethionate, calcium gluconate, and magnesium gluconate, respectively. For glucose-rich solution, an additional 20 mM D-glucose was added to regular BR bathing solution on the mucosal side. To determine the possible osmotic effect of excess osmolytes, net fluid secretion rate was measured in the presence of mucosal 20 mM Dmannitol, in lieu of 20 mM D-glucose. Fluidflu measurements: Fluid flow across an excised pigmented rabbit conjunctiva was measured by a pair of capacitance probes (ASP-lo-CTAISP, Mechanical Technology, Inc., Latham, NY) similar to that described by Edelman and Miller (2). The entire apparatus was placed in an enclosed cabinet maintained at 37’C and at a relative humidity of 70%. Prior to each experiment, the capacitance probes were also equilibrated at 37°C inside the enclosed cabinet. In pilot experiments, we found that a relative humidity of at least 60% was required to prevent detectable evaporation of fluid during the course of an experiment. One mM 8-Br CAMP (mucosal), 10 pM UTP (mucosal), 20 mM D-glucose (mucosal), 20 mM D-mannitol (mucosal), or 0.5 mM ouabain (serosal) was applied to the reservoir using the concurrent feed and drain method with a pair of 10 pl Hamilton syringes joint at the end of the plungers. Instillation of a drug solution by one syringe was counterbalanced by the withdrawal of an equal volume of buffer from the inner chamber by the opposite syringe, thus keeping a constant fluid volume (3.5 ml on each side) during drug application in the reservoir. The concentration used was based on the I,, measurements reported previously (7,9,10). To demonstrate a role for active Cl secretion in transconjunctival fluid flow, water flux was measured under the Cl- free condition, whereby a Cl--free solution was used in both the serosal and mucosal reservoirs. Each tissue was treated with only one agent/condition. Except
Vol. 66, No. 7,200O
Conjunctival Fluid Secretion
PL-107
for the Cl-free condition, the baseline fluid flux was measured for 30 min across all tissues prior to the drug application. The fluid flux was then monitored for another 30 min from the time of drug instillation. The capacitance of the air gap between the probe and the fluid surface (less than 3 mm in distance) on each side of the chamber was measured. The voltage output from the capacitance probe indicates the loss of fluid from one side of the chamber, while gain of fluid on the opposite side. Under fluid secreting conditions, the voltage output from the mucosal probe would decrease, and a symmetrical increase in voltage output would be observed on the serosal side. The voltage output from each probe was equal but opposite. Non-symmetrical voltage changes were considered an artifact and not acquired. The fluid transport rate (pl/hr/cm*) was calculated based on the slope of the voltage vs. time and conversion of voltage changes to fluid volume displacement based on a calibration table provided by the manufacturer. Statistical analysis: Results were expressed as mean * s.e.m. Comparisons between the group means of two data sets were analyzed with unpaired Student’s t-test. One way analysis of variance and post-hoc comparisons based on Fisher’s least squared difference approach were used to determine statistical significance among group means of more than two data sets. A p value of less than or equal to 0.05 was considered significant.
A net serosal (s) to mucosal (m) fluid secretion rate (JV) of 4.3 f 0.2 pl/hr/cm* (n = 27) was observed in the freshly excised conjunctiva under baseline condition (Figs. 1 and 2). Mucosal application of 1 mM 8-Br CAMP approximately doubled the baseline fluid secretion (p < 0.05) to 8.4 rt 0.4 @/hr/cm* (n = 5) (Figs. 1 and 2). Moreover, the increase in conjunctival fluid secretion afforded by 8-Br CAMP was concentration-dependent for up to 3 mM range studied (Fig. 3). The observed maximal change in fluid secretion rate (AI,) was 5.1 k 0.3 pl/hr/cm* and the halfmaximal 8-Br CAMP concentration was 0.32 f 0.06 mM. This latter value is comparable to the previously reported effective half-maximal concentration (ECso) values for stimulating Isc (0.28 mM) and net Cl- secretion (0.30 mM) by 8-Br CAMP (6). By contrast, 10 pM UTP increased fluid secretion to 9.8 f 0.6 pl/hr/cm* (n = 6) (p c 0.05) (Fig. 2) transiently for 10 min (Fig. 1). The baseline J, was significantly (p < 0.05) reduced by serosally applied 0.5 mM ouabain to 0.3 + 0.1 pl/hr/cm* (n = 4), not different from zero (p > 0.05) (Figs. 1 and 2). In addition, net fluid transport was abolished (p < 0.05) when Cl was removed from both sides of the conjunctiva (0.06 + 0.04 p&r/cm*, n = 4) (Fig. 2). Twenty mM D-glucose was previously determined to induce a maximal increase in transconjunctival Isc(10). Mucosal addition of 20 mM D-glucose to BR significantly reduced (p < 0.05) the baseline J, by 77% to 1.0 f 0.5 pl/hr/cm* (n = 4) (Figs 1 and 2). By contrast, mucosally applied 20 mM D-mannitol (which elevated the osmolality by 22 f 1.2 mOsm/kg) did not affect the baseline J, significantly (p > 0.05). A good correlation (3 = 0.98) was observed between the changes in net fluid flux and those in IX induced by agents known to affect active conjunctival Clsecretion (i.e., excluding D-glucose) (Figd).
Conjunctid
PL-108
Vol. 66, No. 7,200O
Fluid Secretion
,
1 mM 8-Br CAMP 10 @l UTP
20 mM D-glucose 0.5mMOuabain
30 Time (min)
60
Fig. 1 Typical voltage vs. time tracings, demonstrating the effect of 1 mM S-Br CAMP, 10 @I UTP, 0.5 mM ouabain, and 20 mM D-glucose on transconjunctival fluid secretion. Agents were applied after 30 n-tin of baseline measurement (as indicated by the arrow). One volt (V) deflection is equivalent to 127 pm of fluid displacement according to the calibration table provided by the probe manufacturer (Mechanical Technology, Inc., Latham, NY). The positive slope of the linear segment denotes the transconjunctival fluid secretion rate.
12.OT
*
I *
-;
8.0--
g t G-
4.0--
t t
0.07
Ls
t
I
Fig. 2 Conjunctival fluid secretion rate at baseline (n = 35) and in response to mucosal 1 mM 8Br CAMP (n = 5), mucosal 10 pM UTP (n = 6), serosa10.5 mM ouabain (n = 4), mucosal 20 mM D-glucose (n = 4), mucosal 20 mM D-mannitol (n = 3), and U-free (both sides, n = 4) conditions. Error bars represent s.e.m. * indicates a significantly higher fluid flow (p < 0.05) than baseline. t indicates a significantly lower (p < 0.05) fluid flow than baseline.
Vol. 66, No. 7,200O
Con-
Fluid
Secretion
[&Br CAMP] (mM) Fig. 3 Changes in net transconjunctival fluid secretion rate as a function of mucosal 8-Br CAMP concentration. Error bars denote s.e.m. for 4-6 tissues.
6, 3 mM &Br CAMP 1 mlU 5-Br CAMP iu^ 5
0.5 mM &Br
2
0.1 mY EBr
5 a
CAMP
CAMP
O >
2 20 mM D-glucose
-6-1 -14.0
0.0
I 14.0
AL (Wcm2) Fig. 4 Relation between change in net fluid secretion rate (AJV”‘) and the corresponding a, under conditions known to affect active Cl- secretion and Na+ absorption. Error bars denote s.e.m. for 4-6 tissues. The & values for 8-Br CAMP (6), Cl--free (5), ouabain (5), D-glucose (lo), and UTP (7) were from our previous reports. The linear correlation (3 = 0.98) was determined only for conditions that affect active Cl- secretion.
PL-110
Conjunctival Fluid Secretion
Vol. 66, No. 7,200O
Discussion We have demonstrated that the pigmented rabbit conjunctiva secretes fluid at a baseline J, of 4.3 f 0.2 pVhr/cm’. This value is on the same order of magnitude reported for frog retinal pigment epithelium (3.9 pl/hr/cm2) (1 I), bovine lens epithelium (4.7 pl/hr/cm2) (12), and human tracheal gland cells (2 @/hr/cm*) (3). By contrast, the fluid secretion rate in the rabbit cornea is 0.2 @/hr/cm2 (l), only 5% of that in the pigmented rabbit conjunctiva. Conjunctival fluid secretion appears to be coupled to active Cl secretion. This is indicated by the abolishment of fluid secretion both in the presence of serosa10.5 mM ouabain and upon removal of Cl from the bathing solutions (Fig 2). It is, therefore, not surprising that compounds known to stimulate Cl- secretion in the conjunctiva, such as 8-Br CAMP (6) and UTP (7), also stimulate conjunctival fluid secretion (Figs. 1 and 2). Indeed, there exists an excellent correlation between the changes in fluid secretion and those in I,, in the conjunctiva under conditions that affect active Cl secretion (Fig. 4). The 128% enhancement of fluid secretion by 10 p.M UTP may be of potential therapeutic importance in the management of dry eye state, when CAMP-sensitive Clsecretion may have been compromised. In fact, Jiang ef al. (3) reported that human cystic fibrosis (CF) tracheal gland cells maintain normal fluid secretory response to UTP, despite impairment of the CAMP-dependent Cl secretion. Conjunctival fluid flow is also sensitive to conditions that affect active Na+ absorption. For instance, increasing mucosal D-glucose concentration from 5 to 25 mM reduced conjunctival fluid flow by 77% (Fig. 2), presumably secondary to stimulation of Na+ uptake via the Na+-coupled glucose transporter (10). The lack of significant influence by 20 mM D-mannitol on J, indicates that the osmotic contribution of 20 mM D-glucose on the observed J, was negligible. There may exist two possible pathways for conjunctival fluid flow: paracellular (13) and transcellular (14). Both pathways are probably utilized by fluid flow driven either by active Cl secretion or an osmotic gradient. Transcellular fluid flow, in addition, is probably mediated by aquaporins (AQP). Aquaporin type 3 (AQP3) has been demonstrated in the human and rat conjunctival epithelium, using immunoblot analysis of membrane proteins, high-resolution immunocytochemistry, and immunoelectron microscopy (15). The relative proportion of the tear volume attributable to the fluid secreted by the conjunctiva is not known. Given that the surface area of the rabbit conjunctiva is 13 cm2 and assuming that all of it participates in fluid secretion at the observed rate, the transconjunctival fluid secretion rate would amount to 56 pl/hr, 175% of the tear turnover rate reported by Chrai ef al. (16). This seemingly contradictory observation may not be so, however, considering that the majority of the tear fluid is held under both the upper and lower eyelids (16). Conceivably, the fluid volume may not have been accounted for in the radioactive technetium dilution technique used to estimate tear turnover rate (16). An alternative scenario is that not all of the conjunctiva participates in fluid secretion or that fluid secretion rate varies from one region of the conjunctiva to another. Whether maintenance of fluid balance in the conjunctival sac is a primary function of the conjunctiva remains to be seen. It is tempting, however, to speculate that conjunctival fluid flow may play an important role in hydrating the mucus secreted by goblet cells (17). Such a possibility has been reported in the nasal mucociliary system (18), whereby fluid secreted by the nasal epithelial cells played a key role in hydrating mucus secreted by goblet cells. A significant decrease in goblet cell density was reported in patients with various dry eye syndromes, such as keratitis sicca, Stevens-Johnson syndrome, ocular pemphigoid, and acute alkali bum (19). This pathology
Vol. 66, No. 7,200O
ConjunctivalFluid secretion
PL-111
was hypothesized to be the result, rather than the cause, of dryness in the eye (20). It would be interesting, therefore, to determine whether transconjunctival fluid flow is affected in the dry eye state and to determine whether the formation of mucus threads at the conjunctival surface of dry eye patients (20) is a manifestation of reduced transconjunctival fluid secretion. In summary, pigmented rabbit conjunctiva is capable of secreting fluid that is subject to CAMP and purinergic modulation. The stage is set for addressing the physiological function of transconjunctival fluid flow; its possible alteration in pathological conditions affecting the conjunctiva, such as inflammation, infection, and dry eye; and its restoration by pharmacologic intervention. Acknowledgments The authors are grateful to Drs. Ernest M. Wright and Donald D.F. Loo (Department of Physiology, University of California at Los Angeles) for their assistance. This work was supported in part by the National Institutes of Health research grants EY10421 (VHLL), I-IL38658 (KJK), I-IL46943 (IUK), and University of Southern California Charles and Charlotte Krown Fellowship (MHIS). References 1. S.D. KLYCE, Am. J. Physiol. 228 1446-1452 (1975). 2. J.L. EDELMAN and S.S. MILLER, Invest. Ophthalmol. Vis. Sci. 32 3033-3040 (1991). 3. C. JIANG, W.E. FINKBEINER, J.H. WIDDICOMBE, and S.S. MILLER, J. Physiol. (Lond) 501637-647 (1997). 4. Y. MONSEREENUSORN, J. Pharmacobiodyn. 3 631-635 (1980). 5. U.B. KOMPELLA, K.J. KIM, and V.H.L. LEE, Cm-r. Eye Res. 12 1041-1048 (1993). 6. M.H.I. SHIUE, K.J. KIM, and V.H.L. LEE, Exp. Eye Res. 66 275-282 (1998). 7. K. HOSOYA, H. UEDA, K.J. KIM, and V.H.L. LEE, J. Pharmacol. Exp. Ther. (in press) (1999). 8. O.A. CANDIA, X.P. SHI, and L.J. ALVAREZ, Exp. Eye Res. 66 615-624 (1998). 9. U.B. KOMPELLA, K.J. KIM, M.H.I. SHIUE, and V.H.L. LEE, J. Ocular Pharmacol. Therap. 12 281-287 (1996). 10. K. HOSOYA, U.B. KOMPELLA, K.J. KIM, and V.H.L. LEE, Cm-r. Eye Res. 15 447-45 1 (1996). 11. B.A. HUGHES, S.S. MILLER, and T.E. MACHEN, J. Gen. Physiol. 83 875-899 (1984). 12. J. FISCHBARG, F.P. DIECKE, K. KUANG, B. YU, F. KANG, P. ISEROVICH, Y. LI, H. ROSSKOTHEN, and J.P. KONIAREK, Am. J. Physiol. 276 C548-C557 (1999). 13. K. LOESCHKE and C.J. BENTZEL, Am. J. Physiol. 266 G722-G730 (1994). 14. P. CARPI-MEDlNA and G. WHITTEMBURY, Pflugers Arch. 412 66-74 (1988). 15. S. HAMANN, T. ZEUTHEN, M. LA COUR, E.A. NAGELHUS, O.P. OTTERSEN, P. AGRE, and S. NIELSEN, Am. J. Physiol. 274 C1332-Cl345 (1998). 16. S.S. CHRAI, T.F. PATTON, A. MEHTA, and J.R. ROBINSON, J. Pharm. Sci. 62 1112-l 121 (1973). 17. T.L. KESSLER, H.J. MERCER, J.D. ZIESKE, D.M. MCCARTHY, and D.A. DARTT, Curr. Eye Res. 14 985-992 (1995). 18. B. PETRUSON, H.A. HANSSON, and G. KARLSSON, Arch.Otolaryngol. 110 576-581 (1984). 19. R.A. RALPH, Invest. Ophthalmol. 14 299-302 (1975). 20. S. LIOTET, O.P. VAN BIJSTERVELD, 0. KOGBE, and L. LAROCHE, Ophthalmologica 195 119-124 (1987).
PI1 SOO24-3205@9)00638-4
ELSEVIER
Vol. 66. No. 7, Pp. PL LOS-1II,2000 cowligM02OMlElasviaSCiihC. Pri&diniheUSA. ASrightareacmd 00%32OYOO/ssee thmt matter
PfL4RM4COLoGYmm Accelerated Communicatbn
PHARMACOLOGICAL
MODULATION OF FLUID SECRETION IN THE PIGMENTED RABBIT CONJUNCTIVA
Michael H. I. Shiue,’ Ashutosh A. Kulkarni,’ Hovhannes J. Gukasyan,’ Jennifer B. Swisher,’ Kwang-Jin Kim,2 and Vincent H. L. Leelv3
Departments of ‘Pharmaceutical Sciences, ‘Ophthalmology, 2Medicine, 2Physiology and Biophysics, 2Molecular Pharmacology and Toxicology, and 2Biomedical Engineering and 2Will Rogers Institute Pulmonary Research Center, Schools of Pharmacy, Medicine, and Engineering, University of Southern California, Los Angeles, CA 90033 (Submitted July 6,1999; accepted August 16, 1999; received in final form September 22,1999)
Abstract. We determined net fluid secretion rate across the pigmented rabbit conjunctiva in the presence and absence of pharmacological agents known to affect active Cl secretion and Na+ absorption. Fluid flow across a freshly excised pigmented rabbit conjunctiva mounted between two Lucite half chambers was measured by a pair of capacitance probes in an enclosed cabinet maintained at 37 “C and a relative humidity of 70%. Fluid transport was also measured in the presence of compounds known to affect active Cl secretion (CAMP, UTP, and ouabain), Na+ absorption (D-glucose), or under the Cl-free condition on both sides of the tissue. Net fluid secretion rate across the pigmented rabbit conjunctiva in the serosal-to-mucosal direction at baseline was 4.3 + 0.2 pl/hr/cm2 (mean f s.e.m.). Net fluid secretion rate was increased approximately two-fold by mucosally applied 1 mM 8-Br CAMP (8.4 + 0.4 pl/hr/cm2) and 10 pM UTP (9.8 f 0.6 p.l/hr/cm2), but was abolished by either serosally applied 0.5 mM ouabain (0.3 k 0.1 pl/hr/cm2) or under the Cl-free conditions (0.06 f 0.04 pl/hr/cm2). Mucosal addition of 20 mM Dglucose decreased net fluid secretion rate to 1.0 If: 0.5 pl/hr/cm2. In conclusion, the pigmented rabbit conjunctiva appears to secrete fluid secondary to active Cl‘ secretion. This net fluid secretion is subject to modulation by changes in active Cl secretion rate and in mucosal fluid composition such as glucose concentration. 8 2000ElsevierScienceInc. Key Words: conjuuctiva, uTP,capacitanceprobe
net fluid
secretion,
active
~hlori&
secretion,
Na+-coupled
glucose
transport,
CAMP,
Introduction Transtissue fluid flow has been demonstrated in epithelia such as rabbit comeal epithelium (l), bovine retinal pigment epithelium (2), human tracheal gland cells (3), and rat intestinal epithelium (4). Typically, fluid secretion across epithelia is driven by active Cl secretion (1). The pigmented rabbit conjunctiva is predominantly a Cl-secreting tissue (5), subject to modulation by CAMP, Corresponding Author: Vincent H.L. Lee, Ph.D., University of Southern California, School of Pharmacy, Department of Pharmaceutical Sciences, 1985 Zonal Avenue, Los Angeles, CA 90033 USA. (Voice: 323-442-1368; FAX: 323-442-1390; E-mail: [email protected]).
PL-106
ConjunctivalFluid Secretion
Vol. 66, No. 7, 2000
Ca*+, protein kinase C (PKC), and purine and pyrimidine nucleotides (6,7). The short circuit current (1,J was stimulated 133% by 1 mM 8-Br CAMP, 107% by 10 l.tM A23187 (a Ca2+ ionophore), 87% by 1 pM phorbol lZmyristate-13-acetate (PMA) (6), and 180% by 10 pM UTP (7). The only report on conjunctival water permeation was a study done by Candia ef al. (8) investigating the effect of hypotonicity on the mucosal to serosal diffusional 3H-water flux across the excised albino rabbit conjunctiva. Thus, the purpose of the present study was to determine whether the pigmented rabbit conjunctiva indeed is capable of secreting fluid and whether such fluid secretion was coupled with active Cl secretion. 8-Br CAMP and UTP were chosen for study, since they stimulated active Cl- secretion the most in our previous studies (6,7). Materials and Methods Reagents: 8-bromoadenosine-3’,5’-cyclic monophosphate (8-Br CAMP), ouabain, D-glucose, Dmannitol, and uridine 5’-triphosphate (UTP) were purchased from Sigma Chemicals Co. (St. Louis, MO). Animals and tissue preparation: Male, Dutch-belted pigmented rabbits (2.5-3.0 kg) were purchased from Irish Farms (Norco, CA) and were treated in compliance with The Guiding Principles in the Care and Use of Animals (DI-IEW Publication, NM 80-23). The detailed procedure for conjunctival tissue preparation was reported previously (5). In brief, rabbits were euthanized with an overdose of sodium pentobarbital solution (325 mg/kg) via a marginal ear vein. Both eyeballs were excised, and the conjunctival tissues were trimmed for mounting as flat sheets between two Lucite half chambers for water flux measurements. Approximately the same region of the conjunctival tissue was used for each experiment. The mucosal and the serosal sides of the tissue (0.385 cm*) were each bathed in bicarbonated Ringer’s solution (BR). Regular BR contained 111.55 n&l NaCl, 4.82 mM KCI, 0.86 mM NaI-IzPOd, 29.20 mM NaHC03, 1.04 mM CaC12, 0.74 mM MgC12, and 5 mM D-glucose. The buffer pH was adjusted to 7.4 by bubbling with 5% CO2 in air. Cl free condition was established by replacing, on an equimolar basis, NaCI, KCl, CaC12, and MgCl2 with sodium isethionate, potassium isethionate, calcium gluconate, and magnesium gluconate, respectively. For glucose-rich solution, an additional 20 mM D-glucose was added to regular BR bathing solution on the mucosal side. To determine the possible osmotic effect of excess osmolytes, net fluid secretion rate was measured in the presence of mucosal 20 mM Dmannitol, in lieu of 20 mM D-glucose. Fluidflu measurements: Fluid flow across an excised pigmented rabbit conjunctiva was measured by a pair of capacitance probes (ASP-lo-CTAISP, Mechanical Technology, Inc., Latham, NY) similar to that described by Edelman and Miller (2). The entire apparatus was placed in an enclosed cabinet maintained at 37’C and at a relative humidity of 70%. Prior to each experiment, the capacitance probes were also equilibrated at 37°C inside the enclosed cabinet. In pilot experiments, we found that a relative humidity of at least 60% was required to prevent detectable evaporation of fluid during the course of an experiment. One mM 8-Br CAMP (mucosal), 10 pM UTP (mucosal), 20 mM D-glucose (mucosal), 20 mM D-mannitol (mucosal), or 0.5 mM ouabain (serosal) was applied to the reservoir using the concurrent feed and drain method with a pair of 10 pl Hamilton syringes joint at the end of the plungers. Instillation of a drug solution by one syringe was counterbalanced by the withdrawal of an equal volume of buffer from the inner chamber by the opposite syringe, thus keeping a constant fluid volume (3.5 ml on each side) during drug application in the reservoir. The concentration used was based on the I,, measurements reported previously (7,9,10). To demonstrate a role for active Cl secretion in transconjunctival fluid flow, water flux was measured under the Cl- free condition, whereby a Cl--free solution was used in both the serosal and mucosal reservoirs. Each tissue was treated with only one agent/condition. Except
Vol. 66, No. 7,200O
Conjunctival Fluid Secretion
PL-107
for the Cl-free condition, the baseline fluid flux was measured for 30 min across all tissues prior to the drug application. The fluid flux was then monitored for another 30 min from the time of drug instillation. The capacitance of the air gap between the probe and the fluid surface (less than 3 mm in distance) on each side of the chamber was measured. The voltage output from the capacitance probe indicates the loss of fluid from one side of the chamber, while gain of fluid on the opposite side. Under fluid secreting conditions, the voltage output from the mucosal probe would decrease, and a symmetrical increase in voltage output would be observed on the serosal side. The voltage output from each probe was equal but opposite. Non-symmetrical voltage changes were considered an artifact and not acquired. The fluid transport rate (pl/hr/cm*) was calculated based on the slope of the voltage vs. time and conversion of voltage changes to fluid volume displacement based on a calibration table provided by the manufacturer. Statistical analysis: Results were expressed as mean * s.e.m. Comparisons between the group means of two data sets were analyzed with unpaired Student’s t-test. One way analysis of variance and post-hoc comparisons based on Fisher’s least squared difference approach were used to determine statistical significance among group means of more than two data sets. A p value of less than or equal to 0.05 was considered significant.
A net serosal (s) to mucosal (m) fluid secretion rate (JV) of 4.3 f 0.2 pl/hr/cm* (n = 27) was observed in the freshly excised conjunctiva under baseline condition (Figs. 1 and 2). Mucosal application of 1 mM 8-Br CAMP approximately doubled the baseline fluid secretion (p < 0.05) to 8.4 rt 0.4 @/hr/cm* (n = 5) (Figs. 1 and 2). Moreover, the increase in conjunctival fluid secretion afforded by 8-Br CAMP was concentration-dependent for up to 3 mM range studied (Fig. 3). The observed maximal change in fluid secretion rate (AI,) was 5.1 k 0.3 pl/hr/cm* and the halfmaximal 8-Br CAMP concentration was 0.32 f 0.06 mM. This latter value is comparable to the previously reported effective half-maximal concentration (ECso) values for stimulating Isc (0.28 mM) and net Cl- secretion (0.30 mM) by 8-Br CAMP (6). By contrast, 10 pM UTP increased fluid secretion to 9.8 f 0.6 pl/hr/cm* (n = 6) (p c 0.05) (Fig. 2) transiently for 10 min (Fig. 1). The baseline J, was significantly (p < 0.05) reduced by serosally applied 0.5 mM ouabain to 0.3 + 0.1 pl/hr/cm* (n = 4), not different from zero (p > 0.05) (Figs. 1 and 2). In addition, net fluid transport was abolished (p < 0.05) when Cl was removed from both sides of the conjunctiva (0.06 + 0.04 p&r/cm*, n = 4) (Fig. 2). Twenty mM D-glucose was previously determined to induce a maximal increase in transconjunctival Isc(10). Mucosal addition of 20 mM D-glucose to BR significantly reduced (p < 0.05) the baseline J, by 77% to 1.0 f 0.5 pl/hr/cm* (n = 4) (Figs 1 and 2). By contrast, mucosally applied 20 mM D-mannitol (which elevated the osmolality by 22 f 1.2 mOsm/kg) did not affect the baseline J, significantly (p > 0.05). A good correlation (3 = 0.98) was observed between the changes in net fluid flux and those in IX induced by agents known to affect active conjunctival Clsecretion (i.e., excluding D-glucose) (Figd).
Conjunctid
PL-108
Vol. 66, No. 7,200O
Fluid Secretion
,
1 mM 8-Br CAMP 10 @l UTP
20 mM D-glucose 0.5mMOuabain
30 Time (min)
60
Fig. 1 Typical voltage vs. time tracings, demonstrating the effect of 1 mM S-Br CAMP, 10 @I UTP, 0.5 mM ouabain, and 20 mM D-glucose on transconjunctival fluid secretion. Agents were applied after 30 n-tin of baseline measurement (as indicated by the arrow). One volt (V) deflection is equivalent to 127 pm of fluid displacement according to the calibration table provided by the probe manufacturer (Mechanical Technology, Inc., Latham, NY). The positive slope of the linear segment denotes the transconjunctival fluid secretion rate.
12.OT
*
I *
-;
8.0--
g t G-
4.0--
t t
0.07
Ls
t
I
Fig. 2 Conjunctival fluid secretion rate at baseline (n = 35) and in response to mucosal 1 mM 8Br CAMP (n = 5), mucosal 10 pM UTP (n = 6), serosa10.5 mM ouabain (n = 4), mucosal 20 mM D-glucose (n = 4), mucosal 20 mM D-mannitol (n = 3), and U-free (both sides, n = 4) conditions. Error bars represent s.e.m. * indicates a significantly higher fluid flow (p < 0.05) than baseline. t indicates a significantly lower (p < 0.05) fluid flow than baseline.
Vol. 66, No. 7,200O
Con-
Fluid
Secretion
[&Br CAMP] (mM) Fig. 3 Changes in net transconjunctival fluid secretion rate as a function of mucosal 8-Br CAMP concentration. Error bars denote s.e.m. for 4-6 tissues.
6, 3 mM &Br CAMP 1 mlU 5-Br CAMP iu^ 5
0.5 mM &Br
2
0.1 mY EBr
5 a
CAMP
CAMP
O >
2 20 mM D-glucose
-6-1 -14.0
0.0
I 14.0
AL (Wcm2) Fig. 4 Relation between change in net fluid secretion rate (AJV”‘) and the corresponding a, under conditions known to affect active Cl- secretion and Na+ absorption. Error bars denote s.e.m. for 4-6 tissues. The & values for 8-Br CAMP (6), Cl--free (5), ouabain (5), D-glucose (lo), and UTP (7) were from our previous reports. The linear correlation (3 = 0.98) was determined only for conditions that affect active Cl- secretion.
PL-110
Conjunctival Fluid Secretion
Vol. 66, No. 7,200O
Discussion We have demonstrated that the pigmented rabbit conjunctiva secretes fluid at a baseline J, of 4.3 f 0.2 pVhr/cm’. This value is on the same order of magnitude reported for frog retinal pigment epithelium (3.9 pl/hr/cm2) (1 I), bovine lens epithelium (4.7 pl/hr/cm2) (12), and human tracheal gland cells (2 @/hr/cm*) (3). By contrast, the fluid secretion rate in the rabbit cornea is 0.2 @/hr/cm2 (l), only 5% of that in the pigmented rabbit conjunctiva. Conjunctival fluid secretion appears to be coupled to active Cl secretion. This is indicated by the abolishment of fluid secretion both in the presence of serosa10.5 mM ouabain and upon removal of Cl from the bathing solutions (Fig 2). It is, therefore, not surprising that compounds known to stimulate Cl- secretion in the conjunctiva, such as 8-Br CAMP (6) and UTP (7), also stimulate conjunctival fluid secretion (Figs. 1 and 2). Indeed, there exists an excellent correlation between the changes in fluid secretion and those in I,, in the conjunctiva under conditions that affect active Cl secretion (Fig. 4). The 128% enhancement of fluid secretion by 10 p.M UTP may be of potential therapeutic importance in the management of dry eye state, when CAMP-sensitive Clsecretion may have been compromised. In fact, Jiang ef al. (3) reported that human cystic fibrosis (CF) tracheal gland cells maintain normal fluid secretory response to UTP, despite impairment of the CAMP-dependent Cl secretion. Conjunctival fluid flow is also sensitive to conditions that affect active Na+ absorption. For instance, increasing mucosal D-glucose concentration from 5 to 25 mM reduced conjunctival fluid flow by 77% (Fig. 2), presumably secondary to stimulation of Na+ uptake via the Na+-coupled glucose transporter (10). The lack of significant influence by 20 mM D-mannitol on J, indicates that the osmotic contribution of 20 mM D-glucose on the observed J, was negligible. There may exist two possible pathways for conjunctival fluid flow: paracellular (13) and transcellular (14). Both pathways are probably utilized by fluid flow driven either by active Cl secretion or an osmotic gradient. Transcellular fluid flow, in addition, is probably mediated by aquaporins (AQP). Aquaporin type 3 (AQP3) has been demonstrated in the human and rat conjunctival epithelium, using immunoblot analysis of membrane proteins, high-resolution immunocytochemistry, and immunoelectron microscopy (15). The relative proportion of the tear volume attributable to the fluid secreted by the conjunctiva is not known. Given that the surface area of the rabbit conjunctiva is 13 cm2 and assuming that all of it participates in fluid secretion at the observed rate, the transconjunctival fluid secretion rate would amount to 56 pl/hr, 175% of the tear turnover rate reported by Chrai ef al. (16). This seemingly contradictory observation may not be so, however, considering that the majority of the tear fluid is held under both the upper and lower eyelids (16). Conceivably, the fluid volume may not have been accounted for in the radioactive technetium dilution technique used to estimate tear turnover rate (16). An alternative scenario is that not all of the conjunctiva participates in fluid secretion or that fluid secretion rate varies from one region of the conjunctiva to another. Whether maintenance of fluid balance in the conjunctival sac is a primary function of the conjunctiva remains to be seen. It is tempting, however, to speculate that conjunctival fluid flow may play an important role in hydrating the mucus secreted by goblet cells (17). Such a possibility has been reported in the nasal mucociliary system (18), whereby fluid secreted by the nasal epithelial cells played a key role in hydrating mucus secreted by goblet cells. A significant decrease in goblet cell density was reported in patients with various dry eye syndromes, such as keratitis sicca, Stevens-Johnson syndrome, ocular pemphigoid, and acute alkali bum (19). This pathology
Vol. 66, No. 7,200O
ConjunctivalFluid secretion
PL-111
was hypothesized to be the result, rather than the cause, of dryness in the eye (20). It would be interesting, therefore, to determine whether transconjunctival fluid flow is affected in the dry eye state and to determine whether the formation of mucus threads at the conjunctival surface of dry eye patients (20) is a manifestation of reduced transconjunctival fluid secretion. In summary, pigmented rabbit conjunctiva is capable of secreting fluid that is subject to CAMP and purinergic modulation. The stage is set for addressing the physiological function of transconjunctival fluid flow; its possible alteration in pathological conditions affecting the conjunctiva, such as inflammation, infection, and dry eye; and its restoration by pharmacologic intervention. Acknowledgments The authors are grateful to Drs. Ernest M. Wright and Donald D.F. Loo (Department of Physiology, University of California at Los Angeles) for their assistance. This work was supported in part by the National Institutes of Health research grants EY10421 (VHLL), I-IL38658 (KJK), I-IL46943 (IUK), and University of Southern California Charles and Charlotte Krown Fellowship (MHIS). References 1. S.D. KLYCE, Am. J. Physiol. 228 1446-1452 (1975). 2. J.L. EDELMAN and S.S. MILLER, Invest. Ophthalmol. Vis. Sci. 32 3033-3040 (1991). 3. C. JIANG, W.E. FINKBEINER, J.H. WIDDICOMBE, and S.S. MILLER, J. Physiol. (Lond) 501637-647 (1997). 4. Y. MONSEREENUSORN, J. Pharmacobiodyn. 3 631-635 (1980). 5. U.B. KOMPELLA, K.J. KIM, and V.H.L. LEE, Cm-r. Eye Res. 12 1041-1048 (1993). 6. M.H.I. SHIUE, K.J. KIM, and V.H.L. LEE, Exp. Eye Res. 66 275-282 (1998). 7. K. HOSOYA, H. UEDA, K.J. KIM, and V.H.L. LEE, J. Pharmacol. Exp. Ther. (in press) (1999). 8. O.A. CANDIA, X.P. SHI, and L.J. ALVAREZ, Exp. Eye Res. 66 615-624 (1998). 9. U.B. KOMPELLA, K.J. KIM, M.H.I. SHIUE, and V.H.L. LEE, J. Ocular Pharmacol. Therap. 12 281-287 (1996). 10. K. HOSOYA, U.B. KOMPELLA, K.J. KIM, and V.H.L. LEE, Cm-r. Eye Res. 15 447-45 1 (1996). 11. B.A. HUGHES, S.S. MILLER, and T.E. MACHEN, J. Gen. Physiol. 83 875-899 (1984). 12. J. FISCHBARG, F.P. DIECKE, K. KUANG, B. YU, F. KANG, P. ISEROVICH, Y. LI, H. ROSSKOTHEN, and J.P. KONIAREK, Am. J. Physiol. 276 C548-C557 (1999). 13. K. LOESCHKE and C.J. BENTZEL, Am. J. Physiol. 266 G722-G730 (1994). 14. P. CARPI-MEDlNA and G. WHITTEMBURY, Pflugers Arch. 412 66-74 (1988). 15. S. HAMANN, T. ZEUTHEN, M. LA COUR, E.A. NAGELHUS, O.P. OTTERSEN, P. AGRE, and S. NIELSEN, Am. J. Physiol. 274 C1332-Cl345 (1998). 16. S.S. CHRAI, T.F. PATTON, A. MEHTA, and J.R. ROBINSON, J. Pharm. Sci. 62 1112-l 121 (1973). 17. T.L. KESSLER, H.J. MERCER, J.D. ZIESKE, D.M. MCCARTHY, and D.A. DARTT, Curr. Eye Res. 14 985-992 (1995). 18. B. PETRUSON, H.A. HANSSON, and G. KARLSSON, Arch.Otolaryngol. 110 576-581 (1984). 19. R.A. RALPH, Invest. Ophthalmol. 14 299-302 (1975). 20. S. LIOTET, O.P. VAN BIJSTERVELD, 0. KOGBE, and L. LAROCHE, Ophthalmologica 195 119-124 (1987).