BIOCHEMICAL SOCIETY TRANSACTIONS
388
240000 (Tanabe et al., 1975; Mackall & Lane, 1977; Brownsey
et al., 1977; Witters et al., 1979). We found that acetyl-CoA
carboxylase from lactating rabbit mammary gland has a slightly
larger subunit mol.wt., 250000 (Hardie & Cohen, 1978, 1979).
We now report that brief incubation of this enzyme with
proteinases converts it into 235 000-mol.wt. fragment without
affecting the activity. However, this conversion may alter the
regulation of the enzyme, because at least one of the phosphorylation sites is located on the small fragments that are released.
Acetyl-CoA carboxylase was purified from rabbit mammary
gland by the method of Hardie & Cohen (1978). Digestion was
carried out at 25OC in Tris/HCl buffer, 10.05, pH7.0,
containing 1mM-EDTA, 0.1% (v/v) 8-mercaptoethanol, 1mg of
acetyl-CoA carboxylase/ml and 0.1 p g of proteinase/ml.
All four proteinases tested produced similar patterns of
digestion, but did so at different rates: trypsin > elastase = subtilisin > chymotrypsin. During the first lOmin of incubation with
trypsin, the intact subunit of mo1.M. 250000 was converted
quantitatively into a fragment of mo1.w. 235000, with a
transient intermediate of mol.wt. 242000. the 242000- and
235 000-mol.wt. fragments correspond in mobility to the two
minor components of rabbit mammary acetyl-CoA carboxylase
observed previously (Hardie & Cohen, 1978). We have been
unable to detect the small fragments released, as they appear to
be rapidly degraded into small peptides. After much longer
digestion times (30-60min with trypsin) the 235 000-mol.wt.
fragment is further degraded to subfragments of mol.wts.
120000 and 11OOOO. This corresponds to the proteolytic
cleavage previously described for the rat liver enzyme (Tanabe
et al., 1975). Both cleavages occurred more slowly in the
presence of the allosteric activator citrate (IOmM), an effect
previously noticed for the secondary cleavage (Tanabe et al.,
1977). One can conclude that the citrate-induced polymerization of the enzyme (Gregolin et al., 19686) renders it less
accessible for the proteinase.
Conversion of the enzyme into the 235 000-mol.wt. fragment
by trypsin had no detectable effect on the activity of rabbit
mammary acetyl-CoA carboxylase in the optimal assay or on
the concentration of citrate required for half-maximal activation. Phosphorylation of the enzyme by cyclic AMP-dependent protein kinase or acetyl-CoA carboxylase kinase-2 (Hardie
& Cohen, 1978) did not affect the rate of tryptic cleavage.
However, if acetyl-CoA carboxylase was phosphorylated by
using [ Y - ~ ~ P I A Tand
P acetyl-CoA carboxylase kinase-2, and
subjected to tryptic digestion, only a few per cent of the
radioactivity was recovered in the 235 000-mol.wt. fragment.
Similar experiments using cyclic AMP-dependent protein kinase
showed that about 50% of the radioactivity was recovered in the
235 000-mol.wt. fragment. Since cyclic AMP-dependent protein
kinase phosphorylates the enzyme at two sites, one of which
appears to be identical with the site phosphorylated by
acetyl-CoA carboxylase kinase-2, we conclude that this site is
located in the small terminal fragment(s) that is rapidly released
by proteinases (Hardie, 1980).
This study should sound a warning note in that it is possible
to prepare a fully active form of acetyl-CoA carboxylase which
lacks a phosphorylation site of possible regulatory significance.
Phosphorylation sites would be expected to be located on
exposed portions of the protein surface and may therefore be
susceptible to proteolysis. This has been well documented for
phosphorylase (Fischer et al., 1959), glycogen synthase (Takeda
& Larner, 1975) and pyruvate kinase (Bergstrom et al., 1978).
This possibility should be borne in mind before concluding that a
protein is not a substrate for a protein kinase.
This work was supported by project grants from the Medical
Research Council and the British Diabetic Association.
Bergstrom, G., Ekman, P., Humble, E. & Engstrom, L. (1978)
Biochim. Biophys. Acra 532,259-267
Bloch, K. & Vance, D. (1977) Annu. Reu. Biochem. 46,263-298
Brownsey, R. W., Hughes, W. A., Denton, R. M. & Mayer, R. J.
(1977) Biochem. J. 168,441-445
Carlson, C. A. & Kim, K. (1973)J. Biol. Chem. 248,378-380
Fischer, E. H., Graves, D. J., Crittenden, E. R. S.& Krebs, E. (1959)J.
Biol. Chem. 234,1698-1704
Gregolin, C., Ryder, E. & Lane, M. D. (19680) J. Biol. Chem. 243,
4227-4235
Gregolin, C., Ryder, E., Warner, R. C., Kleinschmidt, A. K. & Lane,
M.D. (1968b)P~ro~.
Null. Acad. SCi. U S A A56,1751-1758
.
Hardie, D. G. (1980) in Molecular Aspects of Cellular Regularion
(Cohen, P., ed.), vol. 1. Elsevier/North-Holland, Amsterdam, in the
press
Hardie, D. G. & Cohen, P. (1978) FEBS Lett. 91, 1-7
Hardie, D. G. & Cohen, P. (1979) Eur. J. Biochem. 92,25-34
Inoue, H. & Lowenstein, J. M. (1973)J. Biol. Chem. 247,4825-4832
Mackall, J. & Lane, M.D. (1977) Bfochem.J. 162,635-642
Takeda, Y. & Lamer, J. (1975) J. Biol. Chem. 250,895 1-8956
Tanabe, T., Wada, K., Okayaki, T. & Numa, S. (1975) Eur. J.
Biochem. 57,15-24
Tanabe, T., Wada, K., Ogiwara, H. & Numa, S. (1977) FEBS Lett. 82,
85-88
Witters, L. A., Kowaloff, E. M. & Avruch, J. (1979) J. Biol. Chem.
254,245-248
A protein of 70000 molecular weight is joined by disulphide bridges to pig gastric-mucus
glycoprotein
JEFFREY P. PEARSON and ADRIAN ALLEN
Department of Physiology, The Medical School, University of
Newcastle upon o n e , Newcastle upon o n e NEI 7R U,U.K.
The glycoprotein (molecular weight 2 x lo6) from pig gastric
mucus was purified free of non-covalently bound protein by gel
filtration on Sepharose 4B followed by equilibrium centrifugation in a CsCl density gradient (Starkey et al., 1974). N o
protein components could be detected in the glycoprotein by
polyacrylamide-gel electrophoresis in 1% SDS*. In contrast,
protein bands were clearly visible after gel electrophoresis of the
glycoprotein in 1% SDS containing 0.2 M-mercaptoethanol.
There was a strongly staining major protein band of molecular
weight 70000, together with another weakly stained band of
Abbreviation: SDS, sodium dodecyl sulphate.
60000 molecular weight. There were also suggestions of one or
two very faint bands in the region of 90000-100000 and
30000-40000 molecular weight. Gel filtration on Sepharose 2B
showed that the glycoprotein was all undegraded polymer,
excluded by the gel, whereas after reduction in 0.2 M-mercaptoethanol the glycoprotein was all subunit, of molecular weight
5 x los and included on the gel (Pearson et al., 1979). The
totally included fraction from the eluate of the reduced
glycoprotein on Sepharose 2B, although containing no detectable glycoprotein, did contain protein, which on SDS/polyacrylamide-gel electrophoresis showed the same major protein
band of 70000 molecular weight.
A second fractionation of the glycoprotein in a CsCl density
gradient which contained 0.2 M-mercaptoethanol separated the
protein in the low-density fractions 1 and 2 (Fig. 1) from the
reduced glycoprotein in the high-density fractions 5, 6 and 7.
1980
587th MEETING, DUNDEE
389
2.0 r
h
B
.3 1.0 -
v
rE
16
0Y
. 8
.-C
2
e
n
- 4
8
8
-0
1
2
3
4
5
6
7
8
Fraction no.
Fig. I. Equilibrium centrifugation of pig gastric glycoprotein in
a CsCl density gradient in the presence or absence of
0.2 M-mercaptoethanol
Pure glycoprotein was dissolved in CsCl at a starting density of
1.42g.ml-1; after centrifugation (1.5 x lO'g, 5"C, 48h) the
densities in fractions 1 and 8 were 1.38 and 1.56g.ml-I
respectively. Four tubes contained 0.2 M-mercaptoethanol and
CsCl (A,W) and four tubes contained CsCl only (40).Protein
(A, A) was measured by the method of Lowry et a f . (I95 I ) and
glycoprotein (H, 0)by the method of Mantle & Allen (1978).
of the glycoprotein in SDS in non-reducing conditions, was
evidence that the protein removed on reduction was joined
covalently by disulphide bridges to the glycoprotein.
Proteolytic digests of the glycoprotein with limiting amounts
of enzyme were examined, as a function of time, by SDS/polyacrylamide-gel electrophoresis in the absence of reducing agent.
Digestion of the glycoprotein with papain, trypsin or pepsin for
periods up to 6 h gave a band of molecular weight 7000080000 as the major protein band on electrophoresis. Thus the
protein that is relnased by proteolytic digestion is itself
somewhat resistant to further digestion, although after longer
incubation times smaller proteins, of molecular weight 20000
and below, replaced the higher-molecular-weight band. Similarly
protein bands of less than 20000 molecular weight only were
observed during limited papain digestion of the reduced
glycoprotein subunit, which had been fractionated by gel
filtration from the protein released on reduction.
The protein isolated from the glycoprotein after reduction
(Fig. 1) represents about 3.7% by weight of the total
glycoprotein, and this would be just sufficient for there to be one
molecule of the protein of molecular weight 70000 per
glycoprotein molecule. Previous work has shown that exhaustive
proteolytic digestion removes about 4% by weight of the purified
glycoprotein, 25% of its total protein content (Scawen & Allen,
1977) and it follows that much of this is the 70000-molecularweight protein. Further, this protein, which is joined to the
glycoprotein by disulphide bridges, is released from it by
reduction or proteolysis, both of which result in the formation of
glycoprotein subunits. Immunodiffusion studies have shown that
this 70000-molecular-weight protein is not serum albumin; but it
is interesting to note its similarities in molecular size, resistance
to peptic digestion and dependence on disulphide bridges to the
secretory component which is present in mucus secretions.
We thank Dr. Steve Parry for the immunoelectrophoresis and the
Medical Research Council for financial support.
SDS/polyacrylamide-gel electrophoresis of the protein from
fractions I and 2 showed the presence of the band at 70000
molecular weight as the main component. In the absence of
0.2~-mercaptoethanol, no protein was released from the
glycoprotein in the 3.5M/CSCI gradient (Fig. 1) even when
4 M-guanidinium chloride was included. This, together with the
complete absence of protein bands on gels from electrophoresis
Lowry, 0. H., Rosebrough, N. J.. Farr, A. L. & Randall, R. J. (195 1) J.
Biol. Chem. 193,265-275
Mantle, M.& Allen, A. (1978) Biochem. Soc. Trans. 6,607-609
Pearson, J. P., Allen, A. & Venables, C. W. (1979) Biochem. SOC.
Trans. 1,904-905
Scawen, M.S. & Allen, A. (1977) Biochem. J. 163,363-368
Starkey, B. J., Snary, D. & Allen, A. (1974) Biochem. J. 141,633-639
Studies on the purification of glucose 6-phosphatase from rabbit liver microsomal fraction
GORDON F. BICKERSTAFF and BRIAN BURCHELL
Department of Biochemistry, Medical Sciences Institute,
University of Dundee, Dundee DD1 4HN,Scotland, U.K.
Glucose 6-phosphatase (EC 3.1.3.9), a microsomal enzyme
found in liver, kidney and small intestine, has not been purified
extensively, although the enzyme has been known for more than
35 years (Nordlie, 1971). The lack of success in purification
procedures is due mainly to the instability of the enzyme, which
has hindered attempts to obtain a solubilized preparation with
good retention of enzyme activity. In this communication we
report a satisfactory method for solubilizing the enzyme, and
some preliminary purification steps.
The liver from an adult rabbit was homogenized in 5vol. of
ice-cold 0.25 M-sucrose solution. The isolation of microsomal
fraction from the homogenate was performed essentially as
described by Burchell (1977). The microsomal pellets were
resuspended in a volume of buffer A 10.1 M-Tridacetate buffer,
pH 7.5, containing 1.0% sodium cholate and 20% (v/v) glycerol]
q u a 1 to 5.0% of the volume of the homogenate. The
VOl.
8
solubilization was assisted by gentle homogenization by hand,
then the mixture was left at 4OC for 20min. The mixture was
then centrifuged at 105000g at 4OC for 1.0h before the
supernatant was collected. The supernatant contained over 95%
of the microsomal protein and all of the microsomal glucose
6-phosphatase activity (Table 1). The solubilized preparation
could be stored at -2OOC for a few months without significant
loss of activity.
The protein concentration of the solubilized preparation was
adjusted to 20mg/ml by dilution with further buffer A, and the
tempercture of the solution lowered to 0-3OC in an ice-bath. To
the gently stirred solution was added 25% (w/v) poly(ethy1ene
glycol) so!ution (average mol.wt. 6000) slowly to a final
concentration [of poly(ethy1ene glycol)] of 3%. The mixture was
stirred for a further IOmin before the precipitate was collected
by centrifugation at 25000g for 20min at 4OC. The precipitate
was resuspended in a volume of buffer A equal to 3 times the
volume of poly(ethy1ene glycol) added, and the mixture
dispersed by gentle homogenization by hand. The preparation
can be stored at -2OOC for a few months without loss of