Exp. Eye Res. (1992) 55, 637-640
The Microchemical Detection of Carbonic Anhydrase in Corneal Epithelia C U R T I S W. C O N R O Y , R I C H A R D H. B U C K AND T H O M A S H. M A R E N
University of Florida Health Science Center, College of Medicine, Department of Pharmacology and Therapeutics, P.O. Box 100267, Gainesville, FL 32610, U.S.A. (Received Chicago 7 January 1992 and accepted in revised form 23 March 1992) Carbonic anhydrase (CA) activity has been detected and quantified in the corneal epithelia of various mammalian species using a microchemical assay. The highest levels were found in the rabbit, followed by man, dog, sheep, and cat. Enzyme levels in the rabbit epithelium were approximately one-third the amount found in the endothelium, perhaps explaining earlier failures to find CA in the epithelium. Selective inactivation of CA-[[ in tissue homogenates with bromopyruvic acid allowed determination of CA-I/CA-II ratios. The ratios in corneal epithelia from two species (rabbit and dog) approximated that in the erythrocyte. Although CA levels in the epithelium are low they are perhaps functional in the transfer of CO2 across the cornea. Key words: carbonic anhydrase; isoenzyme distribution; corneal epithelium; bromopyruvic acid; CO2 transport.
1. Introduction Carbonic anhydrase (CA) has not generally been regarded as being present in the corneal epithelium. Using Warburg manometers and a modification of the method of Krebs and Roughton, Gloster (1955) failed to demonstrate CA activity in corneal epithelial extracts of rabbits. This method, however, was not able to measure epithelial CA concentrations less than 19/0 of blood CA. L6nnerholm (1974) examined for CA activity histochemically in the corneas of adult man, h u m a n fetus, monkey, rabbit, rat, cow, pig and American bullfrog. Unfixed epithelial cells of all species invariably appeared unstained, but individual cells in human, cow and bullfrog showed positive staining in fixed tissue specimens. A changing pH method was also used by this investigator to determine epithelial CA activity is dissected cow and rabbit corneas. No enzyme was detected in the cow; results from rabbit were equivocal. Silverman and Gerster (1973), using an 180 exchange method, failed to find CA in rabbit corneal epithelium. Wistrand, Schenholm and L6nnerholm (1986) examined for the presence of CA in the h u m a n cornea using monospecific antisera raised against isoenzymes CA-I and CA-II, in addition to a modified version of the Hansson method. Good agreement was seen between histochemical and immunocytochemical findings in all eye tissue examined with the exception of the epithelium, where histochemical staining was variable and always weaker than immunottuorescence. Birndorf et al. {1990) have recently reported the presence of CA-I and CA-II in the h u m a n corneal epithelium in embryonic through adult stages using 0014-4835/92/100637+04 $08.00/0
an immunohistochemical technique in which formalin fixed deparaffinized tissue sections were labeled with anti-human CA-I and CA-If antibodies. In addition, indirect evidence for CA in corneal epithelium is provided by the effect of methazolamide on the shortcircuit current across the isolated frog corneal epithelium (Candia, 1990). In the present study total CA activities in corneal epithelia of several species were measured using a microenzymatic assay (Maren, 1960) in combination with a differential inactivation procedure using bromopyruvic acid which permitted determination of CA-I/CA-II ratios (Conroy and Maren, 1985). 2. Materials and Methods Enucleated eyes from various species were obtained as donor material from several sources. H u m a n eye bank corneas unsuitable for transplantation were used. Male New Zealand White rabbits weighing 2 - 3 kg were killed by intramuscular injection of ketamine followed by pentobarbital. Corneas were excised and rinsed with distilled water. The epithelial layer from the central portion of each cornea was carefully scraped with a surgical blade and transferred to a small test tube for weighing. The typical weights of tissues are shown in Table IIL Tissue samples were homogenized by needle aspiration in 0.2 ml distilled water prior to assay for CA activity. Total tissue CA concentrations were measured using a microchemical technique based on the catalytic hydration reaction of CO~ and the associated pH change (Maren, 1960). All tissue samples were homogenized in 0.2 ml distilled water. Aliquots were added to 0.5 ml distilled water containing 50 mg 1-1 of phenol red in a reaction vessel maintained at O°C. Pre© 1992 Academic Press Limited
638
C.W. C O N R O Y ET AL.
TABLE I
Relations among enzyme units and molarity in the several analytical systems at O°C and 100% C02: carbonic anhydrase I1"*
Carbonate (Maren, 1960, 1967) Barbital (Maren et al., 1960)
Molar (M) equivalent in 7 ml reaction volume
Molar (M) equivalent in 0'7 ml reaction volume
1 eu
1 eu g-~ tissue
1 eu
1 eu g-~ tissue
2'4 x 10 -9
17 x 10 -9
2"4 x 10 -9
1"7 x 10 -~
0"7 x 10 -9
5 x 10 -9
0"7 x 10 -9
0"5 x 10 9
* For carbonic anhydrase I, multiply the barbital values by 8.5. See relations between eqns (3) and (4). Carbonate is inhibitory to CA-I, so relations in this system are not given.
cooled 100% CO2 was bubbled through the chamber at a rate of 60 ml min -1. The reaction was initiated by the addition of 0-2 ml of 50 mM barbital buffer at its pK~ (pH 8-1) and the time (T~) recorded to obtain an end-point at pH 7'2. [The visual end-point in the 0"7 ml system can be enhanced by the use of bromothymol blue (50 mg 1-1) instead of phenol red. Since the end-point for this buffer (pH 6"6) is lower than for phenol red, slightly longer reaction times are obtained. The isoenzyme molarities for 1 enzyme unit (eu) in this system, however, are unchanged.] Barbital buffer was chosen because it does not inhibit either CA isoenzyme. The total number of eu were computed from the relationship in which 1 eu doubles the uncatalyzed rate. Thus eu--
~°-~ T~
~° T~
1
where /~ represents the uncatalyzed rate of CO~ hydration. The carbonic anhydrase I and II isoenzyme content of tissue homogenates was determined by incubation of equal volumes of homogenate and 40 mM bromopyruvic acid (pH 7'5) for 50 rain prior to assay for CA activity. Under these conditions greater than 95% of CA-I is irreversibly inactivated with less than a 5% loss of CA-II (Gothe and Nyman, 1972; Conroy and Maren, 1985). The molar equivalences of 1 eu for CA II and CA I at O°C and 100% CO2 (70 raM) may be derived from the rate expressions for the two isozymes. V~ -
koat "S" E
(1)
Sq-K m
where S is the substrate concentration in the cell. At 0°C and for median pH of 7.6, kc~t and K~ are 270 000 sec -1 and 10 raM, respectively, for CA II, and 29 650 sec -1 and 4 mM, respectively, for CA I (Sanyal and Maren, 1981). Rearrangement of eqn (1) to give E yields F - Gds
+ Kin)
S" k0~t
(2)
For the case of 1 eu in the system V0~t= V,,~c~t. Titration of 0.2 ml 50 mM barbital from a starting pH of 8"2 to pH 7.2 (phenol red end-point) represents acidification of 5 8 5 0 # M of the added 14300,uM buffer. This concentration divided by (/~c) gives a velocity ibr the system of 167 #M sec -1. Substituting this V,n~t for Vc~t in eqn (2) gives the molar equivalences for 1 eu for the two isoenzymes as follows: CA-II = 167 #M sec-1 (70 mM + 10 raM) 70 mM X 270 000 sec -1 = 0"71x 10 -9 M
(3)
CA-I = 167 #M sec -1 (70 r a M + 4 raM) 70 mM X 29 650 sec -1 = 6.0x 10 9M.
(4)
Note that this conversion of eu to molarity is independent of the reaction volume of the assay but depends on temperature and substrate concentration. The conversion of eu g-1 tissue to molarity of enzyme, however, does depend on the reaction volume. From eqn (3): 1 eu = 0"71 x 10 -9 tool -1 = 5 x 10 -13 mol (in 0.7 ml reaction volume)
(5) 5 x 10 -la tool Then (taking g -~ ml) 1 eu g-~ 10 -31
(6)
= 0"5 x 10 -9 mol 1-1 (M). Thus, 100 eu gm -~ of CA-II in the 0.7 ml system would correspond to 100 (0.5 x 10 -9 M) = 0"05 #M CA II. The equivalency of eqn (3) has also been made by titration of enzyme with powerful inhibitors such as ethoxzolamide, yielding for CA II 1 eu = 1 x 10 -9 M (Maren, Parcell and Malik, 1960), now modified by recent titration to 0"7 x 10 -9 M, the same as obtained from the rate of eqn (3). Because enzyme units differ between buffer systems and the units g-1 tissue depend on the volume of the reaction, we show the relations (Table f) among these
CARB'ONICANHYDRASE IN CORNEAL EPITHELIUM
639
TABLE H
Carbonic anhydrase activities in selected tissues*
Source Red Cell Lens Ciliary process Corneal endothelium Corneal epithelium
CA-II (eu gm 1)
CA-I (eu gm -1)
CA-II
Species
Total CA activity (eu gm -1)
Human Rabbit Dog Rabbit Rabbit Rabbit Rabbit
45 800 20 750 31800 6000 1600 450 150
24 000 12 700 23 100 6000 1600 450 80
21 800 8050 8700 0 0 0 65
12"0 6"4 11"6 3"0 0"8 0"23 0"04
CA-I
Total CA
~M) 91"6 33"8 36"5 0 0 0 0'27
103"6 40"2 48"1 3'0 0'8 0'23 0'31
* See Materials and Methods section for conversion of eu gm 1 to #M.
TABLE HI
Isoenzyme distribution in corneal epithelium
Species Rabbit Hum an Dog Cat Sheep Sheep (fetal)
Average weight of epithelium (rag)
n
10 10 15 20 20 10
4 5 4 2 2 2
Total CA activity (eu g-l)
CA-II* (eu g-* tissue)
CA-I (eu g-1 tissue)
CA-II (riM)
145 _+38 88 __+9 45 + 14 12 15
79+27 67_+8 35_+6
65_+36 21_+5 10_+3
40 34 18
N.D.J"
CA-I (riM)
273 88 42
--
--
6
--
--
--
7
--
.
.
.
.
Total CA (riM) 313 122 60 6 7 0
* Activity remaining after bromopyruvate inactivation. t Not detected.
systems, including our original 7 ml carbonate procedure (see Table 16, Maren, 1967). It will be seen that the relation between eu g-1 tissue in the large cell carbonate system and the present small cell barbital system is 17/0.5 or 34, i.e. the numbers in Maren (1967) x 34 equal the present values.
3.
Results
and
Discussion
The CA activity in selected tissues of the rabbit eye is shown in Table II along with the content in red cell hemolysate from three species. The levels reported here in the barbital system for blood, lens, and ciliary process are similar to those previously reported by Ballintine and Maren (1955) using bicarbonate buffer when allowances are made for inhibition caused by this buffer. The ciliary process and corneal endothelium were found to contain only the high activity CA-II isoenzyme in agreement with previous studies. Although CA levels in these tissues are low compared to erythrocyte CA, the enzyme has a well defined function in both, In general, CA-II m a y be found as soluble cytoplasmic enzyme in a number of electrolyte transporting epithelia. It facilitates the m o v e m e n t of
water coupled to Na + and HCO a- (Maren, 1988), the latter being generated from COy In the case of the corneal endothelium, the movement of fluid from the stroma into the aqueous h u m o r is thought to maintain the clarity of the cornea. In the ciliary process the enzyme serves a secretory role giving rise to the aqueous humor. Incubation of lens homogenate with bromopyruvic acid did not result in a decrease in CA activity, consistent with an absence of CA-I in this tissue. This is in disagreement with a previous immunohistochemical study which found both CA-I and CA-II staining in the lens (Wistrand et al., 1986). Friedland and Maren (1981 ) have previously discussed the possible role of CA in the elimination of CO,, from the lens across narrow concentration gradients. The CA content and isoenzyme distribution of epithelial homogenates from several species are shown in Table III. Only the first three of the homogenates showed sufficient CA activity to permit estimation of CA-I and CA-II ratios. In rabbit and dog the activity ratio of CA-I to CA-II approximated that found in the red cell of that species. In all cases inclusion of 10 -7 M methazolamide in epithelial homogenates reduced the CO2 hydration activity to the uncatalyzed rate. The role played by CA in corneal epithelia is difficult
640
to clarify, although its function is probably unrelated to fluid secretion or ion transport. Even more difficult to explain is the presence of CA-I in this tissue since an exact role for CA-I in erythrocyte§ has never been found. The most plausible hypothesis is that epithelial CA-I and CA-II m a y facilitate the outward movement of metabolic CO~ across small concentration gradients. This facilitated diffusion occurs through the catalytic conversion of CO2 to HCO3-, as described for the lens (Friedland and Maren, 1981). According to this scheme inhibition of epithelial CA should lead to the establishment of a new and higher gradient for CO~ diffusion to ambient air. In support of this are recent findings by Candia (1990, 1991) which showed that lO-4M methazolamide reduced the flux of [14C]CO~ across the isolated corneal epithelium by 30 %. Further studies which involve measurement of CO2 gradients across the cornea during inhibition by topical CA inhibitors m a y provide additional evidence for the participation of epithelial CA in CO~ transport.
Acknowledgement This research was supported by National Institutes of Health grant EY-02227.
References Ballintine, E.J. and Maren, T.H. (1955). Carbonic anhydrase activity and the distribution of Diamox in the rabbit eye. Am. J. Ophthahnol. 40, 148-54. Birndorf, L., Kitada, S., Shapourifar-Tehrani, S. and Lee, D. A. (1990). Developmental expression of carbonic anhydrase isoenzymes in human eyes. Invest. Ophthalrnol. Vis. Sci. 31 (Suppl.), 155.
C.W. C O N R O Y ET AL.
Candia, O.A. (1990). Forskofin-induced HCO3- current across apical membrane of the frog corneal epithelium. Am. 1. PhysioL 259, C215-33. Candia, O.A. (1991). The movement of CO~ and HCO~across frog epithelium. Evidence for the presence of carbonic anhydrase. Invest. Ophthalrnol. Vis. Sci. 32 (Suppl.), 977. Conroy, C. W. and Maren, T. H. (1985). The determination of osteopetrotic phenotypes by selective inactivation of red cell carbonic anhydrase isoenzymes. Clin. Chim. Acta 152, 347-54. Friedland, B.R. and Maren, T.H. (1981). The relation between carbonic anhydrase activity and ion transport in elasmobranch and rabbit lens. Exp. Eye. Res. 33, 545-61. Gloster, J. (1955). Investigation of the carbonic anhydrase content of the cornea of the rabbit. Br. ]. Ophthalmol. 39, 743-6. Gothe, P.O. and Nyman, P.O. (1972). Inactivation of human erythrocyte carbonic anhydrase by bromopyruvate. FEBS Lett. 21, 159-64. L6nnerholm, G. (1974). Carbonic anhydrase in the cornea. Acta Physiol. Scand. 90, 143-52. Maren, T.H. (1960). A simplified micromethod for the determination of carbonic anhydrase and its inhibitors. J. PharmacoL Exp. Ther. 130, 26-9. Maren, T.H. (1967). Carbonic anhydrase: chemistry, physiology, and inhibition. Phgsiol. Rev. 47. 595-781. Maren, T. H. (1988). The kinetics of HCQ synthesis related to fluid secretion, pH control, and CO~ elimination. Ann. Rev. Physiol. 50, 695-717. Maren, T.H., Parcefi, A.L. and Mafik, M.N. (1960). A kinetic analysis of carbonic anhydrase inhibition. J. Pharmacol. Exp. Ther. 130, 389-400. Sanyal, G. and Maren, T.H. (1981). Thermodynamics of carbonic anhydrase catalysis, a comparison between human isoenzymes B and C. J. Biol. Chem. 256,608-12. Silverman, D. N. and Gerster, R. (1973). The detection and localization of carbonic anhydrase in the rabbit cornea. Exp. Eye Res. 17, 129-36. Wistrand, P. J., Schenholm, M. and L6nnerholm, G. (1986). Carbonic anhydrase isoenzymes CA I and CA II in the human eye. Invest. Ophthalmol. Vis. Sci. 27, 419-28.
The Microchemical Detection of Carbonic Anhydrase in Corneal Epithelia C U R T I S W. C O N R O Y , R I C H A R D H. B U C K AND T H O M A S H. M A R E N
University of Florida Health Science Center, College of Medicine, Department of Pharmacology and Therapeutics, P.O. Box 100267, Gainesville, FL 32610, U.S.A. (Received Chicago 7 January 1992 and accepted in revised form 23 March 1992) Carbonic anhydrase (CA) activity has been detected and quantified in the corneal epithelia of various mammalian species using a microchemical assay. The highest levels were found in the rabbit, followed by man, dog, sheep, and cat. Enzyme levels in the rabbit epithelium were approximately one-third the amount found in the endothelium, perhaps explaining earlier failures to find CA in the epithelium. Selective inactivation of CA-[[ in tissue homogenates with bromopyruvic acid allowed determination of CA-I/CA-II ratios. The ratios in corneal epithelia from two species (rabbit and dog) approximated that in the erythrocyte. Although CA levels in the epithelium are low they are perhaps functional in the transfer of CO2 across the cornea. Key words: carbonic anhydrase; isoenzyme distribution; corneal epithelium; bromopyruvic acid; CO2 transport.
1. Introduction Carbonic anhydrase (CA) has not generally been regarded as being present in the corneal epithelium. Using Warburg manometers and a modification of the method of Krebs and Roughton, Gloster (1955) failed to demonstrate CA activity in corneal epithelial extracts of rabbits. This method, however, was not able to measure epithelial CA concentrations less than 19/0 of blood CA. L6nnerholm (1974) examined for CA activity histochemically in the corneas of adult man, h u m a n fetus, monkey, rabbit, rat, cow, pig and American bullfrog. Unfixed epithelial cells of all species invariably appeared unstained, but individual cells in human, cow and bullfrog showed positive staining in fixed tissue specimens. A changing pH method was also used by this investigator to determine epithelial CA activity is dissected cow and rabbit corneas. No enzyme was detected in the cow; results from rabbit were equivocal. Silverman and Gerster (1973), using an 180 exchange method, failed to find CA in rabbit corneal epithelium. Wistrand, Schenholm and L6nnerholm (1986) examined for the presence of CA in the h u m a n cornea using monospecific antisera raised against isoenzymes CA-I and CA-II, in addition to a modified version of the Hansson method. Good agreement was seen between histochemical and immunocytochemical findings in all eye tissue examined with the exception of the epithelium, where histochemical staining was variable and always weaker than immunottuorescence. Birndorf et al. {1990) have recently reported the presence of CA-I and CA-II in the h u m a n corneal epithelium in embryonic through adult stages using 0014-4835/92/100637+04 $08.00/0
an immunohistochemical technique in which formalin fixed deparaffinized tissue sections were labeled with anti-human CA-I and CA-If antibodies. In addition, indirect evidence for CA in corneal epithelium is provided by the effect of methazolamide on the shortcircuit current across the isolated frog corneal epithelium (Candia, 1990). In the present study total CA activities in corneal epithelia of several species were measured using a microenzymatic assay (Maren, 1960) in combination with a differential inactivation procedure using bromopyruvic acid which permitted determination of CA-I/CA-II ratios (Conroy and Maren, 1985). 2. Materials and Methods Enucleated eyes from various species were obtained as donor material from several sources. H u m a n eye bank corneas unsuitable for transplantation were used. Male New Zealand White rabbits weighing 2 - 3 kg were killed by intramuscular injection of ketamine followed by pentobarbital. Corneas were excised and rinsed with distilled water. The epithelial layer from the central portion of each cornea was carefully scraped with a surgical blade and transferred to a small test tube for weighing. The typical weights of tissues are shown in Table IIL Tissue samples were homogenized by needle aspiration in 0.2 ml distilled water prior to assay for CA activity. Total tissue CA concentrations were measured using a microchemical technique based on the catalytic hydration reaction of CO~ and the associated pH change (Maren, 1960). All tissue samples were homogenized in 0.2 ml distilled water. Aliquots were added to 0.5 ml distilled water containing 50 mg 1-1 of phenol red in a reaction vessel maintained at O°C. Pre© 1992 Academic Press Limited
638
C.W. C O N R O Y ET AL.
TABLE I
Relations among enzyme units and molarity in the several analytical systems at O°C and 100% C02: carbonic anhydrase I1"*
Carbonate (Maren, 1960, 1967) Barbital (Maren et al., 1960)
Molar (M) equivalent in 7 ml reaction volume
Molar (M) equivalent in 0'7 ml reaction volume
1 eu
1 eu g-~ tissue
1 eu
1 eu g-~ tissue
2'4 x 10 -9
17 x 10 -9
2"4 x 10 -9
1"7 x 10 -~
0"7 x 10 -9
5 x 10 -9
0"7 x 10 -9
0"5 x 10 9
* For carbonic anhydrase I, multiply the barbital values by 8.5. See relations between eqns (3) and (4). Carbonate is inhibitory to CA-I, so relations in this system are not given.
cooled 100% CO2 was bubbled through the chamber at a rate of 60 ml min -1. The reaction was initiated by the addition of 0-2 ml of 50 mM barbital buffer at its pK~ (pH 8-1) and the time (T~) recorded to obtain an end-point at pH 7'2. [The visual end-point in the 0"7 ml system can be enhanced by the use of bromothymol blue (50 mg 1-1) instead of phenol red. Since the end-point for this buffer (pH 6"6) is lower than for phenol red, slightly longer reaction times are obtained. The isoenzyme molarities for 1 enzyme unit (eu) in this system, however, are unchanged.] Barbital buffer was chosen because it does not inhibit either CA isoenzyme. The total number of eu were computed from the relationship in which 1 eu doubles the uncatalyzed rate. Thus eu--
~°-~ T~
~° T~
1
where /~ represents the uncatalyzed rate of CO~ hydration. The carbonic anhydrase I and II isoenzyme content of tissue homogenates was determined by incubation of equal volumes of homogenate and 40 mM bromopyruvic acid (pH 7'5) for 50 rain prior to assay for CA activity. Under these conditions greater than 95% of CA-I is irreversibly inactivated with less than a 5% loss of CA-II (Gothe and Nyman, 1972; Conroy and Maren, 1985). The molar equivalences of 1 eu for CA II and CA I at O°C and 100% CO2 (70 raM) may be derived from the rate expressions for the two isozymes. V~ -
koat "S" E
(1)
Sq-K m
where S is the substrate concentration in the cell. At 0°C and for median pH of 7.6, kc~t and K~ are 270 000 sec -1 and 10 raM, respectively, for CA II, and 29 650 sec -1 and 4 mM, respectively, for CA I (Sanyal and Maren, 1981). Rearrangement of eqn (1) to give E yields F - Gds
+ Kin)
S" k0~t
(2)
For the case of 1 eu in the system V0~t= V,,~c~t. Titration of 0.2 ml 50 mM barbital from a starting pH of 8"2 to pH 7.2 (phenol red end-point) represents acidification of 5 8 5 0 # M of the added 14300,uM buffer. This concentration divided by (/~c) gives a velocity ibr the system of 167 #M sec -1. Substituting this V,n~t for Vc~t in eqn (2) gives the molar equivalences for 1 eu for the two isoenzymes as follows: CA-II = 167 #M sec-1 (70 mM + 10 raM) 70 mM X 270 000 sec -1 = 0"71x 10 -9 M
(3)
CA-I = 167 #M sec -1 (70 r a M + 4 raM) 70 mM X 29 650 sec -1 = 6.0x 10 9M.
(4)
Note that this conversion of eu to molarity is independent of the reaction volume of the assay but depends on temperature and substrate concentration. The conversion of eu g-1 tissue to molarity of enzyme, however, does depend on the reaction volume. From eqn (3): 1 eu = 0"71 x 10 -9 tool -1 = 5 x 10 -13 mol (in 0.7 ml reaction volume)
(5) 5 x 10 -la tool Then (taking g -~ ml) 1 eu g-~ 10 -31
(6)
= 0"5 x 10 -9 mol 1-1 (M). Thus, 100 eu gm -~ of CA-II in the 0.7 ml system would correspond to 100 (0.5 x 10 -9 M) = 0"05 #M CA II. The equivalency of eqn (3) has also been made by titration of enzyme with powerful inhibitors such as ethoxzolamide, yielding for CA II 1 eu = 1 x 10 -9 M (Maren, Parcell and Malik, 1960), now modified by recent titration to 0"7 x 10 -9 M, the same as obtained from the rate of eqn (3). Because enzyme units differ between buffer systems and the units g-1 tissue depend on the volume of the reaction, we show the relations (Table f) among these
CARB'ONICANHYDRASE IN CORNEAL EPITHELIUM
639
TABLE H
Carbonic anhydrase activities in selected tissues*
Source Red Cell Lens Ciliary process Corneal endothelium Corneal epithelium
CA-II (eu gm 1)
CA-I (eu gm -1)
CA-II
Species
Total CA activity (eu gm -1)
Human Rabbit Dog Rabbit Rabbit Rabbit Rabbit
45 800 20 750 31800 6000 1600 450 150
24 000 12 700 23 100 6000 1600 450 80
21 800 8050 8700 0 0 0 65
12"0 6"4 11"6 3"0 0"8 0"23 0"04
CA-I
Total CA
~M) 91"6 33"8 36"5 0 0 0 0'27
103"6 40"2 48"1 3'0 0'8 0'23 0'31
* See Materials and Methods section for conversion of eu gm 1 to #M.
TABLE HI
Isoenzyme distribution in corneal epithelium
Species Rabbit Hum an Dog Cat Sheep Sheep (fetal)
Average weight of epithelium (rag)
n
10 10 15 20 20 10
4 5 4 2 2 2
Total CA activity (eu g-l)
CA-II* (eu g-* tissue)
CA-I (eu g-1 tissue)
CA-II (riM)
145 _+38 88 __+9 45 + 14 12 15
79+27 67_+8 35_+6
65_+36 21_+5 10_+3
40 34 18
N.D.J"
CA-I (riM)
273 88 42
--
--
6
--
--
--
7
--
.
.
.
.
Total CA (riM) 313 122 60 6 7 0
* Activity remaining after bromopyruvate inactivation. t Not detected.
systems, including our original 7 ml carbonate procedure (see Table 16, Maren, 1967). It will be seen that the relation between eu g-1 tissue in the large cell carbonate system and the present small cell barbital system is 17/0.5 or 34, i.e. the numbers in Maren (1967) x 34 equal the present values.
3.
Results
and
Discussion
The CA activity in selected tissues of the rabbit eye is shown in Table II along with the content in red cell hemolysate from three species. The levels reported here in the barbital system for blood, lens, and ciliary process are similar to those previously reported by Ballintine and Maren (1955) using bicarbonate buffer when allowances are made for inhibition caused by this buffer. The ciliary process and corneal endothelium were found to contain only the high activity CA-II isoenzyme in agreement with previous studies. Although CA levels in these tissues are low compared to erythrocyte CA, the enzyme has a well defined function in both, In general, CA-II m a y be found as soluble cytoplasmic enzyme in a number of electrolyte transporting epithelia. It facilitates the m o v e m e n t of
water coupled to Na + and HCO a- (Maren, 1988), the latter being generated from COy In the case of the corneal endothelium, the movement of fluid from the stroma into the aqueous h u m o r is thought to maintain the clarity of the cornea. In the ciliary process the enzyme serves a secretory role giving rise to the aqueous humor. Incubation of lens homogenate with bromopyruvic acid did not result in a decrease in CA activity, consistent with an absence of CA-I in this tissue. This is in disagreement with a previous immunohistochemical study which found both CA-I and CA-II staining in the lens (Wistrand et al., 1986). Friedland and Maren (1981 ) have previously discussed the possible role of CA in the elimination of CO,, from the lens across narrow concentration gradients. The CA content and isoenzyme distribution of epithelial homogenates from several species are shown in Table III. Only the first three of the homogenates showed sufficient CA activity to permit estimation of CA-I and CA-II ratios. In rabbit and dog the activity ratio of CA-I to CA-II approximated that found in the red cell of that species. In all cases inclusion of 10 -7 M methazolamide in epithelial homogenates reduced the CO2 hydration activity to the uncatalyzed rate. The role played by CA in corneal epithelia is difficult
640
to clarify, although its function is probably unrelated to fluid secretion or ion transport. Even more difficult to explain is the presence of CA-I in this tissue since an exact role for CA-I in erythrocyte§ has never been found. The most plausible hypothesis is that epithelial CA-I and CA-II m a y facilitate the outward movement of metabolic CO~ across small concentration gradients. This facilitated diffusion occurs through the catalytic conversion of CO2 to HCO3-, as described for the lens (Friedland and Maren, 1981). According to this scheme inhibition of epithelial CA should lead to the establishment of a new and higher gradient for CO~ diffusion to ambient air. In support of this are recent findings by Candia (1990, 1991) which showed that lO-4M methazolamide reduced the flux of [14C]CO~ across the isolated corneal epithelium by 30 %. Further studies which involve measurement of CO2 gradients across the cornea during inhibition by topical CA inhibitors m a y provide additional evidence for the participation of epithelial CA in CO~ transport.
Acknowledgement This research was supported by National Institutes of Health grant EY-02227.
References Ballintine, E.J. and Maren, T.H. (1955). Carbonic anhydrase activity and the distribution of Diamox in the rabbit eye. Am. J. Ophthahnol. 40, 148-54. Birndorf, L., Kitada, S., Shapourifar-Tehrani, S. and Lee, D. A. (1990). Developmental expression of carbonic anhydrase isoenzymes in human eyes. Invest. Ophthalrnol. Vis. Sci. 31 (Suppl.), 155.
C.W. C O N R O Y ET AL.
Candia, O.A. (1990). Forskofin-induced HCO3- current across apical membrane of the frog corneal epithelium. Am. 1. PhysioL 259, C215-33. Candia, O.A. (1991). The movement of CO~ and HCO~across frog epithelium. Evidence for the presence of carbonic anhydrase. Invest. Ophthalrnol. Vis. Sci. 32 (Suppl.), 977. Conroy, C. W. and Maren, T. H. (1985). The determination of osteopetrotic phenotypes by selective inactivation of red cell carbonic anhydrase isoenzymes. Clin. Chim. Acta 152, 347-54. Friedland, B.R. and Maren, T.H. (1981). The relation between carbonic anhydrase activity and ion transport in elasmobranch and rabbit lens. Exp. Eye. Res. 33, 545-61. Gloster, J. (1955). Investigation of the carbonic anhydrase content of the cornea of the rabbit. Br. ]. Ophthalmol. 39, 743-6. Gothe, P.O. and Nyman, P.O. (1972). Inactivation of human erythrocyte carbonic anhydrase by bromopyruvate. FEBS Lett. 21, 159-64. L6nnerholm, G. (1974). Carbonic anhydrase in the cornea. Acta Physiol. Scand. 90, 143-52. Maren, T.H. (1960). A simplified micromethod for the determination of carbonic anhydrase and its inhibitors. J. PharmacoL Exp. Ther. 130, 26-9. Maren, T.H. (1967). Carbonic anhydrase: chemistry, physiology, and inhibition. Phgsiol. Rev. 47. 595-781. Maren, T. H. (1988). The kinetics of HCQ synthesis related to fluid secretion, pH control, and CO~ elimination. Ann. Rev. Physiol. 50, 695-717. Maren, T.H., Parcefi, A.L. and Mafik, M.N. (1960). A kinetic analysis of carbonic anhydrase inhibition. J. Pharmacol. Exp. Ther. 130, 389-400. Sanyal, G. and Maren, T.H. (1981). Thermodynamics of carbonic anhydrase catalysis, a comparison between human isoenzymes B and C. J. Biol. Chem. 256,608-12. Silverman, D. N. and Gerster, R. (1973). The detection and localization of carbonic anhydrase in the rabbit cornea. Exp. Eye Res. 17, 129-36. Wistrand, P. J., Schenholm, M. and L6nnerholm, G. (1986). Carbonic anhydrase isoenzymes CA I and CA II in the human eye. Invest. Ophthalmol. Vis. Sci. 27, 419-28.
