THE JOURNAL OF BIOLOGICAL CHEMISTRY 0 1991 by The American Society for Biochemistry and Molecular Biology, Inc. Vol. 266, No. 34, Issue of December 5, pp. 22954-22959, 1991

Printed in U. S. A.

ATP-dependent Regulation of Flagellar Adenylylcyclase in Gametes of Chlamydomonas reinhardtii”

(Received for publication, August 6, 1991)

Yuhua ZhangS, Elliott M. Ross#, and William J. SnellSV From the iDemrtment of Cell Biolom and Neuroscience and $Department of Pharmacology, University of Texas Southwestern Medical School, Dallas, Teras 7523,f”

Adenylylcyclase activity in the flagella of gametes of Chlamydomonas reinhardtii was inhibited by prior incubation at or below 30 OC in the presence of ATP. This decrease did not occur in the absence of ATP, in the presence of the ATP analog 6’-adenylylimidodi- phosphate (App(NH)p), or in the presence of ATP plus the protein kinase inhibitor staurosporine (2 PM). If ATP treatment was performed in the absence of an ATP-regenerating system, activity initially declined and subsequently recovered. Incubation of flagella at 45 “C in the absence of ATP or incubation at lower temperatures in the presence of either App(NH)p or staurosporine both increased adenylylcyclase activity (over 10-fold) and blocked subsequent ATP-dependent loss of activity at 30 OC. This heat-induced activation was prevented by the presence of ATP plus an ATP- regenerating system. Incubation of flagella with [y- 3aP]ATP followed by gel electrophoresis in sodium do- decyl sulfate indicated the presence of endogenous pro- tein kinase and protein phosphatase activities. These data suggest that the flagellar adenylylcyclase in Chlamydomonas gametes is inhibited by phosphoryla- tion and stimulated by dephosphorylation. This mech- anism for regulating adenylylcyclase may underlie the rapid increase in cyclic AMP that is induced by flagel- lar adhesion during fertilization in Chlamydomonas.

We have begun to study cell contact-mediated signal trans- duction during the mating reaction in the unicellular, bi flagellated green alga, Chlamydomonas. When gametes of opposite mating types (mt’ and mt-) are mixed together, they adhere to each other via adhesion molecules, agglutinins, on the surfaces of their flagella (1). The adhesive interaction between mt+ and mt- flagellar agglutinins causes an imme- diate 10-fold increase in intracellular cyclic AMP (2,3), which leads to several events in fertilization that can be mimicked by addition of exogenous dibutyryl cyclic AMP. These events that accompany sexual signaling include: (i) secretion of a serine protease, which converts a prometalloprotease to an active enzyme that hydrolyzes the cell wall (4, 5); (ii) move- ment of preexisting agglutinins from the plasma membrane of the cell body onto the continguous flagellar membrane (6- 8); (iii) formation of an actin-filled mating structure (9, 10); and (iv) cell fusion to form a zygote.

With the goal of learning how the interaction between mt+

* This work was supported by National Institutes of Health Grant GM-25661 (to W. J. S.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

2349. ll To whom correspondence should be addressed. Tel.: 214-688-

and mt- agglutinins is coupled to the increase in intracellular cyclic AMP, we have decided first to study the regulation of flagellar adenylylcyclase in gametes of a single mating type. In the studies reported here we show that the activity of flagellar adenylylcyclase of mt+ gametes is increased in vitro either by heating or by an inhibitor of protein kinases. Fur- thermore, the adenylylcyclase can be inhibited by pretreat- ment with ATP, but not with the nonhydrolyzable ATP analog, App(NH)p.’ Our results suggest that phosphorylation and dephosphorylation may play an important role in regu- lation of flagellar adenylylcyclase activity in Chlamydomonas.

EXPERIMENTAL PROCEDURES

Materials-HEPES was from Research Organics Inc., Cleveland, OH; pyruvate kinase was from Boehringer Mannheim GmbH, Federal Republic of Germany; H-7 and H-8 were from Seikagaku Kogyo Co., Tokyo, Japan; W-7 was from Calbiochem; tyrphostin was a gift from Dr. Alain Schrieber of Rorer Central Research, King of Prussia, PA; [3H]cyclic AMP was from ICN Biomedicals Inc., Boston, MA; di- methyl sulfoxide was from J. T. Baker Inc.; R020-1724 was from Biomol (Plymouth Meeting, PA); EDTA, Tris, sucrose, BSA, PEP, imidazole, cyclic AMP, ATP, App(NH)p, GTP, papaverine, leupeptin, lima bean trypsin inhibitor, TPCK, TLCK, phenylmethanesulfonyl fluoride and staurosporine were from Sigma.

Cells and Cell Culture-Chlamydomonas reinhardtii strains 21GR (mt’) and 6145C (mt-) (available from the Chlamydomonm Genetics Center, Duke University) were cultured at room temperature in medium I or medium I1 of Sager and Granick (11) on a 12-h light- dark cycle. Gametic cells were obtained as previously described (12) by transferring vegetative cells (4-7 X lo6 cells/ml) 6 h after the beginning of the light period into nitrogen-free medium (12) modified to contain 0.15 g/liter KH2P04 and 0.3 g/liter K,HPO,.

Isolation of Flagella-Gametic flagella were separated from cell bodies by a modification of the pH shock method of Witman et al. (13). Gametic cells were harvested by centrifugation at 3,000 rpm for 3 min resuspended in ice-cold 10 mM Tris-HC1, 7% sucrose (pH 7.2). The pH of a stirred suspension of cells was rapidly lowered to 4.4 by the dropwise addition of 0.5 M acetic acid. The sample was held at this pH for about 1 min, and then the pH rapidly was raised to 7.2- 7.4 by the dropwise addition of 0.5 M KOH. The suspension was centrifuged at 3,000 rpm for 3 min at 4 “C in a conical polystyrene centrifuge tube (IEC centrifuge, rotor 253). The supernatant, which contained the flagella, was removed by aspiration and transferred to another centrifuge tube. The suspension was underlaid with ice-cold 10 mM Tris-HC1, 25% sucrose (pH 7.2) and was centrifuged at 3,000 rpm for 10 min at 4 “C. After the centrifugation, all of the supernatant above the 7-25% sucrose interface was collected by aspiration. In some cases, this procedure was repeated once. The supernatant was

The abbreviations used are: App(NH)p, 5’-adenylylimidodiphos- phate; H-8, N-[2-(methylamino)ethyl]-5-isoquinolinesulfonamide; H- 7, l-(5-isoquinolinylsulfonyl)-2-methylpiperazine; W-7, N46-amino- hexyl)-5-chloro-l-naphthalenesulfonamide; BSA, bovine serum al- bumin; TLCK, N“-p-tosyl-L-lysine chloromethyl ketone; TPCK, N- tosyl-L-phenylalanine chloromethyl ketone; R020-1724,4-(3-butoxy- 4-methoxybenzyl)-2-imidazolidinone; PEP, phosphoenolpyruvate; HEPES, 4-(2-hydroxyethyl)-l-piperazineethanesulfonate; GTPrS, guanosine 5’-O-(thiotriphosphate).

22954

Regulation of Adenylylcyclase in Chlamydomonas 22955

centrifuged at 15,000 rpm for 20 min at 4 "C in a Du Pont-Sorvall SA 600 rotor, and the sedimented flagella were resuspended in a buffer containing 20 mM HEPES, pH 7.2, 4% sucrose, 1 mM EDTA, 0.5% BSA, 2.5 mM MgCIZ, 0.05 mM GTP, and 0.1 mM papaverine or 0.1 mM R020-1724. After addition of a mixture of protease inhibitors (7 p~ leupeptin, 3.2 pg/ml trypsin inhibitor from lima bean, 60 p M TPCK, 60 p~ TLCK, 0.13 mM phenylmethanesulfonyl fluoride), the flagella were divided in small aliquots, frozen, and stored in liquid nitrogen until use.

Protein Determination-Protein was determined with the Coo- massie Blue protein assay reagent of Pierce Chemical Co. using crystalline BSA as standard.

Adenylylcyclase Assay-Adenylylcyclase activity was measured es- sentially according to the methods of Ross et al. (14). The 10O-pl final assay volume contained 20 mM Na-HEPES (pH 7.2), 1 mM EDTA, 2.5 mM MgCl2, 4% sucrose (w/v), 0.5% BSA (w/v), 50 p~ GTP, 30 pg/ml pyruvate kinase, 3 mM K2PEP, 0.1 mM R020-1724 or 0.1 mM papaverine, 0.5 mM ATP, and IO7 cpm/ml [a-"PIATP. GTP was included in initial experiments because of its known effects on other adenylylcyclases. Subsequently it was found to have no effect on Chlamydomonas flagellar adenylylcyclase, but for consistency it was included in later experiments as well. Unless otherwise noted, reac- tions were started by the addition of flagella (0.4-0.9 mg/ml final protein concentration) and were incubated at 30 "C for the times shown. Reactions were stopped and cyclic AMP was measured accord- ing to Salomon et al. (15). The zero time points represent samples with no protein. Results shown are the average of duplicate samples, which usually varied by less than 10%.

Polyacrylamide Gel Electrophoresis-SDS-PAGE was performed on 4-16% acrylamide gradient gels as previously described (4). Pre- stained, low molecular weight protein markers were from Bio-Rad.

RESULTS

Heat-induced Activation of Flagellar Adenylylcyclase-To determine the optimal temperature for assaying adenylylcy- clase in the flagella of Chlamydomonas gametes, activity ini- tially was assayed at different temperatures. Fig. 1 shows that adenylylcyclase activity was not linear with time at 25 or 30 "C but declined by 4 min after initiation of the assay. In contrast, the decline was less at 37 "C. To test whether this behavior might reflect the presence of a temperature-sensitive inhibitor, flagella were pretreated at various temperatures before being assayed a t 30 "C. As shown in Fig. 2, preincuba- tion of flagella at 40 "C both caused an absolute increase in the initial rate of adenylylcyclase activity and also caused activity to remain nearly linear over 10 min. The increase in initial rate was about 50%, but the nonlinear activity in the untreated sample caused the effective stimulation by the 40 "C treatment to approach 5-fold by the end of the 10-min assay. Flagellar adenylylcyclase activity was essentially un-

I 1

T L P

Y ' I I I I I I I 0 4 8 12 16 20 24 28

Time (rnin)

FIG. 1. Effect of assay temperature on adenylylcyclase ac- tivity. Adenylylcyclase activity in Chlamydomonas flagella was as- sayed as described under "Experimental Procedures" at 25 "C (O), 30 "C (O), and 37 "C (A). The protein contained in each assay was 60 pg.

Time (min)

FIG. 2. Effect of pretreatment at different temperatures on adenylylcyclase activity. Flagella (72 pg/assay) were incubated in buffer that contained 20 mM Na-HEPES, pH 7.2, 4% sucrose, 1 mM EDTA, 0.5% BSA, 2.5 mM MgCl,, 0.05 mM GTP, and 0.1 mM R020- 1724 for 15 min at the following temperatures: 40 "C (O), 18 "C (O), 10 "C (O), 4 "C (A), or 0 "C (0). Samples were then diluted 10-fold in the same medium, and the assay reaction was initiated at 30 "C by the addition of [cu-:'2P]ATP, PEP, and pyruvate kinase to the concen- trations given under "Experimental Procedures."

changed by preincubation at 4 or 10 "C, but preincubation at 18 "C actually caused a net decrease in activity

To determine the optimally activating temperature of pre- treatment, flagella were incubated at various temperatures before being assayed for adenylylcyclase activity. The results (data not shown) indicated that pretreatment at 45 "C for 10- 15 min produced the greatest activation. For these reasons a 12-min pretreatment at 45 "C was used for heat activation for most of the experiments presented below.

Several potential regulators of activity were tested in order to learn more about activation and inhibition of adenylylcy- clase. Compounds that displayed no marked effect during pretreatment at 30 or 45 "C included cyclic AMP; the phos- phodiesterase inhibitors papaverine and R020- 1724; BSA and several protease inhibitors (leupeptin, lima bean trypsin in- hibitor, TPCK, TLCK, and phenylmethanesulfonyl fluoride); and GTP and GTPyS. Low concentrations of the detergents Lubrol and digitonin did not cause a marked change in aden- ylylcyclase activity during the assay at 30 "C, indicating that accessibility of substrate was not limiting. In addition, thin layer chromatography followed by autoradiography showed that the concentration of [a-"PIATP in the assay medium was unchanged during the first 30 min of the assay.

Effect of ATP and Protein Kinase Inhibitors on Adenylyl- cyclase Activity-The data of Fig. 3 show that the presence of 0.5 mM ATP plus an ATP-regenerating system during the pretreatment at 45 "C blocked the heat-induced activation of adenylylcyclase almost completely. Although pretreatment at 45 "C for 12 min produced an 11-fold increase in activity over the non-pretreated sample, the inclusion of 0.5 mM ATP with an ATP-regenerating system virtually eliminated activation. Similar results were observed with pretreatment for 20 min at 30 "C. Such incubation without ATP caused about &fold activation, while similar treatment with ATP and a regener- ating system caused a 2-fold reduction of adenylylcyclase activity. In addition to its ability both to block activation and to reduce the adenylylcyclase activity in non-pretreated fla- gella, incubation with ATP also reduced the activity of par- tially activated adenylylcyclase. The data of Fig. 4 show that flagellar adenylylcyclase that was activated by pretreatment at 30 "C subsequently could be deactivated by incubation with ATP at 30 "C, indicating that activation was reversible.

22956 Regulation of Adenylylcyclase in Chlamydomonas

13

P

Time (min)

FIG. 3. Effect of ATP on the heat-induced activation of adenylylcyclase. Flagella were incubated in buffer containing 20 mM Na-HEPES, pH 7.2, 4% sucrose, 1 mM EDTA, 0.5% BSA, 2.5 mM MgCI,, 0.05 mM GTP, 0.1 mM R020-1724, 3 mM PEP, and 30 pg/ml PK either in the absence (0, A) or presence (0, A) of 0.5 mM ATP and an ATP-regenerating system (3 mM PEP and 30 pg/ml pyruvate kinase) either a t 30°C for 20 min (0, 0) or at 45°C for 12 min (A, A). Adenylylcyclase was then assayed at 30°C for the indi- cated times. Assays were initiated by the addition of [a-3ZP]ATP and unlabeled ATP to yield the same final specific acitivity. One sample ( 0 ) was held at 0°C prior to assay. Each assay contained 90 pg of protein.

I

0

01 I I I I I I 10 20 30 40 50 6(

Time of Pretreatment (min)

FIG. 4. ATP-dependent inhibition of partially activated ad- enylylcyclase. Aliquots of flagella were incuated a t 30°C in the buffer described in the legend to Fig. 3 without ATP or the regener- ating system (0). At 0 min (0) and 25 min (0) after the beginning of the incubation, 1 mM ATP, 10 pg/ml pyruvate kinase, and 6 mM PEP were added to some samples, and incubation was continued at 30°C. At the indicated times the samples were diluted 2-fold with the same buffer and then assayed for adenylylcyclase activity at 30°C for 8 min as described in Fig. 3. The protein content of each sample was 80 kg.

The effects of the protein kinase inhibitor staurosporine on activation and deactivation of adenylylcyclase are shown in Fig. 5 . Staurosporine (1 p ~ ) prevented the decline in ade nylylcyclase during the assay of untreated flagella. Adenylyl- cyclase activity was linear with time in the presence of 1 p M staurosporine and, at 8 min, was about 2-fold higher than in the absence of inhibitor. As shown in Fig. 6 ,2 p~ staurospor-

Time (min) FIG. 5. Effect of staurosporine on adenylylcyclase activity.

Untreated flagella were assayed for adenylylcyclase activity a t 30°C in the presence (0) or absence (0) of 1 p M staurosporine. For comparison, flagella also were incubated in the medium described in the legend to Fig. 3 a t 45°C for 12 min before assay (A). Each assay contained 80 pg of protein.

7

FIG. 6. Effect of staurosporine on the ATP-dependent in- hibition of adenylylcyclase activity. Flagella were incubated a t 30°C for the indicated t.imes in the buffer described in the legend to Fig. 3 in the presence of 1.5 mM ATP (0) or 1.5 mM ATP and 1 pM staurosporine (0), diluted 2-fold with the same buffer, and then assayed for adenylylcyclase activity a t 30°C for 8 min. The assay was initiated by the addition of [a-"P]ATP (lo6 cpm/sample). The con- centration of ATP in each sample during the assay was 0.75 mM, and the protein concentration was 0.8 mg/ml.

ine both prevented the ATP-dependent inhibition of ade nylylcyclase, and incubation at 30 "C in staurosporine under these conditions caused an increase of adenylylcyclase activity of about 70%. Stimulation by staurosporine at 30 "C was not as great as that caused by pretreatment at 45 "C without staurosporine (Fig. 5 ) , and the extent of activation at 45 "C was unaltered by the addition of 2 p~ staurosporine (not shown).

The ability of several other protein kinase inhibitors to block the ATP-dependent inhibition of adenylylcyclase was tested by pretreating flagella at 30 "C in the presence of ATP or ATP plus inhibitor for various times before assaying aden- ylylcyclase activity. Protamine (50 pg/ml) and heparin (10 pg/ml) blocked inhibition by 95 and 50%, respectively, but protamine also inhibited adenylylcyclase activity directly. Neither H-8 (40 p M ) (16), H-7 (10 pM) (161, w - 7 (10 pM) (17), tyrphostin (0.1 mM) (18), or NaCl (0.2 M) blocked the ATP-dependent inhibition by more than 15%.

App(NH)p Does Not Support Deactivation of Adenylylcy- clase-App(NH)p is a good substrate for mammalian ade-

nylylcyclase (19) but i s not a substrate for protein kinases. The experiments shown in Fig. 7 indicate that AppSNHIp did not support the ATP-dependent decrease in adenylylcyclase that routinely was observed at 30 "C. App(NH)p also permit- ted activation at either 30 or 45 "C. For these experiments flagella were preincubated with App(NH)p rather than ATP, and then tracer amounts of [(u-~*P]ATP were added to allow assay of adenylylcyclase. When flagella were assayed for 10 rnin in the presence of App(NH)p plus [a-"PJATP, but with no preincubation, adenylylcyclase activity was elevated al- most 2-fold. Preincubation at 45 "C in the presence of 0.5 mM App(NH)p caused a further increase in adenylylcyclase activ- ity, up to 7-fold, equivalent to that observed after 45 "C incubation in the absence of the nucleotide. The effects of pretreatment with App(NH)p at 30 "C for 20 min and at 45 "C were equivalent. Thus, in contrast to ATP, App(NH~p did not block the heat-induced activation of flagellar adenylylcy- clase and allowed maximal activation at 30 "C.

These results, along with the increase in activity observed when flagella were incubated with ATP plus staurosporine (Fig. 6), suggested that the activation of adenylylcyclase caused by preincubation at 45 "C could not be attributed only to inactivation of an inhibitor. For this reason, several inhib- itors of protein phosphatases were tested for their ability to block the activation that occurred at 45 "C. Microcystin (1 PM) and NaPPi (5 mM) blocked about 50% of the increase in activity, and EDTA blocked the increase by 70%. The bloc- kade by EDTA was partially prevented by inclusion of Mi2+ during the preincubation, whereas M P did not prevent the blockade of activation by NaPPi. Okadaic acid ( 2 PM), vanadate (50 pM), NaF (50 mM), phosphate buffer (10 mM), and &glycerol phosphate (10 mM) had no effect on activation.

ATP-dependent Inactivation of Adenylylcyclase Is Reversi- ble-When flagella were pretreated at 30 "C in the presence of 1 mM ATP but in the absence of a regenerating system, adenylylcyclase activity decreased at the same rate that was observed with ATP plus PEP and pyruvate kinase (Fig. 8). After about 20 min, however, activity began to recover. After 60 min, the adenylylcyclase activity reached nearly the same level of activity as in the original untreated sample. In eon- trast, recovery did not occur if ATP levels were maintained

Time (mini

FIG. 7. Effect of App(NH)p on the activation of flagellar adenylylcyclase. Flagella were incubated at 30 "C for 20 min (A) or at 45 "C for 12 min (0) in the buffer described in the legend to Fig. 3 in the presence of 0.5 mM App(NH)p and then assayed at 30 "C. The assays were initiated by the addition of Icx-"~P)ATP ( lo6 cpm/samplef. Untreated flagella were assayed in the same medium using either 0.5 mM A p p ( N ~ ) p (E!) or 0.5 mM ATP (0) plus a trace amount of [t.- ,'"P]ATP (10' cpmi as substrate. The protein content in each sample was 90 pg.

5 0 ~ 1 L I I I 1 I 0 10 20 30 40 50 60

Time of Pretreatment (rnin)

tivation with ATP at 30 "C. Ftagelia were incubated at 30 "C in FIG. 8. Reac~~vation of adenylylcycl~e activity after deae-

the buffer described in the legend to Fig. 2 in the presence of either 1 mM ATP (0) or 1 mM ATP, 10 pg/ml pyruvate kinase, and 5 mM PEP (0). At the indicated times, adenylylcyclase assays were initiated by the addition of an equal volume of buffer containing [a-"'P]ATP (10' cpm/sample) and the appropriate amount of PK and PEP to yield final concentrations of 10 pg/ml and 5 mM, respectively. Assays were carried out at 30 "C for 6 min. The protein contained in each sample was 90 pg.

by the ATP-regenerating system. Protein Kinase and Phosphatase Activities in Flagella-

Flagella were tested directly for protein phosphatase activity and protein kinase activity by assaying phosphorylation and desphosphorylation of flagellar proteins using f-y-"'P]ATP. Fig. 9A shows that, several proteins were phosphorylated after incubation of flagella with [y3*P]ATP for 10 min at 30 "C. Phosphorylation was reduced substantially by 2 I.IM stauro- sporine or by pretreatment at 45 "C before the incubation with [ y-32P]ATP. These results indicated that Chlamydomo- nas flagella contain heat-labile, staurosporine-sensitive pro- tein kinase activity.

To determine if there was protein phosphatase activity in these preparations, flagella were preincubated as above with [T-~*P~ATP a t 30 "C for 10 min, washed and resuspended in fresh buffer without [-y-:"P]ATP, and incubated for an addi- tional 30 min. Fig. 9B shows that this additional incubation at 30 "C in the absence of [yY2PPJATP led to a s ~ g n i ~ c a n t decrease in phosphory~ated pol~ept ides (right lane) com- pared with the washed sample that was kept on ice for 30 min (left lane), indicating that the flagella contained protein phos- phatase activity.

DISCUSSION

One of the earliest cellular responses to interactions be- tween flagellar agglutinins during the mating reaction in C h l a m y d o m o n ~ ~ is a rapid rise in the intracellular concentra- tion of cyclic AMP ( 2 , 3). One possibility for regulation of this adenylylcyclase is that, as in many other signaling sys- tems, G proteins play a prominent role. To date, however, there has been no evidence to implicate these ubiquitous signaling molecules in signal transduction during fertilization in Chlamydomonas. Chlamydomonas does contain G proteins (20) , some of which are enriched in the eyespot (21). Pasquale and Goodenough (2), however, reported that GTP, GTP+, AlF;, and forskolin failed to stimulate flagellar adenylylcy- clase, results that we have confirmed in preliminary experi- ments (not shown). Instead the resuIts presented here indicate that ~ h ~ a m y d o m o n a s flagellar adenylylcyclase exhibits a novel regulatory mechanism.

One model is that phosphorylation by a heat-labile protein

22958 Regulation of Adenylylcyclase in Chlamydomonas

A

110 -

84 - 47 -

33 -

24 -

16 -

FIG. 9. Phosphorylat ion and dephosphorylat ion of flagellar proteins. A, flagella (40 pg) were incubated in 90 pl of buffer containing 20 mM Na-HEPES, pH 7.2, 4% sucrose, 1 mM EDTA,

p~ sodium orthovanadate, and [y-:''P]ATP (specific activity, 3000 Ci/mM) for 10 min on ice ( Ice ) or a t 30 "C without (30 " C ) or with prior treatment a t 45 "C (PT, 30 "C) or in the presence of 2 pM staurosporine (30 "C + Staurosporine). At the end of the incubation samples were analyzed by SDS-PAGE and autoradiography. B, non- pretreated flagella were incubated with [y-'"P]ATP as above for 10 min at 30 "C without staurosporine, washed by centrifugation a t 40,000 rpm in a Beckman TL-100 ultracentrifuge (rotor TL-100.1), and then incubated for 30 min on ice or at 30 "C before being analyzed by SDS-PAGE and autoradiography. The migration of prestained molecular weight markers is indicated on the left.

0.0.56h RSA, 2.5 mM MgCI,, 0.05 mM GTP, 0.1 mM R020-1724, 50

kinase inhibits the flagellar adenylylcyclase and that inhibi- tion can be reversed by a thermostable activator, possibly a protein phosphatase. The data suggest this mechanism relates to the decline in activity that occurred when adenylylcyclase was assayed at 30 "C or lower and the increase in activity that occurred upon incubation a t higher temperatures. The decline was not detected in earlier studies (2, 22, 23) because the assays were carried out a t 37 "C, where the decline in activity is less striking than at lower temperatures. In their studies on adenylylcyclase in the flagella of Chlamydomonas eugametos, Kooijman et al. (23) also did not note any inhibition of activity during assays at 23 "C. Kooijman et al. (23) and Pasquale and Goodenough (2), however, used flagella that had been soni- cated prior to the assay, and preliminary experiments (not shown) suggest that sonication diminishes the regulatory effects described here.

Evidence That a Heat-labile Protein Kinase Inhibits Flagel- lar Adenylylcyclase-Several of our results are consistent with the idea that an endogenous protein kinase inhibits adenylylcyclase. First, activity was significantly inhibited by pretreatment at 30 "C in the presence of ATP but was not inhibited by pretreatment with the nonhydrolyzable analogue App(NH)p. Moreover, untreated samples assayed in App(NH)p at 30 "C did not lose activity during the assay,

whereas the sample assayed with ATP showed a progressive loss of activity during the 10-min assay. Finally, the relatively non-specific protein kinase inhibitor staurosporine (24, 25) prevented the decline in activity (Fig. 5) and blocked the ATP-dependent inhibition during pretreatment a t 30 "C (Fig. 6). Because this putative protein kinase is insensitive to more selective inhibitors, its identity is unknown.

Consistent with other reports (26-28), in vitro phosphoryl- ation experiments using [-y-:'2P]ATP indicated that flagella contain protein kinase($). Analysis by SDS-PAGE and auto- radiography showed that several proteins were phosphoryl- ated after incubation of flagella with [-y-:"P]ATP a t 30 "C for 10 min. This protein kinase activity was significantly inhib- ited by 2 PM staurosporine or by prior incubation of the flagella a t 45 "C. Thus, the major protein kinase activity in flagella resembles adenylylcyclase inhibitory activity with re- spect to staurosporine sensitivity and thermal lability.

Evidence for a Thermostable Activator of Adenylylcyclase- Although it was possible that the increased adenylylcyclase activity observed after pretreatment without ATP a t 45 "C (Fig. 2) reflected only inactivation of an inhibitory protein kinase, several of our results are consistent with the presence of an endogenous activator. First, in the absence of an ATP- regenerating system, treatment at 30 "C with ATP led to a 50% loss of activity after 20 min, but continued incubation for 40 min caused nearly full recovery (Fig. 8). Because other experiments (not shown) indicate the presence of endogenous ATPase in the preparations of flagella, it is likely that contin- ued incubation caused depletion of ATP and subsequent reactivation of the cyclase. Second, pretreating flagella for 20 min a t 30 "C or for 12 min a t 45 "C with App(NH)p caused a 3-fold increase in activity. Finally, preincubation at 30 "C with staurosporine did not merely block the loss of activity of adenylylcyclase but caused nearly a 2-fold increase in activity. Although other mechanisms cannot be ruled out, we feel it is likely that activation of the adenylylcyclase is a consequence of dephosphorylation by a protein phosphatase. The in vitro dephosphorylation experiments indicated that flagella con- tain protein phosphatase activity, consistent with recent re- sults from Bloodgood's laboratory (28).

Other workers have presented evidence that protein kinases can bring about modest changes in activity of adenylylcyclase in multicellular organisms (29-33). The adenylylcyclases in these systems, however, are also regulated by the G protein Gs. On the other hand, invertebrate and mammalian sperm adenylylcyclases are not regulated by G, (34,35). Bookbinder et al. (36) recently reported that a M , 190,000 adenylylcyclase in sea urchin sperm is highly enriched in the flagellum and may be phosphorylated. Our results suggest that Chlamydo- monas may use a novel mechanism for regulating adenylyl- cyclase in the flagellum. One possibility is that interaction between agglutinin molecules inhibits a kinase or activates a phosphatase, thereby stimulating adenylylcyclase.

REFERENCES 1. Adair, W. S. (1985) J. Cell Sci. Suppl. 2, 233-260 2. Pasquale, S. M., and Goodenough, U. W. (1987) J. Cell Biol. 105,

3. Pijst, H. L. A., van Driel, R., Janssens, P. M. W., Musgrave, A., and van den Ende, H. (1984) FEBS Lett. 174,132-136

4. Buchanan, M. J., Imam, S. H., Eskue, W. A., and Snell, W. J. (1989) J. Cell Biol. 108, 199-207

5. Snell, W. J., Eskue, W. A., and Buchanan, M. J. (1989) J. Cell Biol. 109, 1689-1694

6. Hunnicutt, G. R., Kosfiszer, M. G., and Snell, W. J. (1990) J. Cell Biol. 11 1, 1605-1616

7. Tomson, A.M., Demets, R., Musgrave, A., Kooijman, R., S t epee , D., and Van Den Ende, H. (1990) J . Cell Sci. 95,293-301

2279-2292

Regulation of Adenylylcyclase in Chlamydomonas 22959

8. 9.

10.

11. 12. 13.

14.

15.

16.

17.

18.

19.

20. 21.

22.

Goodenough, U. W. (1989) J. Cell Biol. 109,247-252 Detmers, P. A,, Goodenough, U. W., and Condeelis, J. (1983) J .

Goodenough, U. W., and Weiss, R. L. (1975) J . Cell Bwl. 67,

Sager, R., and Granick, S. (1954) J. Gen. Physiol. 37, 729-742 Snell, W. J. (1976) J. Cell Biol. 68, 48-69 Witman, G., Carlson, K., Berliner, J., and Rosenbaum, J. L.

Ross, E. M., Maguire, M. E., Sturgill, T. W., Biltonen, R. L., and

Salomon, Y., Londos, C., and Rodbell, M. (1974) Anal. Biochem.

Hidaka, H., Inagaki, M., Kawamoto, S., and Sasaki, Y. (1984)

Hidaka, H., Yamaki, T., Naka, M., Tanaka, T., Hayashi, H., and

Gazit, A., Yaish, P., Gilon, C., and Levitzke, A. (1989) J. Med.

Rodbell, M., Birnbaumer, L., Pohl, S. L., and Krans, H. M. J.

Schloss, J. (1990) Mol. Gen. Genet. 221, 443-452 Korolkov, S. N., Garnovskaya, M. N., Basov, A. S., Chunaev, A.

Pasquale, S. M., and Goodenough, U. W. (1988) Bot. Acta 101,

Cell Biol. 97, 522-532

623-637

(1972) J. Cell Biol. 54, 507-539

Gilman, A. G. (1977) J. Biol. Chem. 252,5761-5775

58,541-548

Biochemistry 23,5036-5041

Kobayashi, R. (1980) Mol. Pharmacol. 17,66-72

Chem. 32, 2344-2352

(1971) J. Bid. Chem. 246, 1877-1882

S., and Dumler, I. L. (1990) FEBS Lett. 270,132-134

24.

25.

26. 27.

28.

29. 30.

31.

32.

33.

34.

35.

A., and van den Ende, H. (1990) Planta 181.529-537 Tamaoki, T., Nomoto, H., Takahashi, I., Kato, Y., Morimoto, M.,

and Tomita, F. (1986) Biochem. Biophys. Res. Commun. 135,

Ruegg, U. T., and Burgess, G. M. (1989) Trends Pharmacol. Sci.

Segal, R. A., and Luck, D. J. (1985) J. Cell Biol. 101, 1702-1712 Hasegawa, E., Hayashi, H., Asakura, S., and Kamiya, R. (1987)

Bloodgood, R. A., and Salomonsky, N. L. (1989) J. Cell Biol. 109,

Wiener, E., and Scarpa, A. (1989) J. Bwl. Chem. 264,4324-4328 Anand-Srivastava. M. B.. and Srivastava, A. K. (1990) Mol. Cell.

397-402

10,218-220

Cell Motil. Cytoskeleton 8, 302-311

178a

Biochem. 92,91-98 '

. .

Otte. A. P.. van Run. P.. Heideveld. M.. van Driel. R.. and Durston, A. J. (1989) Ceb 58, 641-648 .

Beckner, S. K., and Farrar,.L. (1987) Biochem. Biophys. Res. Commun. 146,176-182

Yoshimasa, T., Sibley, D. R., Bouvier, M., Lef'kowitz, R. J., and Caron, M. G. (1987) Nature 327, 67-73

Hildebrandt, J. D., Codina, J., Tash, J. S., Kirchick, H. J., Lipschultz, L., Sekura, R. D., and Birnbaumer, L. (1985) En- docrinology 116, 1357-1366

Kopf, G. S. (1988) in Meiotic Inhibition: Molecular Control of Meiosis (Haseltine, F., and First, N., eds) pp. 357-386, Alan R. Liss. Inc.. New York

, ,

118-122 36. Bookdinder, L. H., Moy, G. W., and Vacquier, V. D. (1990) J. 23. Kooijman, R., de Wildt, P., van den Briel, W., Tan, S., Musgrave, Cell Biol. 11 1, 1859-1866