International Immunopharmacology 10 (2010) 611–618

Contents lists available at ScienceDirect

International Immunopharmacology

j ourna l homepage: www.e lsev ie r.com/ locate / in t imp

Administered CpG oligodeoxynucleotide induces mRNA expression of CXC and CCchemokines at the intestinal mucosa and PBMCs in piglets

Qing Cheng 1, Chenchao Xu 1, Linghua Zhang ⁎, Jiaoqing Li, Ting Cao, Meirong ZhangCollege of Life Sciences, South China Agricultural University, Guangzhou 510642, China

⁎ Corresponding author. Microbiological Staff Room,China Agricultural University, Wushan Road, Tianhe Di510642, China. Tel.: +8620 85281389; fax: +8620 852

E-mail address: lhzhang73212000@yahoo.com.cn (L1 These authors contributed equally to this work.

1567-5769/$ – see front matter © 2010 Elsevier B.V. Aldoi:10.1016/j.intimp.2010.02.013

a b s t r a c t

a r t i c l e i n f o

Article history:Received 16 September 2009Received in revised form 9 January 2010Accepted 22 February 2010

Keywords:PigletsCpG ODNChemokinesInnate immunity

Oligonucleotides containing CpG motifs (CpG ODN) are known to be potent stimulators of the innateimmune system in vitro and in vivo. We therefore investigated if intranasal (IN)–mucosal or intramuscular(IM)–systemic administration of CpG ODN could enhance innate immunity in the intestinal mucosa andperipheral blood mononuclear cells (PBMCs) in piglets. Repeated IN or IM administration of CpG ODNsignificantly increased local/systemic mRNA expression of the CC chemokines macrophage inflammatoryprotein 1β (MIP-1β) and monocyte chemoattractant protein-1 (MCP-1) and CXC chemokine gammainterferon-inducible protein 10 (IP-10) and percentages of macrophages and cDCs in the intestine (jejunum,caecum and colon) and PBMCs by different kinetics. IN delivery of CpG ODN induced much strongerchemokine responses than IM delivery at intestinal mucosas, whereas IN delivery of CpG ODN induced someweaker chemokine responses than IM delivery in PBMCs. These findings suggest that IN administration of100 μg/kg-CpG ODN without antigen codelivery may represent a valuable strategy for induction of innateimmunity against infection.

College of Life Sciences, Southstrict, Guangzhou, GuangDong82180.. Zhang).

l rights reserved.

© 2010 Elsevier B.V. All rights reserved.

1. Introduction

A recently discovered role of host cell receptors and the dangersignals (present in bacteria, viruses and fungi) in the activation of innateimmune responses has revolutionized the understanding of immunol-ogy. Toll like receptors (TLRs) are one of the cellular receptors whichinteract with different components of pathogens and activate innateimmune responses [15,22,23]. Of the known TLRs (10 members inhuman and 13 in mice), TLR9 recognizes unmethylated oligodinucleo-tide (CpG ODN) motifs present in bacterial and viral DNA and leads tothe activation of an array of innate immune responses characterized bythe induction of interferons andother cytokines [8,18,22]. In vitro and invivo studies have shown that CpGODN is potent activators of the innateimmune system in humans, non-human primates and other animals.

Many studies have shown that systemic administration of CpG ODNwithout antigen codelivery can induce nonspecific Th1-like protectiveinnate immune responses. Compared with humans and mice, relativelylimited work has been performed to characterize responses induced byCpGODNs in domestic animals. In pigs, early in vitro studies byKamstrupet al. [16] indicated that CpG ODN induced IL-6 and IL-12 mRNAexpression and proliferation in PBMCs [16]. More recently Dar et al. [9]reported an extensive study evaluating activity of the various classes of

CpG ODNs in PBMCs and also in lymph node cells (LNC) [9]. It wasrecently demonstrated in our laboratory that in vivo administration ofCpG ODN in piglets could enhance antigen specific immune responses asindicated by increased proliferation in PBMCs and increased IFN-γ level[30,31]. These investigations suggest that CpGODNshave potential in thecontrol or treatment of infectious diseases in swine.

Innate immunity against pathogens is in part orchestrated by theordered release of different chemokines that function as chemoattrac-tants and activators of various immune cells, a property that enablesimmune cells to serve as afirst line of cell-mediatedhost defense againstinfections. Since chemokines play an important role in innate immunity,of different chemokines studied to date, members of the CXC and CCchemokine subfamilies appear to have greatest effects on therecruitment of various immune cells, such as dendritic cells, macro-phages, polymorphonuclear cells and lymphocytes, to the site ofinfection. Thus, theCXC chemokinegamma interferon-inducibleprotein10 (IP-10), produced by a variety of cell types, including fibroblasts,endothelial cells, and mononuclear cells, through binding to its specificreceptor, CXCR3, chemoattracts dendritic cells, macrophages, andlymphocyte cells to the site of infection. The CC chemokines MIP-1βand MCP-1, both produced by leukocytes in response to proinflamma-tory cytokines or endotoxin, can induce the influx of macrophages, andimmature dendritic cells [20]. Therefore, in the present study CXCchemokine IP-10 and CC chemokines MCP-1 and MIP-1β were used asbiomarkers for activation of innate immunity, either systemically (sera)or locally (intestine tissues).

It is well known that IN delivery is a kind of mucosal admin-istered route, while IM delivery is a systemic administered route. And

612 Q. Cheng et al. / International Immunopharmacology 10 (2010) 611–618

administered route is important for the induction of effective local orsystemic immune responses. So, the present study was undertaken toinvestigate the effects of intranasal (IN)–mucosal or intramuscular(IM)–systemic administration of CpG ODN on induction of innateimmune responses in intestinal mucosas and PBMCs. We show herefor the first time that administration of CpG ODN in piglets rapidlyelicits mRNA expression of the CC chemokines MCP-1 and MIP-1βand the CXC chemokine IP-10 in the porcine intestinal mucosas andPBMCs.

2. Materials and methods

2.1. Reagents

CpG ODN was synthesized in the TaKaRa Biotech Co., a porcine-specific motif, which was used in the present study, was the equi-valent of the sequence D19 used by Kamstrup et al. [16], in whichthe phosphodiester nucleotides were shown in upper case andphosphorothioate nucleotides were shown in lower case. CpG ODN(5′-ggTGCATCGATTTATCGATGCAGggggg-3′) was resuspended inphosphate buffer saline (PBS) at a concentration of 2 mg/mL, andstored at−20 °C. Commercial kit— SV Total RNA Isolation Systemwaspurchased from MRC, Ltd., and RevertAid First Strand cDNA synthesiskit was purchased from MBI Fermentas. Real-time PCR commercialreagent kit was purchased from Applied Biosystems, Ficoll-Paque wasfrom Amersham Biotech, Uppsala, Sweden. Halothane was fromSigma, Ltd. Primary antibody for flow cytometry analysis anti-swineSWC3-biotin and anti-swine CD11R1 were respectively from South-ern Biotech and Serotec, Ltd. Secondary antibody streptavidin-PerCPand rat anti-mouse IgG1-APC were both from BD Biosciences, Ltd.

2.2. Animals and administration of CpG ODN

14-day-old Landrace×Yorkshire×Durok pigswere used in this study.All pigswere from the Swine Breeding Center of Guangdong LiZhu. Sevengroups of 14 piglets (average 7 kg per piglet) were divided according totheir dams.GroupAwas administered IMwithphosphate-buffered saline(PBS), groupBwasadministered IMwith10 μg/kgbodyweightCpGODN,group C was administered IM with 100 μg/kg body weight CpG ODN,group D was administered IM with 1000 μg/kg body weight CpG ODN,group Ewas administered INwith 10 μg/kg bodyweight CpGODN, groupF was administered IN with 100 μg/kg body weight CpG ODN, group Gwas administered IN with 1000 μg/kg body weight CpG ODN. 7 d later,piglets were re-administered with the same formulations. All formula-tions were delivered in a total volume of 1 mL. All piglets were kept withtheir dams, and the animals were allowed to suckle. The experimentalprotocol was approved by the Science and Technology Bureau ofGuangdong Province. All animals were manipulated according to theprocedures and consistent with the policies of the local animal carecommittee. The dosage of 10–1000 μg/kg body weight CpG ODNs wasused for IM or IN immunization since we have previously found this toyield effective immune responses in piglets in our experiments [32,33].

Table 1Sequences of primer pairs for the quantitative real-time sequence detection system used in

Target gene Primer probe Sequence (5′–3′)

IP-10 Forward primer CTGTTCGCTGTACCTGCATCAReverse primer TCAGACATCTTTTCTCCCCAT

MCP-1 Forward primer CGGCTGATGAGCTACAGAAGReverse primer TATGGAGTCCTGGACCCACT

MIP-1β Forward primer TCCCACCTCCTGCTGCTTCReverse primer GACCTGCCTGCCCTTTTTG

β-actin Forward primer ATCGTGCGGGACATCAAGGReverse primer GGCAGCTCGTAGCTCTTCTC

2.3. Sample collection

A complete blood count was performed regularly to confirm normalhematological status. Blood samples were collected in evacuated testtubes with heparin (150 USP units) by venipuncture of jugular vein(pigs) using sterile equipments respectively on 0 h, 5 h, 24 h, 3 d, 7 d,7 d+5 h, 7 d+24 h, and 10 d post first immunization. At these time-points, piglets were euthanized by bleeding under anaesthesia byintravenous injection of Halothane (24 mg/kg). Intestinal tissues(jejunum, caecum and colon) were collected respectively at the sametime-points. Some tissue samples were stored at −70 °C until assayedby quantitative real-time sequence detection system for MCP-1, IP-10and MIP-β mRNA expression in intestinal tissues using β-actin geneexpression as housekeeping gene. Other tissue samples were directlyused for the mononuclear cell (MNC) isolation.

2.4. Cell preparation

Peripheral blood mononuclear cells (PBMCs) were prepared bydensity gradient centrifugation of heparinized peripheral blood samplesobtained from pigs as previously reported [21]. Briefly, a volume ofperipheral blood was diluted with an equal volume of PBS and layeredon Ficoll-Hypaque PLUS. After centrifugation at 350×g at roomtemperature for 40 min, the layer containing the PBMC fraction wasobtained and washed once with PBS. Red blood cells were removed byhypotonic shock and the remaining cellswerewashed twice in cold PBS.Then the cells were resuspended in PBS at a cell concentration of 1×106

cells/mL for RNA isolation.

2.5. A quantitative real-time sequence detection system to measureporcine IP-10, MCP-1 and MIP-β mRNA expression

A real-time sequence detection system was applied for thequantitative measurement of chemokine mRNAs as previouslyreported (Asmi Citra Malina, 2006).Total RNA was isolated fromcells or tissues using commercial kit — SV Total RNA Isolation System,and reverse-transcribed into cDNA with RevertAid First Strand cDNAsynthesis kit. Expression of mRNA was quantified with real-time PCR(Opticon-II, MJ, Germany) using a commercial reagent kit. For each ofthe targeted genes, a pair of oligonucleotide primers were designed byPrimer Premier 5.0 software (Table 1), based on the sequencesregistered in GenBank database (GenBank/EMBL/DDBJ accessionnumber: IP-10, DQ372065; MIP-1β , NM_213779; MCP-1,NM_214214; and β-actin, U07786). Relative quantities of chemokinemRNAs were determined by relative quantitative RT-PCR using β-actin gene expression as housekeeping gene. Values for each targetgene were normalized using porine β-actin. Expression values werecalculated using the 2−ΔΔCt method [19], i.e., the amount ofchemokine mRNAs, normalized to the endogenous reference β-actinmRNA and relative to the control sample, is given by:

2− chemokine Ct−β�actin Ct of CpG�stimulated sampleð Þ− chemokine Ct�β�actin Ct of the control sampleð Þð Þ

this study.

Nucleotide position Length of the PCR product(bp)

137–158 125TC 283–261AG 191–212 123TC 292–313

118–136 130229–247256–274 109345–364

613Q. Cheng et al. / International Immunopharmacology 10 (2010) 611–618

(If the samples were from group A (control samples), values of groupA can be calculated with:

2− chemokine Ct−β�actin Ct of the control sampleð Þ− chemokine Ct�β�actin Ct of the control sampleð Þð Þ = 20 = 1Þ:

2.6. Isolation of the mononuclear cells (MNC) and flow cytometryanalysis of monocytes/macrophages and dendritic cells (DCs)

The jejunum, caecum and colon were collected from each pig ateuthanasia and the mononuclear cells (MNC) were isolated aspreviously described [28,29]. Following cell isolation, cell stainingwas performed on the same day. The freshly isolated MNC werewashed with staining buffer [3% FBS and 0.09% sodium azide in PBS(0.2 mg/mL KCI, 0.2 mg/mL KH2PO4, 8 mg/mL NaCl and 2.16 mg/mLNa2HPO4, and pH 7.2–7.4)] and stained with primary antibodiesagainst SWC3 conjugated with biotin, CD11R1, followed by secondaryreagent streptavidin conjugated with peridinine chlorophyll protein(PerCP) or antibodies conjugated with allophycocyanin (APC). Isotypematched antigen-irrelevant control monoclonal antibody IgG1 (forCD11R1) was used in each staining as negative control to set thequadrantmarkers for the bivariate dot plots. At least 10,000 cells were

Fig. 1. IP-10 relative gene expressions in (A) jejunum, (B) caecum, (C) colon and (D) PBMC foto their dams were treated intramuscularly (IM) or intranasally (IN) or orally. 7 d later, piglmRNA was determined by relative quantitative RT-PCR using β-actin gene expression as h*pb0.05 and **pb0.01 vs IM-10 μg/kg-CpG group, #pb0.05 and ##pb0.01 vs IN-10 μg/kg-C

analyzed per sample. Stained cells were analyzed by cytofluorometry(FACSCalibur, Becton & Dickinson, USA), and the data were analyzedwith Cellquest V3.3 software. Results were reported as percentage ofcell populations expressing the markers of interest among gated cells.

2.7. Statistical analysis

Data were analyzed using the statistical software program Systat 10(SPSS). Distribution of data was determined using descriptive statistics.Datawhichwerenotnormallydistributedwere transformedby ranking.Means of the rank were compared using Tukey's multiple comparisontests. A p-value of b0.05 was considered significant.

3. Results

3.1. Local expression of the CXC chemokine IP-10 mRNA post CpG ODNadministration

To test local and systemic chemokine responses post CpG ODNadministration, piglets were administered (A) PBS, (B) IM-10 μg/kg-CpG,(C) IM-100 μg/kg-CpG, (D) IM-1000 μg/kg-CpG, (E) IN-10 μg/kg-CpG,(F) IN-100 μg/kg-CpG, and (G) IN-1000 μg/kg-CpG on days 0 and 7 post

llowing CpG or PBS treatment. Seven groups of 14 piglets which were divided accordingets were re-administered with the same formulations. Relative quantity of chemokinesousekeeping gene. Mean relative quantity of chemokines mRNA +/−SEM is shown.pG group.

614 Q. Cheng et al. / International Immunopharmacology 10 (2010) 611–618

first administration and measured CXC and CC chemokines in theintestine and PBMCs at various time-points.

Since IP-10 plays an important role in protection against a variety ofbacterial and viral infections, we investigated whether IN- or IM-CpGadministration induced IP-10 mRNA expression in the intestinal tractmucosas (jejunum, caecum, and colon). As shown in Fig. 1, basal mRNAexpression levelsof IP-10weredetected in the jejunum, caecumandcolonof piglets before CpG treatment. The expression levels of IP-10 mRNAelevated sharply at 5 h post IN-100 μg/kg-CpG or IM-100 μg/kg-CpGprime delivery in the jejunum, while some moderately in the caecumandcolon. Theexpression levels of IP-10mRNAwanedat 24 h; thesewerefollowed by a second wave of IP-10 mRNA expression at day 3 inthe jejunum, caecum and colon. After boost, CpG delivery induced similar(but not stronger) IP-10 responses in the jejunum, caecum and colon ofpiglets. Throughout the experiment, IN-100 μg/kg-CpG elicited muchstronger IP-10 responses than IM-100 μg/kg-CpG in the intestinalmucosas, other deliveries such as IM-10 μg/kg-CpG, IM-1000 μg/kg-CpGand IN-10 μg/kg-CpG, IN-1000 μg/kg-CpG all induced someweaker IP-10responses than IM-100 μg/kg-CpG and IN-100 μg/kg-CpG respectively.

Fig. 2. MCP-1 relative gene expressions in (A) jejunum, (B) caecum, (C) colon and (D) PBaccording to their dams were treated intramuscularly (IM) or intranasally (IN) or orally. 7chemokines mRNA was determined by relative quantitative RT-PCR using β-actin gene exprshown. *pb0.05 and **pb0.01 vs IM-10 μg/kg-CpG group, #pb0.05 and ## pb0.01 vs IN-10

3.2. Local expression of the CC chemokines MCP-1 and MIP-1β mRNApost CpG ODN administration

To examine the effects of IN or IM administration of CpG ODN onlocalmucosalmRNA expression levels of the CC chemokinesMCP-1 andMIP-1β, we determined the mRNA expression levels of the MCP-1 andMIP-1β in the intestine (jejunum, caecum and colon) at various time-points post CpG ODN administration. As shown in Fig. 2, a strongexpression of theMCP-1mRNA at 5 h post CpG administrationwas onlyobserved in the jejunum, but weaker in the caecum and colon. Theexpression of theMCP-1mRNA alsowaned at 24 h, and peaked again atday 3 post CpG prime or boost in the jejunum. But in the caecum andcolon, the expression of the MCP-1 mRNA only peaked at 3 d post CpGprime and boost. Similar with IP-10 responses, IN-100 μg/kg-CpG alsoelicited much stronger MCP-1 responses than IM-100 μg/kg-CpG in theintestinal mucosas. However, in the colon, IN-10 μg/kg-CpG inducedsimilar MCP-1 mRNA expression compared with IN-100 μg/kg-CpG.

MIP-1β responses induced by CpG treatment in the jejunumweresimilar with that in the caecum (Fig. 3). Expression levels of MIP-1β

MC following CpG or PBS treatment. Seven groups of 14 piglets which were dividedd later, piglets were re-administered with the same formulations. Relative quantity ofession as housekeeping gene. Mean relative quantity of chemokines mRNA +/−SEM isμg/kg-CpG group.

Fig. 3. MIP-1 beta relative gene expressions in (A) jejunum, (B) caecum, (C) colon and (D) PBMC following CpG or PBS treatment. Seven groups of 14 piglets which were dividedaccording to their dams were treated intramuscularly (IM) or intranasally (IN) or orally. 7 d later, piglets were re-administered with the same formulations. Relative quantity ofchemokines mRNA was determined by relative quantitative RT-PCR using β-actin gene expression as housekeeping gene. Mean relative quantity of chemokines mRNA +/−SEM isshown. *pb0.05 and **pb0.01 vs IM-10 μg/kg-CpG group, #pb0.05 and ##pb0.01 vs IN-10 μg/kg-CpG group.

615Q. Cheng et al. / International Immunopharmacology 10 (2010) 611–618

mRNA peaked respectively at 5 h and 3 d post prime and boost, IN-100 μg/kg-CpG delivery induced strongerMIP-1β responses than IM-100 μg/kg-CpG delivery. However, a strong expression of the MIP-1βmRNA in colon was observed at 5 h post prime and boost, and thisexpression level was maintained until 3 d post prime and boost,and then waned at 7 d post prime. In this intestine section (colon),similar MIP-1β mRNA expression levels were observed among IN-10 μg/kg-CpG, IN-100 μg/kg-CpG, IM-10 μg/kg-CpG, and IM-100 μg/kg-CpG delivery (pN0.05).

3.3. Systemic expression of the CXC chemokine IP-10 mRNA post CpGODN administration

We further tested the ability of CpG ODN to induce expression ofchemokines in PBMCs. As shown in Fig. 1D, the expression levels of IP-10 mRNA in PBMCs increased at 5 h post CpG ODN administration,and waned at 24 h, and peaked further at 3 d post delivery. In contrastto intestinal mucosa chemokine responses, CpG-IM delivery inducedmuch stronger IP-10 responses than IN delivery in PBMCs at 5 h postCpG administration. Expression levels of IP-10 mRNA induced by IN

delivery were delayed until 3 d post administration, when the levelscould be comparative with IM delivery.

3.4. Systemic expression of the CC chemokines MCP-1 and MIP-1βmRNApost CpG ODN administration

Except for the CXC chemokine, the systemic expression of CCchemokines also was determined (Figs. 2D, 3D). Similar with CXCchemokine expression in PBMCs, the CC chemokines also peaked twicepost each administration. AlthoughmRNAexpression levels of theMCP-1and MIP-1β induced by IM delivery (specially IM-100 μg/kg-CpG) werehigher than that induced by IN delivery (specially IN-100 µg/kg-CpG) at5 h post CpG ODN administration, interestingly, MCP-1 and MIP-1βresponses elicited by IN delivery at 3 d were similar with IM delivery.

3.5. Percentages of monocytes/macrophages and DCs

In pigs, the most frequently detected marker on porcine mono-cytes/macrophages and DCs is SWC3 (CD172a) [5–7,14]. It is amember of the signal regulatory protein family and associates with

Fig. 4. Flow cytometry analysis of monocytes/macrophages and cDCs in pig intestine.Forward and side scatter of MNC R1 area was used for electronic gating (A). SWC3/CD11R1 dot plot was performed within R1 area to define monocytes/macrophages andcDCs with (SWC3+CD11R1−) and (SWC3+CD11R1+) areas, respectively (B).

616 Q. Cheng et al. / International Immunopharmacology 10 (2010) 611–618

protein–tyrosine phosphatase SHP-1 [2]. In addition to SWC3, CD11b(CD11R1) is a marker specifically and differentially expressed onporcine DCs, but not on monocytes/macrophages [6,7]. In this study, inorder toprovidemechanistic insights to our observationsweusedSWC3and CD11R1 to define cDCs as SWC3+CD11R1+ and monocytes/macrophages as SWC3+CD11R1−, and detected their distributions inthe jejunum, caecum and colon (Figs. 4–5).

In ileum, at 5 h or 3 d post first immunization, percentages ofmonocytes/macrophages in all treatment groups were higher than PBScontrol, and those in the IN-100 μg/kg and IM-100 μg/kg groups wererespectively significantly higher than in the PBS group (pb0.01,pb0.05). On the other hand, the highest percentages of cDCs in ileumwere also observed in the IN-100 μg/kg groupand significantly higher inboth IN-100 μg/kg and IM-100 μg/kg groups than control group.However, at 24 h and 7 d post first immunization there were nosignificant differences for the percentages of both monocytes/macro-phages and cDCs among all treatment. Similar results were alsoobserved in the caecum and colon (Fig. 5). These results suggested

that dosage and route both played a major role in increasing thefrequencies of monocytes/macrophages and cDCs in the intestine.

4. Discussion

The ability of CpG ODN to activate the innate immune system andits use as potent immunomodulator against P. yoelii, L. monocytogenes,H. pylori and L. major had clearly been shown in mice in vivo[10,12,17,25,34]. But only limited studies examined in vivo the effectsof CpG ODN on piglets. Kamstrup et al. [16] showed that porcinePBMCs could respond to palindromic hexamer motifs containing anunmethylated CpG ODN. Moreover, they found that ATCGAT was theoptimal sequence to stimulate the proliferation and the expression ofcytokine (IL-6, IL-12, and TNF-α) mRNA. Our previous in vivo studyalso had shown that CpG ODN could stimulate innate immune systemin newborn piglets by preventing the reduction of the proportion ofCD4+ T cells and upregulation of MHC-II antigen and IFN-γ [32].However, there is a lack of information concerning the effects of CpGODN on chemokine expression in piglets. The data presented heredemonstrated that the intranasal (IN)–mucosal or intramuscular(IM)–systemic administration of CpG ODN could both enhance theinnate immunity in the intestinal mucosa and peripheral bloodmononuclear cells (PBMCs) in piglets. To our knowledge, this is thefirst time chemokine responses mediated by CpG ODN have beendemonstrated at intestinal mucosas and PBMCs in piglets.

Following IN–mucosal or IM–systemic delivery of CpG ODN, weobserved a rapid and transient induction of IP-10 mRNA expression inthe intestine (jejunum, caecum, and colon). Since the IP-10 receptor ismainly expressed on Th1 cells [1,11], it seems reasonable to speculatethat IP-10 preferentially recruits Th1 cells to the intestinal mucosa,which may in turn contribute to induction of protective immunityagainst pathogen infection. In addition to IP-10,we found that IN and IMadministrations of CpG ODN in piglets also triggered local mRNAexpression of the CC chemokines MIP-1β and MCP-1 in intestinaltissues. These CC chemokines can be secreted by a variety of cell types,includingmacrophages, activated NK cells, and T cells [4]. The commonreceptor for all three CC chemokines, CCR5 [24], is expressed onmacrophages, dendritic cells, NK cells, and activates Th1 cells and canthus recruit these immunocompetent cells into the intestinalmucosas ofpiglets that received CpG ODN [26,27]. Recently, it was shown that asingle vaginal–mucosal dose of CpG ODN elicited rapid Th1-likecytokine and chemokine responses in the female genital tract mucosa[13]. Thus, it appears that IN or IM delivery of CpG ODN is capable ofinducing strong innate immunity at mucosal (intestine) surfaces.

In the present study, two waves (peaked respectively at 5 h and 3 dpost CpG administration) were all observed in the mRNA expression ofIP-10, MCP-1 andMIP-1β and percentages of macrophages and cDCs inthree sections of the intestine tissues throughout this experiment,which is consistent with a previous report showing that intragastricadministration of a single dose of CpG ODN induced two waves ofsignificantly higher levels of the local CC chemokines MIP-1β, MIP-1α,and RANTES and the CXC chemokine IP-10 in the stomach and/or thesmall intestine [25]. In this study we observed the marked enhancedpercentages of macrophages and dendritic cells in the intestine tissuespost CpG administration, since IP-10 and MIP-1β, MCP-1 receptors aremainly expressed on macrophages, dendritic cells, T cells and NK cells[20] and the induction of IP-10 and MIP-1β, MCP-1 in the intestinetissues suggests that these chemokines can recruit innate immune cellsto the intestinemucosas, whichmay contribute to induce strong innateimmunity against pathogen invasion. But the mechanism whichmediated the biphasic nature of the chemokine responses is still notcompletely clear. Since many kinds of immune cells (such as macro-phages, dendritic cells, epithelial and endothelial cells) all can expressthe TLR9, which can be activated by CpG ODN directly, we think theprimary peak or phase of the chemokine responses was due to eitherdirectly by activated immune cells in the tissue or indirectly through

Fig. 5. Distribution of monocytes/macrophages and cDCs in (A) jejunum, (B) caecum, and (C) colon following CpG or PBS treatment. Seven groups of 14 piglets which were dividedaccording to their dams were treated intramuscularly (IM) or intranasally (IN) or orally. 7 d later, piglets were re-administered with the same formulations. Mean percentage ofmonocytes/macrophages and cDCs +/−SEM is shown. *pb0.05 and **pb0.01 vs PBS group.

617Q. Cheng et al. / International Immunopharmacology 10 (2010) 611–618

activation of epithelial and endothelial cells by CpG ODN inducedcytokines, however, whether the secondary peak or phase of thechemokine responses was due to recruited immune cells, still neededfurther investigation.

Furthermore, we found that IN-CpG delivery could induce muchstronger chemokine (CC and CXC chemokines) responses in theintestine than IM-CpG delivery. Whereas IM-CpG delivery could elicita stronger IP-10 in PBMCs than IN-CpG delivery at 5 h post CpGadministration, this suggests that CpG ODN administered IM passesrapidly through the injected tissues and passes directly into the bloodwhere systemic activation occurs. In contrast, following IN delivery CpGODN remains in the intestinal mucosal area where it activates localimmune cells. This is also supported by previous findings that local, butnot systemic, delivery of CpG ODN induced rapid proliferation and

thickening of the genital epithelium and caused significant recruitmentof inflammatory cells to submucosa [3].

Surprisingly, we observed that 100 μg/kg was an optimal dosagewhich could induce strongest chemokine responses in both CpG-IN-treated and CpG-IM-treated groups, this is not consistent with ourprevious study, where CpG ODN in PBS or emulsigen could enhance thelevels of MHC-II expression and IFN-γ in a dose-dependent manner inpiglets. These differences in dosage responses could have been due tobreed and age-related changes. Our previous study used 1-day-oldLandrace×Large White piglets while in the present study some olderconventional Landrace×Yorkshire×Durok pigs (2 weeks at the momentof first immunization) were used.

In conclusion, we report in this study that IN–mucosal and IM–

systemic administrations of CpG ODN both induce local and systemic

618 Q. Cheng et al. / International Immunopharmacology 10 (2010) 611–618

mRNA expression of the CC chemokines MCP-1 and MIP-1β and theCXC chemokine IP-10 in the intestinal tissues (jejunum, caecum andcolon) and PBMCs. Our data show that IN-route can result in muchstronger chemokine responses in the gut than IM-route. That CpGODN IN treatment associates with increased expression of CC and CXCchemokines further confirms that CpG ODN induces a Th1-typeresponse potentially leading to improved host defenses in the gut.These findings suggest that IN–mucosal delivery of CpG ODNmay be avaluable supplement to the therapy for control of intestinal pathogen.

Acknowledgments

This study was supported by Guangdong Province AgriculturalTechnological Project (no: 2009B020307010) and Key Laboratory ofAnimal Disease Control and Prevention of the Ministry of Agriculture.

References

[1] Akhiani AA, Pappo J, Kabok Z, Schon K, Gao W, Franzen LE, Lycke N. Protectionagainst Helicobacter pylori infection following immunization is IL-12-dependentand mediated by Th1 cells. J Immunol 2002;169:6977–84.

[2] Alvarez B, Sanchez C, Bullido R, Marina A, Lunney J, Alonso F, Ezquerra A,Dominguez J. A porcine cell surface receptor identified by monoclonal antibodiesto SWC3 is a member of the signal regulatory protein family and associates withprotein– tyrosine phosphatase SHP-1. Tissue Antigens 2000;55:342–51.

[3] Ashkar AA, Bauer S, Mitchell WJ, Vieira J, Rosenthal KL. Local delivery of CpGoligodeoxynucleotides induces rapid changes in the genital mucosa and inhibitsreplication, but not entry, of herpes simplex virus type 2. J Virol 2003;77(16):8948–56.

[4] Baggiolini M, Dewald B, Moser B. Interleukin-8 and related chemotactic cytokines—CXC and CC chemokines. Adv Immunol 1994;55:97–179.

[5] Bautista EM, Gregg D, Golde WT. Characterization and functional analysis of skin-derived dendritic cells from swine without a requirement for in vitro propagation.Vet Immunol Immunopathol 2002;88:131–48.

[6] Bimczok D, Post A, Tschernig T, Rothkotter HJ. Phenotype and distribution ofdendritic cells in the porcine small intestinal and tracheal mucosa and their spatialrelationship to epithelial cells. Cell Tissue Res 2006;325:461–8.

[7] Bimczok D, Sowa EN, Faber-Zuschratter H, Pabst R, Rothkotter HJ. Site-specificexpression of CD11b and SIRPalpha (CD172a) on dendritic cells: implications fortheirmigrationpatterns in the gut immune system. Eur J Immunol 2005;35:1418–27.

[8] Coban C, Ishii KJ, Kawai T, Hemmi H, Sato S, Uematsu S. Toll-like receptor 9mediates innate immune activation by the malaria pigment hemozoin. J Exp Med2005;201:19–25.

[9] Dar A, Nichani AK, Benjamin P, Lai K, Soita H, Krieg AM. Attenuated cytokineresponses in porcine lymph node cells stimulated with CpG DNA are associatedwith low frequency of IFNalpha-producing cells and TLR9 mRNA expression. VetImmunol Immunopathol 2008;123:324–36.

[10] Elkins KL, Rhinehart-Jones TR, Stibitz S, Conover JS, Klinman DM. Bacterial DNAcontaining CpGmotifs stimulates lymphocyte-dependent protection ofmice againstlethal infection with intracellular bacteria. J Immunol 1999;162(4):2291–8.

[11] Garhart CA, Redline RW, Nedrud JG, Czinn SJ. Clearance of Helicobacter pyloriinfection and resolution of postimmunization gastritis in a kinetic study ofprophylactically immunized mice. Infect Immun 2002;70:3529–38.

[12] Gramzinski RA, Doolan DL, SedegahM, Davis HL, Krieg AM, Hoffman SL. Interleukin-12- and gamma interferon-dependent protection against malaria conferred by CpGoligodeoxynucleotide in mice. Infect Immun 2001;69(3):643–1649.

[13] Harandi AM, Eriksson K, Holmgren J. A protective role of locally administeredimmunostimulatory CpG oligodeoxynucleotide in a mouse model of genitalherpes infection. J Virol 2003;77:953–62.

[14] Haverson K, Bailey M, Higgins VR, Bland PW, Stokes CR. Characterization ofmonoclonal antibodies specific for monocytes, macrophages and granulocytes fromporcineperipheral blood andmucosal tissues. J ImmunolMethods1994;170:233–45.

[15] Janeway CA, Medzhitov RJ. Innate immune recognition. Annu Rev Immunol2002;20:197–216.

[16] Kamstrup S, Daniela V, DennisMK. Response of porcine peripheral bloodmononuclearcells to CpG-containing oligodeoxynucleotides. Vet Microbiol 2001;78(4):353–62.

[17] Krieg AM, Love-Homan L, Yi AK, Harty JT. CpG DNA induces sustained IL-12expression in vivo and resistance to Listeria monocytogenes challenge. J Immunol1998;161(5):2428–34.

[18] Krug A, French AR, Barchet W, Fischer JA, Dzionek A, Pingel JT. TLR9-dependentrecognition of MCMV by IPC and DC generates coordinated cytokine responsesthat activate antiviral NK cell function. Immunity 2004;21:107–19.

[19] Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real-timequantitative PCR and the 2-ΔΔCt method. Methods 2001;25:402–8.

[20] Luster AD. The role of chemokines in linking innate and adaptive immunity. CurrOpin Immunol 2002;14:129–35.

[21] Masuda K, Sakaguchi M, Saito S, Deboer DJ, Fujiwara S, Kurata K, Yamashita K,Hasegawa A, Ohno K, Tsujimoto H. In vivo and in vitro tests showing sensitizationto Japanese cedar (Cryptomeria japonica) pollen allergen in atopic dogs. J Vet MedSci 2000;62:995–1000.

[22] Matzinger P. Friendly and dangerous signals: is the tissue in control? Nat Immunol2007;8:11–3.

[23] Medzhitov R, Preston-Hurlburt PJ. A human homologue of the Drosophila Tollprotein signals activation of adaptive immunity. Nature 1997;388:394–7.

[24] Moser B, Loetscher P. Lymphocyte traffic control by chemokines. Nat Immunol2001;2:123–8.

[25] Raghavan S, Nyström J, FredrikssonM, Holmgren J, Harandi AM. Orally administeredCpG oligodeoxynucleotide induces production of CXC and CC chemokines in thegastric mucosa and suppresses bacterial colonization in a mouse model ofHelicobacter pylori infection. Infect Immun 2003;71(12):7014–22.

[26] Taub DD, Conlon K, Lloyd AR, Oppenheim JJ, Kelvin DJ. Preferential migration ofactivated CD4 and CD8 T cells in response to MIP-1 alpha and MIP-1 beta. Science1993;260:355–8.

[27] Uguccioni M, D'Apuzzo M, Loetscher M, Dewald B, Baggiolini M. Actions of thechemotactic cytokines MCP-1, MCP-2, MCP-3, RANTES, MIP-1 alpha and MIP-1beta on human monocytes. Eur J Immunol 1995;25:64–8.

[28] Ward LA, Rosen BI, Yuan L, Saif LJ. Pathogenesis of an attenuated and a virulentstrain of group A human rotavirus in neonatal gnotobiotic pigs. J Gen Virol1996;77(Pt 7):1431–41.

[29] Yuan L, Ward LA, Rosen BI, To TL, Saif LJ. Systematic and intestinal antibody-secreting cell responses and correlates of protective immunity to human rotavirusin a gnotobiotic pig model of disease. J Virol 1996;70:3075–83.

[30] Zhang LH, Guo Y, Tian XS, Zhou FZ. Co-administration of porcine-specific CpGoligodeoxynucleotide enhances the immune responses to pseudorabies attenuatedvirus vaccine in newborn piglets in vivo. Dev Comp Immunol 2006;30:589–96.

[31] Zhang LH, Tian XS, Zhou FZ. Vaccination with Newcastle disease vaccine and CpGoligodeoxynucleotides induces specific immunity and protection against Newcastledisease virus in SPF chicken. Vet Immunol Immunopathol 2007;115:216–22.

[32] Zhang LH, Tian XS, Zhou FZ. In vivo immunostimulatory effects of CpG ODN innewborn piglets. Mol Immunol 2007;44(6):1238–44.

[33] Zhang LH, Tian XS, Zhou FZ. In vivo effects of oligodeoxynucleotides containingsynthetic immunostimulatory motifs in the immune response to swine strepto-coccic septicemia vaccine in weaned piglets. Mol Immunol 2007;44(6):1141–9.

[34] Zimmermann S, Egeter O, Hausmann S, Lipford GB, Rocken M, Wagner H, Heeg K.CpG oligodeoxynucleotides trigger protective and curative Th1 responses in lethalmurine leishmaniasis. J Immunol 1998;160(8):3627–30.