Eur. J. Biochem.

236, 796-805 (1996)

0 FEBS 1996

Cellular stresses and cytokines

activate multiple mitogen-activated-protein kinase kinase homologues
in PC12 and KB cells
Roger MEIER', John ROUSE', Ana CUENDA', Angel R. NEBREDA2 and Philip COHEN'
' MRC Protein Phosphorylation Unit, Department of Biochemistry, University of Dundee, Scotland
lmperial Cancer Research Fund, Clare Hall Laboratories, South Mimms, England

(Received 2,November 1995) - EJB 95 1807/1

The identities of the upstream activators of the mitogen-activated protein (MAP) kinase homologues
termed stress-activated-protein (SAP) kinase-I (also known as JNK or SAPK) and SAP kinase-2 (also
known as p38, RK and CSBP) were investigated in rat PC12 cells and human KB cells after exposure to
cellular stresses and cytokines. In PC12 cells, the same two upstream activators, SAP kinase kinase-I
(SAPKK-1) and SAPKK-2 were activated after exposure to osmotic shock, ultraviolet irradiation or the
protein synthesis inhibitor anisomycin, and more weakly in response to sodium arsenite. SAPKK-1 was
capable of activating both SAP kinase-1 and SAP kinase-2 and was similar, if not identical, to the
previously described MAP kinase kinase homologue MKK4, as judged by immunological criteria and by
its ability to be activated by MEK kinase in vitro. In contrast, SAPKK-2 activated SAP kinase-2, but not
SAP kinase-1 in vitro. In KB cells, five distinct upstream activators of SAP kinase-1 and SAP kinase-2
were induced, namely SAPKK-I, SAPKK-2, SAPKK-3, SAPKK-4 and SAPKK-5, whose appearance
depended on the nature of the stimulus. SAPKK-3, which was strongly induced by every stimulus tested
(osmotic shock, ultraviolet irradiation, anisomycin or IL-l), accounted for about 95% of the SAP
kinase-2 activator activity in these cells, did not activate SAP kinase-l and eluted from Mono S at a
lower salt concentration than SAPKK-2. SAPKK-4 and SAPKK-5 were also eluted from Mono S at
higher NaCl concentrations than SAPKK-3 and these enzymes activated SAP kinase-I but not SAP
kinase-2. SAPKK-4 was the only SAP kinase-I activator induced by interleukin-1 or ultraviolet irradia-
tion, while two SAP kinase-I activators, SAPKK-1 and SAPKK-5, were induced by osmotic shock or
anisomycin. SAPKK-2, SAPKK-3, SAPKK-4 and SAPKK-5 were not activated by MEK kinase in vitro,
were separable from the major activator(s) of p42 MAP kinase, and were not recognised by anti-MKK4
antibodies. At least two of these enzymes are likely to be novel MAP kinase kinase homologues. Our
results demonstrate unexpected complexity in the upstream regulation of stress and cytokine-stimulated
kinase cascades and indicate that the selection of the appropriate SAPKK varies with both the stimulus
and the cell type.
Keywords: mitogen-activated-protein kinase ; stress-activated protein kinase ; HOG1 ; cellular stress ;
cytokine.

Two mitogen-activated protein (MAP) kinase homologues
have been identified in mammalian cells which are activated
by cellular stresses (such as chemical, heat and osmotic shock,
ultraviolet radiation or inhibitors of protein synthesis), by bacte-
Correspondence to P. Cohen, MRC Protein Phosphorylation Unit,
Department of Biochemistry, University of Dundee, DD1 4HN, Scot-
land
Abbreviations. MAP, mitogen-activated protein; SAP, stress-acti-
vated protein; SAPK, SAP kinase; SAPKK, SAPK kinase; MAPKAP,
MAP-kinase-activated protein; MAPKK, MAP kinase kinase; MEK and
MKK, homologues of MAPK; MEKK, MEK kinase; 1L-I, interleukin
1a ; PP2A, protein-serinelthreonine phosphatase 2A; PTP, protein-tyro-
sine phosphatase ; XMEK, Xenopus homologue of MEK; XMpk2, Xeno-
pus homologue of SAPK-2; GSH, glutathione; GST, glutathione S-trans-
ferase ; PKI, specific protein inhibitor of cyclic AMP-dependent protein
kinase.
rial endotoxin and by the cytokines interleukin-1 (IL-I) and
tumour necrosis factor-a [I] and have therefore been termed
stress-activated protein (SAP) kinases. Isoforms of SAP
kinase-l (also termed JNKs) phosphorylate the transcription
factor c-Jun, which is thought to stimulate its transactivation
function [2-51, while isoforms of SAP kinase-2 (previously
termed p38 161, p40 171, RK [8] or CSBP 191) activate MAP-
kinase-activated protein (MAPKAP) kinase-2 [S], one of whose
intracellular targets is the small heat shock protein HSP27
[ 10- 121.
The MAP kinases that participate in the classical growth-
factor-stimulated pathway (p42 and p44 MAP kinases) become
active only when a Thr and a Tyr residue [ 131 in a Thr-Glu-Tyr
[14] motif are phosphorylated, these residues being located in
subdomain VIII of the kinase catalytic domain, just N-terminal
to the conserved Ala-Pro-Glu motif. The activation of SAP
kinases requires a similar dual Thr/Tyr phosphotylation, the Thr-

Meier et al. (Eur: J. Biochern. 236)
Pro-Tyr [4, 5, IS] (SAP kinase-I) and Thr-Gly-Tyr [6, 8, 16, 171
(SAP kinase-2) motifs being situated in the same region of these
enzymes. The activation of p42 and p44 MAP kinases is cata-
lysed by dual specificity enzymes, termed MAP kinase
kinase-l (MAPKK- 1) and MAPKK-2, which phosphorylate the
Thr and the Tyr residues [IS], although the latter residue is
targetted preferentially [191. These enzymes do not phosphoryl-
ate SAP kinase-1 [4, S] or SAP kinase-2 (7, 81. Conversely,
activators of SAP kinase-2 detected in chemically stressed PC12
cells [8] or IL-I-stimulated KB cells [7] failed to activate p42
MAP kinase, and their inactivation by the protein-serinehhreo-
nine phosphatase PP2A, but not by protein-tyrosine phospha-
tases [8] indicated that they were probably MAPKK homo-
logue(s).
Recently, cDNAs encoding two novel MAPKK homologues
were isolated, termed SEKl [20] (or MKK4) [21] and MKK3
[21]. MKK4, and its Xenopus homologue XMEK2 1221 were
found to activate both SAP kinase-I [20, 211 and SAP kinase-2
[16, 211 in vitro. In contrast, MKK3 activated SAP kinase-2 in
vitro, but not SAP kinase-l 1211. MKK4 and XMEK2 phosphor-
ylated the Tyr residue much faster than the Thr residue in the
Thr-Pro-Tyr and Thr-Gly-Tyr motifs of SAP kinase-l [20] and
SAP kinase-2 [16]. In 293 cells transfected with cDNA encoding
MKK4, subsequent stimulation with tumor necrosis factor-a
or anisomycin (a protein synthesis inhibitor) activated the
transfected MKK4 and enhanced activation of cotransfected
SAP kinase-I. Moreover, the activation of transfected SAP
kinase-I by anisomycin was prevented by cotransfection of a
dominant negative (inactive) form of MKK4 [20]. Cotransfec-
tion of MKK4 and SAP kinase-l also activated the latter in
COS-1 cells [21].
MEK kinase (MEKK) [23] is a mammalian homologue of
STE11, a protein kinase that activates the MAPKK homologue
STE7 in a pathway which coordinates the physiological changes
that allow yeast cells to mate. Although originally found to acti-
vate MAPKK-1 in vitro, and therefore thought to be involved in
the classical MAP kinase pathway, MEKK was shown subse-
quently to activate MKK4 in vitro 124, 251 and to induce activa-
tion of cotransfected SAP kinase-1 at DNA concentrations 100-
fold lower than those which activate cotransfected p42 MAP
kinase [24, 261. Moreover, MEKK is activated by osmotic shock
in rat 3Y1 fibroblasts and PC12 cells [2S], and a dominant nega-
tive mutant of MKK4 blocked the activation of SAP kinase-l
induced by MEKK in L929 cells [24]. In contrast, MEKK was
reported to cause little or no activation of SAP kinase-2 when
cotransfected with MKK4 in COS-1 or HeLa cells [21, 261.
These experiments therefore suggested that, although transfec-
tion of COS-I cells with MKK4 induces the activation of co-
transfected SAP kinase-2 [21], MKK4 may be specific for SAP
kinase-I in viva In contrast, transfection of COS-I cells with
MKK3 activated cotransfected SAP kinase-2, but not SAP ki-
nase-I, consistent with the specificity of this enzyme observed
in vitro [21].
Although MKK4 activates SAP kinase-I and SAP kinase-2
in vitro or after transfection of its DNA into cells, and MKK3
activates SAP kinase-2 in vitro or after transfection, these
experiments are not physiological, and it remains possible that
other (or additional) MAPKK homologues activate SAP ki-
nase-1 and SAP kinase-2 in vivo. Here we have studied which
upstream activators of SAP kinase-I and SAP kinase-2 are in-
duced when PC12 or KB cells are exposed to a variety of stim-
uli. These experiments indicate that distinct activators of SAP
kinase-I and SAP kinase-2 are induced in different cells, even
by the same stimuli, and that at least one (and probably several)
activators distinct from MKK3 and MKK4 are activated in re-
sponse to cellular stresses and cytokines.
14321033, 1996, 3, Downloaded from https://febs.onlinelibrary.wiley.com/doi/10.1111/j.1432-1033.1996.00796.x by Nat Prov Indonesia, Wiley Online Library on [15/07/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
797
MATERIALS AND METHODS
Materials. Anisomycin, sorbitol and glutathione(GSH)-
agarose were purchased from Sigma, amylose-agarose from
New England Biolabs, recombinant human interleukin-la (IL-1)
from Boehringer, microcystin-LR from Gibco and sodium meta-
arsenite from Fluka Chemicals. Protein-tyrosine phosphatase
(PTP) 1B expressed in Escherichia coli was a gift from Dr D.
Barford (Department of Biochemistry, University of Oxford).
MAPKAP kinase-2(46-400) lacking the N-terminal proline-
rich domain [27] provided by Dr R. Ben-Levy and Dr C. J.
Marshall (Institute for Cancer Research, London) and the N-
terminal 194 residues of c-Jun [c-Jun(1 - 194)] provided by Dr
Richard Treisman (Imperial Cancer Research Fund, London)
were expressed in E. coli as glutathione S-transferase (GST) fu-
sion proteins and purified on GSH-agarose. The Xenopus homo-
logue of SAP kinase-2 (XMpk2) was expressed in E. coli as a
Mal-E fusion protein and purified on amylose-agarose 11 61,
while MEKK (provided by Dr G. L. Johnson, National Jewish
Center for Immunology and Respiratory Medicine, Denver) was
expressed in E. coli as a fusion protein containing six histidine
residues at the C-terminus and purified on nickel-nitrilotriace-
tate-agarose (Hybaid, Middlesex, UK). The MEKK cDNA was
amplified by PCR with oligonucleotides designed to create an
NdeI site at Met16 in the reported sequence [23] and a BuinHI
site in front of the TAG stop codon, which changed the C-termi-
nal Trp to Asp and removed the stop codon. The PCR product
was digested with NdeI and BamHI and ligated into the NdeI-
BumHI site of the pET2lb vector (Novagen). Protein phospha-
tase 2A, purified from rabbit skeletal muscle [28] was provided
by Dr N. Morrice in the Dundee laboratory. PKI, the specific
peptide inhibitor of CAMP-dependent protein kinase (TTYADFI-
ASGRTGRRNAIHD) and all other peptides were synthesized
by Mr F. B. Caudwell in the MRC Protein Phosphorylation Unit.
The peptide GPIGPPASGHPAVSS corresponding to residues
20-34 of murine MKK4 1201 was conjugated to bovine serum
albumin using glutaraldehyde and injected into sheep at the
Scottish Antibody Production Unit (Carluke, Ayrshire). Anti-
MKK4 antibodies were obtained by affinity purification of the
antiserum on a peptide-CH-Sepharose column. The peptide
EQMPVSPSSPMYVD, corresponding to the C-terminal 14 resi-
dues of XMEK2 [22], a Xenopus homologue of MKK4, was
conjugated to keyhole limpet haemocyanin and injected into
rabbits, to raise anti-XMEK2 antibodies.
Purification of SAP kinase-1. Wistar rats (200 g) were in-
jected intraperitoneally with cycloheximide ( 5 mg/loO g of body
mass) [29]. After 1 h, the rats were killed by stunning and the
livers removed, homogenised in 2.5 ml/g of 15 mM sodium
pyrophosphate pH6.8, 5 mM EDTA, 25mM NaF, 0.4mM
phenylmethylsulfonyl fluoride, 1 mM benzamidine, 2 mM so-
dium orthovanadate, 5 pM leupeptin and centrifuged for 20 min
at 2OOOOXg. The supernatant was filtered through glass wool
and loaded on to a column of Q-Sepharose (10X1.6 cm) equili-
brated at 0-4°C in SO mM Tris/HCl pH 7.4, 2 mM EDTA,
2 mM EGTA, 0.03% (by mass) Brij-35, 5 % (by vol) glycerol,
1 mM benzamidine, 0.1 % (by vol.) 2-mercaptoethanol, 0.3 mM
sodium orthovanadate (buffer A). The column was developed
with a 400-ml linear gradient of 0-0.7 M NaCl in buffer A at
a flow rate of 3 ml/min and fractions of 5 ml were collected.
Two peaks of c-Jun kinase activity were eluted, the major one
peaking at 0.09 M NaCl and the minor one at 0.18 M NaCi. In-
gel kinase assays revealed that their apparent molecular masses
were 50 kDa and 55 kDa, indicating that they correspond to the
two forms of SAP kinase-ln purified from IL-I-stimulated rab-
bit liver [30].The fractions from 50-130 mM NaCl. which con-
tained the major 50-kDa form of SAP kinase-la activity, were

798 Meier et al. ( E M J. Biochem. 236)
pooled and applied to a 3-ml column of phenyl-Sepharose equil-
ibrated in 50 mM Tris/HCl, 20 mM sodium glycerol 2-phosphate
pH 7.4, 0.1 % (by vol.) 2-mercaptoethanol, 1 mM benzamidine
(buffer B) containing 0.5 M NaC1. After washing with equilibra-
tion buffer until the absorbance at 280 nm was <0.01 and then
with buffer B, SAP kinase-1 was eluted with buffer B plus 50%
ethylene glycol. The active fractions were pooled, dialysed
against buffer A, concentrated by ultrafiltration through a
Centricon-30 membrane (Amicon) and stored at -70°C. The
SAP kinase-1 was free of p42 or p44 MAP kinase which eluted
from Q-Sepharose at higher NaCl concentrations because, after
inactivation by PTP-1B (see below), no reactivation occurred
after subsequent incubation with MgATP and MAPKK-1.
Culture and stimulation of cells. PC12 [31] and KB [7]
cells were cultured as described previously and then either ex-
posed to chemical (0.5 mM sodium arsenite) or osmotic (0.5 M
sorbitol) stress, or stimulated with IL-1 (20 ng/ml) or aniso-
mycin (10 pg/ml) for different times. In other experiments cells
were exposed to ultraviolet radiation (60 J/mz) using a Spectro-
linker XL-1500 UV Crosslinker (Spectronics Corporation, sup-
plied by Stratagene) fitted with an ultraviolet-C source and then
incubated at 37°C for up to 30 min before cell lysis.
Chromatography of cell lysates on Mono S. Cells were
lysed as described [8], except that 2 pM microcystin was also
present in the lysis buffer. The lysates were then chromato-
graphed on Mono S [8] except that 400 pg protein was
loaded on to a column (5X0.16 cm) of Mono S attached to a
SMART FPLC System (Pharmacia). The volume of the NaCl
gradient was 4 ml and fractions of 0.05 ml or 0.1 ml were
collected.
Protein kinase assays. SAP kinuse-1. A solution (40 pl)
containing SAP kinase-l and 200 pg/ml GST-c-Jun(1-194) in
25 mM Tris/HCI, 20 mM sodium glycerol 2-phosphate pH 7.4,
0.1 mM EGTA, 1 mM sodium orthovanadate, 2.5 pM PKI, was
incubated for 3 min at 30°C and the assay initiated with 10 p1
50 mM MgCl,, 0.5 mM [y-'2P]ATP (5x10' cpdnmol). After a
further 20 min, a 40-p1 aliquot was withdrawn, spotted on to a
square (1 SX1.5 cm) of Whatman P81 phosphocellulose paper
and immersed immediately in 75 mM phosphoric acid. After
washing the papers [32), "P radioactivity incorporated into c-
Jun(1-194) was measured. One unit (U) of SAP kinase-1 activ-
ity was that amount which catalysed the phosphorylation of
1 nmol GST-c-Jun(l - 194) in 1 min.
Inactivation of SAP kinase-1. SAP kinase-1 (100 U/ml) was
incubated for 60 min at 30°C with PTP-1B (0.3 mg/ml) and
chromatographed on a column (5X0.16cm) of Mono Q
equilibrated in buffer A. SAP kinase-1 was eluted with
buffer A plus 175 mM NaCl and separated from PTP-1B
which eluted at 0.3 M NaCl. The fractions containing SAP
kinase-l were pooled, concentrated to 0.1 5 ml by centrifugation
through a Centricon-30 membrane and stored in aliquots at
-70°C.
Assay of SAP kinase-l activators. 10 pl of a solution con-
taining SAP kinase activator and inactive SAP kinase-I (equiva-
lent to 100 mU before inactivation with PTP 1B) in 20 mM
sodium glycerol 2-phosphate pH 7.4, 1.0 mM EGTA, 0.03% (by
mass) Brij-35, 5 % (by vol.) glycerol, 1 mM benzamidine, 0.1 %
(by vol.) 2-mercaptoethanol, 1 mM sodium orthovanadate, 1 pM
microcystin-LR was incubated for 3 min at 30"C, and the reac-
tion initiated with 2 p1 60 mM MgCI,, 0.6 mM unlabeled ATP.
After 30 min at 3O"C, a 5-pl aliquot was withdrawn and assayed
for SAP kinase-1 as above. Control incubations were performed
in which SAP kinase-l activator or inactive SAP kinase-l were
omitted. Fractions containing SAP kinase-1 activator were
diluted appropriately in buffer A to ensure that initial rate
14321033, 1996, 3, Downloaded from https://febs.onlinelibrary.wiley.com/doi/10.1111/j.1432-1033.1996.00796.x by Nat Prov Indonesia, Wiley Online Library on [15/07/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
conditions were met. One unit (U) of SAP kinase-1 activator
was that amount which increased SAP kinase-I activity by
1 U/min.
Assay of SAP kinase-1 in cell extracts. 10 pg GST-cJun( 1-
194) bound to 5 p1 packed GSH-agarose beads was incubated
with cell lysate (0.1 mg) for 3 h at 4°C to allow SAP kinase-1
to bind to its substrate c-Jun(1-194). The beads were then
washed four times in phosphate-buffered saline, once in 25 mM
Tris/HCl, 20 mM sodium glycerol 2-phosphate pH 7.4, contain-
ing 10 mM MgCl,, 0.1 mM EGTA, 1 mM sodium orthovanadate
(buffer C) and resuspended in 27 p1 buffer C plus 2.5 pM PKI.
After 3 min at 30"C, the reaction was initiated with 3 pl 1 mM
[y-'2P]ATP (2X lo6 cpdnmol) and terminated after a further
20 min at 30°C by adding 5 pl 6% (by mass) SDS. A 30-p1
aliquot was withdrawn, electrophoresed on a 10% SDS/
polyacrylamide gel and autoradiographed. The '2P-labelled band
was excised and '*P measured.
Assay of SAP kinase-2. SAP kinase-2 was measured by the
activation of bacterially expressed inactive GST-MAPKAP
kinase-2(46-400). An aliquot (6 pl) of inactive MAPKAP
kinase-2 (0.4 pM) was incubated with 4 yl SAP kinase-2, and ,
after incubation for 3 min at 30"C, activation was initiated by
the addition of 2 pl 60 mM magnesium acetate, 0.6 mM unla-
belled ATP. After 30 min at 30"C, a 5-pl aliquot was removed
and assayed for MAPKAP kinase-2 activity at a further 10-fold
dilution using the peptide KKLNRTLSVA as substrate as
described previously [32]. One unit (U) of SAP kinase-2 was
that amount which increased the activity of MAPKAP ki~ase-2
by 1 U/min. One unit of MAPKAP kinase-2 was that amount
which catalysed the phosphorylation of 1 nmol peptide
KKLNRTLSVA/min.
Assuy of SAP kinase-2 activator: SAP kinase-2 activator
(3 p1 Mono S fraction) was incubated for 3 min at 30°C' with
6 y1 0.4 pM Mal-E-XMpk2 in 100mM Tris/HCI pK 7.4,
0.2 mM EGTA, 0.5 mM sodium orthovanadate, 0.1 % (b! vol.)
2-mercaptoethanol, and the reaction was initiated with 3 p1 of
40 mM magnesium acetate, 0.8 mM unlabelled ATP. After
30 min, 24 p1 ice-cold 50 mM Tris/HCl pH 7.4, 0.1 mM E:GTA,
0.5 mM sodium orthovanadate, 1 mg/ml bovine serum albumin,
0.1% (by vol.) 2-mercaptoethanol was added to stop the reac-
tion. An aliquot (4 pl) was withdrawn and assayed for SAP
kinase-2 activity as described above. SAP kinase-2 activator or
XMpk2 were omitted in control incubations, and these blanks
were subtracted from the value obtained in the presence of both
SAP kinase-2 activator and XMpk2. Fractions containing SAP
kinase-2 activator were diluted appropriately in buffer A to
ensure that initial rate conditions were met. One unit (U) of SAP
kinase-2 activator was that amount which increased the activity
of SAP kinase-2 by 1 U/min.
Immunoprecipitation of SAP kinase activators. 4 111 anti-
XMEK2 antiserum or 5 1-18affinity purified anti-MKK4 antibody
was bound to 5 pI (settled volume) protein-G-Sepharose equili-
brated in 20 mM sodium glycerol 2-phosphate pH 7.4, 1.5 mM
EGTA, 0.03% (by mass) Brij 35, 5 % (by vol.) glycerol, 1 mM
benzamidine, 0.1 mM phenylmethylsulphonyl fluoride and 0.1 %
(by vol.) 2-mercaptoethanol (buffer D) by incubation for 30 min
at 4°C and the beads washed three times in buffer D. An aliquot
(30 pl) of Mono-S-purified SAP kinase activator was added,
and, after mixing for 90min at 4"C, the suspension was
centrifuged for 2 min at 13000Xg and the supernatant assayed
for its ability to activate SAP kinase-1 and/or SAP kinase-2.
Control immunoprecipitations were performed in which the anti-
bodies were incubated with the relevant peptide immunogen
(1 mg/ml) for 30 min at 4°C prior to the addition of protein-G-
Sepharose.
 14321033, 1996, 3, Downloaded from https://febs.onlinelibrary.wiley.com/doi/10.1111/j.1432-1033.1996.00796.x by Nat Prov Indonesia, Wiley Online Library on [15/07/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Meier et al. ( E M J. Biochem. 236) 799
100 15
A sorbitol B uv
SAPKK-1

0.4 0.6 0.8 1.0 1 3 1.4 1.6 1.8 0.4 0.6 0.8 1.0 1 2 1.4 1.6 1.8

elution volume (ml) elution volume (ml)

15 0.5
C anisomycin D arsenite
0.4
8 Oa4
2 10 g 3,
; 0.3
v
:: 0.3
8
I
i t
g ;
0.08
'E
N
0.2

5 ; p!
0.2 g
$ 0.04
0.1
;4 0.1 $ 2
0 0 0 0
0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8

elution volume (ml) elution volume (ml)

Fig. 1. Two SAP kinase kinases are activated in PC12 cells in response to cellular stresses. Cell lysates (0.4 mg protein) from cells osmotically
shocked for 5 min with 0.5 M sorbitol (A), ultraviolet irradiated at 60 J/m*and then left for a further 30 min (B), stimulated for 30 min with 10 pg/
ml anisomycin (C), or stimulated for 15 min with 0.5 mM sodium arsenite (D), were chromatographed on Mono S; activators of SAP kinase-1 ( 0 )
and SAP lunase-2 (0)were assayed as described under Materials and Methods. The two major SAP kinase (SAPK) activators, SAPKK-1 and
SAPKK-2, are marked. The NaCl gradient (- - - -) was the same in each chromatography and is indicated only in D. Lysates from unstimukdted
cells were chromatographed separately on Mono S and assayed for SAPK-1 activator (m, A) and SAPK-2 activator (0, D). Similar results were
obtained in at least three different experiments for each stimulus.

RESULTS pitated almost quantitatively with either the antibody raised

Identification of SAP kinase kinases in PC12 cells. PC12 cells
are a rat adrenal tumour line which differentiate from immature
chromaffin cells to a sympathetic neurone-like phenotype in re-
sponse to nerve growth factor. Activation of the classical MAP
kinase pathway is required and sufficient to induce differentia-
tion 1331. SAP kinase-2 becomes activated when PC12 cells are
stimulated with sodium arsenite [8], a compound that mimics
many of the effects of heat shock on protein phosphorylation
and gene transcription. In the present study, osmotic shock, ul-
traviolet irradiation or exposure to the protein synthesis inhibitor
anisomycin were found to be much more potent inducers of
SAPKK activities than arsenite (Fig. 1). Activation by each ago-
nist was detectable after 5-15 min and sustained for at least
30 min (data not shown). When cell lysates were chromato-
graphed on Mono S, the same two peaks of SAPKK activity
were identified, irrespective of the stimulus to which the cells
had been exposed. The activity peaking at 0.9 ml was termed
SAPKK-1 while that eluting at 1.4 ml was termed SAPKK-2.
No SAPKK activity was detectable in unstimulated cells
(Fig. 1 A and D).
SAPKK-1 was capable of activating both SAP kinase-1 and
SAP kinase-2 and the activity ratio of SAP kinase-2 activator/
SAP kinase-1 activator in the peak tubes was about 40 with each
stimulus (Fig. 1). SAPKK-1 from PC12 cells was immunopreci-
against a peptide near the N-terminus of murine MKK4 (anti-
body N) or the antibody raised against the C-terminal peptide of
XMEK2 (a Xenopus homologue of MKK4) (antibody C). In
both cases, immunoprecipitation was specific because it was
prevented by prior incubation of the antibody with the appropri-
ate peptide immunogen (Fig. 2). A similar proportion of activity
was immunoprecipitated whether SAPKK-1 was assayed by the
activation of SAP kinase-2 (Fig. 2 A and B) or by the activation
of SAP kinase-1 (Fig. 2 C and D). The results are shown in
Fig. 2 for the SAPKK-1 activity stimulated by osmotic shock
and ultraviolet irradiation, but essentially the same results were
obtained after stimulation with anisomycin (data not shown).
These results demonstrate that SAPKK-1 is MKK4 or a very
closely related isoform. Consistent with its identity as MKK4,
the enzyme from osmotically shocked PC12 cells was inacti-
vated by PP2A, and its activity towards both SAP kinase-1 and
SAP kinase-2 was restored by incubation with MgATP and
MEKK (Fig. 3).
In contrast to SAPKK-1, SAPKK-2 activated SAP kinase-2
but not SAP kinase-1 (Fig. 1) and was not immunoprecipitated
by antibody C or antibody N (Fig. 2). The results are shown for
the SAPKK-2 activity stimulated by osmotic shock (Fig. 2A)
and ultraviolet irradiation (Fig. 2B), but essentially the same
results were obtained after stimulation with anisomycin (data not
 14321033, 1996, 3, Downloaded from https://febs.onlinelibrary.wiley.com/doi/10.1111/j.1432-1033.1996.00796.x by Nat Prov Indonesia, Wiley Online Library on [15/07/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
800 Meier et al. ( E m J. Biochern. 236)

peptideimmunogen - + - + .+ - + - + - + - + - +
antibody N N CC N N C C N N C C N N cc
form of SAPKK 1 1 2 2 1 1 2 2

sorbitol D w

peptide immunogen

antibody

form of SAPKK
N
1
L
N C

1
+
C N
-

1
+
N
-
c c
1
Fig. 2. Anti-MKK4 antibodies immunoprecipitate SAPKK-1, but not SAPKK-2, from PC12 cells. Aliquots of Mono-S-purified SAPKK-1 or
+

SAPKK-2 from osmotically shocked or ultraviolet-irradiated PC12 cells (Fig. 1) were immunoprecipitated with the anti-MKK4 antibodies. termed
antibody N and antibody C, as described under Materials and Methods. The figure shows the percentage of SAP kinase-2 activator (A, B) or SAP
kinase-1 activator (C, D) remaining in the supernatant after immunoprecipitation (average of four experiments) relative to an incubation in which
anti-MKK4 antibody was omitted. The figure also shows further experiments in which the antibodies were incubated with the appropriate peptide
immunogen prior to immunoprecipitation.

shown). SAPKK-2 was also inactivated by PP2A, but, in
contrast to SAPKK-1, it could not be reactivated by MEKK
(Fig. 3).
When the Mono S fractions obtained after chromatography
of lysates from unstimulated PC12 cells were immunoblotted
with antibody C (Fig. 4A) or antibody N (data not shown), a
single immunoreactive band was observed whose apparent mo-
lecular mass (43 kDa) was similar to that of MKK4. However,
the 43-kDa band eluted at 1.1 -1.2 ml, later than the SAPKK-1
activity from stimulated cells which peaked at 0.9 ml (Fig. 1).
The immunoreactive 43-kDa protein eluting at 1.1 - 1.2 ml
appears to be inactive MKK4BAPKK-1 because it can be acti-
vated by incubation with MgATP and MEKK (Fig. 4B) and
activity can then be immunoprecipitated with antibody-C
(Fig. 4C). It should also be noted that, after activation of the
latent SAPKK-1 by MEKK, the activity towards SAP kinase-2
was 7-20-fold higher than that observed after osmotic shock
(Fig. 1 and data not shown) and 25-500-fold higher than that
observed after ultraviolet irradiation, or stimulation with aniso-
mycin or arsenite (compare Figs 1 and 4). This finding and im-
munoblotting experiments (data not shown) indicate that very
little SAPKK-1 is activated in response to cellular stress.
Identification of SAP kinase kinases in KB cells. KB cells are
a human oral epidermal carcinoma cell line which are particu-
larly responsive to IL-I; SAP kinase-1 [34, 351 and SAP
kinase-2 [7] are both activated when these cells are stimulated
with IL-1 or exposed to cellular stresses. The experiments de-
scribed above in PC12 cells were therefore repeated using both
stressed and IL-1-stimulated KB cells.
As observed in PC12 cells, almost no SAPKK activity was
detected after Mono S chromatography, unless the cells had been
stimulated with IL-1 or exposed to cellular stresses (Fig. 5). The
Mono S profiles obtained after chromatography of lysates from
IL-1-stimulated or ultraviolet-irradiated cells were similar
(Fig. 5A and B). In each case, a single major peak of SAP ki-
nase-2 activator was detected, termed SAPKK-3. SAPKK-3
eluted at 0.85-0.9 ml, similar to activated SAPKK-1 in PC12
cells, but it was not immunoprecipitated by antibody C (Fig. 6).
Moreover, after inactivation by PP2A,, no reactivation of
SAPKK-3 occurred following incubation with MgATP and
MEKK (data not shown). These experiments demonstrate that
SAPKK-3 is different from SAPKK-1. Moreover, SAPKK-3
activity did not coelute with the major peak of SAP kinase-1
activator induced by IL-1 or ultraviolet irradiation, termed
 14321033, 1996, 3, Downloaded from https://febs.onlinelibrary.wiley.com/doi/10.1111/j.1432-1033.1996.00796.x by Nat Prov Indonesia, Wiley Online Library on [15/07/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Meier et al. ( E m J. Biochem. 236) 801

SAPKK-1 SAPKK-1 SAPKK-2 A

T elutionvolume 0.9 1.0 1.1 1.2 1.3 1.4

-43
-29
20

15
- + + - + + - + + 3e
- - + _ _ - - + b,
Q 10
SAPK-1 SAPK-2 SAPK-2 $
Fig. 3. MEK kinase reactivates SAPKK-1, but not SAPKK-2, from
PC12 cells. Mono-S-purified SAPKK-1 or SAPKK-2 from osmotically
g 25

shocked PC12 cells (30 pl) were incubated for 45 min at 30°C with or
2
without 100 mU/ml PP2A, (see [28] for definition of units of PP2A 0
activity). The solutions were then made 2 pM in microcystin-LR to inac- 0.4 0.6 0.8 1.0 1.2 1.4 1.6 13
tivate PP2A,, incubated for a further 30 min at 30°C in the presence or
absence of 50 pg/ml MEKK, 10 mM MgCl, and 0.2 mM unlabelled ATP, elution volume (ml)
and then assayed for SAP kinase-I (SAPK-1) reactivating and/or SAP
kinase-2 (SAPK-2) reactivating activity. The results are the average 150
C
(mean t SD) of three separate experiments. There was no reactivation if
MEKK was replaced by buffer.
.5 I

SAPKK-4. SAPKK-4 consistently eluted one fraction later than

SAPKK-3 in six separate experiments (Fig. 5A and B), and the
leading fraction of the peak of SAPKK-3 (eluting at 0.8 ml) was
hardly able to activate SAP kinase-1, while the tailing fraction
of SAPKK-4 (eluting at 1.0 ml) was unable to activate SAP ki-
nase-2 (see Fig. 5A). These results demonstrate that SAPKK-3
and SAPKK-4 are separate enzymes. Like SAPKK-3, SAPKK-
4 was not immunoprecipitated by antibody C (Fig. 6), was inac-
tivated by PP2A, but was not reactivated by MEKK. These
properties, its high degree of selectivity for SAP kinase-1, and
its chromatographic properties distinguish SAPKK-4 from
SAPKK-1.
Similar to the effects of IL-I and ultraviolet radiation,
SAPKK-3 was the dominant SAP kinase-2 activator generated
when KB cells were osmotically shocked or stimulated with ani-
somycin (Fig. 5 C and D). Surprisingly, however, no peak eluting
in the position of SAPKK-4 was generated by these stimuli. In-
stead, two chromatographically distinct SAP kinase-1 activators
were detected. One eluted just before SAPKK-3, while the other
eluted later than SAPKK-4, and was termed SAPKK-5. The first
peak of SAP kinase-I activator from osmotically shocked or ani-
somycin-stimulated KB cells was immunoprecipitated by anti-
body C (Figs 5C and 6) and, after inactivation with PP2A, could
be reactivated by MEK kinase (data not shown), indicating that
it was SAPKK-1 or a closely related isoform. In contrast,
SAPKK-5 was not immunoprecipitated by antibody C (Figs 5C
and 6) and, after inactivation by PP2A, was not reactivated by
MEK kinase. Antibody N could not be used in these experiments
because it is specific for SAPKK-1 from rodent cells, the peptide
to which it was raised differing at three positions from human
SAPKK-1.
It should also be noted that SAPKK-3 activity towards SAP
kinase-2 in stimulated KB cells is extremely high. Its activity is
over 10-fold higher than that of SAPKK-1 or SAPKK-2 activity
in PC12 cells generated in response to any cellular stress (com-
pare Figs 1 and 5). The high activity of SAPKK-3 in KB cells
obscures the much weaker activity of SAPKK-1 towards SAP
kinase-2. In fact, the activity ratio SAP kinase-2 activatodSAP
peptide immunogen - + - +
antibody c c c c
substrate SAPK-1 SAPK-2
Fig.4. Identification of the latent form of SAPKK-1 in PC12 cells.
(A) Lysates from unstimulated PC12 cells (0.4 mg protein) were chro-
matographed on Mono S as described under Materials and Methods and
immunoblotted with antibody C as described [38].No immunoreactive
band was observed in any of the 40 fractions from the column other than
those eluting at 1.1 ml and 1.2 ml. (B) The fractions from A were as-
sayed for SAP kinase-1 (SAPK-1) activator ( 0 ,). or SAP kinase-2
(SAPK-2) activator (0,0)either before,.( 0) or after (0,0) incuba-
tion with MEKK and MgATP as described in the legend to Fig. 3. No
SAPK-1 activator or SAPK-2 activator was generated by incubation
with MEKK and MgATP in any of the 40 fractions other than those
shown. The NaCl gradient was the same as in Fig. 1 D. (C) The peak of
SAPK-USAPK-2 activator obtained after incubation with MEKK and
MgATP in B was immunoprecipitated with antibody C as described un-
der Materials and Methods. The figure shows the percentage of SAPK-
1 activator and SAPK-2 activator remaining in the supernatant relative
to an incubation in which the anti-MKK4 antibody was omitted. The
figure also shows further experiments in which the antibodies were
incubated with the appropriate peptide immunogen prior to immuno-
precipitation. The results are presented as averages (meant SD) for
three separate experiments.
kinase-1 activator at the leading edge of the peak of SAPKK-1
(see fraction eluting at 0.8 ml in Fig. SD) is about 100, at least
as high as that of activated SAPKK-1 in PC12 cells (Fig. 1).
The extremely high activity of SAPKK-3 also swamps the much
weaker activity of SAPKK-2 (Fig. 5A-D), which is only detect-
able if the fractions are assayed at 25-fold lower dilutions
 14321033, 1996, 3, Downloaded from https://febs.onlinelibrary.wiley.com/doi/10.1111/j.1432-1033.1996.00796.x by Nat Prov Indonesia, Wiley Online Library on [15/07/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
802 Meier et al. (Eur J. Biochem. 236)

100

z
30 15 120
h
75
3 E
", I

2 20 10 3 80
P 50
.-* *
.m

h Y
N
2a 10 5 2 40
2s
3 3
0 0 0 0
0.6 0.8 1.0 1.2 1.4 0.6 0.8 1.0 1.2 1.4

elution volume (ml) elution volume (ml)

50
C anisomycin
30
60 1" sorbitol 30

c5
9
s 40
E
$
.-
30 20 40
20
--
;
0

.-
Y 5
q3 20
c2 10
10 20 10 7
LU
v1

0 0 0 0
0.6 0.8 1.0 1.2 1.4 0.6 0.8 1 1.2 1.4

elution volume (ml) elution volume (ml)

6
E SAPKK-2
5
f -t- ILl
: 4
s -0- uv
$ 3 ---c anisomycin
."
+ sorbitol
c
7
4
2
-E-- control
W l

0
1.1 1.2 1.3 1.4 1.S

elution volume (ml)

Fig.5. Several SAP kinase kinases are activated in KB cells in response to cellular stresses or IL-1. (A-D) Cell lysates (0.4 mg protein) from
cells stimulated for 15 min with 20 ng/ml 1L-I (A), ultraviolet irradiated at 60 J/m' and then left for 30 min (B), stimulated for 30 min with 10 pg/
ml anisomycin (C), or osmotically shocked for 5 min with 0.5 M sorbitol (D) were chromatographed on Mono S and activators of SAP kinase-l
(SAPK-I, 0 ) and SAP kinase-2 (SAPK-2, 0)were assayed as described under Materials and Methods. The NaCl gradient was as in Fig. 1 D.
Lysates from unstimulated cells were chromatographed separately on Mono S and assayed for SAP kinase-1 activator (m, A) and SAP kinase-2
activator (0, D). In C, SAP kinase-1 activator (0-- -0) was assayed after immunodepletion of each fraction with antibody C. The positions of
SAPKK-1, SAPKK-2, SAPKK-3, SAPKK-4 and SAPKK-5 are indicated; similar results were obtained in at least three different experimcnts with
each stimulus. (E) The fractions eluting between 1.0-1.5 ml were assayed for SAP kinase-2 activator at a 25-fold lower dilution than i n 4 - D :
( 0 )1L-l stimulation; (0)ultraviolet irradiation; (A)anisomycin stimulation; (A) osmotic shock (0.5 M sorbitol); (U) unstimulated cells

(Fig. 5E). Like the SAPKK-2 in PC12 cells, SAPKK-2 in KB
cells is not immunoprecipitated by antibody C (data not shown).
In contrast to the very high activity of SAPKK-3 towards
SAP kinase-2, the activity of SAPKK-4 or SAPKK-5 towards
SAP kinase-l in KB cells is similar to the activity of SAPKK-1
towards SAP kinase-1 in PC12 cells, following exposure to the
same stresses (compare Figs 1 and 5).
In summary, these experiments indicate that the major SAP
kinase-l activators (SAPKK-4 and SAPKK-5) and the major
SAP kinase-2 activator (SAPKK-3) induced in KB cells by cel-
lular stresses and IL-1 are distinct from SAPKK-l, SAPKK-2
and the major MAPKK which activates p42 and p44 MAP
kinases. MAPKK activity can be detected after IL-1 stimulation
of KB cells, and elutes from Mono S at lower concentrations of
NaCi than the SAPKKs, peaking at 0.4 ml (data not shown).
Activation of SAP kinase-1 in K 3 cells and PC12 cells. The
finding that no SAPKK-1 activity was generated after I L l stim-
ulation or ultraviolet irradiation of KB cells prompted us to
study whether SAP kinase-l was activated under these condi-
tions. As shown in Fig. 7A, IL-1 stimulation, ultraviolet irradia-
tion, as well as .osmotic shock or treatment with anisomycin,
produced a robust activation of SAP kinase-l. SAP kinase-I was
also activated by these stimuli in PC12 cells, the extent of acti-
vation being similar to that expected from the extent of activa-
tion of SAPKK-1 (Fig. 7B).
 14321033, 1996, 3, Downloaded from https://febs.onlinelibrary.wiley.com/doi/10.1111/j.1432-1033.1996.00796.x by Nat Prov Indonesia, Wiley Online Library on [15/07/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Meier et al. (Eur: J. Biochem. 236) 803
125 A
A w
1 2 3 4 5 6

peptide immunogen

form of SAPKK

substrate
0 -

-
3

SAPK-2
+
3
II -
4

SAPK-I
+
4
Jun-

Jun-
1 2 3
6
4 5

Fig.7. Activation of SAP kinase-1 by different stimuli in KB and

PC12 cells. (A) Extracts prepared from KB cells after exposure to dif-
ferent agonists were assayed for SAP kinase-1 activity by the phosphory-
lation of GST-c-Jun(1-194) (Jun) as described under Materials and
Methods. The figure shows an autoradiograph of the SDS/polyacrylam-
ide gel. Lane 1 , unstimulated cells; lane 2, ultraviolet-irradiated cells
(60 Urn2): lane 3, cells stimulated for 15 min with 20 ng/ml IL-1; lane
4, cells stimulated for 30 min with 10 pg/ml anisomycin; lane 5 , un-
stimulated cells; lane 6, cells osmotically shocked for 15 min with 0.5 M
150 sorbitol. (B) Same as A, except that extracts from PC12 cells were as-
B sorbitol sayed for SAP kinase-1 activity. Lane 1, unstimulated cells: lane 2, cells
osmotically shocked for 15 min with 0.5 M sorbitol; lane 3, ultraviolet-

11
irradiated cells ; lane 4, anisomycin-stimulated cells; lane 5 , arsenite-
stimulated cells.

In PC12 cells, SAPKK-1 was the only activator of SAP

0
peptide immunogen - + + - +
form of SAPKK 1 1 3 5 5
substrate SAPK-1 SAPK-2 SAPK-1
Fig. 6. Anti-MKK4 (antibody C) immunoprecipitates SAPKK-1, but
not SAPKK-3, SAPKK-4 or SAPKK-5 from KB cells. Aliquots of
Mono-S-purified SAPKKs from ultraviolet-irradiated (A) or osmotically
shocked (B) KB cells see (Fig. 5 ) were immunoprecipitated with anti-
body C as described under Materials and Methods. The figure shows the
percentage of SAPK-1 activator or SAPK-2 activator remaining in the
supernatant (average of four experiments) relative to a control incubation
in which antibody C was omitted. The figure also shows further experi-
ments in which the antibodies were incubated with the peptide immuno-
gen prior to immunoprecipitation.
DISCUSSION
The results presented in this paper demonstrate that the acti-
vation of SAP kinase-1 and SAP kinase-2 is more complex than
has hitherto been realised. Mammalian cells clearly contain
SAPKKs which can activate SAP kinase-1 and not SAP ki-
nase-2, SAPKKs which can activate SAP kinase-2 and not SAP
kinase-1, as well as SAPKK-I which is capable of activating
both SAP kinases in vitro. All three classes of SAPKK can be
activated by cellular stresses, but our studies indicate that the
SAPKKs which are activated can vary between cells and (in KB
cells at least) with the nature of the cellular stress. It is also
possible that different SAPKKs may be activated depending on
the degree of stress applied, although in preliminary experiments
we found that the same SAPKKs were activated in PC12 and
KB cells irrespective of the degree of osmotic shock. The activa-
tion of SAPKK-1 and SAPKK-2 increased progressively as the
concentration of sorbitol in the medium was increased from
0.1 M to 0.5 M (Meier, R., unpublished observations).
kinase-1 detected after exposure to a variety of stresses. More-
over, the extent of activation of SAPKK-1 showed a reasonable
correlation with the extent of activation of SAP kinase-1
(Fig. 7 B). These observations indicate that SAPKK-1 is likely
to be the major activator of SAP kinase-1 in PC12 cells. How-
ever, it should be noted that the SAP kinase-l substrate used in
our studies was a form of p54 SAP kinase which had been inac-
tivated using PTP-1B. This leaves open the possibility that the
Thr residue in SAP kinase-1 might be phosphorylated by a dif-
ferent enzyme(s). Since SAP kinases, like MAP kinases, phos-
phorylate Ser and Thr residues that are followed by Pro, the Thr
residue in SAP kinase-I (being followed by a Pro) might well
be phosphorylated by SAP kinase-1 itself after the Tyr residue
has been phosphorylated by a SAPKK.
Even when PC12 cells had been exposed to osmotic shock,
which caused the greatest activation of SAPKK-1 (Fig. I), the
conversion of SAPKKl to the activated state was still very low,
ranging over 6-15% in a number of experiments (compare
Figs 1 and 4). Moreover, exposure of PC12 cells to sodium arse-
nite activates much less than 1% of the SAPKK-1, yet causes
40-50% activation of MAPKAP kinase-2 IS]. Similar observa-
tions have been made in the classical MAP kinase pathway. For
example, in Swiss 3T3 cells, stimuli which activate < 1% of the
MAPKK-1 cause 10-20% activation of p42 MAP kinase and
30-80% activation of its downstream target MAPKAP kinase-1
13.51. These observations emphasise the enormous amplification
potential of MAP kinase cascades.
In contrast to PC12 cells, SAPKK-1 was not activated to a
detectable extent in KB cells after stimulation with 1L-1 or ultra-
violet irradiation. SAP kinase-1, which is activated strongly un-
der these conditions (Fig. 7A), must therefore be phosphorylated
by another enzyme. This is presumably SAPKK-4, the major
SAP kinase-I activator detected under these conditions (Fig. 5).
SAPKK-1 was activated in KB cells after exposure to osmotic
shock or anisomycin but, even under these conditions, another
SAP kinase-1 activator was induced (SAPKK-5). SAPKK-5
eluted from Mono S at higher NaCl than SAPKK-4 and was
therefore resolved from SAPKK-3 (Fig. 5 D). The elution profile

804 Meier et al. (Eur J. Biochern. 236)
of SAPKKs observed after exposing KB cells to osmotic shock
demonstrates that SAPKK-3 is highly specific for SAP ki-
nase-2, while SAPKK-5 is specific for SAP kinase-I.
One enzyme (SAPKK-3) was induced by every cellular
stress tested or by IL-1 in KB cells, and accounted for > 95 %
of the SAP kinase-2 activator detected (Fig. 5 ) , the contribution
of SAPKK-1 and SAPKK-2 being negligible. SAPKK-3 is
therefore likely to be responsible for the activation of SAP ki-
nase-2 [7] by each of these stimuli in KB cells. In PC12 cells,
SAPKK-3 was absent and SAPKK-1 and SAPKK-2 were the
only SAP kinase-2 activators induced by cellular stresses. How-
ever SAPKK-l, but not SAPKK-2, is activated by MEK kinase
in vitro (Fig. 4) [24, 251, and in COS-1 and HeLa cells cotrans-
fection of MEK kinase with SAPKK-1 leads to activation of
transfected SAP kinase-1, but not SAP kinase-2 [21, 261. Simi-
larly, transfection of COS-1 or 293 cells with germinal centre
kinase a homologue of the Saccharomyces cerevisiae protein
kinase STE20 activates SAP kinase-I, but not SAP kinase-2,
and the activation of SAP kinase-I is prevented by a dominant
negative mutant of SAPKK-1 [36]. These observations strongly
suggest that SAPKK-1 is a specific SAP kinase-I activator in
vivo. The activation of SAP kinase-2 is therefore likely to be
catalysed by SAPKK-2 in PC12 cells and by SAPKK-3 (and
perhaps SAPKK-2) in KB cells.
Very recently, Nishida and coworkers also identified multiple
activators of SAP kinase-I in extracts prepared from rat 3Y1
fibroblasts that had been exposed to 0.7 M NaCl for 60 min [37].
A minor activity which was retained by Q-Sepharose and recog-
nised by antibodies raised against XMEK2, is likely to be
SAPKK-1. However, the major activator of SAP kinase-I was
not retained by Q-Sepharose or recognised by anti-XMEK2 anti-
bodies and was resolved into two components by subsequent
chromatography on heparin-Sepharose, that might be SAPKK-4
and SAPKK-5. The second peak coeluted with the major protein
kinase that phosphorylated XMpk2 (the Xenopus homologue of
RK), but this XMpk2 kinase was largely resolved from the major
SAPK-1 activator by chromatography on hydroxyapatite. This
enzyme could be SAPKK-3, although as it was only measured
by the phosphorylation and not the activation of XMpk2, the
possibility remains that it phosphorylates a residue(s) outside the
Thr-Gly-Tyr sequence.
In summary, our results, and those of Nishida and coworkers
[37], demonstrate that cellular stresses and IL-1 induce several
SAP kinase activators in mammalian cells besides SAPKK-1,
but further work is needed to establish the identities of these
enzymes. SAPKK-4 and SAPKK-5 in KB cells activate SAP
kinase-1 specifically and are likely to be novel MAPKK homo-
logues, but we cannot exclude the possibility that they are the
same enzyme, their distinct chromatographic behaviour reflect-
ing different degrees of phosphorylation andlor association with
another protein(s). SAPKK-2 and SAPKK-3 were eluted from
Mono S at different NaCl concentrations, the activity of the lat-
ter being at least 20-fold higher than the former. We have there-
fore treated them as distinct entities in this paper although, as
they activate SAP kinase-2, have little or no activity towards
SAP kinase-I , are inactivated by PP2A, and not reactivated by
MEKK, they may also be related. The isolation and characterisa-
tion of SAPKK-2, SAPKK-3, SAPKK-4 and SAPKK-5 are in
progress.
A. R. N. thanks Dr. Tim Hunt for his advice and support. R. M.
acknowledges a postdoctoral fellowship from the Swiss Government,
A. C. a postdoctoral fellowship from the Spanish Government and J.
R. a postgraduate studentship from The Royal Society. This work was
supported by the Medical Research Council and Royal Society (P, C.)
and the Imperial Cancer Research Fund (A. R. N.)
14321033, 1996, 3, Downloaded from https://febs.onlinelibrary.wiley.com/doi/10.1111/j.1432-1033.1996.00796.x by Nat Prov Indonesia, Wiley Online Library on [15/07/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
REFERENCES
1. Cohen, P. (1996) Trends Cell Biol., in the press.
2. Pulverer, B. J., Kyriakis, J. M., Avruch, J., Nikolakaki, E. & Wood-
gett, J. R. (1991) Nature 353, 670-674
3. Hibi, M., Lin, A,, Smeal, T., Minden, A. & Karin, M. (1993) Genes
Dev. 7, 2135-2148.
4. Kyriakis, J. M., Banerjee, P., Nikolakaki, E., Dai, T., Rubie, E. A.,
Ahmad, M. F., Avruch, J. & Woodgett, J. R. (1994) Nature 369,
156-160.
5. Derijard, B., Hibi, M., Wu, I.-H., Barrett, T., Su, B., Deng, T.. Karin,
M. & Davis R. J. (1994) Cell 76, 1025-1037.
6. Han,J., Lee, J. D, Bibbs, L. & Ulevitch, R. J. (1994) Scien1.e 265,
808-81 1.
7. Freshney, N. W., Rawlinson, L., Guesdon, F., Jones, E., Conley, S . ,
Hsuan, J. & Saklatvala, J. (1994) Cell 78, 1039-1049.
8. Rouse, J., Cohen, P., Trigon, S . , Morange, M., Alonso-Llan-iazares,
A,, Zamanillo, D., Hunt, T. & Nebreda, A. (1994) Cell 78,1027-
1037.
9. Lee, 5. C., Laydon, J. T., McDonnell, P. C., Gallagher, T. E, Kumar,
S., Green, D., McNulty, D., Blumenthal, M. J., Heys, R. J , Land-
vatter, S. W., Strickler, J. E., McLaughlin, M. M., Siemens, I. R.,
Fisher, S. M., Livi, G. P., White, J. R., Adams, J. L. & Young, P.
R. (1994) Nature 372, 739-746
10. Stokoe, D., Engel, K., Campbell, D. G., Cohen, P. & Gae.;tel, M.
(1992) FEBS Lett. 313, 307-313.
11. Cuenda, A,, Rouse, J., Doza, Y. N., Meier, R., Cohen, P., Gsllagher,
T. F., Young, P. R. & Lee, J. C. (1995) FEBS Lett. 364, 229-
233.
12. Huot, J., Lambert, H., Lavoie, J. N., Guimond, A,, Houle, F. &
Landry, J. (1995) Eur: J. Biochem. 227, 416-427.
13. Anderson, N. G., Maller, J. L., Tonks, N. K. & Sturgill, T. U. (1990)
Nature 343, 651-653
14. Payne, D. M., Rossomando, A. J., Martino, P., Erickson, A. K.. Her,
J.-H., Shabanowitz, J., Hunt, D. E, Weber, M. J. & Sturgill, T.
W. (1991) EMBO J. 10, 885-892
15. Kyriakis, J. M., Brautigan, D. L., Ingebritsen, T. S. & Akmch, J.
(1991) J. Biol. Chem. 266, 10043-10046.
16. Doza, Y. N., Cuenda, A,, Thomas, G. M., Cohen, P. & Kebreda, A.
R. (1995) FEBS Lett. 364, 223-228.
17. Raingeaud, J., Gupta, S., Rogers, J. S., Dickens, M., Han, J.,
Ulevitch, R. J. & Davis, R. J. (1995) J. Bid. Chem. 27(/, 7420-
7426.
18. Nakielny, S., Cohen, P., Wu, J. & Sturgill, T. W. (1992) EMBO J.
11, 2123-2129.
19. Haystead, T. A. J., Dent, P., Wu, J., Haystead, C. M. M. & Sturgill,
T. W. (1993) FEBS Lett. 306, 17-22.
20. Sanchez, I., Hughes, R. T., Mayer, B. J., Yee, K., Woodgett. J. R.,
Avruch, J., Kyriakis, J. M. & Zon, L. I. (1994) Nature 372, 794-
798.
21. Derijard, B., Raingeaud, J., Barrett, T., Wu, I.-H., Han, J., Ulevitch,
R. J. & Davis, R. J. (1995) Science 267, 682-684.
22. Yashar, B. M., Kelley, C., Yee, K., Errede, B. & Zon, L. I. (1993)
Mol. Cell. Biol. 13, 5738-5748.
23. Lange-Carter, C. A,, Pleiman, C. M., Gardner, A. M., Blumer, K.
J. &Johnson, G. L. (1993) Science 260, 315-319.
24. Yan, M., Dai, T., Deak, J. C., Kyriakis, J. M., Zon, L. I., Woodgett,
J. R. & Templeton, D. J. (1994) Nature 372, 798-800.
25. Matsuda, S., Kawasaki, H., Moriguchi, T., Gotoh, Y. & Nmhida, E.
(1995) J. Bid. Chem. 270, 12781-12786.
26. Lin, A,, Minden, A,, Martinetto, H., Claret, E-X., Lange-Carter, C.,
Mercurio, F., Johnson, G. L. & Karin, M. (1995) Science 268,
286-290.
27. Ben-Levy, R., Leighton, I. A., Doza, Y. N., Morrice, N.. Attwood.
P., Marshall, C. J. & Cohen, P. (1995) EMBO J . 14, 101-
110.
28. Cohen, P., Alemany, S., Hemmings, B. A., Resink, T.. Stralfors, P. &
Tung, H. Y. L. (1988) Methods Enzymol. 159, 390-408.
29. Kyriakis, J. & Avruch, J. (1990) J. Biol. Chem. 265, 17355-17363.
30. Kracht, M., Truong, O., Totty, N. F., Shiroo, M. & Saklatvala, J.
(1994) J. Exp. Med. 180, 2017-2025.
31. Gomez, N., Tonks, N. K., Morrison, C., Harmar, A. J. & Cohen, P.
(1990) FERS Lett. 271, 1 19- 122.
 14321033, 1996, 3, Downloaded from https://febs.onlinelibrary.wiley.com/doi/10.1111/j.1432-1033.1996.00796.x by Nat Prov Indonesia, Wiley Online Library on [15/07/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Meier et al. ( E w J. Biochem. 236) 805
32. Stokoe, D., Caudwell, F. B., Cohen, P. T. W. & Cohen, P. (1993) 36. Pombo, C. M., Kehrl, J. H., Sanchez, I., Katz, P., Avruch, J., Zon,
Biochem. J. 296, 842-849. L. I., Woodgett, J. R., Force, T. & Kyriakis, J. M. (1995) Nature
33. Marshall, C. J. (1995) Cell 80, 179-185. 377, 750-754.
34. Kracht, M., Shiroo, M., Marshall, C. J., Hsuan, J. J. & Saklatvala, 37. Moriguchi, T., Kawasaki, H., Matsuda, S., Gotoh, Y. & Nishida, E.
J. (1994) Biochem. J. 302, 897-905. (1995) J. Biol. Chem. 270, 12969-12972.
35. Alessi, D. R., Cuenda, A., Cohen, P., Dudley, D. & Saltiel, A. (2995) 38. Traverse, S., Cohen, P., Paterson, H., Marshall, C., Rapp, U. &
J. Biol. Chem. 270, 27489-27494. Grand, R. J. A. (1993) Oncogene 8, 3175-3181.

Note added in proof

We have recently found that two different antipeptide-antibodies
raised against human MKK3 (and which recognize human
MKK3 in E. coli) immunoprecipitate SAPKK2 in both rat PC12
cells and human KB cells, but do not immunoprecipitate
SAPKK3 from KB cells (Meier, R., unpublished results). Thus,
SAPKK2 appears to be identical to MKK3.