Eur. J. Biochem. 255, 4922500 (1998)
 FEBS 1998
The ubiquityl-calmodulin synthetase system from rabbit reticulocytes:
isolation of the calmodulin-binding second component and enzymatic properties
Matthias MAJETSCHAK, Markus LAUB, Christoph KLOCKE, Johannes A. STEPPUHN and Herbert P. JENNISSEN
Institut für Physiologische Chemie, AG Biochemische Endokrinologie, Universität-GHS-Essen, Germany
(Received 13 October/31 December 1997) 2 EJB 97 1462/4
Ubiquitin2calmodulin ligase (uCaM synthetase; EC 6.3.2.21), which has been detected in all tissues
so far examined, catalyzes the Ca 21-dependent reversible synthesis of ubiquityl-calmodulin which is not
directed to degradation by the ATP-dependent 26-S protease [Laub, M. & Jennissen, H. P. (1997) Biochim. Biophys. Acta 1357, 1732191]. As has been shown in the preceding paper in this journal, the
uCaM synthetase holosystem can be separated into two essential protein components : uCaM Syn-F1, a
ubiquitin-binding protein belonging to the ubiquitin-activating enzyme family (E1) and uCaM Syn-F2
which bestows the reaction specificity leading to the covalent modification of calmodulin with ubiquitin.
UCaM Syn-F2, which binds to calmodulin2Sepharose in a Ca 21-dependent manner, has been purified
over 3500-fold in seven steps from rabbit reticulocytes and has a native molecular mass of <620 kDa. It
binds calmodulin with a Km of 5 µM and to uCaM Syn-F1, i.e. ubiquitin-activating enzyme (E1), with a
Km of 3 nM. The maximal specific activity obtained in enriched uCaM Syn-F2 is 628 pkat/mg. The pH
optimum of uCaM synthetase lies at pH 8.5. In kinetic experiments the Km values for 125I-ubiquitin and
ATP/Mg21 were determined to be 8 µM and 16 nM, respectively, for the uCaM synthetase holosystem.
The existence of a third separable protein component of uCaM synthetase, as is the case in E1, E2, E3
systems, is very unlikely since affinity chromatography on calmodulin2Sepharose, two ion-exchange
chromatography steps and finally a gel-filtration step failed to indicate any additional protein component
essential for synthetase activity. We therefore propose a two-component model for uCaM synthetase. This
model is also supported by simple hyperbolic velocity curves in kinetic experiments based on the variation
of these two components. The data suggests that uCaM Syn-F2 is neither an E2 nor an E3 but evidently
combines the properties of both, making the Ca21-dependent uCaM synthetase the member of a group of
two-component ubiquitin ligase systems.
Keywords : calmodulin; ubiquitin; ubiquityl2calmodulin synthetase ; non-catabolic protein ubiquitylation;
calcium.
Ubiquitin is a 76-amino-acid 8.5-kDa protein which, similarly to calmodulin, plays an essential role in eukaryotic cells.
Ubiquitin exerts its many functions through covalent conjugation to itself or other proteins (for review see [1]). Two types
of protein ubiquitylation can be distinguished on the basis of
metabolism [1]: (a) catabolic protein ubiquitylation, which involves the ATP-ubiquitin-dependent proteolytic pathway and is
important for the degradation of abnormal, misfolded and some
short-lived proteins (e.g. cyclins, p53); (b) non-catabolic protein
ubiquitylation where other functions of the thus modified proteins (e.g. histone 2A, calmodulin) are primarily involved withCorrespondence to H. P. Jennissen, Institut für Physiologische
Chemie, AG Biochemische Endokrinologie Universität-GHS-Essen,
Hufelandstr. 55, D-45122 Essen, Germany
Fax: 149 201 723 5944.
Abbreviations. CaM, calmodulin ; 125I-CT-ubiquitin, ubiquitin labelled with 125I by reaction with chloramine T; 125I-BH-ubiquitin, ubiquitin labelled with 125I by the Bolton-Hunter reagent ; 125I-IC-ubiquitin,
ubiquitin labelled with 125I by the iodomonochloride method ; APF II,
fraction of reticulocyte lysate eluted by KCl from DEAE ion exchanger ;
uCaM, ubiquityl-calmodulin; uCaM Syn-F1, uCaM synthetase protein
factor 1; uCaM Syn-F2, uCaM synthetase protein factor 2.
Enzymes. ATP-dependent 26-S protease (26-S proteasome (EC
3.4.99.2); ubiquitin2calmodulin ligase, ubiquityl2calmodulin synthetase, (EC 6.3.2.21) ; ubiquitin2protein ligase E1, E2, E3 (EC 6.3.2.19).
out terminating in degradation by the 26-S protease. Large-scale
catabolic protein ubiquitylation appears to be restricted to a few
specialized cells, e.g. reticulocytes, since no significant activity
of the ATP-ubiquitin-dependent proteolytic pathway could be
detected in any other rabbit tissues [2]. In contrast, non-catabolic
protein ubiquitylation, e.g. of calmodulin, is conspicuously present in all of these tissues [224].
The ubiquitylation of vertebrate calmodulin was first reported in 1987/88 [5, 6] and found to be catalyzed by ubiquitin2
calmodulin ligase (uCaM synthetase) [729] in a Ca21-dependent
manner at Lys21 of calmodulin [10]. The following equation
describes the reaction catalyzed by uCaM synthetase (n is an
integer) [9] :
Ca21
calmodulin 1 n ubiquitin 1 n ATP Y==y
(ubiquitin)n-calmodulin 1 n AMP 1 n PPi .
(1)
In the preceding paper [11], uCaM synthetase from rabbit
reticulocytes was separated into two components called uCaM
Syn-F1 and uCaM Syn-F2. The first component, uCaM SynF1, was purified 3500-fold to homogeneity. This protein, which
contained no other enzymes of ubiquitin metabolism, was shown
to belong to the ubiquitin-activating enzyme (E1) family and to
be 56 amino acids shorter than the rabbit mRNA transcript.
Majetschak et al. (Eur. J. Biochem. 255)
In this paper, we describe the purification and characterization of the second component uCaM Syn-F2 and elucidate the
interplay of the two components in generating calmodulin-conjugating activity. UCaM Syn-F2 could be purified <3500-fold
to a stage where other components of the ubiquitin system, e.g.
ubiquitin-activating enzyme and ubiquitin-conjugating enzymes,
had been removed. Strong evidence is presented that the uCaM
synthetase system only consists of two separable components.
MATERIALS AND METHODS
Materials. For the materials employed see the preceding paper [11] and [2].
Preparative methods. The preparation of rabbit reticulocytes, ATP-depleted reticulocyte lysate and the enrichment either
by ion-exchange chromatography on DEAE-Sephacel to Sephacel-APFII or on Fractogel EMD-DEAE 650 S under sample displacement conditions to Fractogel-APFII were all performed as
described in the preceding paper [11].
Ubiquitin-activating enzyme (uCaM Syn-F1). uCaM Syn-F1
and uCaM Syn-F2 were separated by affinity chromatography
of APF II on ubiquitin2Sepharose [11]. The uCaM Syn-F1 desorbed from the column at alkaline pH 9.0 by 10 mM dithioerythritol [11] is <800-fold purified (i.e. affinity purified uCaM
Syn-F1). UCaM Syn-F1 was purified <3500-fold to homogeneity (homogenous uCaM Syn-F1) and corresponds to ubiquitin-activating enzyme (E1) in the structural form described [11].
Ubiquitin. This was prepared as described [12] or purchased
from Sigma [11]. 125I-labeled ubiquitin was prepared according
to several methods. 125I-CT-ubiquitin was synthesized according
to the chloramine T procedure (0.1283108 cpm/mg) [12, 13].
For reduction of the specific radioactivity 125I-CT-ubiquitin was
diluted with native ubiquitin [13] in classical activity measurements [8], whereas for Km determinations 125I-CT-ubiquitin was
diluted with non-radioactive CT-ubiquitin. 125I-BH-ubiquitin
(42383106 cpm/mg) was synthesized according to the BoltonHunter method [14, 15]. 125I-IC-ubiquitin (172933106 cpm/mg)
was synthesized according to the iodomonochloride method [16,
17]. In the latter, the incubation mixture (total volume 140 µl)
contained 0.8 M glycine, 0.8 M NaCl, pH 9.0, 3.7 mg/ml ubiquitin, 36 µl Na125I (100 mCi/ml, 15.8 mCi 125I/µg iodine), and a
twofold molar excess of iodomonochloride. Incubation was
stopped after 15 min at room temperature by lowering the pH to
7.0. Free Na125I and iodomonchloride were separated by dialyzing (versus 1000-fold buffer volume) the reaction mixture for
four 1-h periods against a total of four buffer changes of 20 mM
Tris/HCl, 1 mM dithioerythritol, pH 7.5. Typical yields for the
incorporation of 125I were 70280%. The specific radioactivity
of 125I-BH-ubiquitin and 125I-IC-ubiquitin was always diluted
with their respective non-radioactive ubiquitin derivatives.
Calmodulin. Calmodulin was prepared from bovine testis
and tested for biological activity as described [18]. Calmodulin
was radioactively labeled either by the chloramine T procedure
(125I-CT-calmodulin) [19] or the Bolton-Hunter method (125I-BHcalmodulin) [14, 15, 20]. For reduction of the specific radioactivity, labeled calmodulin was diluted with the corresponding
non-radioactive derivative.
Phosphorylase b and phosphorylase kinase were purified according to [21] and [18] respectively.
Coupled Sepharose. Fluphenazine-Sepharose was prepared
from Fluphenazine and Sepharose 6B (Pharmacia) as described
in [18] and employed for the purification of calmodulin and for
assaying uCaM synthetase.
Ubiquitin-Sepharose was prepared by coupling Sepharose
4B by the divinylsulfone method [22, 23] with the modifications
493
given in the preceding paper [11]. Gels of low and high degrees
of substitution (627 mg/ml and 17218 mg/ml packed gel, respectively) were prepared. The gels were regenerated after use
and stored as described in the preceding paper [11].
Calmodulin-Sepharose 4B, with 425 mg CaM immobilized/
ml packed gel, was synthesized according to [18]. Washing and
regeneration of the affinity gel are described in [18].
Chromatographic methods. Gel filtration of Fractogel-APF
II and uCaM Syn-F2 was performed on a column of Superose
6 pg (0.4 internal diameter326.6 cm, Pharmacia) in 20 mM Tris/
HCl, 1.0 mM dithioerythritol, 150 mM NaCl, 5 µg/ml leupeptin,
pH 7.2 (buffer A) at 5 °C. The column had been previously calibrated using proteins of known molecular mass: phosphorylase
kinase 1.2 MDa, thyroglobulin 668 kDa, ferritin 446 kDa, catalase 232 kDa, phosphorylase b 192 kDa, aldolase 158 kDa, IgG
160 kDa, BSA 67 kDa, ovalbumin 45 kDa, trypsin inhibitor
20.1 kDa, myoglobin 17 kDa, ubiquitin 8.5 kDa. Molecular
masses of the enzyme fractions were calculated by non-linear
regression based on the respective calibration function.
Affinity chromatography of reticulocyte Fractogel-APF II on
ubiquitin2Sepharose was performed as described in the preceding paper [11] and the fractions of uCaM Syn-F1 and uCaM
Syn-F2 obtained were stored at 280°C. UCaM Syn-F2 was
purified by affinity chromatography on calmodulin2Sepharose
[18] (425 mg/ml packed gel, column : 2.7 cm internal diameter
33.5 cm, flow rate 35 ml/h, fraction volume 2.524.0 ml). The
column was equilibrated with 20 column volumes 50 mM sodium sn-glycerol 2-phosphate, 10% sucrose, 2 mM EDTA,
5 mM Mg21, 10 mM 2-mercaptoethanol, 2 mM Ca21, 5 µg/ml
leupeptin, pH 7.0 (buffer B). 60285 mg uCaM Syn-F2 from the
ubiquitin2Sepharose step in 20 mM Tris/HCl, 1 mM dithioerythritol, 5 µg/ml leupeptin, pH 7.6 (buffer C) containing 2 mM
Ca21 was applied to the column. After washing the calmodulin2
Sepharose with several volumes buffer D, the column was eluted
with 50 mM sodium sn-glycerol 2-phosphate, 10% sucrose,
2 mM EDTA, 5 mM Mg21, 10 mM 2-mercaptoethanol, 5 mM
EGTA, 5 µg/ml leupeptin, pH 7.0 (buffer D). Ca21 was added
to the eluate in a final concentration of 5 mM followed by a
concentration and dialysis on Centricon 30 tubes against buffer
C and stored frozen at 280°C.
Affinity purified uCaM Syn-F2 was further purified by anion-exchange chromatography (UV Detector L-4000, ChromatoIntegrator D-2500, Intelligent Pump L-6210, Merck-Hitachi) on
a quaternary aminomethyl MonoQ HR 5/5 column [exchanger
group: 2CH2N1(CH3)3 ; 0.5 cm internal diameter35 cm gel
height; 30 ml/h, 1.522.2 Mpa, fraction volume 0.521.5 ml) in
buffer C at 124°C. 728 ml of sample (122 mg/ml) was applied
in 10 ml Superloop (Pharmacia-LKB) to the column ; column
and Superloop were placed on ice during the run. The column
was eluted with increasing concentrations (021 M) of KCl in
50 mM Tris/HCl, 0.5 mM dithioerythritol, 5 µg/ml leupeptin,
pH 7.2 (buffer E). The salt gradients were checked by conductance measurements (see below) in the eluted fractions.
Analytical methods. Ubiquityl-calmodulin synthetase.
UCaM synthetase activity measurements were based on the Fluphenazine-Sepharose affinity test [8, 24] as described in the preceding paper [11]. When the activity of the single components
uCaM Syn-F1 and uCaM Syn-F2 was determined, the system
was reconstituted with the respective missing component. In this
case mixtures with the single component served as control. The
Ca21-activity ratio is defined as 2Ca 21 activity/1Ca21 activity
[24].
Testing for labile (thiolester) ubiquitin conjugates and essential SH-groups. Synthesis of labile thiolester ubiquitin conjugates was based on the procedure of [25] and is described
extensively in the preceding paper [11]. For detection of thiol
494
Majetschak et al. (Eur. J. Biochem. 255)
Fig. 1. Affinity chromatography of uCaM Syn-F2 on calmodulin−Sepharose. A sample of 40 ml (run-through of ubiquitin2Sepharose) was applied to a column of calmodulin2Sepharose (degree of substitution 4 mg/ml packed gel, column internal diameter 2.7 cm33.5 cm
gel height, flow rate 35 ml/h, fraction volume 3.0 ml) in buffer B at 5°C.
The column was eluted with 5 mM EGTA in buffer D. The incubation
mixtures for activity assays contained 10 µg/ml uCaM Syn-F1 (E1). Arrow 1, application of sample ; arrow 2, start of wash with buffer B; arrow
EGTA, elution with buffer D. For further details see Materials and Methods and Results. (d) uCaM synthetase activity 1Ca21 ; (s) uCaM synthetase activity 2Ca 21 (3) A 280 nm.
labile conjugates (thiolesters), 25-µl aliquots of the incubation
mixtures were added to 40 µl 10% glycerol, 0.5% SDS, 0.001%
bromphenol blue, 65 mM Tris/HCl, 10 mM EGTA, pH 6.8 (thiol
ester sample buffer) in the absence of mercaptoethanol and immediately placed on ice for 20 min. For thiolysis 25-µl aliquots
were added to 40 µl 10% glycerol, 0.5 % SDS, 0.001% bromphenol blue, 65 mM Tris/HCl, 10 mM EGTA, 1.0 M 2-mercaptoethanol, pH 6.8 (thiolysis sample buffer) and heated to 100°C
for 10 min and then placed on ice for 5 min. If the conjugates
did not disappear in thiolysis buffer after 10 min they were
classed as stable. For further analysis of these conjugates see
SDS/PAGE below. UCaM Syn-F2 was tested for an essential
thiol group by incubation in 5.0 mM N-ethylmaleimide [12] for
10 min at 30°C followed by dilution into 10 mM dithioerythritol
buffer of the Fluphenazine-Sepharose affinity test (see above).
Protein was determined after precipitation with trichloroacetic acid (5%), washing and resolubilization of the pellet in
0.1 M NaOH, 1% SDS according to the method of [26] on an
AutoAnalyzer (Technicon) employing BSA as standard.
SDS/PAGE autoradiography. SDS/PAGE at 5°C and autoradiography of 10215% gels [27] was described in the preceding paper [11]. The incubation mixtures for the autoradiographic
analysis of uCaM on polyacrylamide gels were the same as those
for the Fluphenazine-Sepharose-affinity test. Unless otherwise
stated, the sample buffer for the Laemmli system [27] additionally contained 10 mM EGTA [8]. The molecular mass standards
for SDS/PAGE are given above. Electrophoresis of labile ubiquitin adducts (E1~125I-CT-ubiquitin or E2~125I-CT-ubiquitin) [25]
was performed on 10215% polyacrylamide gels as described in
the preceding paper [11].
Conductivity measurements. The KCl gradients were
checked by conductivity measurements on a CDM 80 conductivity meter (electrode CDC 364 ; Radiometer Copenhagen).
Data analysis and curve fitting. The data of the enzyme kinetics were analyzed and fitted (for further details see [18]) to a
rectangular hyperbola employing the PC program GraphPad
Prism for Windows (GraphPad Software Inc.) The program
yields the fitted constants as mean values with the standard errors of the mean (SEM). Analyses of analytical gel filtration and
SDS/PAGE are described in the accompanying article [11].
RESULTS
Purification of uCaM Syn-F2. For the purification of uCaM
Syn-F2, enriched reticulocyte Fractogel-APF II [11] was first
passed over ubiquitin2Sepharose for the selective adsorption of
Table 1. Purification of uCaM Syn-F2. uCaM synthetase activity was measured in each pool with the FP-Test. After the ubiquitin2Sepharose
step, uCaM Syn-F2 had to be reconstituted to uCaM synthetase by addition of affinity-purified uCaM Syn-F1 in saturating amounts, which was
generally achieved by adding 10 µg/ml. The amount of protein in the total volume of 50 µl was: lysate 320 µg, APF II 45 µg, CaM-Sepharose
eluate 15 µg, MonoQ I eluate 2 µg, MonoQ II eluate 0.25 µg. The gel filtration step is taken from Fig. 2 B and corresponds to a second separate
preparation of similar purity and yield up to the MonoQ II step. Only the purification factor (original value 4600-fold) has been normalized to the
data (steps 1 to 6-F2) of the first preparation. All other data in step 7-F2, including the total yield, correspond to the original values of the second
preparation. For further details see Materials and Methods, Results and Fig. 2 B. QAM, quaternary aminomethyl.
Step
Volume
ml
uCaM Syn-F2
concentration
specific
activity
total
activity
pkat/ml
pkat/mg
pkat
0.025
164
Purification
Yield
Protein
-fold
%
mg/ml
100
72
1. Reticulocyte lysate
91
1.8
2. Ion-exchange chromatography
on Fractogel EMD-DEAE
1.0
21
17.1
1.9
359
78
226
9
3. Passive affinity chromatography
on ubiquitin-Sepharose (run-through)
90
2.7
1.7
236
68
143
1.6
4-F2. Affinity chromatography
on CaM-Sepharose
8
10.5
5.3
84
214
52
2.0
5-F2. Ion-exchange I chromatography
on QAM-MonoQ
9
3.0
12.1
27
489
20
0.25
6-F2. Ion-exchange II chromatography
on QAM-MonoQ
1.5
3.8
74.9
5.6
3018
3.5
0.05
7-F2. Gel filtration on Superose 6 pg
0.9
3.8
87.4
3.4
3496
2.0
0.04
Majetschak et al. (Eur. J. Biochem. 255)
ubiquitin-activating (E1, uCaM Syn-F1) and ubiquitin-conjugating (E2) enzymes and the passive purification of uCaM Syn-F2.
The uCaM Syn-F2 thus obtained in the run-through is totally
inactive due to the complete removal of uCaM Syn-F1, i.e.
ubiquitin-activating enzyme. The activity of uCaM Syn-F2 was
measured in the reconstitution assay mixture with uCaM SynF1 (see preceding paper [11]).
In the next step uCaM Syn-F2 was affinity-adsorbed to its
immobilized substrate calmodulin (Fig. 1), a major aim being to
remove additional components of ubiquitin metabolism which
may have escaped adsorption to ubiquitin2Sepharose. UCaM
Syn-F2 is characteristically adsorbed in the presence of Ca 21
and eluted with EGTA (5 affinity purified uCaM Syn-F2) ; some
activity runs unadsorbed through the column. In averaging over
10 affinity preparations, it was found that a 5.5-fold purification
at a 30240% yield is typical for the calmodulin2Sepharose step
(see also [18]). Reconstitution experiments with the fractions of
the calmodulin2Sepharose affinity step provided no evidence
for any additional components. Total enrichment after affinity
chromatography on calmodulin2Sepharose was over 200-fold
(Table 1). Additional purification of uCaM Syn-F2 was achieved
by ion-exchange chromatography on a quaternary aminomethyl
MonoQ HR 5/5 column. In the first cycle (not shown) a single
symmetrical activity peak was eluted at a KCl concentration of
450 mM. In the rechromatography step on the same column
(Fig. 2A) the activity again eluted at 450 mM KCl in a single
symmetrical peak with a <3000-fold purification and a protein
yield of 80 µg being achieved. Analysis of this pooled peak by
SDS/PAGE (see insert to Fig. 2 A) led to five major protein
bands of 248 kDa, 145 kDa, 101 kDa, 73.5 kDa and 38.8 kDa.
In a second preparation of similar purity, uCaM Syn-F2 was
obtained for gel filtration (included as final step in Table 1) and
for determining the native molecular mass. As shown in Fig. 2 B,
uCaM Syn-F2 elutes from Superose 6 pg as a single symmetric
activity peak with a molecular mass of 6236 56 kDa, a specific
activity of 87.4 pkat/mg and a yield of <50% (see legend to
Fig. 2 B). It runs slightly behind the major protein peak, indicating that the preparation is not yet homogeneous. With the gelfiltration step an overall purification of <3500-fold with a yield
of <2% can be calculated for the purification of uCaM Syn-F2.
Since the seven-step purification scheme (Table 1) had failed
to indicate any other essential protein component, it was highly
probable that uCaM synthetase consisted essentially of only two
components. However, it could be that uCaM Syn-F2 displayed
properties similar to an E2 by forming thiol-labile ubiquitin conjugates. Therefore, uCaM synthetase was reconstituted with homogenous uCaM Syn-F1 (ubiquitin-activating enzyme, E1) and
the 3500-fold purified uCaM Syn-F2 followed by incubation
with 125I-CT-ubiquitin and ATP/Mg21 (Fig. 3). No formation of
thiol-labile ubiquitin conjugates with the exception of those expected for the pure ubiquitin-activating enzyme itself (lanes 1
and 5) could be detected. In MonoQ fractions of a lower purity
state (i.e. with a specific catalytic activity of <12220 pkat/mg)
usually a strong stable ubiquitylation band, which has not been
identified, around 2502270 kDa, can be observed (Fig. 3, lanes
4 and 8). Sometimes a faint thiol-labile band could also be observed (not shown). In the purer fractions of uCaM Syn-F2 with
higher specific catalytic activities of 70290 pkat/mg (Fig. 3,
lanes 3 and 7) these bands were lost.
Enzymological characteristics of uCaM synthetase. Influence
of protease inhibitors and pH. Prior to the enzymological characterization of uCaM synthetase, the possible action of proteases
under test conditions had to be excluded. The 502100-fold purified Fractogel-APF II pool was utilized which contained the
physiological mixture of the synthetase components. Leupeptin
495
Fig. 2. Purification of uCaM Syn-F2 by ion-exchange chromatography on MonoQ HR 5/5 (A) and molecular mass estimation of the
purified active component by gel-filtration chromatography on Superose 6 pg (B). (A) Rechromatography of uCaM Syn-F2 on MonoQ
HR 5/5 (0.5 cm internal diameter35 cm gel height; 30 ml/h, 1.52
2.2 MPa, fraction volume 0.5 ml). Protein was eluted by a (021 M) KCl
gradient. The active fractions of the first MonoQ run were pooled (8 ml,
0.25 mg/ml), dialyzed for 2 h against buffer C in a BSA-coated dialysis
membrane (Visking) at 5°C and applied to the column in a Superloop.
Fractions of 0.5 ml were collected and aliquots of 5 µl were tested for
activity in the Fluphenazine-Sepharose affinity test. The incubation mixtures (50 µl) containing 0.01 mg/ml affinity purified uCaM Syn-F1 were
incubated for 30 min at 37 °C. The first peak at fraction 40 corresponds
to residual BSA of the dialysis tubing. For further details see Materials
and Methods and Results. Insert: SDS/PAGE of 5 µg of the pooled activity peak (6-F2); K, contaminants in sample buffer. (d) Ca21 -dependent
activity; (– – –) A280 nm for estimation of protein concentration ; (· · · · ·)
KCl gradient. (B) Analytical gel filtration of uCaM Syn-F2. A sample of
200 µl uCaM Syn-F2 (1.3 mg/ml, total activity 7.1 pkat) from a separate
MonoQ-II step (stage 6-F2) was applied to a Superose 6 pg column
(0.4 cm internal diameter 326.6 cm gel height, flow rate 3 ml/h, fraction
volume 0.3 ml) at 5°C. Activity was measured in the FluphenazineSepharose-affinity test (40 min at 37°C) supplemented with 10 µg/ml
affinity purified uCaM Syn-F1. The total regained activity was 3.4 pkat.
See Table 1. (d) uCaM Syn-F2 activity: (–––) (A 280 nm ; (3) molecular
mass (m) standard proteins (phosphorylase kinase 1.3 MDa, thyroglobulin 668 kDa, ferritin 446 kDa, catalase 232 kDa) with non-linear regression fitted calibration line. The term relative A280 nm indicates that
the given values were not corrected for signal enhancement in the photometer.
and aprotinin show no effect on calmodulin ubiquitylation,
whereas the influence of phenylmethanesulfonyl fluoride and
trypsin inhibitor are within the error of the test, thus making an
effect of proteases on the synthetase test unlikely (Table 2). However inhibitors reacting with sulfhydryl groups such
as iodoacetamide and cis-platin strongly inhibit the system (Table 2). In this case a direct inhibition via the essential SH-group
of ubiquitin-activating enzyme (E1) is most likely. In addition,
uCaM Syn-F2 could also have been inhibited if it possessed
496
Majetschak et al. (Eur. J. Biochem. 255)
Fig. 3. Synthesis of labile ubiquitin conjugates in the purified fractions of uCaM Syn-F2. (A) For detection of thiol labile conjugates (thiolesters)
25-µl aliquots of the mixtures [containing 125I-CT-ubiquitin, ATP/Mg21 and pure uCaM Syn-F1 (stage 6-F1, pure E1) 10 µg/ml and uCaM Syn-F2
(gel filtration pool, stage 7-F2) 30 µg/ml incubated for 15 min at 37 °C as described in Materials and Methods] were added to 40 µl thiol ester
sample buffer (10% glycerol, 0.5 % SDS, 0.001% bromphenol blue, 65 mM Tris/HCl, 10 mM EGTA, pH 6.8) in the absence of mercaptoethanol
and immediately placed on ice for 20 min. (B) For thiolysis 25-µl aliquots of the same composition as described above were added to 40 µl thiolysis
sample buffer (10 % glycerol, 0.5 % SDS, 0.001% bromphenol blue, 65 mM Tris/HCl, 10 mM EGTA, 1.0 M mercaptoethanol, pH 6.8) and heated
to 100°C for 30 min. The samples which contained <0.320.8 µg protein were analyzed by SDS/PAGE according to [27] on 12.5% polyacrylamide
gels followed by autoradiography. For further details see Materials and Methods and Results. Lanes 124, no mercaptoethanol; lanes 528,
1 merpactoethanol; lanes 1 and 5, pure uCaM Syn-F1 (E1) ; lanes 2 and 6, pure uCaM Syn-F1 (E1) 1 uCaM Syn-F2 (gel-filtration pool) ; lanes 3
and 7, uCaM Syn-F2 (gel-filtration pool); lanes 4 and 8, pure uCaM Syn-F1 (E1) 1 uCaM Syn-F2 MonoQ II. For further details see Table 1,
Fig. 2, Materials and Methods, Results and the previous paper [11].
Table 2. Influence of inhibitors on the ubiquitylation of calmodulin.
The incorporation of ubiquitin into calmodulin (CaM) was measured
with the Fluphenazine test (see Materials and Methods). The incubation
mixture contained in a final volume of 50 µl : 50 mM Tris/HCl, 1 mM
dithioerythritol, 5 mM MgCl2, 1 mM ATP, 10 mM phosphocreatine,
0.1 mM CaCl2, 0.1 mg/ml creatine kinase, 0.9 mg/ml reticulocyte APF
II and the indicated amounts of protease inhibitors. After 15 min of incubation 0.5 mg/ml calmodulin and 1 mg/ml 125I-CT-ubiquitin were added
and the incubation was continued for 30 min. Controls were run without
ATP. Incubation was stopped by heating to 96 °C for 4 min.
Inhibitor
uCaM synthetase
activity
Control
Aprotinin (0.1 mg/ml)
Leupeptin (0.1 mg/ml)
Phenylmethanesulfonyl fluoride (1 mM)
Trypsin inhibitor (0.1 mg/ml)
Iodoacetamide (5 mM)
cis-Platin (800 µM)
Mn21 (10 mM)
pkat/mg
%
1.5
1.5
1.5
1.3
1.9
0
0.16
0.12
100
100
100
86
127
0
11
8
an essential SH-group. Therefore, purified uCaM Syn-F2 was
incubated for 10 min with 5 mM N-ethylmaleimide in a separate
experiment at 30°C (not shown) before quenching the latter with
dithioerythritol. After reconstituting the test system with purified
ubiquitin activating enzyme, over 95% of the uCaM synthetase
activity disappeared, strongly indicating the presence of an
essential SH-group not only in the first component of the synthetase system but also in the second component, uCaM Syn-F2.
Mn21 also leads to an inhibition of enzyme activity probably by
competitive displacement of Mg21 from ATP.
The pH dependence of uCaM synthetase activity in APF II
is shown in Fig. 4. The pH optimum was reached in the far-
Fig. 4. pH profile of the uCaM synthetase system in APF II. The
uCaM synthetase activity (n) was measured with the FluphenazineSepharose affinity test (see Materials and Methods). The incubation mixture contained in a final volume of 50 µl : 60 mM Tris/HCl, 60 mM snglycerol 2-phosphate of the respective pH, 1 mM dithioerythritol, 5 mM
MgCl2 , 1 mM ATP, 0.1 mM CaCl2, 0.9 mg/ml Fractogel APF II and
1.0 mg/ml 125I-CT-ubiquitin (9300 cpm/µg). Before each measurement
the final pH was checked.
alkaline region at pH 8.5. A significant activity (over 50%) was
even found at pH 9.5 with zero activity being reached over
pH 10.5. The skewed profile of the curve indicates a complex
pH dependence of the uCaM synthetase system.
Ca 2+ dependence and two-component interplay. In a first
series of experiments the quantitative relationship between the
two components generating calmodulin-conjugating activity was
determined (Fig. 5) employing the affinity-enriched uCaM SynF1 and uCaM Syn-F2 components, which contain Ca21-dependent and independent conjugating activities. The concentrations
of ubiquitin and calmodulin were held at saturating levels. First
the Ca21-dependent and Ca21-independent uCaM synthetase
activities were measured by saturating a constant amount of en-
Majetschak et al. (Eur. J. Biochem. 255)
Fig. 5. Reciprocal activation of uCaM Syn-F1 and uCaM Syn-F2.
(A) Generation of uCaM-Syn-F2-based synthetase activity by adding
increasing concentrations of affinity-purified uCaM Syn-F1 (E1) to a
constant concentration (0.12 mg/ml) of affinity-purified uCaM Syn-F2 in
the incubation mixture (50 µl). The mixtures containing 125I-CT-ubiquitin
(25 cpm/pmol) were incubated for 30 min at 37 °C. The data were fitted
non-linearly to a rectangular hyperbola (1Ca21, r2 5 0.98 ; 2Ca21, r2 5
0.86). Insert (m): generation of uCaM-Syn-F2-based synthetase activity
by adding increasing amounts of homogenous uCaM Syn-F1 (ubiquitinactivating enzyme) to highly purified uCaM Syn-F2 (MonoQ II fraction)
in the presence of Ca21. (d) Activity 1Ca21 ; (s) activity 2Ca21. (B)
Generation of uCaM-Syn-F1-based synthetase activity by adding
increasing amounts of affinity-purified uCaM Syn-F2 to a constant concentration (10 µg/ml) of affinity-purified uCaM Syn-F1 (E1) in the incubation mixture (50 µl). The mixtures containing 125I-CT-ubiquitin
(25 cpm/pmol) were incubated for 60 min at 37 °C. The data were fitted
non-linearly to a rectangular hyperbola (1Ca21 , r2 5 0.99 ; 2Ca21, r2 5
0.97). Insert (m): decrease of Ca21-activity ratio as a function of increasing uCaM Syn-F2 in the mixture. (d) Activity 1Ca21 ; (s) activity
2Ca21.
riched uCaM Syn-F2 with uCaM Syn-F1 (Fig. 5A). The kinetic
curve of uCaM Syn-F2 activation follows a classical rectangular
hyperbola. Half-saturation of uCaM Syn-F2 with uCaM Syn-F1
was obtained at <7 µg/ml (K0.5 1Ca21, 7.5 6 1.5 µg/ml ; K0.5
2Ca21, 6.0 63.5 µg/ml). A maximal rate of <9.0 pkat/mg
uCaM Syn-F2 was obtained. The final 2Ca21/1Ca21 activity
ratio reached at Vmax corresponds to 0.116. Considering the purity of uCaM Syn-F1 as <25%, the half-saturation value can be
estimated to be in the nanomolar range. To obtain an unambiguous value, the experiment was repeated (see insert to Fig. 5A)
with homogenous ubiquitin-activating enzyme (uCaM Syn-F1)
and highly purified uCaM Syn-F2 (MonoQ step 5-F2). The data
497
Fig. 6. Substrate dependence of uCaM synthetase activity. (A) Dependence of uCaM synthetase activity on the concentration of native
calmodulin. The incubation mixture (502100 µl) contained 50 mM Tris/
HCl, 1 mM dithioerythritol, 5 mM MgCl2 , 1 mM ATP, 10 mM phosphocreatine, 0.1 mg/ml creatine kinase, 1.0 mg/ml 125I-CT-ubiquitin
(117 µM), varying concentrations of calmodulin, 4 µg/ml homogenous
uCaM Syn-F1 (E1) and 130 µg/ml affinity purified uCaM Syn-F2. The
mixtures contained either 1.1 mM CaCl2 and 1 mM EGTA (1Ca21 ) or
1 mM EGTA alone (2Ca21). Enzyme activity was calculated from the
linear kinetics obtained in the first 10215 min at 37 °C. Insert: The incubation mixture as above contained instead varying concentrations of native calmodulin or calmodulin derivatives, 0.05 mg/ml 125I-CT-ubiquitin
(6 µM), 0.9 mg/ml Sephacel-APF II [8]. The mixtures were incubated at
37° for 60 min. (d) Native calmodulin 1Ca21 ; (s) native calmodulin
2Ca21 ; (m) Bolton Hunter derivative of calmodulin; (j) chloramine T
derivative of calmodulin; (B) Dependence of uCaM synthetase on ubiquitin derivatives. The incubation mixture (502100 µl) contained
50 mM Tris/HCl, 1 mM dithioerythritol, 5 mM MgCl2, 1 mM ATP,
10 mM phosphocreatine, 0.1 mg/ml creatine kinase, 500 µg/ml calmodulin (30 µM), varying concentrations of 125I-ubiquitin derivatives as indicated, 10 µg/ml affinity purified uCaM Syn-F1 and 75 µg affinity purified uCaM Syn-F2. All mixtures contained 1.1 mM CaCl2 and 1 mM
EGTA (i.e. 1Ca21) and were incubated at 37 °C for 15 min. (m) Bolton
Hunter derivative of ubiquitin; (j) chloramine T derivative of ubiquitin;
(r) iodomonochloride derivative of ubiquitin. (C) Dependence of uCaM
synthetase activity on ATP concentration (d). The incubation mixture
(502100 µl) contained 50 mM Tris/HCl, 1 mM dithioerythritol, 5 mM
MgCl2 , varying concentrations of ATP, 10 mM phosphocreatine, 0.1 mg/
ml creatine kinase, 500 µg/ml calmodulin, 1.0 mg/ml 125I-CT-ubiquitin,
0.9 mg/ml APF II. All mixtures contained 1.1 mM CaCl2 and 1 mM
EGTA (i.e. 1Ca21). Enzyme activity was calculated from the linear kinetics obtained in the first 15220 min at 37°C. Insert : double-reciprocal
plot of the data. The solid line corresponds to the linearized statistically
fitted curve to the data. In all cases the data were fitted non-linearly to
a rectangular hyperbola (Graph Pad Prism) as depicted by the solid line
(r2 5 0.9020.99). For further details see Table 3, Materials and Methods, Results and the previous paper [11].
498
Majetschak et al. (Eur. J. Biochem. 255)
Table 3. Apparent kinetic constants of uCaM synthetase for substrate derivatives. Affinity purified 5 uCaM Syn-F2 affinity purified on CaMSepharose, uCaM Syn-F1 affinity purified on ubiquitin-Sepharose or homogenous [11] at saturating concentrations. The second substrate, ubiquitin
or calmodulin, respectively, was at saturating concentrations (5210-fold Km). APF II (Sephacel) 5 uCaM synthetase enriched and measured in the
standard assay according to [8]. APF II (Fractogel) 5 uCaM synthetase enriched according to [11, 24], calmodulin and ubiquitin 5210 fold Km .
Mean values and standard errors (SEM) are given, the numbers in parentheses indicate the data points. For further details see legend to Fig. 6.
uCaM synthetase preparation
Substrate
Affinity-purified components
APF II (Sephacel)
Calmodulin (Fig. 6A)
CaM (native)
CaM (native)
CT-CaM
BH-CaM
Affinity-purified components
APF II (Fractogel)
Ubiquitin (Fig. 6B)
I-CT-ubiquitin (control)
125
I-BH-ubiquitin
125
I-IC-ubiquitin
ATP (Fig. 6C)
125
could again be fitted to a Michaelis-Menten-type hyperbola allowing the direct determination of the Km value to 3.36 0.7 nM
(see insert to Fig. 5A and Table 3) for the binding of ubiquitinactivating enzyme to uCaM Syn-F2. Thus, the two proteins bind
in a simple non-cooperative manner without requirement for an
additional component.
In the second series of experiments a given amount of
ubiquitin-activating enzyme (affinity-enriched uCaM Syn-F1,
2Ca21/1Ca21 activity ratio 5 1.1) was saturated with uCaM
Syn-F2 (Fig. 5 B). The Michaelis-Menten curve can be measured
to <50% saturation with the concentrations of enriched uCaM
Syn-F2 available. Again all the data points lay on a rectangular
hyperbola. In this case, due to the calcium-independent intrinsic
synthetase activity of affinity-enriched uCaM Syn-F1 (see preceding paper [11]), the 2Ca21/1Ca21 activity ratio initially lies
at 1.1 but then decreases 10-fold to 0.099 at Vmax (see insert to
Fig. 5 B). In the presence of Ca21 the highest specific activity
measured was <350 pkat/mg calculated on a mass basis of enriched uCaM Syn-F2. The maximal catalytic activity (Vmax) was
determined to 756 pkat/mg uCaM Syn-F2. If, however, these
activities are calculated to the constant amount of uCaM SynF1 (10 µg/ml) employed in the test, specific activities of 35.0
and 75.6 nkat/mg uCaM Syn-F1 result. The specific activity of
35 nkat/mg is already <9-fold higher than the hitherto reported
value (4.3 nkat/mg) for purified ubiquitin-activating enzyme
(E1) in the pyrophosphate test [25]. Since the uCaM Syn-F1
preparation employed for these calculations (Fig. 5B) was only
25230% pure, the true maximal rate (Vmax) of pure uCaM SynF1 (E1) saturated with uCaM Syn-F2 can be estimated to be in
the order of 0.220.3 µkat/mg uCaM Syn-F1.
Substrate saturation kinetics of uCaM synthetase. Classical
saturation kinetics of Ca21-dependent uCaM synthetase activity
with the substrates calmodulin, ubiquitin and ATP/Mg21 are
shown in Fig. 6. In all cases classical hyperbolas are obtained.
The kinetics with calmodulin were measured with pure uCaM
Syn-F1 (E1) and affinity purified uCaM Syn-F2 (Fig. 6A). A
Km value of 4.9 µM was obtained (Table 3). Similar values were
obtained with Sephacel-APF II (see insert to Fig. 6A) indicating
that the Km is not significantly influenced by the purity of the
enzyme preparation. The purity is moreover reflected in the Vmax
values. A significant result of the kinetics with pure E1 is that
a Ca21-independent activity can no longer be detected (Fig. 6A),
which indicates that the Ca21-independent activity in uCaM SynF1 is physically removed during the purification of uCaM Syn-
Apparent affinity
constants Km
Apparent maximal rate
constants Vmax
µM
pkat/mg
4.9 6 1.2
6.8 6 1.4
132.96 1.4
4.0 6 1.2
(6)
(8)
(8)
(8)
3.57 6 0.28
0.22 6 0.01
0.29 6 0.06
0.15 6 0.01
(6)
(8)
(8)
(8)
7.66 2.7
43.26 8.0
2.56 0.7
15.66 3.2
(8)
(8)
(8)
(24)
5.75 6 0.45
9.52 6 0.63
4.47 6 0.16
2.79 6 0.13
(8)
(8)
(8)
(24)
F1 (E1) to homogeneity. A chemical modification of calmodulin
however can show a strong influence on synthetase activity. Derivitization of calmodulin with chloramine T led to <20-fold
increase of Km, indicating a dramatic decrease in affinity towards
this substrate (insert in Fig. 6A, Table 3). On the other hand,
derivitization of calmodulin with the Bolton-Hunter reagent has
no influence on the Km and only reduced Vmax by 30% (Table 3).
Bolton-Hunter-derivatized calmodulin is therefore nearly as
good a substrate as native calmodulin.
Enzyme kinetics with ubiquitin have a certain limitation
since the generally accepted substrate for measuring ubiquitinconjugating activity and the substrate used in the FluphenazineSepharose-affinity test is 125I-CT-ubiquitin (see [1]), making it
very difficult to obtain the kinetic constants for native ubiquitin.
Typical values of the Km lie in the range of 8 µM (Table 3). The
iodomonochloride derivative, 125I-IC-ubiquitin, has a lower Km
value of 2.5 µM (Table 3), and thus appears to be a better substrate than the chloramine T derivative. In contrast, derivitization
of ubiquitin with the Bolton-Hunter reagent increases the Km
value to 43 µM, indicating a large (fivefold) decrease in affinity
of ubiquitin for the enzyme system.
The dependence of uCaM synthetase activity on ATP/Mg21
is shown in Fig. 6 C. Quite surprisingly, the Km value was determined to 15.6 nM (Fig. 6 C, Table 3), indicating an affinity constant far below the physiological intracellular ATP concentrations.
DISCUSSION
Up to now only a few ligases of ubiquitin metabolism have
been identified and characterized (for review see [1]). A wellknown and defined representative is the fairly unspecific ubiquitin2protein ligase (EC 6.3.2.19) [28, 29] of reticulocytes for
which multiple forms have been described [29] and which has
been maximally enriched <350-fold from fraction II [28]. In the
form ubiquitylating the substrate 125I-BSA, it consists of three
soluble separate components acting in sequence: (a) ubiquitinactivating enzyme (E1, 110 kDa), (b) ubiquitin-conjugating enzyme (E2, 14 kDa) and (c) the ligase component 3 (E3,
350 kDa). Its ligation products are channeled to the ATP-dependent 26-S protease (26-S proteasome) for proteolytic degradation. Other ubiquitin2protein ligases such as E6-AP ubiquitin2
protein ligase [30], cyclin ubiquitin2protein ligase [31, 32] and
Majetschak et al. (Eur. J. Biochem. 255)
an enzyme processing a transcriptional activator precursor, the
p105 ubiquitin2protein ligase [33], have also been reported. All
of these ligases appear to consist of three soluble separate enzyme components similar to the unspecific ubiquitin2protein
ligase described above. As far as is known all of these systems
have identical first components, i.e. the ubiquitin2activating enzyme E1. However, although the second and third components
are often denoted by the same terms (i.e. E2 and E3), they vary
in structure and specificity, which is in most cases not known in
detail. The three-component systems have in common the connection to the ATP-dependent 26-S protease allowing a classification under catabolic protein ubiquitylation in the form of the
ubiquitin-proteasome pathway.
It is therefore of great interest as to whether other general
types of ubiquitin ligase systems exist. Ubiquitin2calmodulin
ligase (uCaM synthetase, EC 6.3.2.21) was the first specific
ubiquitin ligase described (see [1]). In contrast to the above ligases of the ubiquitin proteasome pathway, ubiquitin2calmodulin ligase does not channel calmodulin to degradation by the
ATP-dependent 26-S protease in agreement with the fact that its
blocked N-terminus is not a recognition signal for the N-end
rule. Moreover, uCaM synthetase catalyzes a reaction readily
reversible in cell-free extracts [2], thus allowing a classification
under the group of non-catabolic protein ubiquitylations. Present
indications are that ubiquitylation strongly decreases the biological activity of calmodulin [1, 34]. The binding of calmodulin to
the target uCaM Syn-F2 has been demonstrated here on calmodulin2Sepharose (see also [35, 36]) and also by kinetic experiments with various calmodulin substrates (Km ca. 527 µM, Table 3). This Km allows a second classification of uCaM synthetase also under the calmodulin-binding proteins, specifically the
group of low-affinity targets of calmodulin similar to caldesmon
and B50/neuromodulin (for review see ([37]).
In the present papers we have succeeded in purifying uCaM
synthetase 3500-fold. It could be demonstrated that the first fraction, uCaM Syn-F1, corresponding to the first component of
uCaM synthetase, is identical to ubiquitin-activating enzyme E1
(see preceding paper [11]). The purity of this first component
and the absence of any contaminating ubiquitin-conjugating enzymes could be conclusively demonstrated [11]. Therefore, all
other putative components of the uCaM synthetase system could
only have been in the uCaM Syn-F2 fraction. However, no such
additional component (E2, E3) essential for activity could be
detected in this fraction during seven purification steps. The
following experiments also strongly support this two-component
hypothesis : (a) a single activity peak was demonstrated for both
components in four chromatographic steps to 3500-fold purification with the gel-filtration step having the greatest weight; (b)
the activation of purified uCaM Syn-F2 (stage 5-F2) by pure
(E2-free) ubiquitin-activating enzyme E1 follows one simple hyperbolic velocity curve. Thus we are left with the question as to
what kind of a protein this second component of uCaM synthetase might be.
Is uCaM Syn-F2 a ubiquitin-conjugating enzyme (E2) ? In
an attempt to define E2s on a functional basis, Klemperer et al.
[38] considered a protein of 230 kDa to belong to the family of
E2s on the grounds that (a) the catalysis of ubiquitin transfer is
E1-dependent but occurs without involvement of an E3 and (b)
the preparation contains a protein that accepts ubiquitin, in labile
linkage, from the E1 ubiquitin thiolester. Another characteristic
property has been noted to be the ability of an E2 to bind to
ubiquitin-activating enzyme [39]. Probably the most stringent
proof for identifying a protein as an E2 is obtained from the
amino-acid sequence. In the sequence data banks <80 ubiquitinconjugating enzymes have been deposited ranging in monomer
molecular mass over 13.2234.1 kDa. Since some of these en-
499
Fig. 7. Model of the interplay of uCaM Syn-F1 (E1) and uCaM SynF2 in the conjugation of calmodulin with ubiquitin.
zymes occur as oligomers, maximal molecular masses of
120 kDa as for the tetramer of 32-kDa E2 can be expected [39].
Thus it is unclear how the 230-kDa protein of Klemperer et al.
[38] fits into this structural class of E2s. Turning to uCaM SynF2, some properties are in agreement with the above definitions :
(a) it binds to E1 with high affinity (Km 5 3.3 nM) as suggested
by the kinetic experiments; (b) it transfers ubiquitin in an E1dependent manner to a substrate without involvement of a third
component, i.e. an E3. On the other hand, the molecular mass
of 623 kDa again makes it very improbable that uCaM Syn-F2
belongs structurally to the family of E2s. In addition, the required labile ubiquitin linkage of ubiquitin to uCaM Syn-F2 has
eluded detection. Is uCaM Syn-F2 therefore an E3? This can
also not be affirmed, since there are only two components and,
in addition, an E3 does not bind to an E1 [40]. Also the basic
pH optimum (pH 829, Fig. 4) of uCaM synthetase is incompatible with the neutral pH optimum reported for the only characterized E3, i.e. that of the ubiquitin2protein ligase system [41].
We therefore conclude that uCaM Syn-F2 is neither an E2 nor
an E3 but evidently combines the properties of both, making
Ca21-dependent uCaM synthetase the member of a novel group
of two-component ubiquitin ligase systems.
Although the search for labile ubiquitin conjugates in uCaM
Syn-F2 fractions of highest purity was negative, the experiments
with N-ethylmaleimide indicate the existence of an essential SH
group on uCaM Syn-F2, suggesting the possibility that this protein or the putative oligomer contains a subunit having the homologous function of a ubiquitin-conjugating enzyme. For some
reason, the synthesis of labile ubiquitin conjugates may be kinetically unfavorable under the conditions we have chosen, so that
the protein in uCaM Syn-F2 which might be involved in ubiquitin transfer to calmodulin could not be identified. Other groups
[30, 42] have reported that an essential SH-group may be present
on the third component (E3) of ubiquitin2protein ligases, indicating a still more complex mechanism of ubiquitin transfer also
in other ubiquitin2protein ligases.
Not much kinetic information has been published on the substrates of ubiquitin ligase systems. The best examined protein is
ubiquitin-activating enzyme [43] where equilibrium and dissociation constants for various partial reactions (six different Michaelis complexes) were determined. For ATP and ubiquitin
binding to the thiol ester enzyme, dissociation constants of
0.45 µM and 0.58 µM were reported, respectively [43]. However, the concentrations of ATP necessary for half-maximal
velocity of PPi exchange and of AMP exchange were reported
500
Majetschak et al. (Eur. J. Biochem. 255)
as 36 µM and 38 µM, respectively [43], demonstrating the difficulty of comparing isolated dissociation constants with putative
Km values in such systems. The latter values however compare
in magnitude with the Km value for ATP (15.6 µM) as determined in this paper (Table 3). By similar reasoning, it is difficult
to reconcile the low dissociation constants above for ubiquitin
[43] with the Km values reported here. In the uCaM synthetase
system the Km value of ubiquitin depends strongly on the isotope
labeling procedure, with the iodomonochloride method yielding
the lowest (2.5 µM) and the Bolton-Hunter method the highest
(43 µM) Km values. Similar values have been derived from activity measurements with the complete ATP-ubiquitin-dependent
proteolytic system (i.e. ubiquitin2protein ligase 1 ATP-dependent 26-S protease) in reticulocyte APF II, where the half-saturation constant for ubiquitin was estimated to 2.8 µM [12]. The
strong dependency of the Km on the chemistry of the isotopic
labeling procedure, as reported here, cannot however be easily
explained by the conclusion of Haas and Rose [43] that native
and iodinated 125I-CT-ubiquitins are functionally identical for
ubiquitin-activating enzyme (E1). It may therefore be that the
differential reactivities of the modified ubiquitins, as shown
here, only become effective in multicomponent systems further
down-stream [44246], e.g. at the level of ubiquitin-conjugating
enzymes [46] or the ligase component. This again bears out the
fact that the measurement of kinetic constants in holo-enzyme
systems is essential, especially if the numbers are to be relevant
for biological systems.
The experiments in this paper also allowed the estimation of
the maximal catalytic activity of the uCaM synthetase holo-system and the individual components for the first time. Interestingly, the highest specific activity (Vmax of purified ubiquitinactivating enzyme (E1) in this system was estimated to lie in the
range of 0.2 µkat/mg (see above). From this, and a monomer
molecular mass of 112 kDa, a kcat of 2.2 s21 can be calculated.
Taking a mean Km value of 8 µM for ubiquitin (Table 3), an
estimate of the kcat/Km ratio for ubiquitin transfer yields a value
of 2.83105 s21 M21.
On the basis of the preceding paper [11] and this work, a
plausible working model of the interaction of uCaM Syn-F1
(E1) and uCaM Syn-F2 can be made (Fig. 7). In this model
uCaM Syn-F1 (E1) binds ubiquitin in the presence of ATP via
a thioester bond (for review see [1]). UCaM Syn-F2, on the
other hand, has two binding sites as shown in this paper : one
for Ca21-calmodulin and a second site for binding the uCaM
Syn-F1(E1)-ubiquitin complex. When the two loaded components of uCaM synthetase themselves bind to form a higher order complex, ubiquitin is transferred to calmodulin, forming
ubiquityl-calmodulin.
This work was supported by grants from the Deutsche Forschungsgemeinschaft (Je 84/8-1 and Je 84/9-1. We also thank Dr R. Ziegenhagen
for specific aspects of this work and Mrs G. Botzet and Mrs G. Hessler
for excellent technical help.
REFERENCES
1. Jennissen, H. P. (1995) Eur. J. Biochem. 231, 1230.
2. Laub, M. & Jennissen, H. P. (1997) Biochim. Biophys. Acta 1357,
1732191.
3. Laub, M. & Jennissen, H. P. (1989) Biol. Chem. Hoppe-Seyler 370,
9262927.
4. Laub, M. & Jennissen, H. P. (1991) FEBS Lett. 294, 2292233.
5. Gehrke, P. P., Ziegenhagen, R. & Jennissen, H. P. (1987) Biol. Chem.
Hoppe-Seyler 368, 744.
6. Ziegenhagen, R., Gehrke, P. & Jennissen, H. P. (1988) FEBS Lett.
237, 1032107.
7. Jennissen, H. P. & Laub, M. (1988) Biol. Chem. Hoppe-Seyler 369,
132521330.
8. Ziegenhagen, R. & Jennissen, H. P. (1988) Biol. Chem. HoppeSeyler 369, 131721324.
9. Webb, E. C. (1990) Eur. J. Biochem. 187, 2632281.
10. Laub, M., Steppuhn, J. A., Blüggel, M., Immler, D., Meyer, H. E. &
Jennissen, H. P. (1998) Eur. J. Biochem. 255, 4222431.
11. Majetschak, M., Laub, M., Meyer, H. E. & Jennissen, H. P. (1998)
Eur. J. Biochem. 255, 4822491.
12. Gehrke, P. P. & Jennissen, H. P. (1987) Biol. Chem. Hoppe-Seyler
368, 6912708.
13. Ciechanover, A., Heller, H., Elias, S., Haas, A. L. & Hershko, A.
(1980) Proc. Natl Acad. Sci. USA 77, 136521368.
14. Bolton, A. E. & Hunter, W. M. (1973) Biochem. J. 133, 5292539.
15. Ziegenhagen, R. & Jennissen, H. P. (1990) FEBS Lett. 273, 2532
256.
16. McFarlane, A. S. (1958) Nature 182, 53.
17. Ly, B. & Kierulf, P. (1975) Thromb. Res. 6, 3872398.
18. Jennissen, H. P. & Botzet, G. (1993) J. Mol. Recogn. 6, 1172130.
19. van Eldik, L. J. & Watterson, D. M. (1981) J. Biol. Chem. 256,
420524210.
20. Chafouleas, J. G., Dedman, J. R., Munjaal, R. P. & Means, A. R.
(1979) J. Biol. Chem. 254, 10 262210 267.
21. Fischer, E. H. & Krebs, E. G. (1958) J. Biol. Chem. 231, 65271.
22. Sairam, M. R. & Porath, J. (1976) Biochem. Biophys. Res. Commun.
69, 1902196.
23. Sharma, K. R., Wang, T. H., Wirch, E. & Wang, J. H. (1980) J. Biol.
Chem. 255, 591625923.
24. Jennissen, H. P., Botzet, G., Majetschak, M., Laub, M., Ziegenhagen, R. & Demiroglou, A. (1992) FEBS Lett. 296, 51256.
25. Ciechanover, A., Elias, S., Heller, H. & Hershko, A. (1982) J. Biol.
Chem. 257, 253722542.
26. Lowry, O. H., Rosebrough, N. J., Farr, A. L. & Randall, R. J. (1951)
J. Biol. Chem. 193, 2652275.
27. Laemmli, U. K. (1970) Nature 227, 6802685.
28. Reiss, Y. & Hershko, A. (1990) J. Biol. Chem. 265, 368523690.
29. Heller, H. & Hershko, A. (1990) J. Biol. Chem. 265, 653226535.
30. Scheffner, M., Nuber, U. & Huibregtse, J. M. (1995) Nature 373,
81283.
31. Sudakin, V., Ganoth, D., Dahan, A., Heller, H., Hershko, J., Luca,
F. C., Ruderman, J. V. & Hershko, A. (1995) Mol. Biol. Cell 6,
1852197.
32. King, R. W., Peters, J. M., Tugendreich, S., Rolfe, M., Hieter, P. &
Kirschner, M. W. (1995) Cell 81, 2792288.
33. Orian, A., Whiteside, S., Israel, A., Stancovski, I., Schwartz, A. L. &
Ciechanover, A. (1995) J. Biol. Chem. 270, 21 707221 714.
34. Laub, M. & Jennissen, H. P. (1993) Biol. Chem. Hoppe-Seyler 374,
774.
35. Majetschak, M., Laub, M. & Jennissen, H. P. (1992) Biol. Chem.
Hoppe-Seyler 373, 7972798.
36. Parag, H. A., Dimitrovsky, D., Raboy, B. & Kulka, R. G. (1993)
FEBS Lett. 325, 2422246.
37. Vogel, H. J. & Zhang, M. (1995) Mol. Cell. Biochem. 149(150), 32
15.
38. Klemperer, N. S., Berleth, E. S. & Pickart, C. M. (1989) Biochemistry 28, 603526041.
39. Haas, A. L. & Bright, P. M. (1988) J. Biol. Chem. 263, 13 2582
13 267.
40. Hershko, A., Heller, H., Elias, S. & Ciechanover, A. (1983) J. Biol.
Chem. 258, 820628214.
41. Lee, P. L., Midelfort, C. F., Murakami, K. & Hatcher, V. B. (1986)
Biochemistry 25, 313423138.
42. Gonen, H., Stancovski, I., Shkedy, D., Hadari, T., Bercovich, B.,
Bengal, E., Mesilati, S., Abu-Hatoum, O., Schwartz, A. L. &
Ciechanover, A. (1996) J. Biol. Chem. 271, 3022310.
43. Haas, A. L. & Rose, I. A. (1982) J. Biol. Chem. 257, 10 329210 337.
44. Cox, M. J., Haas, A. L. & Wilkinson, K. D. (1986) Arch. Biochem.
Biophys. 250, 4002409.
45. Bamezai, S., Banez, M. A. T. & Breslow, E. (1990) Biochemistry
29, 538925396.
46. Pickart, C. M., Haldeman, M. T., Kasperek, E. M. & Chen, Z. (1992)
J. Biol. Chem. 267, 14 418214 423.