ARCHIVES
OF BIOCHEMISTRY
AND
BIOPHYSICS
Vol. 297, No. 1, August 15, pp. 28-34,1992
Modulation of Calmodulin Levels, Calmodulin Methylation, and Calmodulin Binding Proteins during Carrot Cell Growth and Embryogenesis’ Suk-Heung Oh,*y2 Henry-York
Steiner,? Donald K. Dougall,? and Daniel M. Roberts*p3 Department of Biochemistry, and iDepartment of Botany,
*Center for Legume Research and The University of Tennessee, Knoxville,
Received January
Tennessee 37996
21, 1992, and in revised form April 15, 1992
Carrot cell cultures were used to study the dynamics of calmodulin protein levels, calmodulin methylation, and calmodulin-binding proteins during plant growth and development. Comparisons of proliferating and nonproliferating wild carrot cells show that, while calmodulin protein levels does not vary significantly, substantial variation in post-translational methylation of calmodulin on lysine- 115 is observed. Calmodulin methylation is low during the lag and early exponential stages, but increases substantially as exponential growth proceeds and becomes maximal in the postexponential phase. Unmethylated calmodulin quickly reappears within 12 h of reinoculation of cells into fresh media, suggesting that the process is regulated according to the cell growth state. Calmodulin and calmodulin-binding proteins were also analyzed during the formation and germination of domestic carrot embryos in culture. Neither calmodulin methylation nor calmodulin protein levels varied significantly during somatic embryogenesis. However, upon germination of embryos, the level of calmodulin protein doubled. By calmodulin overlay analysis, we have detected a major 54,000 M, calmodulin-binding protein that also increased during embryo germination. This protein was purified from carrot embryo extracts by calmodulinSepharose chromatography. Overall, the data suggest that calmodulin methylation is regulated depending upon the state of cell growth and that calmodulin and its target proteins are modulated during early plant development. 0 1992
Academic
Press,
Inc.
Calmodulin is a ubiquitous calcium-modulated protein that interacts with and stimulates the activity of numer1 Supported by USDA Grants 88-37261-3521 and 91-37305-6752. ’ Present address: Dept. of Biotechnology, Chonju Woo Suk University, Chonju, Korea. ’ To whom correspondence should be addressed.
ous enzymes, and has been proposed to be an integral target of calcium-dependent regulation in eukaryotes (reviewed in (1, 2)). Many calmodulins undergo post-translational trimethylation at lysine-115 that is catalyzed by a calmodulin-specific S-adenosyl-L-methionine:calmodulin-lysine N-methyltransferase (3-5). Although this modification is commonly found in a number of calmodulins, the question of its in uiuo role in calmodulin cell function has yet to be answered. However, based on in uitro studies with unmethylated and methylated calmodulin, it has been shown that the activation of certain enzymes, such as plant NAD kinase, is reduced by calmodulin trimethylation (6), whereas other calmodulinstimulated enzyme activities are not similarly affected (3, 7-11). Thus, it has been suggested that calmodulin methylation may be a mechanism to attenuate the activation level of certain enzymes (12). In order to address the potential significance of calmodulin methylation in plants, we have been analyzing the variations of calmodulin methylation during plant growth and development. In previous studies (5, 12), it was shown that calmodulin isolated from apical regions of roots has a lower level of methylation at position 115 than calmodulin from differentiated, mature root tissues. Since the apical region contains the proliferating meristematic cells and calmodulin homeostasis has been shown to be altered in dividing and nondividing animal cells (1315), it is tempting to propose that post-translational methylation varies in proliferating and nonproliferating cells. However, it is unclear whether the unmethylated calmodulin in the root apex is preferentially associated with a particular cell type, since the apical regions examined in previous experiments (5) consisted of several different tissues (e.g., the root cap, the meristem, epiderma1 cells, cortical cells, and cells that are beginning to differentiate into vascular elements). Subcellular local-
28 All
0003.9861/92 $5.00 Copyright 0 1992 by Academic Press, Inc. rights of reproduction in any form reserved.
CALMODULIN
HOMEOSTASIS
ization studies of root apices show that calmodulin protein levels vary considerably among these various cell types within the root apex (16, 17). In the present study we have utilized carrot cell cultures to expand our study of calmodulin methylation in proliferating and nonproliferating tissues. In addition, since studies with animal cells have indicated differences in the levels of calmodulin during cell division, we have used a calmodulin-specific RIA4 to analyze the levels of calmodulin protein during carrot cell proliferation and during embryogenesis. Finally, we have used calmodulin gel overlay and calmodulin-Sepharose methods to identify potential targets of calmodulin regulation in these tissues. MATERIALS
AND
METHODS
Carrot cell cultures. Two separate lines of carrot cell cultures, a nonembryogenic wild carrot (Daucus carota L.) cell line (25) and an embryogenic domestic carrot (Daucw car&a L. c.v. Danvers half-long) cell line (18), were used in the present study. Wild carrot cell line 13-l was grown in WCM-4 medium (19) containing 2.2 mg/liter 2,4-D. Sevenday-old 13-l cells maintained on WCM-1 medium (20) were inoculated into WCM-4 medium at a ratio of 1:lO (v/v). Cultures were grown at 25°C by shaking at 120 rpm on a gyrotory shaker. Samples were harvested at 2, 4, 5, 6,8, 11, and 14 days after inoculation. For the second growth cycle, cells from the 14-day-old culture were inoculated into fresh WCM4 medium and growth was continued under identical conditions. Samples were taken at 0.5, 1, 2, 3, 4, and 8 days after reinoculation. The cells were collected by filtration on a 70-mm Whatman No. 2 filter paper, were weighed, and were frozen in liquid nitrogen and stored at -80°C. Embryogenic cultures of domestic carrot (18) were a gift of Dr. Lynn Zimmerman (University of Maryland, Baltimore County). Embroygenic carrot cultures were maintained by serial passage of a 10% (v/v) inoculum at weekly intervals. The maintenance medium contained MurashigeSkoog salts and vitamins supplemented with 30 g/liter sucrose, 146 mg/ liter glutamine, 0.1 mg/liter y,y-dimethylallylamino purine, and 0.5 mg/ liter 2.4-D (MS medium). Cultures were grown by shaking at 25°C in the dark on a gyrotory shaker (150 rpm). The cells from 44 to 105 pm were selected by screening l-week-old cultures. This material was cultured at 4 to 5 ~1 packed tissue/ml embryogenesis medium (MS medium without 2,4-D) and embryogenesis was allowed to proceed as previously described (18). After 4 days the cultures were again diluted 1:5 with fresh embryogenesis medium. At this stage, by microscopic examination, the formation of globular embryos had begun. After six additional days of incubation (lo-day-old embryos), the cultures were diluted again 1:lO with embryogenesis medium. At this stage a variety of early stage embryos were present, including globular and heart-shaped embryos. After seven additional days (l7-day-old embryos), the cultures were again diluted 1:lO with fresh embryogenesis medium. Torpedo-shaped embryos were now also present. After seven additional days of incubation (24. day-old embryos) the cultures consisted of many large, germinated plantlets and a fewer number of torpedo-shaped embryos. All samples were frozen in liquid nitrogen immediately after harvest and stored at -8O’C until analyzed.
’ Abbreviations used: AdoMet, S-adenosylmethionine; BSA, bovine serum albumin; CaM, calmodulin; DTT, dithiothreitol; 2,4-D, 2,4-dichlorophenoxyacetic acid, EGTA, ethylene glycol bis(P-aminoethyl ether) N,N’-tetraacetic acid; @ME, &mercaptoethanol; PMSF, phenylmethylsulfonyl fluoride; PVPP, polyvinylpolypyrrolidone; RIA, radioimmunoassay; SDS-PAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis.
IN CARROT
29
CULTURES
Measurement of calmodulin and calmodulin methyl&ion. Calmodulin was extracted from the carrot samples by using a modification of the procedures described by Oh and Roberts (5). The extraction buffer consisted of 100 mM potassium phosphate, 5 mM EDTA, 20 mM /3ME, 1 pg/ml leupeptin, 1 mM PMSF, 10 mM sodium isoascorbate, 10% (w/v) PVPP, pH 7.4. Carrot samples were frozen in liquid nitrogen and ground in a mortar with a pestle. The ground samples were thawed in a volume of extraction buffer that was three times the frozen weight of the samples. The mixture was stirred and then centrifuged at 30,OOOgat 4°C for 15 min. The supernatant was removed and the pellets were reextracted in a volume of the extraction buffer that was the same as that used in the first extraction. Further extraction with chaotropic agents, such as 8 M urea, did not release significant additional amounts of calmodulin (Table I), suggesting that the protocol used resulted in a nearly quantitative extraction of calmodulin from the tissue. Further, in order to verify that the recovery of extracted calmodulin was the same for samples throughout the growth curve, we added ‘251-calmodulin to cell samples and conducted a test extraction by using the protocol described above. The recovery of radiolabeled calmodulin was identical regardless of the age or growth state of the carrot cells (data not shown). For the determination of calmodulin, we used the extraction protocol described above, and the supernatants from the first two extractions were combined. The calmodulin content was measured by using a competitive RIA (5). Calmodulin levels were standardized to the amount of total cellular protein determined by the method of Bradford (21). For methylation analyses, carrot calmodulin was fractionated from the crude extracts by using differential ethanol precipitation (15) or by chromatography on phenyl-Sepharose (22). Both approaches yielded identical results. To determine the degree of calmodulin methylation, a radiometric assay (5) based on the purified calmodulin methyltransferase was used. Homogeneous sheep brain calmodulin methyltransferase was prepared by the purification protocol previously described by Oh and Roberts (5). Based on previous kinetic measurements with plant calmodulins (5,12) this enzyme recognizes plant and animal calmodulins with equal affinity and is suitable for use in methylation analyses of plant calmodulin. Carrot calmodulin (0.1 ng) was incubated with 0.1 Gg of calmodulin methyltransferase and 12 pM [methyl-aH]AdoMet (0.5 &!i total) at 37’C for 4 h, in 100 ~1 of the methyltransferase enzyme reaction mixture as described previously (5). The reaction was terminated by heating the mixture at 90°C for 3 min. [3H]Calmodulin was isolated by phenyl-Sepharose chromatography and was counted by liquid scintillation spectrometry. SDS-PAGE (23) and fluorography were done as previously described (5). Identification and purification of calmodulin-binding proteins. Calmodulin-binding proteins were detected on SDS-PAGE gels by an iz51calmodulin overlay protocol (24) with the modifications described below. After the washing and renaturation steps, the gels were incubated over-
TABLE Extraction
Extraction” I II III
I
of Calmodulin Cell Cultures
Efficiency
Calmodulin 511 231 66
from Carrot
(pmol) *
% 63 29 8
0 Two-day-old carrot cells (1.5 g) were extracted as discussed under Materials and Methods. I, first supernatant fraction obtained with extraction buffer; II, second supernatant fraction obtained with extraction buffer; III, supernatant fraction obtained with extraction buffer containing 8 M urea. * Calmodulin amount was determined by RIA.
30
OH ET AL.
night with ‘251-calmodulin ( lo7 cpm/gel) in 50 mM Tris-HCl, 0.2 M NaCl, 1 mg/ml BSA, pH 7.6, containing either 1 mM CaCl, or 5 mM EDTA. The gels were subsequently washed with 100 ml of 50 mM Tris-HCl, 0.2 M NaCl, pH 7.6, containing either 0.5 mM CaCl, or 2 mM EDTA, for 3 h with three buffer changes. All buffers were filtered through millipore filters (0.2 pm) into acid-washed glassware. We found that these modifications reduce the nonspecific background staining of the gel matrix. The washed gels were stained in Coomassie blue, were dried, and were exposed to X-ray film at -8O’C with a Cronex intensifying screen. In order to purify calmodulin-binding proteins from carrot embryos, 24-day-old germinated embryo samples (8 g) were frozen in liquid nitrogen and ground in a mortar with a pestle. The ground samples were thawed in 25 ml of 50 mM Tris-HCl, pH 7.5, containing 5 mM EDTA, 10% (w/v) PVPP, 20 mM @ME, 1 pg/ml leupeptin, 1 mM PMSF, 10 mM sodium isoascorbate, and 0.2 M NaCl. The mixture was stirred and then centrifuged at 30,500g at 4°C for 15 min. The supernatant was applied to a DEAE-cellulose column (1.5 X 6 cm) equilibrated with 50 mM TrisHCl, 0.2 M NaCl, 3 mM MgClz, 1 mM EGTA, pH 7.5, and the effluent was collected. CaCl, was added to a final concentration of 1 mM and the samples were applied to a bovine brain calmodulin-Sepharose column (1.0 X 1.0 cm) equilibrated with 25 mM Tris-HCl, 0.25 mM CaC&, 3 mM MgClx, 0.15 M NaCl, 1 mM DTT, pH 7.5. The column was washed with 1.5 mM 50 ml of the same buffer and was eluted with 25 mM Tris-HCl, EGTA, 3 mM MgClr, 0.15 M NaCl, 1 mM DTT, pH 7.5. The elutant was concentrated and stored at -80°C until ‘%I-calmodulin overlay and SDSPAGE analyses.
RESULTS Analyses of Calmodulin in Proliferating Nonproliferating Wild Carrot Cells
and
To analyze the levels of calmodulin and the state of calmodulin methylation during the course of plant cell growth, we have used a wild carrot cell culture system (25). As shown in Fig. 1, carrot cells exhibit a typical growth curve, including an exponential growth phase between 2 and 8 days after inoculation and an apparent stationary phase occurring after 10 days. Upon reinoculation into fresh media, cells reenter the exponential phase within 4 days (Fig. 1). Calmodulin was extracted from the
25
carrot cell samples collected during various stages of growth and the levels of calmodulin were determined by RIA. Although slight fluctuations were observed from sample to sample, there was no clear trend of calmodulin increase, and the overall levels of calmodulin were similar throughout the growth curve (Fig. 1A). In contrast, there were substantial differences in the state of post-translational methylation of calmodulins during the course of cell growth (Fig. 1B). In order to assay the level of methylation, calmodulin was isolated and was incubated with the calmodulin methyltransferase in the presence of [methyl-3H]AdoMet. This analysis is based on the ability of unmethylated or undermethylated calmodulins to be methylated by this enzyme (5). Thus, if the endogenous calmodulin is not methylated at lysine115 in Go, it will show a high ability to accept methyl3H-groups in the in vitro methyltransferase assay. Calmodulin isolated from the early exponential stages of the growth curve (i.e., 2- and 4-day) incorporates as high as 1.4 mol methyl-3H-group/mol calmodulin. As growth proceeds the level of endogenous calmodulin methylation increases. This is reflected by the fact that the methyl-3Hgroup incorporation into calmodulin isolated from postexponential ll-day-old cultures was sevenfold lower (0.22 mol/mol calmodulin). Fluorographic analyses support these findings and show that calmodulin was the only detectable product of methylation (Fig. 2). In order to address whether rapid changes in the state of post-translational calmodulin methylation were associated with reentry into mitosis, we reinnoculated cells from the stationary phase (Day 14) into fresh media (Day 15). Methylation analyses showed that within 12 h after reinoculation, unmethylated calmodulin quickly reaccummulated, independent of changes in calmodulin protein concentration, and reached a maximum level 1 day
E;If2QI l,.6f@, .$?. -20 !j do ?8_ 5 F
-z
!F
A
i : 5 0 . 0’. . . . . . . . . . . 0
-30
,/"
20-
5
10 3
000 _,.,
15
10
days
;
..,.
20
=
250 days
FIG. 1. Analysis of the levels of calmodulin protein and the methylation state of calmodulin in wild carrot cell cultures. (A) The levels of calmodulin protein (expressed as pmol CaM/mg extracted protein) determined by RIA are indicated by the closed circles. Open circles represent the fresh weight of the carrot cells (mg packed cells/ml culture); (B) The calmodulin methylation state (expressed as mol methyl-3H-groups incorporated/mol calmodulin) is represented by the closed triangles. The open circles represent the carrot cell fresh weight. All values are the averages of two determinations, with error bars showing standard errors. The absence of error bars indicates that the range is smaller than the size of the symbol. At Day 14 the cells were inoculated into fresh media as described under Materials and Methods.
CALMODULIN
HOMEOSTASIS
IN CARROT
ZO 0
12
5
after reinoculation. Overall, the data demonstrate that calmodulin is highly unmethylated at position 115 in the early exponential stage of carrot cell growth and that the level of endogenous calmodulin methylation increases as growth proceeds and cells enter the stationary phase. Further, the process seems to be subject to rapid regulation since unmethylated calmodulins reappear quickly upon reinoculation into fresh media.
10
15
20
25
days
34
FIG. 2. Fluorographic analysis of 3H-methylated wild carrot calmodulin. Carrot calmodulin samples were subjected to methylation analysis by using the calmodulin methyltransferase and [nethyl-3H]AdoMet as described under Materials and Methods. The samples (0.15 pg of each) were resolved by SDS-PAGE on a 15% polyacrylamide gel and were analyzed fluorographically. Samples are from 2-day (lane l), 4-day (lane 2), &day (lane 3), and ll-day (lane 4) cultures.
31
CULTURES
FIG. 3. Analysis of the levels of calmodulin protein and the methylation state of calmodulin during domestic carrot somatic embryogenesis. The calmodulin protein concentration (pmol CaM/mg protein) is represented by the open circles and the calmodulin methylation state (mol methyl-3H-groups/mol CaM) is represented by the closed circles. Values are the averages of two determinations, with error bars showing the standard errors. The absence of error bars indicates that the range is smaller than the size of the symbol.
tected in embryo extracts (Fig. 4). Since no bands were observed on the autoradiogram of a duplicate gel exposed to ‘251-calmodulin in the presence of EDTA or EGTA (data not shown), the bands appeared to represent calcium-dependent calmodulin-binding proteins. Although several calmodulin-binding proteins were detected in undifferentiated cells as well as in developing and germinated embryos, some proteins appeared to vary during embryo-
Analysis of Calmodulin and Calmodulin-Binding Proteins during Carrot Embryogenesis In order to further investigate the calmodulin system during plant development, we have utilized carrot embryogenesis as a model system. Domestic carrot cell cultures transferred to media devoid of 2,4-D (O-day sample) developed into embryos within 17 days. By the end of the 24th day the embryos germinated. Carrot samples from the O-day, lo-day, 17-day, and 24-day stages were harvested and analyzed using the same protocols as described above. Significant differences in the levels of calmodulin protein were not observed among the embryo samples from O-day, lo-day, and 17-day cultures. However, after passing the 17-day point, the level of calmodulin began to increase and doubled by the end of the 24th day (Fig. 3). Unlike the results obtained with proliferating vs nonproliferating wild carrot cells, the calmodulin methylation levels showed less dramatic changes during embryogenesis, and methyl-3H-group incorporation ranged from 0.3 to 0.2 mol/mol calmodulin (Fig. 3). Calmodulin overlay analysis was used to identify potential target proteins of calmodulin action and to determine whether differences in these calmodulin-binding proteins occur during the formation and germination of embryos. Several calmodulin-binding proteins were de-
MW
X
10-3
946743-
30-
2012341234 FIG. 4. 1251-calmodulin overlay analysis of extracts of cultured domestic carrot cells undergoing embryogenesis. Samples representing Oday (lane l), lo-day (lane 2), 17.day (lane 3), and 24.day (lane 4) carrot samples (16.7 pg of protein in each lane) were subjected to SDS-PAGE on a 12.5% polyacrylamide gel. (A) Coomassie blue-stained gel. (B) Autoradiogram of the gel exposed to ‘251-calmodulin in the presence of 1 mM
Cd&.
32
OH ET AL.
genesis and germination. For example, a 26,000 M, calmodulin-binding protein appeared to be preferentially associated with developing embryos and was present at a very low level in undifferentiated cells. In addition, calmodulin binding to a 54,000 M, protein increased during embryo germination and this protein was the major calmodulin-binding protein detected in extracts of germinated carrot embryos. This protein is similar in properties to a major calmodulin binding protein described in carrot protoplasts (37). As shown in Fig. 4, the increase in the 54,000 M, protein in carrot embryos occurred between 17 and 24 days after the initiation of embryogenesis and paralleled the increase observed in calmodulin levels in the germinated embryos (Fig. 3). In an effort to purify the 54,000 M, protein, we used calmodulin-Sepharose chromatography. The extract of 24-day-old carrot embryos was first chromatographed on DEAE-cellulose to separate the endogenous carrot calmodulin from the 54,000 M, calmodulin-binding protein. The resulting sample was adsorbed to calmodulin-sepharose in the presence of calcium and was eluted with EGTA. As shown in Fig. 5, the only polypeptide detected in the calmodulin-Sepharose eluate by calmodulin overlay or by protein staining was the 54,000 M, polypeptide. This result verifies that the native, undenatured 54,000 M, protein binds to calmodulin and, together with the overlay data, provides strong evidence that this protein is a bona fide calmodulin-binding protein. DISCUSSION In the present study we have analyzed calmodulin homeostasis during plant cell proliferation and embryo formation by using carrot culture model systems. During the growth of nonembryogenic wild carrot cultures, large differences in the levels of post-translational methylation of calmodulin were seen depending upon the stage of cell growth, with the highest levels of unmethylated calmodulin associated with the early exponential phase. We previously have shown that transient pools of undermethylated calmodulin exist in certain plant tissues, such as apical and young lateral root segments, that contain meristematic regions (5, 12). The present findings support this previous work and suggest that undermethylated calmodulin is preferentially associated with proliferating cell populations. The changes in methylation were not paralleled by changes in calmodulin levels, which did not appear to vary significantly during the growth of the carrot cells. Similarly, the calmodulin levels in developing domestic carrot embryos remained constant. However, the calmodulin protein levels doubled upon embryo germination. This increase was accompanied by an increase in a 54,000 M, calmodulin-binding protein which is the major apparent calmodulin-receptor protein detected in the germinated carrot embryos.
96 67-
30-
20-
I
2
1
2
FIG. 5. SDS-PAGE and ‘251-calmodulin overlay analysis of purified calmodulin-binding proteins from 24-day-old domestic carrot somatic embryos. Proteins were resolved by SDS-PAGE on 12.5% polyacrylamide gels. (A) Coomassie blue-stained gel. (B) Autoradiogram of the overlay gel exposed to ‘251-calmodulin in the presence of 1 mM CaCl,. Lane 1, crude extract; lane 2, 54,000 M, calmodulin-binding protein purified by calmodulin-Sepharose.
The data suggest that the levels of unmethylated calmodulin are regulated independent of calmodulin protein concentrations. For example, upon inoculation of postexponential phase cells into fresh media, unmethylated calmodulin quickly reaccumulates while there is little change in overall calmodulin protein levels. This reaccumulation is striking since it happens within 12 h of reinoculation, suggesting that a substantial turnover of methylated calmodulin occurs during this period. Although, it is currently unknown what factors control calmodulin methylation in plants, changes in the levels of the calmodulin methyltransferase present in the cells could be involved. For example, based on previous observations with animal systems, it has been shown that the levels of calmodulin methyltransferase enzyme can vary depending on the source of the tissue (reviewed in (26)). Little is currently known regarding the plant calmodulin methyltransferase. Further investigations of the properties of the plant enzyme are warranted to determine whether it is regulated during plant cell growth. In addition to the potential regulation of methyltransferase levels, calmodulin methylation could also be influenced by calcium binding to calmodulin. For example, Roberts et al. (12) observed that calmodulin methylation is stimulated by micromolar calcium concentrations, and it has been suggested that the calcium-bound form of calmodulin may be preferred by the calmodulin methyltransferase. Thus, calcium fluxes during cell growth could influence the rate of calmodulin methylation. The degree
CALMODULIN
HOMEOSTASIS
of calmodulin methylation also could be controlled by the availability of the methyl donor, AdoMet. Recently, Minocha et al. (27) observed that the specific activity of AdoMet synthetase in the extract of the &day-old carrot cells is more than 25-fold higher than that in the extract of the l-day-old carrot cells. Thus, the increase in AdoMet synthetase levels during carrot growth appears to parallel the increase in methylated calmodulin seen in our studies. The methylation of calmodulin has a significant effect on its ability to activate plant NAD kinases, with methylated calmodulins exhibiting a lower level of activation compared to unmethylated calmodulins (6, 12). In addition, there is evidence that lysine-115 but not trimethyllysine-115 is a site for ubiquitination of calmodulin by animal cell extracts under in vitro conditions, and it has been suggested that methylation may protect calmodulin from intracellular proteolytic turnover (38). It is tempting to suggest that variations in calmodulin methylation may be a mechanism for attenuating its enzyme activator properties or could influence cellular levels of calmodulin by affecting its susceptibility to proteolytic turnover. However, there is not yet any support for these roles for calmodulin methylation in uiuo. Further, as discussed below, there is no clear trend in the fluctuation of calmodulin levels during carrot cell growth that may reflect a change in calmodulin turnover. However, future work with the carrot culture system including the investigation of nicotinamide coenzyme homeostasis and calmodulin turnover during cell growth may provide insight into these potential roles of post-translational methylation of lysine115 of calmodulin. Our findings that the overall levels of calmodulin are not enhanced in exponential and postexponential wild carrot culture cells contrast with those of other studies with transformed animal cell models that show higher levels of calmodulin in rapidly proliferating cells compared to those in quiescent cells (13-15). Although the exact role of this increase in calmodulin during cell growth has not yet been established, it has been proposed that a transient increase in calmodulin during the G,/S boundary of the cell cycle can control the rate of cell cycle progression (28-30) by influencing the duration of the G, phase (29, 30). Since the carrot cultures used in the present study were not synchronized, it is not yet clear whether transient increases in calmodulin occurred at specific cell cycle stages that may have evaded detection. In plant tissues, it also has been proposed that calmodulin levels may be preferentially enhanced in dividing cells, since meristematic regions appear to have higher calmodulin protein levels compared to mature tissues based on immunocytochemical(16,17), enzyme activator (31), and RIA (32) analyses. This trend has also been found with calmodulin mRNA levels (33). However, there are fundamental differences among these systems of study. In the present study calmodulin homeostasis in
IN CARROT
CULTURES
33
dividing and nondividing undifferentiated cells in culture was investigated. In contrast, studies with apical and mature root and shoot tissues differ since heterogeneous populations of cells that are in the process of differentiating into specialized tissues and cell types are being compared, and thus these are not simple comparisons of proliferating and nonproliferating cells. In contrast to our findings with undifferentiated carrot cells, we observed that the germination of carrot embryos generated in culture resulted in a twofold increase in the concentration of calmodulin protein. Similar increases in the levels of calmodulin have also been observed during the germination of certain seed embryos, such as radish (34,35) and chick pea embryos (36). Currently, the question of the relationship between enhanced calmodulin levels and embryo germination is unclear, but the elevation of calmodulin levels could play a role in calciumsignal transduction processes during germination. In this regard it is of interest that the apparent levels of a major 54,000 M, calmodulin-binding protein also increase substantially during the germination of carrot embryos. Future studies, including the identification and characterization of this major calmodulin-binding protein, may provide insight into the roles of calmodulin in calciumsignal transduction in the germinating embryo. REFERENCES 1. Cohen, P., and Klee, C. B. (1988) Molecular Aspects of Cellular Regulation: Calmodulin, Elsevier, Amsterdam. 2. Roberts, D. M., and Harmon, A. C. (1992) Annu. Reu. Plant Physiol. Mol. Biol. 43, 375-414. 3. Rowe, P. M., Wright, L. S., and Siegel, F. L. (1986) J. Biol. Chem. 261,7060-7069. 4. Morino, H., Kawamoto, T., Miyake, M., and Kakimoto, Y. (1987) J. Neurochem. 48, 1201-1208. 5. Oh, S.-H., and Roberts, D. M. (1990) Plant Physiol. 93,880-887. 6. Roberts, D. M., Rowe, P. M., Siegel, F. L., Lukas, T. J., and Watterson, D. M. (1986) J. Biol. C&m. 261, 1491-1494. 7. Roberts, D. M., Burgess, W. H., and Watterson, D. M. (1984) Plant Physiol. 75, 796-798. a. Marshak, D. R., Clarke, M., Roberts, D. M., and Watterson, D. M. (1984) Biochemistry 23,2891-2899. G., Chi9. Roberts, D. M., Crea, R., Malecha, M., Alvarado-Urbina, arello, R. H., and Watterson, D. M. (1985) Biochemistry 24,50905098. 10. Putkey, J. A., Slaughter, G. R., and Means, A. R. (1985) J. Biol. Chem. 260,4704-4712. 11. Putkey, J. A., Draetta, G. F., Slaughter, G. R., Klee, C. B., Cohen, P., Stull, J. T., and Means, A. R. (1986) J. Biol. Chem. 261,98969903. 12. Roberts, D. M., Oh, S.-H., Besl, L., Weaver, C. D., and Stacey, G. (1990) in Current Topics in Plant Biochemistry and Physiology (Randall, D. D., and Blevins, D. G., Eds.), Vol. 9, pp. 67-84, Univ. of Missouri Press, Columbia, MO. 13. Watterson, D. M., Van Eldik, L. J., Smith, R. E., and Vanaman, T. C. (1976) Proc. Natl. Acad. Sci. USA 73, 2711-2715. 14. Chafouleas, J. G., Bolton, W. E., Boyd, A. E., and Means, A. R. (1981) Proc. Natl. Acad. Sci. USA 78, 996-1000.
34
OH ET AL.
15. Zendegui, J. G., Zielinski, R. E., Watterson, L. J. (1984) Mol. Cell. Biol. 4, 883-889.
D. M., and Van Eldik,
16. Lin, C.-T., Sun, D., Song, G.-X., and Wu, J.-Y. (1986) J. Ifistochem. Cytochem. 34,561-567. 17. Dauwalder, 461-470.
M., Roux, S. J., and Hardison,
18. Zimmerman, J. L., Apuya, N., Darwish, Plant Cell 1,1137-1146. 19. Dougall, D. K., and Weyrauch,
L. (1986) Plunta
K., and O’Carroll,
C. (1989)
K. W. (1980) In Vitro 16,969-975.
20. Wetherell,
D. F. (1969) Plant Physiol. 44,1734-1737.
21. Bradford,
M. M. (1976) And. Biochem. 72, 248-254.
22. Gopalakrishna, R., and Anderson, Res. Commun. 104,830-836. 23. Laemmli,
168,
W. B. (1982) Biochem. Biophys.
U. K. (1970) Nature 227, 680-685.
24. Burgess, W. H., Watterson, Cell Biol. 99, 550-557.
D. M., and Van Eldik, L. J. (1984) J.
25. Dougall, D. K., Johnson, J. M., and Whitten,
149,292-297.
G. H. (1980) Planta
26. Siegel, F. L., Vincent, P. L., Neal, T. L., Wright, L. S., Heth, A. A., and Rowe, P. M. (1990) in Protein Methylation (Paik, W. K., and Kim, S., Eds.), pp. 33-58, CRC Press, Boca Raton, FL. 27. Minocha, S. C., Minocha, R., and Komamine, A. (1991) Plant Physiol. Biochem. 29, 231-237. 28. Chafouleas, J. G., Bolton, W. E., Boyd, A. E., and Means, A. R. (1982) Cell 28, 41-50. 29. Rasmussen, C. D., and Means, A. R. (1987) EMBO J. 6,3961-3968. 30. Rasmussen, C. D., and Means, A. R. (1989) EMBO J. 8, 73-82. 31. Allan, E. F., and Trewavas, A. J. (1985) Planta 165,493-501. 32. Muto, S., and Miyachi, S. (1984) 2. Pflanzenphysiol. Bd. 114,421431. 33. Zielinski, R. E. (1987) Plant Physiol. 84,937-943. 34. Cocucci, M. (1984) Plant Cell Enuiron. 7,215-221. 35. Cocucci, M., and Negrini, N. (1988) Plant Physiol. 88, 910-914. J., Rodriguez, D., Nicolas, G., and Aldasoro, 36. Hernandez-Nistal, J. J. (1989) Physiol. Plant. 75, 255-260. 37. Cyr, R. J. (1991) J. Cell Sci. 100, 311-317. 38. Gregori, L., Marriott, D., Putkey, J. A., Means, A. R., and Chau, V. (1987) J. Biol. Chem. 262, 2562-2567.
OF BIOCHEMISTRY
AND
BIOPHYSICS
Vol. 297, No. 1, August 15, pp. 28-34,1992
Modulation of Calmodulin Levels, Calmodulin Methylation, and Calmodulin Binding Proteins during Carrot Cell Growth and Embryogenesis’ Suk-Heung Oh,*y2 Henry-York
Steiner,? Donald K. Dougall,? and Daniel M. Roberts*p3 Department of Biochemistry, and iDepartment of Botany,
*Center for Legume Research and The University of Tennessee, Knoxville,
Received January
Tennessee 37996
21, 1992, and in revised form April 15, 1992
Carrot cell cultures were used to study the dynamics of calmodulin protein levels, calmodulin methylation, and calmodulin-binding proteins during plant growth and development. Comparisons of proliferating and nonproliferating wild carrot cells show that, while calmodulin protein levels does not vary significantly, substantial variation in post-translational methylation of calmodulin on lysine- 115 is observed. Calmodulin methylation is low during the lag and early exponential stages, but increases substantially as exponential growth proceeds and becomes maximal in the postexponential phase. Unmethylated calmodulin quickly reappears within 12 h of reinoculation of cells into fresh media, suggesting that the process is regulated according to the cell growth state. Calmodulin and calmodulin-binding proteins were also analyzed during the formation and germination of domestic carrot embryos in culture. Neither calmodulin methylation nor calmodulin protein levels varied significantly during somatic embryogenesis. However, upon germination of embryos, the level of calmodulin protein doubled. By calmodulin overlay analysis, we have detected a major 54,000 M, calmodulin-binding protein that also increased during embryo germination. This protein was purified from carrot embryo extracts by calmodulinSepharose chromatography. Overall, the data suggest that calmodulin methylation is regulated depending upon the state of cell growth and that calmodulin and its target proteins are modulated during early plant development. 0 1992
Academic
Press,
Inc.
Calmodulin is a ubiquitous calcium-modulated protein that interacts with and stimulates the activity of numer1 Supported by USDA Grants 88-37261-3521 and 91-37305-6752. ’ Present address: Dept. of Biotechnology, Chonju Woo Suk University, Chonju, Korea. ’ To whom correspondence should be addressed.
ous enzymes, and has been proposed to be an integral target of calcium-dependent regulation in eukaryotes (reviewed in (1, 2)). Many calmodulins undergo post-translational trimethylation at lysine-115 that is catalyzed by a calmodulin-specific S-adenosyl-L-methionine:calmodulin-lysine N-methyltransferase (3-5). Although this modification is commonly found in a number of calmodulins, the question of its in uiuo role in calmodulin cell function has yet to be answered. However, based on in uitro studies with unmethylated and methylated calmodulin, it has been shown that the activation of certain enzymes, such as plant NAD kinase, is reduced by calmodulin trimethylation (6), whereas other calmodulinstimulated enzyme activities are not similarly affected (3, 7-11). Thus, it has been suggested that calmodulin methylation may be a mechanism to attenuate the activation level of certain enzymes (12). In order to address the potential significance of calmodulin methylation in plants, we have been analyzing the variations of calmodulin methylation during plant growth and development. In previous studies (5, 12), it was shown that calmodulin isolated from apical regions of roots has a lower level of methylation at position 115 than calmodulin from differentiated, mature root tissues. Since the apical region contains the proliferating meristematic cells and calmodulin homeostasis has been shown to be altered in dividing and nondividing animal cells (1315), it is tempting to propose that post-translational methylation varies in proliferating and nonproliferating cells. However, it is unclear whether the unmethylated calmodulin in the root apex is preferentially associated with a particular cell type, since the apical regions examined in previous experiments (5) consisted of several different tissues (e.g., the root cap, the meristem, epiderma1 cells, cortical cells, and cells that are beginning to differentiate into vascular elements). Subcellular local-
28 All
0003.9861/92 $5.00 Copyright 0 1992 by Academic Press, Inc. rights of reproduction in any form reserved.
CALMODULIN
HOMEOSTASIS
ization studies of root apices show that calmodulin protein levels vary considerably among these various cell types within the root apex (16, 17). In the present study we have utilized carrot cell cultures to expand our study of calmodulin methylation in proliferating and nonproliferating tissues. In addition, since studies with animal cells have indicated differences in the levels of calmodulin during cell division, we have used a calmodulin-specific RIA4 to analyze the levels of calmodulin protein during carrot cell proliferation and during embryogenesis. Finally, we have used calmodulin gel overlay and calmodulin-Sepharose methods to identify potential targets of calmodulin regulation in these tissues. MATERIALS
AND
METHODS
Carrot cell cultures. Two separate lines of carrot cell cultures, a nonembryogenic wild carrot (Daucus carota L.) cell line (25) and an embryogenic domestic carrot (Daucw car&a L. c.v. Danvers half-long) cell line (18), were used in the present study. Wild carrot cell line 13-l was grown in WCM-4 medium (19) containing 2.2 mg/liter 2,4-D. Sevenday-old 13-l cells maintained on WCM-1 medium (20) were inoculated into WCM-4 medium at a ratio of 1:lO (v/v). Cultures were grown at 25°C by shaking at 120 rpm on a gyrotory shaker. Samples were harvested at 2, 4, 5, 6,8, 11, and 14 days after inoculation. For the second growth cycle, cells from the 14-day-old culture were inoculated into fresh WCM4 medium and growth was continued under identical conditions. Samples were taken at 0.5, 1, 2, 3, 4, and 8 days after reinoculation. The cells were collected by filtration on a 70-mm Whatman No. 2 filter paper, were weighed, and were frozen in liquid nitrogen and stored at -80°C. Embryogenic cultures of domestic carrot (18) were a gift of Dr. Lynn Zimmerman (University of Maryland, Baltimore County). Embroygenic carrot cultures were maintained by serial passage of a 10% (v/v) inoculum at weekly intervals. The maintenance medium contained MurashigeSkoog salts and vitamins supplemented with 30 g/liter sucrose, 146 mg/ liter glutamine, 0.1 mg/liter y,y-dimethylallylamino purine, and 0.5 mg/ liter 2.4-D (MS medium). Cultures were grown by shaking at 25°C in the dark on a gyrotory shaker (150 rpm). The cells from 44 to 105 pm were selected by screening l-week-old cultures. This material was cultured at 4 to 5 ~1 packed tissue/ml embryogenesis medium (MS medium without 2,4-D) and embryogenesis was allowed to proceed as previously described (18). After 4 days the cultures were again diluted 1:5 with fresh embryogenesis medium. At this stage, by microscopic examination, the formation of globular embryos had begun. After six additional days of incubation (lo-day-old embryos), the cultures were diluted again 1:lO with embryogenesis medium. At this stage a variety of early stage embryos were present, including globular and heart-shaped embryos. After seven additional days (l7-day-old embryos), the cultures were again diluted 1:lO with fresh embryogenesis medium. Torpedo-shaped embryos were now also present. After seven additional days of incubation (24. day-old embryos) the cultures consisted of many large, germinated plantlets and a fewer number of torpedo-shaped embryos. All samples were frozen in liquid nitrogen immediately after harvest and stored at -8O’C until analyzed.
’ Abbreviations used: AdoMet, S-adenosylmethionine; BSA, bovine serum albumin; CaM, calmodulin; DTT, dithiothreitol; 2,4-D, 2,4-dichlorophenoxyacetic acid, EGTA, ethylene glycol bis(P-aminoethyl ether) N,N’-tetraacetic acid; @ME, &mercaptoethanol; PMSF, phenylmethylsulfonyl fluoride; PVPP, polyvinylpolypyrrolidone; RIA, radioimmunoassay; SDS-PAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis.
IN CARROT
29
CULTURES
Measurement of calmodulin and calmodulin methyl&ion. Calmodulin was extracted from the carrot samples by using a modification of the procedures described by Oh and Roberts (5). The extraction buffer consisted of 100 mM potassium phosphate, 5 mM EDTA, 20 mM /3ME, 1 pg/ml leupeptin, 1 mM PMSF, 10 mM sodium isoascorbate, 10% (w/v) PVPP, pH 7.4. Carrot samples were frozen in liquid nitrogen and ground in a mortar with a pestle. The ground samples were thawed in a volume of extraction buffer that was three times the frozen weight of the samples. The mixture was stirred and then centrifuged at 30,OOOgat 4°C for 15 min. The supernatant was removed and the pellets were reextracted in a volume of the extraction buffer that was the same as that used in the first extraction. Further extraction with chaotropic agents, such as 8 M urea, did not release significant additional amounts of calmodulin (Table I), suggesting that the protocol used resulted in a nearly quantitative extraction of calmodulin from the tissue. Further, in order to verify that the recovery of extracted calmodulin was the same for samples throughout the growth curve, we added ‘251-calmodulin to cell samples and conducted a test extraction by using the protocol described above. The recovery of radiolabeled calmodulin was identical regardless of the age or growth state of the carrot cells (data not shown). For the determination of calmodulin, we used the extraction protocol described above, and the supernatants from the first two extractions were combined. The calmodulin content was measured by using a competitive RIA (5). Calmodulin levels were standardized to the amount of total cellular protein determined by the method of Bradford (21). For methylation analyses, carrot calmodulin was fractionated from the crude extracts by using differential ethanol precipitation (15) or by chromatography on phenyl-Sepharose (22). Both approaches yielded identical results. To determine the degree of calmodulin methylation, a radiometric assay (5) based on the purified calmodulin methyltransferase was used. Homogeneous sheep brain calmodulin methyltransferase was prepared by the purification protocol previously described by Oh and Roberts (5). Based on previous kinetic measurements with plant calmodulins (5,12) this enzyme recognizes plant and animal calmodulins with equal affinity and is suitable for use in methylation analyses of plant calmodulin. Carrot calmodulin (0.1 ng) was incubated with 0.1 Gg of calmodulin methyltransferase and 12 pM [methyl-aH]AdoMet (0.5 &!i total) at 37’C for 4 h, in 100 ~1 of the methyltransferase enzyme reaction mixture as described previously (5). The reaction was terminated by heating the mixture at 90°C for 3 min. [3H]Calmodulin was isolated by phenyl-Sepharose chromatography and was counted by liquid scintillation spectrometry. SDS-PAGE (23) and fluorography were done as previously described (5). Identification and purification of calmodulin-binding proteins. Calmodulin-binding proteins were detected on SDS-PAGE gels by an iz51calmodulin overlay protocol (24) with the modifications described below. After the washing and renaturation steps, the gels were incubated over-
TABLE Extraction
Extraction” I II III
I
of Calmodulin Cell Cultures
Efficiency
Calmodulin 511 231 66
from Carrot
(pmol) *
% 63 29 8
0 Two-day-old carrot cells (1.5 g) were extracted as discussed under Materials and Methods. I, first supernatant fraction obtained with extraction buffer; II, second supernatant fraction obtained with extraction buffer; III, supernatant fraction obtained with extraction buffer containing 8 M urea. * Calmodulin amount was determined by RIA.
30
OH ET AL.
night with ‘251-calmodulin ( lo7 cpm/gel) in 50 mM Tris-HCl, 0.2 M NaCl, 1 mg/ml BSA, pH 7.6, containing either 1 mM CaCl, or 5 mM EDTA. The gels were subsequently washed with 100 ml of 50 mM Tris-HCl, 0.2 M NaCl, pH 7.6, containing either 0.5 mM CaCl, or 2 mM EDTA, for 3 h with three buffer changes. All buffers were filtered through millipore filters (0.2 pm) into acid-washed glassware. We found that these modifications reduce the nonspecific background staining of the gel matrix. The washed gels were stained in Coomassie blue, were dried, and were exposed to X-ray film at -8O’C with a Cronex intensifying screen. In order to purify calmodulin-binding proteins from carrot embryos, 24-day-old germinated embryo samples (8 g) were frozen in liquid nitrogen and ground in a mortar with a pestle. The ground samples were thawed in 25 ml of 50 mM Tris-HCl, pH 7.5, containing 5 mM EDTA, 10% (w/v) PVPP, 20 mM @ME, 1 pg/ml leupeptin, 1 mM PMSF, 10 mM sodium isoascorbate, and 0.2 M NaCl. The mixture was stirred and then centrifuged at 30,500g at 4°C for 15 min. The supernatant was applied to a DEAE-cellulose column (1.5 X 6 cm) equilibrated with 50 mM TrisHCl, 0.2 M NaCl, 3 mM MgClz, 1 mM EGTA, pH 7.5, and the effluent was collected. CaCl, was added to a final concentration of 1 mM and the samples were applied to a bovine brain calmodulin-Sepharose column (1.0 X 1.0 cm) equilibrated with 25 mM Tris-HCl, 0.25 mM CaC&, 3 mM MgClx, 0.15 M NaCl, 1 mM DTT, pH 7.5. The column was washed with 1.5 mM 50 ml of the same buffer and was eluted with 25 mM Tris-HCl, EGTA, 3 mM MgClr, 0.15 M NaCl, 1 mM DTT, pH 7.5. The elutant was concentrated and stored at -80°C until ‘%I-calmodulin overlay and SDSPAGE analyses.
RESULTS Analyses of Calmodulin in Proliferating Nonproliferating Wild Carrot Cells
and
To analyze the levels of calmodulin and the state of calmodulin methylation during the course of plant cell growth, we have used a wild carrot cell culture system (25). As shown in Fig. 1, carrot cells exhibit a typical growth curve, including an exponential growth phase between 2 and 8 days after inoculation and an apparent stationary phase occurring after 10 days. Upon reinoculation into fresh media, cells reenter the exponential phase within 4 days (Fig. 1). Calmodulin was extracted from the
25
carrot cell samples collected during various stages of growth and the levels of calmodulin were determined by RIA. Although slight fluctuations were observed from sample to sample, there was no clear trend of calmodulin increase, and the overall levels of calmodulin were similar throughout the growth curve (Fig. 1A). In contrast, there were substantial differences in the state of post-translational methylation of calmodulins during the course of cell growth (Fig. 1B). In order to assay the level of methylation, calmodulin was isolated and was incubated with the calmodulin methyltransferase in the presence of [methyl-3H]AdoMet. This analysis is based on the ability of unmethylated or undermethylated calmodulins to be methylated by this enzyme (5). Thus, if the endogenous calmodulin is not methylated at lysine115 in Go, it will show a high ability to accept methyl3H-groups in the in vitro methyltransferase assay. Calmodulin isolated from the early exponential stages of the growth curve (i.e., 2- and 4-day) incorporates as high as 1.4 mol methyl-3H-group/mol calmodulin. As growth proceeds the level of endogenous calmodulin methylation increases. This is reflected by the fact that the methyl-3Hgroup incorporation into calmodulin isolated from postexponential ll-day-old cultures was sevenfold lower (0.22 mol/mol calmodulin). Fluorographic analyses support these findings and show that calmodulin was the only detectable product of methylation (Fig. 2). In order to address whether rapid changes in the state of post-translational calmodulin methylation were associated with reentry into mitosis, we reinnoculated cells from the stationary phase (Day 14) into fresh media (Day 15). Methylation analyses showed that within 12 h after reinoculation, unmethylated calmodulin quickly reaccummulated, independent of changes in calmodulin protein concentration, and reached a maximum level 1 day
E;If2QI l,.6f@, .$?. -20 !j do ?8_ 5 F
-z
!F
A
i : 5 0 . 0’. . . . . . . . . . . 0
-30
,/"
20-
5
10 3
000 _,.,
15
10
days
;
..,.
20
=
250 days
FIG. 1. Analysis of the levels of calmodulin protein and the methylation state of calmodulin in wild carrot cell cultures. (A) The levels of calmodulin protein (expressed as pmol CaM/mg extracted protein) determined by RIA are indicated by the closed circles. Open circles represent the fresh weight of the carrot cells (mg packed cells/ml culture); (B) The calmodulin methylation state (expressed as mol methyl-3H-groups incorporated/mol calmodulin) is represented by the closed triangles. The open circles represent the carrot cell fresh weight. All values are the averages of two determinations, with error bars showing standard errors. The absence of error bars indicates that the range is smaller than the size of the symbol. At Day 14 the cells were inoculated into fresh media as described under Materials and Methods.
CALMODULIN
HOMEOSTASIS
IN CARROT
ZO 0
12
5
after reinoculation. Overall, the data demonstrate that calmodulin is highly unmethylated at position 115 in the early exponential stage of carrot cell growth and that the level of endogenous calmodulin methylation increases as growth proceeds and cells enter the stationary phase. Further, the process seems to be subject to rapid regulation since unmethylated calmodulins reappear quickly upon reinoculation into fresh media.
10
15
20
25
days
34
FIG. 2. Fluorographic analysis of 3H-methylated wild carrot calmodulin. Carrot calmodulin samples were subjected to methylation analysis by using the calmodulin methyltransferase and [nethyl-3H]AdoMet as described under Materials and Methods. The samples (0.15 pg of each) were resolved by SDS-PAGE on a 15% polyacrylamide gel and were analyzed fluorographically. Samples are from 2-day (lane l), 4-day (lane 2), &day (lane 3), and ll-day (lane 4) cultures.
31
CULTURES
FIG. 3. Analysis of the levels of calmodulin protein and the methylation state of calmodulin during domestic carrot somatic embryogenesis. The calmodulin protein concentration (pmol CaM/mg protein) is represented by the open circles and the calmodulin methylation state (mol methyl-3H-groups/mol CaM) is represented by the closed circles. Values are the averages of two determinations, with error bars showing the standard errors. The absence of error bars indicates that the range is smaller than the size of the symbol.
tected in embryo extracts (Fig. 4). Since no bands were observed on the autoradiogram of a duplicate gel exposed to ‘251-calmodulin in the presence of EDTA or EGTA (data not shown), the bands appeared to represent calcium-dependent calmodulin-binding proteins. Although several calmodulin-binding proteins were detected in undifferentiated cells as well as in developing and germinated embryos, some proteins appeared to vary during embryo-
Analysis of Calmodulin and Calmodulin-Binding Proteins during Carrot Embryogenesis In order to further investigate the calmodulin system during plant development, we have utilized carrot embryogenesis as a model system. Domestic carrot cell cultures transferred to media devoid of 2,4-D (O-day sample) developed into embryos within 17 days. By the end of the 24th day the embryos germinated. Carrot samples from the O-day, lo-day, 17-day, and 24-day stages were harvested and analyzed using the same protocols as described above. Significant differences in the levels of calmodulin protein were not observed among the embryo samples from O-day, lo-day, and 17-day cultures. However, after passing the 17-day point, the level of calmodulin began to increase and doubled by the end of the 24th day (Fig. 3). Unlike the results obtained with proliferating vs nonproliferating wild carrot cells, the calmodulin methylation levels showed less dramatic changes during embryogenesis, and methyl-3H-group incorporation ranged from 0.3 to 0.2 mol/mol calmodulin (Fig. 3). Calmodulin overlay analysis was used to identify potential target proteins of calmodulin action and to determine whether differences in these calmodulin-binding proteins occur during the formation and germination of embryos. Several calmodulin-binding proteins were de-
MW
X
10-3
946743-
30-
2012341234 FIG. 4. 1251-calmodulin overlay analysis of extracts of cultured domestic carrot cells undergoing embryogenesis. Samples representing Oday (lane l), lo-day (lane 2), 17.day (lane 3), and 24.day (lane 4) carrot samples (16.7 pg of protein in each lane) were subjected to SDS-PAGE on a 12.5% polyacrylamide gel. (A) Coomassie blue-stained gel. (B) Autoradiogram of the gel exposed to ‘251-calmodulin in the presence of 1 mM
Cd&.
32
OH ET AL.
genesis and germination. For example, a 26,000 M, calmodulin-binding protein appeared to be preferentially associated with developing embryos and was present at a very low level in undifferentiated cells. In addition, calmodulin binding to a 54,000 M, protein increased during embryo germination and this protein was the major calmodulin-binding protein detected in extracts of germinated carrot embryos. This protein is similar in properties to a major calmodulin binding protein described in carrot protoplasts (37). As shown in Fig. 4, the increase in the 54,000 M, protein in carrot embryos occurred between 17 and 24 days after the initiation of embryogenesis and paralleled the increase observed in calmodulin levels in the germinated embryos (Fig. 3). In an effort to purify the 54,000 M, protein, we used calmodulin-Sepharose chromatography. The extract of 24-day-old carrot embryos was first chromatographed on DEAE-cellulose to separate the endogenous carrot calmodulin from the 54,000 M, calmodulin-binding protein. The resulting sample was adsorbed to calmodulin-sepharose in the presence of calcium and was eluted with EGTA. As shown in Fig. 5, the only polypeptide detected in the calmodulin-Sepharose eluate by calmodulin overlay or by protein staining was the 54,000 M, polypeptide. This result verifies that the native, undenatured 54,000 M, protein binds to calmodulin and, together with the overlay data, provides strong evidence that this protein is a bona fide calmodulin-binding protein. DISCUSSION In the present study we have analyzed calmodulin homeostasis during plant cell proliferation and embryo formation by using carrot culture model systems. During the growth of nonembryogenic wild carrot cultures, large differences in the levels of post-translational methylation of calmodulin were seen depending upon the stage of cell growth, with the highest levels of unmethylated calmodulin associated with the early exponential phase. We previously have shown that transient pools of undermethylated calmodulin exist in certain plant tissues, such as apical and young lateral root segments, that contain meristematic regions (5, 12). The present findings support this previous work and suggest that undermethylated calmodulin is preferentially associated with proliferating cell populations. The changes in methylation were not paralleled by changes in calmodulin levels, which did not appear to vary significantly during the growth of the carrot cells. Similarly, the calmodulin levels in developing domestic carrot embryos remained constant. However, the calmodulin protein levels doubled upon embryo germination. This increase was accompanied by an increase in a 54,000 M, calmodulin-binding protein which is the major apparent calmodulin-receptor protein detected in the germinated carrot embryos.
96 67-
30-
20-
I
2
1
2
FIG. 5. SDS-PAGE and ‘251-calmodulin overlay analysis of purified calmodulin-binding proteins from 24-day-old domestic carrot somatic embryos. Proteins were resolved by SDS-PAGE on 12.5% polyacrylamide gels. (A) Coomassie blue-stained gel. (B) Autoradiogram of the overlay gel exposed to ‘251-calmodulin in the presence of 1 mM CaCl,. Lane 1, crude extract; lane 2, 54,000 M, calmodulin-binding protein purified by calmodulin-Sepharose.
The data suggest that the levels of unmethylated calmodulin are regulated independent of calmodulin protein concentrations. For example, upon inoculation of postexponential phase cells into fresh media, unmethylated calmodulin quickly reaccumulates while there is little change in overall calmodulin protein levels. This reaccumulation is striking since it happens within 12 h of reinoculation, suggesting that a substantial turnover of methylated calmodulin occurs during this period. Although, it is currently unknown what factors control calmodulin methylation in plants, changes in the levels of the calmodulin methyltransferase present in the cells could be involved. For example, based on previous observations with animal systems, it has been shown that the levels of calmodulin methyltransferase enzyme can vary depending on the source of the tissue (reviewed in (26)). Little is currently known regarding the plant calmodulin methyltransferase. Further investigations of the properties of the plant enzyme are warranted to determine whether it is regulated during plant cell growth. In addition to the potential regulation of methyltransferase levels, calmodulin methylation could also be influenced by calcium binding to calmodulin. For example, Roberts et al. (12) observed that calmodulin methylation is stimulated by micromolar calcium concentrations, and it has been suggested that the calcium-bound form of calmodulin may be preferred by the calmodulin methyltransferase. Thus, calcium fluxes during cell growth could influence the rate of calmodulin methylation. The degree
CALMODULIN
HOMEOSTASIS
of calmodulin methylation also could be controlled by the availability of the methyl donor, AdoMet. Recently, Minocha et al. (27) observed that the specific activity of AdoMet synthetase in the extract of the &day-old carrot cells is more than 25-fold higher than that in the extract of the l-day-old carrot cells. Thus, the increase in AdoMet synthetase levels during carrot growth appears to parallel the increase in methylated calmodulin seen in our studies. The methylation of calmodulin has a significant effect on its ability to activate plant NAD kinases, with methylated calmodulins exhibiting a lower level of activation compared to unmethylated calmodulins (6, 12). In addition, there is evidence that lysine-115 but not trimethyllysine-115 is a site for ubiquitination of calmodulin by animal cell extracts under in vitro conditions, and it has been suggested that methylation may protect calmodulin from intracellular proteolytic turnover (38). It is tempting to suggest that variations in calmodulin methylation may be a mechanism for attenuating its enzyme activator properties or could influence cellular levels of calmodulin by affecting its susceptibility to proteolytic turnover. However, there is not yet any support for these roles for calmodulin methylation in uiuo. Further, as discussed below, there is no clear trend in the fluctuation of calmodulin levels during carrot cell growth that may reflect a change in calmodulin turnover. However, future work with the carrot culture system including the investigation of nicotinamide coenzyme homeostasis and calmodulin turnover during cell growth may provide insight into these potential roles of post-translational methylation of lysine115 of calmodulin. Our findings that the overall levels of calmodulin are not enhanced in exponential and postexponential wild carrot culture cells contrast with those of other studies with transformed animal cell models that show higher levels of calmodulin in rapidly proliferating cells compared to those in quiescent cells (13-15). Although the exact role of this increase in calmodulin during cell growth has not yet been established, it has been proposed that a transient increase in calmodulin during the G,/S boundary of the cell cycle can control the rate of cell cycle progression (28-30) by influencing the duration of the G, phase (29, 30). Since the carrot cultures used in the present study were not synchronized, it is not yet clear whether transient increases in calmodulin occurred at specific cell cycle stages that may have evaded detection. In plant tissues, it also has been proposed that calmodulin levels may be preferentially enhanced in dividing cells, since meristematic regions appear to have higher calmodulin protein levels compared to mature tissues based on immunocytochemical(16,17), enzyme activator (31), and RIA (32) analyses. This trend has also been found with calmodulin mRNA levels (33). However, there are fundamental differences among these systems of study. In the present study calmodulin homeostasis in
IN CARROT
CULTURES
33
dividing and nondividing undifferentiated cells in culture was investigated. In contrast, studies with apical and mature root and shoot tissues differ since heterogeneous populations of cells that are in the process of differentiating into specialized tissues and cell types are being compared, and thus these are not simple comparisons of proliferating and nonproliferating cells. In contrast to our findings with undifferentiated carrot cells, we observed that the germination of carrot embryos generated in culture resulted in a twofold increase in the concentration of calmodulin protein. Similar increases in the levels of calmodulin have also been observed during the germination of certain seed embryos, such as radish (34,35) and chick pea embryos (36). Currently, the question of the relationship between enhanced calmodulin levels and embryo germination is unclear, but the elevation of calmodulin levels could play a role in calciumsignal transduction processes during germination. In this regard it is of interest that the apparent levels of a major 54,000 M, calmodulin-binding protein also increase substantially during the germination of carrot embryos. Future studies, including the identification and characterization of this major calmodulin-binding protein, may provide insight into the roles of calmodulin in calciumsignal transduction in the germinating embryo. REFERENCES 1. Cohen, P., and Klee, C. B. (1988) Molecular Aspects of Cellular Regulation: Calmodulin, Elsevier, Amsterdam. 2. Roberts, D. M., and Harmon, A. C. (1992) Annu. Reu. Plant Physiol. Mol. Biol. 43, 375-414. 3. Rowe, P. M., Wright, L. S., and Siegel, F. L. (1986) J. Biol. Chem. 261,7060-7069. 4. Morino, H., Kawamoto, T., Miyake, M., and Kakimoto, Y. (1987) J. Neurochem. 48, 1201-1208. 5. Oh, S.-H., and Roberts, D. M. (1990) Plant Physiol. 93,880-887. 6. Roberts, D. M., Rowe, P. M., Siegel, F. L., Lukas, T. J., and Watterson, D. M. (1986) J. Biol. C&m. 261, 1491-1494. 7. Roberts, D. M., Burgess, W. H., and Watterson, D. M. (1984) Plant Physiol. 75, 796-798. a. Marshak, D. R., Clarke, M., Roberts, D. M., and Watterson, D. M. (1984) Biochemistry 23,2891-2899. G., Chi9. Roberts, D. M., Crea, R., Malecha, M., Alvarado-Urbina, arello, R. H., and Watterson, D. M. (1985) Biochemistry 24,50905098. 10. Putkey, J. A., Slaughter, G. R., and Means, A. R. (1985) J. Biol. Chem. 260,4704-4712. 11. Putkey, J. A., Draetta, G. F., Slaughter, G. R., Klee, C. B., Cohen, P., Stull, J. T., and Means, A. R. (1986) J. Biol. Chem. 261,98969903. 12. Roberts, D. M., Oh, S.-H., Besl, L., Weaver, C. D., and Stacey, G. (1990) in Current Topics in Plant Biochemistry and Physiology (Randall, D. D., and Blevins, D. G., Eds.), Vol. 9, pp. 67-84, Univ. of Missouri Press, Columbia, MO. 13. Watterson, D. M., Van Eldik, L. J., Smith, R. E., and Vanaman, T. C. (1976) Proc. Natl. Acad. Sci. USA 73, 2711-2715. 14. Chafouleas, J. G., Bolton, W. E., Boyd, A. E., and Means, A. R. (1981) Proc. Natl. Acad. Sci. USA 78, 996-1000.
34
OH ET AL.
15. Zendegui, J. G., Zielinski, R. E., Watterson, L. J. (1984) Mol. Cell. Biol. 4, 883-889.
D. M., and Van Eldik,
16. Lin, C.-T., Sun, D., Song, G.-X., and Wu, J.-Y. (1986) J. Ifistochem. Cytochem. 34,561-567. 17. Dauwalder, 461-470.
M., Roux, S. J., and Hardison,
18. Zimmerman, J. L., Apuya, N., Darwish, Plant Cell 1,1137-1146. 19. Dougall, D. K., and Weyrauch,
L. (1986) Plunta
K., and O’Carroll,
C. (1989)
K. W. (1980) In Vitro 16,969-975.
20. Wetherell,
D. F. (1969) Plant Physiol. 44,1734-1737.
21. Bradford,
M. M. (1976) And. Biochem. 72, 248-254.
22. Gopalakrishna, R., and Anderson, Res. Commun. 104,830-836. 23. Laemmli,
168,
W. B. (1982) Biochem. Biophys.
U. K. (1970) Nature 227, 680-685.
24. Burgess, W. H., Watterson, Cell Biol. 99, 550-557.
D. M., and Van Eldik, L. J. (1984) J.
25. Dougall, D. K., Johnson, J. M., and Whitten,
149,292-297.
G. H. (1980) Planta
26. Siegel, F. L., Vincent, P. L., Neal, T. L., Wright, L. S., Heth, A. A., and Rowe, P. M. (1990) in Protein Methylation (Paik, W. K., and Kim, S., Eds.), pp. 33-58, CRC Press, Boca Raton, FL. 27. Minocha, S. C., Minocha, R., and Komamine, A. (1991) Plant Physiol. Biochem. 29, 231-237. 28. Chafouleas, J. G., Bolton, W. E., Boyd, A. E., and Means, A. R. (1982) Cell 28, 41-50. 29. Rasmussen, C. D., and Means, A. R. (1987) EMBO J. 6,3961-3968. 30. Rasmussen, C. D., and Means, A. R. (1989) EMBO J. 8, 73-82. 31. Allan, E. F., and Trewavas, A. J. (1985) Planta 165,493-501. 32. Muto, S., and Miyachi, S. (1984) 2. Pflanzenphysiol. Bd. 114,421431. 33. Zielinski, R. E. (1987) Plant Physiol. 84,937-943. 34. Cocucci, M. (1984) Plant Cell Enuiron. 7,215-221. 35. Cocucci, M., and Negrini, N. (1988) Plant Physiol. 88, 910-914. J., Rodriguez, D., Nicolas, G., and Aldasoro, 36. Hernandez-Nistal, J. J. (1989) Physiol. Plant. 75, 255-260. 37. Cyr, R. J. (1991) J. Cell Sci. 100, 311-317. 38. Gregori, L., Marriott, D., Putkey, J. A., Means, A. R., and Chau, V. (1987) J. Biol. Chem. 262, 2562-2567.
