Physiological Levels of 5α-Dihydrotestosterone Depress Wound Immune Function and Impair Wound Healing Following Trauma-Hemorrhage

Hypothesis  Studies indicate that a depressed wound immune function contributes to an increased rate of wound complications and impaired wound healing following trauma-hemorrhage (T-H). Androgen, ie, 5α-dihydrotestosterone, is responsible for producing the depressed systemic cell-mediated immune responses following T-H in males. The aim of the present study was to determine whether depletion of 5α-dihydrotestosterone in males before T-H has any salutary effects on wound immune cell function and wound healing in male mice following T-H.

Design  Mice were castrated or sham castrated 14 days before midline laparotomy (ie, tissue trauma) and subcutaneous polyvinyl sponge implantation, followed by hemorrhage (mean ± SEM blood pressure, 35 ± 5 mm Hg for 90 minutes and resuscitation) or sham operation. At 24 hours thereafter, wound immune cells from the sponges were harvested and cultured with lipopolysaccharide A. Release of interleukin 1β (IL-1β) and IL-6 (in picograms per milliliter) was determined in the supernatants by enzyme-linked immunosorbent assay. In addition, IL-6 was assessed at the wound site by immunohistochemistry. Ten days after T-H, wound-breaking strength was measured.

Results  Precastration prevented the significantly suppressed capacity of wound immune cells to release IL-1β and IL-6. In addition, precastration normalized the elevated IL-6 expression at the wound site in the T-H mice. Moreover, wound-breaking strength was improved in castrated mice 10 days after T-H.

Conclusions  Male sex steroids appear to be responsible for wound immune cell dysfunction following trauma and severe blood loss. Because decreasing androgen levels resulted in improved wound healing, our results suggest that the use of androgen receptor–blocking agents, eg, flutamide, following T-H might represent a novel adjunct for decreasing the rate of wound complications under those conditions.

Recent studies^1,2 indicate that trauma and hemorrhagic shock lead to profound alterations in wound immune cell function. These changes persist for up to 3 days after resuscitation and result in an increased rate of wound complications.^1-3 An increased inflammatory response at the wound site has been postulated to be involved in the immunosuppression following severe blood loss.² Similarly, systemic cell-mediated immune responses are suppressed in male mice following trauma and hemorrhagic (T-H) shock.^4,5 In this regard, androgen has been shown to be responsible for producing this immunosuppression in male mice.^4,5 Support for this notion comes from studies^4,6 that showed that depletion of male sex steroids by castration 2 weeks before T-H prevented the depression of immune responses typically observed in normal males under those conditions. Similarly, administration of an androgen receptor blocker, eg, flutamide, restored depressed immune responses and increased the survival rate of hemorrhaged animals subjected to subsequent sepsis.^7,8 Conversely, administration of physiological levels of androgen in female mice resulted in depressed splenic and peritoneal macrophage immune response following T-H to levels comparable to those of males under those conditions.^9,10 Therefore, physiological levels of androgen suppress systemic cell-mediated immune responses in males following T-H.

Because salutary effects of androgen depletion on systemic cell-mediated immune responses have been demonstrated following hemorrhage, we hypothesized that depleting androgens in mice by castration 2 weeks before T-H might improve depressed local wound immune cell function by decreasing the local release of inflammatory cytokines at the wound site. In addition, it was the aim of the present study to determine whether any salutary effects of androgen depletion in male mice result in improved wound healing following T-H.

Inbred male C3H/HeN mice (Charles River, Sulzfeld, Germany) aged 5 to 7 weeks were used. All procedures were carried out in accord with the guidelines set forth in the Animal Welfare Act and the Guide for the Care and Use of Laboratory Animals by the National Institutes of Health, Bethesda, Md. The Institutional Animal Care and Use Committee of the Regierung von Oberbayern, Munich, and Ludwig-Maximilians University approved this study.

Mice were randomized into 4 groups (6-8 per group). The mice in groups 1 and 2 underwent sham operation alone (sham animals), while the mice in groups 3 and 4 were subjected to the T-H procedure. Groups 2 and 4 were castrated 2 weeks before the experiment, as previously described.^5,7

Mice were lightly anesthetized with methoxyflurane (Metofane; Pitman-Moore, Mundelein, Ill), they were restrained in a supine position, and the skin was disinfected using 75% ethanol. A 2.5-cm midline laparotomy was performed (ie, trauma induced), and the muscular layer was then closed aseptically using 6-0 sutures (Ethilon; Ethicon, Inc, Somerville, NJ). Following this, polyvinyl sponges (0.5 cm in diameter) (M-PACT Worldwide, LLC, Eudora, Kan) were aseptically implanted subcutaneously (3 per animal) next to the incision site, avoiding contamination or infection of the wound site. The skin incision was closed using 4-0 sutures.

Following the midline laparotomy and sponge implantation, both femoral arteries were aseptically cannulated with polyethylene 10 tubing (Clay-Adams, Parsippany, NJ) using a minimal dissection technique. Blood pressure was constantly monitored by attaching one of the catheters to a blood pressure analyzer (Digi-Med, Louisville, Ky). On awakening, the animals were bled rapidly through the other catheter to a mean ± SEM arterial blood pressure of 35 ± 5 mm Hg (blood pressure before hemorrhage was 95 ± 5 mm Hg), which was maintained for 90 minutes. At the end of that period, the animals were resuscitated intra-arterially with lactated Ringer solution (4 times the shed blood volume for 30 minutes) to provide adequate fluid resuscitation. Lidocaine hydrochloride was applied to the groin incision sites, the catheters removed, the vessels ligated, and the groin incisions closed. Sham-operated animals underwent the same groin dissection, which included ligation of both femoral arteries; however, neither hemorrhage nor fluid resuscitation was carried out. No mortality was observed in this model of T-H.

The animals were humanely killed by methoxyflurane overdose 24 hours after T-H and resuscitation or sham operation, and sponges were harvested by dissecting them free from the surrounding connective tissue using sterile surgical methods. In addition, wound skin was harvested and placed in 4% paraformaldehyde. Microscopic sections were obtained and stained for interleukin (IL) 6 by immunohistochemistry.

The sponges were transferred into a plastic bag with ice-cold phosphate-buffered saline solution. The sponges were repeatedly compressed, and the resultant cell suspension was centrifuged at 300g for 15 minutes at 4°C. The cell pellet of wound immune cells was diluted to 2 × 10⁶ cells/mL in RPMI 1640 medium containing 10% fetal calf serum and gentamicin sulfate. One milliliter of this cell suspension was cultured on a 24-well plate for 24 hours at 37°C, under conditions of 5% carbon dioxide and 90% humidity, in the presence of 10 µg/mL of lipopolysaccharide A. After incubation, the cell suspension was centrifuged at 300g for 15 minutes at 4°C, and the supernatants were harvested and stored at −80°C until assayed for IL-1β and IL-6 levels.

Interleukin 1β and IL-6 levels in the splenocyte supernatants were determined using the enzyme-linked immunosorbent assay technique described by Mosmann et al.¹¹ In brief, 96-well plates (Nunc-Immuno; MaxiSorp, Roskilde, Denmark) were coated overnight with 4 µg of rat anti-mouse IL-1β or IL-6 capture antibody (clone JES-5; Pharmingen, San Diego, Calif) per milliliter of 0.1M sodium hydrogen carbonate, pH 8.2. The plates were washed 3 times with phosphate-buffered saline containing 0.05% Tween-20 (Sigma-Aldrich Co, St Louis, Mo) and blocked with phosphate-buffered saline containing 20% fetal calf serum for 2 hours. Following washing, 100 µL of the samples and standard (10 ng/mL of murine IL-1β or IL-6 [Pharmingen]) was added to the plate and then incubated overnight (at 4°C). After repeated washings, the plates were incubated for 1 hour with 100 µL of biotinylated monoclonal rat anti-mouse IL-1β or IL-6 (clone SXC-1, Pharmingen) at a concentration of 2 µg/mL at room temperature. Following washing, the plates were incubated at room temperature for 30 minutes with avidin peroxidase (diluted 1:400, Sigma-Aldrich Co). The plates were then washed, and 100 µL of 2,2′-azino-di-(3-ethylbenzthiazoline)-6 sulfonic acid hydrogen peroxide substrate buffer was added to each well to initiate color development. The optical density at 405 nm for each well was determined on a microplate reader. The concentration of IL-1β or IL-6 present in the samples was determined by interpolation using a standard curve produced with murine IL-1β or IL-6.

Sections of the wound site were fixed in 4% paraformaldehyde. Paraffin blocks of specimens were cut at 1 to 2 µm thickness. The specimens were stained using avidin-biotin complex. The sections were then incubated with primary polyclonal goat IL-6 anti-mouse antibody (R&D Systems Inc, Minneapolis, Minn) at a dilution of 1:200 for 1 hour at room temperature. Following this, a biotinylated anti-goat antibody (Dianova, Hamburg, Germany) was added at a dilution of 1:4000 for 30 minutes at room temperature. The sections were then incubated at room temperature for 30 minutes with a streptavidin peroxidase–conjugated antibody at a dilution of 1:200. To initiate color development, 3′-amino-9′-ethyl carbazole (Sigma-Aldrich Co) was added for 8 minutes. The slides were evaluated double-blindedly by 2 investigators (S.M.N. and M.K.A.).

In a separate group of animals, the mice were humanely killed on the 10th postoperative day by methoxyflurane overdose for measurement of wound-breaking strength. Ink tattoo lines were placed 2 cm apart symmetrically to the incision (laparotomy wound or back incision), and skin specimens were harvested. Using a specifically designed cutting device, specimens were cut into sections that were at least 3 cm long and exactly 1 cm wide, with a variable thickness of approximately 2 mm. The specimens were transported on ice to the test room. Each specimen was secured in the Machine Tool Services (Munich, Germany) machine using cryoclamps cooled to approximately −40°C (dry ice). The clamps were tightened on the tattoo lines while the specimen was completely loose. Once secured, the specimens were stretched to a preload of 0.5 N, which returned the specimen gauge length to approximately 2 cm. Each specimen was stretched, and displacement-controlled testing was carried out at a rate of 2 mm per second to a maximum of 30 mm. Load and displacement data were collected, and a graph was printed for each test. Maximal-breaking strength as determined by the maximal failure load was calculated. Moreover, the stiffness of the back skin was calculated.

Results are presented as mean ± SEM. One-way analysis of variance, followed by the Newman-Keuls test or Tukey test as a post hoc test for multiple comparisons, was used to determine the significance of the differences between experimental means. P<.05 was considered significant.

The number of wound immune cells and the percentage of macrophages and polymorphonuclear leukocytes were similar in T-H and sham animals (Table 1). Moreover, castration before sham operation or T-H did not affect the number of wound immune cells or the distribution of polymorphonuclear leukocytes and macrophages.

Hemorrhage did not affect the plasma levels of 5α-dihydrotestosterone (DHT) in intact male mice: sham animals, 2.10 ± 0.20 ng/mL; sham castrated, 0.10 ± 0.06 ng/mL; T-H animals, 1.70 ± 0.30 ng/mL; and T-H castrated, 0.08 ± 0.05 ng/mL. Castration, however, resulted in markedly decreased plasma levels of DHT in sham-operated and castrated mice. Trauma-hemorrhage had no effect on plasma DHT levels in castrated mice.

Castration of mice did not affect the capacity of wound immune cells to release IL-1β and IL-6 in response to in vitro lipopolysaccharide A stimulation following sham operation (Figure 1). Following T-H, significantly suppressed IL-1β and IL-6 release by wound immune cells was evident in sham-castrated mice. In contrast, castrated mice displayed maintained IL-1β and IL-6 release capacities by wound immune cells following T-H.

The expression at the wound site as measured immunohistochemically was similar following sham operation in castrated and noncastrated animals at the wound site (Figure 2). Following T-H, the expression of IL-6 markedly increased at the wound site. Depletion of DHT by castration attenuated the increased IL-6 expression at the wound site following T-H.

In a separate set of animals, wound skin sections were harvested 10 days after T-H or sham operation. The typical load displacement curves are shown in Figure 3. Six animals per group were analyzed. An evaluation of the maximal breaking strength is shown in Figure 4.

The typical load displacement curve of the abdominal wall shows a double peak: the first one represents the breakage of the skin and the second peak represents the muscular layer (Figure 3).

The maximal breaking strength (failure load) of the abdominal wound was significantly decreased 10 days following T-H (Figure 4). Depletion of DHT before T-H restored the depressed maximal breaking strength following T-H to a level comparable to that of sham-operated mice.

Several clinical and experimental studies^12-15 demonstrate impaired wound healing in the presence of concurrent trauma, ie, fracture, and thermal injury. In this respect, the nonhealing wound becomes a potential source of considerable mortality and morbidity as a result of subsequent tissue inflammation and infection, the latter being especially common in the presence of compromised immune functions and altered wound healing in critically ill patients.^16,17 Tissue repair has been shown to require a sequence of synchronized events in which multiple cell types interact successfully to restore tissue integrity.¹⁸ These events include the release of cytokines, growth factors, and other bioactive molecules.¹⁷ In this respect, impaired wound immune cell function has been found following severe blood loss.¹ Furthermore, this impairment has been associated with an increased release of proinflammatory cytokines and a decreased production of transforming growth factor β at the wound site.¹ Studies^5,9,10 indicate that DHT is responsible for producing the depressed systemic immune responses following severe blood loss in male mice. However, although it is tempting to speculate that DHT depletion improves local wound immune cell function and results in improved wound healing, this has not yet been examined. To study this, mice were castrated 2 weeks before T-H or sham operation. Similar to previous findings, castration resulted in significantly suppressed DHT plasma levels.^4,19 In addition, T-H did not alter plasma sex steroid levels. The androgen DHT was measured in this study because it is considered to be the primary androgenic hormone in males.

The results of the present study indicate suppressed cytokine release capacities by wound immune cells harvested from intact male mice following T-H. This is in agreement with previous findings demonstrating altered cytokine release capacities for up to 3 days following T-H.¹ Depletion of androgen by castration prevented the depression in IL-1β and IL-6 release capacity by wound immune cells. Because the percentage of macrophages and polymorphonuclear leukocytes within the wound immune cells was unaffected by T-H or castration, the observed improvement in cytokine release in castrated animals does not appear to be due to variations in the distribution of wound immune cells. Interestingly, castration of mice did not affect cytokine release capacities of wound immune cells in sham mice. Similarly, depletion of DHT by castration restored the depressed cytokine release capacity of splenic and peritoneal macrophages following T-H, whereas cell-mediated immune response in sham mice was not affected.^4,19 These findings suggest that physiological levels of testosterone are only harmful in an immunologically compromised host, not in normal animals. In this respect, steroid synthesis has been shown to be altered following hemorrhage.^20,21 Those studies demonstrate an increased DHT synthesis and a decreased catabolism of this steroid hormone following T-H due to changes in the activity of enzymes involved in steroid synthesis. In particular, 5α-reductase activity was increased, whereas 17β-hydroxysteroid dehydrogenase activity decreased following T-H in lymphocytes and splenic macrophages harvested from male mice. Although steroid synthesis pathways in wound cells following hemorrhage have not been studied, those findings suggest that an enhanced intracellular DHT synthesis following T-H might explain the immunodepressive effects of physiological DHT plasma levels. Similar to our findings, studies^22-25 demonstrate that androgens do not depress macrophage cytokine release in normal animals. For instance, studies by Ahmed et al²⁶ showed that administration of testosterone at physiological levels in normal C57BL/6 mice for 2 to 4 weeks did not alter splenocyte IL-2 release.

Although susceptibility to wound infection has not been determined in the present study, previous findings suggest that normalization of wound immune cell function is associated with a decreased rate of wound infections.¹ In this respect, suppressed wound immune cell cytokine release capacities following hemorrhage correlated with increased rates of wound infection.¹ Moreover, normalization of suppressed cell-mediated immune responses by androgen depletion decreased the susceptibility to subsequent polymicrobial sepsis.⁷ Those studies support the hypothesis that restoration of wound immune cell function following T-H in precastrated mice decreases the susceptibility to wound infection.

In contrast to the depressed in vitro release of proinflammatory cytokines by wound immune cells following hemorrhage, the expression of IL-6 was increased at the wound site under those conditions. Similarly, the expression of IL-6 is increased in splenic macrophages following hemorrhage, whereas the in vitro response of splenic macrophages to lipopolysaccharide A is suppressed under those conditions.²⁷ Conversely, depletion of testosterone resulted in an attenuation of the increased inflammatory response at the wound site and restoration of the in vitro release capacity. The excessive inflammatory response in vivo appears to be responsible for the exhaustion of wound immune cells to respond to a second stimulus, ie, lipopolysaccharide A, in vitro. Further support for this notion comes from previous findings demonstrating an increased proinflammatory cytokine response in the wound fluid of hemorrhaged mice.² The proinflammatory cytokine IL-6 has been proposed to be one of the major contributors to the inhibitory effects of wound fluid on fibroblast proliferation.²⁸ Therefore, androgen depletion might restore the depressed wound immune cell function following T-H by decreasing the inflammatory immune response at the wound site. This hypothesis, however, needs to be proven in further studies.

The results of the present study further demonstrate decreased breaking strength of the excised wound specimen on the 10th postoperative day following T-H. Castration of mice 2 weeks before T-H, however, attenuated the depressed wound-breaking strength on the 10th postoperative day. Although the decrease in wound-breaking strength in hemorrhaged intact male mice was statistically significant, the change was modest. Therefore, it remains unknown whether the diminished breaking strength following hemorrhagic shock is of biological significance, in particular because the number of animals used in the present study was small. Previously, similar breaking-strength measurements 10 days after sham operation or hemorrhagic shock in intact male mice have been shown.²⁹ This demonstrates that the breaking-strength measurements are consistent and reproducible. The present findings indicate, however, that normalization of the depressed wound immune cell function and attenuation of the increased inflammatory response at the wound site are associated with restoration of the depressed wound-breaking strength following T-H. In this regard, studies by Salomon et al³⁰ indicate that topical administration of the proinflammatory cytokine tumor necrosis factor α decreased mechanical strength in incisional wounds in rats. Conversely, decreased levels of proinflammatory cytokines in the wound fluid of endotoxin-resistant mice were associated with increased wound-breaking strength within the first 7 days after wounding.³¹ This suggests that androgens contribute to the impaired wound healing in male mice following hemorrhage by increasing the inflammatory response at the wound site and diminishing wound immune cell function. Because the number of wound immune cells was unaltered in hemorrhaged mice, the diminished wound healing is not a result of a decreased number of wound immune cells at the wound site.

Wound-strength measurements in the present study were limited to a single time point. Therefore, our findings do not allow us to distinguish whether impairment of wound healing following severe blood loss and its improvement in androgen-depleted animals are a result of changes in wound outcome or alterations in the wound-healing process. Further studies investigating wound-breaking strength at various time points following hemorrhage in castrated and intact male mice will be conducted to address this issue. Studies by Levenson et al³² demonstrate an abnormal histological appearance of the wound at 7 days in the presence of concurrent burn injury, while at 14 days the wound appeared normal in both groups. Similarly, wounded animals subjected to polymicrobial sepsis exhibited decreased breaking strength of the wound for up to 14 days.³³ At 3 to 5 weeks, however, wounds from septic and nonseptic animals were indistinguishable in terms of breaking strength.³³ Therefore, the wound-healing process following severe blood loss might attain a normal-healing process at a later time point in noncastrated mice. Nonetheless, prevention of diminished wound healing early following T-H may reduce comorbid events with potentially catastrophic effects for trauma patients (ie, increased infections).

In summary, our results indicate that wound immune cell function and maximal breaking strength of the wound were impaired following T-H shock in intact male mice. Castration of mice 2 weeks before T-H prevented wound immune cell dysfunction and improved maximal breaking strength following T-H. In addition, the increased expression of IL-6 at the wound site was attenuated by androgen depletion. Therefore, depletion of androgen restores wound immune cell function and improves wound healing, potentially by decreasing IL-6 release at the wound site. Although the effect of an androgen receptor blocker on wound immune cell function has not been evaluated, previous findings suggest that short-term treatment with flutamide following T-H might similarly beneficially affect local wound immune cell function.^7,8 Therefore, attempts to improve wound immune cell function at the wound site by treatment with androgen receptor blockers might represent a useful adjunct for improving wound healing and decreasing the incidence of wound infections in trauma patients.

Box Section Ref ID

Corresponding author: Martin K. Angele, MD, Department of Surgery, Klinikum Grosshadern, Ludwig-Maximilians University, Marchioninistrasse 15, 81377 Munich, Germany (e-mail: mangele@aol.com).

Accepted for publication July 12, 2003.