Cutaneous induction of corticotropin releasing hormone by Propionibacterium acnes extracts

Abstract

The skin commensal bacillus Propionibacterium acnes is known to play a major role in the development of acne vulgaris and it is established that this bacteria is involved both in the induction and maintenance of the inflammatory phase of acne. The corticotropin releasing hormone (CRH), a neuropeptide originally isolated from the hypothalamus, is also produced by the skin. CRH has been reported to play a role in the inflammation, the production of sebum and finally the differentiation of keratinocytes. At the therapeutic level, zinc is known to act specifically on inflammatory lesions with still partially known mechanisms and thus could play an important role in the development of inflammatory acne lesions. Our objective was to study the modulation of CRH expression by keratinocytes induced by P. acnes extracts. CRH expression was examined using immunohistochemistry technique on deep-frozen sections of normal human skin explants incubated with two different extracts of P. acnes and with or without zinc salts. We observed that the membrane fraction (FM) of P. acnes increased the CRH expression in the epidermis. This result indicates that P. acnes, by stimulating the production of CRH, can both modulate the differentiation of keratinocytes and increase the local inflammation, arguing that this bacterium plays a role not only in the development of inflammatory acne lesions but also in the formation of the microcomedo in the early stages of acne.

Introduction

Acne vulgaris, the most common pathology of the skin, is a chronic inflammatory disease of the pilosebaceous unit (PSU). Several pathogenic factors such as ductal hypercornification, increased sebum production, abnormality of the microbial flora within the PSU and secretion of mediators of inflammation contribute to the aetiology of this multifactorial disease.¹ Although P. acnes has been associated with acne, the precise mechanisms governing the development and progression of acne remain unclear.

In response to systemic stress, the hypothalamic-pituitary-adrenal (HPA) axis is activated.² The process begins with the hypothalamic release of corticotropin releasing hormone (CRH) which stimulates the production of adrenocorticotropic hormone (ACTH) and other proopiomelanocortin (POMC) peptides via the activation of CRH receptor type 1 (CRH-R1). It was reported that ACTH is also produced outside the pituitary tissue, including the skin.^3–6 Interestingly it has been proposed that skin has an equivalent of the central HPA axis.^5,7–10 There is increasing evidence that the cutaneous nervous system modulates physiologic and pathophysiologic effects in the skin.¹¹ Although similar to its systemic equivalent, the cutaneous HPA axis is responsive to local stressors (solar, thermal, chemical, biological, etc.,) and resultantly activates the neuronal, endocrine and immune systems in the skin.

As acne is obviously exacerbated under acute or chronic psychological stress^11–13 the role of CRH in this pathology appears as an important point in the development of lesions. In humans, CRH is synthesized among others by keratinocytes,¹⁴ immune cells¹⁵ and human mast cells^16,17 under the influence of a stress. Interestingly, CRH is also reported to play a role in the regulation of keratinocytes proliferation and differentiation representing an important step in the early stages of the development of acne lesions. Moreover CRH is known to act on inflammation by inducing the degranulation of mast cells, the release of inflammatory cytokines and the modulation of immune cells.^18–20 And finally CRH acts also on sebum production by promoting lipogenesis in human sebocytes and thus by increasing the seborrhoea.^21,22 Consequently, CRH could be strongly implicated in the development of acne.

Furthermore, several studies have underlined an important role of P. acnes in the pathophysiology of inflammatory lesions.^23–29

In this context, we hypothesized that P. acnes is able to increase CRH expression by keratinocytes, representing a local stress in acne. We thus evaluated the modulation of CRH expression in the epidermis induced by the microbial stressor P. acnes.

Results

In all cases, we observed that the mean expression of CRH was moderate (1.25 ± 0.50) in control medium and was significantly increased (p < 0.05) in presence of P. acnes FM (3.50 ± 1.00). However the slight overexpression detected in the presence of SA (1.75 ± 0.96) was not significant (p > 0.05). This increased expression of CRH was observed after an incubation of three hours with P. acnes extracts.

Concerning LPS which is a proinflammatory substance, a slight increase (1.88 ± 0.25) in the expression of CRH was noted while zinc gluconate decreased this expression (0.50 ± 0.50) in comparison with the control medium, for all donors. However the modulations observed were not significant (Figs. 1 and 2).

Figure 1.

Evaluation of the expression of corticotrophin-releasing hormone in the epidermis of 4 donors after an incubation of 3 hours in the presence of Medium (Ctrl), SA (P. acnes supernatant A), FM (P. acnes membrane fraction), LPS (E. coli Lipopolysaccharide), Zn (Zinc Gluconate) (scale: null labelling (0), very weak labelling (1), weak labelling (2), moderate labelling (3), strong labelling (4) and very strong labelling (5). (***p value < 0.05).

Figure 2.

Detection of the expression of corticotrophin-releasing hormone in the skin after an incubation of three hours in the presence of Medium (Ctrl) (A), SA (P. acnes supernatant A) (B), FM (P. acnes membrane fraction) (C), LPS (E. coli Lipopolysaccharide) (D), Zn (Zinc Gluconate) (E). (Magnification ×40).

Discussion

In the present work, we demonstrate that P. acnes stimulates the production of CRH by keratinocytes. In addition we observe that zinc gluconate used in inflammatory acne decreases the epidermal CRH production.

Concerning P. acnes extracts, the stimulating effect is specifically obtained with FM and not with SA. While FM contains P. acnes membrane components, in particular peptidoglycan and lipoteichoic acid, Supernatant A (SA) contains only cytosolic proteins. Our result thus confirms that the CRH overexpression is mediated by at least one component of FM, suggesting that P. acnes acts on the CRH production through a direct contact with keratinocytes membranes. Notably, the induction of CRH in the presence of FM is more important than this obtained in the presence of LPS which is considered as a reference among proinflammatory substances.

It is widely accepted that inflammatory acne may be mainly mediated by different factors secreted by P. acnes, such as lipases which induce the liberation of proinflammatory free fatty acids from sebum.^30,31 Recently, it has also been described that P. acnes triggers anti-microbial peptide and cytokine secretion of keratinocytes in vitro.^23–25 P. acnes, itself, is able to induce cytokines-like such as TNFα, IL-1α and IL-8.^26,27 Moreover the genome sequence of P. acnes is known to encode many factors that may have inflammatory potential.²⁸ But, recently it has also been shown that P. acnes was able to induce keratinocytes proliferation and filaggrin expression by keratinocytes,²⁹ indicating that this bacterium could play a role not only in the development of inflammation in acne lesions but also in the formation of the comedo.

The role of zinc salts has already been reported in several studies and beneficial effects were described on inflammatory lesions in mild to moderate acne. It has been shown until now that its anti-inflammatory activity has different targets.³² Indeed zinc acts via an inhibition of polynuclear cells chemotaxis.³³ The anti-inflammatory effects of zinc could be attributed to its ability to decrease TNFα and IL-6 production^34,35 or to inhibit TLR-2 expression by keratinocytes.³⁶ The modulation of integrins expression by zinc is also described.^37,38 At the same time, zinc is known to have a direct influence on P. acnes development by its antimicrobial activities³⁹ and its ability to reduce the resistance of P. acnes strains to erythromycin.⁴⁰ Finally, zinc is also able to modulate the 5α-reductase, a key enzyme implicated in the transformation of testosterone leading to an exacerbated seborrhoea.⁴¹ By these many actions, zinc is able to improve inflammatory acne. In this study, we suggest another mechanism by which zinc salts could be able to improve acne by decreasing the epidermal CRH expression. The fact that the modulation observed with zinc salts was not significant can be mainly due to the low level of basal CRH production in skin control but interestingly the decreased expression of CRH was noted for all the four donors studied. Moreover, the model used was very close to in vivo. In this context, zinc salts could represent a potential new way to counteract acne by targeting CRH.

The potential interest of CRH in acne was recently suggested by the discovery of an overexpression of CRH in the PSU.²¹ Indeed, CRH has been reported to play several roles potentially strongly implicated in the development of acne lesions. First, as CRH was reported to promote lipogenesis in human sebocytes and thus to increase the seborrhoea, it could represent a key factor in the development of acne.^21,22 Moreover, CRH is also known to act as a growth factor in the skin by activating the CRH-R1. CRH is thus described as an activator of keratinocytes differentiation.⁴² It is also an inhibitor of the early and late apoptosis of many skin cell types such as keratinocytes, dermal fibroblasts and melanocytes.⁴³ Finally CRH has also a pro inflammatory activity.¹⁸ Interestingly, it was suggested that CRH could act as a local endocrine mediator that enhances inflammatory responses to bacterial antigens.⁴⁴

In this study, we demonstrate for the first time that P. acnes acts as a strong local stressor at the origin of CRH overexpression by keratinocytes, in the epidermis. Thus, as CRH modifies the differentiation of keratinocytes, we confirm that this bacterium has a spectrum of action, not limited to inflammation but also in the early stages of the formation of acne lesions.²⁹ This confirms the fundamental role played by this bacterium which can be detected as a local stressor by the skin and could potentially lead to the development of acne lesions.

In conclusion, CRH represents a new target for P. acnes in the formation of both retentional and inflammatory lesions. Zinc gluconate seems to be able to decrease the production of CRH by keratinocytes, thus explaining its role in the improvement of acne. But this point has to be confirmed by further studies.

Materials and Methods

Bacterial extracts.

Two extracts of P. acnes IP53113T (Pierre Fabre, Toulouse, France) were made available to us. The strain was first described in 1968. The membrane fraction (FM) contained peptidoglycan and lipoteichoic acid. Supernatant A (SA) contained cytosolic proteins. The membrane fractions of the bacteria were resuspended in DMEM (Dulbecco’s Modified Eagle Medium) (Sigma-Aldrich, Saint-Quentin Fallavier, France).

LPS extracted from E. coli 0111:B4 (Sigma, St. Louis, USA), was reconstituted in PBS (Phosphate-Bufferd Saline). As LPS is a pro inflammatory substance, it was used as positive control of the inflammation. LPS was diluted in DMEM (Dulbecco’s Modified Eagle Medium) (Sigma-Aldrich) medium with a final concentration of 1 µg/mL and incubated with cutaneous explants.

Trace element.

As zinc salts are an anti inflammatory treatment of acne lesions it was used as negative control. Zinc gluconate (Labcatal, Montrouge, France), was diluted in DMEM with a final zinc concentration of 1 µg/mL and incubated with cutaneous explants.

Skin explants technique.

Punches (4 mm in diameter) from abdominal skin of 6 healthy donors, considered as an healthy skin model, were incubated at 37°C in a moist atmosphere in the presence of 5% CO₂ for 3, 6 or 24 hours in DMEM. The medium contained P. acnes extracts at the following concentrations: FM ½ or SA 1/5 or Zn 1 µg/mL. Medium alone was used as a control. After incubation of 3, 6 or 24 h, explants were removed from the culture medium and frozen at −80°C.

Immunoperoxidase.

Sections (5 µm thick) were then cut with a cryostat, fixed in acetone at 4°C for 10 min and frozen at −20°C. The non-specific sites were saturated for 30 mn with TBS (Tris-Buffered Saline) 0.05% Tween20 w/v (Sigma-Aldrich), 0.1% BSA w/v (Bovine Serum Albumin) (Sigma-Aldrich). CRH was detected with an immunoperoxidase technique, using PBS for dilution and TBS for washes. The rabbit anti-CRH Serum (1/50) (Phoenix Pharmaceuticals, Strasbourg, France) was deposited upon the slides for one hour in a humid environment at room temperature. Rinses of 10 mn were made between each stage using TBS, 0.1% BSA, 0.1%. The slides were incubated successively with a secondary biotinylated antibody (DAKO ChemTek detection kit peroxidase/AEC, Rabbit/Mouse, Trappes, France) (30 mn) and streptavidin coupled with peroxidase (30 mn). Reaction was stopped with distilled water (10 min) and counter-staining was done with Mayer haemalun (VWR International, Fontenay-sous-Bois, France) for about two minutes. Slides were rinsed with distilled water, and mounted in an aqueous medium. The control section included omission of primary antibody. Two different examiners read the slides. Labelling intensity was scored on a five-point scale: null labelling (0), very weak labelling (1), weak labelling (2), moderate labelling (3), strong labelling (4) and very strong labelling (5).

Statistical analysis.

CRH protein levels were expressed as the mean ± SD. To determine the differences among all experimental groups (Ctrl, SA, FM, LPS, Zn), a Kruskal-Wallis test was performed, followed by Dunnett’s test to isolate the group(s) that differed from the control group. Significance for stimulating effects of SA, FM, LPS or Zn compared with controls, was assessed at the p-value <0.05.

Abbreviations

P. acnes

Propionibacterium acnes

E. coli

Escherichia coli

CRH

corticotropin releasing hormone

FM

membrane fraction

SA

supernatant A

Zn

zinc

LPS

lipopolysaccharide

Ctrl

control

HPA

hypothalamic-pituitary-adrenal

ACTH

adrenocorticotropic hormone

POMC

proopiomelanocortin

CRH-R1

corticotropin receptor type 1

PSU

pilosebaceous unit

PBS

phosphate-buffered saline

TBS

tris-buffered saline

DMEM

dulbecco’s modified eagle medium

TNFα

tumor necrosis factor alpha

IL-1α

interleukin-1 alpha

IL-8

interleukin-8

TLR-2

toll-like receptor-2