The Cellular Proteins Which Can Associate Specifically with
Polyomavirus Middle
T
Antigen in
Human 293
Cells Include
the
Major Human
70-Kilodalton Heat Shock Proteins
DAVIDC. PALLAS,* WILLIAM MORGAN, ANDTHOMAS M. ROBERTS
Dana-Farber CancerInstituteandDepartmentof Pathology, Harvard Medical School, Boston, Massachusetts 02115
Received 24April 1989/Accepted22July1989
We compared the proteins which associate with middle T antigen (MT) of polyomavirus in human cells
infected with Ad5(pymT), a recombinant adenovirus which directs the overexpression of MT, with the
MT-associated proteins (MTAPs) previouslyidentified in murine fibroblasts expressing MT. MTAPs of 27, 29,
36, and 63 kilodaltons (kDa) appeared to be fairly well conserved between the two species, asjudged by
comigrationontwo-dimensional gels. Several 61-kDa MTAP species detected in MT immunoprecipitates from
bothcell sourcesalso comigrated onthese gels. However, noprotein comigrating precisely with the murine
85-kDa MTAP could be detected in the human cells. Furthermore, twoproteins of 72 and 74 kDa associated
with wild-type MT in the infected humancells but not in murine fibroblasts expressing MT. It had been
previously reported for murine cells that the 70-kDa heat shock protein associates withaparticularmutantMT
but not with wild-type MT (G. Walter, A. Carbone, and W. J. Welch, J. Virol. 61:405-410, 1987). By the
criteria ofcomigrationontwo-dimensional gels,tryptic peptide mapping, and immunoblotting,weshowed that
the 72- and 74-kDa proteins that associate with wild-type MT in humancells arethe major human 70-kDa heat
shockproteins.
Rodent cells expressing polyomavirus middle T antigen (MT) have been utilized extensively for the identificationand analysis of proteins which associate with MT. To date,
cellularproteins of approximately 27, 29, 36, 51, 61, 63, and
85 kilodaltons (kDa) that specifically coimmunoprecipitate withwild-type MT in these cells have been identified (4, 8, 12-14, 18-21). One of the proteins in the 61-kDa range is
known to be pp6Oc-src, a cellular tyrosine kinase (4). The
85-kDa protein has been tentatively identifiedas a
3'-phos-phatidylinositol kinase, since its presence in MT
immuno-precipitates correlates with MT-associated 3'-phosphatidyl-inositol kinase activity (3, 9, 14). The molecular functions of the remainder of the MT-associated proteins (MTAPs), however, are unknown.
Human293 cells expressing MT have been utilized more
recentlywiththedevelopmentofarecombinant adenovirus,
Ad5(pymT), which facilitates the expression ofMT in these
cells atlevels much higher thanthose found in
polyomavi-rus-infected or -transformed rodent cells (1, 21). The MT synthesizedin the human293 cellsappearstobe thesame as
the MTexpressedinrodentcells: it localizestomembranes,
forms acomplexwithpp60csrc, and ismodified in thesame
wayas MTproduced inpolyomavirus-infected mousecells (21). It has been previously reported that when MT is
obtained from human 293 cells by immunoaffinity
purifica-tion, three cellular proteins of 63, 72, and 74 kDa that specifically copurify with MT can be visualized by
Coomassie blue staining (21). In this report we describe
further the identification and analysis of proteins which
associate with MT in human 293 cells. We show thatjustas
in rodent cells, proteins of 27, 29, 36, 61, and 63 kDa associate with MT. The 85-kDa MTAP identified in rodent
cells couldnotreadily be identifiedin this system, although
the MT-associated3'-phosphatidylinositol kinase activityis
*Correspondingauthor.
knowntobepresent(D. Kaplanand T. M. Roberts,
unpub-lished data). We also show that the 27-, 29-, 72-, and 74-kDa proteinsareabundant cellular proteins and thatthe 72- and
74-kDa proteins are the major human 70-kDa heat shock
proteins.
MATERIALS AND METHODS
Cell culture. Murine cells whichexpresswild-typeMTand
G418 resistanceoronlyG418 resistance have been described
previously (2). All rodent cell lines were maintained in
Dulbecco modified eagle medium (DMEM) supplemented
with10% calfserum.Human 293 cells(7)weremaintainedin
DMEMsupplemented with10% fetal calfserum.
Radiolabeling andextractionof cells. For metabolic label-ing of rodent cells with methionine, subconfluent dishes of
cellswerelabeled for5 h with[35S]methionine (300
,uCi/ml)
in DMEM lackingmethionine but supplemented with0.5%
dialyzed fetal bovine serum. For metabolic labeling of293
cells, cells were labeled with
[35S]methionine
(100,XCi/ml)
for 24 h postinfection in DMEM lacking methionine but
supplementedwith2.5% fetal calfserumand5% the normal
level of methionine. Celllysateswerepreparedasdescribed
previously (25).
Immunoprecipitation and protein kinase assays.
Immuno-precipitates were prepared as described previously with
rabbit anti-polyomavirus tumor antigen sera raised against
purified small t antigen (15). Protein kinase assays were performedinvitroasdescribedpreviously (10).
One-dimensional and 2D gel electrophoresis and
fluorog-raphy. Sodium dodecyl sulfate-polyacrylamide gel
electro-phoresis (10% acrylamide) was performed as described by
Laemmli (11). Two-dimensional (2D) gel analysis was
per-formedasdescribedpreviously (13). GradientsofpH4.8 to
7.4wereobtainedby mixing400 ,ulofampholine (pH 3.5 to
10;LKB Instruments, Inc.) with 150
,ul
ofampholine(pH6to 8; LKB) per 10 ml ofgel solution. Gels of
methionine-4533
0022-538X/89/114533-07$02.00/0
CopyrightC 1989, American Society for Microbiology
on November 10, 2019 by guest
http://jvi.asm.org/
Ad5[pymT]
infected
pre
imm
Ad5[wt]
infected
pre
imm
antibodyspecific for both constitutive and inducible formsof hsp70 (17; W. J. Welch, personal communication). They
were washed again and probed with alkaline
phosphatase-conjugated second antibody.Theblotsweredevelopedwith anappropriate substrate and air dried, and the radiolabeled
proteins were visualized by autoradiography. All antibody
solutions incubated with the blots contained 0.5% bovine
serum albumintoreduce nonspecific binding.
74 72
63-9
mT -4
actin
am
369-e
29-)
27-e Om.
FIG. 1. Identification of MTAPs in human 293cells. Celllysates
from 293 cells infected with either recombinant adenovirus
Ad5(pymT) or wild-type adenovirus Ad5(wt) and labeled with
[35S]methioninefor24 h wereimmunoprecipitated withanti-tumor
antigenserum(imm)orpreimmuneantiserum(pre) (13).The
immu-noprecipitates wereanalyzedonsodiumdodecylsulfate-10%
poly-acrylamidegels, and the radiolabeled proteins were visualized by
fluorographyandautoradiography.Thepositions ofMT(mT), actin,
andMTAPsareindicatedbyarrows. Proteins thoughttobe
break-downproductsof MTarenotindicated.
labeled proteins were preincubated with En3Hance before
exposure; gels ofphosphate-labeled proteins were exposed
withintensifyingscreens. Allexposuresweredoneon
XAR-5 film(EastmanKodak Co.) at -70°C.
Peptide mapping. Samples for maps of limit digests of
methionine-labeled proteins were prepared as described
previously (13). The peptides were analyzed by published
procedures (5, 6). The final samples (2
,ul)
were spottedonplastic-backed cellulose-coated thin-layer chromatography plates (10 by 10cm; EMLaboratories). Electrophoresis and
chromatography were carried out as described previously
(13).
Immunoblotting.Immunoblottingwascarriedoutby
stan-dard procedures (23). Briefly, [35S]methionine-labeled
pro-teins in 2D gels were electrophoretically transferred to
nitrocellulose membranes at0.5 A for 3.5 h in an
electro-blotting apparatus (Hoefer). The membranes were blocked
with3%bovineserumalbumin, washed,and probed for 1 h
at22°C with a 1:500 dilution of N27, a mouse monoclonal
RESULTS
Identification of MTAPs in human 293 cells. To
identify
proteins
in
humancells
thatassociate
either
directly
orindirectly
withwild-type
polyomavirus MT,
weanalyzed
immunoprecipitates of
MTfrom
Ad5(pymT) recombinant
adenovirus-infected
(MT-expressing)
(1)
andAd5(wt)
wild-type
adenovirus-infected
(control)
human 293 cellsby
so-dium
dodecyl
sulfate-polyacrylamide gel
electrophoresis
(Fig.
1). A fewbands,
e.g.,actin,
werepresent at the samelevel
inimmunoprecipitates
prepared
fromMT-expressing
or
control cells
and therefore representpolypeptides
non-specifically sticking
toprotein
A-Sepharose beads
oranti-body.
Manybands, however,
were present inimmunopre-cipitates
prepared
from
MT-expressing
cells butonly
foundin trace amounts in
immunoprecipitates
prepared from
con-trol cells
with immune serum(Fig. 1).
Todetermine if
anyof
these
proteins
werenonspecifically
sticking proteins induced
in cellsexpressing
MT, weprepared
immunoprecipitates
from
MT-expressing
cellswith
preimmune
serum andana-lyzed
them inparallel
withthose
prepared
withimmune
serum
(Fig. 1). Any MT-induced proteins
nonspecifically
associating with
immunoprecipitates should
be presentatthesame levels in
immunoprecipitates prepared
with immuneand
preimmune
sera.However,
if theproteins
specifically
associate
directly
orindirectly with
MT,they should be
present at
significantly higher
levelsinimmunoprecipitates
prepared
with immune serum. Thelatterpossibility
was thecase; the
proteins
specific
for
thepresenceof
MTwerealsospecific
for the useof
immune serum.They
arelikely
tobeMTAPs. The
possibility
that somebands
represented
pro-teins that bothcross-reacted
with immune sera and were inducedby
MT cannot be ruled out. Bands below MTpotentially
could be breakdownproducts of
MT, whereasbands
above
MT wereall
mostlikely cellular
proteins.
Thenumbered
arrows inFig.
1indicate
theapproximate
[image:2.612.85.270.75.429.2]posi-tions of
proteins
that we havedetermined
arenotbreakdownTABLE 1. MTAPsidentifiedon2Dgelsa
Presence(+)orabsence(-) in Size of immunoprecipitates of MT from: associated Approx
associatein
AkpI
NIH 3T3 lines 293 cellsprotein (kDa) expressing infectedwith MT Ad5(pymT)
27 5.3 + +
29 5.1 + +
36 6.1 + +
61 6.6-7.4 + +b
63 5.6 + +
72 6.2 - +
74 6.15 - +
85 6.8 +
aThe 51-kDa MTAPidentified inrodent cellsisnotincluded,since it may be a breakdownproductof oneofthe 61-kDaproteins.
bAsubsetofthe61-kDaMTAPspecies foundin NIH 3T3 cells hasbeen
found in 293 cells.
on November 10, 2019 by guest
http://jvi.asm.org/
[image:2.612.319.557.566.697.2]r~~~~~~~~~~~~~~~
.,
.. IhAP
36 294
27
a
H*4
aT
4. 4
s - =
~~
a4
Ta~~~~~~~~O
AA'
4
I
b
4M'. -1
~~4ft.~~~1
d
c
A
A
AA
_ 9
i*4
7,
g
o..
_t~
A.Ma
0 * *
4
IA
[image:3.612.166.450.113.629.2]e
a
FIG. 2. Comparisonon2Dgels of the MTAPs from human and rodent cells.MTandcontrol immunoprecipitates werepreparedfrom
"S-labeled
Ad5(pymT)-infected293cells and from"'S-labeled
MT-expressingNIH 3T3cellsasdescribed in Materialsand Methods. Theseimmunoprecipitates anduntreatedlysatefrom
"S-labeled
uninfected293cellswereanalyzedonparallel2Dgels.(a)MT-expressingNIH 3T3cells, immune serum; (b) MT-expressing NIH 3T3 cells, preimmune serum; (c)
Ad5(pymT)-infected
293 cells, immune serum; (d)Ad5(pymT)-infected
293cells, preimmuneserum; (e)uninfected293 celllysate. Thepositions ofthecomigrating MTAPs areindicatedbynonstarred arrowheadsin panelsa tod. The starred arrowheadsindicatethepositions ofthe human 72-and74-kDa MTAPs. Thepositions
of actin (A) and tubulin (T)areindicated for reference.
{w
i
T, A*
.4t...4 ..'I,*
-ok
AW 'i '16
WI "', 'O i.
I
on November 10, 2019 by guest
http://jvi.asm.org/
a
.4
b
*:-1
-..A:p._
*IIIo"
:::i.:SE P
C
d
W
4W
.Aw * *.4
.4.
.44w
[image:4.568.51.256.63.234.2].4.
FIG. 3. Comparison by complete tryptic peptide mapping of the
27-and29-kDa MTAPs and their comigrating counterparts from 293
cell lysates. MT and control immunoprecipitates were prepared
from 35S-labeled Ad5(pymT)-infected 293 cells as described in
Materials and Methods. These immunoprecipitates and untreated
lysate from 35S-labeled uninfected 293 cells were analyzed on
parallel 2D gels similar to those in Fig. 2c to e. The 27- and 29-kDa proteins were excised and exhaustively digested with trypsin. The
peptideswere prepared as described in Materials and Methods and
analyzed onthin-layer plates by electrophoresis (horizontal axis)
followed by chromatography (vertical axis). (a) Peptides of the
27-kDa protein from MT immunoprecipitates; (b) peptides of the
29-kDa protein from MT immunoprecipitates; (c) peptides of the
27-kDa protein from 293 cell lysates: (d) peptides of the 29-kDa
protein from 293 cell lysates. The peptides in panel d were
electro-phoresed 5 min longer than those in the other panels and are
therefore spread out more in the horizontal direction.
products
of
MT butrather are cellular proteins that associaterwith
MT(seebelow).
Theseinclude proteins of
27, 29, 36,63, 72,
and 74 kDa. Twoother
proteins of
approximately
93and 110
kDa, indicated by arrowheads in Fig.
1,probably
arealso cellular
proteins
buthave
notbeen analyzed
further.To determine
which
proteins from
the human cells maycorrespond
totherodent cell
MTAPspreviously identified
(14)
(Table
1), we analyzed MTimmunoprecipitates from
both
cell sourcesin
parallel
on 2Dgels
(Fig.
2). Thenonstarred arrowheads
inFig.
2aand cindicate
the MTAPsfrom
the two sourceswhich
comigrated. Comigration
wasdetermined initially by
acareful
comparison of the parallel
2D
gels but
wasconfirmed
by
anexperiment in which the
immunoprecipitates
from the
twosources weremixed
prior
to
analysis
on 2Dgels (data
notshown).
Thecomigrating
MTAPs
included
the27-, 29-, 36-, and 63-kDa
proteins
and atleast
oneof the five 61-kDa
species identified previously in
mouse
cells (14).
Thestarred arrowheads
inFig.
2indicate
the
human
72-and 74-kDa MTAPspecies
thatappearto bespecifically associated with
MTin human cells
but notrodent
cells.Because
adenovirus shuts down host
protein synthesis
around
12hpostinfection (22),
some61-kDa
species
may belabeled less
well in human cells than in rodentcells,
making
them
difficult
todetect
by metabolic labeling.
Therefore,
since
severalof the 61-kDa
species
canbephosphorylated
efficiently
in MTimmunoprecipitates by incubation with
[_y-32P]ATP
(unpublished data),
MTimmunoprecipitates
from human and
mousecells
werephosphorylated in vitro
and
analyzed
onone-dimensional and
2Dgels (data
notshown). On examination of
thesegels, several
more61-kDa
MTAP
species from
the human and mouse cells thatcomi-grated
onthe 2Dgels
couldeasily
be identified. Inaddition,
the rodent 85-kDa MTAP could be seen
clearly,
but no precisely comigrating band could be detected in the MTimmunoprecipitates from
human cells(data
notshown).
However,
onone-dimensionalgels,
severalminor bands thatmigrated slightly higher or lower than the rodent 85-kDa protein were detected in MT
immunoprecipitates
from hu-man cells, and one of these may represent the human equivalent of thisprotein.
Identification of the 27-, 29-, 72-, and 74-kDa proteins directly in cell lysates analyzed on 2D gels. To determine whether any of the human MTAPs were major cellular proteins, we analyzed Nonidet P-40 lysates of 293
cells
metabolically labeled with methionine in parallel with MTimmunoprecipitates
on 2D gels. The 27-, 29-, 72-, and74-kDaproteins but not the 36-, 61-, or 63-kDaprotein were found to comigrate with predominant cellular
polypeptides
from the lysates (compare panels c and e of Fig. 2). Each of the proteins was excised from these gels, and the identity ofthe
immunoprecipitated
andcomigrating
lysate species wasconfirmedby 2D separation of complete trypticdigests
(Fig.
3 and4). Of note, the 72- and 74-kDa proteins were notonly
identical to theircomigrating counterparts but also appeared to bepartially
related to each other(Fig.
4).The 72- and 74-kDaproteins are the major human 70-kDa heat shock proteins. We suspected that the 72- and 74-kDa proteins might be the human 70-kDa heat shock proteins because of theirpositions of migration on 2D
gels,
the fact thatthey were related to oneanother,
and their abundancein cell lysates. To test thishypothesis, we analyzed celllysates
from heat-shocked
and control 293 cells on 2Dgels.
Thebrackets
inFig.
5indicate
thepositions of
themajor
heat-inducible 70-kDa proteins in these cells. These were the sameproteins that we found to comigrate with the 72-and
a
b
C
d
FIG. 4. Comparisonbycomplete trypticpeptidemappingofthe
72-and74-kDaproteins and theircomigratingcounterpartsfrom 293
cell lysates. MT and control immunoprecipitates were prepared
from 35S-labeled Ad5(pymT)-infected 293 cells as described in
Materials and Methods. These immunoprecipitates and untreated
lysate from 35S-labeled uninfected 293 cells were analyzed on
parallel2Dgels similartothose inFig.2cto e.The72-and74-kDa
proteins wereexcised andexhaustivelydigestedwithtrypsin. The
peptideswerepreparedasdescribed in Materials and Methods and
analyzed on thin-layer plates by electrophoresis (horizontal axis)
followed by chromatography (vertical axis). (a) Peptides of the
72-kDa protein from MT immunoprecipitates; (b) peptides ofthe
74-kDa protein from MT immunoprecipitates; (c) peptides of the
72-kDa protein from 293 cell lysates; (d) peptides ofthe 74-kDa
protein from 293 cell lysates.
on November 10, 2019 by guest
http://jvi.asm.org/
[image:4.568.295.533.438.584.2]37
3C
430C90->
55->- Tt
A
A29
-i
FIG. 5. Identification of the 72- and 74-kDa MTAPsasthe major human70-kDa heat shock proteins. Parallel 60% confluent dishes of human 293 cellswereincubatedat37or43°C for 3 h and labeled with [35S]methionine (100,uCi/mlinDMEMlacking methionine) for 1 hat the same temperatures. Cells were harvested, washed with phos-phate-buffered saline, and boiled in sample buffer (1.5% sodium dodecyl sulfate, 80 mM dithiothreitol, 8% glycerol). The lysates weresonicated andtreated for 20min with DNaseI(50 ,ug/ml) and
RNase(35,ug/ml)at37°C. Finally, the sampleswereanalyzedon2D gels, and the radiolabeled proteins were visualized by autoradiog-raphy. The brackets indicate the major 70-kDa heat shock proteins. A, Actin; T, tubulin. The numbers correspondtomolecularweights
inthousands.
74-kDa MTAPs and thatweshowedby tryptic mappingwere
equivalentto these MTAPs(Fig. 4). The 90-kDa heat shock proteinwas also visible in Fig. 5, althoughit did notfocus
well andwas retardedatseveralpositions in the gel. Other
proteins could also be seen to vary with the change in
temperature, but those variations were not seen
consis-tently.
Finally,weexamined therelationshipbetween the 72- and
74-kDa MTAPs and the 70-kDa heat shockproteins
immu-nologically. 2D gels of MT immunoprecipitates prepared
with immune and preimmune sera from 35S-labeled cells
were immunoblotted with antisera specific for the 70-kDa
heat shock proteins. Figure 6a and b show the profile of
35S-labeled proteins from these immunoprecipitates after
transfer to nitrocellulose membranes. The 72- and 74-kDa
MTAPsareindicatedby arrowheads. Figure6c and d show
the results of probing these membranes with anti-hsp70
monoclonal antibody N27 (17). The antiserum specifically
recognizedthe 72- and 74-kDa MTAPs butnoneof the other proteins in the immunoprecipitates, further supportingthe
hypothesis that these are the major human 70-kDa heat
shockproteins.
DISCUSSION
We haveanalyzedtheproteins thatcanassociate
specifi-callywith MT in human 293 cellsinfected withAd5(pymT),
a
recombinant adenovirus
which
expressesMT athigh levels
in
these cells. The
proteins identified
arefor the
most partthe
same as thosefound associated with
MTin rodent celllysates. Proteins
of 27, 29, 36, 61, and 63 kDa appear tobe
fairly
well conserved between human and rodentcells,
as
judged by
comigration
on 2Dgels.
Therefore,
theAdS(pymT)-human 293 cell
system maybe
agood
sourceof
these
proteins
forbiochemical
analysis
andfor
use asimmunogens.
Initial
attemptsby
ourlaboratory
topurify
the 36- and63-kDa MTAPs from these cells with anMTimmu-noaffinity
columnfollowed
by preparative
2Dgels
haveyielded microgram
amountsof these proteins (unpublished
data).
Aprotein
precisely
comigrating
with the rodent 85-kDa MTAP couldnotbedetected, raising the possibility
that the humanequivalent of
thisprotein,
ifit
exists, might
notbeas
conserved
as theother
MTAPs. Lackof
detection,
however, certainly does
notproveits absence. Among
otherexplanations,
it ispossible
that the shutdownof
hostprotein
synthesis
around
12h
postinfection
with
adenovirus
(22)
reduces
itsabundance
in thesecells.
Two
additional
proteins
of
approximately
72and
74 kDawhich
have notpreviously
beenshown
toassociate
withwild-type
MTin rodent cells
werefound associated with
MTin
Ad5(pymT)-infected
human
293cell
lysates (21;
Fig.
1).
These
proteins
werefirst
reported by Schaffhausen and
co-workers
tocopurify
with
MT when MT wasimmunoaf-finity purified from these
same cells(21).
We have nowidentified the
72- and 74-kDaproteins
as themajor
human70-kDa
heat shockproteins
by
threecriteria:
comigration
on2D
gels
(Fig.
2and
5),
2Danalysis
of
complete
tryptic digests
(Fig.
4),
andimmunoblotting (Fig.
6).
The 72-kDaprotein
probably
corresponds
tohsx70, also
calledhsp70
or72K,
while
the 74-kDaprotein probably
corresponds
tohsc70,
also called
73K(for
a summaryof
thenomenclature,
seereference
16).
Bothof these
proteins
are present athigh
basal levels in
growing
cells,
and the 72Kprotein
is inducible
by
theadenovirus
ElA
protein,
which
293cells express(for
a
review,
seereference
16).
This isconsistent with
thehigh
level
of
theseproteins
found in
293cells
grownat37°C (Fig.
5).
Incontrast tothe results with human 293
cells,
neither
the73Knorthe 72K
protein
is found
specifically
associated
withimmunoprecipitates
of
wild-type
MTfrom
cultured mousecells
expressing
MT(14, 24).
Theabsence of
the 72Kprotein
is
expected
because there is nodirect
rodentequivalent
ofthis
protein (16). However,
therodent
equivalent
of
the 73Kprotein
is
presentat ahigh
level in celllysates (16)
and yet is notspecifically
associated with
MT(24).
Thereare anumberof
possible explanations
for this result.One
suchexplanation
is
suggested by
thefact that the 73Kprotein
doesassociate
in mouse cells with a
transformation-defective
mutantof
MT, NG59
MT(24),
which
isdeficient in
binding
mostof
theMTAPs
(14).
Itispossible
that the heatshockproteins
bind toMTwhen MTis not associated with itsnormal
arrayof
cellular
proteins.
This wouldbe consistent
with the modelpostulated by
Pelham(16)
which
suggests that the 70-kDa heatshockproteins
haveanaffinity
for "abnormal
proteins"
andbind
tightly
tohydrophobic
surfaces such asthose thatmight
be presentatprotein-protein
binding
sites. Asite thatcan interact with the 70-kDa heat shock
proteins might
be availableonly
whencertain
MTAPsarenotboundtoMT.In the case of NG59MT,
the NG59 mutationprevents
themutant MT from
binding
the36-,
60- and61-,
and 63-kDaMTAPs,
allowing, perhaps,
the lessstringent
binding
of theheatshock
proteins.
Inthecaseof
the293cells,
in which 10to100
times
moreMTissynthesized
than inrodent
cells,
the-06
Ti
lkll.*
, 11 11Or%
on November 10, 2019 by guest
http://jvi.asm.org/
[image:5.612.62.302.76.324.2]Pre-imnu
ne
I
[image:6.612.121.493.79.442.2]a
...
14
Blot
c
m
rn
u
n(i
*:ae>: *-1~~~~~~~~~~~~~~~~~~~~~~~~~
b
d
FIG. 6. Immunoblotting ofMTimmunoprecipitateswithmonoclonalantibodyto70-kDa heatshockproteins.Celllysatesfrom 293 cells
infected withrecombinant adenovirus Ad5(pymT) and labeled with [35S]methionine for 24 h were immunoprecipitated withpreimmune
antiserum (a and c) oranti-tumor antigen serum (band d) (13). The immunoprecipitates were analyzedon 2Dgels, electrophoretically
transferred tonitrocellulose membranes, andprobed with N27, amonoclonal antibody specific forthe 70-kDaheat shockproteins. The
radiolabeled proteinson the membranes were visualizedby autoradiography (aand b). Immunoblots are shownin panels c and d. The
positions ofthe 72- and 74-kDaMTAPsareindicatedbyarrowheads.
MT may
have
saturated some of the associatedproteins,
leaving
somefraction
of MT free to bind the 70-kDa heatshock
proteins.
Attempts
to varythe amountof
MT inthe 293 cells to determine if some of theassociated proteins are saturatedwith
MT and if theassociation of
70-kDa heat shockproteins
correlateswith saturation
have thusfar
beenunsuccessful.
There arealternative explanations
aswell. InHeLa cells
coinfected
with theMT-expressing
recombinant adenovirus andwild-type
helper virus, the association of MT with 70-kDa heat shock proteins can also be detected,indicating
thatthis association is not unique to the 293 cellline (unpublished data). However, both 293 cells and the
wild-type
helper virus express the adenovirusElA
andE1Bproteins,
and the effects of these proteins on 293 or HeLacells
might
in some way induce the binding of heat shockproteins
to MT.ACKNOWLEDGMENTS
Wethank BrianDrukerandLilian Shahrikfor critical review of
themanuscriptandWilliamWelch for hisgenerous gift of antiserum
directed against hsp70.
ThisinvestigationwassupportedbyPublicHealth Servicegrants
CA30002(T.M.R.)and CA45285(D.C.P.)awardedbythe National Cancer Institute. D.C.P. was a Leukemia Society of America Fellowduringaportionof this work.
LITERATURECITED
1. Berkner, K.L., B. S. Schaffhausen, T. M. Roberts, and P. A.
Sharp. 1987. Abundant expression of polyomavirus middle T
antigenanddihydrofolatereductase in anadenovirus
recombi-nant. J.Virol.61:1213-1220.
2. Cherington, V., B. Morgan, B. M. Spiegelman, and T. M.
Roberts.1986. Recombinant retroviruses that transduce
individ-ualpolyomatumorantigens: effectsongrowth and
differentia-tion. Proc. Natl. Acad. Sci. USA 83:4307-4311.
3. Courtneidge, S. A., and A. Heber. 1987. An 81kd protein
complexed with middle T antigen and pp6O -src: a possible
phosphatidylinositol kinase. Cell 50:1031-1037.
4. Courtneidge, S. A., and A. E. Smith. 1983. Polyoma virus
transforming protein associates with the productof the c-src
cellulargene. Nature(London) 303:435-439.
5. Elder,J. H., F. C.Jensen,M. L.Bryant, and R. A. Lerner. 1977.
Polymorphism of the major envelope glycoprotein (gp 70) of
-4
on November 10, 2019 by guest
http://jvi.asm.org/
murine C-type viruses: virion associated and differentiation
antigens encoded by a multi-gene family. Nature (London)
267:23-28.
6. Gibson, W. 1974. Polyomavirus proteins: adescription ofthe
structural proteins of the virion based on polyacrylamide gel
electrophoresis and peptide analysis. Virology62:319-336.
7. Graham, F. L., J. Smiley, W. C. Russell, and R. Nairn. 1977.
Characteristicsofahumancell linetransformed by DNAfrom
humanadenovirustype 5. J.Gen. Virol.36:59-72.
8. Grussenmeyer, T., K. H. Scheidtmann, M. A. Hutchinson, W. Eckhart, and G. Walter. 1985. Complexes of polyoma virus
mediumTantigen and cellularproteins.Proc. Natl. Acad. Sci.
USA 82:7952-7954.
9. Kaplan, D. R., M. Whitman, B. Schaffhausen, D. C. Pallas, M.
White,L.Cantley,and T. M. Roberts. 1987. Commonelements
in growth factor stimulation and oncogenic transformation: 85
kd phosphoprotein and phosphatidylinositol kinase activity.
Cell 50:1021-1029.
10. Kaplan, D. R., M. Whitman, B. Schaffhausen, L. Raptis, R. L. Garcea, D.Pallas, T. M.Roberts, and L. Cantley. 1986. Phos-phatidylinositol metabolism and polyoma-mediated
transforma-tion. Proc. Natl.Acad. Sci. USA83:3624-3628.
11. Laemmli,U. K. 1970.Cleavageofstructuralproteinsduring the
assembly of the head of bacteriophage T4. Nature (London)
227:680-685.
12. Noda, T., M.Satake, T. Robins, and Y. Ito. 1986.Isolation and
characterization of NIH 3T3 cells expressing polyomavirus
smallTantigen. J.Virol. 60:105-113.
13. Pallas, D., and F. Solomon. 1982. Cytoplasmic microtubule
associatedproteins:phosphorylationatnovel sitesis correlated
with theirincorporation into assembledmicrotubules. Cell30:
407-414.
14. Pallas,D.C., V.Cherington, W. Morgan, J. DeAnda, D. Kaplan,
B.Schaffhausen,and T.M. Roberts.1988.Cellularproteins that
associate with the middle and smallTantigens of polyomavirus.
J. Virol. 62:3934-3940.
15. Pallas, D. C., C. Schley, M. Mahoney, E. Harlow, B. S.
Schaff-hausen, and T. M.Roberts. 1986. Polyomavirus small t antigen:
overproduction in bacteria, purification, and utilization for
monoclonal and polyclonal antibody production. J. Virol. 60: 1075-1084.
16. Pelham, H. R. B. 1986. Speculations on the functions ofthe
major heat shock and glucose-regulated proteins. Cell
46:959-961.
17. Riabowol, K. T., L. A. Mizzen, and W. J. Welch. 1988. Heat
shock is lethal to fibroblasts microinjected with antibodies
againsthsp70. Science242:433-436.
18. Rundell, K., E.0.Major, and M. Lampert. 1981.Association of
cellular56,000-and32,000-molecular-weight proteins with BK
virusand polyomavirust-antigens. J. Virol.37:1090-1093.
19. Schaffhausen, B., J. Silver, and T. Benjamin. 1978. Tumor
antigen(s) in cellsproductively infected by wild-type polyoma
virusand mutantNG18.Proc. Natl. Acad.Sci. USA75:79-83.
20. Schaffhausen, B. S., and T. L. Benjamin. 1981. Polyoma virus
middle Tantigenassociated protein kinase. Cold Spring Harbor Conf. Cell Proliferation 8:1281-1297.
21. Schaffhausen, B. S., B. Bockus, K. Berkner, D. Kaplan, and T. M. Roberts. 1987. Characterization of middle T antigen
expressed byusinganadenovirus expressionsystem. J. Virol.
61:1221-1225.
22. Tooze, J.(ed.). 1981. Molecularbiology oftumorviruses, 2nd ed., part 2, revised. Cold Spring Harbor Laboratory, Cold
Spring Harbor,N.Y.
23. Towbin, H., T.Staehelin, and J. Gordon. 1979.Electrophoretic
transfer of proteins from polyacrylamidegels tonitrocellulose
sheets:procedure andsomeapplications.Proc.Natl. Acad.Sci.
USA76:4350-4354.
24. Walter, G., A.Carbone, and W. J. Welch. 1987.Mediumtumor
antigen ofpolyomavirus transformation-defectivemutantNG59
is associated with 73-kilodalton heat shock protein. J. Virol.
61:405-410.
25. Whitman, M., D. R.Kaplan, B.Schaffhausen,L. Cantley,and T.M.Roberts.1985. Association ofphosphatidylinositol kinase
activity withpolyoma middle-T competentfor transformation.
Nature (London)315:239-242.