Dysregulation of signaling pathways
in CD45-deficient NK cells leads to differentially
regulated cytotoxicity and cytokine production
David G. T. Hesslein*, Rayna Takaki*†, Michelle L. Hermiston‡, Arthur Weiss§¶, and Lewis L. Lanier*?
*Department of Microbiology and Immunology and Cancer Research Institute,†Biomedical Sciences Graduate Program,‡Department of Pediatrics,
and§Department of Medicine, The Rosalind Russell Medical Research Center for Arthritis and Howard Hughes Medical Institute, University
of California, San Francisco, CA 94143
Contributed by Arthur Weiss, March 14, 2006
CD45, a protein tyrosine phosphatase that regulates Src family
kinases, is important for regulating T cell and B cell receptor
signaling; however, little is known about how CD45 regulates
immunoreceptor tyrosine-based activation motif (ITAM)-depen-
dent natural killer (NK) cell receptor signaling and the resulting
effector functions. NK cells from CD45-deficient mice are relatively
competent for ITAM receptor-induced cell-mediated cytotoxicity,
yet completely deficient for cytokine secretion after stimulation
with ligands to or antibodies against NK1.1, CD16, Ly49H, Ly49D,
and NKG2D. This deficiency in cytokine?chemokine production
occurs at the level of mRNA expression. After receptor engage-
ment, extracellular signal-regulated kinase and c-Jun N-terminal
kinase activation was markedly perturbed, whereas p38 activation
was not substantially affected. The pattern and amounts of basal
tyrosine phosphorylation were altered in freshly isolated NK cells
and were surprisingly and markedly increased in IL-2-expanded NK
cells from CD45?/? mice. These findings indicate that CD45-
dependent regulation of ITAM-dependent signaling pathways is
essential for NK cell-mediated cytokine production but not cyto-
target cells by means of activating cell-surface receptors, which
upon ligation induce target cell lysis and cytokine production.
Although some ligands of the activating receptors are known,
many are still undiscovered. The known ligands include Ig
constant regions, viral antigens, adhesion molecules, and stress-
induced glycoproteins, which are induced after viral infection or
cellular transformation (1). Inhibitory receptors, which recog-
nize self-MHC class I molecules, modulate the signals of the
activating receptors to prevent damage of normal tissues.
Many of the known NK cell-activating receptors, such as
CD16, NK1.1, Ly49H, Ly49D, and an isoform of NKG2D,
associate with the CD3?, Fc?RI?, and?or DAP12 adapter
proteins, which contain immunoreceptor tyrosine-based activa-
tion motifs (ITAMs) (2). Ligation of an ITAM-associated NK
cell receptor results in cytokine production and killing of the
target cell. Src family kinases phosphorylate the tyrosine resi-
dues in ITAMs and initiate the activating signal cascade (3). The
tyrosine kinases Syk and Zap-70 then bind to the doubly
phosphorylated ITAMs via their tandem SH2 domains, resulting
in the activation of downstream molecules, such as phospho-
lipase C-?1 and -?2, Vav-1, Vav-2, Vav3, Ras-family members,
and the mitogen-activated protein kinases (MAPK) c-Jun N-
terminal kinase (JNK), extracellular signal-regulated kinase
(ERK), and p38. MAPKs are important regulators of NK
cell-mediated cytotoxicity and cytokine production. ERK ap-
pears to play a major role in these effector functions (4–8),
whereas JNK and p38 have more limited roles (4, 5, 9, 10).
atural killer (NK) cells confer immune protection against
certain intracellular pathogens and tumors. They recognize
The requirement for Src kinase activity for ITAM-based
signaling implies that the regulation of Src kinases likely affects
NK cell effector functions (11–13). Src kinases are modulated, in
part, by the protein tyrosine phosphatase CD45. By dephospho-
rylating the negative regulatory tyrosine of Src kinases, CD45
generates a pool of ‘‘primed’’ Src family kinases capable of rapid
activation upon receptor stimulation (14). Depending on a cell’s
activation status and developmental stage, CD45 can also de-
phosphorylate the catalytic tyrosine, leading to kinase inactiva-
tion (14). CD45?/? mice suffer from severe defects in B and T
cell development and function because of deficient antigen
receptor signaling, demonstrating the importance of this phos-
phatase (15, 16). When we began our studies, little was known
about the role of CD45 in NK cell function. mAbs against CD45
had been reported to activate or inhibit human and rat NK cell
cytokine production or target cell lysis (17–22). NK cells from
CD45?/? mice had normal cytotoxic activity against the pro-
totypic tumor cell line Yac-1 (23, 24); NK cells primarily use the
NKG2D receptor to kill Yac-1 (25). A study showed that
CD45?/? NK cells have a reduced capacity for IFN? production
when stimulated by anti-NK1.1 mAb (23). Here we examine the
function of CD45 in cytokine and chemokine production and
target cell killing initiated by well characterized ITAM-based
ITAM-Based Receptor-Induced Cytotoxicity Is Intact in CD45?/? NK
Cells. We examined the role of CD45 in ITAM receptor-
dependent NK cell-mediated cytotoxicity. Splenic and liver cells
from WT and CD45?/? mice were stained with mAbs against
NK cell receptors. CD3?, NK1.1?cells were increased in both
percentage and number in CD45?/? mice (2- to 3-fold increase
in percentage compared with WT mice; data not shown), in
accordance with prior reports (23, 24). Expression of NKG2D,
Ly49H, CD16, Ly49G2, VLA-2 (DX5), and NK1.1 was similar
between genotypes whether the NK cells were freshly isolated or
IL-2 expanded (data not shown). However, the percentage of
NK cells expressing Ly49A and Ly49D was reduced to 40% and
55% of WT levels, respectively.
IL-2-expanded WT and CD45?/? NK cells were compared in
Conflict of interest statement: No conflicts declared.
Abbreviations: ITAM, immunoreceptor tyrosine-based activation motif; MAPK, mitogen-
activated protein kinase; ERK, extracellular signal-regulated kinase; JNK, c-Jun N-terminal
dent cell-mediated cytotoxicity.
¶To whom correspondence may be addressed at: University of California and Howard
?To whom correspondence may be addressed at: Department of Microbiology and Immu-
nology, University of California, 513 Parnassus Avenue, Box 0414, San Francisco, CA
94143-0414. E-mail: email@example.com.
© 2006 by The National Academy of Sciences of the USA
May 2, 2006 ?
vol. 103 ?
their ability to kill target cells through specific NK cell receptors.
m157 is a murine CMV-encoded ligand for the activating Ly49H
receptor (26, 27). CD45?/? NK cells killed Ba?F3 mouse pro-B
cells transduced with the m157 similarly to WT NK cells (71%
of WT) (Fig. 1A). CD45?/? cells also efficiently killed target
cells bearing RAE-1?, a ligand for NKG2D (73% of WT) (Fig.
1B). Ly49H signals by using the ITAM-bearing DAP12 adapter
(28), and NKG2D signals via the phosphatidylinositol 3-kinase
activating DAP10 adapter (29), as well as through DAP12 (30).
To extend our findings to other ITAM adapter?receptor
receptor signals through Fc?RI? (31, 32). We tested the function
of the Fc?RI??CD16 receptor complex in CD45?/? NK cells by
testing antibody-dependent cell-mediated cytotoxicity (ADCC)
against mAb-coated RMA tumor cells.51Cr-labeled RMA cells
were coated with an anti-CD90 mAb [previously shown to
mediate efficient ADCC (33)] and then incubated with NK
effector cells. Both WT and CD45?/? NK cells efficiently killed
the anti-CD90 mAb-coated RMA cells at equivalent levels (Fig.
1C). This activity depended on signaling via CD16 because
ADCC was completely blocked by the presence of a soluble
neutralizing anti-CD16?32 mAb. Although ITAM-based recep-
with WT NK cells, there was nonetheless significant and easily
detectable lytic activity induced by either DAP12- or Fc?RI?-
associated receptors. Therefore, NK cells lacking CD45 can still
mediate cytolytic activity via ITAM-dependent receptors.
Impaired in CD45?/? NK Cells. Besides directly killing target cells,
NK cells also function as the main innate immune cell type
producing IFN?, as well as other cytokines and chemokines,
including granulocyte–macrophage colony-stimulating factor,
TNF?, macrophage inflammatory protein (MIP) 1?, MIP-1?,
and RANTES. We tested the ability of CD45?/? NK cells to
produce cytokines and chemokines in response to stimulation of
their ITAM-based receptors. CD45?/? NK cells were defective
in secreting IFN? when stimulated with plate-bound mAbs
against NKG2D, Ly49D, or NK1.1 (Fig. 2A). In contrast, WT
of CD45?/? NK cells with ligands for CD16 (plate-bound IgG)
or NKG2D (soluble H60) resulted in no detectable IFN?
production (Fig. 2A and data not shown). CD45?/? NK cells
were also defective in MIP-1? and granulocyte–macrophage
colony-stimulating factor production (Fig. 2B and data not
shown). However, CD45?/? NK cells were capable of cytokine
and chemokine production because abundant IFN? and MIP-1?
were produced when receptor-mediated signals were bypassed
with PMA and ionomycin stimulation (Fig. 2C). Thus, CD45?/?
NK cells are defective in cytokine and chemokine production
when stimulated by ITAM-based receptors.
CD45E613R mice harbor a single amino acid mutation in the
CD45 juxtamembrane wedge domain thought to play a regula-
tory role by inhibiting phosphatase activity upon receptor ho-
modimerization. This point mutation results in aberrant T and B
cell activation, leading to a lymphoproliferative disorder and
mouse Ba?F3 cells transduced with murine CMV m157 (A) or RAE-1? (B). Untransduced Ba?F3 cells served as a measure of background killing. (C) ADCC against
RMA or anti-CD90 mAb-coated RMA targets in the presence or absence of a soluble neutralizing anti-CD16?32 mAb. Results are representative of two to four
CD45 is not required for ITAM receptor-induced NK cell-mediated cytotoxicity. (A and B) Comparison of the ability of WT and CD45?/? NK cells to kill
Hesslein et al.PNAS ?
May 2, 2006 ?
vol. 103 ?
no. 18 ?
autoimmunity (34). We hypothesized that if CD45?/? NK cells
are unable to produce IFN?, then CD45E613R NK cells might
exhibit enhanced IFN? production. However, comparison of
CD45E613R and WT NK cells stimulated via CD16, NK1.1,
NKG2D, or Ly49D revealed no difference in cytokine produc-
tion (data not shown).
Cells. To determine the cause for the defect in cytokine and
chemokine production, we isolated RNA from WT and
CD45?/? NK cells stimulated with anti-NK1.1 or control mAb
and performed quantitative RT-PCR. IFN? and MIP-1? tran-
scripts were greatly increased upon stimulation in WT NK cells
but unchanged from unstimulated controls in CD45?/? NK
cells (Fig. 3). Thus, the cytokine and chemokine production
defect in CD45?/? NK cells is explained by the failure to
accumulate mRNA from these genes.
MAPK Activation Is Deficient upon Receptor Engagement in CD45?/?
NK Cells. The MAPKs ERK, JNK, and p38 function at important
downstream signaling branch points, where they serve as pen-
ultimate regulators of cytokine and chemokine gene expression
by controlling the activation of transcription factors such as
Elk-1, ATF2, and c-Jun. These MAPKs have been shown to be
involved in the regulation of IFN?, granulocyte–macrophage
colony-stimulating factor, MIP-1?, MIP-1?, and TNF? gene
expression (22, 35–37). Thus, we examined the activation?
phosphorylation state of these MAPKs after receptor engage-
ment in CD45?/? NK cells. ERK was markedly activated upon
NK1.1, NKG2D, or CD16 stimulation in WT NK cells but was
minimally activated in CD45?/? cells (Fig. 4). After NK1.1
stimulation, JNK phosphorylation was decreased in CD45?/?
NK cells relative to WT controls (Fig. 4C). p38 phosphorylation
was minimally (if at all) increased in WT or CD45?/? NK cells
in response to anti-NK1.1 (Fig. 4D). CD45?/? NK cells are
capable of activating ERK and JNK because stimulation with
PMA yielded phosphorylation levels equivalent to WT NK cells
(data not shown).
Basal Tyrosine Phosphorylation Is Increased in CD45?/? NK Cells.
Because CD45 is a transmembrane protein tyrosine phosphatase
and is important for the initiation and regulation of signaling
pathways, we examined CD45?/? NK cells for any alteration in
tyrosine phosphorylation. Freshly isolated CD45?/? NK cells
contained a number of proteins that were hyperphosphorylated
signaling by CD45?/? NK cells. WT and CD45?/? NK cells were incubated on
plate-bound mAbs for NKG2D and Ly49D (Left) or NK1.1 and mouse IgG2a (a
natural ligand of CD16) (Right). C, control mAb. Supernatants were harvested
performed in the presence of soluble anti-CD16?32 mAb (2.4G2) to block Fc
receptor activation, except in experiments in which NK cells were stimulated
via CD16 by plate-bound mouse IgG2a. (C) IFN? (Left) and MIP-1? (Right)
secretion after PMA (25 ng?ml) and ionomycin (1 ?g?ml) stimulation. Results
are representative of two to five independent experiments.
Deficient cytokine and chemokine production in response to ITAM
cells. Quantitative RT-PCR was performed on RNA harvested from WT and
CD45?/? NK cells stimulated with plate-bound anti-NK1.1 mAb. IFN? (A) and
MIP-1? (B) mRNA levels were analyzed in two independent experiments with
Cytokine and chemokine mRNA levels are reduced in CD45?/? NK
www.pnas.org?cgi?doi?10.1073?pnas.0601851103 Hesslein et al.
relative to WT cells (Fig. 5). Examination of IL-2-expanded NK
cells revealed an even more surprising and striking increase in
the number of tyrosine phosphoproteins and level of tyrosine
phosphorylation in CD45?/? NK cells compared with WT NK
cells (Fig. 5). These changes were noted in the unstimulated
state, suggesting a profound effect of CD45 in regulating the
basal state of tyrosine phosphoproteins in freshly isolated and
IL-2-expanded cells even before ITAM receptor-dependent
pathways are activated.
We have shown that CD45 is not required for ITAM-dependent
target cell killing induced by the CD16 or Ly49H receptors,
which pair and signal through Fc?RI? or DAP12 ITAM-bearing
adapters, respectively, and is not required for NKG2D-mediated
cytotoxicity. Because CD45 deficiency had only a modest effect
do not require CD45 for activating cytotoxicity. Although
ITAM-dependent killing was somewhat reduced in CD45?/?
NK cells, it was still readily detectable.
In contrast, cytokine and chemokine production and mRNA
accumulation are severely affected by CD45 deficiency. IFN?
production was virtually undetectable when CD45?/? NK cells
were stimulated via NK1.1, CD16, NKG2D, or Ly49D. Because
these receptors signal via ITAM adapters, we conclude that
CD45 is essential for ITAM-based activation of cytokine and
chemokine production. Although we observed a defect in
dependent signaling may also regulate cytokine translation,
posttranslational modifications, or secretion. If there were ad-
ditional stages of protein expression that CD45 controls, these
would be masked by the severe defect in mRNA levels.
These findings reveal an unexpected separation in the require-
cell functions. It appears that this tyrosine phosphatase is
essential for ITAM receptor-induced cytokine and chemokine
production but is not absolutely required for cytotoxicity. CD45
tors. (A) ERK activation after stimulation of WT and CD45?/? NK cells with
with plate-bound mouse IgG2a to activate via CD16. (C) JNK activation after
stimulation with anti-NK1.1 mAb. Two exposures (short and long) are shown.
(D) p38 activation after stimulation with anti-NK1.1 mAb. Data are represen-
tative of two to seven experiments.
Defective MAPK signaling downstream of ITAM-based NK cell recep-
prepared from WT and CD45?/? NK cells that were freshly isolated or ex-
Left and Right are short and long exposures, respectively, of the same blot.
actin. Data are representative of two to nine independent experiments.
Hesslein et al. PNAS ?
May 2, 2006 ?
vol. 103 ?
no. 18 ?
is the only molecule to date that has this activity. In contrast,
Vav-1, Vav-2, and Vav-3 are required for NK cell-mediated
killing but not for cytokine production (38, 39). As we were
concluding our experiments, a study was published describing
the same dichotomy between target cell killing and cytokine and
chemokine production in CD45?/? NK cells (4).
We explored the molecular basis for this apparent dichotomy
in function, focusing first on MAPKs because of their known
importance in both functional outcomes. ERK1 and ERK2 have
been well studied in NK cells and are activated after receptor
stimulation (6–8, 40). ERK regulates both NK cell-mediated
cytotoxicity and cytokine production, as demonstrated through
the use of dominant-negative ERK constructs and MEK inhib-
itors (5–8). MEK is the upstream kinase that activates ERK. We
found deficient MAPK activation in CD45?/? NK cells. ERK
phosphorylation was mildly affected. Interestingly, we found
upon closer comparison between various published studies that
the IC50for the MEK inhibitor, PD098059, was much higher for
inhibition of NK cell target killing than was required to block
IFN? mRNA production or protein secretion (5–8). This finding
suggests a relative difference in the requirement for ERK in
cytotoxicity versus cytokine production. Thus, in CD45?/? NK
cells, the defect in ERK activation may explain, in part, the
defect seen in cytokine?chemokine production.
The role for p38 and JNK in NK cell effector function is less
clear. p38 activation has been noted upon stimulation of NK cells
(4, 9, 10). However, exposure of NK cells to p38 inhibitors
resulted in either no or partial inhibition of target killing (9, 10),
with inhibition occurring only at extremely high levels of inhib-
itor (10). The p38 inhibitors did not affect IFN? secretion, and
mRNA levels were only partially reduced at high inhibitor
concentrations (5, 10). Thus, p38 appears to not be required for
target killing or cytokine production. Although little is known
about its role in cytokine secretion, JNK activation has been
observed after NK cell receptor engagement (4, 10) (Fig. 4C).
However, expression of a dominant-negative JNK had no effect
on human NK cell-mediated cytotoxicity (10). In T cells, JNK
regulates IFN? expression (36, 37), raising the possibility that
JNK performs the same function in NK cells.
To explore the basis for the dichotomous defects in CD45?/?
phosphoprotein expression. We expected to see an effect mostly
limited to Src family kinases, based on previous studies of T and
B cells. Instead, we were surprised to observe an increased basal
tyrosine phosphorylation of many proteins when freshly isolated
CD45?/? NK cells were examined and an even more impressive
effect when these cells were expanded in IL-2. Freshly isolated
T cells from CD45?/? mice also demonstrate elevated basal
phosphorylation of several proteins, similar to NK cells (41)
(data not shown). How can one explain this effect on multiple
proteins? There are several possible explanations that remain to
be explored. First, some of these phosphorylated proteins are
most likely substrates of CD45, such as the Src family kinases,
known regulators of ITAM signaling. Second, because these Src
kinases are important in inhibitory receptor signaling, it is
possible that impairments or alterations in Src kinase function
lead to loss of tonic inhibitory signals, resulting in higher
amounts of basal activation. Finally, it is possible that the defects
in MAPK activation and in cytokine production are the result of
adaptation of the cells to elevated constitutive signaling and
compensating inhibitory feedback loops.
Stimulation of the same receptor?ITAM adapter complex
results in differential requirements for CD45 for target cytotox-
icity and cytokine production. It is likely that strength and?or
duration of signal may determine the requirement for CD45.
Release of cytotoxic granules occurs near the cell surface, in
close proximity to many cell signaling components that are
activated upon receptor?ITAM activation. This process requires
little time (minutes) between receptor crosslinking and granule
release. Conversely, the induction of cytokines (42) by receptor
engagement is a lengthier process (hours), involving signal
transduction, gene transcription, RNA processing, translation,
and secretion. Sustained signaling may be needed for induction
of cytokines, whereas brief stimulation may be all that is needed
for a robust cytotoxicity response by NK cells (43).
Materials and Methods
Mice. Inbred C57BL?6 mice were purchased from Charles River
Laboratories or the National Cancer Institute. CD45?/? mice
(exon 6 disruption) were obtained from E. Brown (University of
California, San Francisco) (15). All mice, including CD45E613R
(34) (mice having a point mutation substituting an R for E at
position 613 in CD45), were backcrossed eight generations onto
pathogen-free facility, and experiments were performed accord-
ing to University of California, San Francisco, Institutional
Animal Care and Use Committee guidelines.
Flow Cytometry. Freshly isolated NK cells and IL-2-cultured NK
cells were stained with mAbs against CD3, NK1.1, Ly49D,
Ly49G2, Ly49A, and DX5 (VLA-2) (BD Biosciences); NKG2D
(CX5) (eBioscience); and CD16?32 (2.4G2) (Harlan). Anti-
Ly49H mAb (3D10) was a generous gift from W. Yokoyama
(Washington University, St. Louis). Phycoerythrin-conjugated
goat anti-rabbit and donkey anti-mouse IgG F(ab?)2fragments
were used as second-step reagents (Jackson ImmunoResearch).
Flow cytometry was performed by using a FACScan (BD
Preparation of NK Cells. Cells from spleen or liver were isolated,
and red blood cells were lysed with ACK buffer (BioWhittaker).
For the isolation of fresh cells, spleen cells were first stained with
anti-CD16?32 mAb to block Fc receptors and then stained with
fluorescent-labeled anti-CD3 and anti-NK1.1. NK1.1?CD3?
cells were isolated by using a FACSAria (BD Biosciences). For
the generation of IL-2-expanded NK cells, splenocytes were
stained with anti-CD4 and anti-CD8 mAbs and then incubated
with goat anti-rat IgG-coated and goat anti-mouse IgG-coated
magnetic beads (Qiagen) to deplete T and B cells. NK cells were
stained with phycoerythrin-conjugated DX5 mAb and positively
selected by using an anti-phycoerythrin mAb-conjugated mag-
netic bead system (Miltenyi Biotec or StemCell Technologies).
The resulting NK cells were cultured with 4,000 units?ml re-
combinant human IL-2 (National Institutes of Health Biological
Resources Branch Preclinical Repository) and used after 6–8
51Cr Release Assay. Mouse Ba?F3 pro-B cells, Ba?F3 cells trans-
duced with Rae-1? (44) or murine CMV m157 (27), and RMA
T cell lymphoma cells were cultured in RPMI medium 1640
containing 10% FCS, 2 mM glutamine, 50 units?ml penicillin,
and 50 ?g?ml streptomycin. IL-2-activated NK cells were used as
effector cells in a 4- to 6-h51Cr cytotoxicity assay (45). For
ADCC, RMA cells were incubated with or without 10 ?g?ml
anti-CD90 (Thy1) (provided by Salim Dhanji and Hung-sai Teh,
University of Vancouver, Vancouver) (33), spun down, and
10 ?g?ml anti-CD16?32 mAb (2.4G2) (to block ADCC) for 30
min and then mixed with RMA or anti-CD90 mAb-coated RMA
target cells at various effector:target ratios.
Stimulation of NK Cells. Ninety-six-well plates (BD Biosciences)
were coated with mAb or ligand as described (45). H60 extra-
cellular domain–human IgG1 Fc fusion protein was generously
provided by J. P. Houchins (R & D Systems) (46). CD16
www.pnas.org?cgi?doi?10.1073?pnas.0601851103 Hesslein et al.
stimulation was performed by incubating NK cells on tissue Download full-text
culture plates coated with a nonspecific mouse IgG mAb in the
gave robust cytokine production when cultured on IgG-coated
plates. Addition of soluble anti-CD16?32 mAb binds CD16 and
prevents binding to the IgG-coated plastic plates, thereby block-
ing CD16-dependent NK cell activation. A total of 2 ? 105cells
per well were plated, centrifuged, and incubated for 4–18 h for
cytokine and chemokine secretion assays, 4–8 h for RNA
analysis, and 15, 30, 45, or 60 min for preparation of protein
lysates for Western blot analysis (all time points yielded robust
signals). ELISAs for IFN? (R & D Systems), MIP-1? (BD
Biosciences), and granulocyte–macrophage colony-stimulating
factor (eBioscience) were performed according to the manufac-
mRNA Quantitation. NK cells were stimulated with plate-bound
mAb as described above. NK cell RNA was DNase-treated and
converted to cDNA by using standard methods. Quantitative
PCR was performed as described (48) and normalized to the
amount of hypoxanthine phosphoribosyltransferase transcripts.
Probe sequences were FAM-CACAGGTCCAGCGCCAAG-
CATTC-TAMRA (IFN?) and FAM-CTCTGACCCTC-
CCACTTCCTGCTGTTT-TAMRA (MIP-1?). PCR primers
were ATGCATTCATGAGTATTGCCAAGT (IFN? sense),
GCTGGATTCCGGCAACAG (IFN? antisense), CCAGGGT-
TCTCAGCACCAAT (MIP-1? sense), and GCTGCCGGGAG-
GTGTAAGA (MIP-1? antisense).
MAPK Activation and Tyrosine Phosphorylation. IL-2-expanded NK
cells were removed from IL-2 ?18 h before stimulation. NK cells
were serum-starved for 1–2 h before lysis and analysis for
tyrosine phosphorylation. NK cells stimulated with plate-bound
mAb were lysed directly in the 96-well plate by the addition of
2?-reducing SDS?PAGE loading dye. Lysates were loaded onto
12% polyacrylamide gels and transferred to polyvinylidene
fluoride membranes by using standard Western blotting tech-
niques. The blots were incubated with antibodies to actin (Santa
Cruz Biotechnology), phosphotyrosine (4G10), ERK, phospho-
ERK, JNK, phospho-JNK, p38, and phospho-p38 (Cell Signaling
Technology), probed with horseradish peroxidase-conjugated
anti-rabbit IgG (Cell Signaling Technology), and developed by
using ECL? reagents (Amersham Pharmacia Bioscience).
We thank Jay Ryan, Tony DeFranco, Cliff Lowell, Clare Abram, Mark
Klinger, Jessica Hamerman, Melissa Lodoen, and Susan Watson for
advice; the L.L.L. and A.W. laboratories for support; and Allison Tan
and Jessica Jarjoura for expert technical assistance. This work was
supported by National Institutes of Health Grants AI068129 (to L.L.L.),
K08 CA098418-01 (to M.L.H.), and AI35297 (to A.W.) and by the
Howard Hughes Medical Research Institute (A.W.). D.G.T.H. is sup-
ported by a Cancer Research Institute Fellowship, and L.L.L. is an
American Cancer Society Research Professor.
1. Lanier, L. L. (2005) Annu. Rev. Immunol. 23, 225–274.
2. Lanier, L. L. (2003) Curr. Opin. Immunol. 15, 308–314.
3. Samelson, L. E. (2002) Annu. Rev. Immunol. 20, 371–394.
4. Huntington, N. D., Xu, Y., Nutt, S. L. & Tarlinton, D. M. (2005) J. Exp. Med.
5. Ortaldo, J. R., Bere, E. W., Hodge, D. & Young, H. A. (2001) J. Immunol. 166,
6. Trotta, R., Puorro, K. A., Paroli, M., Azzoni, L., Abebe, B., Eisenlohr, L. C.
& Perussia, B. (1998) J. Immunol. 161, 6648–6656.
7. Wei, S., Gamero, A. M., Liu, J. H., Daulton, A. A., Valkov, N. I., Trapani, J. A.,
Larner, A. C., Weber, M. J. & Djeu, J. Y. (1998) J. Exp. Med. 187, 1753–1765.
8. Wei, S., Gilvary, D. L., Corliss, B. C., Sebti, S., Sun, J., Straus, D. B., Leibson,
P. J., Trapani, J. A., Hamilton, A. D., Weber, M. J. & Djeu, J. Y. (2000)
J. Immunol. 165, 3811–3819.
9. Chini, C. C., Boos, M. D., Dick, C. J., Schoon, R. A. & Leibson, P. J. (2000)
Eur. J. Immunol. 30, 2791–2798.
& Perussia, B. (2000) J. Immunol. 165, 1782–1789.
11. Einspahr, K. J., Abraham, R. T., Binstadt, B. A., Uehara, Y. & Leibson, P. J.
(1991) Proc. Natl. Acad. Sci. USA 88, 6279–6283.
12. Colucci, F., Schweighoffer, E., Tomasello, E., Turner, M., Ortaldo, J. R.,
Vivier, E., Tybulewicz, V. L. & Di Santo, J. P. (2002)Nat. Immunol. 3, 288–294.
13. Lowin-Kropf, B., Kunz, B., Schneider, P. & Held, W. (2002) Eur. J. Immunol.
14. Hermiston, M. L., Xu, Z. & Weiss, A. (2001) Annu. Rev. Immunol. 21, 107–137.
15. Kishihara, K., Penninger, J., Wallace, V. A., Kundig, T. M., Kawai, K.,
Wakeham, A., Timms, E., Pfeffer, K., Ohashi, P. S., Thomas, M. L., et al. (1993)
Cell 74, 143–156.
16. Byth, K. F., Conroy, L. A., Howlett, S., Smith, A. J., May, J., Alexander, D. R.
& Holmes, N. (1996) J. Exp. Med. 183, 1707–1718.
17. Poggi, A., Pardi, R., Pella, N., Morelli, L., Sivori, S., Vitale, M., Revello, V.,
Moretta, A. & Moretta, L. (1993) Eur. J. Immunol. 23, 2454–2463.
18. Giezeman-Smits, K. M., Gorter, A., van Vlierberghe, R. L., v Eendenburg,
J. D., Eggermont, A. M., Fleuren, G. J. & Kuppen, P. J. (1999) J. Immunol. 163,
19. Starling, G. C., Davidson, S. E., McKenzie, J. L. & Hart, D. N. (1987)
Immunology 61, 351–356.
20. Starling, G. C. & Hart, D. N. (1990) Immunology 71, 190–195.
21. Xu, X. & Chong, A. S. (1995) J. Immunol. 155, 5241–5248.
23. Martin, S. M., Mehta, I. K., Yokoyama, W. M., Thomas, M. L. & Lorenz, R. G.
(2001) J. Immunol. 166, 6066–6073.
24. Yamada, H., Kishihara, K., Kong, Y.-Y. & Nomoto, K. (1996) J. Immunol. 157,
25. Zompi, S., Hamerman, J. A., Ogasawara, K., Schweighoffer, E., Tybulewicz,
V. L., Santo, J. P., Lanier, L. L. & Colucci, F. (2003) Nat. Immunol. 4, 565–572.
26. Smith, H. R., Heusel, J. W., Mehta, I. K., Kim, S., Dorner, B. G., Naidenko,
O. V., Iizuka, K., Furukawa, H., Beckman, D. L., Pingel, J. T., et al. (2002) Proc.
Natl. Acad. Sci. USA 99, 8826–8831.
27. Arase, H., Mocarski, E. S., Campbell, A. E., Hill, A. B. & Lanier, L. L. (2002)
Science 296, 1323–1326.
28. Smith, K. M., Wu, J., Bakker, A. B. H., Phillips, J. H. & Lanier, L. L. (1998)
J. Immunol. 161, 7–10.
29. Wu, J., Song, Y., Bakker, A. B. H., Bauer, S., Groh, V., Spies, T., Lanier, L. L.
& Phillips, J. H. (1999) Science 285, 730–732.
30. Diefenbach, A., Tomasello, E., Lucas, M., Jamieson, A. M., Hsia, J. K., Vivier,
E. & Raulet, D. H. (2002) Nat. Immunol. 3, 1142–1149.
31. Arase, H., Arase, N. & Saito, T. (1996) J. Exp. Med. 183, 2391–2396.
32. Lanier, L. L., Yu, G. & Phillips, J. H. (1991) J. Immunol. 146, 1571–1576.
33. Dhanji, S., Tse, K. & Teh, H. S. (2005) J. Immunol. 174, 1253–1258.
34. Majeti, R., Zheng, X., Parslow, T. G., Olson, J. L., Daikh, D. I., Killeen, N. &
Weiss, A. (2000) Cell 103, 1059–1070.
35. Melchjorsen, J., Sorensen, L. N. & Paludan, S. R. (2003) J. Leukocyte Biol. 74,
36. Dong, C., Davis, R. J. & Flavell, R. A. (2002) Annu. Rev. Immunol. 20, 55–72.
37. Szabo, S. J., Sullivan, B. M., Peng, S. L. & Glimcher, L. H. (2003) Annu. Rev.
Immunol. 21, 713–758.
38. Cella, M., Fujikawa, K., Tassi, I., Kim, S., Latinis, K., Nishi, S., Yokoyama,
W. M., Colonna, M. & Swat, W. (2004) J. Exp. Med. 200, 817–823.
39. Colucci, F., Rosmaraki, E., Bregenholt, S., Samson, S. I., Di Bartolo, V.,
Turner, M., Vanes, L., Tybulewicz, V. & Di Santo, J. P. (2001) J. Exp. Med. 193,
40. McVicar, D. W., Taylor, L. S., Gosselin, P., Willette-Brown, J., Mikhael, A. I.,
Geahlen, R. L., Nakamura, M. C., Linnemeyer, P., Seaman, W. E., Anderson,
S. K., et al. (1998) J. Biol. Chem. 273, 32934–32942.
41. Stone, J. D., Conroy, L. A., Byth, K. F., Hederer, R. A., Howlett, S., Takemoto,
Y., Holmes, N. & Alexander, D. R. (1997) J. Immunol. 158, 5773–5782.
42. Valitutti, S., Dessing, M., Aktories, K., Gallati, H. & Lanzavecchia, A. (1995)
J. Exp. Med. 181, 577–584.
44. Cerwenka, A., Baron, J. L. & Lanier, L. L. (2001) Proc. Natl. Acad. Sci. USA
45. Ogasawara, K., Yoshinaga, S. K. & Lanier, L. L. (2002) J. Immunol. 169,
46. O’Callaghan, C. A., Cerwenka, A., Willcox, B. E., Lanier, L. L. & Bjorkman,
P. J. (2001) Immunity 15, 201–211.
47. Unkeless, J. C. (1979) J. Exp. Med. 150, 580–596.
48. Ogasawara, K., Hamerman, J. A., Hsin, H., Chikuma, S., Bour-Jordan, H.,
Chen, T., Pertel, T., Carnaud, C., Bluestone, J. A. & Lanier, L. L. (2003)
Immunity 18, 41–51.
Hesslein et al.PNAS ?
May 2, 2006 ?
vol. 103 ?
no. 18 ?