Granulocyte chemotactic protein-2 mediates adaptive
immunity in part through IL-8R interactions
Udai P. Singh,* Shailesh Singh,* Prosper N. Boyaka,† Jerry R. McGhee,† and
James W. Lillard Jr.*,†,1
*Department of Microbiology and Immunology, Morehouse School of Medicine, Atlanta, Georgia; and †University of
Alabama at Birmingham
Abstract: Chemokines constitute a large family of migration of lymphocytes to sites of disease, an important
structurally related proteins that play a role in prerequisite for host defense. Previous studies from our labo-
leukocyte migration and differentiation. Indeed, ratory suggest that chemokines such as lymphotactin , reg-
the early expression of human CXC chemokine ulated on activation, normal T expressed and secreted (RAN-
receptor 1 (hCXCR1) and hCXCR2 [homologous TES) , macrophage inﬂammatory protein-1 and -1
to mouse interleukin (IL)-8R ] ligands by the epi- (MIP-1 and MIP-1 ) , as well as interferon- (IFN- )-
thelium is a hallmark of the mucosal host defense. inducible protein-10, monokine induced by IFN- , and IFN-
Mice lack IL-8; however, granulocyte chemotactic inducible T cell -chemoattractant  can modulate mucosal
protein-2 (GCP-2)/lipopolysaccharide-induced CXC adaptive immunity. Hence, chemokines may play important
chemokine, a murine homologue of human GCP-2, roles in bridging innate and early inﬂammatory responses with
has 32% and 61% sequence identity to human IL-8 the adaptive immune system.
and GCP-2, respectively, and binds hCXCR1, The inﬂammatory condition is composed of multiple medi-
hCXCR2, and mouse IL-8R . To better understand ators that regulate leukocyte functions (e.g., activation and
the role of GCP-2 in adaptive immunity and as a nasal migration). Interleukin (IL)-8 is secreted by epithelial and
adjuvant, we characterized the exogenous effects of endothelial cells as well as by leukocytes and is chemotatic for
this CXC chemokine on cellular and humoral muco- neutrophils . Indeed, this CXC chemokine plays a crucial
sal immune responses. GCP-2 signiﬁcantly enhanced role in neutrophil migration during many inﬂammatory condi-
serum immunoglobulin G (IgG) and mucosal IgA an- tions [8, 9]. IL-8 binds with equal afﬁnity to human CXC
tibodies through increased cytokine secretion by chemokine receptor 1 (hCXCR1) and hCXCR2. It is important
CD4 T cells. These alterations in humoral and cel- to understand the cellular and molecular mechanism of
lular responses were preceded by an increase in the hCXCR1 and hCXCR2 ligands in leukocyte activation and
number of B cells in the nasal tract, a decrease in the differentiation. Mice lack a clear-cut homologue of hIL-8, but
number of CD4 T cells in the nasal tract as well as mouse granulocyte chemotactic protein-2 (GCP-2) is 61%
cervical lymph nodes, and an increase in the number identical to hGCP-2  and is a functional murine homologue
of neutrophils in the nasal tract 12 h after GCP-2 for hIL-8, with 32% identity [11–13].
immunization. This chemokine also modulated CD28 GCP-2 acts as a potent chemoattractant for neutrophils in
expression by CD4 T cells during CD3 stimulation the course of acute inﬂammation. Endothelial and bronchial
of wild-type mice. GCP-2 increased CD80 and CD86 epithelial cells produce GCP-2 after lipopolysaccharide (LPS)
expression on B cells during in vitro stimulation in a and/or IL-1 exposure [14 –16]. Moreover, GCP-2 is produced
dose-dependent manner. In contrast, cytokine and during peritonsillar abscess . Similarly, GCP-2 is ex-
costimulatory molecule enhancement by GCP-2 was pressed in the synovial tissue during rheumatoid arthritis .
not induced by lymphocytes from IL-8R / mice, The recurring expression of GCP-2 coincides with the relaps-
suggesting that GCP-2 modulates cellular immunity ing nature of these inﬂammatory diseases, which are also T
in part through IL-8R interactions. J. Leukoc. Biol.
cell-dependent. Hence, we have used a mouse model, which
76: 1240 –1247; 2004.
mucosally administers GCP-2 (multiple times) along with a T
cell-dependent antigen, to study this chemokine’s role in mu-
Key Words: adjuvant Th1/Th2 B7
cosal adaptive immune and inﬂammatory responses.
The multiple biological activities related to the immuno-
pathogenesis of GCP-2 are poorly understood. Effective leuko-
The mucosa serves as the ﬁrst innate defense against mucosal 1
pathogens and is a source of numerous effector molecules, Correspondence: Morehouse School of Medicine, Department of Microbi-
ology, Biochemistry, and Immunology, 720 Westview Drive, Atlanta, GA
namely, defensins, chemokines, and cytokines, which initiate 30310-1495. E-mail: Lillard@msm.edu.
mucosal adaptive immune responses [1– 6]. Numerous studies Received September 26, 2003; revised August 4, 2004; accepted August 17,
have demonstrated the ability of chemokines to regulate the 2004; doi: 10.1189/jlb.0903444.
1240 Journal of Leukocyte Biology Volume 76, December 2004 http://www.jleukbio.org
cyte inﬁltration along with activation can initiate immuno- Chemical Co.), and 10% of fetal calf serum (FCS; Atlanta Biologicals,
stimulating cascades that help to bridge innate and adaptive Norcross, GA).
host responses. The present study has determined some of the
Cytokine and OVA-speciﬁc Ab detection by
cellular and molecular mechanisms that GCP-2 uses to mod-
ulate adaptive immunity.
For the assessment of cytokine production by the spleen, lungs, nasal tract, and
CLNs, culture supernatants were harvested after 3 days of ex vivo restimula-
tion. The presence of T helper cytokines, IL-2, IL-4, IL-5, IL-6, IL-10, IFN- ,
MATERIALS AND METHODS and tumor necrosis factor (TNF- ), in cell culture supernatants was deter-
mined by ELISA following the manufacturer’s instructions (E-Biosciences, San
Immunogens Diego, CA). Fecal and serum sample levels of OVA-speciﬁc Ab were measured
by ELISA, as previously described . Brieﬂy, 96-well Falcon 3912 ﬂexible
GCP-2/LPS-induced CXC chemokine  was purchased from PeproTech ELISA plates (Fisher Scientiﬁc, Pittsburgh, PA) were coated with 100 l 1
(Rocky Hill, NJ). The potential level of endotoxin contamination was quantiﬁed mg/ml OVA in PBS overnight (O/N) at 4°C and blocked with 10% FCS (Atlanta
by the chromogenic Limulus amebocyte lysate assay (Cape Cod Inc., East Biologicals) in PBS (B-PBS) for 3 h at room temperature. Individual samples
Falmouth, MA) to be 5 endotoxin units/mg. Chicken egg albumin (OVA) and (100 l) were added and serially diluted in B-PBS. After O/N incubation at
bovine serum albumin (BSA) were purchased from Sigma Chemical Co. (St. 4°C and three washes using PBS containing 0.05% Tween 20 (PBS-T), titers
Louis, MO). of IgM, IgG, IgA, or IgG subclasses were determined by the addition of 100 l
biotinylated detection Ab (BD PharMingen, San Diego, CA). After incubation
Mice and immunizations and wash steps, 100 l 1:3000 dilution of antibiotin horseradish peroxidase Ab
(Vector Laboratories, Burlingame, CA) in B-PBS-T was added to IgG subclass
Female BALB/c and IL-8R / on BALB/c background mice, aged 6 – 8 detection wells and incubated for 1 h at room temperature. Following incuba-
weeks, were purchased from Jackson Laboratories (Bar Harbor, ME). All mice tion, all plates were washed six times, and the color reaction was developed by
were housed in horizontal laminar ﬂow barrier cabinets free of microbial adding 100 l 1.1 mM 2,2 -azino-bis(3)-ethylbenzthiazoline-6-sulfonic acid
pathogens. Routine antibody (Ab) screenings for a large panel of pathogens and (ABTS; Sigma Chemical Co.) in 0.1 M citrate-phosphate buffer (pH 4.2)
histological analyses of organs and tissues were performed to ensure that mice containing 0.01% H2O2 (ABTS solution). The plates were read at 415 nm after
were pathogen-free. Following anesthesia, mice were nasally immunized on 10 min.
days 0, 7, and 14 with 75 g OVA alone or OVA plus 1 g GCP-2 in 15 l
phosphate-buffered saline (PBS; 7.5 l per nare). Experimental groups con- T cell proliferation assay
sisted of ﬁve mice, and studies were repeated three times. The guidelines
proposed by the Committee for the Care of Laboratory Animal Resources Antigen-speciﬁc lymphocyte proliferation was measured by a 5-bromo-2 -
Commission of Life Sciences, National Research Council, were followed to deoxy uridine (BrdU) absorption-detection kit, according to the manufacturer’s
minimize animal pain and distress. ¨
instructions (Roche Diagnostics, Dusseldorf, Germany). Subsequently, BrdU
incorporation was detected using a scanning multiwell spectrophotometer
Sample and tissue collection (SpectraMax 250 ELISA reader, Molecular Devices, Sunnyvale, CA). In brief,
after 2 days of culture with OVA (1 mg/ml), CD4 T cells at the density of 5
Fecal samples were weighed and dissolved in 1 ml PBS containing 0.1% 106 cells/ml with 106 cells/ml -irradiated feeder splenocytes were transferred
sodium azide per 100 mg fecal pellet. Following suspension by vortexing for 10 to polystyrene 96-well plates (Corning Glass Work, Midland, MI). BrdU
min, fecal samples were centrifuged, and supernatants were collected for labeling solution (10 l; 10 M ﬁnal concentration per well) was added, and
analysis. Blood samples were collected by supra-orbital capillary puncture, cells were incubated for 18 h at 37°C with 5% CO2. The cells were then ﬁxed
and serum was obtained following centrifugation. Serum and mucosal secre- and incubated with 100 l nuclease in each well for 30 min at 37°C. The cells
tions were collected 1 week after the last immunization and analyzed for were washed with complete media and again incubated with BrdU-peroxidase
OVA-speciﬁc Ab responses by enzyme-linked immunosorbent assay (ELISA). solution for 30 min at 37°C. The incorporation was developed with a ABTS
Mice were killed by CO2 inhalation 1 week after the last immunization to solution, and the change in optical density was read at 450 nm (OD450).
quantify the OVA-speciﬁc CD4 T cell responses present in immune com-
partments. Flow cytometry analysis of costimulatory
Lymphocytes were isolated from the spleens of normal and IL-8R / mice
After nasal immunization with PBS and/or 75 g OVA alone or OVA plus 1 g and added at a density of 106 cells/ml in complete medium containing 0, 1, 10,
GCP-2, leukocytes were obtained from single-cell suspensions of spleen, lung, 100, or 1000 ng/ml GCP-2. Anti-CD3ε Ab-coated plates were used to activate
cervical lymph node (CLN), and nasal tract [3–5]. To isolate lower respiratory primary CD4 T cells from normal or IL-8R / mice. After incubation for 3
tract lymphocytes, lungs were injected with 10 ml cold PBS to remove blood, days, the cells were stained with phycoerythrin (PE)-conjugated rat anti-mouse
dissected into small pieces, and digested in collagenase type IV (Sigma CD28, CD80, or CD86 plus ﬂuorescein isothiocyanate (FITC)-conjugated rat
Chemical Co.) in RPMI 1640 (collagenase solution) for 45 min with stirring at anti-mouse CD4 or B220 monoclonal Ab (mAb; BD PharMingen) for 30 min
37°C [2–5]. Nasal tract lymphocytes were isolated by gently washing nasal with shaking. Lymphocytes were then washed with ﬂuorescein-activated cell
cavities with 200 l cold PBS to remove blood. Next, the nasal tract mucosal sorter (FACS) buffer (PBS with 1% BSA), ﬁxed in 2% paraformaldehyde in
tissue was removed by scrapping. Cell suspensions were washed twice in RPMI PBS, and analyzed by ﬂow cytometry (Becton Dickinson, San Diego, CA). The
1640. Lung and nasal tract lymphocytes were further puriﬁed using a discon- percent increase (or decrease) of the costimulatory molecule expression by
tinuous Percoll (Pharmacia, Uppsala, Sweden) gradient, collecting at the resting or CD3ε-activated splenocytes from normal and IL-8R / mice in
40 –75% interface. cultures with supernatant containing GCP-2 was calculated as CD28 or CD80
T cell fractions were obtained by passing single-cell suspensions over nylon and CD86 expression on CD3 CD4 or CD3– B220 cells, respectively,
wool for 1 h at 37°C ( 98% purity). Subsequently, CD4 T cells were cultured with GCP-2 ligands minus the percent gated of double-positive cells
enriched ( 98% purity) using Mouse CD4 Cellect plus columns, according to in cultures without GCP-2, divided by the latter.
the manufacturer’s protocols (Biotex Laboratories, Edmonton, Alberta, Can-
ada). Lymphocytes were maintained in complete medium, which consisted of Flow cytometry analysis of leukocyte migration
RPMI 1640 supplemented with 10 ml/L nonessential amino acids (Mediatech,
Washington, DC), 1 mM sodium pyruvate (Sigma Chemical Co.), 10 mM Mice were immunized with OVA alone or OVA plus GCP-2, as before. After
HEPES (Mediatech), 100 U/ml penicillin, 100 g/ml streptomycin, 40 g/ml 12 h, leukocytes from the spleen or mucosal tissues were stained with CY5-
gentamycin (Elkins-Sinn, Cherry Hill, NJ), 50 M mercaptoethanol (Sigma conjugated CD4, FITC-conjugated CD8, and/or B220 mAb along with PE-
Singh et al. GCP-2-mediated immunity 1241
conjugated CD11b, CD11c, NK1.1, and LY6.G (BD PharMingen) for 30 min
with occasional shaking. Labeled cells were washed with FACS staining buffer
(PBS with 1% BSA), ﬁxed in 2% paraformaldehyde in PBS, and 104 cells were
analyzed using a FACScan™ ﬂow cytometer and Cellquest™ software (BD
The data are expressed as the mean SEM and compared using a two-tailed
Student’s t-test or an unpaired Mann Whitney U-test. The results were ana-
lyzed using the Microsoft Excel program (Seattle, WA) and were considered
statistically signiﬁcant if P values were 0.05. When cytokine levels were
below the detection limit (BD), they were recorded as one-half the lower
detection limit (e.g., 10 pg/ml for IL-10) for statistical analysis.
GCP-2 stimulates OVA-speciﬁc, systemic Ab
The optimal doses required for chemokines to induce chemo-
taxis have been well-documented [20 –24]; however, previous
studies from our laboratory yield consistent and signiﬁcant
increases when 1 g chemokine(s) is given as mucosal adju- Fig. 2. OVA-speciﬁc serum and fecal Ab responses following nasal immu-
vants [4 – 6]. To determine the optimal dose of GCP-2 as nization. Groups of ﬁve BALB/c mice were nasally immunized on days 0, 7,
and 14 with 75 g OVA alone or 1.0 g GCP-2 in 15 l PBS. (A and B) Ig
adjuvant, which would affect antigen-speciﬁc serum Ab re-
isotype and IgG subclass Ab titers, respectively, in the serum. (C) IgA and IgG
sponses, mice were nasally administered (three times at weekly Ab titers in fecal secretions. OVA-speciﬁc serum and fecal Ab titers on day 21
intervals) with 75 g OVA alone or in the presence of increas- were determined by ELISA, and data presented are the mean Ab titers SEM
ing concentrations of GCP-2 (e.g., 0.0, 0.5, 1.0, 2.5, and 5.0 or those BD of four separate experiments. ( ) Statistically signiﬁcant differ-
g). Accordingly, we analyzed OVA-speciﬁc IgA, IgG, and ence (i.e., P 0.05) from Ab titers of mice immunized with OVA alone.
IgM Ab isotypes in sera. Signiﬁcant titers of OVA-speciﬁc Ab
responses were elicited when mice received 1 g GCP-2 as
adjuvant (Fig. 1). Although higher doses of GCP-2 also en- Therefore, subsequent studies used 75 g OVA plus 1 g
hanced humoral responses, there was no signiﬁcant increase in GCP-2 as the immunization regimen.
host responses when 1 g GCP-2 was used as adjuvant. Mice nasally immunized three times with 75 g OVA plus
1 g GCP-2 displayed signiﬁcant increases in antigen-speciﬁc
serum IgG responses when compared with mice receiving OVA
alone (Figs. 1 and 2A). The humoral adjuvant activity of
GCP-2 induced a signiﬁcant increase in serum IgG1 responses
followed by IgG2b responses (Fig. 2B). We next asked whether
the adjuvant activity of nasally coadministered GCP-2 could
promote mucosal secretory IgA (S-IgA) Ab responses. Analysis
of OVA-speciﬁc S-IgA responses in mucosal secretions re-
vealed signiﬁcant S-IgA Ab titers in fecal extracts (Fig. 2C).
Proliferation and cytokine responses induced by
As GCP-2 enhanced mucosal and systemic Ab responses, we
next examined the pattern of the T helper cytokine responses it
promoted. CD4 T cells isolated from the spleen, nasal tract,
CLNs, or lungs of mice immunized with OVA plus GCP-2
exhibited marked increases in OVA-speciﬁc, proliferative re-
sponses as compared with CD4 T cells from mice immunized
Fig. 1. Dose response of OVA-speciﬁc serum Ab responses following immu- with OVA alone (Fig. 3). CD4 T cells from the spleen, nasal
nization using GCP-2 as adjuvant. Groups of ﬁve BALB/c mice were nasally tract, CLNs, and lungs of mice immunized with OVA plus
immunized on days 0, 7, and 14 with 75 g OVA alone or 0.5, 1.0, 2.5, and GCP-2 also showed signiﬁcant increases in IL-2, TNF- , and
5.0 g GCP-2 in 15 l PBS. OVA-speciﬁc IgA (●), IgG ( ), and IgM (Œ) Ab IFN- secretions by OVA-restimulated T cells compared with
titers in the serum on day 21 were determined by ELISA, and data presented
are the mean Ab titers SEM of four separate experiments. ( ) Statistically
controls (Fig. 3). As an adjuvant, GCP-2 also increased Th2
signiﬁcant differences (i.e., P 0.05) from OVA-speciﬁc IgG Ab titers of mice responses; most notably, IL-5 and IL-6 secretions by ex vivo-
immunized with OVA alone. restimulated T lymphocytes were dramatically elevated in im-
1242 Journal of Leukocyte Biology Volume 76, December 2004 http://www.jleukbio.org
Fig. 3. Proliferation and T helper cell type 1 (Th1)/
Th2-type cytokine secretion by OVA-restimulated
CD4 T cells from previously immunized mice. Groups
of ﬁve BALB/c mice were nasally immunized on days 0,
7, and 14 with 75 g OVA alone or with 1.0 g GCP-2
in 15 l PBS. One week after the last immunization with
OVA alone (open bar) or OVA plus GCP-2 (solid bar),
lung-, spleen-, nasal tract-, and CLN-derived CD4 T
cells were puriﬁed and cultured at a density of 5 106
cells/ml with 1 mg/ml OVA for 3 days. Cytokine ELISA
was determined in culture supernatant productions. Pro-
liferation was measured by BrdU incorporation. The
data presented are the mean – OD450 for proliferative
responses or IL-2, TNF- , IL-4, IL-5, IL-6, IL-10, and
IFN- secretion (pg/ml) SEM of quadruplicate cul-
tures. ( ) Statistically signiﬁcant difference (P 0.05)
between OVA alone and OVA plus GCP-2-immunized
munized mice (Fig. 3). GCP-2 promoted relatively low IL-4 or also modestly increased the expression of CD80 and CD86 by
IL-10 responses. Taken together, these results show that B220 B cells from normal mice but not from IL-8R / mice
GCP-2 enhanced IL-2, IL-5, and IL-6 as well as TNF- and costimulated by anti-CD3ε, mAb-treated T cells in culture. As
IFN- T cell responses. described previously by others [20 –24] and our laboratory
[4 – 6], chemokines optimally induce chemotaxis at 10 ng/ml
Primary Th cell responses induced by GCP-2 and affect costimulatory molecule expression and leukocyte
When GCP-2 was used as an adjuvant, it up-regulated sys- activation at 10 –50 ng/ml. In conﬁrmation, we also show that
temic and mucosal Ab responses as well as the responses of GCP-2 behaves in a similar manner to increase CD28 and B7
CD4 T cells from the spleen, nasal tract, CLNs, and lungs. expression in part through IL-8R interactions.
We next examined whether these effects were mediated
GCP-2-mediated in vivo migration of leukocyte
through IL-8R interactions. CD3ε stimulation was required to
increase Th1 and Th2 cytokine secretion by wild-type or
IL-8R / T cells. IL-2 secretion patterns of CD3ε-stimulated To further establish the effect of GCP-2 on the modulation of
CD4 T cells from normal and IL-8R / mice were not cellular and humoral immunity, mice were nasally immunized
affected by GCP-2, which increased TNF- and IFN- primary as before with OVA plus GCP-2 or with PBS and/or with OVA
Th1 responses after CD3ε stimulation in normal mice but not alone. Nasal immunization (with OVA alone) did not signiﬁ-
by IL-8R / CD4 T cells, treated in a similar manner (Fig. cantly alter the number of leukocytes in the spleen or lungs
4). GCP-2 leads to a robust increase in the secretion of IL-5 (Table 1). When compared with OVA alone, GCP-2 plus OVA
and IL-6 cytokines by wild-type, primary, CD3ε-stimulated T signiﬁcantly decreased the number of CD4 T cells in the
cells but not by similarly treated IL-8R / T cells. It is nasal tract and CLNs but caused a modest increase in the
interesting that GCP-2 induced IL-4 and IL-10 secretion by number of B cells in the CLNs and a signiﬁcant increase in the
CD4 T cells from IL-8R / mice after CD3 stimulation number of nasal tract B220 B cells. A modest yet statistically
when compared with resting CD4 T cells or CD3ε-stimulated, signiﬁcant increase in the number of CD11c dendritic cells in
wild-type Th cells. Taken together, these results show that CLNs was observed 12 h after OVA plus GCP-2 immunization.
GCP-2 increases IL-4 and IL-10 secretion patterns by CD3ε- We also noted changes in the number of LY-6G neutro-
stimulated, IL-8R / naıve CD4 T cells but enhances IL-5,
¨ phils after nasal administration of OVA alone when compared
IL-6, TNF- , and IFN- production by CD3ε-stimulated, wild- with OVA plus GCP-2; a signiﬁcant increase in neutrophils
type, primary CD4 T cells. was noticed in the nasal tract 12 h after immunization. Taken
together, these data suggest that GCP-2 ( OVA) mediates
GCP-2 modulates CD28 and B7 expression leukocyte recruitment to and from the nasal tract and CLNs,
Earlier studies have shown that chemokines can differentially 12 h after nasal immunization when compared with groups
modulate the expression of costimulatory molecules by lym- given OVA alone.
phocytes [4 – 6, 20 –24]. To better elucidate the effects of
GCP-2 on adaptive immune responses, we assessed its poten-
tial to modulate the expression of costimulatory molecules by DISCUSSION
lymphocytes from normal and IL-8R / mice. GCP-2 had
minimal or no effects on resting wild-type or IL-8R / lym- GCP-2 acts as a functional murine homologue of hIL-8 and
phocytes (Fig. 4 and data not shown). However, GCP-2 signif- GCP-2 [11–13], and its ability to interact with CXCR1 and
icantly increased the expression of CD28 by CD3ε-stimulated CXCR2 provided the rationale(s) to test our hypothesis that
lymphocytes from wild-type mice but not from IL-8R / mice GCP-2 enhances adaptive immunity. Previous studies from our
(Fig. 5). Similar to the induction of CD28 expression, GCP-2 laboratory showed that RANTES, when used as an adjuvant,
Singh et al. GCP-2-mediated immunity 1243
Fig. 4. Cytokine secretion by CD3ε-stimulated, naıve T cells from IL-8R / and normal mice. CD4 T cells from IL-8R / (open and checkered bars) and
normal mice (solid and dotted bars) were cultured at a density of 5 106 cells/ml with 0, 1, 10, 100, or 1000 ng/ml GCP-2 on uncoated (checkered and dotted
bars) or anti-mouse CD3ε Ab-coated plates (solid and open bars). IL-2, TNF- , and IFN- as well as IL-4, IL-5, IL-6, and IL-10 production was determined by
ELISA of cultured supernatants. (✰) Statistically signiﬁcant differences (P 0.05) between cultured, resting, naıve lymphocytes from IL-8R / or normal mice.
( ) Statistically signiﬁcant differences (P 0.05) between cultured, CD3ε-stimulated, naıve lymphocytes from IL-8 R / or normal mice.
induces predominantly Th1-driven, antigen-speciﬁc IgG2a, immunized with OVA alone. IFN- production is often asso-
followed by IgG2b, IgG3, and IgG1 Ab , and that MIP-1 ciated with IgG2a and IgG3 Ab production . The low doses
functions as a Th1 inducer to propagate cytotoxic T cell re- of IFN- (1500 units) have been shown to increase IgG2a
sponses and serum IgG2a and mucosal IgA Ab responses . production in vivo, and considerably higher doses of IFN-
Lymphotactin also acts as an innate mucosal adjuvant to in- (12,500 units) are required to induce decreases in IgG1 and
duce Th2 Th1 responses and dramatically increased serum IgE responses . Taken together, the analysis of the OVA-
IgG subclasses with robust mucosal IgA Ab responses . In speciﬁc, humoral responses was supported by mixed Th2/Th1
the present study, we have shown that GCP-2 fosters Th2 and cytokine help.
Th1 responses as well as CD28-B7 expression, in part through The precise cytokine signals required for S-IgA production
IL-8R ligation. and for mucosal immunity in general are not completely un-
The increases in IgG1 and the subsequent expression of derstood. It has been shown that mucosal IgA responses re-
IgG2b and IgG2a OVA-speciﬁc Ab titers were most likely a quire Th2-type, cell-derived cytokines (e.g., IL-5, IL-6, and
result of the mixed Th2/Th1 cytokine help provided by CD4 IL-10) . Studies have supported that Th1- and Th2-type,
T cells as well as by use of a soluble protein antigen [25, 26]. cell-derived cytokines are important for S-IgA responses [34 –
In this regard, Th2 cytokines support IgG1 [26 –28] and IgG2b 36]. We have previously shown that lymphotactin, MIP-1 ,
 Ab generation, and the levels of anti-OVA IgG1 and MIP-1 , and RANTES induce mucosal IgA [3–5]. The level of
IgG2b Ab induced by GCP-2 plus OVA were consistent with the particular cytokines(s) required for B cells to express the
the cytokine secretion patterns of IL-5 and IL-6 expression by IgA isotype was also provided in our experimental model.
CD4 T cells from mice immunized with GCP-2 as compared Although it has been reported that IL-4, IL-5, and IL-6 do not
with mice immunized with OVA alone. Indeed, IL-5 has been induce IgA switching, IL-5 and IL-6 induce surface IgA B
shown to increase the Ig secretion of IgA-, IgG1-, and IgE- cells to secrete IgA . The cytokine produced by CD4 T
committed B cells [30, 31]. cells in systemic and mucosal compartments after GCP-2 im-
GCP-2, as an adjuvant, also increased antigen-speciﬁc IL-2, munization explains why an increase in OVA-speciﬁc IgA
TNF- , and IFN- CD4 T cell responses compared with mice occurred in mucosal secretions. The heightened mucosal Ab
1244 Journal of Leukocyte Biology Volume 76, December 2004 http://www.jleukbio.org
responses generated by GCP-2 also correlated with the Th2
and Th1 responses displayed by OVA-restimulated CD4 T
cells from immunized mice.
The end result of the immunization strategy we used was
increased humoral and cell-mediated, adaptive immune re-
sponses. However, the mechanism of GCP-2 adjuvantcy re-
mained uncertain. We have previously shown that chemokines
can modulate cytokine and costimulatory molecule expression
by activated lymphocytes [3–5]. Now, we show that GCP-2
modulates cytokine and costimulatory molecule expression by
activated T cells (Figs. 4 and 5). However, cytokine secretion
alone does not completely explain the adjuvant effects of
CD28 is equally expressed by CD4 and CD8 T cells,
which cooperatively regulate T cell activation through B7 and
T cell receptor stimulation . CD28 supplies a coactivation
signal for T cell activation [38, 39] and is required for mucosal
and T cell-mediated immunity [40, 41]. We have previously
shown that RANTES and MIP-1 act as mucosal adjuvants,
partly through CD28 up-regulation [4, 5]. Similarly, CXCR3
ligands may enhance adaptive immune responses through
CD28 modulation . GCP-2 signiﬁcantly increased CD28
expression by CD3ε-stimulated wild-type CD4 T cells in a
Fig. 5. Modulation of CD28 CD80, and CD86 expression by GCP-2. CD3ε- dose-dependent manner but not by IL-8R / Th cells treated
stimulated, naıve lymphocytes from normal (●) or IL-8R / ( ) mice were
¨ in a similar manner.
incubated with 0, 1, 10, 100, and 1000 ng/ml GCP-2 in 96-well culture plates.
The percent increase (or decrease) in the expression of the costimulatory
To address another potential mechanism of adjuvant activ-
molecules by normal or IL-8R / lymphocytes was calculated as the percent ity, we investigated how GCP-2 affects the expression of B7
of double-positive (CD4 CD28 , B220 CD80 , or B220 CD86 ) cells in molecules on B cells, which also express IL-8Rs . Previous
cultures containing GCP-2 minus the percent gated of double-positive cells in studies ﬁrst showed that the CD28 binds B7-1, B7-2, and
cultures without GCP-2, divided by the latter. Studies were repeated four B7-H1 on antigen-presenting cells [43, 44]. It has been re-
times, and the data presented are the mean percent change SEM of these
ported that the mucosal adjuvanticity of CT involves the se-
lective up-regulation of CD86 expression  and that MIP-1
signiﬁcantly up-regulates the expression of CD80 , like
RANTES , but also increases the CD86 surface level on
TABLE 1. GCP-2 Effects on In Vivo Leukocyte Migration
Cells per 104 of gated leukocytes from the:
subpopulation Additions Spleen Nasal tract Cervical lymph nodes Lungs
CD3 CD4 PBS alone 2170 325 470 70 2660 452 550 81
OVA alone 2271 340 380 57 2060 309 440 66
GCP-2 OVA 2380 355 245 43( )
1440 231( )
CD3 CD8 PBS alone 490 73 80 13 790 118 90 14
OVA alone 540 81 90 10 530 79 70 10
GCP-2 OVA 570 85 50 7 418 62 110 16
B220 PBS alone 5550 832 4110 751 4020 603 4160 624
OVA alone 5620 843 4012 769 4580 687 4190 628
GCP-2 OVA 5840 876 5740 831 4710 667 4450 831
CD11b PBS alone 60 9 70 11 40 6 30 5
OVA alone 90 13 60 9 30 5 50 8
GCP-2 OVA 101 8 80 9 50 4 60 9
CD11c PBS alone 140 21 40 6 29 4 90 13
OVA alone 150 22 50 7 30 4 110 16
GCP-2 OVA 120 18 60 9 88 9 120 18
LY-6G PBS alone 750 111 98 15 290 43 839 125
OVA alone 860 129 108 16 210 31 503 75
GCP-2 OVA 910 136 401 59 310 46 500 75
Wild-type BALB/c mice were nasally administered PBS, OVA alone, or OVA along with GCP-2 in a volume of 7.5 l PBS. Spleen, nasal tract, CLN, and lung
lymphocytes were puriﬁed and prepared for FACS analysis 12 h after administration. There were no statistically signiﬁcant differences between PBS and OVA in
PBS-treated groups. Differences between OVA alone and OVA GCP-2-immunized groups were considered signiﬁcant when P 0.05 ( ).
Singh et al. GCP-2-mediated immunity 1245
resting B cells. Similarly, our data suggest that GCP-2 mod- 3. Lillard Jr., J. W., Boyaka, P. N., Hedrick, J. A., Zlotnik, A., McGhee, J. R.
(1999) Lymphotactin acts as an innate mucosal adjuvant. J. Immunol.
estly enhances B7-1 and B7-2 surface expression on B cells in 162, 1959 –1965.
a dose- and IL-8R -dependent manner to presumably support 4. Lillard Jr., J. W., Boyaka, P. N., Taub, D. D., McGhee, J. R. (2001)
primary adaptive immune responses. Hence, local GCP-2 ex- RANTES potentiates antigen-speciﬁc mucosal immune responses. J. Im-
pression during the innate host response could act on adjacent munol. 166, 162–169.
5. Lillard Jr., J. W., Singh, U. P., Boyaka, P. N., Singh, S., Taub, D. D.,
or nearby lymphocytes to enhance the expression of IL-5, IL-6, McGhee, J. R. (2003) MIP-1 and MIP-1 differentially mediate mucosal
TNF- , and IFN- after initial antigen stimulation. Together, and systemic adaptive immunity. Blood 101, 807– 814.
this would help initiate and direct adaptive immune response. 6. Singh, U. P., Singh, S., Weaver, C. T., Iqbal, N., McGhee, J. R., Lillard Jr.,
J. W. (2003) IFN- -inducible chemokines enhance adaptive immunity
It is generally accepted that neutrophils are among the ﬁrst and volitis. J. Interferon Cytokine Res. 23, 591– 600.
leukocytes recruited to sites of injury or infection. Indeed, 7. Hoch, R. C., Schraufstatter, I. U., Cochrane, C. G. (1996) In vivo, in vitro,
increases in IL-8R ligands are an indication of neutrophil and molecular aspects of interleukin-8 and the interleukin-8 receptors.
J. Lab. Clin. Med. 128, 134 –145.
recruitment to inﬂammatory loci. In the present study, we 8. Mitsuyama, K., Toyonaga, A., Sasaki, E., Watanabe, K., Tateishi, H.,
have shown that GCP-2 provides signals to bridge innate Nishiyama, T., Saiki, T., Ikeda, H., Tsuruta, O., Tanikawa, K. (1994) IL-8
and adaptive immunity. Potentially, these characteristics as an important chemoattractant for neutrophils in ulcerative colitis and
Crohn’s disease. Clin. Exp. Immunol. 96, 432– 436.
would allow GCP-2-activated leukocytes to contain and 9. Raab, Y., Gerdin, B., Ahlstedt, S., Hallgren, R. (1993) Neutrophil mucosal
engulf mucosal pathogens and subsequently, present foreign involvement is accompanied by enhanced local production of interleu-
antigens to naıve lymphocytes to support protective and kin-8 in ulcerative colitis. Gut 34, 1203–1206.
adaptive immune responses. Correspondingly, these mech- 10. Van Damme, J., Wuyts, A., Froyen, G., Van Coillie, E., Struyf, S., Billiau,
A., Proost, P., Wang, J. M., Opdenakker, G. (1997) Granulocyte chemo-
anisms would also permit GCP-2 to markedly enhance an- tactic protein-2 and related CXC chemokines: from gene regulation to
tigen-speciﬁc host responses to microbes that enter the receptor usage. J. Leukoc. Biol. 62, 563–569.
sterile, peripheral environment. 11. Proost, P., De Wolf-Peeters, C., Conings, R., Opdenakker, G., Billiau, A.,
Van Damme, J. (1993) Identiﬁcation of a novel granulocyte chemotactic
GCP-2 may use several mechanisms to facilitate the induc- protein (GCP-2) from human tumor cells. In vitro and in vivo comparison
tion of adaptive mucosal immune responses. Our results show with natural forms of GRO, IP-10, and IL-8. J. Immunol. 150, 1000 –
that GCP-2 enhances the recognition phase of the adaptive host 1010.
12. Froyen, G., Proost, P., Ronsse, I., Mitera, T., Haelens, A., Wuyts, A.,
response by modestly increasing the number of B cells, de- Opdenakker, G., Van Damme, J., Billiau, A. (1997) Cloning, bacterial
creasing the number of CD4 T cells from CLNs and the nasal expression and biological characterization of recombinant human granu-
tract, and most importantly, increasing neutrophils in the nasal locyte chemotactic protein-2 and differential expression of granulocyte
chemotactic protein-2 and epithelial cell-derived neutrophil activating
tract 12 h after nasal immunization. The variances observed in
peptide-78 mRNAs. Eur. J. Biochem. 243, 762–769.
humoral and cellular immunity induced by GCP-2 coincide 13. Wuyts, A., Van Osselaer, N., Haelens, A., Samson, I., Herdewijn, P.,
with its ability to modulate leukocyte, especially neutrophil, Ben-Baruch, A., Oppenheim, J. J., Proost, P., Van Damme, J. (1997)
migration to and from mucosal effector sites. This innate ability Characterization of synthetic human granulocyte chemotactic protein 2:
usage of chemokine receptors CXCR1 and CXCR2 and in vivo inﬂamma-
of GCP-2 to induce chemotactic and costimulatory signals has tory properties. Biochemistry 36, 2716 –2723.
implications in mechanisms of immunity and the priming and 14. Mine, S., Nasu, K., Fukuda, J., Sun, B., Miyakawa, I. (2003) Secretion of
maintenance of chronic inﬂammatory responses. IL-8R inter- granulocyte chemotactic protein-2 by cultured human endometrial stromal
cells. Fertil. Steril. 79, 146 –150.
nalization may represent another mechanism that regulates the 15. Prause, O., Laan, M., Lotvall, J., Linden, A. (2003) Pharmacological
migration and retention of neutrophils to inﬂammatory sites as modulation of interleukin-17-induced GCP-2-, GRO- - and interleukin-8
well as controlling the immune or inﬂammation-enhancing release in human bronchial epithelial cells. Eur. J. Pharmacol. 462,
function of GCP-2 . Our results suggest that GCP-2 dic- 16. Beck, G. C., Yard, B. A., Breedijk, A. J., Van Ackern, K., Van Der
tates immune responses on concomitant exposure with leuko- Woude, F. J. (1999) Release of CXC-chemokines by human lung micro-
cytes. Although further studies will be needed to elucidate the vascular endothelial cells (LMVEC) compared with macrovascular umbil-
ical vein endothelial cells. Clin. Exp. Immunol. 118, 298 –303.
precise cellular and molecular contributions that GCP-2 makes
17. Rudack, C., Jorg, S., Sachse, F. (2004) Biologically active neutrophil
toward host immune responses, our results have helped to chemokine pattern in tonsillitis. Clin. Exp. Immunol. 135, 511–518.
clarify some of the cellular and molecular mechanisms that this 18. Pierer, M., Rethage, J., Seibl, R., Lauener, R., Brentano, F., Wagner, U.,
chemokine uses to affect mucosal and systemic adaptive im- Hantzschel, H., Michel, B. A., Gay, R. E., Gay, S., Kybruz, D. (2004)
Chemokine secretion of rheumatoid arthritis synovial ﬁbroblasts stimu-
munity and chronic inﬂammation. lated by Toll-like receptor 2 ligands. J. Immunol. 172, 1256 –1265.
19. Smith, J. B., Herschman, H.R. (1995) Glucocorticoid-attenuated response
genes encode intercellular mediators, including a new C-X-C chemokine.
J. Biol. Chem. 270, 16756 –16765.
ACKNOWLEDGMENTS 20. Bacon, K. B., Premack, B. A., Gardner, P., Schall, T. J. (1995) Activation
of dual T cell signaling pathways by the chemokine RANTES. Science
U. P. S. and S. S. contributed equally to this manuscript. 21. Bacon, K. B., Flores-Romo, L., Life, P. F., Taub, D. D., Premack, B. A.,
Arkinstall, S. J., Wells, T. N., Schall, T. J., Power, C. A. (1995) IL-8-
induced signal transduction in T lymphocytes involves receptor-mediated
activation of phospholipases C and D. J. Immunol. 154, 3654 –3666.
REFERENCES 22. Xu, L., Kelvin, D. J., Ye, G. Q., Taub, D. D., Ben-Baruch, A., Oppenheim,
J. J., Wang, J. M. (1995) Modulation of IL-8 receptor expression on
1. Lillard, J. W., McGhee, J. R. (1997) Adjuvants or live delivery systems for puriﬁed human T lymphocytes is associated with changed chemotactic
the characterization of mucosal T helper subset responses. Res. Immunol. responses to IL-8. J. Leukoc. Biol. 57, 335–342.
148, 520 –527. 23. Taub, D. D., Turcovski-Corrales, S. M., Key, M. L., Longo, D. L., Murphy,
2. Lillard, J. W. J., Boyaka, P. N., Chertov, O., Oppenheim, J. J., McGhee, W. J. (1996) Chemokines and T lymphocyte activation: I. Chemokines
J. R. (1999) Mechanism for induction of acquired host immunity by costimulate human T lymphocyte activation in vitro. J. Immunol. 156,
neutrophil peptide defensins. Proc. Natl. Acad. Sci. USA 96, 651– 656. 2095–2103.
1246 Journal of Leukocyte Biology Volume 76, December 2004 http://www.jleukbio.org
24. Taub, D. D., Anver, M., Oppenheim, J. J., Longo, D. L., Murphy, W. J. 36. VanCott, J. L., Staats, H. F., Pascual, D. W., Roberts, M., Chatﬁeld, S. N.,
(1996) T lymphocyte recruitment by interleukin-8 (IL-8). IL-8-induced Yamamoto, M., Coste, M., Carter, P. B., Kiyono, H., McGhee, J. R. (1996)
degranulation of neutrophils releases potent chemoattractants for human T Regulation of mucosal and systemic antibody responses by T helper cell
lymphocytes both in vitro and in vivo. J. Clin. Invest. 97, 1931–1941. subsets, macrophages, and derived cytokines following oral immunization
25. Stevens, T. L., Bossie, A., Sanders, V. M., Fernandez-Botran, R., Coffman, with live recombinant Salmonella. J. Immunol. 156, 1504 –1514.
R. L., Mosmann, T. R., Vitetta, E. S. (1988) Regulation of antibody isotype 37. Azuma, M., Ito, D., Yagita, H., Okumura, K., Phillips, J. H., Lanier, L. L.,
secretion by subsets of antigen-speciﬁc helper T cells. Nature 334, Somoza, C. (1993) B7 antigen is a second ligand for CTLA-4 and CD28.
255–258. Nature 366, 76 –79.
26. Snapper, C. M., Finkelman, F. D., Paul, W. E. (1988) Differential regu- 38. Linsley, P. S., Clark, E. A., Ledbetter, J. A. (1990) T-cell antigen CD28
lation of IgG1 and IgE synthesis by interleukin 4. J. Exp. Med. 167, mediates adhesion with B cells by interacting with activation antigen
B7/BB-1. Proc. Natl. Acad. Sci. USA 87, 5031–5035.
27. Coffman, R. L., Seymour, B. W., Lebman, D. A., Hiraki, D. D., Chris-
39. Freedman, A. S., Freeman, G. J., Rhynhart, K., Nadler, L. M. (1991)
tiansen, J. A., Shrader, B., Cherwinski, H. M., Savelkoul, H. F., Finkel-
man, F. D., Bond, M. W., et al. (1988) The role of helper T cell products Selective induction of B7/BB-1 on interferon- -stimulated monocytes: a
in mouse B cell differentiation and isotype regulation. Immunol. Rev. potential mechanism for ampliﬁcation of T cell activation through the
102, 5–28. CD28 pathway. Cell. Immunol. 137, 429 – 437.
28. Finkelman, F. D., Katona, I. M., Urban Jr., J. F., Holmes, J., Ohara, J., 40. Linsley, P. S., Brady, W., Grosmaire, L., Aruffo, A., Damle, N. K.,
Tung, A. S., Sample, J. V., Paul, W. E. (1988) IL-4 is required to generate Ledbetter, J. A. (1991) Binding of the B cell activation antigen B7 to CD28
and sustain in vivo IgE responses. J. Immunol. 141, 2335–2341. costimulates T cell proliferation and interleukin 2 mRNA accumulation. J.
29. McGhee, J. R., Mestecky, J., Elson, C. O., Kiyono, H. (1989) Regulation Exp. Med. 173, 721–730.
of IgA synthesis and immune response by T cells and interleukins. J. Clin. 41. Gardby, E., Lane, P., Lycke, N. Y. (1998) Requirements for B7-CD28
Immunol. 9, 175–199. costimulation in mucosal IgA responses: paradoxes observed in CTLA4-H
30. Coffman, R. L., Seymour, B. W., Hudak, S., Jackson, J., Rennick, D. 1 transgenic mice. J. Immunol. 161, 49 –59.
(1989) Antibody to interleukin-5 inhibits helminth-induced eosinophilia 42. Tan, J. Q., Moller, B., Storgaard, M., Mukaida, N., Bonde, J., Grunnet, N.,
in mice. Science 245, 308 –310. Black, F. T., Larsen, C. G., Matsushima, K. K., Thestruppedersen, K., Xu,
31. McKenzie, D. T., Filutowicz, H. I., Swain, S. L., Dutton, R. W. (1987) L. L., Warren, M. K., Rose, W. L., Gong, W, Wang, J. M. (1997)
Puriﬁcation and partial sequence analysis of murine B cell growth factor Chemotaxis and IL-8 receptor expression in B cells from normal and
II (interleukin 5). J. Immunol. 139, 2661–2668. HIV-infected subjects. Human recombinant monocyte chemotactic protein
32. Coffman, R. L., Varkila, K., Scott, P., Chatelain, R. (1991) Role of and other C-C chemokines bind and induce directional migration of
cytokines in the differentiation of CD4 T-cell subsets in vivo. Immunol. dendritic cells in vitro. J. Immunol. 158, 475– 484.
Rev. 123, 189 –207. 43. June, C. H., Bluestone, J. A., Nadler, L. M., Thompson, C. B. (1994) The
33. Finkelman, F. D., Holmes, J., Katona, I. M., Urban Jr., J. F., Beckmann,
B7 and CD28 receptor families. Immunol. Today 15, 321–331.
M. P., Park, L. S., Schooley, K. A., Coffman, R. L., Mosmann, T. R., Paul,
44. Dong, H., Zhu, G., Tamada, K., Chen, L. (1999) B7–H1, a third member
W. E. (1990) Lymphokine control of in vivo immunoglobulin isotype
selection. Annu. Rev. Immunol. 8, 303–333. of the B7 family, co-stimulates T-cell proliferation and interleukin-10
34. Beagley, K. W., Eldridge, J. H., Lee, F., Kiyono, H., Everson, M. P., secretion. Nat. Med. 5, 1365–1369.
Koopman, W. J., Hirano, T., Kishimoto, T., McGhee, J. R. (1989) Inter- 45. Cong, Y. Z., Weaver, C. T., Elson, C. O. (1997) The mucosal adjuvanticity
leukins and IgA synthesis. Human and murine interleukin-6 induce high of cholera toxin involves enhancement of costimulatory activity by selec-
rate IgA secretion in IgA-committed B cells. J. Exp. Med. 169, 2133– tive up-regulation of B7.2 expression. J. Immunol. 159, 5301–5308.
2148. 46. Feniger-Barish, R., Belkin, D., Zaslaver, A., Gal, S., Dori, M., Ran, M.,
35. Beagley, K. W., Eldridge, J. H., Kiyono, H., Everson, M. P., Koopman, Ben-Baruch, A. (2000) GCP-2-induced internalization of IL-8 receptors:
W. J., Honjo, T., McGhee, J. R. (1988) Recombinant murine IL-5 induces hierarchical relationships between GCP-2 and other ELR( )-CXC che-
high rate IgA synthesis in cycling IgA-positive Peyer’s patch B cells. mokines and mechanisms regulating CXCR2 internalization and recy-
J. Immunol. 141, 2035–2042. cling. Blood 95, 1551–1559.
Singh et al. GCP-2-mediated immunity 1247