83

J Physiology (1992) 86, 83-88 © Elsevier, Paris

Calmodulin and cell cycle control C D R a s m u s s e n a, K P L u b, R L M e a n s b, A R M e a n s b aDepartment of Anatomy and Cell Biology and the NCI Group in Molecular Mechanisms of Growth Control, University of Alberta, Edmonton, AB, Canada T6G 2H7; bDepartment of Pharmacology, Duke University Medical Center, Durham, NC, USA

Summary - Previous studies have indicated a r01e for the calcium receptor calmodulin in the control of eukaryotic cell proliferation. Using a molecular genetic approach in the filamentous fungus Aspergitlus niduIans we have shown that CaM is required for cell cycle progression at multiple points in the cell cycle. Construction of an A nidutans strain conditional for calmodulin expression reveals-that this protein is required during G~/S and for the initiation of mitosis, A lack of calmodulin results in cell cycle arrest, and a failure in polar growth that accompanies germination of A nidulans spores. In addition, increased expression of calmodulin in this organism permits growth at suboptimal calcium concentrations, indicating that cell growth is coordinately regulated by calcium and calmodulin, Together these results indicate that calmodulin-dependent processes may be conserved between A nidulans and vertebrate cells, and suggest that this approach may allow us to elucidate the molecular mechanism underlying calmodulin-regulated control of cell proliferation. calmodulin t cell cycle central l second messenger

Introduction Second messenger signalling pathways are important in the control of many cellular functions. As a member of this class of cellular regulators, Ca 2+ is involved in processes ranging from the control of cyclic nucleotide levels to cell growth (Means, 1988). All CaZ+-dependent functions are mediated by proteins that bind this ion, called CaZ+-binding proteins. These CaZ+/CaZ+-binding protein complexes interact with other cellular targets thus propagating the calcium signal. One of these proteins is the calcium receptor, calmodulin (CAM). In all cells except striated muscle, CaM is the major CaZ+-binding protein and a primary mediator of a variety of Ca 2+ regulated processes. In vitro studies have suggested that more than 20 known enzymes are regulated by CaM in a Ca z+dependent manner (Means, 1988). Over the past several years we have been investigating the control of cell growth by CaM. Several previous studies had suggested that Ca 2+ and CaM were involved in the regulation of the

cell cycle in eukaryotes. Alteration of Ca 2+ levels had been shown to alter progression through mitosis, and both Ca 2+ and CaM had been demonstrated to be localized to the mitotic spindle during cell division (Welsh et al, 1978; Wolniak et al, 1980). Calcium was also known to be required for cell cycle progression prior to the onset of DNA replication (Hazelton et al, 1979). The intracellular levels of both Ca 2+ and CaM change in cell-cycle-dependent manner (Christenson and Means, unpublished observations), and drugs that selectively inhibit CaM function were found to cause reversible cell cycle arrest (Sasaki and Hidaka, 1981; Chafouleas et al, 1982, 1984). Recently, using eukaryotic expression vectors to specifically manipulate CaM levels, we have found that the rate of cell proliferation in mouse C127 cells is directly related to intraceltular CaM concentration. Increased CaM accelerates cell proliferation, while reduced CaM results in a blockage of cell cycle progression (Rasmussen and Means, 1987, 1989). These studies also demonstrated that CaM is specifically required

84 during the cell cycle for the initiation of DNA synthesis, the initiation of mitosis, and the onset of chromosome movement at the metaphaseanaphase transition of mitosis. While the role of CaM in growth control undoubtedly requires interaction with specific target proteins, a precise mechanism has yet to be elucidated. In order to investigate these important pathways, we and others have begun to combine molecular biology and genetic approaches. Calmodulin genes have been cloned from three different organisms in which genetic analysis may be used to study cell cycle control, Saccharomyces cerevisiae, Schizosaccharomyces pombe, and Aspergillus niduIans (Davis et al, 1986; Takeda and Yamamoto, 1987; Rasmussen et al, 1990). In each case it has been possible to demonstrate that CaM is essential for cell growth and division. In this paper we report some of our recent efforts to more clearly define specific events in the cell cycle for which CaM is required. This approach was made possible by the construction and utilization of a strain of the filamentous fungus A nidutans in which CaM expression has been placed under the control of a regulatable promoter. This strain allows us to examine the phenotypic consequences of alterations in CaM levels on cell function in a genetically manipulable eukaryote.

Materials and methods Growth and manipulation of AspergilIus were as previously described (Rasmussen et al, 1990). Aspergillus CaM mRNA was analysed by Northern blot and CaM levels determined by radioimmunoassay using established procedures (Rasmussen et al, 1990).

Results The Aspergillus nidulans calmodulin gene We previously cloned and characterized the CaM gene from Aspergillus nidulans (Rasmussen et al, 1990). The gene contains five introns, three of which are in conserved positions as compared to CaM genes from other species. The protein encoded by this gene is more than 80% identical to vertebrate CaM (93% when conservative amino acid differences are considered), making it more similar to its vertebrate counterpart than are the CaMs from either S cerevisiae or S pombe. By

constructing a bacterial expression plasmid containing the complete A nidulans CaM cDNA we were able to express and purify large quantities of Aspergillus CaM for functional studies. Enzyme activation experiments showed that Aspergillus CaM will activate vertebrate CaM-dependent enzymes with kinetics similar to CaM from vertebrate sources (Rasmussen et al, 1990). The levels of CaM mRNA and protein were found to vary in a ceil-cycle dependent manner in Aspergillus, the highest level of CaM associated with progression through M-phase. Finally, disruption of the CaM gene in Aspergillus by site-specific homologous recombination was lethal, revealing that CaM is essential for the growth of Aspergillus. Studies using similar approaches in the yeasts S cerevisiae and S pombe have also demonstrated CaM to be essential (Davis et al, 1986; Takeda and Yamamoto, 1987).

Construction of a strain conditional f o r calmodulin expression Since our previous studies had shown that CaM acts at specific points in the cell cycle, and a lack of CaM is lethal, we wished to generate a strain of A nidulans that was conditional for CaM expression. This was accomplished by constructing a plasmid in which a portion of the CaM gene was placed 3" to the alcA gene promoter of A nidulans (fig 1). Expression from this promoter is regulated by the carbon source present in the growth medium (Waring et al, 1989). Glucose or acetate repress expression, glycerol is non-repressing but also non-inducing (ie allows a low basal rate of transcription), while ethanol or threonine induce high levels of expression. Thus it is possible to express a gene at three different levels when regulated by the alcA promoter. In Aspergillus as in yeasts, transformation occurs preferentially by homologous recombination (May, 1989). Therefore, integration of the pALCaM(RP) plasmid at the CaM gene locus in Aspergillus results in the simultaneous disruption of the endogenous CaM gene, and its replacement with a chimeric CaM gene under control of the alcA promoter (fig 1). The GR5 strain was transformed with the pAL-CaM(RP) plasmid, and transformants able to grow in the absence of uridine + uracil were selected. From these transformants, spores were isolated and tested for growth in media in which the alcA promoter was expected to function. An isolate (named alcCaM)

85 Construction of an Aspergillus nidulans strain Conditional for Calmodulin Expression.

g r e a t e r than 100-fold but protein levels i n c r e a s e d a b o u t 3 - 4 - f o l d (fig 3). We h a v e p r e v i o u s l y obs e r v e d this s a m e p h e n o m e n o n in m o u s e C127 cells c o n s t i t u t i v e l y o v e r e x p r e s s i n g high l e v e l s o f C a M m R N A , w h e r e a 100-fold i n c r e a s e in m R N A o n l y resulted in a 4 - 5 i n c r e a s e in the level o f CaM. These results s u g g e s t that C a M m a y negat i v e l y r e g u l a t e its o w n s y n t h e s i s at the transl a t i o n a l level. F u r t h e r studies will be n e e d e d to determine if translational control of CaM does i n d e e d occur. In o t h e r studies we o b s e r v e d that A s p e r g i l l u s n i d u l a n s r e q u i r e s a d d i t i o n a l C a 2÷ for o p t i m a l

~r~ pyr4 selectable marker m

CalmodulinGene (3' a)

I ~ IdeAPromoter

HomologousRecombination EndogenousCaM Gene

CaM Gene (3' a) Non-functional

pyr4

alcA/CaM Hybrid Conditional Expression

i

Calmodulin Expression Carbon S o u r c e

Exvression Level

Acetate

Repressed

Glycerol

Basal

Threonine

Induced

Growth

+ 4-++

Fig 1. Diagram of strategy for creation of a conditional strain. The plasmid pAL-AspCaM(RP) was constructed by placing a fragment of the A nidulans CaM gene deleted at the 3' end behind the alcA geue promoter. This plasmid was used to transform A nidulans, and transformants assayed by Southern blot. Those which gave a result consistent with that expected for a single integration of the plasmid at the CaM gene locus were kept for analysis.

was o b t a i n e d w h i c h g r e w on p e r m i s s i v e m e d i u m (either g l y c e r o l or g l y c e r o l + threonine) but f a i l e d to g r o w in m e d i u m c o n t a i n i n g g l u c o s e or acetate. Southern a n a l y s i s c o n f i r m e d that the p A L C a M ( K O ) p l a s m i d was i n t e g r a t e d at the C a M gene locus. E x a m i n a t i o n o f C a M m R N A levels b y N o r t h e r n blot a n a l y s i s s h o w e d that the a l c A - C a M h y b r i d was a p p r o p r i a t e l y r e g u l a t e d in r e s p o n s e to c h a n g e s in the c a r b o n source p r e s e n t in the m e dium. In acetate ( a l c A - r e p r e s s i n g ) no C a M m R N A was o b s e r v e d , in g l y c e r o l ( p e r m i s s i v e but non-ind u c i n g ) l o w levels o f C a M m R N A were present, and high level e x p r e s s i o n was a t t a i n e d during g r o w t h in threonine (inducing) (fig 2). R a d i o i m m u n o a s s a y s h o w e d that C a M p r o t e i n l e v e l s also c h a n g e d c o o r d i n a t e l y with the c h a n g e s in C a M m R N A although full i n d u c t i o n raised C a M m R N A

Fig 2. mRNA levels in alcCaM. Northern blot analysis was used to examine CaM mRNA levels in the alcCaM strain after overnight growth in medium containg either acetate, glycerol, or glycerol + threonine. Equal amounts (5 I.tg/lane) were loaded per lane, and the blot hybridized to an NcolIEcoRI fragment of the A nidulans CaM cDNA (Rasmussen et al, 1990).

CaM Levels in alcCaM Germlings Grown on Different Carbon Sources

~,~

3 • []

Acetate Glycerol

[]

Threonine

0 Carbon Source

Fig 3. CaM levels in the alcCaM strain, alcCaM spores were grown overnight in various media as described in figure 2. CaM levels were determined by radioimmunassay using an antibody raised against purified A nidulans CaM (Lu et al, in press).

86 growth (Lu et al, in press). Since we were able to vary CaM levels in the alcCaM strain, we wished to determine if the overexpression of CaM would result in a change in the optimal external Ca 2+ concentration. Normal or alcCaM strains were grown either in glycerol (permissive but non-inducing medium) or in threonine + glycerol (inducing medium) and the concentration of Ca 2+ required for half-maximal growth rate determined at varying external Ca 2+ concentrations. The different media had negligible effect on growth of the normal strain, but when CaM was overexpressed in the alcCaM strain we found that halfmaximal growth occurred at Ca 2+ levels 10-fold lower than in non-inducing medium. These data clearly demonstrate a coordinate regulation of growth in Aspergillus by both external Ca 2÷ and intracellular CaM. Because we could limit CaM levels by culturing the alcCaM strain in repressing medium, we next wanted to determine the phenotypic consequence that a lack of CaM might have on progression through the cell cycle. Spores were germinated in medium containing either acetate (non-permissive) or glycerol (permissive) as the carbon source, and the extent of growth analysed by staining nuclei with DAPI and counting nuclear number as an indicator of progression past successive mitotic divisions. We found that when CaM expression was prevented, most spores complete only a single mitosis and arrest with two interphase nuclei (fig 4). In contrast, parallel glycerol-fed cultures in which CaM was expressed were able to complete several mitotic divisions. Interestingly, the lack of CaM was not lethal, and germlings could recover and resume progression through the cell cycle if CaM expression was induced by washing growth-arrested germlings out of the repressive medium, and placing into inducing medium. This was evidenced by the emergence of germtubes concomitant with the increase in CaM levels (fig 5), as well as the resumption of mitotic divisions (not shown). Because this effect was reversible, it allowed us to determine where blockage of the cell cycle occurs in the absence of CaM expression. By using a reciprocal shift experiment (Jarvik and Botstein, 1975) in which growth-arrested germlings were shifted into inducing medium containing the DNA synthesis inhibitor hydroxyurea, we found that 20% of germlings were arrested in G1/S, while the remaining 80% were arrested in G2 prior to the onset of mitosis.

Cell Cycle Progression in Normal & Growth Arrested alcCaM Germlings 100-

80" 0

°- [

•

Acetate

[]

Glycerol

1

40"

o-i|

I

O.

0

1

3

2

4

No. of Mitoses

Fig 4. Nuclear number in normal and growth-arrested alcCaM germlings. Conidia were incubated for 12 h in permissive (glycerol) or non-permissive (acetate) medium. After fixation and staining with the nuclear binding fluorochrome DAPI, the number of nuclei per germling were determined. In normal cultures grown in glycerol germlings have predominantly 9 - 1 6 nuclei indicating that they are in the 4th cell cycle following germination. In contrast, conidia culture in acetate medium have predominantly two nuclei per germling (actual average = 1.8) suggesting that celt cycle arrest has occurred after a single mitotic division.

Recovery of alcCaM After Arrest

100

O----

Germ Tubes

~,

CaM Levels

2000

,,O

,d O

[..,

100o

t-

0 0

1

2

3

4

5

6

T i m e After Induction (hrs) Fig 5. Germ tube emergence in previously arrested alcCaM. alcCaM were grown for 12 h in acetate medium to induce arrest. The medium was then replaced with inducing medium (glycerol + threonine) and cultures grown further. The percentage of germlings with germ tubes was determined, and compared to CaM levels following the change in medium. The results show that growth arrest is reversible, and follows the experimentally-induced increase in CaM levels.

87 1st Cycle G0/G1

S

rill 2nd Cycle - - ~ G2 GI S G2

~///////,'~

GI/S t

~////////A

G2/M

CellCycleArrest after CaM repression

Fig 6. Points of CaM-dependent cell cycle control in A nidulans. The results of our studies have suggested that CaM is required for progression through the cell cycle both in G1/S and the G2/M boundary of the cell cycle. In the absence of CaM expression, A niduIans are only able to carry out a single mitotic division. The ability to progress through one cell cycle is probably due to the dowry of CaM and CaM mRNAin the conidium which becomes diluted by initial growth in the absence of new expression.

Discussion The results of this study support the hypothesis that CaM is an important component of the regulatory mechanisms that control cellular proliferation in eukaryotic cells. The observation that limiting CaM results in growth arrest at multiple points in the cell cycle is consistent with our earlier studies in mouse cell lines (Rasmussen and Means, 1987, 1989). Of greater significance, these results demonstrate that CaM functions at similar points during the cell cycle both in mammalian cells and in AspergitIus suggesting functional conservation of CaM-dependent regulation of growth control. Combined with the facility of genetic analysis in Aspergillus, this system affords ideal opportunities to elucidate the molecular mechanism through which CaM is involved in the control of cell proliferation. Previous studies have suggested that the effects of CaM on cell cycle progression could be mediated by a variety of target proteins. Calcium and CaM are localized to the mitotic spindle (Keith et al, 1985; Welsh et al, 1978, 1979). Calmodulin could potentially affect spindle function via its ability to enhance Ca2+-dependent depolymerization of microtubules (Keith et al, 1983), or by modification of microtubule-associated proteins. Recent studies reveal that a 62-kDa protein in sea urchin mitotic apparatus is phosphorylated by a Ca2+-CaM dependent protein kinase activity. However, neither the identity of the kinase nor its substrate have been reported (Dinsmore and Sloboda, 1988).

Calmodulin is also thought to play a role in regulating the initiation of DNA synthesis. The increase in CaM-binding proteins in the nuclear fraction of ceils stimulated to proliferate (Bachs et al, 1990), and the reported association of CaM with CaM-binding proteins that co-purify with DNA polymerase c~ (Hammond et al, 1989) support our previous studies in mammalian ceils, and the results discussed in this paper that demonstrate a block in G1/S progression in A s p e r g i l l u s that lack CaM. Our future efforts will include isolation of genes that encode CaM-binding proteins from A nidulans, and using the molecular genetic techniques available for use in this organism, determine which of these CaM targets are critical for regulating cell proliferation. Two potential candidate genes have been cloned to date: the CaM-dependent protein kinase homologue of CaM kinase II (by Dr Dianna Bartelt, St John's University) and a CaM-dependent protein phosphatase, by our laboratories. The remarkable similarities of these gene products with their mammalian counterparts suggest that information gained from studies in A nidulans may well be relevant to an understanding of the role CaM plays in mammalian growth control.

Acknowledgments This work has been supported by grants from the American Cancer Society (ARM) the National Cancer Institute of Canada (CDR) and by an Establishment Grant and Scholarship from the Alberta Heritage Foundation for Medical Research (CDR). The assistance of Greg Morrison in the preparation of the figures is appreciated.

References Bachs O, Lanina L, Serratosa J, Coil MJ, Bastos R, Aligue R, Rius E, Carafoli E (1990) J Biol Chem 265, 18595-18600 Chafouleas JG, Bolton WE, Boyd III AE, Means AR (1984) Changes in calmodulin and its mRNA accompany reentry of quiescent (GO) cells into the cell cycle. Cell 36, 73-81 Chafouleas JG, Bolton WE, Hidaka H, Boyd AE, I Means AR (1982) Calmodulin and the cell cycle: Involvement in regulation of cell cycle progression. Cell 28, 41-50

88 Davis TN, Urdea MS, Masiarz FR, Thorner J (1986) Isolation of the yeast calmodulin gene: calmodulin is an essential protein. Cell 47, 423-431 Dinsmore JH, Sloboda RD (1988) Cell 53, 769-780 Hammond RA, Foster KA, Berchtold MW, Gassman M, Holmes AM, Hubscher U, Brown NC (1989) Calcium-dependent calmodulin-binding proteins associated with mammalian DNA polymerase ~. Biochem Biophys Acta 951, 315-321 Hazelton B, Mitchell B, Tupper J (1979) Calcium, magnesium and growth control in the WI-38 fibroblast cell. J Cell Biol 83, 487-498 Jarvik J, Botstein D (1975) Conditional lethal mutations that suppress genetic defects in morphogenesis by altering structural proteins. Proc Natl Acad Sci USA 72, 2738-2742 Keith CH, DiPaola M, Maxfield MR, Shelanski ML (1983) Microinjection of CaZ+-calmodulin causes a localized depolymerization of microtubules. J Cell Biol 97, 1918-1924 Keith CH, Ratan R, Maxfield FR, Bajer A, Shelanski ML (1985) Local cytoplasmic calcium gradients in living mitotic cells. Nature 316, 848-850 May GS (1989) The highly divergent ~-tubulins of Aspergillus nidulans are functionally interchangeable. J Cell Biol 109, 2267-2274 Means AR (1988) Molecular mechanism of action of calmodulin. Rec Prog Horm Res 44, 223-262 Rasmussen CD, Means AR (1987) Calmodulin is involved in regulation of cell proliferation. EMBO J 6, 3961-3968

Rasmussen CD, Means AR (1989) Calmodulin is required for cell cycle progression during G1 and mitosis. EMBO J 8, 73-82 Rasmussen CD, Means RL, Lu KE May GS, Means AR (1990) Expression of the unique calmodulin gene of aspergillus nidulans is essential for cell cycle progression. J Biol Chem 265, 13767-13775 Sasaki Y, Hidaka H (1981) Catmodulin and cell proliferation. Biochem Biophys Res Commun 104, 451456 Takeda T, Yamamoto M (1987) Analysis and in vivo disruption of the gene coding for calmodulin in Schizosaccharomyces pombe. Proc Natl Acad Sci USA

84, 3580-3584 Waring RB, May GS, Morris NR (1989) Characterization of an inducible expression system in Aspergillus nidulans using alcA and tubulin genes. Gene 79, 119-130 Welsh MJ, Dedman JR, Brinkley BR, Means AR (1978) Calcium-dependent regulator protein: localization in the mitotic spindle of eucaryotic cells. Proc Natl Acad Sci USA 75, 1867-1871 Welsh MJ, Dedman JR, Brinkley BR, Means AR (1979) Tubulin and calcium-dependent regulator protein: effects of microtubule and microfilament inhibitors on localization in the mitotic apparatus. J Cell Biol 81, 624-634 Wolniak SM, Hepler PK, Jackson WT (1980) Detection of the membrane-calcium distributon during mitosis in Haemanthus endosperm with chlorotetracycline. J Cell Biol 87, 23-32