Toll-Like Receptor 3 Ligand Polyinosinic:Polycytidylic Acid Promotes Wound Healing in Human and Murine Skin

See related commentary on pg 1955

ORIGINAL ARTICLE

Toll-Like Receptor 3 Ligand Polyinosinic:Polycytidylic Acid Promotes Wound Healing in Human and Murine Skin Qing Lin1,2,6, Li Wang3,6, Youkun Lin4, Xialin Liu1, Xiangrong Ren1, Sijian Wen4, Xiaolin Du3, Tao Lu3, Sarah Y. Su2, Xiaoping Yang2, Wenlin Huang5, Shiyou Zhou1, Feng Wen1 and Shao Bo Su1 Toll-like receptors (TLRs) are pattern-recognition receptors and have a critical role in both innate and adaptive responses to tissue injury. Our previous study showed that wound healing was impaired in TLR3-deficient mice. In this study, we investigated the capacity of the TLR3 agonist polyriboinosinic-polyribocytidylic acid (poly(I:C)) to promote the healing of skin wounds in humans and mice. We found that topical application with poly(I:C) accelerated the closure of wounds in patients with laser plastic surgery. In a mouse model, topical application of poly(I:C) markedly enhanced re-epithelialization, granulation, and neovascularization required for wound closure. Further studies revealed that poly(I:C) treatment resulted in enhanced recruitment of neutrophils and macrophages in association with upregulation of a chemokine, macrophage inflammatory protein-2 (MIP-2/ CXCL2), in the wounds. The effect of poly(I:C) was abolished in TLR3-deficient mice or by treatment with MIP-2/ CXCL2-neutralizing antibodies. These results suggest a potential therapeutic value of the TLR3 activator poly(I:C) for wound healing. Journal of Investigative Dermatology (2012) 132, 2085–2092; doi:10.1038/jid.2012.120; published online 10 May 2012

INTRODUCTION Wound healing of the skin represents a dynamic process involving re-epithelialization, fibroplasias, and angiogenesis (Singer and Clark, 1999). These processes are anteceded by recruitment of leukocytes spatiotemporally and differentially regulated by chemokines. The expression of chemokine receptors by resident cells such as keratinocytes and endothelial cells suggests that chemokines may also contribute to the regulation of epithelialization, tissue remodeling, and angiogenesis (McKay and Leigh, 1991). Thus, chemokines are in an exclusive position to integrate inflammatory events and reparative processes (Gillitzer and Goebeler, 2001). In addition to chemokines, the repair process is executed and promoted by complicated signaling network that involves numerous growth 1

The State Key Laboratory of Ophthalmology, Zhongshan Ophthalmic Center, Sun Yat-sen University, Guangzhou, China; 2Johns Hopkins University School of Medicine, Baltimore, Maryland, USA; 3Department of Dermatology, First Affiliated Hospital, Shantou University Medical College, Shantou, China; 4Department of Dermatology, First Affiliated Hospital, Guangxi Medical University, Nanning, China and 5The State Key Laboratory of Oncology in South China, Cancer Center, Sun Yat-sen University, Guangzhou, China

6

These authors contributed equally to this work.

Correspondence: Shao Bo Su, The State Key Laboratory of Ophthalmology, Zhongshan Ophthalmic Center, Sun Yat-sen University, 54 S Xianlie Road, Guangzhou, China. E-mail: [email protected] Abbreviations: Abs, antibodies; MIP-2, macrophage inflammatory protein-2; poly(I:C), polyriboinosinic-polyribocytidylic acid; TLR, Toll-like receptor; TRIF, Toll/IL-1R domain-containing adaptor-inducing IFN-b Received 28 October 2011; revised 27 January 2012; accepted 17 February 2012; published online 10 May 2012

& 2012 The Society for Investigative Dermatology

factors and cytokines to regulate the growth, differentiation, and metabolism of a target cell via paracrine, autocrine, juxtacrine, or endocrine mechanisms. The usage of transgenic and gene knockout mice provides important results that reveal the in vivo function of mediators in wound repair (Werner and Grose, 2003). The importance of cognate receptors responding to these mediators in cutaneous wound repair, such as TCRg (Jameson et al., 2002), TNFR p55 (Mori et al., 2002), chemokine receptor CXCR2 (Devalaraja et al., 2000), and CX3CR1 (Ishida et al., 2008), as well as Toll-like receptors (TLRs), have been demonstrated. As members of the pattern-recognition receptors that build the first defense line of the body, mammalian TLRs are linkers of innate and adaptive responses. TLRs recognize microbial, as well as endogenous, ligands from necrotic cells to induce inflammatory responses (Lin et al., 2011b). For example, TLR3 recognizes double-stranded RNAs derived from viruses (Matsumoto and Seya, 2008), and RNA released from damaged tissue. Ligands contained within endocytosed cells also activate TLR3 from within the cells (Kariko et al., 2004). Studies revealed that TLRs are crucial in the response to tissue injuries. A markedly slower healing of wounds has been observed in MyD88-deficient mice (Macedo et al., 2007), whereas the absence of TLR2 decreases inflammation and improves wound healing (Dasu et al., 2010). A recent study reported that TLR9 activation by its ligand CpG ODN accelerated wound healing in mice (Sato et al., 2010). In addition to MyD88-dependent signaling, Toll/IL-1R domaincontaining adaptor-inducing IFN-b (TRIF) activated by TLR3 was also involved in wound healing based on our previous www.jidonline.org 2085

Q Lin et al. Poly(I:C) Promotes Wound Healing

Table 1. Effects of topical poly(I:C) application on patient skin wounds Poly(I:C) Patient number

Age

Wound of

1

43

2

11

3 4

Saline Complete healing (day)

Patient number

Nevus

5

1

5

Nevus

Nevus of Ota

5

2

18

Nevus of Ota

14

Nevus

6

3

14

Tattoo

7

17

Nevus of Ota

5

4

19

Nevus of Ota

7

5

9

Nevus

5

5

2

Nevus

5

6

22

Nevus

5

6

35

Nevus

7

7

20

Nevus

4

7

31

Nevus

9

8

48

Nevus

7

8

25

Nevus

9

9

38

Freckle

6

9

17

Nevus

7

Average: 5.3 (days)

Age

Wound of

Complete healing (day) 7 10

Average: 7.6 (days)

Abbreviation: Poly(I:C), polyriboinosinic-polyribocytidylic acid. Wounds in patients receiving laser surgery were treated with poly(I:C) or saline daily until healing.

findings that skin healing was impaired in TLR3-deficient mice with defective expression of chemokines and recruitment of myeloid cells (Lin et al., 2011a). These results raised the possibility that activation of TLR3 in vivo with its exogenous agonists may promote wound closure. To test this hypothesis, we investigated the effect of topical application with TLR3 agonist polyriboinosinic-polyribocytidylic acid (poly(I:C)), a common injection solution for the treatment of viral diseases such as chronic viral hepatitis in clinic, on skin restoration, as well as the underlying mechanisms. Our results showed that topical application with poly(I:C) accelerated wound closure in patients with post laser surgery injury. We also found that in a mouse model activation of the TLR3 pathway by poly(I:C) promoted the healing of excisional skin wounds by enhancement of the production of a chemokine macrophage inflammatory protein-2 (MIP-2/CXCL2), which recruited myeloid cells to the wounds for repair. These results indicate a therapeutic potential for poly(I:C) to promote wound healing in humans. RESULTS Effects of poly(I:C) on skin wound healing in patients

To study the clinical effect of poly(I:C) on wound healing, 18 patients with laser surgery skin wounds were enrolled into the study. Eighteen patients with skin wounds caused by laser plastic surgery were enrolled into the study. Ethics approval for the study was granted by the local research ethics committee. The patients with laser wounds were randomly divided into saline-treated and poly(I:C)-treated groups. The solution of poly(I:C; Guangdong Bangmin Pharmaceutical) or saline were topically applied to wounds once a day, and no patient showed systemic or local side effects during or after treatment. Among 18 patients with laser-caused wounds, 12 patients underwent laser surgery to remove nevus, 4 patients were operated to remove nevus of Ota, 1 patient had freckles, and 2086 Journal of Investigative Dermatology (2012), Volume 132

1 had a tattoo (Table 1). Poly(I:C) application markedly enhanced skin wound healing as compared with treatment with saline solution (Table 1 and Figure 1). In the group of 12 patients of nevus surgery (Table 1), 6 patients who received poly(I:C) treatment had wound closure time of 5.35±1.03 days, whereas the other 6 patients treated with saline needed 7.35±1.51 days to achieve wound closure. The difference in 2 days was statistically significant with a P-value of 0.023. Topical administration of poly(I:C) enhanced skin wound healing in mice

To study the basis of the mechanism for the capacity of poly(I:C) to promote skin wound healing, we used a mouse model. Four excision wounds with 4 mm diameter were performed symmetrically on the back skin of mice. Two right side wounds were treated with 1 mg ml1 poly(I:C) (Sigma) and two left side wounds were treated with delivery vehicle (Figure 2a). Wound areas were analyzed throughout the healing process. The results showed that treatment with poly(I:C) decreased wound area beginning on day 3 and became significant on day 5 after injury. Healing was completed on day 10 following surgery in poly(I:C)-treated wounds, which was 2 days earlier than in vehicle-treated wounds (Figure 2b and c). Microscopic examination on day 3 revealed that although the control skin wound displayed similar granulation tissue and wound gap area as compared with the poly(I:C)-treated wound (Figure 2d and e), poly(I:C) markedly enhanced re-epithelialization (Figure 2h and i), organized granulation tissue formation (Figure 2f and g), and neovascularization (Figure 2j and k), with a more promoted effect on day 5. The enhancing effect of poly(I:C) was not observed after denaturation (Supplementary Figure S1 online), confirming the specificity of the poly(I:C)-promoted wound healing. However, poly(I:C) failed to enhance wound healing in TLR3/ mice throughout the 14-day period

Q Lin et al. Poly(I:C) Promotes Wound Healing

Day 0

(Supplementary Figure S2a and b online) and failed to enhance re-epithelialization, granulation tissue formation, and neovascularization (Supplementary Figure S2c online). We also examined the effect of poly(I:C) on the expression of TLR3 and its adaptor protein TRIF in wounded skin. Quantitative PCR showed that TLR3 and TRIF mRNA was significantly upregulated in wounded skins of wild-type mice (Figure 2l and Supplementary Figure S3 online). Wounds with poly(I:C) treatment showed an increase in the expression of TLR3 and TRIF as compared with samples tissues without poly(I:C) treatment (Figure 2l). These findings indicate that poly(I:C) accelerates skin wound healing in a TLR3-dependent manner.

Day 5

Saline

Poly(I:C)

Figure 1. Therapeutic effects of polyriboinosinic-polyribocytidylic acid (poly(I:C)) topical application on human skin wounds after laser surgery. Wounds after laser surgery of nevus on the cheek of patients were shown. The wounds were treated daily with saline (upper panels) or poly(I:C) (lower panels) solution for 5 days. Shown is a representative of nine patients from each group.

Poly(I:C) enhanced leukocyte infiltration in skin wounds in mice

We next examined the effect of poly(I:C) on leukocyte infiltration at the excisional wound sites. The influx of neutrophils reached the peak on day 1 and gradually reduced until day 6 in all wounds without poly(I:C) treatment.

(I:

y ol

BS

100 % Original wound area

Post injury (day)

0

Poly(I:C)

c

P

P

PBS

)

C

b

a

3 5 7

80

PBS Poly(I:C)

60

*

40

*

20

*

0 0

10 PBS

Poly(I:C)

d d

e i

h

f

g

5 f

g

h

i

16

4 6 8 10 12 14 Post injury (day)

PBS

*

Poly(I:C)

12

*

8

4

7

j 10

l

Fold change

Post injury (day)

3

e

2

k

0

1

3 5 Post injury (day)

Figure 2. Polyriboinosinic-polyribocytidylic acid (poly(I:C)) accelerated skin wound healing in mice. (a) Topical administration of poly(I:C). Wounds were created on day 0 and immediately treated with topical application of vehicle on the two wounds on the left or poly(I:C) on the wounds on the right. Thereafter, animals received the same administration each day until complete skin wound healing. The analyses of the wound area are determined at the indicated time point. (b) Representative images of wounds treated with phosphate-buffered saline (PBS) or poly(I:C) post injury. (c) Kinetics of wound closure. Wound area was quantified and is expressed as the percentage of the original wound area. Data are the mean±SEM of n ¼ 6 and representative of three independent experiments. (d–k) Histopathology of wounds. The wounds were harvested for histopathology on days 3, 5, 7, and 10 post injury. On day 3, boxed areas showed granulation formation, and enlarged images are shown in d and e. (f, g) On day 5, boxed areas showed further granulation formation and (h, i) re-epithelialization. (j,k) Sections from wounds on day 5 were stained with anti-CD31 antibodies. Representative results were from three independent experiments with five animals in each group. Bar ¼ 100 mm. (l) Quantitative reverse transcriptase–PCR (RT-PCR) to assess Toll-like receptor 3 (TLR3) gene expression in wounded skin at the indicated time points. The relative changes were normalized against the housekeeping gene GAPDH and calculated using the 2–DDCT method. Results were from three separate experiments performed in duplicate and expressed as mean±SEM of fold increase over control (intact skin).*Po0.05 poly(I:C)-treated wounds versus PBS-treated wounds.

www.jidonline.org 2087

Q Lin et al. Poly(I:C) Promotes Wound Healing

a

b

c

g PBS

Poly(l:C)

Poly(l:C)

Cell number per hpf

PBS

d * 80 *

e

*

40

f

*

*

h PBS Poly(l:C)

F4/80

Ym1

Merge

F4/80

Fizz1

Merge

PBS

Poly(l:C) 0

1

3

6

1 3 6 1 Post injury (day)

3

6

Figure 3. Polyriboinosinic-polyribocytidylic acid (poly(I:C)) enhanced leukocyte accumulation in skin wounds in mice. Immunofluorescence analysis was performed on wound samples with antibodies (Abs) against myeloperoxidase (MPO; (a), day 1), F4/80 ((b), day 3), and CD3 ((c), day 6). Bar ¼ 100 mm. The numbers of (d) neutrophils, (e) macrophages, and (f) T cells per high-power field (hpf) were counted. *Po0.05, poly(I:C)-treated wounds in comparison with phosphate-buffered saline (PBS)–treated wounds. (g) Day 6 wounds treated with vehicle (upper panels) or poly(I:C) (lower panels) were double stained for F4/80 (green) and Ym1 (red) in granulation tissue. (h) Double labeling for F4/80 and Fizz1 in granulation tissue of day 6 wounds. Bar ¼ 100 mm. Representative results from three independent experiments with four animals in each group are shown.

Neutrophil infiltration was significantly enhanced by poly(I:C) treatment (Figure 3a and d). The difference was pronounced between days 1 and 3 post surgery, with a 41% increase on day 1 as compared with the wounds treated with phosphate-buffered saline (PBS; Figure 3a and d). Macrophages began to accumulate on day 1 after injury and reached maximal levels on day 6 (Figure 3b and e). The number of macrophages was notably increased in poly(I:C)treated wounds on days 3 and 6 (Figure 3b and e). In contrast, CD3 þ T cells in the wound site were in small numbers after injury. The kinetics and extent of CD3 þ T-cell recruitment were similar in poly(I:C)-treated and control wounds (Figure 3c and f). Thus, poly(I:C) specifically promoted the recruitment of neutrophils and macrophages, but not T cells, into the wounds. Macrophages in wounds can be divided into alternatively activated macrophages and ‘‘classically’’ activated macrophages (Stein et al., 1992). The classically activated macrophages tend to elicit tissue injury, whereas alternatively activated macrophages exhibit a high endocytic and phagocytotic capacity (Goerdt and Orfanos, 1999), promote angiogenesis (Kodelja et al., 1997), and contribute to wound healing (Schebesch et al., 1997). A number of markers for the alternatively activated macrophages including FIZZ1/ RELMa and YM1 have been identified (Raes et al., 2002). In our study, there were numerous F4/80 and Ym1 or F4/80 and FIZZ1 double-positive macrophages in both control and poly(I:C)-treated wound tissue 6 days after injury (Figure 3g and h). The number of Ym1- or FIZZ1-positive macrophages in poly(I:C)-treated wounds was higher than that in control wounds (Figure 3g and h). However, poly(I:C) treatment failed to enhance leukocyte accumulation in the wounds of TLR3/ animals (Supplementary Figure S4 online). These data indicate that activation of TLR3 by poly(I:C) enhanced 2088 Journal of Investigative Dermatology (2012), Volume 132

the recruitment of wound healing/alternatively activated/M2 macrophages into the wounds. Poly(I:C) enhanced chemokine MIP-2/CXCL2 expression in skin wounds of mice

Markedly enhanced infiltration of neutrophils and macrophages implied that chemokine expression might be upregulated in poly(I:C)-treated mouse skin wounds. Quantitative PCR arrays revealed that poly(I:C) treatment markedly increased the expression of the genes for MIP-2/CXCL2, MIP-1a/CCL3, MCP-1/CCL2, and RANTES/CCL5, with an elevated expression of MIP-2/CXCL2 up to 10.5-fold on day 1 (Figure 4a). The protein level of MIP-2/CXCL2 was also determined. In intact skin, MIP-2/CXCL2 protein was below detectable level (data not shown). The level of MIP-2/CXCL2 protein was significantly increased, reaching the peak on day 1, declining on day 3, and was still detectable until day 6 after injury (Figure 4b). At every checked point, poly(I:C) significantly enhanced MIP2/CXCL2 production. The most notable increase was seen on day 1, with up to a 2.5-fold increase above the control level (Figure 4b). MIP-2/CXCL2 is a ligand for the receptor CXCR2 (Jerva et al., 1997). CXCR2 expression was therefore examined in wounded skin by immunohistochemical staining. CXCR2 protein was detectable in normal skin, with increased expression in both the dermis and subcutaneous tissue of wounds particularly after poly(I:C) treatment (Figure 4c). Similar to protein expression, the CXCR2 mRNA was highly expressed in wounds on day 1 after injury, which was further enhanced by poly(I:C) treatment (Figure 4d). However, poly(I:C) treatment failed to upregulate MIP-2/CXCL2 and CXCR2 expression in TLR3/ mouse wounds (Supplementary Figure S5 online). These data suggest that MIP-2/CXCL2 induced by poly(I:C) has an important role in the recruitment of myeloid cells in TLR3-mediated wound healing.

Q Lin et al. Poly(I:C) Promotes Wound Healing

a I-TAC TARC SLC MIG MDC I-309 IP-10 KC Eotaxin RANTES MIP-3β MIP-2 MIP-1β MIP-1α MCP-1

–2

* *

*

* * 0

8

4

12

I-TAC TARC SLC MIG MDC I-309 IP-10 KC Eotaxin RANTES MIP-3β MIP-2 MIP-1β MIP-1α MCP-1

–2

Day 1

*

0.6 0.4 0.2 0

5 1 3 Post injury (day)

0

4 Fold change Day 3

* 8

10

c

–2

* * * * 0

4

d

PBS Poly(I:C)

10 8 6

PBS (day 1) Poly(I:C) (day 1)

* *

4 2 0

Normal (day 0)

8

Day 5

Fold change

0.8

*

PBS Poly(I:C)

*

Dermis

MIP-2 (ng mg–1)

1.0

*

Subcutaneous

b

*

I-TAC TARC SLC MIG MDC I-309 IP-10 KC Eotaxin RANTES MIP-3β MIP-2 MIP-1β MIP-1α MCP-1

5 1 3 Post injury (day)

Figure 4. Polyriboinosinic-polyribocytidylic acid (poly(I:C)) upregulated chemokine expression in mouse wounds. (a) Skin samples were obtained 0, 1, 3, and 5 days after the injury and chemokine gene expressions were quantified by real-time PCR using RNA from phosphate-buffered saline (PBS)- or poly(I:C)-treated wounded skin. Results are from three separate experiments performed in duplicate and expressed as the mean±SEM of fold increase over control (n ¼ 5 per group).*Po0.05 poly(I:C)-treated wounds in comparison with PBS-treated wounds as baseline. (b) Macrophage inflammatory protein-2 (MIP-2/CXCL2) in skin wounds on day 1, 3, and 5 post injury measured by ELISA. All results are expressed as ng chemokine per mg total protein recovered after homogenization of 8 mm punch biopsy. (c) Immunohistochemical analysis of CXCR2 levels in the dermis (upper panels) and subcutaneous tissue (lower panels) of skin wounds on days 0 and 1 post injury. Bar ¼ 100 mm. Representative results from three independent experiments with four animals in each group are shown. (d) Quantitative reverse transcriptase–PCR (RT-PCR) to assess the expression of CXCR2 mRNA in wounded skin at the indicated time points. Results are expressed as mean±SEM fold of control (intact skin; n ¼ 5 per group). *Po0.05, poly(I:C)-treated wounds versus PBS-treated wounds.

The effect of neutralizing MIP-2/CXCL2 antibody on poly(I:C)-treated wounds in mice

To further confirm the role of MIP-2/CXCL2 in wound healing, we treated mouse wounds with anti-MIP-2/CXCL2 antibodies (Abs) before poly(I:C) application. As shown in Figure 5a, the effect of poly(I:C) on wound closure was abrogated by neutralizing MIP-2/CXCL2 Ab treatment. Notably, in poly(I:C)-treated wounds, MIP-2/CXCL2 neutralization also reduced the mRNA expression of monocytespecific chemokine MCP-1/CCL2, MIP-1a/CCL3, and RANTES/CCL5 (Figure 5b), as well as epithelialization, granulation tissue formation (Figure 5c), angiogenesis, and macrophage accumulation in the wounds (Figure 5c and d). These results confirm that MIP-2/CXCL2 is a critical mediator for poly(I:C)-promoted wound healing. DISCUSSION The primary function of the skin is to serve as a protective barrier against environment insults. Loss of the integrity of large portions of the skin as a result of injury or illness may lead to severe disability or even death. The primary goals of the treatment of wounds are promoting rapid wound closure and the formation of a functional and aesthetically satisfactory scar (Singer and Clark, 1999). The present study demonstrates that in the clinic topical treatment with TLR3 agonist poly(I:C) accelerated wound closure in patients. The

mouse model study further showed that topical application of poly(I:C) promoted wound healing by enhancing re-epithelialization, granulation tissue formation, and angiogenesis during the proliferative phase. Moreover, poly(I:C) treatment enhanced inflammatory cell infiltration associated with MIP2/CXCL2 upregulation in wild-type mice. Poly(I:C) was an effective adjuvant for Th1/17 cell responses (Longhi et al., 2009; Ren et al., 2011). Poly(I:C) also has important regulatory activity on the production of proinflammatory cytokines and chemokines (Cavassani et al., 2008; Liu et al., 2008). In human skin, poly(I:C) induced MIP1a or IL-8 production by keratinocytes via TLR3 (Tohyama et al., 2005; Matsukura et al., 2006). In an in vivo study on skin wounds, it was found that poly(I:C) was a potent inducer of proinflammatory cytokines IL-6 and tumor necrosis factor-a (Lai et al., 2009). Recent reports showed that poly(I:C) binds to cytosolic receptors including PKR, retinoic acid–inducible gene-I (RIG-I), and melanoma differentiation–associated antigen 5 (MDA5) in addition to TLR3. In this study, we found that the stimulatory effect of poly(I:C) was absent in TLR3/ mice, indicating that TLR3 signaling is critical in poly(I:C)-enhanced wound healing. Our findings are consistent with the study by Maheshwari’s group showing that poly(I:C) enhanced wound healing in rats and mice by upregulating adhesion molecules, extracellular matrix proteins, and TGFb1 (Bhartiya et al., 1992; Sidhu et al., 1996). Our data also www.jidonline.org 2089

Q Lin et al. Poly(I:C) Promotes Wound Healing

b 80 60 40 20

* * *

0

c

Control lgG Anti-MIP-2

MIP-1α

RANTES

Mcp-1 6 4

*

*

*

*

*

2 0

0 2 4 6 8 10 12 Post injury (day) H&E staining

* 1

3

5

CD31

Control lgG

AntiMIP-2

5 1 3 Post injury (day)

F4/80

d Cell number per hpf

100

Control lgG Anti-MIP-2 Fold change

% Original wound area

a

1

3

5

Control lgG Anti-MIP-2

80 60 40

* *

20 0

6 3 1 Post injury (day)

Figure 5. The effect of neutralization of macrophage inflammatory protein-2 (MIP-2/CXCL2) on polyriboinosinic-polyribocytidylic acid (poly(I:C)) promoted wound healing in mice. (a) After punching, wounds on the right side of the dorsal skin were treated with 20 ml anti-MIP-2/CXCL2 antibody (Ab; 100 mg ml1) for 30 min, followed by poly(I:C) topical treatment. Wounds on the opposite side were pretreated with control IgG. Wound areas were digitally photographed and quantified at the indicated time intervals. (b) MIP-2/CXCL2 neutralization inhibited the expression of chemokine mRNA in poly(I:C)-treated wounds in wild-type (WT) mice. The ratios of MCP-1, RANTES, and MIP-1a to GAPDH in poly(I:C)-treated wounds with control IgG or anti-MIP-2/CXCL2 Ab pretreatment were determined by real-time PCR at the indicated time intervals. Results are the means (±SEM) of three separate experiments performed in duplicate and expressed as the fold over control (intact skin; n ¼ 4 per group). *Po0.05 versus wounds treated with control IgG þ poly(I:C). (c) Histopathology and immunofluorescence staining (CD31 and F4/80) of anti-MIP-2/CXCL2- and control IgG-pretreated wounds on day 5 post injury. Bar ¼ 100 mm. (d) The number of macrophages per high-power field (hpf) was counted in anti-MIP-2/CXCL2-treated and control IgG–treated wounds. Representative results from three experiments with four individual animals in each group are shown. *Po0.05 in comparison between anti-MIP-2/CXCL2 þ poly(I:C) and control IgG þ poly(I:C) treatment. H&E, hematoxylin and eosin.

demonstrate that poly(I:C) promotes wound healing by activating TLR3 and its signal adaptor TRIF to upregulate the expression of type I INF (data not shown), as well as the chemokine MIP-2/CXCL2 to enhance macrophage accumulation in the wounds. To our knowledge, these observations delineate a previously unreported mechanism by which poly(I:C) accelerates wound healing. Leukocyte infiltration is a hallmark of the inflammatory phase of skin wound healing (Lingen, 2001). The impaired cutaneous wound healing in mice deficient in CXCR2 (Devalaraja et al., 2000) or mice with morphine administration were linked to the defective neutrophil recruitment (Martin et al., 2010). Consistently, our data show that the stimulatory effect of poly(I:C) on early immune cell recruitment, which might have resulted from elevated MIP-2/CXCL2 expression, leads to a net acceleration in wound healing. We speculate that as neutrophils migrate into the site by 24 hours post injury more MIP-1a/CCL3, MCP-1/CCL2, and RANTES/ CCL5 were produced, resulting in a chemoattractant gradient that is essential for the subsequent recruitment of macrophages. It is generally agreed that wound macrophages have a pivotal role in the transition between inflammation and repair during cutaneous healing, by virtue of their function as a rich source of mediators, which are necessary for the initiation and propagation of new tissue formation in wounds (Singer and Clark, 1999; Werner and Grose, 2003; Eming et al., 2007; Schurmann et al., 2010). The notion was supported by in vivo studies in which depletion of macro2090 Journal of Investigative Dermatology (2012), Volume 132

phages caused impaired collagen deposition (van Amerongen et al., 2007) and defective wound repair (Leibovich and Ross, 1975). Other studies demonstrated that injection of macrophages into healing cutaneous wounds augments the repair process (Danon et al., 1989). Our study also showed that macrophages could sustain the angiogenesis cascade under inflammatory conditions in a positive feedback loop through the activation of the HMGB1-TLR4 signaling pathway (Lin et al., 2011c). It is known that alternatively activated macrophages are a main subpopulation with beneficial effect on wound healing (Martinez et al., 2009). Our study showed an increase of Fizz1 and Ym1 double-positive woundhealing/alternatively activated/M2 macrophages (Raes et al., 2002; Nair et al., 2003) in poly(I:C)-treated wounds, suggesting that the enhanced macrophage recruitment may be a key step in the wound healing promoted by poly(I:C). Furthermore, we found that neutralization of MIP-2/CXCL2 abolished the effect of poly(I:C) on wound healing, confirming that poly(I:C)-enhanced macrophage accumulation was dependent on elevated MIP-2/CXCL2. Chronic skin wounds affect 6.5 million patients in the United States (Sen et al., 2009). The immense economic and social impact of wounds calls for a better understanding of the biological mechanisms underlying cutaneous wound complications. In the clinic, the poly(I:C) injection is currently used as an IFN-a/b inducer for the treatment of viral diseases such as chronic hepatitis, flu, and herpes zoster. In our study, topical administration of poly(I:C) solution in

Q Lin et al. Poly(I:C) Promotes Wound Healing

patients accelerated skin wound healing as compared with the saline treatment. These observations indicate the therapeutic effect of poly(I:C) topical administration in human skin wound and suggest a potential therapeutic value for other abnormal wound healing. MATERIALS AND METHODS Reagents and Abs Poly(I:C), goat anti-rabbit and anti-rat IgG-peroxidase Abs, and rabbit anti-goat IgG-peroxidase Ab were purchased from Sigma. TRIzol reagent and Hoechst 33342 were from Invitrogen (Carlsbad, CA). The ExScript RT reagent kit was from TaKaRa (TaKaRa Biotechnology, DaLian, China). Brilliant SYBR Green QPCR Master Mix was from Stratagene (La Jolla, CA). FITC-conjugated anti-mouse F4/80 and FITC-conjugated anti-mouse CD3 were from eBioscience (San Diego, CA). Rabbit antimouse CXCR2, goat anti-mouse PECAM-1 (CD31), goat anti-mouse myeloperoxidase, FITC-conjugated donkey anti-goat, and PE-conjugated donkey anti-rabbit IgG polyclonal antibodies were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Rabbit anti-mouse Ym1 polyclonal antibody was from StemCell Technologies (Vancouver, British Columbia, Canada). Rabbit anti-mouse Fizz1 was from PeproTech (London, UK). Anti-mouse MIP-2/CXCL2 mAb and the mouse MIP-2/CXCL2 ELISA kit were from R&D Systems (Minneapolis, MN). For the clinical trial, the poly(I:C) injection solution (1 mg ml1) was purchased from Guangdong Bangmin Pharmaceutical (JiangMen, China).

Patients and clinical studies Eighteen patients with skin wounds caused by laser plastic surgery were enrolled into the study. Ethics approval for the study was granted by the local research ethics committee. The study was conducted according to the Declaration of Helsinki Principles and in compliance with the International Conference on Harmonization Harmonized Tripartite Guideline for Good Clinical Practice. All patients provided written informed consent before participation. The radiant flux of the laser is 11 mJ s1, which could create a skin lesion reaching the dermis. It was a double-blind, within-patient, saline-controlled, randomized study to investigate the topical efficacy of poly(I:C) on wound healing. The wounds were digitally photographed to determine the size of the areas. The laser wounds of nine patients were treated daily with topical application of poly(I:C) (2 mg per 10  10 cm2, 1 mg ml1, in sodium chloride injection solution). As controls, the other nine patients were daily treated with saline solution. The wounds were digitally photographed at the indicated time intervals.

Animals TLR3/ mice were purchased from Taconic Farms (Rockville, MD). These mice were backcrossed ten or more generations onto the C57BL/6 background, and were then intercrossed to obtain the knockout genotypes and wild-type mice. Littermates of both sexes aged between 8 and 12 weeks old were used in all experiments. Animals were housed individually in cages under specific pathogenfree conditions and given water and standard laboratory chow ad libitum during the experiments. Genotyping of animals was performed by PCR of DNA obtained from tail biopsies. Primers were created according to the genotyping protocol (Stock number: 005217) from the Jackson Laboratory (Bar Harbor, ME). Animal care and all experimental procedures were approved by the Institutional Laboratory Animal Care and Use Committee.

In vivo wound model Full-thickness wounds were created and analyzed as described previously (Lin et al., 2011a). After wound surgery, 20 ml 1 mg ml1 poly(I:C) in PBS were locally applied to the two wounds on the right side daily, whereas the other two wounds on the left side were treated with an equal volume of PBS alone as control. These conducts were paralleled with another two negative control groups in which the two skin wounds on the left were treated with 20 ml PBS or 1 mg ml1 of denatured poly(I:C), which was heated at 95 1C for 5 min and cooled immediately on ice, whereas the wounds on the right were not treated throughout the entire observation period. In another series of experiments, ahead of being treated with poly(I:C), the two wounds on the right side were topically applied with 20 ml of anti-MIP-2/CXCL2-neutralizing polyclonal Ab (endotoxin level: o0.1 EU per 1 mg of the antibody) or control IgG at a concentration of 100 mg ml1 into wild-type mice for 30 min. Wounds were redressed daily, and fresh control or test solutions were applied.

Histopathological and immunohistochemical analyses and leukocyte infiltration The procedures were described previously (Lin et al., 2011a). Wound specimens were fixed and the sections were stained routinely with hematoxylin and eosin for histological analysis. Immunohistochemical analyses were performed using anti-CXCR2 (or -MIP-2/CXCL2) Abs. Frozen cryostat sections of skin wound tissues were immunostained with anti-myeloperoxidase, -F4/80, and -CD3 Ab for neutrophilic granulocytes, macrophages, and T cells, respectively, and double-color immunofluorescence analysis was performed to identify the types of Ym1- or Fizz1-positive macrophages.

ELISA Wound samples were excised and homogenized in 0.4 ml lysis buffer (10 mM PBS, 0.1% SDS, 1% Nonidet P-40, and 5 mM EDTA) containing Protease Inhibitors. The homogenates were centrifuged at 12,000 r.p.m. for 15 min. Supernatants were used to determine MIP2/CXCL2 levels with commercial ELISA kits according to the manufacturer’s instructions. Total protein in the supernatants was measured following the Bradford method. The data were expressed as the target molecule (ng) per total protein (mg) for each sample.

Total RNA isolation and quantitative real-time PCR Total RNA isolation and quantitative real-time PCR were performed as described previously (Lin et al., 2011a). Oligonucleotide primers (Supplementary Table S1 online) were synthesized by Invitrogen Biotechnology (Shanghai, China). The values for the initial concentrations of unknown samples were calculated using a software (version 1.7) provided with the ABI 7700 system. The tested mRNA expression in each sample was finally determined after correction with GAPDH expression. Each measurement of a sample was recorded in duplicate.

Reproducibility and statistical analysis Laboratory experiments were repeated at least twice. All figures show pooled data from replicate experiments or representative experiments as indicated. Data are expressed as mean±SEM for the indicated number of independently performed duplicated experiments. Statistical significance between means was analyzed by one-way ANOVA or two-tailed Student’s t test using the SPSS 13.0 version (IBM, Armonk, NY). A value of Po0.05 was considered statistically significant. www.jidonline.org 2091

Q Lin et al. Poly(I:C) Promotes Wound Healing

CONFLICT OF INTEREST SBS has applied for patents concerning the activity of polyinosinic:polycytidylic acid including their use in accelerating wound repair (application number: 61589348). The other authors state no conflict of interest.

ACKNOWLEDGMENTS We are grateful to Dr Ji Ming Wang of the National Cancer Institute, NIH, for helpful critique of the manuscript. This project was supported by the National Natural Science Foundation of China (81072483, 31170861, and 30872281). SUPPLEMENTARY MATERIAL

Lingen MW (2001) Role of leukocytes and endothelial cells in the development of angiogenesis in inflammation and wound healing. Arch Pathol Lab Med 125:67–71 Liu Y, Kimura K, Yanai R et al. (2008) Cytokine, chemokine, and adhesion molecule expression mediated by MAPKs in human corneal fibroblasts exposed to poly(I:C). Invest Ophthalmol Vis Sci 49:3336–44 Longhi MP, Trumpfheller C, Idoyaga J et al. (2009) Dendritic cells require a systemic type I interferon response to mature and induce CD4+ Th1 immunity with poly IC as adjuvant. J Exp Med 206:1589–602 Macedo L, Pinhal-Enfield G, Alshits V et al. (2007) Wound healing is impaired in MyD88-deficient mice: a role for MyD88 in the regulation of wound healing by adenosine A2A receptors. Am J Pathol 171:1774–88

Supplementary material is linked to the online version of the paper at http:// www.nature.com/jid

Martin JL, Koodie L, Krishnan AG et al. (2010) Chronic morphine administration delays wound healing by inhibiting immune cell recruitment to the wound site. Am J Pathol 176:786–99

REFERENCES

Martinez FO, Helming L, Gordon S (2009) Alternative activation of macrophages: an immunologic functional perspective. Annu Rev Immunol 27:451–83

Bhartiya D, Sklarsh JW, Maheshwari RK (1992) Enhanced wound healing in animal models by interferon and an interferon inducer. J Cell Physiol 150:312–9 Cavassani KA, Ishii M, Wen H et al. (2008) TLR3 is an endogenous sensor of tissue necrosis during acute inflammatory events. J Exp Med 205: 2609–21 Danon D, Kowatch MA, Roth GS (1989) Promotion of wound repair in old mice by local injection of macrophages. Proc Natl Acad Sci USA 86: 2018–20 Dasu MR, Thangappan RK, Bourgette A et al. (2010) TLR2 expression and signaling-dependent inflammation impair wound healing in diabetic mice. Lab Invest 90:1628–36 Devalaraja RM, Nanney LB, Du J et al. (2000) Delayed wound healing in CXCR2 knockout mice. J Invest Dermatol 115:234–44 Eming SA, Krieg T, Davidson JM (2007) Inflammation in wound repair: molecular and cellular mechanisms. J Invest Dermatol 127:514–25 Gillitzer R, Goebeler M (2001) Chemokines in cutaneous wound healing. J Leukoc Biol 69:513–21 Goerdt S, Orfanos CE (1999) Other functions, other genes: alternative activation of antigen-presenting cells. Immunity 10:137–42 Ishida Y, Gao JL, Murphy PM (2008) Chemokine receptor CX3CR1 mediates skin wound healing by promoting macrophage and fibroblast accumulation and function. J Immunol 180:569–79 Jameson J, Ugarte K, Chen N et al. (2002) A role for skin gammadelta T cells in wound repair. Science 296:747–9 Jerva LF, Sullivan G, Lolis E (1997) Functional and receptor binding characterization of recombinant murine macrophage inflammatory protein 2: sequence analysis and mutagenesis identify receptor binding epitopes. Protein Sci 6:1643–52 Kariko K, Ni H, Capodici J et al. (2004) mRNA is an endogenous ligand for Toll-like receptor 3. J Biol Chem 279:12542–50 Kodelja V, Muller C, Tenorio S et al. (1997) Differences in angiogenic potential of classically versus alternatively activated macrophages. Immunobiology 197:478–93 Lai Y, Di Nardo A, Nakatsuji T et al. (2009) Commensal bacteria regulate Tolllike receptor 3-dependent inflammation after skin injury. Nat Med 15:1377–82 Leibovich SJ, Ross R (1975) The role of the macrophage in wound repair. A study with hydrocortisone and antimacrophage serum. Am J Pathol 78:71–100 Lin Q, Fang D, Fang J et al. (2011a) Impaired wound healing with defective expression of chemokines and recruitment of myeloid cells in TLR3deficient mice. J Immunol 186:3710–7 Lin Q, Li M, Fang D et al. (2011b) The essential roles of Toll-like receptor signaling pathways in sterile inflammatory diseases. Int Immunopharmacol 11:1422–32 Lin Q, Yang XP, Fang D et al. (2011c) High-mobility group box-1 mediates Toll-like receptor 4-dependent angiogenesis. Arterioscler Thromb Vasc Biol 31:1024–32

2092 Journal of Investigative Dermatology (2012), Volume 132

Matsukura S, Kokubu F, Kurokawa M et al. (2006) Synthetic double-stranded RNA induces multiple genes related to inflammation through Toll-like receptor 3 depending on NF-kappaB and/or IRF-3 in airway epithelial cells. Clin Exp Allergy 36:1049–62 Matsumoto M, Seya T (2008) TLR3: interferon induction by double-stranded RNA including poly(I:C). Adv Drug Deliv Rev 60:805–12 McKay IA, Leigh IM (1991) Epidermal cytokines and their roles in cutaneous wound healing. Br J Dermatol 124:513–8 Mori R, Kondo T, Ohshima T et al. (2002) Accelerated wound healing in tumor necrosis factor receptor p55-deficient mice with reduced leukocyte infiltration. FASEB J 16:963–74 Nair MG, Cochrane DW, Allen JE (2003) Macrophages in chronic type 2 inflammation have a novel phenotype characterized by the abundant expression of Ym1 and Fizz1 that can be partly replicated in vitro. Immunol Lett 85:173–80 Raes G, De Baetselier P, Noel W et al. (2002) Differential expression of FIZZ1 and Ym1 in alternatively versus classically activated macrophages. J Leukoc Biol 71:597–602 Ren X, Zhou H, Li B et al. (2011) Toll-like receptor 3 ligand polyinosinic:polycytidylic acid enhances autoimmune disease in a retinal autoimmunity model. Int Immunopharmacol 11:769–73 Sato T, Yamamoto M, Shimosato T et al. (2010) Accelerated wound healing mediated by activation of Toll-like receptor 9. Wound Repair Regen 18:586–93 Schebesch C, Kodelja V, Muller C et al. (1997) Alternatively activated macrophages actively inhibit proliferation of peripheral blood lymphocytes and CD4+ T cells in vitro. Immunology 92:478–86 Schurmann C, Seitz O, Sader R et al. (2010) Role of wound macrophages in skin flap loss or survival in an experimental diabetes model. Br J Surg 97:1437–51 Sen CK, Gordillo GM, Roy S et al. (2009) Human skin wounds: a major and snowballing threat to public health and the economy. Wound Repair Regen 17:763–71 Sidhu GS, Thaloor D, Singh AK et al. (1996) Enhanced biosynthesis of extracellular matrix proteins and TGF-beta 1 by polyinosinic-polycytidylic acid during cutaneous wound healing in vivo. J Cell Physiol 169:108–14 Singer AJ, Clark RA (1999) Cutaneous wound healing. N Engl J Med 341: 738–46 Stein M, Keshav S, Harris N et al. (1992) Interleukin 4 potently enhances murine macrophage mannose receptor activity: a marker of alternative immunologic macrophage activation. J Exp Med 176:287–92 Tohyama M, Dai X, Sayama K et al. (2005) dsRNA-mediated innate immunity of epidermal keratinocytes. Biochem Biophys Res Commun 335:505–11 van Amerongen MJ, Harmsen MC, van Rooijen N et al. (2007) Macrophage depletion impairs wound healing and increases left ventricular remodeling after myocardial injury in mice. Am J Pathol 170:818–29 Werner S, Grose R (2003) Regulation of wound healing by growth factors and cytokines. Physiol Rev 83:835–70