MOLECULAR AND CELLULAR BIOLOGY, Feb. 1994, p. 896-905
Copyright © 1994, American Society for Microbiology
The Mutant Type 1 Protein Phosphatase Encoded by glc7-1
from Saccharomyces cerevisiae Fails To Interact Productively
with the GAC1-Encoded Regulatory Subunit
JOAN SKROCH STUART,t DEBRA L. FREDERICK, CATHERINE M. VARNER,
AND KELLY TATCHELL*
Department ofMicrobiology, North Carolina State University, Raleigh, North Carolina 27695-7615
Received 27 August 1993/Returned for modification 14 October 1993/Accepted 26 October 1993
because of low glycogen synthase activity. Increased dosage of GACI results in increased activity of glycogen
synthase and a corresponding hyperaccumulation of glycogen. The glycogen accumulation phenotype ofgacl
is similar to that ofglc7-1, a type 1 protein phosphatase mutant. We have partially characterized the GACI gene
product (Gaclp) and show that levels of Gaclp increase during growth with the same kinetics as glycogen
accumulation. Gaclp is phosphorylated in vivo and is hyperphosphorylated in a gkc7-1 mutant. Gaclp and the
type 1 protein phosphatase directly interact in vitro, as assayed by coimmunoprecipitation, and in vivo, as
determined by the dihybrid assay described elsewhere (S. Fields and O.-k. Song, Nature [London] 340:245-
246, 1989). The interaction between Gaclp and the glc7-1-encoded form of the type 1 protein phosphatase is
defective, as assayed by either immunoprecipitation or the dihybrid assay. Increased dosage ofGACI partially
suppresses the glycogen defect ofgkc7-1. Collectively, our data support the hypotheses that GACI encodes a
regulatory subunit of type 1 protein phosphatase and that the glycogen accumulation defect ofgkc7-1 is due
at least in part to the inability of the mutant phosphatase to interact with its regulatory subunit.
mutants of Saccharomyces cerevisiae fail to accumulate normal levels of glycogen
Protein phosphorylation is an almost universal mechanism
of posttranslational regulation. In eukaryotes, the transfer of
phosphate to serine, threonine, and tyrosine residues is
carried out by a conserved family of protein kinases. The
large number of protein kinases in the cell (37) reflects, at
least in part, their high degree of substrate and/or functional
specificity. Phosphate removal, carried out by phosphopro-
tein phosphatases, appears to follow a different paradigm.
These enzymes do not show the evolutionary conservation
found among protein kinases. A large family of phosphoty-
rosine phosphatases (62) are unrelated to two families of
phosphoserine/threonine phosphatases represented by type
1, type 2A, and type 2B phosphatases (17) and type 2C
phosphatases (59), respectively. Although a large number
of soluble and membrane-bound phosphotyrosine phos-
phatases have recently been identified (62), the number of
phosphoserine/threonine protein phosphatases appears to be
much smaller than that of the corresponding protein kinases,
and the substrate specificity of protein phosphatases is much
lower than the specificity of protein kinases (17). However,
as has been established for protein kinases, phosphoprotein
phosphatases have recently been demonstrated to play key
roles in many cell cycle and metabolic regulatory processes
(reviewed in references 2, 18, and 20).
The type 1 serine/threonine phosphoprotein phosphatase
is an abundant enzyme that has been implicated in the
regulation of diverse cellular processes, ranging from glyco-
gen synthesis in response to insulin (22) to cell cycle regu-
lation (1, 43, 48) and muscle contractility (15). Functional
specificity appears to be derived from regulatory subunits
*Corresponding author. Mailing address: Department of Micro-
biology, Box 7615, North Carolina State University, Raleigh, NC
27695-7615. Phone: (919) 515-3629. Fax: (919) 515-7867.
t Present address: Glaxo Research Institute, Research Triangle
Park, NC 27709.
that target the catalytic subunit to the appropriate site in the
cell and in some cases alter its substrate specificity. The best
evidence for this model comes from studies of a protein
phosphatase that regulates glycogen synthesis in skeletal
muscle. This phosphatase (PP1G) consists of a 37-kDa cata-
lytic subunit (PP1) and a 125-kDa regulatory subunit(RGI)
that targets the catalytic subunit to the glycogen particle.
Dent et al. (22) found that phosphorylation of RG1 by an
insulin-dependent protein kinase increased the specificity of
the phosphatase towards glycogen synthase and phosphory-
lase kinase but not other substrates. In contrast, phospho-
rylation by cyclic AMP-dependent protein kinase results in
the dissociation of RG1 and the catalytic subunit (41). The
catalytic subunit of PP1G is abundant, highly conserved, and
found in eukaryotes ranging from plants (47, 56), fungi (24),
and yeasts (8, 48) to insects (4, 23) and mammals (5, 19, 48),
but the RGl subunit is restricted to skeletal muscle (60).
In addition to the glycogen-targeting subunit, three low-
molecular-weight inhibitor proteins that block or modify the
activity of PP1 have been identified (7). The role of these
proteins in vivo is not known. Other targeting or regulatory
subunits have been partially characterized (7, 17), but the
details of their subcellular and tissue distributions and their
functions in vivo remain to be established.
Recent evidence suggests an important role for PP1 in
modulating glycogen synthesis in the budding yeast Saccha-
romyces cerevisiae. Glycogen synthase is interconverted
between a phosphorylated, inactive form and a dephospho-
rylated, active form (32, 33, 40, 49) in wild-type cells but
remains almost entirely in the phosphorylated form in the
glc7-1 mutant (34, 49). glc7-1 is a mutant in GLC7/DIS2S1
(12, 27), which encodes the catalytic subunit of PP1 (48).
Although GLC7 is an essential gene (16, 27), the glc7-1
mutation results in a relatively mild defect. glc7-1 mutants
fail to accumulate glycogen (34, 49), show a reduced sporu-
lation efficiency (12), and suppress defects caused by muta-
Vol. 14, No. 2
TYPE 1 PROTEIN PHOSPHATASE FROM S. CEREVISL4E
TABLE 1. Yeast strains used in this study
Reference or source
MATa/MA Ta his3/his3 leu2Ileu2 ura3-52/ura3-52 glc7::HIS3/GLC7
MATa ura3-52 leu2 his4 gacl::URA3
MATa ura3-52 leu2 his3 gacl::ura3 glc7::HIS3 YCp5O(URA3)-Ha-GLC7
MATa ura3-52 Ieu2 his3 glc7::HIS3 YCp5O(URA3)-Ha-GLC7
MATa ura3-52 leu2 his3 gacl::ura3 glc7:JHIS3 YCp5O(URA3)-myc-GLC7
MATa leu2 glc7::HIS3 YCp5O(URA3)-Ha-glc7-1
ALTa ura3 leu2 his3
MATa leu2 ura3-52 his3 gk7::HIS3 pep4-AJ gacl::ura3 YCp5O-Ha-glc7-1
MATaleu2 ura3-52 his3 gc7:HIS3pep4-Al gacl::ura3 YCp5O-Ha-GLC7
MATa leu2 ura3-52 his3 glc7::HIS3pep4-Al gacl::ura3 YCp5O-myc-GLC7
MATa ura3-52 leu2 glc7-1
MATa leu2 ura3-52 his3 glc7::LEU2 YCp5O(URA3)-GLC7
Agal4 Agal80 ura3 his4 leu2 ade-
aCongenic to yeast strain JC482 (13).
tions in the translation initiation factor 2 kinase, GCN2 (63),
but grow with nearly normal doubling times (12, 27, 34).
We have previously identified a gene (GAC1) that may
encode a glycogen-specific regulatory subunit of PP1 (34).
Increased dosage of GACI causes an increase in the ratio of
dephosphorylated to phosphorylated glycogen synthase,
whereas loss-of-function alleles of GACI mimic the pheno-
type of glc7-1 with respect to the activity of glycogen
synthase. Unlike GLC7, however, GAC1 is not essential for
growth; gacl null mutants are viable with no observable
defects other than reduced glycogen levels. The amino acid
sequence of the putative GACJ-encoded protein (Gaclp) is
most similar to the mammalian glycogen binding subunit of
PP1,RGI(60). In this report we identify Gaclp and provide
evidence that it interacts with PP1. The mutant form of PP1,
encoded byglc7-1, is defective in its interaction with Gaclp.
markers were introduced into the JC482 background either
by serial backcrosses to the JC482 background or by trans-
formation. The pep4Al mutation in the JC482 background
was a gift of Jens G. L. Petersen. Yeast transformation was
performed by the lithium acetate method of Ito et al. (42) or
Gietz et al. (35). Standard yeast genetic procedures for
diploid construction, sporulation, tetrad analysis, and me-
dium preparation were as described by Rose et al. (52).
Escherichia coli DH5LF' was used for the propagation of
plasmids. Yeast strains were grown in YEPD (1% yeast
extract, 2% peptone, 2% glucose) or minimal media (0.67%
yeast nitrogen base, 2% glucose) supplemented with amino
acids. Cell density was determined with either a Coulter
model ZM particle counter or a hemacytometer (Brightline;
Plasmid construction. Plasmids are listed in Table 2. The
plasmids pNAS7-1, which contains the N-terminal half of
GACI in the vector YEpl3 (11), and pST93, which contains
the full-length form of GAC1 in vector YEp352 (39), are
described elsewhere (34). The full-length GAC1 gene was
cloned into YEp351 by ligating the N-terminal half ofGACI
MATERIALS AND METHODS
Strains, media, and growth conditions. The genotypes of
the yeast strains used in this work are listed in Table 1. New
TABLE 2. Plasmids used in this study
Reference or source
2,m LEU2 shuttle vector
2i±m LEU2 shuttle vector
LEU2 CEN4 shuttle vector
CEN4 URA3 shuttle vector
2pm LEU2 shuttle vector
2tmHIS3GAL4,J147shuttle vector for dihybrid analysis
2>m LEU2GAL4768 m,,shuttle vector for dihybrid analysis
VOL. 14, 1994
SKROCH STUART ET AL.
(3.8-kb AatII-SstI fragment from pST93) into the AatII-SstI
sites of YEp351, yielding pJS-25. pJS-25 was further di-
gested with SphI-SstI, and the C-terminal half of GAC1
(SphI-SstI fragment of pST93) was inserted to create pJS-26.
To clone GAC1 into a low-copy-number plasmid, the 3.8-kb
BamHI-SalI GACI fragment from pJS-29 (EcoRV-EcoRI
GACI fragment in BSSK+) was ligated into pUN100 to form
pJS-30. The truncated form of GACI in a low-copy-number
plasmid, pJS-31, was generated by ligation of the 2.5-kb
fragment from pNAS7-1
BamHI site of pUN100. pJS-34, a 2,m plasmid that contains
the C-terminal truncation of GACI, was constructed by
cloning the 2.5-kb BglII-BamHI fragment from pNAS7-1
into the BamHI site of YEp351.
Epitope-tagged GLC7 and GAC1
quences encoding the epitopes recognized by the anti-Myc
monoclonal antibody 9E10 (MYC) (26) and the anti-hemag-
glutinin antibody 12CA5 (Ha) (64), EQKLISEEDLN and
YPYDVPYDYAT, respectively, were inserted at a site that
corresponds to the second amino acid of PP1. Both were
gifts from Kim Arndt. The Ha- and Myc-tagged GLC7 genes
were cloned as 2.5-kbXhoI-HindIII fragments into the yeast
shuttle vector YCp5O at the EcoRI site (58).
The Myc tag was introduced into GACI between the
second and third codon by using PCR (55) with one oligo-
nucleotide that contained the Myc tag flanked by GACI
T`ITCTGAAGACTTGATACAAACTGCT-3') and a second
taining GACI sequence. The PCR product was digested with
NcoI and SstI, and the 1.6-kb GACI fragment that contained
the Myc epitope was used to replace the NcoI-SstI fragment
of pJS-26 or pJS-34 to create the full-length Myc-tagged
GACI clone (pJS-33) or truncated GACI clone (pJS-36),
respectively. The constructions were confirmed by sequence
analysis. The full-length (pJS-38) and truncated (pJS-39)
versions of the Myc-tagged GACI were ligated into the
low-copy-number pUN100 vector as BamHI-SalI and SphI-
SstI fragments, respectively.
Since GLC7 is an essential gene, we introduced the
epitope-tagged GLC7 genes into yeast cells by transforming
a diploid strain (JS12) heterozygous for aglc7::HIS3 disrup-
tion with a YCpSO-based plasmid carrying either the Myc- or
the Ha-tagged GLC7 genes or the wild-type GLC7 gene. The
transformants were subjected to tetrad analysis, and spore
clones that contained theglc7::HIS3 gene disruption and one
of the GLC7-containing plasmids
pep4Al mutation was introduced into this background by
The glc7-1 mutation was transferred into the YCp5O-Ha-
GLC7 plasmid by gap repair (53). The YCp5O-Ha-GLC7
plasmid was digested with BglII and Sall and transformed
into aglc7-1 strain selecting for uracil prototrophy. Plasmids
were recovered from yeast cells, amplified in E. coli, and
transformed into yeast cells to test for the presence of the
epitope tag by immunoblot analysis and for the glc7-1
phenotype. The presence of the mutation known to be
responsible for the gkc7-1 phenotype (12) was confirmed by
Immunoprecipitation. Yeast cells were inoculated into 50
ml ofminimal medium at a density of 3 x 106 cells per ml and
incubated with shaking until the culture had reached a
density of 4 x 107 to 6 x 107 (11.5 h after inoculation).
Approximately 6.6 x 10' cells were centrifuged at 2,800 x g
in a RT-600B centrifuge (Sorvall) and resuspended in 10 ml
of minimal medium. Cells were added to flasks containing
genes. The DNA se-
were recovered. The
0.51 mCi of [35S]methionine (New England Nuclear) and
incubated with shaking for 1 h at 30°C. The cells were
harvested by centrifugation at 2,800 x g, washed with
breaking buffer (100 mM Tris, 200 mM NaCl, 1 mM EDTA,
5% glycerol [pH 7.0 to 7.1]), resuspended in 0.25 ml of
breaking buffer containing 1 mM phenylmethylsulfonyl flu-
oride and a 1/300 dilution of protease inhibitor cocktail (5 mg
[each] of chymostatin, leupeptin, antipain, and pepstatin
dissolved in a total of 20 ml of 50% ethanol in water and
stored at -20°C), and broken on a vortex mixer in the
presence of glass beads (Sigma G-9268). All steps in the
immunoprecipitation experiments after harvesting the cells
were performed either on ice or at 4°C. The broken cells
were centrifuged for 8 min at 1,500 x g in a RT-600B
centrifuge, and the supernatant was either used immediately
or frozen in liquid nitrogen and stored at -80°C. Ninety
microliters of the supernatant was mixed with 10 ,ul of
primary antibody solution (0.5 ,ul of ascites and 9.5 ,ul of
breaking buffer plus phenylmethylsulfonyl fluoride and pro-
tease inhibitor cocktail) and left on ice for 60 min. The
samples were centrifuged for 12 min at 11,000 x g at 4°C in
a microcentrifuge, and the supematants were transferred to
a microcentrifuge tube containing 30 plof protein G agarose
beads (Bethesda Research Laboratories) equilibrated in ra-
dioimmunoprecipitation assay buffer without sodium dode-
cyl sulfate (SDS) (50 mM Tris, 1% Triton X-100, 0.5%
Na-deoxycholate, 200 mM NaCl [pH 7.0 to 7.1]). The
solution was rocked on a shaker (Nutator) for 60 min at 4°C
and centrifuged at 9,000 x g in a microcentrifuge for 10 s to
pellet the beads, and the supernatant was removed and
discarded. The protein G beads were washed four times with
a mixture of 25% radioimmunoprecipitation assay buffer
without SDS-75% breaking buffer containing 1 mM phenyl-
methylsulfonyl fluoride and a 1/300 dilution of the protease
inhibitor cocktail and then once with 0.5 M Tris-0.5 M NaCl
(pH 7.0 to 7.1). The beads were resuspended in 30 ,ul of
protein gel loading dye (0.6 M Tris [pH 6.8], 10% glycerol,
2% SDS, 5% ,B-mercaptoethanol, 0.05% bromophenol blue),
boiled for 3 min, and electrophoresed on either 10% or 4 to
15% polyacrylamide-SDS gels (44). After electrophoresis,
the gels were fixed in 20% methanol-10% acetic acid and
soaked in Amplify fluorography solution (Amersham). The
dried gels were fluorographed with X-Omat R film (Kodak)
In vivo32Pilabeling. Approximately 6.6 x 108 cells at a
density of 4 x 107 to 6 x
centrifugation at 2,800 x g in a RT-600B centrifuge. Cells
were washed once with distilled water and then resuspended
in 10 ml of low-phosphate medium. For 500 ml of low-
phosphate medium, 3.3 g of yeast nitrogen base without
amino acids was dissolved in 450 ml of distilled water and 5
ml (each) of 1 M MgSO4 and concentrated NH40H were
added in a fume hood. After thorough mixing, the precipi-
tated phosphate was allowed to settle, and the medium was
then filtered through a 0.22-,um-pore-size cellulose acetate
filter. The pH of the medium was adjusted to 5.8 with
concentrated HCl, and glucose and appropriate amino acid
supplements were added before filter sterilization. Cultures
were incubated for 3 h with shaking at 30°C to deplete
endogenous phosphate stores before the addition of 0.5 mCi
of32Pi(Amersham). Cells were labeled for 1 h with shaking
at 30°C and then harvested for preparation of extracts as
Western immunoblot analysis. Cell extracts were prepared
and electrophoresed as described above. After electrophore-
sis, the gel was equilibrated in transfer buffer (25 mM Tris,
107/ml were harvested by
MOL. CELL. BIOL.
TYPE 1 PROTEIN PHOSPHATASE FROM S. CEREVISIAE
192 mM glycine, 20% methanol) and transferred to nitrocel-
lulose (Schleicher & Schuell) by using a Bio-Rad Mini
Trans-Blot apparatus as per the manufacturer's instructions.
The nitrocellulose filter was incubated for 1 h in TBS (20 mM
Tris, 500 mM NaCl[pH 7.5]) plus 5% nonfat milk (Carna-
tion) followed by incubation for 2 h in TBS plus 1% nonfat
milk and 0.3% Tween 20 containing a 1/2,000 dilution of the
primary antibody. The filter was washed twice for 30 min
TBS plus 3% Tween 20 and once for 30 min
incubated for 2 h in TBS plus 1% nonfat milk and 0.3%
Tween 20 containing a 1/1,000 dilution of a goat anti-mouse
peroxidase-conjugated secondary antibody (Sigma). The
blot was then washed as before, and bound antibody was
visualized by using the enhanced chemiluminescence detec-
tion system (Amersham).
In vitro dephosphorylation. Protein extracts labeled with
35S were immunoprecipitated as described above. Following
immunoprecipitation, beads were resuspended in 20 pl
phosphatase buffer (50 mM Tris-HCl [pH 8.0], 1mM MgCl2,
0.1 mM ZnC12). Reaction mixtures were then incubated with
or without the addition of 18 U of calf intestinal alkaline
phosphatase (Boehringer Mannheim) at 37°C for 15 min with
gentle shaking. Parallel reactions were carried out in phos-
phatase buffer containing two phosphatase inhibitors, 70mM
molybdic acid, and 16 mM PPi. Samples were separated by
SDS-polyacrylamide gel electrophoresis and visualized by
Dihybrid screen. The dihybrid screen for interacting pro-
teins was used as described by Fields and Song (29). The
GLC7 coding sequence was inserted into the BamHI site of
the shuttle vector pMA424 (46) to create an in-frame protein
fusion between the first 147 amino acid residues ofGal4p and
PP1. This was accomplished by PCR amplifying a HindIII
digest of plasmid YCp5O-GLC7 with primers designed to
produceBamHI sites (underlined) at the 5' (5'-CGGGATCC
GAATGGACTCAC-3') and 3' (5'-CGGGiATCCCTAGGAC
GTGAATC-3') ends of the gene. PCR was performed with
the Pfu DNA polymerase (Stratagene). Products were elec-
trophoresed on a 0.8% agarose gel and purified by using the
GeneClean II kit (Bio 101). The products were then digested
with BamHI and cloned into pUC18. The GLC7 BamHI
fragment was then cloned into the BamHI site of pMA424 to
give pCV5. Aglc7-1 fusion (pCV8) was created in a similar
manner, except the BamHI-digested PCR product from
pW13a was ligated directly into pMA424.
The GAC1 coding sequence was inserted into the BamHI
site of vector pGAD2F (28) to create a fusion between
residues 768 and 881 of Gal4p and Gaclp (pCV6). GAC1 was
amplified by PCR from plasmid pST93 by using primers
designed to introduce BamHI sites at the 5' (5'-CCGGGA
of the gene. The PCR product was digested withBamHI and
cloned into pUC18 and then into pGAD2F. Sequence anal-
ysis was used to confirm that all gene fusions were in the
correct frame and to confirm the presence of the glc7-1
mutation in pCV8.
Plasmids pCV6 and pCV5 or pCV8 were transformed into
the yeast strain GGY1::171 (36) selecting for histidine and
leucine prototrophy, respectively, and assaying for ,-galac-
tosidase activity as described for liquid cultures (52) or on
nitrocellulose filters (10). In cases in which only one fusion
plasmid was assayed, the complementary vector plasmid
(pMA424 or pGAD2F) was also included so that all trans-
formants had identical auxotrophies.
Analytical procedures. Protein concentrations were deter-
in TBS and then
E S E
FIG. 1. Immunoblot analysis of yeast strains expressing Myc
epitope-taggedPP1 andGaclp. Cell extracts were made from the
control strain DF11(lane 1), DF12 (myc-GLC7) (lane 2), DF11/
pJS-38 (low-copy-number myc-GACI) (lane 3), DF11/pJS-33 (high-
copy-number myc-GACJ) (lane 4), DF11/pJS-39 (low-copy-number
myc-GACI-AC) (lane 5),andDF11/pJS-36 (high-copy-numbermyc-
GACd-AC) (lane 6).
minedbytheprocedureof Bradford (9) with bovine serum
albumin as a standardby usinga kit from Bio-Rad. Qualita-
tiveglycogen assayswereperformed by using iodine vapors
aspreviouslydescribed (14). Quantitative glycogen assays
wereperformedas previously described (30, 31).
Epitope-tagged forms of PP1 and Gaclp are fuly func-
tional. The GLC7 and GACI gene products were epitope
tagged by insertingthe nucleotide sequences that encode the
epitopefor monoclonal antibodies 12CA5 and 9E10, which
recognizeanepitopein the influenza virus hemagglutinin
protein (64)and the human c-Myc protein (26), respectively,
after the second amino acid codon of each gene. The
wild-typeGLC7genewas replaced with the epitope-tagged
GLC7 constructs on a CEN vector, as described in the
Materials andMethods, and cell growth rates and glycogen
levels were compared. We found no difference between
yeaststrainscarryingthewild-typeGLC7 and those contain-
ingtheepitope-tagged geneswith regard to growth rate and
glycogenaccumulation. Strains carrying the epitope-tagged
versions ofGAC1,described in Table 1, showed an increase
inglycogenlevels similar to that observed for overexpres-
sion of the wild-type gene.
To examineexpressionof the epitope-taggedproteins, cell
extracts were compared by immunoblot analysis from
strainsharboringthe low-copy-number myc-GLC7 plasmid
or from strains harboringeither low- or high-copy-number
myc-GAC1 plasmids.As shown in Fig. 1, lane 2, a protein
with anapparentmolecular mass of 36 kDa from the Myc-
taggedGLC7 strain reacts with the 9E10 monoclonal anti-
body.This is close to the predicted molecular mass of 36.2
kDa for theMyc-PP1.Amajor protein band with an apparent
molecular mass of 110 to 120 kDa is found in strains bearing
thehigh-copy-number myc-GACJ plasmid (Fig. 1, lane 4) but
is not detectable in cells transformed with the low-copy-
numbermyc-GAC1 plasmid.The estimated molecular mass
VOL. 14, 1994
SKROCH STUART ET AL.
of this band is significantly greater than the 89.3-kDa pre-
dicted molecular mass deduced from the nucleotide se-
quence. Tang et al. (60) observed a similarly anomalous
electrophoretic mobility for the 124-kDa RGI subunit from
rabbit skeletal muscle which migrates as a 160-kDa band on
SDS-polyacrylamide gels. The more rapidly migrating bands
observed from strains that express myc-GAC1 apparently
correspond to proteolysis products. Their relative abun-
dance fluctuates from preparation to preparation, and all are
more apparent from cell extracts that have taken longer to
prepare. A pep4A&l deletion was introduced into our strain
background to help reduce proteolysis, and many of our
studies have been done withpep4 strains.
GAC1 was originally identified on a plasmid (pNAS7-1)
which harbored a GACI allele truncated at the C terminus of
Gaclp (34), referred to here as GACI-AC. This plasmid was
found to induce levels of glycogen accumulation as high or
higher than those of plasmids expressing the full-length
GAC1 gene. We have also introduced the Myc epitope into
GACJ-AC and expressed
CEN vectors and high-copy-number 2,um vectors. As in the
case of the full-length clone, the truncated form of GAC1 is
barely detectable when expressed from the CEN vector (Fig.
1, lane 5), but it is clearly detectable when expressed from
the 2p,mplasmids (Fig. 1, lane 6). We observed that like the
full-length Gaclp, the truncated form of Gaclp migrates with
an abnormally slow mobility on the SDS-polyacrylamide gel.
Although the predicted molecular mass
observed a reactive species at 80 kDa. The similar levels of
PP1 expressed from a low-copy-number CEN vector (Fig. 1,
lane 2) and Gaclp expressed from a high-copy-number 2,um
vector (Fig. 1, lane 4) suggest that PP1 is more abundant than
Gaclp under our assay conditions.
Gaclp is maximally expressed at the time that glycogen
begins to accumulate. To assess the levels of PP1 and Gaclp
during growth, a strain that expressed myc-GLC7 (JS28) was
transformed with either a high-copy-number plasmid con-
taining myc-GACI (pJS-33) or a vector control (pAAH5).
Transformants were inoculated into minimal medium lacking
uracil and leucine and incubated at 30°C. Cell number,
glycogen concentration, and levels of Myc-PP1 and Myc-
Gaclp were assayed during the growth curve. As shown in
Fig. 2A, a representative transformant containing the high-
copy-number myc-GACl plasmid begins to accumulate gly-
cogen 8.5 h after inoculation. Peak levels of Gaclp are also
found in the 8.5-h sample (Fig. 2B). This is consistent with
the observation that GACI RNA and glycogen begin to
accumulate at the same point in the growth curve (34).
Gaclp is most abundant as cells begin to accumulate glyco-
gen and then decreases in abundance as cells enter station-
ary phase. This observation is reproducible, but we cannot
eliminate the possibility that the decrease in Gaclp levels is
an artifact due to proteolysis. In contrast to Gaclp, the
levels of the 36-kDa PP1 appear to be nearly constant
throughout the growth curve and are not influenced signifi-
cantly by a high dosage of Gaclp.
The glycogen defect ofglc7-1 can be suppressed by increased
expression of GACI. Unlike a null allele ofglc7ldis2s2, which
is haplolethal (16, 27), glc7-1, an Arg-73-to-Cys missense
mutation (12), results in relatively minor defects. glc7-1
mutants fail to accumulate glycogen but grow with normal
doubling times in our genetic background. Glycogen syn-
thase is largely in the inactive, phosphorylated form inglc7-1
mutants (27, 34), which may account for the failure ofglc7-1
mutants to accumulate glycogen. If the glycogen accumula-
tion deficiency of glc7-1 strains is due to the failure of
it from both low-copy-number
is 61 kDa, we
Hours after inoculation
FIG. 2. Gaclp and PP1 expression during growth in batch cul-
ture. A myc-GLC7 gacl::ura3 strain (JS28) was transformed with
2pLm LEU2 plasmids that contain (pJS-33) or lack (pAAH5) the
myc-GAC1 gene. (A) Transformants were inoculated into synthetic
medium lacking uracil and leucine, and cell number (0, 0) and
glycogen concentration (A, A) were determined at the stated times.
(B) Gaclp and PP1 were visualized by immunoblot analysis with cell
extracts prepared from each time point.
PP1g'c7-1 to interact productively with its regulatory subunit,
we might expect that increased levels of the regulatory
subunit could suppress the glycogen deficiency. We tested
this possibility by assaying glycogen levels in glc7-1 strains
transformed with high- and low-copy-number GACJ-ex-
pressing plasmids and found that the glycogen accumulation
of the glc7-1
mutant can be dramatically increased by
overexpression of GACI (Table 3). Whereas CEN plasmids
expressing GACI (pJS-30) have little influence on glycogen
levels, expression of GACI from a high-copy-number 2,um
plasmid (pJS-26) increases glycogen levels. We also note
that transformants containing Gacl-ACp accumulated signif-
icantly higher levels of glycogen than those expressing the
MOL. CELL. BIOL.
TYPE 1 PROTEIN PHOSPHATASE FROM S. CEREVISUAE
TABLE 3. Suppression ofglc7-1 by increased
expression of GAC1P
glucose /107 celis)
KT1119-pNAS7-1gkc7-1, high-copy-number GACJAC
GLC7, high-copy-number GACIAC
0.34 ± 0.1
0.48 ± 0.01
8.2 ± 0.1
8.4 ± 0.2
16.4 ± 0.1
33.0 ± 1.0
glc7-1, low-copy-number GAC1
glc7-1, high-copy-number GAC1
gkc7-1, low-copy-number GACIAC
aGlycogen determinations were performed on cells grown in minimal
medium 12.5 h after inoculation.
b Glycogen assayresults are theaveragesof at least twoexperiments.
Feng et al. (27) observed that aglc7-1 strain had one-third
of the wild-type levels of PP1 activity when assayed in vitro.
It is therefore possible that Gaclp might stimulate glycogen
synthesis by increasing the stability or expression of PP1. To
eliminate this possibility, we assayed the levels of Ha-
PPlglc7-' by immunoblot analysis in a strain expressing high
levels of Myc-Gaclp. As shown in Fig. 3A, increased
expression ofmyc-GAC1 suppresses the glycogen deficiency
ofglc7-1, but it does not influence the level of PPlglc7-1 (Fig.
3B,lowerpanel).The levels ofPPlgk7-1 are similar in control
and GACl-overproducing strains.
Coimmunopmcipitation ofGaclp and PP1. An immunopre-
cipitation assay was developed to demonstrate a direct
interaction between PP1 and Gaclp. Strains containing
Ha-PP1 and Myc-Gaclp were labeled in vivo with [35S]me-
thionine, cell extracts were prepared, and Ha-PP1 or Myc-
Gaclp was precipitated with the 12CA5 or 9E10 antibody,
respectively. The precipitates were electrophoresed on SDS-
polyacrylamide gels and autoradiographed. As shown in Fig.
4, lanes 2 and 4, the 12CA5 antibody precipitates a 36-kDa
protein with the predicted molecular mass of Ha-PP1. The
9E10 monoclonal antibody specifically precipitates two pro-
teins of approximately 110 kDawith the same mobility as the
Gaclp observed by immunoblot analysis (Fig. 4, lane 3). The
same Gaclp bands are coprecipitated with the Ha-specific
antibody (lane 4). These proteins do not coprecipitate with
the 12CA5 antibody unless Ha-PP1 is expressed in the strain
(data not shown). These results suggest that a significant
fraction of Gaclp in our cell extract is associated with PP1
and coprecipitates in our assay.
Although a clear and reproducible Gaclp-PP1 association
is observed when PP1 is immunoprecipitated, we do not
consistently observe coimmunoprecipitation of PP1 with
antibodies directed against Myc-Gaclp. For example, some
PP1 might have been expected to be present in Fig. 4, lane 3.
Although this was observed in some experiments (for exam-
ple, some PP1 is observed to coprecipitate with Gaclp in
Fig. 6, lane 3), it has not been reproducible. Since the only
epitope for the anti-Myc antibody is at the N terminus of
Gaclp, it is possible that the Gaclp-antibody complex pre-
cludes the binding ofPP1. The irreproducibility could also be
simply due to the low levels of Gaclp present in the cell.
Immunoprecipitation experiments were also performed
with strains containing Ha-glc7-1 and myc-GAC1 (Fig. 4,
lanes 5 to 8). Two interesting and reproducible observations
are evident from these experiments. First, little or no Gaclp
coprecipitates with Ha-PPl91c7
Myc-GAC1 protein expressed in glc7-1 stains has an elec-
trophoretic mobility different from that of GLC7 strains
(Fig. 4, lane 6). Second, the
Hours after inoculation
- ,: A
FIG. 3. Gaclp and PP1 expression in a glc7-1 strain. An Ha-
glc7-1 strain (JS94) was transformed with 2,um LEU2 plasmids that
contain (pJS-33) or lack (pAAH5) the myc-GAC) gene. (A) Trans-
formants were inoculated into synthetic medium lacking uracil and
leucine, and cell number (0, 0) and glycogen concentration (A, A)
were determined at the stated times. (B) Immunoblots with the
anti-Myc (upper panel) or anti-Ha antibody (lower panel) were
prepared from cell extracts for each time point to visualize Gaclp
and PP1, respectively.
(compare lanes 3 and 4 with lane 5). Gaclp from GLC7
strains migrates as a doublet, but Gaclp from glc7-1 strains
migrates as a single band with lower mobility. This differ-
ence is always observed inglc7-1 strains, irrespective of the
epitope tag inglc7-1 or the presence of thepep4 mutation in
the strain background (data not shown).
Gaclp is phosphorylated in vivo. TheRGIsubunit of PP1
from skeletal muscle is phosphorylated in vivo in response to
adrenaline (21) and insulin (22). Protein phosphorylation is
therefore an obvious possible explanation for the changes
observed in electrophoretic mobility of Gaclp (Fig. 4). The
more slowly migrating Gaclp band from the GLC7 strain
would be predicted to be a phosphorylated form. To test this
hypothesis, we treated 35S-labeled immunoprecipitates like
VOL. 14, 1994
SKROCH STUART ET AL.
'myc Halrmyc Ha'1myc Halrmyc Ha'
FIG. 4. Coimmunoprecipitation of Gaclp and PP1. Yeast strains
were labeled with [35S]methionine, cell extracts were prepared, and
immunoprecipitates were electrophoresed on a 10% polyacrylamide
gel as described in Materials and Methods. The autoradiograph of
the dried gel is presented. The relevant genotype of each strain and
the antibody used in the immunoprecipitation are listed above each
lane. The specific strains used are as follows: lanes 1 and 2,
DF11/pAAH5; lanes 3 and 4, DF11/pJS-33; lanes 5 and 6, DF2/pJS-
33; and lanes 7 and 8, DF2/YEp351.
the ones used for Fig. 4, lanes 3 and 5, with alkaline
phosphatase prior to electrophoresis. As shown in Fig. 5,
lane 2, alkaline phosphatase converts the Gaclp doublet into
a single band with the more rapid electrophoretic mobility.
Treatment of immunoprecipitated Gaclp with phosphatase
also reduces the mobility ofGaclp expressed from theglc7-1
strain (Fig. 5, lane 5). These results are consistent with the
hypothesis that the upper Gaclp band is phosphorylated in
vivo and that Gaclp is totally in the phosphorylated form in
To provide additional evidence for in vivo phosphoryla-
tion of Gaclp, we labeled the same strains used in Fig. 4 with
32Piand immunoprecipitated PP1 and Gaclp with 12CA5 and
9E10 antibodies, respectively, and electrophoresed the im-
FIG. 5. Alkaline phosphatase treatment of Gaclp-containing ex-
tracts. Yeast strain DF11/pJS-33, a Ha-GLC7 strain that harbors a
high-copy-number myc-GAC1 plasmid, or DF2/pJS-33, a Ha-glc7-1
strain harboring the same plasmid, was methionine labeled, and
immunoprecipitates were treated with alkaline phosphatase (lanes 2
and 5) or with alkaline phosphatase and the phosphatase inhibitors
molybdic acid and PPi (lanes 3 and 6) prior to electrophoresis and
FIG. 6. 32p incorporation into Gaclp. Yeast strains were 32p
labeled, cell extracts were prepared, and immunoprecipitates were
electrophoresed on a 4 to 15% polyacrylamide gel as described in
Materials and Methods. Proteins were transferred to nitrocellulose,
32p incorporation was detected by autoradiography (upper panel),
and Gaclp and PP1 were detected immunologically with the 12CA5
and 9E10 antibodies (lower panel). The relevant genotypes of each
strain and antibody used in the immunoprecipitation are listed above
each lane. The specific strains used are as follows: lanes 1 and 2,
DF11-pAAH5; lanes 3 and 5, DF11-pJS-33; lanes 4 and 6, DF2-
pJS33; and lanes 7 and 8, DF2-YEp351. The upper and lower arrow
to the right of each panel correspond to the electrophoretic mobili-
ties of Gaclp and PP1, respectively.
munoprecipitates on SDS-polyacrylamide gels. PP1 and
Gacl proteins were identified by immunoblot analysis (Fig.
6, lower panel), and 32P-labeled proteins were identified by
autoradiography (Fig. 6, upper panel). Gaclp is phosphory-
lated in both the GLC7 (Fig. 6, lane 3) andglc7-1 strains (Fig.
6, lane 4), but no 32p incorporation is found associated with
PP1 (Fig. 6, lanes 1, 5, 6, and 7).
Gaclp and PP1 interact in the dihybrid assay. Fields and
Song (29) have developed an in vivo assay that utilizes GAL4
fusion proteins to detect interacting proteins. Gal4p is a
transcriptional activator whose DNA-binding domain re-
sides at its N terminus and whose trans-activating domain
resides in its C terminus. Both domains are required for
trans activation of a GAL4-dependent promoter, but they
need not be linked in cis. To assay for the association of two
proteins, the N-terminal DNA-binding domain of Gal4p is
fused to one protein and the C-terminal trans activation
domain of Gal4p is fused to another. If the two proteins
physically interact in vivo, their association brings the two
halves of Gal4 into proximity, and the GAL4-dependent
promoter is activated. This association is monitored by
assaying p-galactosidase expressed from a GALl-lacZ gene
fusion. We constructed a plasmid containing the DNA-
binding domain of Gal4 (Gal41_147) fused to the GLC7 gene
product (pCV5) and a separate plasmid with the transacti-
vating domain of Gal4 (Gal4768-881)fused to GACI (pCV6).
GGY1::171 (36), and ,B-galactosidase activitywas assayed.
for these experiments were
GAL41-147-SNF1 and GAL4768-881-SNF4 plasmids origi-
tHa MYC myC myc Ha Ha Ha myc
MOL. CELL. BIOL.
TYPE 1 PROTEIN PHOSPHATASE FROM S. CEREVISUE
FIG. 7. P-Galactosidaseactivities from strains expressing Gal4p-protein fusions. Yeast strain GGY::171 was transformed with pCV5 and
pGAD2F to express only the GAL4-147-GLC7 fusion; with pCV6 and pMA424, to expressGA1A76&8_81-GAC1;with both pCV5 and pCV6,
to express GALA14147-GLC7 andGAL4.A6&M1-GAC1;with pCV8 and pGAD2F, to express GAL4A1147-glc7-1; with pCV8 and pCV6, to
express GAL41-147-glc7-1 and GAL47681,,-GALl;and with GAL4(1-147-SNF1) and SNF4-GALA(768-881). ,B-Galactosidase activity was
assayed in triplicate for three independent transformants harboring each set of plasmids.
nally described (29). As shown in Fig. 7, cells transformed
with either GAL41 147-GLC7 orGALA76,8,,-GAC1alone
express low levels of ,B-galactosidase activity (1 to 10 U), but
cells transformed with both plasmids express high levels of
3-galactosidase (200 to 900 U). Levels of ,B-galactosidase
activity from the GLC7 and GAC1 transformants were as
high or higher than those from the SNF1 and SNF4 positive
The dihybrid system was next used to assay for pro-
tein-protein interaction between Gaclp and PPgl9c7-l. A
GAL4-1147-glc7-1 plasmid (pCV8) was constructed as out-
lined in Materials and Methods and tested for activation
in combination withGAL476888,-GAC1(pCV6). As shown
in Fig. 7, strains transformed with both GAL41147-glc7-1
and GAL4768-881-GAC1 produced only marginally higher
levels of ,-galactosidase activity than strains transformed
with only one of the two plasmids, suggesting that the
interaction between PPlg&k7
and Gaclp is very weak com-
pared with the interaction between the wild-type PP1 and
An alternative explanation for the low level of biological
activity from the glc7-1 fusion plasmid is that the protein is
unstable or not adequately expressed. To test this possibil-
ity, we asked whether either theGAL4j,147-glc7-1(pCV8)or
GAL4,_147-GLC7(pCV5)fusion plasmids could perform the
essential functions of GLC7. A haploid strain (KT1132)
containing a lethal glc7::LEU2 deletion but rescued by a
wild-type GLC7-containing plasmid was transformed with
pCV5 or pCV8. Transformants were plated on medium
containing 5-fluoroorotic acid (5-FOA), which selects for
loss of URA3 activity (6). Since the URA3 gene is encoded
on the GLC7-containing plasmid, growth on 5-FOA indicates
that the GLC7-containing plasmid is not required to maintain
cell viability. The untransformed strain or transformants
harboring the a GAL4-GLC7 fusion vector with GLC7 in the
opposite orientation could not grow in the presence of
5-FOA, but transformants harboring either the pCV5 or the
pCV8 plasmid were able to grow on 5-FOA. Cells trans-
formed with either pCV5 or pCV8 were recovered from the
5-FOA media and were shown to have lost the URA3-
containing GLC7 plasmid and to stably maintain either the
HIS3-containing pCV5 or the HIS3-containing pCV8 (data
not shown). These results indicate that both the GAL41-147-
glc7-1 and GAL4114r-GLC7 fusion proteins are able to
carry out all the essential biological roles of GLC7, but they
do not exclude the possibility that GAL41-14-glc7-1
present at lower concentrations in the cell. However, we
believe this is unlikely, since PP1 and PP1g9c7-1 appear to be
expressed at similar levels.
The GACI gene product plays a key regulatory role in
glycogen synthesis. Under our assay conditions, the level of
Gaclp directly correlates with the level of glycogen that
accumulates as cells approach stationary phase. Increased
dosage of GACI results in increased levels of Gaclp and a
corresponding increase in glycogen, whereas gacl null mu-
tants fail to accumulate glycogen. Francoiset al. (34) noted
that GACl RNAwas first observed at the point in the growth
curve that glycogen began to accumulate, and similar obser-
vations have been made for other glycogen metabolic genes
(34, 54, 61). We note that Gaclp shows the same timing of
accumulation and that increased dosage of GACJ does not
alter this timing. We have shown that Gaclp and PP1
interact in vitro and in vivo and that the glycogen accumu-
lation deficiency of a gk7-1 mutant can be overcome by
increased expression of Gaclp. Taken together, the results
presented here strongly support the hypothesis that Gaclp
affects glycogen synthesis by acting as a regulatory subunit
What is the role of PP1 and its associated regulatory
subunit? Glycogen synthase from glc7-1 and gacl strains is
mostly in the inactive phosphorylated form (34, 50), arguing
that PP1 and/or Gaclp directly or indirectly dephosphory-
lates glycogen synthase. Hardy and Roach (38) have isolated
mutant forms of glycogen synthase that lack key sites of
phosphorylation. These mutant alleles of GSY2 suppress the
VOL. 14, 1994
SKROCH STUART ET AL.
glycogen-deficient phenotype of both gacl and gkc7-1, sug-
gesting that the GLC7-encoded phosphatase controls the
phosphorylation state ofglycogen synthase. As in the case of
the skeletal muscle form of PP1, Gaclp could recruit PP1 to
the site of glycogen synthesis and/or alter its activity toward
The glycogen accumulation deficiency of glc7-1 can be
explained by proposing that the mutant PP1gkc7-1 phos-
phatase is unable to interact with the glycogen-specific
regulatory subunit encoded by GACI. This loss of interac-
tion may result in the incorrect targeting of PP1 or the
inability to act on its most likely substrate, glycogen syn-
thase. However, this hypothesis may be an oversimplifica-
tion. If the glycogen deficiency ofglc7-1 is due solely to its
inability to bind Gaclp, one would expect that glc7-1 and
gacl strains would have identical defects in glycogen metab-
olism. However, the glycogen defect in a glc7-1 mutant
strain is more severe than in a gacl deletion mutant,
although this is not apparent when comparing Fig. 2 and 3.
gacl strains will accumulate significantly more glycogen
than glc7-1 strains under some starvation conditions (57).
This difference in phenotype could be explained if PPl9&c"
had an additional defect, such as reduced catalytic activity
towards glycogen synthase, or an inability to interact with
additional regulatory subunits that also have a role in glyco-
gen metabolism. This latter hypothesis would also explain
the pleiotropic phenotype of glc7-1. In addition to the
glycogen defect, glc7-1 results in a partial sporulation defect
(12) and suppresses defects caused by mutations in the
translation initiation factor 2 kinase, GCN2 (63). It is possi-
ble that these traits are caused by the failure of PP1glc7-l to
interact with other regulatory subunits that have roles in
sporulation or translational control.
As in the case of the skeletal muscle regulatory subunit
RGI, Gaclp is phosphorylated in vivo. Phosphorylation of
RG1 by cyclic AMP-dependent protein kinase causes disso-
ciation of catalytic and regulatory subunits (41), whereas
phosphorylation by an insulin-dependent protein kinase in-
creases the activity of the phosphatase towards glycogen
synthase and phosphorylase kinase (22). We do not yet know
the biochemical significance of Gaclp phosphorylation, but
we do note in Fig. 4, comparing lane 3 with lane 4, that the
ratio ofmore slowly to faster migrating Gaclp bands is lower
in Gaclp that coprecipitates with PP1 than in the total
Gaclp. This suggests that PP1 might preferentially associate
with the non- or underphosphorylated form of Gaclp. It will
be of interest to determine whether physiological signals
which modulate the accumulation of glycogen, such as
nutrient starvation (45) or mating pheromone treatment (31),
alter the phosphorylation state of Gaclp. Finally, all detect-
able Gaclp is phosphorylated in a gkc7-1 background. This
suggests that Gaclp could be a better protein kinase sub-
strate in the presence of PP1glc7-1
wild-type PP1, when tight association between PP1 and
Gaclp might prevent phosphorylation. Altemnatively, Gaclp
could be hyperphosphorylated
Gaclp is a substrate for PP1 and PP1g'c7-1 is defective in this
than in the presence of
in glc7-1 strains because
We thank John Cannon, Fong Chyr Lin, Kim Arndt, Peter Roach,
Tom Hardy, Stanley Fields, and Jens Peterson for strains and
reagents; Wendy Gilbert for technical assistance; and Lucy Robin-
son, Tim Petty, and an anonymous reviewer for critically reading
C.M.V. and D.L.F. were supported by a NSF predoctoral train-
eeship and a Patricia Roberts Harris fellowship, respectively. This
work was supported by Public Health Service grant GM47789.
1. Alberts, A. S., A. M. Thorburn, S. Shenolikar, M. C. Mumby,
and J. R. Feramisco. 1993. Regulation of cell cycle progression
and nuclear affinity of the retinoblastoma protein by protein
phosphatases. Proc. Natl. Acad. Sci. USA 90:388-392.
2. Alexander, D. R. 1990. The role of phosphatases in signal
transduction. New Biol. 2:1049-1062.
3. Ammerer, G. 1983. Expression ofgenes in yeast using theADCl
promoter. Methods Enzymol. 101:192-201.
4. Axton, J. M., V. Dombradi, P. T. W. Cohen, and D. M. Glover.
1990. One of the protein phosphatase 1 isoenzymes in Dros-
ophila is essential for mitosis. Cell 63:33-46.
5. Berndt, N., D. G. Campbell, F. B. Caudwell, P. Cohen, E. F. da
Cruz e Silva, 0. B. da Cruz e Silva, and P. T. W. Cohen. 1987.
Isolation and sequence analysis of a cDNA clone encoding a
type-1 protein phosphatase catalytic subunit: homology with
protein phosphatase 2A. FEBS Lett. 223:340-346.
6. Boeke, J., F. Lacroute, and G. R. Fink. 1984. A positive
selection for mutants lacking orotidine-5'-phosphate decarbox-
ylase activity in yeast: 5-fluoroorotic acid resistance. Mol. Gen.
7. Bollen, M., and W. Stalmans. 1992. The structure, role, and
regulation of type 1 protein phosphatases. Crit. Rev. Biochem.
Mol. Biol. 27:227-281.
8. Booher, R., and D. Beach. 1989. Involvement of a type 1 protein
phosphatase encoded by bwsl+ in fission yeast mitotic control.
9. Bradford, M. M. 1976. Rapid and sensitive method for quanti-
tation of microgram quantities of protein utilizing the principle
of protein-dye binding. Anal. Biochem. 72:248-254.
10. Breeden, L., and K. Nasmyth. 1985. Regulation of the yeast HO
gene. Cold Spring Harbor Symp. Quant. Biol. 50:1097-1105.
11. Broach, J. R., J. N. Strathern, and J. B. Hicks. 1979. Transfor-
mation in yeast: development of a hybrid cloning vector and
isolation of the CAN1 gene. Gene 8:121-133.
12. Cannon, J. F., J. R. Pringle, A. Fiechter, and M. Khalil.
Characterization of glycogen-deficient glc mutants of Saccha-
romyces cerevisiae. Genetics, in press.
13. Cannon, J. F., and K. Tatchell. 1987. Characterization of
Saccharomyces cerevisiae genes encoding subunits of cyclic
AMP-dependent protein kinase. Mol. Cell. Biol. 7:2653-2663.
14. Chester, V. E. 1968. Heritable glycogen-storage deficiency in
yeast and its induction by ultra-violet light. J. Gen. Microbiol.
15. Chisholm, A. A. K., and P. Cohen. 1988. The myosin-bound
form of protein phosphatase 1 (PP-1M) is the enzyme that
dephosphorylates native myosin in skeletal and cardiac mus-
cles. Biochim. Biophys. Acta 971:163-169.
16. Clotet, J., F. Posas, A. Casamayor, I. Schaaff-Gerstenshlager,
and J. Arifio. 1991. The gene DIS2S1 is essential in Saccharo-
myces cerevisiae and is involved in glycogen phosphorylase
activation. Curr. Genet. 19:339-342.
17. Cohen, P. 1989. The structure and regulation of protein phos-
phatases. Annu. Rev. Biochem. 58:453-508.
18. Cohen, P., and P. T. W. Cohen. 1989. Protein phosphatases
come of age. J. Biol. Chem. 264:21435-21438.
19. Cohen, P. T. W. 1988. Two isoforms of protein phosphatase 1
may be produced from the same gene. FEBS Lett. 232:17-23.
20. Cyert, M., and J. Thorner. 1991. Putting it on and taking it off:
phosphoprotein phosphatase involvement in cell cycle regula-
tion. Cell 57:891-893.
21. Dent, P., D. G. Campbell, F. B. Caudwell, and P. Cohen. 1990.
Identification of three in vivo phosphorylation sites on the
glycogen binding subunit of protein phosphatase 1 from rabbit
skeletal muscle, and their response to adrenaline. FEBS Lett.
22. Dent, P., A. Lavoinne, S. Nakielny, F. B. Caudwell, P. Watt, and
P. Cohen. 1990. The molecular mechanism by which insulin
stimulates glycogen synthesis in mammalian skeletal muscle.
MOL. CELL. BIOL.
TYPE 1 PROTEIN PHOSPHATASE FROM S. CEREVISL4E Download full-text
Nature (London) 348:302-308.
23. Dombradi, V., J. M. Axton, N. D. Brewis, E. F. da Cruz e Silva,
L. Alphey, and P. T. W. Cohen. 1990. Drosophila contains three
genes that encode distinct isoforms of protein phosphatase 1.
Eur. J. Biochem. 194:739-745.
24. Doonan, J. H., and N. R. Morris. 1989. The bimG gene of
Aspergillus nidulans, required for completion of anaphase,
encodes a homolog of mammalian phosphoprotein phosphatase
1. Cell 57:987-996.
25. Elledge, S. J., and R. W. Davis. 1988. A family of versatile
centromeric vectors designed for use in the sectoring-shuffle
mutagenesis assay in Saccharomyces cerevisiae. Gene 70:303-
26. Evan, G., G. Lewis, G. Ramsay, and J. M. Bishop. 1985.
Isolation of monoclonal antibodies specific for human c-myc
proto-oncogene product. Mol. Cell. Biol. 5:3610-3616.
27. Feng, Z., S. E. Wilson, Z.-Y. Peng, K. K. Schlender, E. M.
Reimann, and R. J. Trumbly. 1991. The yeast GLC7 gene
required for glycogen accumulation encodes a type 1 protein
phosphatase. J. Biol. Chem. 266:23796-23801.
28. Fields, S. Personal communication.
29. Fields, S., and O.-k. Song. 1989. A novel genetic system to
detect protein-protein interactions. Nature (London) 340:245-
30. Fran,ois, J., P. Eraso, and C. Gancedo. 1987. Changes in the
concentration of cAMP, fructose 2,6-bisphosphate and related
metabolites and enzymes in Saccharomyces cerevisiae during
growth on glucose. Eur. J. Biochem. 164:369-373.
31. Francois, J., D. L. Higgins, F. Chang, and K. Tatchell. 1991.
Inhibition of glycogen synthesis in Saccharomyces cerevisiae
by the mating pheromone a-factor. J. Biol. Chem. 266:6174-
32. Francois, J., E. Van Schaftingen, and H.-G. Hers. 1988. Char-
acterization of phosphofructokinase 2 and of enzymes involved
in the degradation of fructose 2,6-bisphosphate in yeast. Eur. J.
33. Francois, J., M. E. Villanueva, and H.-G. Hers. 1988. The
control of glycogen metabolism in yeast. 1. Interconversion in
vivo of glycogen synthase and glycogen phosphorylase induced
by glucose, a nitrogen source, or uncouplers. Eur. J. Biochem.
34.Franqois,J. M., S. Thompson-Jaeger, J. Skroch, U. Zellenka, W.
Spevak, and K. Tatchell. 1992. GACI may encode a regulatory
subunit for protein phosphatase type 1 in Saccharomyces cere-
visiae. EMBO J. 11:87-96.
35. Gietz, D., A. St. Jean, R. A. Woods, and R. H. Schiestl. 1992.
Improved method for high efficiency transformation of intact
yeast cells. Nucleic Acids Res. 20:1425.
36. Gill, G., and M. Ptashne. 1987. Mutants ofGAL4 protein altered
in an activation function. Cell 51:121-126.
37. Hanks, S. K., A. M. Quinn, and T. Hunter. 1988. The protein
kinase family: conserved features and deduced phylogeny of the
catalytic domains. Science 241:42-52.
38. Hardy, T., and P. Roach. 1993. Control of yeast glycogen
synthase-2 by COOH-terminal phosphorylation. J. Biol. Chem.
39. Hill, J. E., A. M. Myers, T. J. Koerner, and A. Tzagoloff. 1986.
Yeast/E. coli shuttle vectors with multiple unique restriction
sites. Yeast 2:163-167.
40. Huang, K. P., and E. Cabib. 1974. Yeast glycogen synthetase in
the glucose 6-phosphate-dependent form. J. Biol. Chem. 249:
41. Hubbard, M., and P. Cohen. 1989. Regulation of protein phos-
phatase-1 from rabbit skeletal muscle. 1. Phosphorylation by
cAMP-dependent protein kinase at site 2 releases catalytic
subunit from glycogen-bound holenzyme. Eur. J. Biochem.
42. Ito, H., Y. Fukuda, K. Murata, and A. Kimura. 1983. Transfor-
mation of intact yeast cells treated with alkali cations. J.
43. Kinoshita, N., H. Ohkura, and M. Yanagida. 1990. Distinct,
essential roles of typel and 2A protein phosphatases in the
control of the fission yeast cell division cycle. Cell 63:405-415.
44. Laemmli, U. K. 1970. Cleavage of structural proteins during the
assembly of the head of bacteriophage T4. Nature (London)
45. Lillie, S. H., and J. R. Pringle. 1980. Reserve carbohydrate
metabolism in Saccharomyces cerevisiae: responses to nutrient
limitation. J. Bacteriol. 143:1384-1394.
46. Ma, J., and M. Ptashne. 1987. A new class of yeast transcrip-
tional activators. Cell 51:113-119.
47. Nitschke, K., U. Fleig, J. Schell, and K. Palme. 1992. Comple-
mentation of the cs dis2-11 cell cycle mutant of Schizosaccha-
romyces pombe by a protein phosphatase from Arabidopsis
thaliana. EMBO J. 11:1327-1333.
48. Ohkura, H., N. Kinoshita,
Yanagida. 1989. The fission yeast dis2+ gene required for
chromosome disjoining encodes one of two putative type 1
protein phosphatases. Cell 57:997-1007.
49. Peng, Z.-Y., R. J. Trumbly, and E. M. Reimann. 1990. Purifi-
cation and characterization of glycogen synthase from a glyco-
strain of Saccharomyces cerevisiae.
50. Peng, Z.-Y., W. Wang, S. E. Wilson, K. K. Schlender, R. J.
Trumbly, and E. M. Reimann. 1991. Identification of a glycogen
synthase phosphatase from yeast Saccharomyces cerevisiae as
protein phosphatase 2A. J. Biol. Chem. 266:10925-10932.
51. Rose, M. D., P. Novick, J. H. Thomas, D. Botstein, and G. R.
Fink. 1987. A Saccharomyces cerevisiae genomic plasmid bank
based on a centromere-containing shuttle vector. Gene 60:237-
52. Rose, M. D., F. Winston, and P. Hieter. 1990. Methods in yeast
genetics: a laboratory course manual. Cold Spring Harbor
Laboratory Press, Plainview, N.Y.
53. Rothstein, R. 1983. One-step gene disruption in yeast. Methods
54. Rowen, D. W., M. Meinke, and D. C. LaPorte. 1992. GLC3 and
of Saccharomyces cerevisiae are allelic and encode the
glycogen branching enzyme. Mol. Cell. Biol. 12:22-29.
55. Saiki, R. K., D. H. Gelfand, S. Stoffel, S. J. Scharf, R. Higuchi,
G. T. Horn, K. B. Mullis, and H. A. Erlich. 1988. Primer-
directed enzymatic amplification of DNA with a thermostable
DNA polymerase. Science 239:487-491.
56. Smith, R. D., and J. C. Walker. 1991. Isolation and expression
of a maize type 1 protein phosphatase. Plant Physiol. 97:677-
57. Street, L., and K. Tatchell. Unpublished data.
58. Sutton, A., F. Lin, M. J. F. Sarabia, and K. T. Arndt. 1991. The
SIT4 protein phosphatase is required in late G1 for progression
into S phase. Cold Spring Harbor Symp. Quant. Biol. 56:75-81.
59. Tamura, S., K. R. Lynch, J. Larner, J. Fox, A. Yasui, K.
Kikuchi, Y. Suzuki, and S. Tsuiki. 1988. Molecular cloning of rat
type 2C (IA) protein phosphatase mRNA. Proc. Natl. Acad.
Sci. USA 82:3785-3789.
60. Tang, P., J. Bandor, K. Swiderek, and A. DePaoli-Roach. 1991.
Molecular cloning and expression of the regulatory(RGI)subunit
of the glycogen-associated protein phosphatase. J. Biol. Chem.
61. Thon, V. J., C. Vigneron-Lesens, T. Marianne-Pepin, J. Mon.
treuil, A. Decq, C. Rachez, S. G. Ball, and J. F. Cannon. 1992.
Coordinate regulation of glycogen metabolism in the yeast
Saccharomyces cerevisiae. J. Biol. Chem. 267:15224-15228.
62. Walton, K. M., and J. E. Dixon. 1993. Protein tyrosine phos-
phatases. Annu. Rev. Biochem. 62:101-120.
63. Wek, R. C., J. F. Cannon, T. E. Dever, and A. G. Hinnebusch.
1992. Truncated protein phosphatase GLC7 restores transla-
tional activation ofGCN4 expression in yeast mutants defective
for the eIF-2a kinase GCN2. Mol. Cell. Biol. 12:5700-5710.
64. Wilson, I. A., H. L. Niman, R. A. Houghten, A. R. Cherenson,
M. L. Connolly, and R. A. Lerner. 1984. The structure of an
antigenic determinant in a protein. Cell 37:767-778.
S. Miyatani, T. Toda, and M.
VOL. 14, 1994