Introduction

  Wound fluid has been implicated to play a critical role in the wound healing process.3,4 Acute wound fluid has been shown to stimulate the growth of fibroblasts and endothelial cells,5-7 induce chemotaxis,8 and increase production of extracellular matrix.6,7

In contrast, wound fluid from chronic venous ulcers (CWF) has been shown to inhibit cellular proliferation,9-11 contributing to the impaired healing of chronic ulcers. CWF inhibits the proliferation of newborn dermal fibroblasts,9-11 inhibits DNA synthesis in human neonatal fibroblasts,12 and arrests cells in the G1 phase of the cell cycle.9 A recent report has suggested that CWF-induced suppression of growth involves modulation of cell cycle-dependent proteins, in particular pRb, cyclin D1, CDK4, and p21Cip1/Waf1.1

  The tightly regulated eukaryotic cell cycle can be broadly divided into a S (DNA synthesis) phase and a M (mitosis) phase, with a gap phase before S phase termed G1, and a gap after M phase termed G213 (Figure 1). One of the most important regulators of G1 phase progression is the product of the retinoblastoma tumor-suppressor gene (Rb). When Rb is phosphorylated (pRb) by the cyclin D1/CDK4 kinase complex,14-16 it is then consequently inactivated, causing a cascade of events allowing passage of the cell through G1 into S phase.17 In recent years, p21Cip1/Waf1 was shown to inhibit activity of the cyclin D1/CDK4 complex,18 thus resulting in inhibition of progression through G1 into S phase. Therefore, p21Cip1/Waf1 is often referred to as a growth inhibitory protein.

  The growth inhibitory activity in CWF was shown to be heat sensitive in that when CWF was heated, there was a temperature-dependent reduction in the growth inhibitory activity.2 These results suggested that a thermal wound therapy might stimulate healing of chronic leg ulcers by counteracting the growth inhibitory activity in CWF. In this report, we examined whether a noncontact thermal wound therapy*, which was shown to be beneficial to chronic venous stasis ulcers,19 could counteract the growth inhibitory activity of CWF by blocking CWF-induced changes in the levels of cell cycle-regulatory proteins.

Materials and Methods
  Fibroblast isolation and culture. Dermal fibroblasts were cultured as previously described.20 Briefly, newborn foreskins were treated for five minutes in a povidone-iodine bath, followed by a five-minute 70-percent ethanol bath. The biopsies were then placed in a 1mg/mL trypsin solution overnight at 4 degrees C to dissociate the dermis from the epidermis and adipose tissues. The isolated dermal tissue was then cut into 1-2mm fragments and placed in etched plastic culture plates. Fibroblasts were cultured in Dulbecco’s modified Eagle’s medium (DMEM) with 10-percent bovine calf serum devoid of any antibiotic or antimycotic agents. Second or third passage cells were used in all the experiments.

  Collection of CWF. Chronic venous ulcer patients, recruited with the approval of the Institutional Review Board at Boston Medical Center, were collected as previously described.9 In brief, foam dressings were placed over the ulcer for 24 to 48 hours. CWF was then extracted from the foam dressings through sterile syringes with 20-gauge stainless steel needles. The collected CWF was immediately diluted (1:10) with DMEM containing the protease inhibitors aprotinin (2µg/mL) and phenylmethylsulfonyl fluoride (10µg/mL). Large debris was removed by centrifugation at 15,000 rpm for 15 minutes. The CWF was then stored at -70 degrees Celsius (C) until use. Just prior to addition to cell cultures, the CWF was filtered through a 0.4mm syringe filter to remove bacteria.

  Heating of CWF using noncontact thermal wound therapy. To evaluate whether a noncontact thermal wound therapy could suppress CWF-induced changes on the levels of cell cycle-regulatory proteins, a wound warming device was employed. Noncontact thermal wound therapy is composed of several parts as previously described:19 (1) a foam collar that adheres to the periwound; (2) a transparent film covering the top of the foam collar; (3) a pocket built into the film covering; (4) an infrared warming card that is inserted into the film pocket; and (5) a temperature control unit that controls the warming card (Figure 2). To warm CWF, sterile tissue culture plates containing CWF were placed on top of the pocket such that the infrared warming card was in contact with the bottom of the tissue culture plate. The foam collar was placed over the tissue culture plate.

  To determine the maximal temperature of CWF that can be reached using noncontact thermal wound therapy, CWF was exposed to noncontact thermal wound therapy for up to 72 hours, and the temperature of CWF was taken at the indicated time (Figure 3). The starting temperature of CWF was usually 25 degrees C, and the maximal temperature of 35 degrees C was reached within one hour of heating and remained unchanged for up to 72 hours. Therefore, CWF was heated using noncontact thermal wound therapy for one to two hours for all experiments.

  Protein extraction and immuno-blot analysis. Paired cultures of NbFb were treated with noncontact thermal wound therapy-heated CWF at a concentration of 250mg/mL, nonheated CWF or noncontact thermal wound therapy-heated Bovine serum albumin (BSA) (250mg/mL) for 24 hours. Cells were washed three times in ice-cold phosphate-buffered saline then harvested in a RIPA lysis buffer (0.25M Tris/HCl, 0.75M NaCl, 2.5% SDS, 0.1% Triton) containing phenylmethylsulfonyl fluoride (1uM), aprotinin (1uM), and okadaic acid (1uM) at 24 hours after treatment. Lysate samples were sonicated briefly and centrifuged at 14,000 rpm to get rid of the particulate fraction. Twenty milligrams of each lysate sample were resolved by SDS-polyacrylamide gel electrophoresis on five- to twelve-percent polyacrylamide gels and transferred onto PVDF membranes (Bio-Rad, Hercules, California). The membranes were blocked in five-percent nonfat milk and blotted with antibody against Rb (Santa Cruz, Santa Cruz, California), cyclin D1 (Pharmingen, San Diego, California), CDK4 (Santa Cruz, Santa Cruz, California), and p21Cip1/Waf1 (BD Transduction, Franklin Lakes, New Jersey). This was followed by incubation with the appropriate secondary antibodies conjugated to horseradish peroxidase. Protein bands were visualized using ECL detection system (Amersham, Piscataway, New Jersey).

Results
  Noncontact thermal wound therapy-heated CWF did not display growth inhibitory activity. CWF was shown to inhibit growth of dermal fibroblasts,9 and this growth inhibitory activity was shown to be heat sensitive.2 To determine if a thermal wound therapy would counteract or reduce the growth inhibitory activity, CWF was heated using a noncontact thermal wound therapy as described under Materials and Methods. Addition of BSA heated with noncontact thermal wound therapy and nonheated BSA showed a similar cell number per plate. Addition of nonheated CWF at the concentration of 250µg/mL of media to the dermal fibroblasts dramatically reduced the total cell number compared to that of the cells treated with nonheated BSA at the end of incubation period, consistent with the previous reports9 (Figure 4A). However, when CWF was warmed to 35 degrees C using noncontact thermal wound therapy and added to the cells, there was no inhibition of cellular proliferation. Total cell numbers at the end of the incubation period in cells treated with noncontact thermal wound therapy-heated CWF were similar to that of the cells treated with noncontact thermal wound therapy-heated BSA.

  Morphologically, cells treated with nonheated CWF displayed a granular appearance and were larger in size, as well as reduced number per square area (Figure 4B) as previously reported.11 However, the cells treated with noncontact thermal wound therapy-heated BSA or CWF displayed similar morphology in that they were spindle-shaped, more numerous per square area, and appeared healthy. Taken together, these results suggest that a thermal wound therapy can counteract the growth inhibitory activity in CWF.

  Effects of noncontact thermal wound therapy-heated CWF on the level of pRb. Rb plays a critical role during the G1 to S phase progression of cell cycle. CWF was previously shown to reduce the level of phosphorylated Rb (pRb) in NbFb, thus blocking progression of G1 into S phase. We, therefore, investigated whether heating of CWF using noncontact thermal wound therapy would block the effect of CWF on the level of pRb. When NbFb was exposed to nonheated CWF for 24 hours, the level of pRb was reduced to an undetectable level, as compared to that of the BSA-treated cells (Figure 5A) consistent with results previously reported.1 However, in cells treated with heated CWF, the level of pRb remained unchanged, indicating that thermal wound therapy can block reduction of pRb level by CWF. The level of unphosphorylated Rb remained unaffected in all conditions.

  Effects of noncontact thermal wound therapy-heated CWF on the levels of cyclin D1 and CDK4. The cyclin D1/CDK4 complex is known to phosphorylate Rb. CWF was shown to reduce the level of cyclin D1, resulting in reduced level of pRb. The level of CDK4 was shown to be unaffected by CWF.1 In cells treated with noncontact thermal wound therapy-heated CWF, the level of cyclin D1 stayed unchanged and was similar to that of the control (panel) cells. In contrast, cells treated with nonheated CWF showed a reduction in the level of cyclin D1 (Figure 5B). The level of CDK4 remained unchanged in all conditions (Figure 5B).

  Effects of noncontact thermal wound therapy-treated CWF on p21Cip1/Waf1. The activity of the cyclin D1/CDK4 complex can also be suppressed by p21Cip1/Waf1. To further explore the ability of thermal wound therapy to counteract the effects of CWF on Rb phosphorylation, the level of p21Cip1/Waf1 was determined in parallel, using the lysates analyzed for cyclin D1/CDK4 content. Immunoblot analyses using monoclonal antibody specific for p21Cip1/Waf1 show that CWF reduced the level of p21Cip1/Waf1 as compared to BSA treatment (lane 1), and this was inhibited by warming the CWF with noncontact thermal wound therapy (Figure 5C).

Discussion CWF has been shown to specifically inhibit proliferation of dermal fibroblasts9 and endothelial cells,10 thus retarding the healing process. However, the factor(s) present in CWF responsible for the observed inhibition of cellular growth is not well understood. The growth inhibitory factor(s) in CWF appears to suppress the growth of fibroblasts, in part, by blocking the progression of the cell cycle from G1 to S phase. When CWF was added to dermal fibroblasts, the levels of pRb and cyclin D1, cell cycle-regulatory proteins critical for progression of G1 to S phase, are dramatically reduced.1 Moreover, cell-cycle analysis demonstrated that CWF-treated dermal fibroblasts growth arrested mostly in the G1 phase.9

  Results presented here indicate that a noncontact thermal wound therapy can indeed counteract growth inhibitory activity in CWF. Heating of CWF in vitro with a thermal wound therapy allowed normal proliferation and morphology of dermal fibroblasts. The maximal temperature of CWF reached by heating CWF with noncontact thermal wound therapy for 72 hours was 35 degrees C, a temperature well below the normal body temperature of 37 degrees C. In this study, CWF was heated in vitro; therefore, it is important to confirm the results by warming the ulcer in vivo, collecting the CWF, and examining how heated CWF in vivo affects the growth of dermal fibroblasts. Wound dressings may also elevate the temperature of CWF in vivo by the insulation effect. However, provision of constant heat would be more efficient in raising and maintaining the temperature of CWF.

  We have previously shown that the growth inhibitory activity in CWF is heat sensitive.2,21 When CWF was exposed to heat, there was a temperature-dependent reduction in the growth inhibitory activity.21 Therefore, the factors that are responsible for the growth inhibitory activity in CWF may be small peptides, such as cytokines or growth inhibitory molecules. Some of the known cytokines, such as TNF-a and TGF-b, were shown to affect proliferation of many cell types and to be involved in wound healing.22,23 TNF-a and TGF-b were shown to be present in CWF,11 suggesting that either one of them or both may be responsible for the growth inhibitory activity of CWF. However, when CWF was reacted with antibodies against TNF-a and TGF-b, which neutralize the activities of these cytokines, and added to dermal fibroblasts, CWF continued to inhibit growth of dermal fibroblasts,11 suggesting that TNF-a and TGF-b are not responsible for the growth inhibitory activity present in CWF.

  Heat-sensitivity of growth inhibitory activity in CWF suggests that a thermal wound therapy that warms the wound fluid may be beneficial in treating leg ulcers. Warming of wound fluid in chronic leg ulcers would counteract the growth inhibitory activity of CWF, allowing normal cellular proliferation in the wound. Our results presented here show that warming of CWF using noncontact thermal wound therapy blocks the CWF-induced suppression of Rb phosphorylation. This was achieved, in part, by sustaining the level of cyclin D1/CDK4 complex that phosphorylates Rb. In addition, warming CWF also blocked CWF-induced increases in the growth inhibitory protein p21Cip1/Waf1. Because p21Cip1/Waf1 prevents cyclin D1/CDK4 complex-mediated phosphorylation of Rb, a decreased level of p21Cip1/Waf1 in cells treated with heated CWF would result in the normal level of pRb, thus allowing proper progression of cells through G1 into S phase during proliferation of dermal fibroblasts.

  Therefore, a noncontact thermal therapy would prevent CWF-induced inhibition of the growth of dermal fibroblasts, resulting in enhanced wound healing. Moreover, our results suggest that the enhanced healing by a thermal wound therapy is not due to a general nonspecific stimulation of fibroblast growth, but it is mediated through specific positive modulations on the levels of cell cycle-regulatory proteins. The maintenance of critical cell cycle-regulatory proteins, such as pRb and cyclin D1/CDK4 complex, may be critical for the proper regeneration and healing of the wounds.

Acknowledgements
  We would like to thank Ysabel Bello, MD, Adriana Rojas, MD, and Jasmine Manzoor, MD, for their help in collecting wound fluid samples for this project.

  *Warm-Up® Wound Therapy, Augustine Medical, Inc., Eden Prairie, Minnesota