Homing and Function of Human Skin γδ T
Cells and NK Cells: Relevance for Tumor
Surveillance
This information is current as
of June 16, 2017.
Subscription
Permissions
Email Alerts
J Immunol 2006; 176:4331-4336; ;
doi: 10.4049/jimmunol.176.7.4331
http://www.jimmunol.org/content/176/7/4331
This article cites 38 articles, 13 of which you can access for free at:
http://www.jimmunol.org/content/176/7/4331.full#ref-list-1
Information about subscribing to The Journal of Immunology is online at:
http://jimmunol.org/subscription
Submit copyright permission requests at:
http://www.aai.org/About/Publications/JI/copyright.html
Receive free email-alerts when new articles cite this article. Sign up at:
http://jimmunol.org/alerts
The Journal of Immunology is published twice each month by
The American Association of Immunologists, Inc.,
1451 Rockville Pike, Suite 650, Rockville, MD 20852
Copyright © 2006 by The American Association of
Immunologists All rights reserved.
Print ISSN: 0022-1767 Online ISSN: 1550-6606.
Downloaded from http://www.jimmunol.org/ by guest on June 16, 2017
References
Lisa M. Ebert, Simone Meuter and Bernhard Moser
The Journal of Immunology
Homing and Function of Human Skin ␥␦ T Cells and NK
Cells: Relevance for Tumor Surveillance
Lisa M. Ebert,1 Simone Meuter, and Bernhard Moser2
U
nder conditions of homeostasis, human skin contains an
extensive network of leukocytes, including dendritic
cells (DCs),3 T cells, NK cells, mast cells, and macrophages (1–3). These leukocytes presumably function together in
maintaining skin integrity by continuously surveying for the presence of foreign Ags or signs of cellular stress. The best-characterized mechanism of skin protection involves cutaneous DCs (Langerhans cells and dermal DCs), which migrate to skin-associated
LNs for the initiation of tolerance or, in the case of skin infection,
antimicrobial immune responses (4, 5). In this manner, microbespecific T cells acquire cell surface molecules for guidance to affected skin to combat local infectious agents. Address cues for
effector T cell migration to inflamed skin include E-selectin ligand
(known as cutaneous lymphocyte-associated Ag/CLA in humans),
as well as certain receptors for chemokines expressed in inflamed
skin, including CCR4, CCR6, and CCR10 (6 –9). Thus, the mechanisms by which effector T cells are recruited from blood to inflamed skin are relatively well understood.
In contrast, we are just beginning to understand the migration
properties and function of T cells and other lymphocytes within
healthy tissue (10, 11). We recently identified the chemokine receptor CCR8 as a specific marker of ␣␤ T cells present in normal
skin and demonstrated that CCL1/I-309, the unique ligand for
CCR8, was constitutively produced at strategic sites in noninflamed skin (12). These cutaneous ␣␤ T cells secreted large
Institute of Cell Biology, University of Bern, Bern, Switzerland
Received for publication August 23, 2005. Accepted for publication January 17, 2006.
The costs of publication of this article were defrayed in part by the payment of page
charges. This article must therefore be hereby marked advertisement in accordance
with 18 U.S.C. Section 1734 solely to indicate this fact.
1
Current address: Ludwig Institute for Cancer Research (Melbourne Branch), Heidelberg, Victoria, 3084 Australia.
2
Address correspondence and reprint requests to Dr. Bernhard Moser, Institute of
Cell Biology, University of Bern, Baltzerstrasse 4, CH-3012 Bern, Switzerland. Email address: [email protected]
3
Abbreviations used in this paper: DC, dendritic cell; DETC, dendritic epidermal T
cell; LN, lymph node; MICA, MHC-I-related protein A; MICB, MHC-I-related protein B.
Copyright © 2006 by The American Association of Immunologists, Inc.
amounts of proinflammatory cytokines and may thus contribute to
memory responses after contact with previously encountered Ags.
However, ␣␤ T cells are unlikely candidates for surveillance
against tumors and stressed tissue cells (12). Therefore, we turned
our attention to ␥␦ T cells and NK cells. In contrast to ␣␤ T cells,
␥␦ T cells recognize Ags without the need for processing and
presentation by classical MHC molecules (13, 14). The major subset of ␥␦ T cells in human peripheral blood expresses the V␦2 gene
segment and recognizes nonpeptide Ags of mostly microbial origin. In contrast, most ␥␦ T cells in epithelial tissues express the
V␦1 gene segment and respond to stress-associated factors, including heat shock proteins and the MHC-related molecules MHC-Irelated protein A (MICA) and MHC-I-related protein B (MICB),
which are induced on tumor and virus-infected cells (13–16). NK
cells similarly recognize abnormal cells using a repertoire of receptors that detect stress-associated molecules as well as the loss
of surface MHC protein (17).
Murine skin harbors a major population of V␦1⫹ ␥␦ T cells,
known as dendritic epidermal T cells (DETCs), which appear to
play a critical role in tumor surveillance, as evidenced in ␥␦ T
cell-deficient mice (18, 19). Surprisingly, however, a counterpart
of DETCs does not exist in human epidermis (1), and it is thus
unclear what types of lymphocyte mediate equivalent immune surveillance functions in human skin. In this study we demonstrate
that normal human dermis contains distinct populations of ␥␦ T
cells and NK cells, both of which express receptors for homing to
noninflamed skin and for recognition of allogeneic tumor cells.
Together, these cells are likely to make a unique contribution to the
elimination of cutaneous tumor and otherwise stressed cells.
Materials and Methods
Abs, ELISA, and flow cytometry
Fluorochrome-conjugated Abs to human proteins were purchased from the
following sources: rat Ab to CLA (HECA-452), mouse Abs to CD3
(UCHT1), CD16 (3G8), CD45 (HI30), CD56 (B159), pan-TCR-␥␦ (11F2),
V␦2 (B6.1), perforin (dG9), CCR4 (1G1), and CCR6 (11A9) were obtained
from BD Pharmingen. Mouse Ab to V␦1 (TS8.2) was purchased from
Endogen (Perbio Sciences), and mouse Abs to CD94 (DX22) and NKG2D
0022-1767/06/$02.00
Downloaded from http://www.jimmunol.org/ by guest on June 16, 2017
Normal (noninflamed) human skin contains a network of lymphocytes, but little is known about the homing and function of these
cells. The majority of ␣␤ T cells in normal skin express CCR8 and produce proinflammatory cytokines. In this study we examined
other subsets of cutaneous lymphocytes, focusing on those with potential function in purging healthy tissue of transformed and
stressed cells. Human dermal cell suspensions contained significant populations of V␦1ⴙ ␥␦ T cells and CD56ⴙCD16ⴚ NK cells,
but lacked the subsets of V␦2ⴙ ␥␦ T cells and CD56ⴙCD16ⴙ NK cells, which predominate in peripheral blood. The skin-homing
receptors CCR8 and CLA were expressed by a large fraction of both cell types, whereas chemokine receptors associated with
lymphocyte migration to inflamed skin were absent. Neither cell type expressed CCR7, although ␥␦ T cells up-regulated this lymph
node-homing receptor upon TCR triggering. Stimulation of cutaneous V␦1ⴙ ␥␦ T cell lines induced secretion of large amounts of
TNF-␣, IFN-␥, and the CCR8 ligand CCL1. In contrast to cutaneous ␣␤ T cells, both cell types had the capacity to produce
intracellular perforin and displayed strong cytotoxic activity against melanoma cells. We therefore propose that ␥␦ T cells and NK
cells are regular constituents of normal human skin with potential function in the clearance of tumor and otherwise stressed tissue
cells. The Journal of Immunology, 2006, 176: 4331– 4336.
4332
IMMUNE SURVEILLANCE LYMPHOCYTES IN NORMAL HUMAN SKIN
Statistical analysis
Data given in the text are presented as the mean ⫾ SEM. Statistical significance was assessed using unpaired Student’s t test, and p ⬍ 0.05 was
considered significant.
Results
V␦1⫹ ␥␦ T cells and CD16⫺ NK cells in normal human skin
(1D11) were obtained from eBioscience. Rat Ab to CCR7 (3D12; M. Lipp,
Department of Tumor Genetics and Immunogenetics, Max-Delbrück-Center for Molecular Medicine, Berlin, Germany) and rabbit Ab to CCR8 (12)
were detected using biotinylated donkey anti-rat IgG or biotinylated donkey anti-rabbit IgG, respectively (Jackson ImmunoResearch Laboratories),
followed by streptavidin-PE (DakoCytomation) or streptavidin-allophycocyanin (BD Biosciences). Isotype-matched control mAbs were purchased
from BD Biosciences. Perforin staining was performed after fixation with
2% paraformaldehyde and permeabilization with 0.5% saponin. Flow cytometric analysis was performed on a FACSCalibur (BD Biosciences),
using propidium iodide to exclude dead cells. OptEIA ELISA kits for the
detection of IFN-␥ and TNF-␣ were obtained from BD Pharmingen and
performed according to the manufacturer’s instructions. The CCL1 ELISA
was performed as described previously (12).
Cell preparations
Split skin was obtained using a dermatome and normal skin from patients
undergoing abdominoplastic or mammareduction surgery. For most experiments, skin was digested in dispase II (1.25 U/ml; Roche) and collagenase
D (1 mg/ml; Roche) for 15–30 min at 37°C, followed by removal of the
epidermis and trypsin/EDTA digestion (30 min, 37°C) and gentle homogenization of remaining dermal fragments. Cells were filtered and either
analyzed immediately or further purified by magnetic isolation of CD45⫹
leukocytes using the MACS system (Miltenyi Biotec). For the isolation of
cells for chemotaxis or the analysis of molecules sensitive to dispase and
trypsin (CCR4 and CCR6), skin was digested in collagenase D only. Skin
␥␦ T cell clones were generated by first enriching bulk dermal cells using
a ␥␦ T cell isolation kit (Miltenyi Biotec), followed by sorting single TCR␥␦⫹ cells into each well of round-bottom, 96-well plates containing 5 ⫻
104 irradiated (40 Gy) allogeneic PBMC, 5 ⫻ 103 irradiated (70 Gy) EBVtransformed B cells, 500 ng/ml PHA, and 400 U/ml IL-2. Growing clones
were expanded in 200 U/ml IL-2 and restimulated every ⬃3 wk. The ␣␤
T cell clones have been described previously (12).
Functional assays
Four-hour Transwell chemotaxis assays were performed using a 5-␮m pore
size polycarbonate filter (Corning Costar) using cells recovered at 37°C for
2 h before the assay. Migrated cells were collected from the lower compartment and stained and quantified by flow cytometry relative to a fixed
number of polystyrene beads. Calcein release cytotoxicity assays were performed as previously described (20). The cytokine production capacity of
T cell clones was determined after stimulation with PMA (20 ng/ml) and
ionomycin (1 ␮M) for 24 h at a cell concentration of 1.6 ⫻ 106/ml.
A subpopulation of cutaneous NK cells expresses receptors for
homing to noninflamed skin
Based on our previous study, CCR8 and its chemokine ligand
CCL1 were proposed to provide a selective homing mechanism for
the entry of ␣␤ T cells to the skin under steady-state conditions
(12). As illustrated in Fig. 2a, a population of skin NK cells also
expressed CCR8. These cells differed from the majority of CCR8⫺
NK cells by their low side-scatter profile, suggesting that the
CCR8⫹ NK cells have reduced granularity. These two NK cell
populations also differed in the expression of CLA, which was
coexpressed with CCR8 on NK cells with low side scatter. The
significance of the inverse relationship between CCR8 and CLA
expression, on the one hand, and side scatter characteristics, on the
other hand, is presently unclear. All skin NK cells, regardless of
side scatter characteristics, lacked the lymph node (LN)-homing
chemokine receptor CCR7. CCR8 on NK cells was functional,
because it enabled freshly isolated cells to migrate toward CCL1 in
a dose-dependent manner (Fig. 2b). This is in clear contrast to NK
cells in peripheral blood, which completely lack CCR8 (Fig. 2c),
but a subset of which express CCR7 (22).
Cutaneous ␥␦ T cells display skin-homing properties
Peripheral blood ␥␦ T cells demonstrate inflammatory migration
features and, thus, differ from the majority of blood ␣␤ T cells,
which express CCR7 for continuous recirculation through secondary lymphoid tissues (23). In human skin, however, we found that
the two T cell subsets largely shared a selection of tissue-homing
receptors. Flow cytometric analysis (Fig. 3a) revealed that approximately half (53.1 ⫾ 6.9%) the ␥␦ T cells in normal skin express
CCR8, and almost all cells (88.9 ⫾ 7.7%) were positive for CLA.
In contrast, CCR7 was not detectable, and two chemokine receptors associated with T cell homing to inflamed skin (CCR4 and
CCR6) (10, 11, 24) were also absent. As an internal control (12),
the majority of cutaneous ␣␤ T cells were positive for CCR8 and
CLA, and only minor fractions, or none at all, expressed CCR4,
CCR6, or CCR7 (Fig. 3a). In comparison with skin, ␥␦ T cells in
peripheral blood lacked CCR8, and only a few cells were positive
Downloaded from http://www.jimmunol.org/ by guest on June 16, 2017
FIGURE 1. Normal human skin contains V␦1⫹ ␥␦ T cells and
CD16⫺CD56⫹ NK cells. Cell suspensions (n ⫽ 5– 6) were prepared from
noninflamed skin and analyzed by flow cytometry, gating on bulk CD45⫹
leukocytes (a), CD3⫹ T cells (b), or CD45⫹CD3⫺ cells (c).
Normal skin tissue from patients undergoing mammareduction or
abdominoplastic surgery was digested to release the dermal cells
into suspension, which was subsequently analyzed by flow cytometry. Among the CD45⫹ leukocytes, populations of ␥␦ T cells and
NK cells could be readily identified (Fig. 1a). On the average, from
six skin samples, 3.8 ⫾ 1.1% of skin leukocytes were (CD3⫹␥␦TCR⫹) ␥␦ T cells, whereas 10.6 ⫾ 0.9% were (CD3⫺CD56⫹) NK
cells. ␥␦ T cells represented 4.7 ⫾ 1.1% of the total cutaneous
CD3⫹ T cells (Fig. 1b), similar to previous estimates based on
immunohistochemical studies (1). The V␦1⫹ subset of ␥␦ T cells
predominated in normal human skin, because an Ab to the V␦1
chain, but not an Ab to the V␦2 chain, stained virtually identical
numbers of cells, as detected with an Ab to pan-TCR-␥␦.
Based on the expression of CD16, NK cells are divided into two
subpopulations (21). The majority of peripheral blood NK cells
express CD16, whereas only a minor subset is CD16⫺. In skin,
however, the CD56⫹ NK cells uniformly lacked CD16 (Fig. 1c),
indicating a difference to the majority of circulating NK cells.
The Journal of Immunology
4333
FIGURE 2. A subset of cutaneous NK cells expresses CCR8 and CLA and migrates toward CCL1. a, Dermal cell suspensions were analyzed by flow
cytometry, gating on CD45⫹CD3⫺ CD56⫹ cells, and the intensity of staining for CCR8, CLA, or CCR7 was plotted against the side-scatter profile (n ⫽
3– 4). b, Skin cells were subject to Transwell migration assays using CCL1 (F) or medium alone (E), and total migrated NK cells (CD3⫺ CD56⫹) were
enumerated using flow cytometry relative to a standard number of polystyrene beads. c, PBMC were analyzed by flow cytometry, gating on CD3⫺CD56⫹
NK cells (n ⫽ 4).
for CLA (Fig. 3b), underscoring the unique nature of skin-resident
␥␦ T cells.
The yield of primary ␥␦ T cells isolated from noninflamed skin
tissue did not allow a detailed investigation; therefore, functional
and phenotypic studies were performed with a panel of skin-derived ␥␦ T cell clones. Similar to freshly isolated skin ␥␦ T cells,
approximately half the clones expressed readily detectable levels
of CCR8 (⬎20% CCR8 positivity; Fig. 4a, left panel). Under the
present culture conditions, CCR8⫺ clones did not up-regulate the
expression of this chemokine receptor, whereas CCR8⫹ clones
maintained their respective levels, suggesting that CCR8 is a stable
feature in distinct cutaneous ␥␦ T cells. As with ␣␤ T cells (12),
the conditions required for induction of CCR8 expression in
CCR8⫺ ␥␦ T cells are not known. Chemotaxis assays confirmed
that CCR8 on receptor-positive ␥␦ T cell clones was functional, as
shown previously for ␣␤ T cell clones (12) (data not shown). In
contrast, only one of the 21 clones examined expressed CCR4
(⬎20% CCR4 positivity; Fig. 4a, right panel), which is in keeping
with the lack of skin-associated inflammatory chemokine receptors
on fresh isolated cutaneous ␥␦ T cells. Interestingly, PHA activation of ␥␦ T cell clones resulted in a transient loss of cell surface
CCR8 (Fig. 4b). This suggests that ␥␦ T cells temporarily alter
their homing properties after local activation and may no longer be
retained in the skin environment, but instead gain responsiveness
to alternative chemokines (25). In keeping with this concept,
CCR7 was strikingly up-regulated on day 3 after activation (Fig.
4b), followed by complete down-regulation as cells returned to a
resting state (not shown). This observation suggests that locally
activated cutaneous ␥␦ T cells can assist adaptive immune responses after relocation to the draining LNs, as proposed recently
for blood ␥␦ T cells (26).
All clones tested produced very high levels of the proinflammatory cytokines TNF-␣ (7– 43 ng/ml) and IFN-␥ (94 –1485 ng/
ml) after stimulation with PMA and ionomycin (Fig. 4c). Additionally, they secreted large amounts of CCL1 (6 – 81 ng/ml),
suggesting that CCL1 was responsible in part for the transient loss
of cell surface CCR8.
Skin ␥␦ T cells and NK cells have the capacity to lyse
melanoma cells
To address the potential role of cutaneous ␥␦ T cells and NK cells
in tumor surveillance, their ability to mount cytolytic responses
against melanoma cells was examined. First, skin-derived ␥␦ T
cell clones were tested in cytotoxicity assays using heterologous
Downloaded from http://www.jimmunol.org/ by guest on June 16, 2017
Migration and functional properties of cultured, skin-derived ␥␦
T cells
melanoma cell lines as targets. Of seven clones tested, five exerted
robust cytotoxic responses against the melanoma cell line SKMel2, resulting in up to 90% target cell death at an E:T cell ratio
of 25:1, whereas two clones failed to do so (Fig. 5a, left panel).
Similar results were observed with the melanoma cell line HS-294
(not shown). In control experiments using skin-derived ␣␤ T cell
clones (six CD8⫹ clones and four CD4⫹ clones), eight clones
completely lacked lytic activity against SK-Mel2 and HS-294 cells
(not shown), and only two clones showed borderline activity (Fig.
5a, right panel). These data are not surprising, because the mode
of Ag recognition differs strikingly between ␣␤ T cells and ␥␦ T
cells (13, 14), and the frequency of memory CD8⫹ ␣␤ T cells with
alloreactivity is relatively low. Major mechanisms contributing to
lymphocyte-mediated cytotoxicity involve perforin, which is released at the contact zone between effector and target cells (27),
and activating NK receptors, such as NKG2D/CD314, which recognizes MHC-I-related MICA and MICB as well as other stressrelated molecules (15, 17). Perforin was not detected in freshly
isolated skin T cells (Fig. 5b). However, all 12 ␥␦ T cell clones
tested contained detectable levels of perforin, whereas only two of
26 skin-derived ␣␤ T cell clones were positive (Fig. 5c). The lack
of perforin in skin ␣␤ T cell clones is in line with their proposed
involvement in the control of memory responses rather than tissue
cell lysis (12). In contrast to ␣␤ T cells, freshly isolated ␥␦ T cells
were uniformly positive for NKG2D and maintained this activating receptor during subsequent cell culture (Fig. 5d). NKG2D signaling in NK cells and ␥␦ T cells leads to direct cell activation that
may result in target cell lysis, whereas cytotoxicity in CD8⫹ ␣␤ T
cells is largely controlled by their clonal TCRs recognizing specific tumor Ags (28, 29). Collectively, these data suggest that ␥␦ T
cells present in normal skin have the potential to eliminate stressed
cells, including tumor cells.
The cytotoxic capacity of cutaneous NK cells was examined in
cultured cells after expansion of CD45⫹ skin leukocytes for 2–3
wk in the presence of IL-2, followed by purification of
CD3⫺CD56⫹ NK cells by cell sorting. These lines demonstrated
clear cytotoxic activity against SK-Mel2 cells, with an average of
75% target cell lysis at an E:T cell ratio of 25:1 (Fig. 6a). Furthermore, these cell lines were uniformly positive for intracellular
perforin (Fig. 6b, right panel). Again, as noted for skin ␥␦ T cells,
freshly isolated skin NK cells completely lacked this lytic marker
(Fig. 6b, left panel). Remarkably, and in clear contrast to ␥␦ T
cells, NKG2D could not be detected on freshly isolated skin NK
cells, and its induction of expression required previous NK cell
culture in the presence of IL-2. Obviously, NK cells in healthy
human skin are devoid of immediate lytic activities, which is in
4334
IMMUNE SURVEILLANCE LYMPHOCYTES IN NORMAL HUMAN SKIN
keeping with a previous study demonstrating that the CD16⫺ subset of NK cells within LNs fully depends on activation for mounting lytic activity (30).
Discussion
Local immune surveillance for tumors and stressed tissue cells is
crucial for the maintenance of skin homeostasis. In mice, this function is largely fulfilled by DETCs in the epidermis (18, 19). However, because human epidermis lacks DETCs, it has remained unclear how these important functions are performed in humans. In
this study we demonstrate that human dermis contains significant
populations of ␥␦ T cells and NK cells that express receptors for
homing to normal skin and, in relation to their state of activation,
receptors for tumor cell killing. Although the presence of V␦1⫹ ␥␦
T cells in the human dermis has been previously documented (1),
their function and migration properties have remained elusive.
Their unique TCR repertoire indicates that they are not a random
sample from circulating ␥␦ T cells (31), a view supported by our
finding that cutaneous, but not blood, ␥␦ T cells expressed the
skin-homing receptors CCR8 and CLA. Human dermal ␥␦ T cells
uniformly expressed the activating receptor NKG2D for killing of
tumor or otherwise stressed cells (17). Although freshly isolated
cutaneous ␥␦ T cells lacked perforin, they acquired it upon cell
activation, in keeping with the observation that murine DETCs
require activation to develop cytotoxic function against melanoma
cells (32). In many respects, dermal ␥␦ T cells resemble V␦1⫹ ␥␦
T cells in human intestinal epithelium, which also killed heterologous intestinal tumor cells by a process involving MHC-related
MICA and MICB and probably other stress-related molecules (15,
33, 34).
Similar to ␥␦ T cells, the nature and function of NK cells present
in normal human skin are ill defined. We show in this study that
these cells constitute a significant fraction of dermal leukocytes. Of
note and in clear contrast to NK cells in peripheral blood and LNs
(17, 30), NKG2D is completely absent in cutaneous NK cells, but
is strongly induced upon in vitro culture. Similarly, perforin is
present in cultured, but not freshly isolated NK cells. Obviously,
healthy human dermis is devoid of NK cells with the capacity for
immediate target cell lysis. The majority of cutaneous NK cells
with a low side-scatter profile expressed both CCR8 and CLA, but
lacked CCR7 and therefore displayed a homing phenotype highly
reminiscent of cutaneous ␣␤ and ␥␦ T cells, but clearly different
from that of NK cells in peripheral blood (21, 22, 30). The major
subset of CD16⫹ NK cells in blood expresses CXCR1 and
CX3CR1 for homing to inflammatory sites and contains perforin
for immediate cytotoxic effector function. In contrast, the minor
subset of CD16⫺ NK cells expresses CD62L and CCR7 for recirculation through LNs and has no intracellular perforin. Cutaneous
NK cells also lack CD16 and perforin expression, but differ from
blood CD16⫺ NK cells with respect to migration properties. Interestingly, normal small-intestinal epithelium was also recently
shown to harbor a population of NK cells (35). Thus, in analogy to
memory ␣␤ T cells (10, 11, 24), subsets of CD16⫺ NK cells with
individual homing preferences for normal peripheral tissues may
exist.
In contrast to the rapid recruitment of effector lymphocytes to
sites of skin inflammation (9, 24), the mechanisms controlling the
steady-state traffic of memory T cells and their maintenance within
healthy tissue are not well understood (10, 11). CCL1, the single
ligand for CCR8, is produced in normal healthy skin, notably by
Downloaded from http://www.jimmunol.org/ by guest on June 16, 2017
FIGURE 3. Cutaneous ␥␦ T cells express receptors for homing to normal, but not inflamed, skin. a, Cell suspensions (n ⫽ 3–5) were prepared
from normal skin (a) or peripheral blood (b) and analyzed by flow cytometry, gating on CD3⫹ T cells (a) or CD3⫹TCR-␥␦⫹ cells (b).
FIGURE 4. Chemokine receptor expression and cytokine production
capacity of skin-derived ␥␦ T cell clones. a, Skin ␥␦ T cell clones were
analyzed in the resting state for the expression of CCR8 and CCR4 by flow
cytometry, and the percentage of cells positive relative to peptide-blocked
Ab (CCR8) or isotype control (CCR4) was determined. b, The expression
of CCR7 and CCR8 was assessed on skin ␥␦ T cell clones (n ⫽ 6) in the
resting state (left panels) or after 3 days of stimulation with PHA (right
panels). c, Cytokine production by skin ␥␦ T cell clones was assessed by
ELISA analysis of supernatants after 24-h stimulation with PMA and ionomycin. Cytokines were ⬍1 ng/ml in the supernatants of unstimulated ␥␦ T
cell cultures. Values are the mean from duplicate wells.
The Journal of Immunology
4335
FIGURE 5. ␥␦ T cells in normal skin express NKG2D and lyse melanoma cells after activation. a, Skin-derived ␥␦ (left panel) or ␣␤ (right
panel) T cell clones were tested in calcein release cytotoxicity assays using
the melanoma cell line SK-Mel2 as targets. Values represent the mean and
SEM calculated from triplicate wells for each clone. b, The expression of
perforin in freshly isolated skin T cells was assessed by flow cytometry
after gating on CD3⫹ cells. c, The expression of intracellular perforin was
determined for ␥␦ (left panel; n ⫽ 12) and ␣␤ (right panel; n ⫽ 26) T cell
clones by flow cytometry (open histograms). d, Flow cytometric analysis of
cell surface NKG2D in ␥␦ T cells and ␣␤ T cells (open histograms), either
immediately after isolation from normal human skin tissue or after polyclonal expansion during cell culture (n ⫽ 2). Filled histograms in c and d
represent staining with isotype-matched control Abs.
vascular and perivascular cells within superficial dermal plexus
and by epidermal Langerhans cells and melanocytes (12). This
chemokine system may regulate the recruitment and cutaneous positioning of blood CCR8⫹ ␣␤ T cells and, as proposed in this
study, a large fraction of cutaneous ␥␦ T cells and NK cells. Yet,
we still have little knowledge about the traffic parameters and the
life span of cutaneous lymphocytes (36, 37). The scarcity of
CCR8⫹ lymphocytes within peripheral blood may indicate that the
skin is their primary site of residence. In preliminary coculture
experiments, we have noticed that dermal fibroblasts supported the
slow proliferation of freshly isolated cutaneous T cells and NK
cells (L. M. Ebert and B. Moser, unpublished observations). Skin
is a rich source of growth factors and, in analogy to bone marrow
(38), may support homeostatic expansion in local lymphocytes. It
is possible that skin lymphocyte numbers are maintained by sev-
eral factors, including rates of skin recruitment and exit as well as
the rates of local survival and proliferation.
Distinct lymphocyte populations within healthy skin are likely
to play a crucial role in local immune surveillance. Skin-tropic
memory ␣␤ T cells, also referred to as peripheral immune surveillance T cells (11, 12), may contribute to memory responses against
previously encountered pathogens. However, cytotoxic peripheral
immune surveillance T cells with selectivity for skin tumors appear to be rare, suggesting that they may not participate in tumor
surveillance. Our present study has characterized for the first time
populations of human skin-homing ␥␦ T cells and NK cells with
the potential to elaborate lytic functions, which may involve molecules commonly associated with situations of tissue stress, such
as those present on transformed or virally infected cells (13–17).
Furthermore, we have shown that these three distinct subsets of
cutaneous lymphocytes share a common address code, enabling
their colocalization within noninflamed skin. Due to their specialization in Ag recognition and function, these diverse cell types
may cooperate in cutaneous immune surveillance against a wide
variety of challenges.
Acknowledgments
We are grateful to Drs. M. Brandes and P. Schaerli for critical discussion
and technical advice, and we thank Dr. A. Banic and his team of surgeons
at the Department of Plastic and Reparation Surgery, University of Bern,
for providing skin tissue.
Disclosures
The authors have no financial conflict of interest.
References
1. Bos, J. D., M. B. Teunissen, I. Cairo, S. R. Krieg, M. L. Kapsenberg, P. K. Das,
and J. Borst. 1990. T-cell receptor ␥␦ bearing cells in normal human skin. J. Invest. Dermatol. 94: 37– 42.
Downloaded from http://www.jimmunol.org/ by guest on June 16, 2017
FIGURE 6. NK cells in normal skin lack NKG2D and perforin, but lyse
melanoma cells after activation. a, NK cells were sorted from IL-2-expanded skin leukocyte cultures and tested in cytotoxicity assays using SKMel2 cells. Values are the mean and SEM calculated from triplicate wells
for cell lines derived from three different donors. b, Expression of intracellular perforin was determined for NK cells, either freshly isolated (open
histograms; left panel) or after expansion in IL-2 (right panel) by flow
cytometry (n ⫽ 3). c, Flow cytometric analysis of cell surface NKG2D
(open histograms) in freshly isolated (left panel) and cultured (right panel)
NK cells (n ⫽ 2). Filled histograms in b and c represent staining with
isotype-matched control Abs.
4336
IMMUNE SURVEILLANCE LYMPHOCYTES IN NORMAL HUMAN SKIN
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
tional analyses on recovered effector cells and supernatants. Clin. Diagn. Lab.
Immunol. 8: 1131–1135.
Cooper, M. A., T. A. Fehniger, and M. A. Caligiuri. 2001. The biology of human
natural killer-cell subsets. Trends Immunol. 22: 633– 640.
Inngjerdingen, M., B. Damaj, and A. A. Maghazachi. 2001. Expression and regulation of chemokine receptors in human natural killer cells. Blood 97: 367–375.
Brandes, M., K. Willimann, A. B. Lang, K. H. Nam, C. Jin, M. B. Brenner,
C. T. Morita, and B. Moser. 2003. Flexible migration program regulates ␥␦ T-cell
involvement in humoral immunity. Blood 102: 3693–3701.
Kunkel, E. J., and E. C. Butcher. 2002. Chemokines and the tissue-specific migration of lymphocytes. Immunity 16: 1– 4.
Foxman, E. F., J. J. Campbell, and E. C. Butcher. 1997. Multistep navigation and
the combinatorial control of leukocyte chemotaxis. J. Cell Biol. 139: 1349 –1360.
Brandes, M., K. Willimann, and B. Moser. 2005. Professional antigen-presentation function by human ␥␦ T cells. Science 309: 264 –268.
Kagi, D., B. Ledermann, K. Burki, R. M. Zinkernagel, and H. Hengartner. 1996.
Molecular mechanisms of lymphocyte-mediated cytotoxicity and their role in
immunological protection and pathogenesis in vivo. Annu. Rev. Immunol. 14:
207–232.
Jamieson, A. M., A. Diefenbach, C. W. McMahon, N. Xiong, J. R. Carlyle, and
D. H. Raulet. 2002. The role of the NKG2D immunoreceptor in immune cell
activation and natural killing. Immunity 17: 19 –29.
Das, H., V. Groh, C. Kuijl, M. Sugita, C. T. Morita, T. Spies, and J. F. Bukowski.
2001. MICA engagement by human V␥2V␦2 T cells enhances their antigendependent effector function. Immunity 15: 83–93.
Ferlazzo, G., D. Thomas, S. L. Lin, K. Goodman, B. Morandi, W. A. Muller,
A. Moretta, and C. Munz. 2004. The abundant NK cells in human secondary
lymphoid tissues require activation to express killer cell Ig-like receptors and
become cytolytic. J. Immunol. 172: 1455–1462.
Holtmeier, W., M. Pfander, A. Hennemann, T. M. Zollner, R. Kaufmann, and
W. F. Caspary. 2001. The TCR-␦ repertoire in normal human skin is restricted
and distinct from the TCR-␦ repertoire in the peripheral blood. J. Invest. Dermatol. 116: 275–280.
Kaminski, M. J., P. D. Cruz, Jr., P. R. Bergstresser, and A. Takashima. 1993.
Killing of skin-derived tumor cells by mouse dendritic epidermal T-cells. Cancer
Res. 53: 4014 – 4019.
Bauer, S., V. Groh, J. Wu, A. Steinle, J. H. Phillips, L. L. Lanier, and T. Spies.
1999. Activation of NK cells and T cells by NKG2D, a receptor for stressinducible MICA. Science 285: 727–729.
Wu, J., V. Groh, and T. Spies. 2002. T cell antigen receptor engagement and
specificity in the recognition of stress-inducible MHC class I-related chains by
human epithelial ␥␦ T cells. J. Immunol. 169: 1236 –1240.
Leon, F., E. Roldan, L. Sanchez, C. Camarero, A. Bootello, and G. Roy. 2003.
Human small-intestinal epithelium contains functional natural killer lymphocytes. Gastroenterology 125: 345–356.
Mackay, C. R. 1993. Immunological memory. Adv. Immunol. 53: 217–265.
Hunger, R. E., N. Yawalkar, L. R. Braathen, and C. U. Brand. 1999. The HECA452 epitope is highly expressed on lymph cells derived from human skin.
Br. J. Dermatol. 141: 565–569.
Di Rosa, F., and R. Pabst. 2005. The bone marrow: a nest for migratory memory
T cells. Trends Immunol. 26: 360 –366.
Downloaded from http://www.jimmunol.org/ by guest on June 16, 2017
2. Bos, J. D., and M. L. Kapsenberg. 1993. The skin immune system: progress in
cutaneous biology. Immunol. Today 14: 75–78.
3. Bangert, C., J. Friedl, G. Stary, G. Stingl, and T. Kopp. 2003. Immunopathologic
features of allergic contact dermatitis in humans: participation of plasmacytoid
dendritic cells in the pathogenesis of the disease? J. Invest. Dermatol. 121:
1409 –1418.
4. Kupper, T. S., and R. C. Fuhlbrigge. 2004. Immune surveillance in the skin:
mechanisms and clinical consequences. Nat. Rev. Immunol. 4: 211–222.
5. Kissenpfennig, A., S. Henri, B. Dubois, C. Laplace-Builhe, P. Perrin, N. Romani,
C. H. Tripp, P. Douillard, L. Leserman, D. Kaiserlian, et al. 2005. Dynamics and
function of Langerhans cells in vivo: dermal dendritic cells colonize lymph node
areas distinct from slower migrating Langerhans cells. Immunity 22: 643– 654.
6. Campbell, J. J., G. Haraldsen, J. Pan, J. Rottman, S. Qin, P. Ponath,
D. P. Andrew, R. Warnke, N. Ruffing, N. Kassam, et al. 1999. The chemokine
receptor CCR4 in vascular recognition by cutaneous but not intestinal memory T
cells. Nature 400: 776 –780.
7. Homey, B., W. Wang, H. Soto, M. E. Buchanan, A. Wiesenborn, D. Catron,
A. Muller, T. K. McClanahan, M. C. Dieu-Nosjean, R. Orozco, et al. 2000.
Cutting edge: the orphan chemokine receptor G protein-coupled receptor-2
(GPR-2, CCR10) binds the skin-associated chemokine CCL27 (CTACK/ALP/
ILC). J. Immunol. 164: 3465–3470.
8. Fitzhugh, D. J., S. Naik, S. W. Caughman, and S. T. Hwang. 2000. Cutting edge:
C-C chemokine receptor 6 is essential for arrest of a subset of memory T cells on
activated dermal microvascular endothelial cells under physiologic flow conditions in vitro. J. Immunol. 165: 6677– 6681.
9. Butcher, E. C., M. Williams, K. Youngman, L. Rott, and M. Briskin. 1999.
Lymphocyte trafficking and regional immunity. Adv. Immunol. 72: 209 –253.
10. Moser, B., M. Wolf, A. Walz, and P. Loetscher. 2004. Chemokines: multiple
levels of leukocyte migration control. Trends Immunol. 25: 75– 84.
11. Schaerli, P., and B. Moser. 2005. Chemokines: control of primary and memory
T-cell traffic. Immunol. Res. 31: 57–74.
12. Schaerli, P., L. Ebert, K. Willimann, A. Blaser, R. S. Roos, P. Loetscher, and
B. Moser. 2004. A skin-selective homing mechanism for human immune surveillance T cells. J. Exp. Med. 199: 1265–1275.
13. Hayday, A. C. 2000. ␥␦ cells: a right time and a right place for a conserved third
way of protection. Annu. Rev. Immunol. 18: 975–1026.
14. Carding, S. R., and P. J. Egan. 2002. ␥␦ T cells: functional plasticity and heterogeneity. Nat. Rev. Immunol. 2: 336 –345.
15. Groh, V., A. Steinle, S. Bauer, and T. Spies. 1998. Recognition of stress-induced
MHC molecules by intestinal epithelial ␥␦ T cells. Science 279: 1737–1740.
16. Ferrarini, M., E. Ferrero, L. Dagna, A. Poggi, and M. R. Zocchi. 2002. Human ␥␦
T cells: a nonredundant system in the immune-surveillance against cancer.
Trends Immunol. 23: 14 –18.
17. Lanier, L. L. 2005. NK cell recognition. Annu. Rev. Immunol. 23: 225–274.
18. Bergstresser, P. R., P. D. Cruz, Jr., and A. Takashima. 1993. Dendritic epidermal
T cells: lessons from mice for humans. J. Invest. Dermatol. 100: 80S– 83S.
19. Girardi, M., D. E. Oppenheim, C. R. Steele, J. M. Lewis, E. Glusac, R. Filler,
P. Hobby, B. Sutton, R. E. Tigelaar, and A. C. Hayday. 2001. Regulation of
cutaneous malignancy by ␥␦ T cells. Science 294: 605– 609.
20. Neri, S., E. Mariani, A. Meneghetti, L. Cattini, and A. Facchini. 2001. Calceinacetyoxymethyl cytotoxicity assay: standardization of a method allowing addi-