NATURE CELL BIOLOGY VOLUME 9 | NUMBER 1 | JANUARY 2007
Regulation of protein tyrosine phosphatase 1B
Shrikrishna Dadke1,6,8, Sophie Cotteret1,8, Shu-Chin Yip1, Zahara M. Jaffer1, Fawaz Haj2,7, Alexey Ivanov3,
Frank Rauscher III3, Ke Shuai4, Tony Ng5, Benjamin G. Neel2 and Jonathan Chernoff1,9
Protein-tyrosine phosphatase 1B (PTP1B) is an ubiquitously
expressed enzyme that negatively regulates growth-
factor signalling and cell proliferation by binding to and
dephosphorylating key receptor tyrosine kinases, such as the
insulin receptor1. It is unclear how the activity of PTP1B is
regulated. Using a yeast two-hybrid assay, a protein inhibitor of
activated STAT1 (PIAS1)2 was isolated as a PTP1B-interacting
protein. Here, we show that PIAS1, which functions as a small
ubiquitin-like modifier (SUMO) E3 ligase, associates with
PTP1B in mammalian fibroblasts and catalyses sumoylation
of PTP1B. Sumoylation of PTP1B reduces its catalytic activity
and inhibits the negative effect of PTP1B on insulin receptor
signalling and on transformation by the oncogene v-crk.
Insulin-stimulated sumoylation of endogenous PTP1B results
in a transient downregulation of the enzyme; this event does
not occur when the endogenous enzyme is replaced with a
sumoylation-resistant mutant of PTP1B. These results suggest
that sumoylation, which has been implicated primarily in
processes in the nucleus and nuclear pore, also modulates a key
enzyme–substrate signalling complex that regulates metabolism
and cell proliferation.
Previously, it has been shown that PTP1B interacts with a number of
proteins through a carboxy-terminal proline-rich motif3–6. Using the
C-terminal region (amino acids 278–400) of PTP1B as bait in a yeast
two-hybrid assay7, we identified PIAS1 as a PTP1B-interacting protein
and confirmed this interaction by coimmunoprecipitation studies using
PTP1B-null and similar mouse embryonic fibroblasts (MEFs) reconsti-
tuted with wild-type PTP1B8 (Fig. 1a).
PIAS proteins have recently been shown to act as SUMO E3 ligases,
thereby promoting sumoylation of target proteins2,9. As we found that
PTP1B interacted with PIAS1, we asked if PTP1B undergoes sumoylation
in cells. To examine this hypothesis, PTP1B-null MEFs were transiently
transfected with HA–PTP1B and T7–SUMO-1, either alone or together,
and cell lysates were subjected to immunoprecipitation using anti-HA
(to immunoprecipitate PTP1B) followed by immunoblotting using anti-
T7 (to visualize sumoylated PTP1B) antibodies. Our results indicated
the presence of a series of slowly migrating bands (because sumoylation
increases the relative molecular mass (Mr) of the target protein by 15,000–
20,000) in cells that were coexpressing PTP1B and SUMO-1, indicating
that PTP1B was sumoylated at multiple sites (Fig. 1b). To confirm that it
was PTP1B, rather than a PTP1B-associated protein, that was sumoylated,
the whole cell lysates were immunoprecipitated using anti-T7 (to immu-
noprecipitate sumoylated proteins) antibodies, followed by immunob-
lotting using anti-HA (to visualize sumoylated PTP1B) antibodies. The
results again indicated the presence of a series of slowly migrating bands
in cells that coexpressed PTP1B and SUMO-1, confirming sumoylation of
PTP1B (Fig. 1b). The putative sumoylated forms of PTP1B were evident
only when cells were lysed in the presence of N-ethylmaleimide (NEM),
a strong inhibitor of SUMO isopeptidases10–13 (Fig. 1c). Sumoylation of
PTP1B was measured under denaturing conditions (which inactivates
SUMO isopeptidases) by transiently transfecting PTP1B-null MEFs with
HA–PTP1B and T7–SUMO-1, followed by lysis in a buffer with or with-
out 1% SDS, and immunoblotting using anti-T7 antibodies. The results
indicated that the sumoylation of PTP1B was preserved to a greater extent
under denaturing conditions (see Supplementary Information, Fig. S1).
Next, we asked whether endogenous PTP1B is subject to sumoylation in
cells. PTP1B-null MEFs, wild-type PTP1B MEFs and native MCF-7 cells
were lysed and the whole-cell lysates were immunoprecipitated using
anti-PTP1B antibodies, followed by immunoblotting using anti-PTP1B
and anti-SUMO-1 antibodies(Fig. 1d). Anti-PTP1B immunocomplexes
from either wild-type PTP1B MEFs or native MCF-7 cells showed
a distinct band migrating above the IgG heavy chain, suggesting that
endogenous PTP1B undergoes sumoylation in cells (Fig. 1d). The level of
1Fox Chase Cancer Center, 333 Cottman Avenue, Philadelphia, PA 19111, USA. 2Cancer Biology Program, Department of Medicine, Beth Israel Deaconess Medical
Center, New Research Building, Rm # 1030, 77 Avenue Louis Pasteur, Boston, MA 02115, USA. 3Wistar Institute, 3601 Spruce St., Philadelphia, PA 19104, USA.
4Division of Hematology-Oncology, University of California Los Angeles, 11-934 Factor Building, 10833 Le Conte Avenue, Los Angeles, CA 90095, USA.
5Randall Division of Cell and Molecular Biophysics, Guy’s Campus, King’s College, London, SE1 1UL, UK. 6Current address: Diabetes Research Program,
Stem Cell Research Center, 9th Floor, Manipol Hospital, Airport Road, Bangalore 560017, Karnataka, India. 7Current address: University of California at Davis,
Nutrition Department, 3135 Meyer Hall, One Shields Avenue, Davis, CA 95616, USA. 8These authors contributed equally to this work.
9Correspondence should be addressed to J.C. (e-mail: J_Chernoff@fccc.edu)
Received 24 August 2006; accepted 20 October 2006; published online 10 December 2006; DOI: 10.1038/ncb1522
NATURE CELL BIOLOGY VOLUME 9 | NUMBER 1 | JANUARY 2007
unmodified PTP1B could not be assessed in these blots as it comigrated
with the IgG heavy chain band. To determine whether PIAS1 is required
for sumoylation of PTP1B, PTP1B sumoylation was examined in PIAS1-
null primary MEFs14. PIAS1-null and wild-type PIAS1 MEFs were tran-
siently transfected with either HA–PTP1B or T7–SUMO-1 alone, or
together, as indicated. Whole-cell lysates were subjected to immunopre-
cipitation using anti-HA antibodies, followed by immunoblotting using
anti-T7 antibodies. PTP1B was not sumoylated in PIAS1-null MEFs,
whereas it was sumoylated in wild-type PIAS1 MEFs (Fig. 1e), indicating
that PIAS1 is required for sumoylation of PTP1B in fibroblasts.
PTP1B localizes to the cytoplasmic face of the endoplasmic reticulum
and nuclear envelope through a targeting motif at its extreme C-termi-
nus15,16. To determine whether this cellular localization of PTP1B affects
its ability to become sumoylated, PTP1B-null MEFs were transiently
transfected with either: full-length, endoplasmic reticulum–nuclear
envelope-anchored PTP1B (PTP); a plasma-membrane-anchored
form of PTP1B (PTPCAAX); a C-terminally truncated, cytoplasmic
mutant (PTPΔC; amino acids 1–403); or a nuclear mutant (PTPNLS; see
Supplementary Information, Fig. S2), together with T7-tagged SUMO-
1. Lysates derived from cells transfected with PTP displayed obvious
evidence of multiple sumoylation, whereas PTPΔC and PTPNLS were
sumoylated to a much lesser degree, and plasma membrane-bound
PTP1B was monosumoylated. These results indicate that endoplas-
mic reticulum–nuclear envelope-localization is required for maximal
sumoylation of PTP1B
To definitively demonstrate the localization of sumoylated PTP1B,
fluorescence lifetime imaging microscopy (FLIM; which has previously
been used to detect posttranslational modifications of proteins, includ-
ing phosphorylation19 and ubiquitination20) was used. In this technique,
fluorescent proteins in close proximity to one another induce a change
in the decay rate of the donor’s fluorescent signal. Cells were transfected
with GFP–PTP1B and mRFP–SUMO-1 and FLIM measurements were
obtained. As shown in Fig. 2a, GFP–PTP1B was primarily distributed in
a cytoplasmic compartment characteristic of the endoplasmic reticulum.
Good colocalization of GFP–PTP1B with wild-type mRFP–SUMO-1 was
observed in the perinuclear region (Fig. 2b). The predominant sites where
GFP–PTP1B undergoes Förster resonance energy transfer (FRET) with
mRFP–SUMO-1 were in punctate structures in the perinuclear region,
some of which extended into more peripheral regions, as indicated by a
reduction in fluorescence lifetime. These measurements were repeated
using SUMO-1QT, a mutant form of SUMO-1 that cannot be conjugated
to sumoylation protein substrates21. Compared with wild-type mRFP1-
SUMO-1, SUMO-1QT was generally found to be more diffusely distributed
in cells. Importantly, in the presence of the mRFP1–SUMOQT variant, no
significant reduction in GFP–PTP1B lifetime was observed (Fig. 2b).
Because the FRET species was mainly confined to punctuate structures
within a small perinuclear region of the cell, whereas the GFP–PTP1B
lifetime was normal in the rest of the cell body, no difference in average
FRET efficiency (Eff) was detected between cells coexpressing either
wild-type mRFP–SUMO-1 (average per cell: Eff ± s.e.m. = 9.2 ± 1.8%) or
its QT variant (average Eff ± s.e.m. = 5.8 ± 1.8%). However, the number
of cell pixels that exhibit a FRET efficiency of ≥20% was significantly
increased in cells expressing wild-type mRFP–SUMO (Fig. 2c) when
compared with the QT variant. These data suggest that the bulk of
sumoylated PTP1B localizes to the perinuclear region of the cell.
To determine the primary site(s) of sumoylation, point mutations were
introduced at four consensus sumoylation sites22 in PTP1B by replacing
individual lysine residues at positions 73, 335, 347 and 389 with arginine
Figure 1 PTP1B undergoes sumoylation in cells. (a) PTP1B-null MEFs
or wild-type PTP1B MEFs were either transfected or not with HA–PIAS1.
Cells were lysed and immunoprecipitated using anti-PTP1B antibodies and
immunoblotted with anti-HA antibodies. (b) PTP1B-null MEFs were either
transfected or not with HA–PTP1B or T7–SUMO-1, alone or together. Cells
were lysed and immunoprecipitated using anti-HA or anti-T7 antibodies,
followed by immunoblotting with anti-T7 or anti-HA antibodies, as indicated.
Arrows indicate either mono (1) or polysumoylated (2–5) forms of PTP1B.
(c) PTP1B-null MEFs were either transfected or not, with either HA–PTP1B
or T7–SUMO-1, alone or together. Cells were lysed either in the presence or
absence of NEM and immunoprecipitated using anti-HA antibodies, followed
by immunoblotting with anti-T7 antibodies. (d) PTP1B-null MEFs (lane 1),
wild-type PTP1B MEFs (lane 2), or native MCF-7 cells (lane 3) were lysed
and immunoprecipitated using anti-PTP1B antibodies and immunoblotted
with anti-PTP1B or anti-SUMO-1 antibodies, as indicated. (e) PIAS1-null
and wild-type PIAS1 MEFs were either mock transfected or transiently
transfected with HA–PTP1B or T7–SUMO-1, alone or together as indicated.
Post-transfection (40 h), cells were lysed and immunoprecipitated using
anti-HA antibodies and immunoblotted with anti-T7 antibodies. To ensure
equal loading of proteins whole-cell lysates were immunoblotted using anti-
β-actin antibodies. Uncropped images of the blots in a and b are shown in
the Supplementary Information, Fig. S4.
NATURE CELL BIOLOGY VOLUME 9 | NUMBER 1 | JANUARY 2007
residues (KR mutants). PTP1B-null MEFs were then transiently trans-
tected with either wild-type PTP1B or the KR mutants, alone or together
with T7–SUMO-1, and assayed for sumoylation. Mutation of Lys 73 or
Lys 389 did not alter the electrophoretic behavior of PTP1B, whereas
mutation of Lys 335 and Lys 347 reduced the number and intensity of
slowly migrating PTP1B species (Fig. 3a). Double substitution of Lys 335
and Lys 347 residues greatly reduced the degree of sumoylation of PTP1B,
consistent with the notion that these are the primary, though not the only,
sites of modification in cells (Fig. 3b). PTP1B-null MEFs were also recon-
stituted with a sumoylation resistant form of PTP1B, K73,335,347,389R
mutant (PTP1BKR) for our further studies. When assessed for sumoyla-
tion, PTP1BKR failed to undergo sumoylation compared with wild-type
PTP1B (Fig. 3c). To ensure that the lysine to arginine mutations did not
result in misfolding or loss of activity of PTP1B, the catalytic activity of
wild-type PTP1B and PTP1BKR was measured. Our results showed that
these mutations did not adversely affect the catalytic activity of PTP1B
(see Supplementary Information, Fig. S3).
Given that PTP1B can be allosterically inhibited by drugs (and per-
haps by endogenous proteins) that bind outside the catalytic pocket23,
we asked whether sumoylation of PTP1B affected its activity. Because
NEM inactivates PTP1B, it was not possible to use PTP1B immunopre-
cipitated from mammalian cells to measure enzymatic activity. Instead,
we used a recently developed bacterial expression system to sumoylate
PTP1B24. Escherichia coli were cotransformed with an expression vec-
tor for wild-type PTP1B or a mutant form of this enzyme lacking the
four lysine residues (PTP1BKR) that lie within potential consensus sites
for sumoylation22, and a second vector encoding sumoylation machin-
ery and SUMO-1 (ref. 24). In the absence of sumoylation, PTP1B
and PTP1BKR have similar enzymatic properties (see Supplementary
Information, Fig. S3). Following protein induction, His-tagged PTP1B
was purified using metal-chelate chromatography. Under these con-
ditions, wild-type PTP1B migrated as a series of bands that differ in
relative molecular mass by about 15K (Fig. 4a), similar to the pattern
observed when PTP1B was immunoprecipitated from mammalian cells
(Fig. 1b), whereas PTP1BKR was not detectably sumoylated. We then
compared the phosphatase activity of wild-type PTP1B and PTP1BKR,
alone or coexpressed with SUMO-1. When p-nitrophenyl phosphate
was used as substrate, wild-type and PTP1BKR rapidly hydrolysed this
compound. Sumoylated wild-type PTP1B, in contrast, had only about
40% activity in this assay (Fig. 4b), whereas the catalytic activity of
sumoylation-resistant PTP1BKR was not affected by SUMO-1. These
results suggest that sumoylation of PTP1B inhibits its catalytic activ-
ity. When the sumoylated PTP1B was treated with the SUMO pro-
tease SENP-1, the banding pattern on SDS–PAGE resembled that of
unmodified PTP1B and the catalytic activity increased, consistent with
the notion that sumoylation inhibits the activity of this phosphatase
(Fig. 4c). Activity was also tested against a physiological substrate, the
insulin receptor. Again, although wild-type PTP1B exhibited robust
activity toward this substrate, sumoylated PTP1B was strongly inhibited
and SENP-1 treatment reversed this inhibition (Fig. 4d). As expected,
the activity of the sumoylation-resistant mutant, PTPKR, was unaffected
by the presence of sumoylation machinery or SUMO proteases.
Multiphoton excitationMultiphoton excitationUV illuminated
Figure 2 FLIM analysis of PTP1B and SUMO-1. (a) Multiphoton and
epifluorescence microscopy images of cells expressing GFP–PTP1B alone.
The fluorescence lifetime (τ) of the GFP signal is presented in ns by a
pseudocolour scale. (b) Multiphoton and epifluorescence microscopy images
of cells expressing GFP–PTP1B, in the presence of wild-type mRFP1–SUMO-1
construct or its QT variant. Intermolecular FRET occurs in the subpopulation
of GFP–PTP1B that is sumoylated, between GFP and mRFP1, resulting in
shortening of the fluorescence lifetime of GFP compared with GFP–PTP1B
alone. (c) Cummulative FRET efficiency histogram compiled from all the data
sets (n = 15 and 7 cells, for wild-type SUMO-1 and SUMO-1QT, respectively)
are shown. FRET efficiency (Eff) is 1-τda/τd, where τda is the pixel-by-pixel
fluorescence lifetime of the donor in the presence of acceptor and τd is the
average lifetime of the donor in the absence of acceptor (GFP alone control). The
scale bars represent 5 µm.
NATURE CELL BIOLOGY VOLUME 9 | NUMBER 1 | JANUARY 2007
If sumoylation represents a regulatory mechanism for controlling the
activity of PTP1B, one might expect growth factors to alter the sumoylation
levels of this phosphatase. We therefore examined whether insulin affects
the sumoylation of PTP1B. Serum-starved MEFs expressing His-tagged
PTP1B were treated with 100 nM insulin for different times and the level
of sumoylated PTP1B was assessed. To preserve sumoylated PTP1B, cells
were lysed in the presence of urea and metal-chelate chromatography was
used to isolate PTP1B. In starved cells, sumoylation of PTP1B was barely
detectable, but in insulin-treated cells there was a progressive increase in
PTP1B sumoylation (to a maximum of ~20% of total PTP1B), peaking
at 5 min and thereafter declining (Fig. 5a). These kinetics match those
reported for insulin-receptor activity, which is maximal at 5 min25. In addi-
tion, insulin receptor autophosphorylation was impaired in PIAS1–/– cells
(Fig. 5b), in which PTP1B is not efficiently sumoylated (Fig. 1e). These
data suggest that PIAS-1-catalysed sumoylation has a significant effect
on endogenous PTP1B activity. To further examine the effect of sumoyla-
tion on PTP1B, insulin receptor autophosphorylation was monitored in
MEFs reconstituted with either wild-type PTP1B or PTP1BKR. Our results
indicated that PTP1BKR was more efficient in dephosphorylating the
insulin receptor compared with wild-type PTP1B (Fig. 5c, d). As the only
difference between these two enzymes is that wild-type PTP1B can
undergo sumoylation in cells whereas PTP1BKR is resistant to sumoylation,
these data strongly support the view that sumoylation of PTP1B suppresses
its ability to downregulate the insulin receptor.
We have previously shown that overexpression of wild-type PTP1B
in rat 3Y1 cells suppresses transformation mediated by v-crk26, and
that this suppression is dependent on the catalytic activity of PTP1B.
− + − + − + − + − + − +
− − − − − − − − − − + +
− − − − − − − − + + − −
− − − − − − + + − − − −
− − − − + + − − − − − −
− − + + − − − − − − − − HA−PTP1B
Figure 3 Mapping sumoylation sites on PTP1B. (a) PTP1B-null MEFs were
either transfected or not with wild-type HA–PTP1B, various HA–PTP1BKR
mutants or T7–SUMO-1, alone or together as indicated. Cells were lysed and
immunoprecipitated using anti-HA antibodies, followed by immunoblotting
with anti-T7 antibodies. (b) PTP1B-null MEFs were cotransfected as in a,
except that the indicated double-point mutant of PTP1B (K335,347R) was
used. (c) PTP1B-null MEFs were reconstituted with a PTP1B sumoylation-
resistant mutant (PTP1BKR) and assayed for sumoylation.
14.8+ − + − +
− + + − −
− − − +
270 240 210
Coomassie PNPPase activity
Figure 4 Sumoylation of PTP1B reduces its catalytic activity. (a, b) BL21
(DE3) cells were cotransformed with pTE1E2S1 (ref. 24) and pET28
bearing wild-type PTP1B or PTP1BKR. Protein expression, as shown in a, was
induced by adding IPTG. His-tagged PTP1B was purified by metal-chelate
chromatography and assayed for phosphatase activity against p-nitrophenyl
phosphate (b). A time course for wild-type PTP1B and sumoylated PTP1B
activity is also shown. (c) Effect of SUMO protease SENP-1 on SDS–PAGE
migration and catalytic activity of PTP1B. Sumoylated PTP1B attached
to Ni-NTA beads was treated with buffer or 100 ng recombinant SENP-1
for 15 min at 37oC. The beads were then washed to remove SENP-1 and
the PTP1B eluted and tested for molecular mass and for activity against
p-nitrophenyl phosphate (PNPPase). An uncropped image of this blot is
shown in the Supplementary Information, Fig. S4. (d) Wild-type His-PTP or
His-PTPKR were purified from bacteria that had been cotransformed with a
control plasmid (– SUMO) or pTE1E2S1 (+ SUMO). The proteins were then
treated with buffer or SENP-1, as indicated, and tested for phosphatase
activity against the autophosphorylated β-subunit of the insulin receptor. The
error bars in b and c represent mean ± s.e.m. (n = 3).
NATURE CELL BIOLOGY VOLUME 9 | NUMBER 1 | JANUARY 2007
If sumoylation suppresses the ability of PTP1B to dephosphorylate
the insulin receptor, presumably through inhibition of the catalytic
activity of PTP1B, then sumoylated PTP1B should be less effective at
suppressing transformation as well. Our data indicated that PTP1BKR
was a more potent suppressor of v-crk-induced anchorage-independent
growth compared with wild-type PTP1B (Fig. 5e, f).
Previously, three regulatory mechanisms have been invoked for
PTP1B: proteolysis, phosphorylation and oxidation1. Here, we suggest
that sumoylation of PTP1B by insulin represents an additional layer
of regulation of PTP1B to insure maximal signalling intensity in the
first few minutes after activation of the insulin receptor. Although ques-
tions remain as to how sumoylated PTP1B, which seems to be confined
mainly to the perinuclear region, affects insulin signalling, it is clear
from the effect of loss of PIAS1 (Fig. 5b), or overexpression of mutant
PTP1B (Fig. 5c), that such is the case. Our data show that sumoylated
PTP1B does not efficiently dephosphorylate artificial and physiological
substrates. The primary sites of sumoylation, Lys 335 and Lys 347, are
C-terminal to the proline-rich domain (amino acids 309–315) that is
responsible for binding the docking protein, p130Cas (ref. 3). Sumoylation
at these sites may alter the conformation of PTP1B such that the catalytic
domain is blocked or rendered non-functional. Although the existing
crystal structures of PTP1B do not include residues beyond amino-
acid 321, steric hindrance may result from sumoylation of PTP1B,
thus reducing interaction with the substrates. Regardless of the precise
molecular mechanism, our data show that sumoylated PTP1B does
not efficiently dephosphorylate its physiological substrate, the insulin
receptor. Interestingly, recent drug studies also suggest that allosteric
mechanisms may regulate PTP1B, although the nature of these puta-
tive endogenous regulators is not known23. Although questions remain
as to how sumoylated PTP1B in the perinuclear region affects insulin
signalling, it is attractive to speculate that sumoylation of residues in
the C-terminus of PTP1B may represent one form of such endogenous
Our findings not only extend the number of known substrates for
sumoylation to include an endoplasmic reticulum-bound signalling
protein, but also broaden the range of important physiological processes
that are regulated by sumoylation. Given that the glucose transporters
GLUT1 and GLUT4 are among the handful of non-nuclear proteins that,
similarly to PTP1B, are known to be sumoylated, it is possible that insulin
signalling in particular is subject to this means of regulation.
Plasmids. pEF-BOS–3HA–PIAS1, pCGT–T7–SUMO-1, pEGFP-C2–SUMO-1
and a baculoviral expression vector containing insulin receptor were kindly
provided by A. Kikuchi (Hiroshima University, Hiroshima, Japan), H.Yokosawa
Hokkaido University, Sapporo, Japan), J. Palvimo (University of Helsinki,
Helsinki, Finland) and S. Hubbard (New York University School of Medicine,
New York, NY), respectively. pmRFP1–SUMO-1 was made by removing EGFP
cDNA from pEGFP-C1 (Clontech, Mountain View, CA) and replacing it with
mRFP1 from pcDNA1–mRFP1 (ref. 27), followed by addition of SUMO-1.
Mutants of PTP1B were generated using a site-directed mutagenesis kit from
Stratagene (La Jolla, CA) and mutations were confirmed by sequencing.
Antibodies. Monoclonal antibodies against PTP1B, SUMO-1, HA and T7 were
purchased from Calbiochem (San Diego, CA), Zymed (South San Francisco, CA),
Babco (Berkeley, CA) and Novagen (Madison, WI), respectively. Polyclonal anti-
Myc agarose conjugated (A14) and polyclonal anti-HA (Y11) antibodies were
obtained from Santa Cruz Biotechnology (Santa Cruz, CA). Polyclonal anti-phos-
pho-insulin receptor antibodies (pY1162/pY1663) were purchased from Biosource
International (Camarillo, CA). Secondary antibodies conjugated to alkaline
phosphatase or rhodamine-X were purchased from Jackson Immunoresearch
Laboratories (West Grove, PA).
5 10 −
5 7.5 10 Time (min)
Number of colonies
Figure 5 Regulation of PTP1B through sumoylation. (a) PTP1B-null MEFs
were transfected with an expression vector for His-tagged PTP1B, serum
starved, then stimulated with 100 nM insulin for the indicated times.
The cells were lysed in a lysis buffer containing 6 M urea and PTP1B was
isolated on Ni-NTA beads. The beads were treated with buffer or recombinant
SENP-1 for 15 min as indicated, and the extracts were separated by
SDS–PAGE and blotted with anti-PTP1B antibodies. (b) Wild-type or PIAS1–
/– MEFs were serum-starved, followed by stimulation with 100 nM insulin
for the indicated times. The phosphorylation status of the β-insulin receptor
(βIR) was monitored using phospho-insulin receptor antibodies. (c) PTP1B-
null, wild-type PTP1B and PTP1BKR MEFs were serum-starved, followed by
stimulation with 100 nM insulin for 5 min. The phosphorylation status of the
βIR was monitored using phospho-insulin receptor antibodies. (d) A graphical
representation of phospho-insulin receptor status is shown. (e) Cells were
assessed for anchorage-independent growth by colony formation in soft
agar26. 3Y1 rat fibroblasts transformed with v-crk oncogene were transfected
with either wild-type PTP1B or PTP1BKR as indicated. Cells were seeded in
DMEM containing 10% FBS, 300 µg of G418 per ml and 0.3% soft agar.
Cells were fed once a week. Two weeks after seeding, colonies larger than
0.1 mm in diameter were scored as positive for growth. (f) Quantification of
transformation suppression by wild-type PTP1B and PTP1BKR (number of
colonies ± s.e.m. are indicated; n = 4).
NATURE CELL BIOLOGY VOLUME 9 | NUMBER 1 | JANUARY 2007
Cell culture. PTP1B-null MEFs and MEFs reconstituted with wild-type PTP1B
have been described previously8. Similarly, PTP1B-null MEFs were reconstituted
with PTP1BKR. PIAS1-null, wild-type PIAS1 MEFs14 and MCF-7 (human breast
carcinoma cells) were cultured in DMEM medium supplemented with 10% FBS,
50 U ml–1 penicillin and 50 μg ml–1 streptomycin. Sf9 cells were maintained in
Sf-900 II serum-free media (Invitrogen, Carlsbad, CA).
Transfection, immunoprecipitation and immunoblotting. Cells were trans-
fected using Lipofectamine PLUS reagent (Invitrogen) as per the manufacturers’
recommendations. Post-transfection (40 h), cells were lysed in buffer (50 mM
Tris–HCl at pH 7.5, 150 mM NaCl, 5 mM EDTA, 15 mM MgCl2, 1% Nonidet P-
40, 0.75% sodium deoxycholate, 1 mM dithiothreitol) supplemented with 1:100
diluted protease inhibitor cocktail and phosphatase inhibitors I and II (Sigma, St
Louis, MO) and 20 mM NEM (Sigma). For immunoprecipitation, 450 μg whole-
cell lysate was subjected to either 5 μl anti-PTP1B monoclonal antibodies,10 μl
anti-HA polyclonal antibodies, 3 μl anti-T7 monoclonal antibodies or 20 μl
anti-Myc agarose-conjugated polyclonal antibodies. Immunocomplexes were
eluted in 2× SDS sample buffer, resolved by SDS–PAGE, transferred onto a
PVDF membrane and visualized using secondary antibodies conjugated to
Sumoylation of PTP1B in E. coli. Sumoylation was performed in bacteria as
previously described24. Briefly, wild-type PTP1B (residues 1–403) or PTP1BKR
was cloned into pET28A and E. coli were cotransformed with this plasmid plus
pTE1E2S1 (a plasmid expressing Aos1, Uba2, Ubc9 and SUMO-1). Purification
of His-tagged PTP1B protein was performed using Ni-NTA resin. The catalytic
activity of PTP1B was measured using p-nitrophenyl phosphate and insulin recep-
tor purified from Sf9 cells. Purified insulin receptor was autophosphorylated in
the presence of γ32P-ATP as previously described28.
Immunofluorescence microscopy. Confocal microscopy was performed using
a Radiance 2000 laser scanning confocal microscope (Bio-Rad laboratories,
Hercules, CA) coupled to a Nikon Eclipse E800 upright microscope (Carl Zeiss,
Fluorescence lifetime imaging microscopy. Time-domain FLIM was performed,
as previously described29, with a multi-photon microscope system, comprising
a solid-state-pumped (8 W Verdi, Coherent), femtosecond self-modelocked Ti:
Sapphire (Mira, Coherent) laser system, an in-house developed scan-head and
an inverted microscope (Nikon TE2000E). Enhanced detection of the scattered
component of the emitted (fluorescence) photons was facilitated by the use of fast
single-photon response (Hamamatsu 7400) non-descanned detectors, developed
in-house, situated in the re-imaged objective-pupil plane. Fluorescence lifetime
imaging capability was provided by time-correlated single-photon counting elec-
tronics (SPC 830; Becker & Hickl, Berlin, Germany). A 40× objective was used
throughout (Nikon, CFI60 Plan Fluor N.A. 1.3) and data collected at 510 ± 10 nm
through a bandpass filter. Laser power was adjusted to give average photon count-
ing rates of the order 104–105 photons s–1 (0.0001–0.001 photon-counts per excita-
tion event) to avoid pulse pile up. Instrumental response was measured by surface
second-harmonic generation in a metal film.
Note: Supplementary Information is available on the Nature Cell Biology website.
We thank V. Guacci, A. Kikuchi, S. Michaelis, J. Palvimo, M. White and H.
Yokosawa for their gifts of reagents, and D. Dadke for her assistance. This work was
supported by grants from the National Institutes of Health (B.G.N., K.S. and J.C),
an endowment fund from the Richard Dimbleby Cancer Fund to King’s College
London (T.N.), a National Cancer Institute CORE grant to the Fox Chase Cancer
Center, and an appropriation from the Commonwealth of Pennsylvania. The
multiphoton FLIM system was built with support from both the Medical Research
Council Co-Operative Group grant (G0100152 ID 56891) and an UK Research
Councils Basic Technology Research Programme grant (GR/R87901/01).
S.D. and J.C. conceived the project and prepared most of the manuscript. S.D., S.C.,
S.-C.Y. and Z.M.J. performed the biochemical and cellular analyses. T.N. carried
out the FLIM experiments shown in Fig. 2. F.H., A.I., F.R., K.S. and B.G.N. provided
key reagents, cell lines and/or advice.
COMPETING FINANCIAL INTERESTS
The authors declare that they have no competing financial interests.
Published online at http://www.nature.com/naturecellbiology/
Reprints and permissions information is available online at http://npg.nature.com/
1. Bourdeau, A., Dube, N. & Tremblay, M. L. Cytoplasmic protein tyrosine phosphatases,
regulation and function: the roles of PTP1B and TC-PTP. Curr. Opin. Cell Biol. 17,
2. Kotaja, N., Karvonen, U., Janne, O. A. & Palvimo, J. J. PIAS proteins modulate tran-
scription factors by functioning as SUMO-1 ligases. Mol. Cell Biol. 22, 5222–5234
3. Liu, F., Hill, D. E. & Chernoff, J. Direct binding of the proline rich region of protein
tyrosine phosphatase 1B to the src homology 3 domain of p130Cas. J. Biol. Chem. 271,
4. Goldstein, B. J., Bittner-Kowalczyk, A., White, M. F. & Harbeck, M. Tyrosine dephospho-
rylation and deactivation of insulin receptor substrate-1 by protein-tyrosine phosphatase
1B. Possible facilitation by the formation of a ternary complex with the Grb2 adaptor
protein. J. Biol. Chem. 275, 4283–4289 (2000).
5. Cheng, A., Bal, G. S., Kennedy, B. P. & Tremblay, M. L. Attenuation of adhesion-
dependent signaling and cell spreading in transformed fibroblasts lacking protein
tyrosine phosphatase-1B. J. Biol. Chem. 276, 25848–25855 (2001).
6. Takino, T. et al. Tyrosine phosphorylation of the CrkII adaptor protein modulates cell
migration. J. Cell Sci. 116, 3145–3155 (2003).
7. Golemis, E. A., Gyuris, J. & Brent, R. in Interaction Trap/Two-Hybrid system to iden-
tify interacting proteins. (eds Ausubel, F. M. et al.) (John Wiley and Sons, New York,
8. Haj, F. G., Markova, B., Klaman, L. D., Bohmer, F. D. & Neel, B. G. Regulation of
receptor tyrosine kinase signaling by protein tyrosine phosphatase-1B. J. Biol. Chem.
278, 739–744 (2003).
9. Sachdev, S. et al. PIASy, a nuclear matrix-associated SUMO E3 ligase, represses LEF1
activity by sequestration into nuclear bodies. Genes Dev. 15, 3088–3103 (2001).
10. Li, S. J. & Hochstrasser, M. A new protease required for cell-cycle progression in yeast.
Nature 398, 246–251 (1999).
11. Bylebyl, G. R., Belichenko, I. & Johnson, E. S. The SUMO isopeptidase Ulp2 prevents
accumulation of SUMO chains in yeast. J. Biol. Chem. 278, 44113–44120 (2003).
12. Johnson, E. S. Protein modification by SUMO. Annu. Rev. Biochem. 73, 355–382
13. Kim, Y. H. et al. Desumoylation of homeodomain-interacting protein kinase 2 (HIPK2)
through the cytoplasmic-nuclear shuttling of the SUMO-specific protease SENP1. FEBS
Lett 579, 6272–6278 (2005).
14. Liu, B. et al. PIAS1 selectively inhibits interferon-inducible genes and is important in
innate immunity. Nature Immunol. 5, 891–898 (2004).
15. Frangioni, J. V., Beahm, P. H., Shifrin, V., Jost, C. A. & Neel, B. G. The nontransmem-
brane tyrosine phosphatase PTP-1B localizes to the endoplasmic reticulum via its amino
acid C-terminal sequence. Cell 68, 545–560 (1992).
16. Woodford-Thomas, T. A., Rhodes, J. D. & Dixon, J. E. Expression of a protein tyrosine
phosphatase in normal and v-src-transformed mouse 3T3 fibroblasts. J. Cell. Biol. 117,
17. Pichler, A., Gast, A., Seeler, J. S., Dejean, A. & Melchior, F. The nucleoporin RanBP2
has SUMO1 E3 ligase activity. Cell 108, 109–120 (2002).
18. Saitoh, N. et al. In situ SUMOylation analysis reveals a modulatory role of RanBP2 in
the nuclear rim and PML bodies. Exp. Cell Res. 312, 1418–1430 (2006).
19. Ng, T. et al. Imaging PKC-α activation in cells. Science 283, 2085–2089 (1999).
20. Ganesan, S., Ameer-beg, S. M., Ng, T. T. C., Vojnovic, B. & Wouters, F. S. A dark
yellow fluorescent protein (YFP)-based Resonance Energy-Accepting Chromoprotein
(REACh) for Forster resonance energy transfer with GFP. Proc. Natl Acad. Sci. USA
103, 4089–4094 (2006).
21. Zhao, Y., Kwon, S. W., Anselmo, A., Kaur, K. & White, M. A. Broad spectrum identifi-
cation of cellular small ubiquitin-related modifier (SUMO) substrate proteins. J. Biol.
Chem. 279, 20999–1002 (2004).
22. Melchior, F. SUMO — nonclassical ubiquitin. Annu. Rev. Cell Dev. Biol. 16, 591–626
23. Wiesmann, C. et al. Allosteric inhibition of protein tyrosine phosphatase 1B. Nature
Struct. Mol. Biol. 11, 730–737 (2004).
24. Uchimura, Y., Nakamura, M., Sugasawa, K., Nakao, M. & Saitoh, H. Overproduction
of eukaryotic SUMO-1- and SUMO-2-conjugated proteins in Escherichia coli. Anal.
Biochem. 331, 204–206 (2004).
25. Dadke, S. & Chernoff, J. Protein-tyrosine phosphatase 1B mediates the effects of
insulin on the actin cytoskeleton in immortalized fibroblasts. J. Biol. Chem. 278,
26. Liu, F. & Chernoff, J. Suppression of oncogene-mediated transformation of rat 3Y1
fibroblasts by protein tyrosine phosphatase 1B requires a functional SH3-ligand. Mol.
Cell. Biol. 18, 250–259 (1998).
27. Campbell, R. E. et al. A monomeric red fluorescent protein. Proc. Natl Acad. Sci. USA
99, 7877–7882 (2002).
28. Hubbard, S. R. Crystal structure of the activated insulin receptor tyrosine kinase in
complex with peptide substrate and ATP analog. EMBO J. 16, 5572–5581 (1997).
29. Peter, M. et al. Multiphoton-FLIM quantification of the EGFP-mRFP1 FRET pair for
localization of membrane receptor-kinase interactions. Biophys J. 88, 1224–1237
Figure S1 Stoichiometry of PTP1B sumoylation. PTP1B-null MEFs were
transiently transfected with HA-PTP1B and T7-SUMO-1. Cells were lysed
in standard lysis buffer or a lysis buffer containing 1% SDS. Samples
were then diluted 10x using the lysis buffer described in the Methods
section, followed by immunoprecipitation using anti-HA antibodies and
immunoblotting using anti-T7 antibodies. Whole cell lysates were also
analyzed for the expression of HA-PTP1B and T7-SUMO-1 as shown in
Figure S2 ER/nuclear envelope localization of PTP1B increases
its sumoylation. PTP1B-null MEFs were either not transfected or
transfected with full-length, ER-anchored, wild-type PTP1B (PTP1B-
WT), a plasma-membrane form of PTP1B (PTP1B-CAAX), a nuclear
localized form (PTP-NLS), or a C-terminal truncation mutant that
localizes to the cytoplasm (PTP1B-ΔC), plus T7-SUMO-1. Cells were
lysed and immunoprecipitated using anti-HA antibodies, followed by
immunoblotting with anti-T7 antibodies.
© 2007 Nature Publishing Group
Figure S3 Activity of PTP1B-WT and PTP1B-KR. BL21 (DE3) cells were
transformed with pET28 bearing wild type PTP1B and PTP1B-KR. Protein
expression was induced by adding IPTG and His-tagged PTP1B purified
by metal-chelate chromatography. Proteins were assayed for phosphatase
activity against p-nitrophenyl phosphate.
full scan for Fig. 4c
+ - +
full scan for Fig. 1b
+ - +
IP: α α-HA
IB: α α-T7
IP: α α-T7
IB: α α-HA
Figure S4 Uncropped scans of figures.
© 2007 Nature Publishing Group
SUPPLEMENTARY INFORMATION Download full-text
Acknowledgements. We thank Vincent Guacci, Akira Kikuchi, Susan Michaelis, Jorma Palvimo, Michael White, and Hideyoshi Yokosawa, for their gifts
of reagents, and Disha Dadke for her assistance. This work was supported by grants from the National Institutes of Health (B.G.N., K.S., and J.C), an
endowment fund from the Richard Dimbleby Cancer Fund to King’s College London (T.N.), a National Cancer Institute CORE grant to the Fox Chase Cancer
Center, and an appropriation from the Commonwealth of Pennsylvania. The multiphoton FLIM system was built with support from both the Medical Research
Council Co-Operative Group grant (G0100152 ID 56891) and an UK Research Councils Basic Technology Research Programme grant (GR/R87901/01).
1. Bourdeau, A., Dube, N. & Tremblay, M. L. Cytoplasmic protein tyrosine phosphatases, regulation and function: the roles of PTP1B and TC-PTP.
Curr Opin Cell Biol 17, 203-9 (2005).
Kotaja, N., Karvonen, U., Janne, O. A. & Palvimo, J. J. PIAS proteins modulate transcription factors by functioning as SUMO-1 ligases. Mol Cell
Biol 22, 5222-34 (2002).
Liu, F., Hill, D. E. & Chernoff, J. Direct binding of the proline rich region of protein tyrosine phosphatase 1B to the src homology 3 domain of
p130Cas. J. Biol. Chem. 271, 31290-31295 (1996).
Goldstein, B. J., Bittner-Kowalczyk, A., White, M. F. & Harbeck, M. Tyrosine dephosphorylation and deactivation of insulin receptor substrate-1
by protein-tyrosine phosphatase 1B. Possible facilitation by the formation of a ternary complex with the Grb2 adaptor protein. J Biol Chem 275,
Cheng, A., Bal, G. S., Kennedy, B. P. & Tremblay, M. L. Attenuation of adhesion-dependent signaling and cell spreading in transformed fibroblasts
lacking protein tyrosine phosphatase-1B. J Biol Chem 276, 25848-25855 (2001).
Takino, T. et al. Tyrosine phosphorylation of the CrkII adaptor protein modulates cell migration. J Cell Sci 116, 3145-3155 (2003).
Golemis, E. A., Gyuris, J. & Brent, R. Interaction Trap/Two-Hybrid system to identify interacting proteins. (ed. F.M. Ausubel, R. B., R. E. Kington,
D. D. Moore, J. G. Seidman, J. A. Smith and K. Struhl, eds.) (John Wiley and sons, New York, 1994).
Haj, F. G., Markova, B., Klaman, L. D., Bohmer, F. D. & Neel, B. G. Regulation of receptor tyrosine kinase signaling by protein tyrosine
phosphatase-1B. J Biol Chem 278, 739-44 (2003).
Sachdev, S. et al. PIASy, a nuclear matrix-associated SUMO E3 ligase, represses LEF1 activity by sequestration into nuclear bodies. Genes Dev
15, 3088-103 (2001).
Li, S. J. & Hochstrasser, M. A new protease required for cell-cycle progression in yeast. Nature 398, 246-51 (1999).
Bylebyl, G. R., Belichenko, I. & Johnson, E. S. The SUMO isopeptidase Ulp2 prevents accumulation of SUMO chains in yeast. J Biol Chem 278,
Johnson, E. S. Protein modification by SUMO. Annu Rev Biochem 73, 355-82 (2004).
Kim, Y. H. et al. Desumoylation of homeodomain-interacting protein kinase 2 (HIPK2) through the cytoplasmic-nuclear shuttling of the SUMO-
specific protease SENP1. FEBS Lett 579, 6272-8 (2005).
Liu, B. et al. PIAS1 selectively inhibits interferon-inducible genes and is important in innate immunity. Nat Immunol 5, 891-8 (2004).
Frangioni, J. V., Beahm, P. H., Shifrin, V., Jost, C. A. & Neel, B. G. The nontransmembrane tyrosine phosphatase PTP-1B localizes to the
endoplasmic reticulum via its amino acid C-terminal sequence. Cell 68, 545-560 (1992).
Woodford-Thomas, T. A., Rhodes, J. D. & Dixon, J. E. Expression of a protein tyrosine phosphatase in normal and v-src-transformed mouse 3T3
fibroblasts. J. Cell. Biol. 117, 401-414 (1992).
Pichler, A., Gast, A., Seeler, J. S., Dejean, A. & Melchior, F. The nucleoporin RanBP2 has SUMO1 E3 ligase activity. Cell 108, 109-20 (2002).
Saitoh, N. et al. In situ SUMOylation analysis reveals a modulatory role of RanBP2 in the nuclear rim and PML bodies. Exp Cell Res 312, 1418-
Ng, T. et al. Imaging PKC alpha activation in cells. Science 283, 2085-2089 (1999).
Ganesan, S., Ameer-beg, S. M., Ng, T. T. C., Vojnovic, B. & Wouters, F. S. A dark yellow fluorescent protein (YFP)-based Resonance Energy-
Accepting Chromoprotein (REACh) for Forster resonance energy transfer with GFP. Proc Natl Acad Sci U S A 103, 4089-4094 (2006).
Zhao, Y., Kwon, S. W., Anselmo, A., Kaur, K. & White, M. A. Broad spectrum identification of cellular small ubiquitin-related modifier (SUMO)
substrate proteins. J Biol Chem 279, 20999-1002 (2004).
Melchior, F. SUMO--nonclassical ubiquitin. Annu Rev Cell Dev Biol 16, 591-626 (2000).
Wiesmann, C. et al. Allosteric inhibition of protein tyrosine phosphatase 1B. Nat Struct Mol Biol 11, 730-737 (2004).
Uchimura, Y., Nakamura, M., Sugasawa, K., Nakao, M. & Saitoh, H. Overproduction of eukaryotic SUMO-1- and SUMO-2-conjugated proteins in
Escherichia coli. Anal Biochem 331, 204-6 (2004).
Dadke, S. & Chernoff, J. Protein-tyrosine phosphatase 1B mediates the effects of insulin on the actin cytoskeleton in immortalized fibroblasts. J
Biol Chem 278, 40607-11 (2003).
Liu, F. & Chernoff, J. Suppression of oncogene-mediated transformation of rat 3Y1 fibroblasts by protein tyrosine phosphatase 1B requires a
functional SH3-ligand. Mol. Cell. Biol. 18, 250-259 (1998).
Campbell, R. E. et al. A monomeric red fluorescent protein. Proc Natl Acad Sci U S A 99, 7877-82 (2002).
Hubbard, S. R. Crystal structure of the activated insulin receptor tyrosine kinase in complex with peptide substrate and ATP analog. EMBO J. 16,
Peter , M. et al. Multiphoton-FLIM quantification of the EGFP-mRFP1 FRET pair for localization of membrane receptor-kinase interactions.
Biophys J. 88, 1224-1237 (2005).
© 2007 Nature Publishing Group