MOLECULAR AND CELLULAR BIOLOGY, Apr. 2002, p. 2716–2727
Copyright © 2002, American Society for Microbiology. All Rights Reserved.
Vol. 22, No. 8
Activation of m-Calpain (Calpain II) by Epidermal Growth Factor Is
Limited by Protein Kinase A Phosphorylation of m-Calpain
Hidenori Shiraha,1Angela Glading,1Jeffrey Chou,1Zongchao Jia,2and Alan Wells1*
Department of Pathology, University of Pittsburgh, Pittsburgh, Pennsylvania 15261,1and Department of Biochemistry,
Queen’s University, Kingston, Ontario, Canada K7L 3N62
Received 10 July 2001/Returned for modification 20 August 2001/Accepted 4 January 2002
We have shown previously that the ELR-negative CXC chemokines interferon-inducible protein 10, mono-
kine induced by gamma interferon, and platelet factor 4 inhibit epidermal growth factor (EGF)-induced
m-calpain activation and thereby EGF-induced fibroblast cell motility (H. Shiraha, A. Glading, K. Gupta, and
A. Wells, J. Cell Biol. 146:243–253, 1999). However, how this cross attenuation could be accomplished remained
unknown since the molecular basis of physiological m-calpain regulation is unknown. As the initial operative
attenuation signal from the CXCR3 receptor was cyclic AMP (cAMP), we verified that this second messenger
blocked EGF-induced motility of fibroblasts (55% ? 4.5% inhibition) by preventing rear release during active
locomotion. EGF-induced calpain activation was inhibited by cAMP activation of protein kinase A (PKA), as
the PKA inhibitors H-89 and Rp-8Br-cAMPS abrogated cAMP inhibition of both motility and calpain activa-
tion. We hypothesized that PKA might negatively modulate m-calpain in an unexpected manner by directly
phosphorylating m-calpain. A mutant human large subunit of m-calpain was genetically engineered to negate
a putative PKA consensus sequence in the regulatory domain III (ST369/370AA) and was expressed in NR6WT
mouse fibroblasts to represent about 30% of total m-calpain in these cells. This construct was not phosphor-
ylated by PKA in vitro while a wild-type construct was, providing proof of the principle that m-calpain can be
directly phosphorylated by PKA at this site. cAMP suppressed EGF-induced calpain activity of cells overex-
pressing a control wild-type human m-calpain (83% ? 3.7% inhibition) but only marginally suppressed that of
cells expressing the PKA-resistant mutant human m-calpain (25% ? 5.5% inhibition). The EGF-induced
motility of the cells expressing the PKA-resistant mutant also was not inhibited by cAMP. Structural modeling
revealed that new constraints resulting from phosphorylation at serine 369 would restrict domain movement
and help “freeze” m-calpain in an inactive state. These data point to a novel mechanism of negative control of
calpain activation, direct phosphorylation by PKA.
Signaling pathways from the epidermal growth factor (EGF)
receptor (EGFR) play important roles in wound healing.
Wound fluid EGFR ligands, including heparin-binding EGF-
like growth factor and transforming growth factor ?, are very
strong stimulatory factors for fibroblast cell migration neces-
sary during the repopulation phase of repair (4, 39, 48). During
this migration, tail deadhesion is postulated to be rate limiting
(34). In experimental models, we and others have demon-
strated that failure to deadhere limits cell motility (24, 42).
Calpain activity is critical to integrin-mediated tail deadhesion
on moderately and highly adhesive substrata (40) and to
growth factor-induced motility (17). This intracellular protease
appears to be a key switch, as calpain inhibitors convert
EGFR-mediated signals from cell motility to matrix contrac-
Calpains (EC 18.104.22.168) are a highly conserved family of
intracellular proteases. The two ubiquitous forms are distin-
guishable by their in vitro requirements for calcium, while the
substrate specificities of these two forms appear to be identical
(44). Calpain I, or ?-calpain, is activated at near-micromolar
calcium; calpain II, or m-calpain, requires millimolar calcium
levels. While calcium fluxes have been postulated to regulate
calpains, the physiologically relevant activators of the m-cal-
pain isoform are unknown, since intracellular calcium levels
fail to reach the near-millimolar concentrations required (21).
m-calpain, which predominates in fibroblastoid cells (17, 24,
44), is required for growth factor receptor-mediated dead-
hesion and motility (17, 42). Interestingly, EGF triggers m-
calpain downstream of extracellular signal-related kinase
(ERK)/mitogen-activated protein kinase signaling and not phos-
pholipase C? signaling, which mobilizes intracellular calcium
(17, 49), suggesting a novel mechanism of activation. The phys-
iological substrates of calpain are not known. However, a num-
ber of in vitro and in vivo substrates provide excellent candi-
dates, as they are present at the inner face of the adhesion
complex. These include the cytoplasmic domain of select ?-in-
tegrins (12), focal adhesion kinase (11), and paxillin and talin
(6). Despite the precise molecular basis for calpain-mediated
deadhesion being unknown, it has been well established that
tail deadhesion requires at least one of these forms to be acting
(24, 40, 42).
This requirement of calpain activity for fibroblast migration
during dermal repair provides a target for negative regulation.
This would both prevent excess fibroplasia and convert the
motile phenotype to one of matrix contraction (2). In our
previous paper (42), we reported that ELR-negative CXC che-
mokines present during the resolution phase of wound repair
(14) limited growth factor-induced cell motility. Interferon-
inducible protein 10 (IP-10), monokine induced by gamma
* Corresponding author. Mailing address: Department of Pathology,
University of Pittsburgh, S713 Scaife, Terrace and Lothrop Street,
Pittsburgh, PA 15261. Phone: (412) 624-0973. Fax: (412) 647-8567.
interferon (MIG), and platelet factor 4 (PF4) prevented
EGFR- and platelet-derived growth factor receptor-mediated
calpain activation and cell deadhesion. Although this was ac-
complished secondarily to cyclic AMP (cAMP) generation, the
steps are unknown that bridge the presumption of protein
kinase A (PKA) being activated and the prevention of m-
calpain proteolytic actions. The only evidence-supported
mechanism for negative regulation of calpains is a stoichiomet-
ric inhibition by the endogenous inhibitor calpastatin (5). How-
ever, it is unclear whether calpastatin suppresses m-calpain
activity, as they do not fully colocalize in cells (33), and/or
whether there are additional mechanisms for m-calpain atten-
uation. Our earlier work (42) demonstrated that the chemo-
kine subclass that binds to the CXCR3 receptor, the ELR-
negative CXC chemokines (IP-10, MIG, PF4, and IP-9),
prevented EGFR-mediated calpain activation through a
cAMP-dependent pathway. Thus, we hypothesized that PKA
directly phosphorylates m-calpain and thereby prevents its ac-
tivation by growth factors. This novel mechanism of attenua-
tion was proposed because (i) one can identify a PKA consen-
sus site in the m-calpain putative regulatory domain III (25)
and (ii) it has been reported elsewhere that m-calpain can be
phosphorylated in vitro by PKA (32). Herein, we report that
PKA phosphorylation prevents EGF-induced m-calpain acti-
vation. This was demonstrated by genetically eliminating this
site in human m-calpain and finding that this rendered the
enzyme PKA resistant and that cells expressing this mutant
calpain were resistant to cAMP inhibition of motility. These
findings provide a novel regulatory mechanism for control of
MATERIALS AND METHODS
Cell culture. The murine fibroblast line NR6WT was used in all experiments.
These cells, lacking endogenous EGFR, express approximately 100,000 human
wild-type EGFRs/cell (7) and respond to EGF with both motility and mitogen-
esis (8). These cells were passaged in minimal essential medium ? (MEM?)
supplemented with 7.5% fetal calf serum, nonessential amino acids, and pyruvate
(all culture reagents from GIBCO/BRL, Rockville, Md.). Since the human
EGFR construct was engineered with neomycin-phosphotransferase as a select-
able marker, 350 ?g of G418/ml was added into the culture medium to maintain
human wild-type EGFR. Prior to all experimentation, the cells were made
quiescent in MEM? supplemented with 0.5% dialyzed fetal calf serum (no
Generation of wild-type and dominant negative human m-calpain large sub-
units. Human m-calpain cDNA was obtained by reverse transcription-PCR-
based cloning from human dermal fibroblasts (Hs68; American Type Culture
Collection, Manassas, Va.). Briefly, total RNA was collected from Hs68 cells
with Trizol (GIBCO/BRL). Reverse transcription was performed with purified
Hs68 total RNA with m-calpain-specific oligonucleotide primer (5?-CCTCGTG
TCCTTTGAGAGCG-3?) and Superscript II reverse transcriptase (GIBCO/
BRL). To generate a poly-His (His)-tagged m-calpain large subunit, cDNA sense
(5?-AGCTAGCGGACCGCAGCATGG) and antisense (5?-GCCTTGCCGGC
including the six-His tag) primers were designed according to an m-calpain
cDNA sequence (GenBank accession no. M23254.1). Purified cDNA was am-
plified by PCR using sense and antisense primer and Elongase (GIBCO/BRL)
and cloned into PCR II TA cloning vector (Invitrogen, Carlsbad, Calif.). The size
of the PCR product was ?2.2 kbp. After confirmation by sequencing, m-calpain
cDNA was subcloned into pCEP4 (Stratagene, La Jolla, Calif.) downstream from
a cytomegalovirus (CMV) promoter; the hygromycin resistance gene conferred
selectability for stable expression. The CMV promoter was replaced with mouse
mammary tumor virus promoter (MMTVp) for inducible expression (8). The
PKA-resistant hCANP mutant clone was generated using a PCR-based mutagen-
esis kit (Stratagene) and primers that encoded the mutation (5?-CTGGAGGC
GGGGCGCAGCTGCGGGAGGTTGCAG-3? and 5?-CTGCAACCTCCCGC
AGCTGCGCCCCGCCTCCAG-3?). This changed amino acids 369 and 370
from ST to AA and thus is referred to as ST369AA. The poly-His tag was
replaced with cDNA for green fluorescent protein (GFP) from pEGFP-C1
(Clontech, Palo Alto, Calif.). Both His-tagged and GFP-tagged hCANP con-
structs were utilized for establishing stable transfected cell lines.
hCANP-expressing cell lines. The human m-calpain (hCANP) constructs were
stably expressed in NR6WT cells by electroporation. Twenty micrograms of
hCANP construct plasmid was added to 500 ?l of cell suspension that contained
approximately 2.0 ? 107cells. The cell suspension was transferred into an
electroporation cuvette (0.2-cm gap) and electroporated (500 ?F, 0.320 kV)
using a Gene Pulser electroporator (Bio-Rad, Hercules, Calif.). Electroporated
cells were plated into a six-well tissue culture plate. At 36 h after electroporation,
cells were selected in complete medium containing 100 ?g of hygromycin (Roche
Diagnostics, Indianapolis, Ind.)/ml. Polyclonal lines consisting of more than 20
colonies were established. At least two independent electroporations and stably
transfected lines were established for each construct.
Cell motility assay. EGF-induced cell migration was assessed by the ability of
the cells to move into an acellular area (7). Cells were made quiescent for 24 h
prior to being denuded by a rubber policeman. The cells were then treated or not
with 10 nM EGF, CPT-cAMPS (1 ?M) (Sigma, St. Louis, Mo.), Rp-8Br-cAMPS
(5 ?M) (Calbiochem, La Jolla, Calif.), Rp-8Br-cGMPS (5 ?M) (Calbiochem),
and/or H-89 (at a specified concentration) (Calbiochem). Cells were incubated at
37°C for 24 h. Photographs were taken at 0 and 24 h, and the relative distance
traveled by the cells at the acellular front was determined.
BOC-LM-CMAC assay to measure calpain activity. EGF-induced calpain
activation was determined by the BOC-LM-CMAC (t-butoxycarbonyl-Leu-Met-
chloromethylaminocoumarin) assay (17, 41). Cleavage of this substrate yields
fluorescence that is selective for calpain; specificity is ensured by blocking fluo-
rescence by calpain-selective inhibitors and by molecular downregulation of
calpains. Cells were plated at 50% confluence in a glass chamber (Labtek II;
Nalge Nunc, Roskilde, Denmark). Cells were treated or not with CPT-cAMPS (1
?M) and/or Rp-8Br-cAMPS (5 ?M) and incubated for 30 min in the presence of
30 ?M BOC-LM-CMAC (Molecular Probes, Eugene, Oreg.). Cells were treated
or not with EGF (10 nM) for 5 min prior to visualization using a cooled
charge-coupled device camera (Spot II; Diagnostic Instruments, Sterling
Heights, Mich.) (17). EGF-induced calpain cleavage of glutathione-conjugated
BOC-LM-CMAC generates glutathione-conjugated CMAC and results in in-
creased fluorescence (excitation, 330 nm; emission, 403 nm). The slides were
observed by fluorescence microscopy with an Olympus (Tokyo, Japan) M-NUA
filter; pictures were taken with the same exposure setting within each experiment.
The signal intensities of cells for each experimental condition were measured in
computer-captured pictures by using Photoshop (Adobe, San Jose, Calif.). The
numerical data are the means ? standard errors of the means (SEM) of more
than 100 cells.
Immunoblotting. Cells were grown to confluence in six-well tissue culture
plastic plates. After 24 h of quiescence, cells were treated or not with dexameth-
asone (2 ?M) (Sigma) for 18 h. Dexamethasone (2 ?M for 18 h) was used to
induce MMTV-driven hCANP expression. Cell lysates were separated on a
sodium dodecyl sulfate (SDS)–10% polyacrylamide gel and transferred to a
polyvinylidene difluoride (PVDF) membrane, Immobilon-P (Millipore, Bedford,
Mass.). Blots were probed with anti-m-calpain (Santa Cruz Biotechnology, Santa
Cruz, Calif.), anti-GFP (Clontech), or anti-poly-His (Santa Cruz) antibodies
before visualization with alkaline phosphatase-conjugated secondary antibodies
(Promega, Madison, Wis.) followed by development with a colorimetric method
In vitro PKA phosphorylation assay. Cells were grown to confluence in a
10-cm-diameter tissue culture plastic plate. After 24 h of quiescence, cells were
treated with dexamethasone (2 ?M) for 18 h. His-tagged m-calpains were puri-
fied by Ni-nitrilotriacetic acid (NTA) agarose (Qiagen, Valencia, Calif.) affinity
chromatography. Calpains were eluted with 50 mM Tris-HCl (pH 7.5) containing
250 mM imidazole and dialyzed with 50 mM Tris-HCl (pH 7.5). Purified His-
tagged m-calpains were incubated for 15 min at 30°C with an assay mixture
containing 50 mM Tris-HCl (pH 7.5), 10 mM MgCl2, 200 ?M [?-32P]ATP (final
specific activity of 200 ?Ci/?mol), and 75 U of cAMP-dependent protein kinase
catalytic subunit (New England Biolabs, Beverly, Mass.). The samples were
separated by SDS–10% polyacrylamide gel electrophoresis, and then samples
were transferred into an Immobilon-P (Millipore) PVDF membrane and ex-
posed to X-ray film in a cassette for 72 h at ?80°C. After the autoradiography the
membrane was blotted with anti-m-calpain antibody (Santa Cruz) to confirm that
the amounts of protein were consistent in each lane. The signals were visualized
using alkaline phosphatase-conjugated secondary antibody (Promega) and a
colorimetric development system (Promega).
VOL. 22, 2002PKA INHIBITS m-CALPAIN ACTIVATION2717
In vitro calpain activity assays. His-tagged hCANP proteins were purified
using a Ni-NTA column as described above. The activity of the hCANP proteins
expressed in cells was tested in vitro using a constitutively active ERK (Upstate
Biotechnology, Inc., Lake Placid, N.Y.) to induce calpain (A. Glading et al.,
unpublished observations). Purified calpain from a confluent 10-cm-diameter
dish was incubated with 0.5 ?g of recombinant Tau protein in 25 mM HEPES
and 1 mM dithiothreitol (DTT) (Panvera, Madison, Wis.) with or without cal-
cium (final concentration of 40 to 41 ?M free Ca2?ion in assay sample) in the
presence or absence of active ERK (300 U in 20 mM morpholinepropanesulfonic
acid [MOPS; pH 7.2], 25 mM ?-glycerol phosphate, 5 mM EGTA, 1 mM
Na3VO4, 1 mM DTT, 75 mM MgCl2, and 500 ?M ATP) for 10 min. The reaction
mixtures were then separated on an SDS–10% polyacrylamide gel, transferred to
a PVDF membrane, and probed for Tau (anti-Tau C-terminal sequence; Zymed,
South San Francisco, Calif.). Reduction in size of the Tau band (65 kDa) was
indicative of calpain activity. While calcium alone did not stimulate substantial
Tau cleavage at this time point, longer incubations with calcium alone did cause
Tau cleavage (data not shown).
To investigate the effect of PKA phosphorylation on in vitro activity of
hCANP, a similar experiment was performed using microtubule-associated pro-
tein 2 (MAP2) as the substrate. The substrate was changed becuase PKA phos-
phorylates Tau and prevents its cleavage by calpain (29, 35). Although PKA
similarly phosphorylates MAP2, longer incubations are able to overcome this
effect (27, 30). Again purified hCANP from a confluent 10-cm-diameter dish was
incubated with 1 ?g of MAP2 protein (Cytoskeleton) dissolved in 25 mM
HEPES–1 mM DTT–CaCl2(final concentration of 23 to 28 ?M free Ca2?ion in
assay sample) with or without ERK (300 U in same buffer as above) in the
presence or absence of PKA catalytic subunit (25 U in PKA assay buffer [Cell
Signaling Technology, Beverly, Mass.] with 75 mM MgCl2) (Cell Signaling Tech-
nology). The reaction mixture was then separated on an SDS–10% polyacryl-
amide gel, transferred to a PVDF membrane, and probed for MAP2 with a
monoclonal anti-MAP2 antibody (Sigma). Loss of cleavage product bands (75 to
120 kDa) was indicative of calpain activity.
Downregulation of endogenous mouse m-calpain. To avoid the influence of
endogenous mouse calpain in NR6WT cells, cells were treated with antisense
phosphothiorate-linked oligonucleotide specific for mouse m-calpain (5?-TGCC
CGCCATGGTAGCGATC-3?) (17). Briefly, cells were treated with oligonucle-
otide and EGF (10 nM) for 6 h to deplete the preexisting calpain. Cells were then
incubated for a further 18 h in the presence of oligonucleotide without EGF to
recover from EGF stimulation but prevent de novo calpain synthesis. At the
same time cells were treated with dexamethasone (2 ?M). Then the cell motility
assay was performed as described above.
Molecular modeling of S369-phosphorylated m-calpain. The modeling tem-
plate is the X-ray crystal structure of m-calpain (22). A phosphate group was
manually placed on the side chain of S369 with the SGI graphics workstation
(Silicon Graphics, Mountain View, Calif.). Appropriate side chain adjustments
were made to eliminate any apparent steric conflict near the phosphorylation
site. The S369-phosphorylated m-calpain was then energy minimized using the
SYBYL software package (Tripos Associates, St. Louis, Mo.), employing a Tri-
pos force field, Gasteger and Marsili charges, and dielectric constant. Energy
minimization continued until the final system energy converged. This process
took approximately 2,000 cycles. The resulting model was carefully examined to
ensure that there was no geometry violation and that the interactions involving
the phosphate group were reasonable (i.e., devoid of any unacceptable short
contacts with other atoms in the vicinity). Diagrams were generated using Mol-
A cAMP-PKA signaling pathway inhibits EGF-induced cell
migration. In our previous paper (42), we demonstrated that
ELR-negative CXC chemokines IP-10, MIG, and PF4 inhibit
EGFR-mediated cell deadhesion and thus motility secondary
to cAMP generation in primary human fibroblasts. To validate
that this occurs in our genetically amenable mouse fibroblast
system, we demonstrated that the cell-permeable cAMP ana-
log CPT-cAMPS had a similar negative effect on human wild-
type EGFR expressing mouse fibroblast NR6WT cells (Fig.
1A). CPT-cAMPS at 1 ?M inhibited 10 nM EGF-induced cell
migration by 55% ? 4.5%. To prove that cAMP inhibits cell
motility through cAMP-dependent PKA, NR6WT cells were
treated with PKA inhibitors. The cell-permeable antagonist of
cAMP, Rp-8Br-cAMPS, prevented CPT-cAMPS inhibition of
motility, whereas the protein kinase G (PKG)-preferential in-
hibitor Rp-8Br-cGMPs had no discernible effect (Fig. 1A). A
second PKA inhibitor, H-89, also eliminated cAMP’s inhibi-
tion in a dose-dependent manner (Fig. 1B).
We postulated that calpain-mediated deadhesion is required
for tail detachment during EGF-induced motility. Thus, cAMP
should prevent this release. Direct visualization of cells by
time-lapse microscopy demonstrated that the effect of attenu-
ated rear release is to prevent productive motility (data not
shown). The vast majority of cells extended protrusions that
adhered to the substratum and pulled the nuclei forward. How-
ever, the tails were not efficiently detached in the presence of
CPT-cAMPS, and the cells were observed to recoil and to
retract their protrusions. With many cells, we could visualize
repeated cycles of extension and retraction with the same point
tail adhesion remaining constant throughout. Over 80% of the
control, EGF-only-treated cells demonstrated productive mo-
tility over the same time period (data not shown), in agreement
with previously published works (36, 47). These findings sup-
port the contention that the rate-limiting step for calpain in
cell motility is at the point of rear release.
cAMP does not inhibit EGF-induced m-calpain activation in
the presence of PKA inhibitor. To demonstrate that this inhi-
bition of motility occurred secondarily to inhibition of calpain
activation, we found that cAMP prevents EGF-induced m-
calpain activation as determined by BOC-LM-CMAC fluores-
cence (Fig. 2). Again, the PKA inhibitor Rp-8Br-cAMPS elim-
inated the effect of CPT-cAMPS (Fig. 2). Cell cytometry
studies determined the numerical value of calpain activity from
the BOC-LM-CMAC assay. CPT-cAMPS inhibited EGF-in-
duced calpain activation by 87% ? 2.3%. This was significantly
reduced by Rp-8Br-cAMPS (15% ? 2.2%) but not by the
control PKG inhibitor Rp-8Br-cGMPS (86% ? 2.3%). These
data demonstrate that the PKA attenuation cross talk of
growth factor-induced m-calpain is functional in these cells,
both extending our understanding of this counterregulatory
pathway and validating the use of these cells in subsequent
PKA, in vitro, phosphorylates wild-type human m-calpain
but not ST369AA m-calpain. The data above show that cAMP
inhibits calpain activation through cAMP-dependent PKA. We
hypothesized that this occurred directly through the novel
mechanism of direct PKA phosphorylation of m-calpain. Hu-
man m-calpain large subunit contains a PKA consensus se-
quence, RRXS/T, in its putative regulatory domain III (46).
Despite the previous report that calpain is not a phosphopro-
tein in vivo (1), we and others find m-calpain isolated from
fibroblasts to be heavily phosphorylated as detected with anti-
serine and antithreonine antibodies (data not shown) and by
standard biochemical assignment of phosphorylation sites (J.
Cong, V. F. Thompson, and D. E. Goll, Abstr. Am. Soc. Cell
Biol. 40th Annu. Meet., 2000, abstr. 2003). In fact, Cong and
colleagues reported that the serine 369 in the PKA consensus
site was phosphorylated. Independent of this report, we mu-
tated this serine 369 to an alanine; the adjacent threonine 370
was simultaneously replaced with an alanine to rule out the
possibility either that both amino acids are targets or that the
elimination of serine 369 would alter the site to RRXXT with
2718 SHIRAHA ET AL.MOL. CELL. BIOL.
a shift to phosphorylation of threonine 370. Both the wild-type
and ST639AA His-tagged constructs were expressed in cells,
isolated using a nickel column, and subjected to in vitro phos-
phorylation by a cAMP-dependent PKA catalytic subunit.
Wild-type hCANP was phosphorylated by PKA, while ST369AA
hCANP was not phosphorylated (Fig. 3A). These findings pro-
vide proof that m-calpain may be directly phosphorylated by
PKA at this site in regulatory domain III.
However, the question remains whether the ST369AA m-
calpain is functional. We expressed both wild-type and ST369AA
hCANP as poly-His-tagged constructs in cells and used nickel
columns to purify them. To calibrate ERK activation and any
FIG. 1. Effects of CPT-cAMPS and PKA inhibitors on EGF-induced cell migration. NR6WT cells were grown to confluence and made
quiescent for 24 h in MEM? with 0.5% dialyzed fetal calf serum before treatment or not with EGF (10 nM), CPT-cAMPS (1 ?M), Rp-8Br-cAMPS
and/or Rp-8Br-cGMPS (5 ?M) (A), and H-89 (at indicated concentrations) (B). The datum for nontreated control cells is labeled as nTx. Basal
and EGF-induced cell migrations were assessed as the ability of the cells to move into an acellular area after 24 h of EGF treatment. The data
are shown as the ratios to the 10 nM EGF-induced cell migration activity. The data are the means ? SEM of at least three independent studies,
each performed in triplicate. Statistical analysis was performed by Student’s t test. n.s., not significant.
VOL. 22, 2002PKA INHIBITS m-CALPAIN ACTIVATION 2719
FIG. 2. EGF-induced calpain activation. NR6WT cells were plated on tissue culture chamber slides (Nunc) and made quiescent for 24 h in
MEM? with 0.5% dialyzed fetal calf serum. Cells were treated or not with CPT-cAMPS (1 ?M), Rp-8Br-cAMPS (5 ?M), and/or Rp-8Br-cGMPS
(5 ?M) 30 min prior to EGF (10 nM) treatment in the presence of BOC-LM-CMAC (Molecular Probes). Then cells were treated or not with EGF
(10 nM) for 5 min. Calpain activation was assessed by fluorescence microscopy. The fluorescence indicates calpain activity. The panel for
nontreated control cells is labeled as nTx. The pictures shown are representative of n ? 9.
inhibition by PKA, we used the physiologically attainable mi-
cromolar concentrations of free calcium rather than the phar-
macological millimolar levels employed for maximal activation
in vitro. This lower level of calcium leads to calpain-dependent
Tau degradation, but that is noticeable only after extended
incubation periods (data not shown). Both wild-type and
ST639AA hCANP were similarly inactive as isolated and sub-
sequently activable by ERK in vitro as determined by the
ability to clip the reporter protein Tau (29) (Fig. 3B). Thus, the
mutation of the serine and threonine in domain III did not
adversely affect either basal or activated enzymatic activity. A
second substrate, MAP2, was used to determine what effect
PKA has on calpain functioning in vitro, as PKA phosphory-
lates Tau to render it calpain resistant (29, 35). Here, too, ERK
activated both wild-type and ST369AA m-calpains to cleave
MAP2 (Fig. 3C). However, addition of the catalytic subunit of
PKA inhibited MAP2 cleavage only by wild-type, not ST369AA,
hCANP (Fig. 3C). This strongly suggests that the elimination
of this PKA phosphorylation site eliminates a negative atten-
Expression of ST369AA m-calpain renders cells resistant to
cAMP inhibition of calpain activation and motility. To deter-
mine the biological role of PKA phosphorylation of m-calpain,
the constructs needed to be expressed in live cells. As there
have been previous reports of instability of expressed calpain in
cells (13), we chose a dexamethasone-inducible system (8). The
expression of both the GFP- and His-tagged clones was tightly
regulated by dexamethasone (Fig. 4). The total level of m-
calpain in the cell increased by 32% ? 5%; the GFP-tagged
m-calpain constructs migrated at ?27 kDa higher. By and
large, the clones are expressed at similar levels in all of the cells
as determined by GFP fluorescence (data not shown). This
FIG. 3. m-Calpain phosphorylation at ST369/370 by PKA limits proteolytic activity. Both His-tagged wild-type- and ST369AA hCANP-
expressing cells were grown and made quiescent for 24 h. Cells were treated with dexamethasone (2 ?M) for 18 h to induce expression of exogenous
hCANP constructs. To determine whether m-calpain was a target of PKA, hCANP was purified by Ni-NTA affinity chromatography and dialyzed
with 50 mM Tris-HCl (pH 7.4) for 24 h. (A) Purified hCANP was treated with constitutively active cAMP-dependent PKA catalytic subunit and
[?-32P]ATP for 15 min. The samples were analyzed by separation through an SDS–10% polyacrylamide gel and transferred to a nylon membrane
(Immobilon-P). The membrane was used for autoradiography and subsequently blotted with anti-m-calpain antibody (Santa Cruz). The results
shown are representative of two independent experiments. (B) Purified hCANP was incubated with recombinant Tau protein with or without
calcium (40 ?M) in the presence or absence of ERK (300 U) for 10 min. Samples were analyzed by SDS–10% polyacrylamide gel electrophoresis
and transferred to a nylon membrane (Immobilon-P). The transferred membrane was blotted with anti-Tau antibody (Zymed). Reduction in the
size of Tau indicates calpain activity. The results shown are representative of two independent experiments. (C) Purified hCANP was incubated
with MAP2 in the presence or absence of ERK (300 U) and PKA catalytic subunit (25 U) for 20 min. Samples were analyzed on an SDS–10%
polyacrylamide gel and transferred to a nylon membrane (Immobilon-P). The membrane was blotted with anti-MAP2 antibody (Sigma). Loss of
cleavage products of 75 to 120 kDa indicated activity. The results shown are representative of two independent experiments.
VOL. 22, 2002PKA INHIBITS m-CALPAIN ACTIVATION2721
expression of the m-calpain constructs in the entire cell pop-
ulation allows for easy investigation of cell responses.
To test our hypothesis that direct phosphorylation of m-
calpain by PKA eliminates EGFR-mediated effects, we in-
duced expression of the calpain constructs in quiescent cells by
using dexamethasone. EGF-induced calpain activity of cells
expressing the control wild-type hCANP was inhibited by 83%
? 3.7% by CPT-cAMPS, similar to the level of inhibition in
mock-transfected cells, while that of cells expressing the PKA-
resistant ST369AA construct was inhibited by only 25% ?
5.5% (Fig. 5). This only partial elimination of inhibition is
likely due to the fact that the mutant construct constitutes only
about one-third of the cellular m-calpain with the endogenous
calpain contributing to the activation level while still being
EGF-induced cell migration activity of cells expressing
ST639AA was unaffected by CPT-cAMPS (?7% ? 12% and
13% ? 11% for His- and GFP-tagged cells, respectively), while
the motility of cells expressing the control wild-type hCANP
constructs was fully eliminated by CPT-cAMPS (63% ? 6.2%
and 75% ? 5.4% for His- and GFP-tagged cells, respectively)
(Fig. 6A; the data for GFP-tagged cells are not shown). In both
His- and GFP-tagged cells, the absolute basal and EGF-in-
duced cell migration activities are quite consistent in the pres-
ence or absence of dexamethasone. The motility parameters of
these cells in the absence of dexamethasone induction were
identical to those for parental cells (Fig. 6B; the data for
GFP-tagged cells are not shown). That this was due to expres-
sion of the PKA-resistant construct was shown both by the
control wild-type construct being inhibited and by the cells that
expressed ST369AA being inhibited by CPT-cAMPS in the
absence of dexamethasone induction.
To eliminate confounding effects of endogenous mouse m-
calpain, antisense oligonucleotides downregulated the murine
m-calpain (17). Antisense m-calpain successfully downregu-
lated endogenous m-calpain expression (Fig. 6E) and inhibited
?60% of EGF-induced cell migration (Fig. 6D). This result is
consistent quantitatively with our previous report of partial
motility elimination by downregulation of m-calpain (17). In
the cells treated with antisense oligonucleotide and dexameth-
asone, hCANP constructs replaced endogenous m-calpain and
restored the full EGF-induced motility response. Only the cell
migration of hCANP wild-type cells was inhibited by CPT-
cAMPS (56% ? 8.2%; P ? 0.05), while that of hCANP
ST369AA was not inhibited by CPT-cAMPS (?7.7% ? 8.7%;
not significant) (Fig. 6C). Thus, the ST369AA construct re-
placed the functionality of endogenous m-calpain while re-
maining PKA insensitive.
S369 phosphorylation is predicted to limit the mobility of
m-calpain domains. Structural studies would provide a molec-
ular basis for this mechanism of inhibition and why it appears
dominant over EGF-induced activation. A structural model of
S369-phosphorylated m-calpain was generated by molecular
modeling. The initial amino acid localizations are based on
high-resolution X-ray procedures. Since S369 of domain III is
located in the interface region between domains III and IV,
the phosphorylated S369 is able to make intimate contacts with
two residues in domain IV. These two residues are R628 and
H634, which can form two and one interaction with S369-P,
respectively (Fig. 7). These new interactions are predicted to
rigidify the mobility of both domains III and IV and prevent
the formation of an active cleft.
We present data demonstrating a novel mechanism for neg-
ative regulation of EGF-induced m-calpain activation–-direct
phosphorylation by PKA. We previously reported that EGF-
induced m-calpain activation is negatively regulated by ELR-
negative CXC chemokines (42). However, the molecular
mechanism for this regulation was undefined. Herein, we re-
port that PKA phosphorylation of m-calpain at amino acids
S369 and T370 blocks EGFR-mediated activation in vivo. The
addition of a negatively charged side chain is modeled to
“freeze” m-calpain in an inactive state. These findings support
an unexpected model of m-calpain regulation that involves
protein phosphorylation of the regulatory domain.
In brief, mutation of a putative consensus PKA site at amino
acids S369 and T370 to alanines generated a calpain molecule
that was resistant to CPT-cAMPS negative attenuation of
EGF-induced calpain activation and cell motility. One concern
about mutagenesis is that it might render the target molecule
inactive. We do not feel that our failure to attenuate EGFR
FIG. 4. hCANP expression in NR6WT cells. MMTV-driven
hCANP constructs were introduced into NR6WT cells by electropo-
ration. Polyclonal stable expression cell lines were selected in the
presence of hygromycin. Cells were treated or not with dexamethasone
(2 ?M) for 18 h. Cells were lysed, and equal amount of proteins were
analyzed by SDS–10% polyacrylamide gel electrophoresis and immu-
noblotted with anti-m-calpain antibody (Santa Cruz), anti-GFP anti-
body (Clontech) (A), or anti-His antibody (Santa Cruz) (B). The
anticalpain antibody recognizes the endogenous mouse m-calpain
(?80 kDa) as well as the exogenously encoded expressed human cal-
pain-GFP (?110-kDa) and His (?80-kDa) fusion proteins. Shown are
representative blots from three independent experiments.
2722 SHIRAHA ET AL.MOL. CELL. BIOL.
FIG. 5. EGF-induced calpain activation in hCANP construct expressing NR6WT cells. NR6WT cells with His-tagged hCANP constructs were
plated on tissue culture chamber slides (Nunc) and made quiescent for 24 h in MEM? with 0.5% dialyzed fetal calf serum. Cells were treated with
dexamethasone (2 ?M) for 18 h. Then cells were treated or not with CPT-cAMPS (1 ?M) 30 min prior to EGF (10 nM) treatment in the presence
of BOC-LM-CMAC (Molecular Probes). Finally cells were treated or not with EGF (10 nM) for 5 min. Calpain activation was assessed by
fluorescence microscopy. The fluorescence indicates calpain activity. nTx, nontreated control cells. The pictures shown are representative of
n ? 9.
signaling is due to such a false-positive result since the
ST369AA mutant calpain remains activable by EGF. The in
vitro activity of the ST369AA mutant was similar to that of
wild-type m-calpain as determined by two different calpain
activity assays, cleavage of Tau and cleavage of MAP2 (Fig. 3).
Our other biologic reporter assay, the BOC-LM-CMAC assay,
relies on the mutant calpain being activable in the face of
normally negative regulation. This finding is buttressed by the
FIG. 6. EGF-induced cell migration activity of hCANP construct-expressing NR6WT cells. NR6WT cells with His-tagged hCANP constructs
were utilized for an in vitro cell migration assay to assess the biological effect of dominant negative m-calpain. Cells were grown to confluence in
MEM? with 7.5% fetal calf serum containing 100 ?g of hygromycin/ml. Cells were made quiescent with MEM? with 0.5% dialyzed fetal calf serum
for 24 h before treatment with EGF. Cells were treated or not with antisense mouse m-calpain to downregulate endogenous mouse m-calpain. Cells
were treated with EGF (10 nM) and antisense m-calpain (20 ?M) for 6 h. Cells were then incubated for a further 12 to 14 h in quiescent medium
with antisense m-calpain. Cells were treated (A and C) or not (B and D) with dexamethasone (2 ?M) 2 h prior to EGF treatment. Then cells were
treated with CPT-cAMPS (1 ?g/ml) and EGF (10 nM). Cell migration activity was measured as the ability of the cells to move into an acellular
area after 24 h of EGF treatment. The data for nontreated control cells are labeled as nTx. The data are shown as ratios to 10 nM EGF-induced
cell migration in the absence of antisense m-calpain oligonucleotide. The data are the means ? SEM of more than three independent studies, each
performed in triplicate. Statistical analysis was performed by Student’s t test. Cell lysates analyzed by immunoblotting (Santa Cruz) demonstrated
downregulation of endogenous calpain (E). The results shown are representative of two independent experiments. n.s., not significant.
2724SHIRAHA ET AL.MOL. CELL. BIOL.
fact that two different tagged constructs acted indistinguish-
ably, expressed either transiently (single-cell calpain activation
[data not shown]) or stably.
A second caveat is that we have not mapped the PKA phos-
phorylation site directly. This was not attempted either in vitro
or in vivo. In vitro, purified m-calpain is multiply phosphory-
lated on both serines and threonines even in its nonactivated
state (data not shown; Cong et al., Abstr. Am. Soc. Cell Biol.
40th Annu. Meet., 2000). In addition, nonphysiological conser-
vative replacement of serine 369 could easily shift the PKA
phosphorylation to the adjacent threonine 370. In vivo, the
multiple, seemingly cotranslational phosphorylation would
confound attempts to metabolically label m-calpain. Further-
more, as EGF activates on the small fraction of m-calpain that
is in the plasma membrane (18), we are likely dealing with a
substantially substoichiometric and potentially short-lived
modification. However, in our preliminary study, we could
detect low-level phosphorylation of wild-type hCANP caused
by PKA activator CPT-cAMPS stimulation, but not in
ST369AA hCANP (data not shown). As J. Cong et al. (Abstr.
Am. Soc. Cell Biol. 40th Annu. Meet., 2000) reported, calpain
can be phosphorylated even in a nonactivated state; the eleva-
tion in signal above the background phosphorylation of
hCANP appeared to be less than twofold. We speculate that
only a low level of phosphorylation is noted because PKA
phosphorylation is a regulatory event that in vivo may both
only involve a small fraction of total cell m-calpain and be
transient or lead to rapid m-calpain degradation. Nevertheless,
PKA will phosphorylate wild-type but not ST369AA m-calpain
both in vitro and in vivo, strongly suggesting that this is the
target site. Additionally, the MAP2 cleavage assay data (Fig. 3)
demonstrate that removal of the PKA target site renders m-
calpain resistant to PKA. Rather, we hypothesized that PKA
directly phosphorylates the putative consensus site in the pro-
posed regulatory domain III (25). Alteration of this site by
replacement with alanines yields a construct resistant to both
PKA phosphorylation and enzymatic repression in vitro and
CPT-cAMPS attenuation in vivo. These findings demonstrate
that PKA phosphorylation of domain III prevents activation of
m-calpain and strongly support the structure-based prediction
that domain III serves as a regulatory domain (46).
A third caveat is that calpain modulation may also occur via
actions on proteins other than m-calpain. cAMP-dependent
PKA also has been reported elsewhere to phosphorylate cal-
pastatin (32), the endogenous calpain inhibitor, with this phos-
phorylation affecting the distribution of calpastatin in neuro-
blastoma cells (3). We do not exclude the possibility, implied in
these reports, that phosphorylation of calpastatin by PKA
FIG. 7. Ribbon diagram of calpain (A) and close-up view of the S369 region (B). Calpain domains are indicated. The box represents the
approximate area which is shown in close-up (B). Interaction distances are also given, along with three key amino acids involved in the interaction
upon phosphorylation of S369.
VOL. 22, 2002PKA INHIBITS m-CALPAIN ACTIVATION2725
might affect the regulation of calpain. However, our hypothesis
that PKA direct phosphorylation of calpain inhibits calpain
activation was sufficiently verified by both in vitro and in vivo
experiments. These findings herein strongly suggest that direct
phosphorylation of m-calpain is the major regulatory mecha-
nism preventing EGF-induced m-calpain activation in fibro-
blasts. Another study reports that in vitro serine phosphoryla-
tion of bovine m-calpain, at the equivalent of S369, by
calmodulin-dependent protein kinase II increases general cal-
pain activity (37). The relevance of this finding to our report is
uncertain for two reasons. First, CaM kinase II phosphorylated
only autoproteolyzed m-calpain and had no effect on full-
length calpain; this is interesting in light of full-length calpain
now being considered to be as active as the autolyzed form,
which might simply be an intermediary of the degradative
attenuation process (28). Second, this report examined calpain
activity only in vitro in the presence of supraphysiological con-
centrations of the activator calcium, whereas our studies were
performed under cytosolic calcium concentrations in vivo; it is
conceivable that opposite effects of phosphorylation at identi-
cal sites could be seen under such diverse circumstances.
Computer modeling of the phosphorylation at S369 supports
our in vivo findings. We chose to focus on S369 since this is the
best consensus PKA site and has been reported elsewhere to
be phosphorylated as determined by phosphopeptide mapping
(J. Cong et al., Abstr. Am. Soc. Cell Biol. 40th Annu. Meet.,
2000); T370 was also mutated to prevent possibly nonphysi-
ological usage of an alternate phosphorylation acceptor. With-
out doubt, the multiple interactions enabled by S369-P add
new constraints in the interface between domain III and do-
main IV. In the proposed activation mode of calpain upon
addition of calcium, various domains of calpain would undergo
domain movement in the process of assembling the active site
(22). Experimental evidence supporting this hypothesis has
been recently reported (23), where disruption of critical inter-
domain constraints resulted in an increase in calcium sensitiv-
ity. By the same reasoning, if the extra constraints were added
then opposite effects would occur. In the case of calpain 3 (or
p94), a thorough structural analysis has again revealed that the
effect on interdomain movement is crucial for the activity (26).
In the case of S369-P, S369 of domain III is strategically lo-
cated at the interface between domains III and IV. Phosphor-
ylation of S369 not only provides a highly charged group but,
more importantly, “extends” the length of the side chain. Thus,
it enables interactions with a couple of residues of domain IV.
As a consequence, these interactions give rise to extra con-
straints in the interface, thereby severely restricting the free-
dom of both domains. The PKA consensus sequence RRxS369
is present only in the closely related ?-calpain (calpain I) and
the testis-specific calpain 11 (45). However, it remains to be
demonstrated experimentally whether this other ubiquitous
calpain is negatively attenuated by PKA since only R628 but
not H643 is present for cross-bridging in ?-calpain. Parenthet-
ically, the supraphysiological concentrations of calcium used to
demonstrate CaM kinase II-induced calpain activity (37) might
either overcome this movement restriction or disrupt salt
bridging. The rigidification imposed by phosphorylation of
S369 would essentially hamper the movement of these domains
in the assembly of the active site and result in the loss of
activity. This modeling, by being theoretical like all modeling,
provides a potential molecular basis for the inhibitory action of
PKA and forwards predictions for both m-calpain structure
and dominance of inhibitory signals that might guide future
experimental studies that lie beyond the scope of the present
Stable transfections yielded only a fractional increase in total
calpain levels. This was not surprising. Initially, we attempted
overexpression of both wild-type and ST369AA calpains using
the strong CMV promoter. In transient transfections, robust
GFP fluorescence was noted shortly after electroporation, but
most of the cells rounded and detached within 24 h (data not
shown). We did not pursue whether this was due to calpain-
mediated deadhesion (6, 11, 12, 17) or actual apoptosis as it lay
beyond the scope of the present study. Calpain has been im-
plicated elsewhere in some mechanisms of apoptosis (43), and
excess calpain activity might trigger caspase-mediated apopto-
sis (38). We established NR6 cell sublines containing the
MMTV-driven calpain constructs. Even in the presence of
dexamethasone, we attained exogenous expression at only
?30% of m-calpain. That this was sufficient to transmit EGFR-
mediated calpain activity and motility in the presence of CPT-
cAMPS suggests that endogenous calpain levels are in excess
of those needed for robust deadhesion during motility. This is
consistent with our ancillary studies that find that EGFR-me-
diated deadhesion requires only the submembrane subset of
In our previous paper, we presented evidence that the coun-
terregulatory ELR-negative CXC chemokines inhibit EGF-
induced cell migration but not proliferation (42). Herein, we
demonstrate that this occurs via direct PKA phosphorylation
of m-calpain. This provides for testable hypotheses concerning
fibroblast functioning during wound repair. In the inflamma-
tory and reparative stages the high levels of EGFR ligands in
the wound bed would promote repopulation through both mo-
tility and mitogenesis. Later in the resolution phase, the pres-
ence of IP-10 from ingrowing endothelial cells (19) and a
related CXCR3-binding chemokine from basal keratinocytes
(IP-9 or I-TAC) (9) would channel the motile phenotype to
matrix contraction (2). As cAMP has been shown elsewhere to
be antiproliferative in fibroblasts (10, 15, 16, 20), a second
PKA-mediated pathway would limit fibroplasia.
We thank Latha Satish, Philip Chang, and Douglas A. Lauffen-
burger for comments and discussions.
This work was supported by a grant from the National Institute of
General Medical Sciences (NIGMS/NIH).
1. Adachi, Y., N. Kobayashi, T. Murachi, and M. Hatanaka. 1986. Ca2?-
dependent cysteine proteinase, calpains I and II are not phosphorylated in
vivo. Biochem. Biophys. Res. Commun. 136:1090–1096.
2. Allen, F. D., C. F. Asnes, P. Chang, E. L. Elson, D. A. Lauffenburger, and, A.
Wells. EGF-induced matrix contraction is modulated by calpain. Wound
Repair Regen., in press.
3. Averna, M., R. de Tullio, M. Passalacqua, F. Salamino, S. Pontremoli, and
E. Melloni. 2001. Changes in intracellular calpastatin localization are medi-
ated by reversible phosphorylation. Biochem. J. 354:25–30.
4. Brown, G. L., L. B. Nanney, J. Griffen, A. B. Cramer, J. M. Yancey, L. J.
Curtsinger, L. Holtzin, G. S. Schultz, M. J. Jurkiewicz, and J. B. Lynch.
1989. Enhancement of wound healing by topical treatment with epidermal
growth factor. N. Engl. J. Med. 321:76–79.
5. Carafoli, E., and M. Molinari. 1998. Calpain: a protease in search of a
function? Biochem. Biophys. Res. Commun. 247:193–203.
2726SHIRAHA ET AL.MOL. CELL. BIOL.
6. Carragher, N. O., B. Levkau, R. Ross, and E. W. Raines. 1999. Degraded Download full-text
collagen fragments promote rapid disassembly of smooth muscle focal ad-
hesions that correlates with cleavage of pp125FAK, paxillin, and talin. J. Cell
7. Chen, P., K. Gupta, and A. Wells. 1994. Cell movement elicited by epidermal
growth factor receptor requires kinase and autophosphorylation but is sep-
arable from mitogenesis. J. Cell Biol. 124:547–555.
8. Chen, P., H. Xie, and A. Wells. 1996. Mitogenic signaling from the EGF
receptor is attenuated by a phospholipase C-?/protein kinase C feedback
mechanism. Mol. Biol. Cell 7:871–881.
9. Cole, K. E., C. A. Strick, T. J. Paradis, K. T. Ogborne, M. Loetscher, R. P.
Gladue, W. Lin, J. G. Boyd, B. Moser, D. E. Wood, B. G. Sahagan, and K.
Neote. 1998. Interferon-inducible T cell alpha chemoattractant (I-TAC): a
novel non-ELR CXC chemokine with potent activity on activated T cells
through selective high affinity binding to CXCR3. J. Exp. Med. 187:2009–
10. Cook, S. J., and F. McCormick. 1993. Inhibition by cAMP of Ras-dependent
activation of Raf. Science 262:1069–1072.
11. Cooray, P., Y. Yuan, S. M. Schoenwaelder, C. A. Mitchell, H. H. Salem, and
S. P. Jackson. 1996. Focal adhesion kinase (pp125FAK) cleavage and regu-
lation by calpain. Biochem. J. 318:41–47.
12. Du, X., T. C. Saido, S. Tsubuki, F. E. Indig, M. J. Williams, and M. H.
Ginsberg. 1995. Calpain cleavage of the cytoplasmic domain of the integrin
?3subunit. J. Biol. Chem. 270:26146–26151.
13. Elce, J. S., C. Hegadorn, and J. S. Arthur. 1997. Autolysis, Ca2?require-
ment, and heterodimer stability in m-calpain. J. Biol. Chem. 272:11268–
14. Engelhardt, E., A. Toksoy, M. Goebeler, S. Debus, E. B. Brocker, and R.
Gillitzer. 1998. Chemokines IL-8, GRO?, MCP-1, IP-10, and Mig are se-
quentially and differentially expressed during phase-specific infiltration of
leukocyte subsets in human wound healing. Am. J. Pathol. 153:1849–1860.
15. Friedman, D. L., R. A. Johnson, and C. E. Zeilig. 1976. The role of cyclic
nucleotides in the cell cycle. Adv. Cyclic Nucleotide Res. 7:69–114.
16. Froehlich, J. E., and M. Rachmeler. 1972. Effect of adenosine 3?-5?-cyclic
monophosphate on cell proliferation. J. Cell Biol. 55:19–31.
17. Glading, A., P. Chang, D. A. Lauffenburger, and A. Wells. 2000. Epidermal
growth factor receptor activation of calpain is required for fibroblast motility
and occurs via an ERK/MAP kinase signaling pathway. J. Biol. Chem. 275:
18. Glading, A., F. Uberall, S. M. Keyse, D. A. Lauffenburger, and A. Wells.
2001. Membrane-proximal ERK signaling is required for M-calpain activa-
tion downstream of epidermal growth factor receptor signaling. J. Biol.
19. Goebeler, M., T. Yoshimura, A. Toksoy, U. Ritter, E. B. Bröcker, and R.
Gillitzer. 1997. The chemokine repertoire of human dermal microvascular
endothelial cells and its regulation by inflammatory cytokines. J. Investig.
20. Halprin, K. M. 1976. William Montagna Lecture. Cyclic nucleotides and
epidermal cell proliferation. J. Investig. Dermatol. 66:339–343.
21. Hirose, K., S. Kadowaki, M. Tanabe, H. Takeshima, and M. Iino. 1999.
Spatiotemporal dynamics of inositol 1,4,5-trisphosphate that underlies com-
plex Ca2?mobilization patterns. Science 284:1527–1530.
22. Hosfield, C. M., J. S. Elce, P. L. Davies, and Z. Jia. 1999. Crystal structure
of calpain reveals the structural basis for Ca2?-dependent protease activity
and a novel mode of enzyme activation. EMBO J. 18:6880–6889.
23. Hosfield, C. M., T. Moldoveanu, P. L. Davies, J. S. Elce, and Z. Jia. 2000.
Calpain mutants with increased Ca2?sensitivity and implications for the role
of the C2-like domain. J. Biol. Chem. 276:7404–7407.
24. Huttenlocher, A., S. P. Palecek, Q. Lu, W. Zhang, R. L. Mellgren, D. A.
Lauffenburger, M. H. Ginsberg, and A. F. Horwitz. 1997. Regulation of cell
migration by the calcium-dependent protease calpain. J. Biol. Chem. 272:
25. Imajoh, S., K. Aoki, S. Ohno, Y. Emori, H. Kawasaki, H. Sugihara, and K.
Suzuki. 1988. Molecular cloning of the cDNA for the large subunit of the
high-Ca2?-requiring form of human Ca2?-activated neutral protease. Bio-
26. Jia, Z., V. Petrounevitch, A. Wong, T. Moldoveanu, P. L. Davies, J. S. Elce,
and J. S. Beckmann. 2001. Mutations in calpain 3 associated with limb girdle
muscular dystrophy: Analysis by molecular modelling and by mutation in
m-calpain. Biophys. J. 80:2590–2596.
27. Johnson, G. V., and V. G. Foley. 1993. Calpain-mediated proteolysis of
microtubule-associated protein 2 (MAP-2) is inhibited by phosphorylation
by cAMP-dependent protein kinase, but not by Ca2?/calmodulin-dependent
protein kinase II. J. Neurosci. Res. 34:642–647.
28. Johnson, G. V., and R. P. Guttmann. 1997. Calpains: intact and active?
29. Johnson, G. V., R. S. Jope, and L. I. Binder. 1989. Proteolysis of tau by
calpain. Biochem. Biophys. Res. Commun. 163:1505–1511.
30. Johnson, G. V., J. M. Litersky, and R. S. Jope. 1991. Degradation of micro-
tubule-associated protein 2 and brain spectrin by calpain: a comparative
study. J. Neurochem. 56:1630–1638.
31. Kraulis, P. J. 1991. MOLSCRIPT: a program to produce both detailed and
schematic plots of protein structures. J. Appl. Crystallogr. 24:946–950.
32. Kuo, W. N., U. Ganesan, D. L. Davis, and D. L. Walbey. 1994. Regulation of
the phosphorylation of calpain II and its inhibitor. Mol. Cell. Biochem.
33. Lane, R. D., D. M. Allan, and R. L. Mellgren. 1992. A comparison of the
intracellular distribution of mu-calpain, m-calpain, and calpastatin in prolif-
erating human A431 cells. Exp. Cell Res. 203:5–16.
34. Lauffenburger, D. A., and A. F. Horwitz. 1996. Cell migration: a physically
integrated molecular process. Cell 84:359–369.
35. Litersky, J. M., and G. V. Johnson. 1992. Phosphorylation by cAMP-depen-
dent protein kinase inhibits the degradation of tau by calpain. J. Biol. Chem.
36. Maheshwari, G., A. Wells, L. G. Griffith, and D. A. Lauffenburger. 1999.
Biophysical integration of effects of epidermal growth factor and fibronectin
on fibroblast migration. Biophys. J. 76:2814–2823.
37. McClelland, P., L. P. Adam, and D. R. Hathaway. 1994. Identification of a
latent Ca2?/calmodulin dependent protein kinase II phosphorylation site in
vascular calpain II. J. Biochem. 115:41–46.
38. Nakagawa, T., and J. Yuan. 2000. Cross-talk between two cysteine protease
families: activation of caspase-12 by calpain in apoptosis. J. Cell Biol. 150:
39. Nanney, L. B., S. Paulsen, M. K. Davidson, N. L. Cardwell, J. S. Whitsitt,
and J. M. Davidson. 2000. Boosting epidermal growth factor receptor ex-
pression by gene gun transfection stimulates epidermal growth in vivo.
Wound Repair Regen. 8:117–127.
40. Palecek, S. P., A. Huttenlocher, A. F. Horwitz, and D. A. Lauffenburger.
1998. Physical and biochemical regulation of integrin release during rear
detachment of migrating cells. J. Cell Sci. 111:929–940.
41. Rosser, B. G., S. P. Powers, and G. J. Gores. 1993. Calpain activity increases
in hepatocytes following addition of ATP. Demonstration by a novel fluo-
rescent approach. J. Biol. Chem. 268:23593–23600.
42. Shiraha, H., A. Glading, K. Gupta, and A. Wells. 1999. IP-10 inhibits epi-
dermal growth factor-induced motility by decreasing epidermal growth fac-
tor receptor-mediated calpain activity. J. Cell Biol. 146:243–253.
43. Solary, E., B. Eymin, N. Droin, and M. Haugg. 1998. Proteases, proteolysis,
and apoptosis. Cell Biol. Toxicol. 14:121–132.
44. Sorimachi, H., S. Ishiura, and K. Suzuki. 1997. Structure and physiological
function of calpains. Biochem. J. 328:721–732.
45. Sorimachi, H., and K. Suzuki. 2001. The structure of calpain. J. Biochem.
46. Strobl, S., C. Fernandez-Catalan, M. Braun, R. Huber, H. Masumoto, K.
Nakagawa, A. Irie, H. Sorimachi, G. Bourenkow, H. Bartunik, K. Suzuki,
and W. Bode. 2000. The crystal structure of calcium-free human m-calpain
suggests an electrostatic switch mechanism for activation by calcium. Proc.
Natl. Acad. Sci. USA 97:588–592.
47. Ware, M. F., A. Wells, and D. A. Lauffenburger. 1998. Epidermal growth
factor alters fibroblast migration speed and directional persistence recipro-
cally and in a matrix-dependent manner. J. Cell Sci. 111:2423–2432.
48. Wells, A., K. Gupta, P. Chang, S. Swindle, A. Glading, and H. Shiraha. 1998.
Epidermal growth factor receptor-mediated motility in fibroblasts. Microsc.
Res. Tech. 43:395–411.
49. Xie, H., M. A. Pallero, K. Gupta, P. Chang, M. F. Ware, W. Witke, D. J.
Kwiatkowski, D. A. Lauffenburger, J. E. Murphy-Ullrich, and A. Wells. 1998.
EGF receptor regulation of cell motility: EGF induces disassembly of focal
adhesions independently of the motility-associated PLC? signaling pathway.
J. Cell Sci. 111:615–624.
VOL. 22, 2002PKA INHIBITS m-CALPAIN ACTIVATION2727