Brain Research, 592 (1992) 170-174 © 1992 Elsevier Science Publishers B.V. All rights reserved 0(O6-8993/92/$05.00
170
BRES 18141
Tyrosine kinase regulation of a molluscan circadian clock Michael H. Roberts, Julie A. Towles and Nancy K. Leader Department of Biology, Clarkson Unicersiry,Potsdam, NY 13699-5805 (USA) (Accepted 12 May 1992)
Key words: Circadian: Bulla gouldiana; Protein kinase; Protein phosphatase; Tyrosine kinase
On a h,rmal level the clocks regulating circadian and cell division cycles are related in that both have been modeled as limit cycle oscillations (S-:e.qce. 211 (1981) 1002-1013; Brain Res., 504 (1989) 211-215; Prec. Natl. Acad. Set. USA, 88 (1991) 7328-7332). Furthermore, in several o,t,aaisms each clock system is able to modulate the other (Science, 211 (1981) 1002-1013). However, in spite of the similarities at the formal level, and the connections al the physiological level, no common cellular elements have been identified linking the two processes. In the current series of experiments we show that one key element of cell cycle regulation, tyrosine phosphorylation/dephosphorylation is intimately associated with circadian rhythm generation in the eye of the marine snail, guile gouldiana. The importance of tyrosine kinase activity in the generation of circadian rhythms provides a possible point of similarity between the fundamental biochemical mechanisms underlying both circadian and cell cycle clocks.
INTRODUCTION
The eyes of the marine snail Belle gouldiana, like the eyes of several other opisthobranchs, contain circadian pacemakers 3, The ocular pacemakers generate a circadian rhythm in the frequency of spontaneously produced optic nerve impulses in viva and in vitro. Investigations carried out using B, gouldiana, or the related snail Aplysia californica, have shown that calcium fluxes~,tt, protein synthesis ~.1z2°, and protein phosphorylation '°'2', may play a role in the regulation or generation of circadian rhythms. However, no model for the clock mechanism in the molluscan eye has been explicitly proposed, in the current series of experiments we show, through the use of several inhibitors, that tyrosine phosphorylation/dephosphorylation is intimately associated with circadian rhythm generation in the eye of B. gouldiana. The inhibition of tyrosine phosphatase activity blocks rhythm expression while the inhibition of serine/threonine phosphatases has no effect, in addition, the tyrosine kinase inhibitor geniste.in, renders eyes arrhythmic when applied chronically, and produces phase-dependent phase shifts when applied for discrete 4 h intervals. The importance of tyrosine kinase activity in the generation of circadian rhythms raises the possibility that there is a similarity
between the fundamental biochemical mechanisms underlying both circadian and cell-cycle clocks. MATERIALS AND METHODS Animal maintenance and rhythm re(:ordlag Animals were obtained ft~)m Marinus Inc, (lain8 Beach, CA), maint,inod in seawater tanks at 15:1:0.1°C, and exposed to light cycles consisting of 12 h of light and 12 h of darkness (L:D; 12:12) with dawn (CT0) at either 10:01l or 22:(X) Eastern Standard Time, Eyes were removed from animals anesthetized by injection of 10 ml of isotonic magnesium chloride (76 8/I) and placed in petri dishes containing 15 ml of filtered artificial seawater (FSW-'lnstant Ocean'; specific 8rarity = 1,024-1,028, 30 mM HEPES, pH 7,75-7.78, 100,000 U / I penicillin, 10•,000 ~tg/t streptomycin). Optic nerves were pulled into glass suction electrodes and spontaneously occurring optic nerve impulses were continuously recorded on a Grass polygraph. Eyes were placed in light-tight boxes in a temperature regulated cold room (15,0±0,1°C), Dissections were timed ~ that eyes were placed in darkness at the time of dusk, Drug trt~ltm¢,ntx Drug treatments were given to eyes by opening boxes under dim red light and replacing 5 ml of the FSW with 5 ml of FSW into which the appropriate concentration of drug was dissolved. Pl'enylarsine oxide (Sigma, St, Louis, Me), 8enistein (Calbiochem, San Diego, CA) and okadaic acid (UBI, Lake Placid, NY) were prepared as stock solutions in DMSO, Control eyes received DMSO at a final concentration of 0,1%, in experiments that involved drug removal, boxes were opened under dim red light and seawater in both experimental and control dishes rinsed out with fresh FSW (dilution factor following rinses -- 0.0003),
Correspondence: M.H, Roberts, Department of Biology, Clarkson University, Potsdam, NY 13699-5805,USA.
171 Effects of genistein on phosphate incoqJoration Eyes were removed from eight animals near dusk (CTI2) and placed in one of two dishes containing 15 ml of FSW in constant darkness at constant temperature (15+0.1"(2). One eye from each animal was treated us experimental and the other as control. Fourteen hours later (CT2) experimental eyes were treated with ! 1.1 p.M genistein in DMSO (final DMSO concentration 0.1%) and 0.5 mCi [S2p]orthophosphate (NEN, Boston, MA), while control eyes were treated with DMSO (0,1% final concentration) and 0.5 mCi [32p]orthophosphate. After 4 h (CT 6), eyes were homogenized in Tris-saline (TBS: l0 mM "Iris, pH 8.6, 0.9% NaCI), 0.1% NP-40, 50 ram EDTA, 0.15 mg/ml benzamidine, and ~ mM sodium-o-vanadate. Each set of 8 eyes was homogenized in a total volume of 100 #1. SDS and DTT were then added to a final concentration of 1% and 10 raM, respectively, and the samples were placed in a boiling water bath for 4 rain. For the determination of total incorporation, 5 Fzl of each sample was precipitated onto filter paper with 10% TCA as 3 replicate sets, rinsed in acetone, and radioactivity measured using a LKB scintillation counter. For immunostaining, eye samples were separated by SDS-PAGE on 10% gels and transferred to a PVDF membrane (Millipore Corp., Bedford, MA). Non-specific binding was blocked with TBS containing 1% Tween-20, and the membrane incubated overnight at 4°C in monoclonal antiphosphotyrosine (UBI, Lake Placid, NY) diluted 1:500 in TBS/Tween. The membrane was rinsed in TBS/Tween and incubated in peroxidase-conjugated goat anti-mouse lgG (Calbiochem, LaJolla, CA) diluted 1:!00 in TBS/Tween for 30 rain. Following this incubation, the membrane was rinsed in TBS/Tween and immunoreactive bands visualized using 4-chloro-l-naphthoi as the chromogen. The membrane was dried and densitometrically scanned using a JAVA Image analysis system (Jandel Scientific, Corte Madera, CA). Relative phosphotyrosine immunoreactivity was expressed as pixel density. An additional replicate gel was stained with Coomassie blue to control for unequal protein in the original homogenates. Total protein was similar in both DMSO and genistein treated samples.
RESULTS Previous studies from our laboratory have shown that a general inhibition of protein kinase activity lengthens the circadian rhythm recorded fronl the isolated B. gouldiana eye in a dose-dependent manner t~. In an attempt to further characterize the requirement for protein phosphorylation/dephosphorylation in the regulation of circadiap rhythms, we surveyed the effects of several phosphatase inhibitors on the period of the ocular circadian rhythm. Okadaic acid (OA) is a potent inhibitor of protein phosphatases 1 and 2A, both serine/threonine phosphatases, that is effective in the 10-100 nM range 7. When chronically applied at 500 nM or 1 /zM to B. gouldiana eyes maintained in darkness at constant temperature (2 eyes for each dose) we found no pronounced changes in the period of the circadian rhythm relative to untreated control eyes (see Fig. la). In contrast, when eyes were chronically treated with phenylarsine oxide (PAO), a tyrosine phosphatase inhibitor effective in the 30-50 /zM range 2, the circadian rhythm was eliminated at cont:entrations as low as 1-5/zM (Fig. lb). The effect of PAO o.n the circadian rhythm was irreversible, although light responses were obtained following chronic treatment at 1/~M. In spite of the interpretive difficulties caused by
L,-
m
4o-
20" 10"
u . . . . . .
t~,
. . . . . . .
~
. . . . . . .
= "
•r i M s ( h r s )
f\
eq ......
A¢ . .......
O"
T" "'~ " "~ .......
TIME (hm) Fig. I. Effect of phosphatase inhibitors on the :ircadian rhythm recorded from the eye of Bulla gouldiana. (a) Effect of 500 nM okadaie acid on the circadian rhythm, Solid line, drug treated eye rhythm; broken line, DMSO (control) eye rhythm. The OA treated eye displays about a 0.25 h increase in period, (b) Effect of 5 /zM phenylarsine oxide, Solid line, drug U'eated; broken line, DMSO treated, Up arrow indicates time of drug introduction, Down arrow indicates time of drug removal, PAO eliminates tile rhythm.
the irreversibility of the treatment, these initial studies with OA and PAO raised the possibility that phosphorylation/dephosphorylation of tyrosine and not serine/threonine residues is a key component of the circadian rhythm-generating mechanism in the B. gouldiana eye. Based upon these preliminary results, we analyzed further the role of tyrosine phosphorylation in the regulation of circadian rhythms. For these experiments, we employed the reversible tyrosine kinase inhibitor, genistein, which has been shown to block the tyrosine autophosphorylation of the EGF receptor with an ID50 of less than 10/zg/ml (approx. 37/zM) I. When applied at 3/~g/ml (11.1 uM) genistein led to an elimination of impulse activity in B. gouldiana eyes. Upon drug removal, 44 h later, spiking resumed and eyes were found to be responsive to light, although the circadian rhythm was absent (n = 4; Fig. 2a). At lower doses (2.2 /~M, 1.1/zM) the rhythm was unaffected (n = 6; Fig. 2B). In order to confirm that tyrosine kinase activity was
172
a,o[
affected by genistein in the B. gouldiana eye, we assayed phosphate incorporation into total phosphoprorein in eyes treated with genistein by radiolabeling with [32p]orthophosphate. We found that overall 32p incorporation was unaffected by genistein (DMSO 14,959 _+ 574 CPM/eye; genistein 16,122+ 1,983 CPM/eye) which was not surprising since phosphotyrosine is a small contributor to total phosphoprotein. Specific incorporation into phosphotyrosine was assayed by separating eye samples by SDS-PAGE, transferring to PVDF membranes, and staining with monoclonal antiphosphotyrosine (UBI, Lake Placid, NY). In the genistein treated sample a large number of proteins were found to be immunoreactive. This would be expected if phosphotyrosine turnover rate was low. However, a number of immunoreactive protein bands were decreased in intensity relative to the DMSO treated control (Fig. 3), with the most prominent decrease in the 30-40 kDa range. Thus, we can conclude that while genistein does not block overall phosphate uptake, phosphorylation of several tyrosine phosphoproteins is reduced in the B. gouldiana eye by gcnistein. The ability of genistein to block both tyrosine phosphorylation and the circadian rhythm in electrical activity recorded from B. gouldiana eyes suggests that tyrosine kinase activity is either involved in rhythm expression or may affect directly or indirectly an important component of the rhythm generating mechanism. In order to distinguish between these possibilities, we attempted to shift the phase of the circadian rhythm with discrete 4-h drug treatments, The ability to alter phase is strong evidence that a process is involved in circadian rhythm regulation ~.
I I
:i
•
,
i'|
50 40 0
o. ,< 30
i
0 :
'.:
ii
i
10
A 24
48
72
96
120
144
TIME (hrs) b
40 120 ~00
'4
:i ~
40
i| 2O
o
0
J
-'
'~
24
48 72 TIME (hrs)
12o
96
Fig. 2, Ocular circadian rhythms from I!, go.hlian, eyes chronictdly treated with Ite.ist~Jin, Expcrintentul cyc rhythms are pk~ttcd us s{~lid lines while control eyes are plotted u,~ dashed lines, (a) Efl'~cts of I1.1 p.M genistcin, Drug treatment heg;m =,! time indicated (up arrow) :lnd was ri.sed or,! 44 h htler (down .rrow). h~set: light r~sl)onse from g~nist,'in treated eye ohlainud after tll~J six dlly r~cordinl~ perind. (h) Effects of I,I p M genist~ill, Drtkq tr~.tltt~nt hugtm at iudic.ted lime (up .rrow) and conti.a~Jd for the entire record,
D
G
a
b 14o.
I
too:
,o
,o
=
=
=,®
.o
®
®
,®.o
MOLECULARWEIGHT (kDa) Fig. 3, Elfeels of gunistein on anti-phosphotyrosine immunoreactivily in Bulla eye homogenates. (a) Phosphotyrosine immunoreactivity in ~canncd western blot, D, DMSO treated sample, G, Genistein treated sample. (b) Densitomctric scan of the Western blot shown in (a). Immunorcactivity (as pixel density) plotted as a function of molecular weight.
173 II0
0.89 h, n = 5; CT 1 0 - I 4 : - 2 . 1 7 + 1,63 h, n = 6, Fig. 3b; CT I4-18: - 3 . 0 0 + 1.73 h, n--4; CT 18-22: -2.88 + 2.02, n - - 4 (an additional 2 eyes became arrhythmic following genistein treatment at this phase); see Fig. 4c for summary of phase shift experiments). When treated with lower doses (3.7 v.M), no phase shifts were produced at any time tested (CT 22-2: 0.25 + 0.29 h, n = 4; CT 2-6:0.33 + 0.76 h, n = 3). The ability to produce phase-dependent phase shifts of the circadian rhythm with discrete 4-h genistein treatments suggests that protein phosphorylation on tyrosine residues is of critical importance in the generation of circadian rhythms in the B. gouldiana eye.
n
8
i
,o
j TIME (hrs)
~) 70, BO
!
DISCUSSION
50.
i w < ¢J 20
vk
10.
\/ TIME (hrs)
C
~,s
O.S 0
,05
,3S
4
!
4,$, ,S
I 0
4
e
la
~8
2O
~4
CIRCADIAN TIME (hrs) Fig. 4. Effects of genistein on the phase of the circadian rhythm recorded from the 13. gouldiana, eye, (a) Oenistein applied from CT22-2. Solid line, treated eye. Dashed line, control eye. Upwards arrowhead, drug in; downwards arrowhead, drug out. Genistein causes a 1.5 h phase advance. (b) Genistein applied from CTI0-14. Solid line, treated eye. Dashed line, control eye. Upwards arrowhead, drug in; downwards arrowhead, drug out. Genistein causes a 2.5 h phase delay, (c) Summary data from phase shift experiments. Horizontal lines indicate average shifts (with 95% confidence intervals - vertical lines)of eye rhythms treated with 11.1/xM genistein.
When applied from circadian time (CT) 22-2 (2 h before dawn of the previous light/dark cycle to 2 h after dawn) genistein (11.1 t~M) induced advance phase shifts (1.36 + 0.56 h; n = 7; Fig. 4a; all values given as mean + standard deviation). When applied at other times the rhythm was phase delayed (CT 2-6: - 3.60:1:
For each of our two observations there are two interpretations. First, the ability of chronic genistein treatment to block the expression of the rhythm, while not affecting impulse production or light responses indicates that inhibition of tyrosine kinase activity either uncouples the clock mechanism from an output pathway, or that some element of the clock mechanism is dependent upon tyrosine phosphorylation. Second, the ability of genistein to produce phase-dependent phase shifts indicates that the inhibition of tyrosine kinase activity affects an input pathway to the clock, or that tyrosine phosphorylation is a critical component of the clock mechanism itself. Given these alternatives, the most parsimonious interpretation of our data is that tyrosine kinase activity is a critical component of the clock mechaxlism in the eye of Bulla gouldiana. Recent studies concerning the regulation of the eukaryotic cell division cycle have some relevance in the further interpretation of our current series of experiments. The cellular and biochemical mechanisms involved in the regulation of the cell cycle have been extensively investigated and found to be surprisingly similar in a wide variety of organisms ta. The generation of the cell cycle has been recently reviewed ts and can be modeled by the interaction of two components t4'1~. The first component is the synthesis, accumulation, and degradation of the protein, cyclin, arm the second is the activation of kinase activity of the p34 'Jez subunit of 'maturation/mitosis promoting factor' (MPF). The kinase activity of the p34 cd~2 subunit is triggered by cyelin accumulation and the dephosphorylation of tyrosine residue 15, resulting in an active kinase and entry into mitosis 6. The active kinase also promotes cyclin degradation t3. At some point in the cycle tyrosine-15 is rephosphorylated through the action of an unspecified tyrosine kinase, although pp60 ~'~rc will phosphorylate p34 cdc2 purified from HeLa cells 4. This rephosphoryla-
174
tion, along with the degradation of cyclin, allows the cycle to continue. Although somewhat simplified, the important points for the current study are that cyclin synthesis and tyrosine phosphorylation/dephosphorylation are key components of the cell cycle clock. By analogy to the cell cycle system, our current results, indicating the critical importance of tyrosine kinase activity in the regulation of circadian rhythms in the B. gouldiana eye, along with previous studies indicating the importance of protein synthesis s'~7'2°, raises the possibility that interactions between tyrosine phosphoproteins and the synthesis of specific proteins may be involved in generating circadian rhythms in the molluscan eye. This possibility is supported by the observation that inhibition of protein synthesis affects the circadian clock just as the prevention of cyclin synthesis halts the cell cycle causing interphase arrest in frog egg extracts ~2. Particularly intriguing, and relevant to this discussion, is our initial observation that one of the phosphotyrosine bands (at about 30 kDa) whose phosphorylation is affected by genistein, is recognized by an antibody raised to the conserved 'PSTAIRE' region of the p34 cdc2 protein kinase. Current studies are aimed at investigating circadian changes in the phosphorylation state of this protein. in conclusion, it may be the case that circadian rhythms generated in post-mitotic cells (B. gouldiana ocular neurons) utilize the identical cellular machinery that was previously used to drive cell division cycles in early nervous system development. Further studies investigatin] the bpecific phases of tyrosine kinase activ. ity and protein synthesis as well as the identification of specific substrates of tyrosine kinase activity will shed light on this matter.
A~knowh, dgemcnts. We would like to thank L. Metola, T. Coskran, L. Kurpiewski and S, Kozlowski for technical assistance, V,M, Cassone, Department of Biology, Texas A&M University, for comments on the manuscript, C. Bishop, Department of Biology, Clarkson University, for the use of his facilities for the 3Zp labeling studies, and the National Institutes of Health (NS26272) for the support of this work.
REFERENCES I Akiyama, T., Ishida, J., Nakagawa, S., Ogawara, H., Watanabe, S-[., Itoh, N., Shibuya, M. and Fukami, Y., Genistein, A specific inhibitor of tyrosinc-specific protein kinases, J. Biol. Chem. 262 (1987) 5592-5595 2 Bernier, M., Laird, D.M. and Lane, M.D., Insulin-activated tyrosine phosphorylation of a 15-kilodalton protein in intact 3T3-L1 adipoc~es, Prec. Nat'. Aead. Sci. USA, 84 (1987) 1844-1848 3 Block, G.D. and Wallace, S.F., Localization of a circadian pacemaker in the eye of a mollusc, Bulla, Science, 217 (1982) 155-157 4 Draetta, G., Piwnica-Worms, H., Morrison, D., Druker, B., Roberts, T. and Beach, D., Human cdc2 protein kinase is a major cell-~vcle regulated tyrosine kinase substrate, Nature, 336 (1988) 738-744 5 Edmunds, L.N. and Adams, K.J., Clocked cell cycle clocks, Science, 211 (1981) 1002-1013 6 Gould, K.L. and Nurse P., Tyrosine phosphorylation of the fission yeast cdc2 + protein kinase regulates ent~ into mitosis, Nature, 342 (1989) 39-45 7 Haystead, T.A.J., Sire, A.T.R., Carling, D., Honnor, R.C., Tsukirant, Y., Cohen, P. and Hardie, D.G., Effects of the tumor promoter okadaic acid on intracellular protein phosphorylation and metabolism, Nature, 337 (1989) 78-81 8 Jacklet, J.W., Neuronal circadian rhythm: phase shifting by a protein synthesis'inhibitor, Science, 198 (1977) 69-71 9 Khalsa, S.B.S. and Block, G.D., Calcium in phase control of the Bulla circadian pacemaker, Brain Res., 506 (1990) 40-45 10 Lotshaw, D.P. and Jacklet, J.W., Serotonin induced protein phosphorylation in the Aplysia eye, Comp. Biochem. Physiol., 86C (1987) 27-32 11 McMahon, D.G. and Block, G.D., The Bulla ocular circadian pacemaker. I. Pacemaker neuron membrane potential controls phase through a calcium-dependent mechanism, J. Comp. Phys. ioL, 161 (1987) 335-346 12 Murray, A.W. and Kirschner, M.W., Cyclin synthesis drives the early embryonic cell cycle, Nature, 339 (1989) 275-280 13 Murray, A,W, and Kirschner M.W,, Dominoes and clocks: the union of two views of the cell cycle, Science, 246 (1989) 614-621 14 Norel, R, and Agur Z., A model for the adjustment of the mitotic clock by cyclin and MPF levels, Science, 251 (1991) 107(')-1078 15 Nurse, P., Universal control mechanism regulating onset of Mphase, Nature, 344 (1990) 503-508 16 Roberts, M.H., Bedian, V, and Chen, Y,, Kinase inhibition lengthens the period of the circadian pacemaker in the eye of
Bulls gouldiana, Brain Res.,504 (1989)211-215 17 Rothman, B.S, and Strumwasser, F., Phase shifting the circadian rhythm of neuronal activity in the isolated Aplysia eye with puromycin and cycloheximide, £ Gen. Physiol., 68 (1976) 359-384 18 Takahashi, J.S. and Zatz, M., Regulation of circadian rhythmicity, Science, 217 (1982) 1104-1111 19 Tyson, JJ,, Modeling the cell division cycle: cdc2 and cyclin interactions, Prec. Natl. Acad, Sci. USA, 88 (1991) 7328-7332 20 Yeung, SJ. and Eskin, A., Responses of the circadian system in the Aplysta eye to inhibitors of protein synthesis, 1. Biol. Rhythms, 3 (1988) 225-236 21 Zwartjes, R.E. and Eskin, A,, Changes in protein phospho~la. tion in the eye of Aplysia associated with circadian rhythm regulation by serotonin, J. Neurobiol., 21 (1990) 376-383
170
BRES 18141
Tyrosine kinase regulation of a molluscan circadian clock Michael H. Roberts, Julie A. Towles and Nancy K. Leader Department of Biology, Clarkson Unicersiry,Potsdam, NY 13699-5805 (USA) (Accepted 12 May 1992)
Key words: Circadian: Bulla gouldiana; Protein kinase; Protein phosphatase; Tyrosine kinase
On a h,rmal level the clocks regulating circadian and cell division cycles are related in that both have been modeled as limit cycle oscillations (S-:e.qce. 211 (1981) 1002-1013; Brain Res., 504 (1989) 211-215; Prec. Natl. Acad. Set. USA, 88 (1991) 7328-7332). Furthermore, in several o,t,aaisms each clock system is able to modulate the other (Science, 211 (1981) 1002-1013). However, in spite of the similarities at the formal level, and the connections al the physiological level, no common cellular elements have been identified linking the two processes. In the current series of experiments we show that one key element of cell cycle regulation, tyrosine phosphorylation/dephosphorylation is intimately associated with circadian rhythm generation in the eye of the marine snail, guile gouldiana. The importance of tyrosine kinase activity in the generation of circadian rhythms provides a possible point of similarity between the fundamental biochemical mechanisms underlying both circadian and cell cycle clocks.
INTRODUCTION
The eyes of the marine snail Belle gouldiana, like the eyes of several other opisthobranchs, contain circadian pacemakers 3, The ocular pacemakers generate a circadian rhythm in the frequency of spontaneously produced optic nerve impulses in viva and in vitro. Investigations carried out using B, gouldiana, or the related snail Aplysia californica, have shown that calcium fluxes~,tt, protein synthesis ~.1z2°, and protein phosphorylation '°'2', may play a role in the regulation or generation of circadian rhythms. However, no model for the clock mechanism in the molluscan eye has been explicitly proposed, in the current series of experiments we show, through the use of several inhibitors, that tyrosine phosphorylation/dephosphorylation is intimately associated with circadian rhythm generation in the eye of B. gouldiana. The inhibition of tyrosine phosphatase activity blocks rhythm expression while the inhibition of serine/threonine phosphatases has no effect, in addition, the tyrosine kinase inhibitor geniste.in, renders eyes arrhythmic when applied chronically, and produces phase-dependent phase shifts when applied for discrete 4 h intervals. The importance of tyrosine kinase activity in the generation of circadian rhythms raises the possibility that there is a similarity
between the fundamental biochemical mechanisms underlying both circadian and cell-cycle clocks. MATERIALS AND METHODS Animal maintenance and rhythm re(:ordlag Animals were obtained ft~)m Marinus Inc, (lain8 Beach, CA), maint,inod in seawater tanks at 15:1:0.1°C, and exposed to light cycles consisting of 12 h of light and 12 h of darkness (L:D; 12:12) with dawn (CT0) at either 10:01l or 22:(X) Eastern Standard Time, Eyes were removed from animals anesthetized by injection of 10 ml of isotonic magnesium chloride (76 8/I) and placed in petri dishes containing 15 ml of filtered artificial seawater (FSW-'lnstant Ocean'; specific 8rarity = 1,024-1,028, 30 mM HEPES, pH 7,75-7.78, 100,000 U / I penicillin, 10•,000 ~tg/t streptomycin). Optic nerves were pulled into glass suction electrodes and spontaneously occurring optic nerve impulses were continuously recorded on a Grass polygraph. Eyes were placed in light-tight boxes in a temperature regulated cold room (15,0±0,1°C), Dissections were timed ~ that eyes were placed in darkness at the time of dusk, Drug trt~ltm¢,ntx Drug treatments were given to eyes by opening boxes under dim red light and replacing 5 ml of the FSW with 5 ml of FSW into which the appropriate concentration of drug was dissolved. Pl'enylarsine oxide (Sigma, St, Louis, Me), 8enistein (Calbiochem, San Diego, CA) and okadaic acid (UBI, Lake Placid, NY) were prepared as stock solutions in DMSO, Control eyes received DMSO at a final concentration of 0,1%, in experiments that involved drug removal, boxes were opened under dim red light and seawater in both experimental and control dishes rinsed out with fresh FSW (dilution factor following rinses -- 0.0003),
Correspondence: M.H, Roberts, Department of Biology, Clarkson University, Potsdam, NY 13699-5805,USA.
171 Effects of genistein on phosphate incoqJoration Eyes were removed from eight animals near dusk (CTI2) and placed in one of two dishes containing 15 ml of FSW in constant darkness at constant temperature (15+0.1"(2). One eye from each animal was treated us experimental and the other as control. Fourteen hours later (CT2) experimental eyes were treated with ! 1.1 p.M genistein in DMSO (final DMSO concentration 0.1%) and 0.5 mCi [S2p]orthophosphate (NEN, Boston, MA), while control eyes were treated with DMSO (0,1% final concentration) and 0.5 mCi [32p]orthophosphate. After 4 h (CT 6), eyes were homogenized in Tris-saline (TBS: l0 mM "Iris, pH 8.6, 0.9% NaCI), 0.1% NP-40, 50 ram EDTA, 0.15 mg/ml benzamidine, and ~ mM sodium-o-vanadate. Each set of 8 eyes was homogenized in a total volume of 100 #1. SDS and DTT were then added to a final concentration of 1% and 10 raM, respectively, and the samples were placed in a boiling water bath for 4 rain. For the determination of total incorporation, 5 Fzl of each sample was precipitated onto filter paper with 10% TCA as 3 replicate sets, rinsed in acetone, and radioactivity measured using a LKB scintillation counter. For immunostaining, eye samples were separated by SDS-PAGE on 10% gels and transferred to a PVDF membrane (Millipore Corp., Bedford, MA). Non-specific binding was blocked with TBS containing 1% Tween-20, and the membrane incubated overnight at 4°C in monoclonal antiphosphotyrosine (UBI, Lake Placid, NY) diluted 1:500 in TBS/Tween. The membrane was rinsed in TBS/Tween and incubated in peroxidase-conjugated goat anti-mouse lgG (Calbiochem, LaJolla, CA) diluted 1:!00 in TBS/Tween for 30 rain. Following this incubation, the membrane was rinsed in TBS/Tween and immunoreactive bands visualized using 4-chloro-l-naphthoi as the chromogen. The membrane was dried and densitometrically scanned using a JAVA Image analysis system (Jandel Scientific, Corte Madera, CA). Relative phosphotyrosine immunoreactivity was expressed as pixel density. An additional replicate gel was stained with Coomassie blue to control for unequal protein in the original homogenates. Total protein was similar in both DMSO and genistein treated samples.
RESULTS Previous studies from our laboratory have shown that a general inhibition of protein kinase activity lengthens the circadian rhythm recorded fronl the isolated B. gouldiana eye in a dose-dependent manner t~. In an attempt to further characterize the requirement for protein phosphorylation/dephosphorylation in the regulation of circadiap rhythms, we surveyed the effects of several phosphatase inhibitors on the period of the ocular circadian rhythm. Okadaic acid (OA) is a potent inhibitor of protein phosphatases 1 and 2A, both serine/threonine phosphatases, that is effective in the 10-100 nM range 7. When chronically applied at 500 nM or 1 /zM to B. gouldiana eyes maintained in darkness at constant temperature (2 eyes for each dose) we found no pronounced changes in the period of the circadian rhythm relative to untreated control eyes (see Fig. la). In contrast, when eyes were chronically treated with phenylarsine oxide (PAO), a tyrosine phosphatase inhibitor effective in the 30-50 /zM range 2, the circadian rhythm was eliminated at cont:entrations as low as 1-5/zM (Fig. lb). The effect of PAO o.n the circadian rhythm was irreversible, although light responses were obtained following chronic treatment at 1/~M. In spite of the interpretive difficulties caused by
L,-
m
4o-
20" 10"
u . . . . . .
t~,
. . . . . . .
~
. . . . . . .
= "
•r i M s ( h r s )
f\
eq ......
A¢ . .......
O"
T" "'~ " "~ .......
TIME (hm) Fig. I. Effect of phosphatase inhibitors on the :ircadian rhythm recorded from the eye of Bulla gouldiana. (a) Effect of 500 nM okadaie acid on the circadian rhythm, Solid line, drug treated eye rhythm; broken line, DMSO (control) eye rhythm. The OA treated eye displays about a 0.25 h increase in period, (b) Effect of 5 /zM phenylarsine oxide, Solid line, drug U'eated; broken line, DMSO treated, Up arrow indicates time of drug introduction, Down arrow indicates time of drug removal, PAO eliminates tile rhythm.
the irreversibility of the treatment, these initial studies with OA and PAO raised the possibility that phosphorylation/dephosphorylation of tyrosine and not serine/threonine residues is a key component of the circadian rhythm-generating mechanism in the B. gouldiana eye. Based upon these preliminary results, we analyzed further the role of tyrosine phosphorylation in the regulation of circadian rhythms. For these experiments, we employed the reversible tyrosine kinase inhibitor, genistein, which has been shown to block the tyrosine autophosphorylation of the EGF receptor with an ID50 of less than 10/zg/ml (approx. 37/zM) I. When applied at 3/~g/ml (11.1 uM) genistein led to an elimination of impulse activity in B. gouldiana eyes. Upon drug removal, 44 h later, spiking resumed and eyes were found to be responsive to light, although the circadian rhythm was absent (n = 4; Fig. 2a). At lower doses (2.2 /~M, 1.1/zM) the rhythm was unaffected (n = 6; Fig. 2B). In order to confirm that tyrosine kinase activity was
172
a,o[
affected by genistein in the B. gouldiana eye, we assayed phosphate incorporation into total phosphoprorein in eyes treated with genistein by radiolabeling with [32p]orthophosphate. We found that overall 32p incorporation was unaffected by genistein (DMSO 14,959 _+ 574 CPM/eye; genistein 16,122+ 1,983 CPM/eye) which was not surprising since phosphotyrosine is a small contributor to total phosphoprotein. Specific incorporation into phosphotyrosine was assayed by separating eye samples by SDS-PAGE, transferring to PVDF membranes, and staining with monoclonal antiphosphotyrosine (UBI, Lake Placid, NY). In the genistein treated sample a large number of proteins were found to be immunoreactive. This would be expected if phosphotyrosine turnover rate was low. However, a number of immunoreactive protein bands were decreased in intensity relative to the DMSO treated control (Fig. 3), with the most prominent decrease in the 30-40 kDa range. Thus, we can conclude that while genistein does not block overall phosphate uptake, phosphorylation of several tyrosine phosphoproteins is reduced in the B. gouldiana eye by gcnistein. The ability of genistein to block both tyrosine phosphorylation and the circadian rhythm in electrical activity recorded from B. gouldiana eyes suggests that tyrosine kinase activity is either involved in rhythm expression or may affect directly or indirectly an important component of the rhythm generating mechanism. In order to distinguish between these possibilities, we attempted to shift the phase of the circadian rhythm with discrete 4-h drug treatments, The ability to alter phase is strong evidence that a process is involved in circadian rhythm regulation ~.
I I
:i
•
,
i'|
50 40 0
o. ,< 30
i
0 :
'.:
ii
i
10
A 24
48
72
96
120
144
TIME (hrs) b
40 120 ~00
'4
:i ~
40
i| 2O
o
0
J
-'
'~
24
48 72 TIME (hrs)
12o
96
Fig. 2, Ocular circadian rhythms from I!, go.hlian, eyes chronictdly treated with Ite.ist~Jin, Expcrintentul cyc rhythms are pk~ttcd us s{~lid lines while control eyes are plotted u,~ dashed lines, (a) Efl'~cts of I1.1 p.M genistcin, Drug treatment heg;m =,! time indicated (up arrow) :lnd was ri.sed or,! 44 h htler (down .rrow). h~set: light r~sl)onse from g~nist,'in treated eye ohlainud after tll~J six dlly r~cordinl~ perind. (h) Effects of I,I p M genist~ill, Drtkq tr~.tltt~nt hugtm at iudic.ted lime (up .rrow) and conti.a~Jd for the entire record,
D
G
a
b 14o.
I
too:
,o
,o
=
=
=,®
.o
®
®
,®.o
MOLECULARWEIGHT (kDa) Fig. 3, Elfeels of gunistein on anti-phosphotyrosine immunoreactivily in Bulla eye homogenates. (a) Phosphotyrosine immunoreactivity in ~canncd western blot, D, DMSO treated sample, G, Genistein treated sample. (b) Densitomctric scan of the Western blot shown in (a). Immunorcactivity (as pixel density) plotted as a function of molecular weight.
173 II0
0.89 h, n = 5; CT 1 0 - I 4 : - 2 . 1 7 + 1,63 h, n = 6, Fig. 3b; CT I4-18: - 3 . 0 0 + 1.73 h, n--4; CT 18-22: -2.88 + 2.02, n - - 4 (an additional 2 eyes became arrhythmic following genistein treatment at this phase); see Fig. 4c for summary of phase shift experiments). When treated with lower doses (3.7 v.M), no phase shifts were produced at any time tested (CT 22-2: 0.25 + 0.29 h, n = 4; CT 2-6:0.33 + 0.76 h, n = 3). The ability to produce phase-dependent phase shifts of the circadian rhythm with discrete 4-h genistein treatments suggests that protein phosphorylation on tyrosine residues is of critical importance in the generation of circadian rhythms in the B. gouldiana eye.
n
8
i
,o
j TIME (hrs)
~) 70, BO
!
DISCUSSION
50.
i w < ¢J 20
vk
10.
\/ TIME (hrs)
C
~,s
O.S 0
,05
,3S
4
!
4,$, ,S
I 0
4
e
la
~8
2O
~4
CIRCADIAN TIME (hrs) Fig. 4. Effects of genistein on the phase of the circadian rhythm recorded from the 13. gouldiana, eye, (a) Oenistein applied from CT22-2. Solid line, treated eye. Dashed line, control eye. Upwards arrowhead, drug in; downwards arrowhead, drug out. Genistein causes a 1.5 h phase advance. (b) Genistein applied from CTI0-14. Solid line, treated eye. Dashed line, control eye. Upwards arrowhead, drug in; downwards arrowhead, drug out. Genistein causes a 2.5 h phase delay, (c) Summary data from phase shift experiments. Horizontal lines indicate average shifts (with 95% confidence intervals - vertical lines)of eye rhythms treated with 11.1/xM genistein.
When applied from circadian time (CT) 22-2 (2 h before dawn of the previous light/dark cycle to 2 h after dawn) genistein (11.1 t~M) induced advance phase shifts (1.36 + 0.56 h; n = 7; Fig. 4a; all values given as mean + standard deviation). When applied at other times the rhythm was phase delayed (CT 2-6: - 3.60:1:
For each of our two observations there are two interpretations. First, the ability of chronic genistein treatment to block the expression of the rhythm, while not affecting impulse production or light responses indicates that inhibition of tyrosine kinase activity either uncouples the clock mechanism from an output pathway, or that some element of the clock mechanism is dependent upon tyrosine phosphorylation. Second, the ability of genistein to produce phase-dependent phase shifts indicates that the inhibition of tyrosine kinase activity affects an input pathway to the clock, or that tyrosine phosphorylation is a critical component of the clock mechanism itself. Given these alternatives, the most parsimonious interpretation of our data is that tyrosine kinase activity is a critical component of the clock mechaxlism in the eye of Bulla gouldiana. Recent studies concerning the regulation of the eukaryotic cell division cycle have some relevance in the further interpretation of our current series of experiments. The cellular and biochemical mechanisms involved in the regulation of the cell cycle have been extensively investigated and found to be surprisingly similar in a wide variety of organisms ta. The generation of the cell cycle has been recently reviewed ts and can be modeled by the interaction of two components t4'1~. The first component is the synthesis, accumulation, and degradation of the protein, cyclin, arm the second is the activation of kinase activity of the p34 'Jez subunit of 'maturation/mitosis promoting factor' (MPF). The kinase activity of the p34 cd~2 subunit is triggered by cyelin accumulation and the dephosphorylation of tyrosine residue 15, resulting in an active kinase and entry into mitosis 6. The active kinase also promotes cyclin degradation t3. At some point in the cycle tyrosine-15 is rephosphorylated through the action of an unspecified tyrosine kinase, although pp60 ~'~rc will phosphorylate p34 cdc2 purified from HeLa cells 4. This rephosphoryla-
174
tion, along with the degradation of cyclin, allows the cycle to continue. Although somewhat simplified, the important points for the current study are that cyclin synthesis and tyrosine phosphorylation/dephosphorylation are key components of the cell cycle clock. By analogy to the cell cycle system, our current results, indicating the critical importance of tyrosine kinase activity in the regulation of circadian rhythms in the B. gouldiana eye, along with previous studies indicating the importance of protein synthesis s'~7'2°, raises the possibility that interactions between tyrosine phosphoproteins and the synthesis of specific proteins may be involved in generating circadian rhythms in the molluscan eye. This possibility is supported by the observation that inhibition of protein synthesis affects the circadian clock just as the prevention of cyclin synthesis halts the cell cycle causing interphase arrest in frog egg extracts ~2. Particularly intriguing, and relevant to this discussion, is our initial observation that one of the phosphotyrosine bands (at about 30 kDa) whose phosphorylation is affected by genistein, is recognized by an antibody raised to the conserved 'PSTAIRE' region of the p34 cdc2 protein kinase. Current studies are aimed at investigating circadian changes in the phosphorylation state of this protein. in conclusion, it may be the case that circadian rhythms generated in post-mitotic cells (B. gouldiana ocular neurons) utilize the identical cellular machinery that was previously used to drive cell division cycles in early nervous system development. Further studies investigatin] the bpecific phases of tyrosine kinase activ. ity and protein synthesis as well as the identification of specific substrates of tyrosine kinase activity will shed light on this matter.
A~knowh, dgemcnts. We would like to thank L. Metola, T. Coskran, L. Kurpiewski and S, Kozlowski for technical assistance, V,M, Cassone, Department of Biology, Texas A&M University, for comments on the manuscript, C. Bishop, Department of Biology, Clarkson University, for the use of his facilities for the 3Zp labeling studies, and the National Institutes of Health (NS26272) for the support of this work.
REFERENCES I Akiyama, T., Ishida, J., Nakagawa, S., Ogawara, H., Watanabe, S-[., Itoh, N., Shibuya, M. and Fukami, Y., Genistein, A specific inhibitor of tyrosinc-specific protein kinases, J. Biol. Chem. 262 (1987) 5592-5595 2 Bernier, M., Laird, D.M. and Lane, M.D., Insulin-activated tyrosine phosphorylation of a 15-kilodalton protein in intact 3T3-L1 adipoc~es, Prec. Nat'. Aead. Sci. USA, 84 (1987) 1844-1848 3 Block, G.D. and Wallace, S.F., Localization of a circadian pacemaker in the eye of a mollusc, Bulla, Science, 217 (1982) 155-157 4 Draetta, G., Piwnica-Worms, H., Morrison, D., Druker, B., Roberts, T. and Beach, D., Human cdc2 protein kinase is a major cell-~vcle regulated tyrosine kinase substrate, Nature, 336 (1988) 738-744 5 Edmunds, L.N. and Adams, K.J., Clocked cell cycle clocks, Science, 211 (1981) 1002-1013 6 Gould, K.L. and Nurse P., Tyrosine phosphorylation of the fission yeast cdc2 + protein kinase regulates ent~ into mitosis, Nature, 342 (1989) 39-45 7 Haystead, T.A.J., Sire, A.T.R., Carling, D., Honnor, R.C., Tsukirant, Y., Cohen, P. and Hardie, D.G., Effects of the tumor promoter okadaic acid on intracellular protein phosphorylation and metabolism, Nature, 337 (1989) 78-81 8 Jacklet, J.W., Neuronal circadian rhythm: phase shifting by a protein synthesis'inhibitor, Science, 198 (1977) 69-71 9 Khalsa, S.B.S. and Block, G.D., Calcium in phase control of the Bulla circadian pacemaker, Brain Res., 506 (1990) 40-45 10 Lotshaw, D.P. and Jacklet, J.W., Serotonin induced protein phosphorylation in the Aplysia eye, Comp. Biochem. Physiol., 86C (1987) 27-32 11 McMahon, D.G. and Block, G.D., The Bulla ocular circadian pacemaker. I. Pacemaker neuron membrane potential controls phase through a calcium-dependent mechanism, J. Comp. Phys. ioL, 161 (1987) 335-346 12 Murray, A.W. and Kirschner, M.W., Cyclin synthesis drives the early embryonic cell cycle, Nature, 339 (1989) 275-280 13 Murray, A,W, and Kirschner M.W,, Dominoes and clocks: the union of two views of the cell cycle, Science, 246 (1989) 614-621 14 Norel, R, and Agur Z., A model for the adjustment of the mitotic clock by cyclin and MPF levels, Science, 251 (1991) 107(')-1078 15 Nurse, P., Universal control mechanism regulating onset of Mphase, Nature, 344 (1990) 503-508 16 Roberts, M.H., Bedian, V, and Chen, Y,, Kinase inhibition lengthens the period of the circadian pacemaker in the eye of
Bulls gouldiana, Brain Res.,504 (1989)211-215 17 Rothman, B.S, and Strumwasser, F., Phase shifting the circadian rhythm of neuronal activity in the isolated Aplysia eye with puromycin and cycloheximide, £ Gen. Physiol., 68 (1976) 359-384 18 Takahashi, J.S. and Zatz, M., Regulation of circadian rhythmicity, Science, 217 (1982) 1104-1111 19 Tyson, JJ,, Modeling the cell division cycle: cdc2 and cyclin interactions, Prec. Natl. Acad, Sci. USA, 88 (1991) 7328-7332 20 Yeung, SJ. and Eskin, A., Responses of the circadian system in the Aplysta eye to inhibitors of protein synthesis, 1. Biol. Rhythms, 3 (1988) 225-236 21 Zwartjes, R.E. and Eskin, A,, Changes in protein phospho~la. tion in the eye of Aplysia associated with circadian rhythm regulation by serotonin, J. Neurobiol., 21 (1990) 376-383
