Calcium-dependent regulation of adenylate cyclase and phosphodiesterase activities in bovine lens: Involvement of lens calmodulin

Exp.

Eye

Res.

(1985),

41, 239-247

Calcium-dependent Regulation of Adenylate Phosphodiesterase Activities in Bovine Involvement of Lens Calmodulin J. C. BIZEC*,

J. KLETHIt

AND

and

P. MANDELt

* Luboratoires H. Faure, 101, route de la Cali~ornie, 07100 t Centre de Neurochimie du CNRS, 5 rue Blaise 67084 Strasbourg Cedez, France (Received 30 January

Cyclase Lens:

Annmay, Pascal,

France,

1985 and accepted 29 May 1985, London)

When calmodulin levels were determined in bovine lens layers at different ages, the values in the epitheliol cell layer were strikingly higher than in the cortical layer and higher than in the nucleus. In the epithelial cell layer, except in very old animals, the calmodulin levels were maintained in adult animals. In the lens nucleus the extremety low level of calmodulin decreased during aging. In the cortical layer there were no systematic changes of calmodulin level during aging. In vitro, low calcium concentrations (micromolar) activated and higher calcium concentrations (over 199 PM) inhibited the lens adenylate cyclase activity. CAMP and cGMP degradation by phosphodie&erase was activated by calcium. It is suggested that calmodulin might be involved in the regulation of both adenylate cyclase and phosphodiesterase activities of lens epithelial cells and that free calcium plays a well-defined role in CAMP synthesis. Key words: calmodulin; bovine lens; aging; epithelial cells; cortex; nucleus: calcium action; CAMP phosphodiesterase; cGMP phosphodiesterase; adenylate cyclase.

1. Introduction Calmodulin is an ubiquitous calcium-binding protein in mammalian cells and is involved in the activity of several enzymes, including phosphodiesterases (Cheung, 1967,1969,1970; Kakiuchi, Yamazaki and Nakajima, 1969; Kakiuchi and Yamazaki, 1970; Lin and Cheung, 1980) adenylate cyclase (Brostrom, Huang, Bruckenridge and Wolff; 1975; Cheung, Bradham, Lynch, Lin and Tallant, 1975), Ca-ATPase (Bond and Clough, 1973) and phosphorylase kinase (Cohen et al., 1978). Using calmodulin-free phosphodiesterase, Liu and Schwartz (1978) demonstrated the presence of calmodulin in the whole lens. However, these authors could not demonstrate the presence of Ca2+-dependent phosphodiesterase activity in the whole lens. More recently, a role of calmodulin in calcium transport at the site of lens fiber junctions (Perracchia, Bernardini and Perracchia, 1981; Hertzberg and Gilula, 1981; Welsh, Aster, Ireland, Alcala and Maisel, 1982; Perracchia, Bernardini and Perracchia, 1982), and in lens ATPase activity (Iwata, Schirasawa and Takehana, 1982) has been suggested. Since in the lens, layers of proliferating and non-proliferating cells can be separated, it seemed to us interesting to assay the calmodulin levels and explore the possible involvement of this molecule in cyclic nucleotide regulation in the two cell types. Variations in both adenylate cyclase and phosphodiesterase activities in the epithelial and fiber cells during aging have previously been demonstrated (Klethi, Bizec and Mandel, 1981; Bizec, Klethi and Mandel, 1982). Correspondence should be addressed Strasbourg Cedex, France.

0014-4835/85/020239+

9 $03.00/O

to: J. C. Bizec, Centre

de Neurochimie,

0 1985 Academic

5, rue Blaise Pascal,

Press Inc. (London)

67984

Limited

240

.1. (‘. K17EC 1

El’

AI,

2. Methods Tiwue preparation

and biochemical assays

Lenses were quickly isolated after animal death, and ciliary bodies and zonula were discarded. The epithelium-adhering capsule was gently removed by an incision on the posterior part of the lens. The cortex was scraped from the superficial decapsulated lens. The deeper cortex lens area was discarded to obtain a central zone, the lens nucleus, well separated from outer fibers. In order to determine the whole calmodulin of lens layers, cortical and nuclear layers were homogenized in 1,530 volumes of 100 mM Tris-HCl (pH 7.8) containing 2 mM EGTA. and epithelial cell extracts were prepared in 2 ml of the same buffer per lens. TABLE

Efect of ageing on calmdulin

I

levels in bovine lens layer Calmodulin

Age Calf Adult Old Very old * Duplicate

Average

lens wet weight

(g)t

1.3 1.9 2.3 2.5

Cortex

101.94a+21.21 10277b + 2.85 [email protected] 65.24d & 1.51

1474e+0.79 1915Pf 1.20 11.92gk1.76 1857”+ 1.44

assays of experiments

of ten lenses each i significantly ab, bc* “1 W- Ib- not significantly different P>O.O5. Average wet weight of ten lenses each. t There is a good correlation between lens wet weight and Muller, 1963).

were centrifuged (EGTA extraction)

ng mg-’

Epithelium

ae, ei. bf. II, C&C, 8k. dh. hl. cd, ef. fg. gh. II. ,k. hl

Homogenates This procedure

levels*

S.E. different

protein) NUCklS

9.87l+ 1.27 6.12’*0.18 2.02’ f 006 0.18’fO.Ol

P
and logarithm

(100000 g) for 20 min, and pellets provides almost all the calmodulin

of age (Ho&win.

were twice (Wallace,

Schmutter

re-extracted. Tallant and

Cheung, 1980). Gathered supernatants were boiled for 5 min and then centrifuged (10000 g). Calmodulin RIA kits (NEN) based on a radioimmunoassay using 1a51-labelled calmodulin were used to determine calmodulin levels in supernatants. Each value was calculated in terms of soluble proteins measured by the method of Lowry, Rosebrough, Farr and Randall (1951). In order to determine enzyme activities, lens epithelial layers were homogenized in 10 mM Tris-HCl buffer pH 7.8. The methods of measurement of adenylate cyclase, and CAMPand cGMP-phosphodiestera activities were carried out as previously described (Klethi et al., 1981; Bizec et al., 1982). Calcium levels of the incubation media were recalculated after Ca2+ measurement on nitric-acid-mineralized extract by atomic absorption. Trifluoperazine and chlorpromazine were purchased from Sigma. Calmidazolium was obtained from Boehringer, Mannheim.

3. Results

and

Discussion

Calmodulin is present in all bovine lens layers. The epithelial cell layer contained strikingly higher calmodulin levels than that found in cortical fiber cells or in the nucleus (Table I). The calmodulin levels in epithelial cells were not found to vary with age except in very old animals (Table I). There were no systematic changes of calmodulin levels in the cortical layer during aging. In the nuclear layer the calmodulin levels decreased during aging (Table I). Adenylate cyclase and phosphodiesterase activities in bovine lens epithelial cells have been demonstrated (Klethi et al., 1981; Bizec et al., 1982). EGTA produced an

CALMODULIN

AND

CYCLIC

l/[ATP]

241

NUCLEOTIDES

(mt.+-‘l

Fm. 1. Effect of EGTA on adenylate cyclsse activity of crude epithelium: kinetic study. Double reciprocal plot; l/specific activity ~8. i/[ATP]. The incubation medium samples contained @I mM calcium. Inset: EGTA dose-effect relationship for adenylate cyclase activity of the particulate fraction of the lens epithelium homogenate. No activity was measured in the soluble fraction of homogenate (100000 8 centrifugation). The calcium concentration of the initial incubation medium was 100 PM before EGTA addition. The activity of 60-PM-EGTA-containing incubation medium did not differ significantly from that of the initial activity (0 PM EGTA) while in presence of @6 m&i EGTA the activity of the treated homogenate w&s much lower. TABLE

Effects of addition

EGTA oal @lo 0.10 0.10 0.10 0.10

(mM)

II

and calcium to the incubation medium cyclaseactivity of bovine lens epithelium

of EGTA

Calcium 004 0.04 0.14 024 0.29 0.44

(mM)*

on adenylate

Adenylate cyclase (pm01 min-’ mg-’ protein)t 5662af 1.28 20.01brt3.58 5904c*451 4048df 1.02 352ge f 8-53 27.67’ +_ 323

(3) (3) (2) (2) (3) (3)

* Calcium content of incubation medium. The initial incubation medium (0 mM EGTA) 0.04 mM calcium. t Activity averages are represented fS.E. with number of assays in parentheses. Activities were estimated on total epithelium homogenate. ab* bc* cd- ef are significantly different P @05.

contained

J. (‘. HI%E(’

I/[ATP]

ET

AI,

h-‘I

FIG. 2. Effect of addition of @4 rn~ calcium on adenylate cyclase activity of crude epithelium homogenate: kinetic study. Lineweaver-Burke plot is used; i/specific activity vs. i/[ATP]. The control incubation medium contained CO9 m&s calcium.

inhibition of adenylate cyclase activity (Fig. 1). A significant decrease of adenylate cyclase activity was noticeable at 96 mM EGTA but not at 606 mM; the calcium content of the incubation medium was 91 mnr (Inset of Fig. 1). At low levels of ATP, the kinetic study (64 mM EGTA) showed a decrease of V,,, (about 85%) without alteration of Km (about 175 pM) (Fig. 1). At high ATP levels the Km increased ( + 78 %) (Fig. 1). The inhibition of adenylate cyclase activity of crude homogenate by 109 ,UM EGTA was reversed by stoichiometric addition of calcium (Table II). Further addition of calcium inhibited enzymatic activity (Table II). The observed calcium effect was analogous to the classical biphasic calcium effect reported for brain cells (Bradham, Holt and Sims, 1970). The kinetic study at high concentrations of calcium demonstrates a decrease of activity and an increase of the apparent Km ( + 63 y0 at high ATP levels) in epithelial lens cells (Fig. 2). Chlorpromazine, a phenothiazine derivative known to block calmodulin activity (Weiss and Hirt, 1977) inhibited adenylate cyclase activity of lens epithelial cells (Fig. 3). Trifluoroperazine was an even more potent inhibitor than chlorpromazine, and the V,, was strikingly lower (50 PM trifluopenazine; - 77 y0 ; 50 pM chlorpromazine ; - 48 %) (Fig. 3). The aflinity of adenylate cyclase for ATP did not change by the above-mentioned phenothiazine derivatives (Km about 175 ,uM) (Fig. 3). It is well known that calmodulin activates adenylate cyclase without changing the Km (Klee, Crouch and Richman, 1980). Calmidazolium, the very potent inhibitor of calmodulin activity, inhibited lens adenylate cyclase activity in a similar way (Fig. 3) CV,, lowered by 52 y’ in presence of 10 pM calmidazolium). When enzymatic degradation of cyclic nucleotides was examined, it appeared that l&100 PM EGTA decreased the activity of cGMP-phosphodiesterase by 80 %. At higher concentrations, EGTA did not significantly decrease this enzymatic activity (Fig. 4). EGTA-inhibited phosphodiesterase activity (CAMP and cGMP degradation) was restored by the addition of calcium (Table III). Increasing quantities of chlorpromazine inhibited both CAMP- and cGMP-phosphodiesterase activities to a

CALMODULIK

AKD

CYCLIC

SUCLEOTIDES

50 I/[ATP]

I00

(rn~-‘l

Fm. 3. Effect of calmodulin inhibitors on adenylate cyclase activity of crude homogenate of bovine lens epithelium: kinetic study. Lineweaver-Burke plot is used; l/specific activity vs. l/[ATP]. The calcium contents of the respective incubation media were 606 mM for chlorpromazine and trifluoperasine experiments and 609 mx for calmidazolium experiment. Inset: Dose-effect relationship of chlorpromazine on adenylate cyclase activity of bovine lens epithelium. The calcium content of the incubation medium was 0.09 mzvr. 4

F’

i

O+k-Tk-Y

IO

EGTA (/aI

Fm. 4. Effect of EGTA content of the incubation

on cGMP-phosphodiesterase medium was 13 8~.

activity

of bovine

lens epithelium.

The calcium

EGTA

(mu) 0.0 @2 0.2 0.2 @2 0.0

Ca (mM)* 0.05 WO5 0.25 045 205 2.05

CAMP-phosphodieate~asr (pm01 ruin-’ mg-’ protein)

t

clomp-I’hosphodiesteras~ (nmoles min-’ mg-’ protrin)t 4%6*Wl5 122+0.14 512+0.40 4YZfti12 $75 * 0.30 519+w24

482 + 23 182&- 4 4852 4 459 + 39 467k 3 456 k 24

* Calcium content of incubation medium. The initial incubation contained @05 rnx t Assays in triplicate + s.E.(M.). EGTA and calcium effects on enzymatic CAMP and cGMP hydrolysis homogenates. Substrate levels were 14 ,UU# cAMP and 125 pM cGMP.

medium

(EGTA

were estimated

non-treated)

on non-dialyzed

lower level (Fig. 5). Chlorpromazine (250 ,UM) produced a 69% decrease in CAMPphosphodiesterase activity whereas only a 40 oh decrease in cGMP-phosphodiesterase activity was observed (Fig. 5). Half-maximal inhibition of the calcium-dependent CAMP- and cGMP-phosphodiesterases activities was observed in concentrations of chlorpromazine ranging between 30 and 40 PM. Investigation of adenylate cyclase and phosphodiesterase activities suggests that calmodulin is involved in the regulation of lens epithelium cyclic nucleotides. The ‘two-step’ action of calcium on adenylate cyclase described in other tissues is also observed in lens epithelium (Bradham et al., 1970). It seems likely that calcium entry into epithelial lens cells plays a fundamental role in the synthesis and degradation of cyclic nucleotides in the different compartments. We have found that lens adenylate cyclase activity is only present in the particulate fraction of epithelial cells whereas phosphodiesterase activity is found in the soluble fraction (data not shown). When calcium was added to calcium-deficient lens epithelium homogenate, by binding to adenylate cyclase and phosphodiesterases were calcium-protein-like calmodulin, reactivated. Moreover, if the free calcium content increases again a secondary effect of inhibition occurred for adenylate cyclase whereas it did not for phosphodiesterase. It appears that free calcium plays an important and well-defined role in the synthesis of CAMP. The calmodulin content in the whole lens is much lower than in other tissues. In the epithelial cells, however, calmodulin levels are the highest but still lower than in brain (Hanbauer and Costa, 1980). During lens epithelial cell differentiation into fiber cells, the decrease in calmodulin levels (Table 1) is comparable to the corresponding loss of activity of enzymes involved in cGMP synthesis (no activity in the cortex) and CAMP or GMP degradation (Klethi et al., 1981; Bizec et al., 1982). We cannot exclude the possibility that calmodulin is involved in this process. We have observed recently that the majority of lens epithelial cell phosphodiesterase isoenzymes are dependent on calmodulin. When adenylate cyclase, CAMP- and cGMP-phosphodiesterase activities were measured in lens epithelium, a significant decrease of activities was observed during aging (Klethi et al., 1981; Bizec et al., 1982). In contrast, epithelium calmodulin levels remained

CALMODULIN

AND CYCLIC

NUCLEOTIDES

245

r/ 5+ 1000

% ti

800 c 0, 5 h ig I, E 0 E

600

a

400

1

\_II

< *\

l\ =\a

200

I

0 ”

7!8

$6

6:6

lb5

Chlorpromazine

31.3

(p)

FIQ. 5. Effect of chlorpromazine on CAMP (-@-a-) diesterase activities of bovine lens epithelium. Activities epithelium homogenate: calcium content 8 ,uL”.

were

2k0

do

lOi

and cGMP (-O-O-)-

phospho-

measured

of crude

on supernatants

practically constant (variations of bovine lens wet weights 1.3-23 g). Thus, the decrease in the enzyme activity cannot be ascribed to variations of calmodulin content. The influence of age on the enzymes may be attributed either to a decrease in the synthesis or to a structural modification of the apoenzymes concerned. Alternatively, changes in the calmodulin binding proteins cannot be excluded. In very old lenses (2.5 g wet weight), a significant decrease of calmodulin content of epithelial cells was observed (Table I). In this case, a further reduction in calmodulin-dependent phosphodiesterase activity cannot be totally ruled out. In the cortical layer, calmodulin levels (Table I) are low and rather variable during aging and cannot account for the loss of CAMP-phosphodiesterase activity (Klethi et al., 1981). In the nucleus, continuous loss of CAMP-phosphodiesterase activity (Klethi et al., 1981) and also a decrease in calmodulin levels (Table I) were observed. Thus, during aging this decreese might be involved in the loss of the phosphodiesterase activity. Finally, although it is likely that during aging variations in calmodulin contents are not solely responsible for the changes in phosphodiesterase activity, it is, however, possible that the decrease in calmodulin levels may be an additional factor involved in the loss of phosphodiesterase activity in the nucleus or in the epithelial cells of very old animals.

246

J.

C. UIZEC

ET

AL

REFERENCES Al&a,

J., Welsh, M. and Maisel, H. (1982). Calmodulin calcium binding protein of the crystallin lens. Fifth Intern. Congress of Eye Res., 84. Int. Sot. for Eye Res., New York. Bizec, J. C., Klethi, J. and Mandel, P. (1982). Cyclic guanosine 3’5’.phosphate, guanylate cyclase and cyclic guanosine 3’5’-phosphate phosphodiesterase. Biochem. Biophys. Res. Commun. 106, 108-12. Bond, G. H. and Clough, D. L. (1973). A soluble protein activator of (Mg2+-Ca2+)-dependent ATPase in human red cell membranes. Biochim. Biophya. Acta 323, 592-99. Bradham, L. S., Holt, D. A. and Sims, M. (1970). The effect of Ca2+ on the adenyl cyclase of calf brain. Biochim. Biophys. Acta 201, 250-60. Brostrom, C. 0.. Huang, Y. C., Bruckenridge, B. M. and Wolff, D. J. (1975). Identification of a calcium-binding protein as a calcium-dependent regulator of brain adenylate cyclase. Proc.

Nat1

Acad.

Sci.

U.S.A.,

72, 64-8.

Cheung, W. Y. (1967). Cyclic 3’5’nucleotide phosphodiesterase: pronounced stimulation by snake venom. Biochem. Biophys. Res. Commun. 29,478-82. Cheung, W. Y. (1969). Cyclic 3’5’.nucleotide phosphodiesterase, preparation of partially inactive enzyme and its subsequent stimulation by snake venom. Biochim. Biophys. Acta 191,303-15. Cheung, W. Y. (1970). Cyclic 3’5’-nucleotide phosphodiesterase. Demonstration of an activator. Biochem. Biophys. Res. Commun. 38, 533-38. Cheung, W. Y., Bradham, L. S., Lynch, T. J., Lin, Y. M. and Tallant, E. A. (1975). Protein activator of CAMP-phosphodiesterase of bovine or rat brain also activates its adenylate cyclase. Biochem. Biophys. Res. Commun. 66, 1055-62. Cohen, P., Burchell, A., Foulkes, J. G., Cohen, P. T. W., Vanaman, T. C. and Nairn, A. C. (1978). Identification of the Ca 2+-dependent modulator protein as the fourth subunit of rabbit skeletal muscle phosphorylase kinase. FEBS Lett. 92, 287-93. Hanbauer, I. and Costa, E. (1980). Role of calmodulin in dopaminergic transmission. In Calcium and Cell Function, Vol. I (Ed. Cheung, W. Y.) Pp. 253-72. Hertzberg, E. L. and Gilula, N. B. (1981). Liver gap junctions and lens fibers junctions: comparative analysis and calmodulin interaction. Cold Spring Harbor Symposia on Quantitative

Biology

46, 639.

Hockwin, O., Schmutter, J. and Muller, H. K. (1963). Untersuchungen uber Gewicht and Volumen verschieden alter Rinderlinsen. Albrecht V. Graefes. Arch. Ophthalmol. 166, 13651. Iwata, S., Shirasawa, E. and Takehana, M. (1982). Calcium-modulator in the lens. Fifth Intern. Congress of Eye Res., 84. Int. Sot. for Eye Res., New York. Kakiuchi, S. and Yamazaki, R. (1970). Calcium-dependent phosphodiesterase activity and its activating factor (PAF) from brain. Studies on cyclic 3’5’-nucleotide phosphodiesterase (III). Biochem. Biophys. Res. Commun. 41, 1104-10. Kakiuchi, S., Yamazaki, R. and Nakajima, H. (1969). Bull. Jap. Neurochem. &‘oc. 8, suppl., 17. Klee, C. B.. Crouch, T. H. and Richman, P. G. (1980). Calmodulin. Ann. Rev. Biochem. 49, 489515. Klethi, J., Bizec, J. C. and Mandel, P. (1981). Adenosine 3’5’-phosphate, adenylate cyclase, AMP cyclique phosphodiesterase du cristallin au tours du vieillissement. C.R. Acad. Sci. Paris 293, 41922. Lin, Y. M. and Cheung, W. Y. (1980). Ca2+-dependent cyclic nucleotide phosphodiesterase. In Calcium and Cell Function, Vol. I. (Ed. Cheung, W. Y.) Pp. 79-107. Academic Press, New York. Liu, Y. P. and Schwartz, H. S. (1978). Protein activator of cyclic AMP phosphodiesterase and cyclic GMP phosphodiesterase in bovine lens and in bovine retina. Activity, subcellular distribution and kinetic parameters. Biochim. Biophys. Acta 526, 186-93. Lowry, 0. H., Rosebrough, N. J., Far-r, A. L. and Randall, R. J. (1951). Protein measurement with the folin phenol reagent. J. Biol. Chem. 193, 265-75. Peracchia, C., Bernardini, G. and Peracchia, L. L. (1981). A calmodulin inhibitor prevents gap junction crystallization and electrical uncoupling. J. Cell Biol. 91, 124a.

CALMODULIN

Peracchia,

AND

CYCLIC

NUCLEOTIDES

247

C., Bernardini, G. and Peraechia, L. L. (1982). Calcium effects on lens junctions. Congress of Eye Res., &1. Int. Sot. for Eye Res, New York. Wallace, R. W., Tallant, E. A. and Cheung, W. Y. (1980). Assay, preparation and properties of calmodulin. In Cd&m and Cell Function, Vol. I. (ed. Cheung, W. Y.). Pp. U-40. Academic Press, New York. Weiss, B. and Hirt, W. N. (1977). Selective cyclic nucleotide phosphodiesterase inhibitors as potential therapeutic agents. Ann. Rev. Pharmacol. Toxicol. 17, 441-77. Welsh, M. J., Aster, J. C!, Ireland, M., Alcala, J. and Maisel, H. (1982). Calmodulin binds to chick lens junction protein in a calcium-independent manner. Science 216, 64243. Fifth

Intern.