Resolution of inflammatory acne vulgaris may involve regulation of CD4+ T-cell responses to Propionibacterium acnes.

BACKGROUND: Propionibacterium acnes has been strongly implicated in inflammatory acne. However, its role in the disease is unclear. It has been hypothesized that an immune response to P. acnes and/or P. acnes heat shock proteins (HSPs) may play a role in the pathogenesis of inflammatory acne. OBJECTIVES: To compare the cell-mediated immune response to P. acnes and HSPs in acne patients, nonacne controls and individuals with resolved acne. METHODS: The proliferative response of peripheral blood mononuclear cells (PBMC) from acne patients, resolved acne donors and healthy controls to P. acnes, P. acnes HSP60 and HSP70, and mycobacterial HSPs was assessed by lymphocyte transformation assay (LTA). The proliferative response of purified CD4+ T cells was further analysed by limiting dilution analysis (LDA). Contingency tables (G-test) were used to analyse the proportion of individuals in each group showing a positive proliferative response for LTA or data fitting single-hit kinetics for LDA. RESULTS: Analysis of stimulation of PBMC with P. acnes, P. acnes HSP60 and HSP70 in the LTA showed the proportion of positive responders to be independent of subject group. However, the proportion of acne patients with a positive response to mycobacterial HSPs was significantly higher than those for the other subject groups. Analysis of LDA data showed the proportion of resolved donors with responses to P. acnes fitting the single-hit kinetics model to be significantly lower than those of the other groups. There were no significant differences in responses to other antigens. CONCLUSIONS: The significantly lower proportion of resolved donors demonstrating a single-hit kinetics response to P. acnes by LDA may represent negative regulation of the CD4+ T-cell response to P. acnes in these subjects.

Engineering of the gut commensal bacterium Bacteroides ovatus to produce and secrete biologically active murine interleukin-2 in response to xylan.

AIMS: The aim of this work was to engineer a gut commensal bacterium, Bacteroidesovatus, to produce and secrete a biologically active cytokine in a regulated manner as a basis for novel immunotherapies for chronic gut disorders. METHODS AND RESULTS: Bacteroides ovatus was engineered to produce murine interleukin-2 (MuIL2) intracellularly in response to xylan in culture media by inserting the MuIL2 gene into the xylanase operon of the organism. A second strain was engineered to secrete MuIL2 by adding Bacteroides fragilis enterotoxin secretion signal sequence to the protein. The recombinant strains produced MuIL2 only in the presence of xylan as determined by ELISA of cell lysates and culture supernatants. The IL2-dependent cell line CTLL-2 was used to demonstrate that MuIL2 produced by both B. ovatus strains was biologically active. This activity could be blocked by an anti-IL2 neutralizing antibody. The xylan-inducible nature of this system was demonstrated by RT-PCR. CONCLUSIONS: Bacteroides ovatus was successfully engineered to produce and secrete biologically active MuIL2 in a xylan-inducible manner. SIGNIFICANCE AND IMPACT OF THE STUDY: The production and secretion of a biologically active mammalian protein by a member of the gut microflora could lead to the development of new long-term immunotherapies for inflammatory gut diseases.

Evidence for diversity within Propionibacterium acnes: a comparison of the T-cell stimulatory activity of isolates from inflammatory acne, endocarditis and the laboratory.

BACKGROUND: Propionibacterium acnes is primarily associated with the pathogenesis of acne vulgaris but reports are increasing in number implicating P. acnes in other diseases such as abscess formation, meningitis and endocarditis. The pathogenicity of P. acnes is thought to be partly due to the interaction of the bacterium with the immune system. Historically, investigations have focused on humoral and cell-mediated immune responses to P. acnes antigens without attention to the possibility that different antigens may be expressed by different isolates. OBJECTIVE: Investigations were performed to determine whether there were differences between a laboratory strain of P. acnes (P-37) and fresh clinical isolates in their ability to stimulate naive and adult lymphocytes. MATERIAL AND METHODS: The fresh isolates were collected from a patient with inflammatory acne and a patient with P. acnes-induced prosthetic valve endocarditis. The lymphocyte transformation assay was used to detect responses to whole-cell suspensions of stationary phase P. acnes isolates during 7 days of incubation. RESULTS: The acne isolate was significantly more stimulatory for cord blood mononuclear cells (CBMNCs) than the laboratory isolate (P. acnes P-37) at day 4 of incubation. There were no significant differences between the three strains at any other time points. However, the isolate cultivated from inflammatory acne was significantly more stimulatory for peripheral blood mononuclear cells (PBMNCs) from acne donors than the endocarditis isolate or the laboratory strain at most time points. There were no significant differences between the endocarditis strain and the laboratory strain. CONCLUSION: It can be hypothesized that in case of P. acnes-induced endocarditis lymphocyte stimulation is a disadvantage for the microorganism and therefore a lack of lymphocyte stimulation may be relevant to the pathogenesis of endocarditis.

The effect of lipids on the adherence of axillary aerobic coryneform bacteria.

AIMS: A wide range of lipids are present on the skin surface of human beings and bacterial lipases are known to modify them. The microflora of the underarm (axilla) is often dominated by aerobic coryneforms and whilst many require lipids for growth, they appear not to be utilized as carbon sources. The aim of this study was to investigate the adherence aerobic coryneforms to lipids present on the skin surface of the human axilla to determine whether they contribute to colonization of the skin. METHODS: Aerobic coryneforms were grown in a defined synthetic medium in the presence of (14)C-glucose to produce radio-labelled cells. Adherence to lipids was tested using a thin layer chromatography plate-based assay. SIGNIFICANCE AND IMPACT OF THE STUDY: The mechanism of bacterial adhesion to skin lipids is unknown. The results of this study show that a significant proportion of cutaneous aerobic coryneform isolates from the axilla interact with skin lipids resulting in increased adherence, which may contribute to skin colonization.

Proinflammatory cytokine production by human keratinocytes stimulated with Propionibacterium acnes and P. acnes GroEL.

BACKGROUND: Keratinocytes form the first line of defence in the skin and alert the host to danger by the production of a number of cytokines and chemokines. However, the interaction of commensal microorganisms with keratinocytes has not been well studied. OBJECTIVES: To investigate the effect of viable and nonviable cells of Propionibacterium acnes in both exponential and stationary growth phases, and of P. acnes GroEL on cytokine production by human primary keratinocytes. METHODS: Actively proliferating or contact-inhibited keratinocytes were cocultured with viable or formaldehyde-killed P. acnes cells in either the exponential or stationary phase of growth. Culture supernatants were assayed by enzyme-linked immunosorbent assay for the cytokines interleukin (IL)-1alpha, tumour necrosis factor (TNF)-alpha and granulocyte/macrophage colony-stimulating factor (GM-CSF). Keratinocytes were also stimulated with different concentrations of P. acnes GroEL and supernatants assayed for cytokines. RESULTS: Viable P. acnes in the stationary phase of growth stimulated keratinocyte monolayers to produce significantly higher amounts of IL-1alpha, TNF-alpha and GM-CSF than unstimulated keratinocytes. Viable exponential-phase bacteria stimulated production of significantly higher amounts of TNF-alpha and GM-CSF but these levels were significantly lower than those for stimulation with stationary-phase bacteria. Nonviable P. acnes from either growth phase was not able to stimulate cytokine production. P. acnes GroEL at concentrations in the range 0.05-1.0 micro g mL(-1) was able to induce increased production of cytokines by keratinocytes in a dose-dependent manner. This was analogous to stimulation with Escherichia coli GroEL. CONCLUSIONS: Stimulation of cytokine production by P. acnes and P. acnes GroEL may be important in the pathogenesis of inflammatory acne vulgaris and may have wider implications for the immunomodulation of the human immune system by commensal skin microorganisms.

Kytococcus sedentarius, the organism associated with pitted keratolysis, produces two keratin-degrading enzymes.

AIMS: To determine characteristics of the extracellular enzyme activity of Kytococcus sedentarius on human callus. METHODS AND RESULTS: A concentrate of a continuous culture supernatant fluid of K. sedentarius, which had callus-degrading activity, was subjected to a series of chromatographic purification procedures. The enzyme activity was found to be attributable to two proteases. These were capable of degrading both native callus and extracted keratin polypeptides and were purified to homogeneity, as shown by SDS-PAGE with silver staining. The enzymes P1 and P2 were 30 kDa and 50 kDa in size with isoelectric points of 4.6 and 2.7, respectively. The optimum conditions for callus-degrading activity were 40 degrees C, pH 7.1 for P1 and 50 degrees C, pH 7.5 for P2. P2 displayed increased activity in the presence of 800 mmol l(-1) NaCl and both enzymes were inhibited by PMSF (1 mmol(-1) Phenylmethylsulphoryl fluoride) and 1 mmol l(-1) EDTA. The main enzyme cleavage sites were Lys-Trp, Val-Lys, Gly-Asp and Asp-Arg, as determined after incubation of P1 and P2 with the beta-chain of insulin. CONCLUSIONS: K. sedentarius produces two extracellular enzymes that independently degrade natural, insoluble human callus. Both enzymes are serine proteases and have cleavage preference sites that are present in a range of human keratins. SIGNIFICANCE AND IMPACT OF THE STUDY: The identification, in K. sedentarius cultures, of two enzymes which can degrade human callus strengthens the hypothesis that this organism is responsible for the pitting in human epidermis observed in pitted keratolysis. These enzymes may be of commercial use in the biodegradation of a range of keratin polymers, biological washing powders and in the treatment of unwanted callus on human skin.

Expression of the S. aureus hysA gene in S. carnosus from a modified E. coli-staphylococcal shuttle vector.

We have modified an E. coli-staphylococcal shuttle vector for use in the general cloning and expression of genes from pathogenic staphylococci in Staphylococcus carnosus. As S. carnosus is non-pathogenic, this expression system will facilitate the study of the roles of individual gene products in the disease process. To evaluate the use of this expression system, a DNA fragment containing the Staphylococcus aureus hyaluronate lyase (hysA) gene was cloned into the modified vector, pNW21, and introduced into S. carnosus. Hyaluronate lyase was both produced and secreted by S. carnosus. In addition, the secreted HysA protein was enzymatically active, as determined using a zymographic assay.

Propionibacterium acnes and inflammation in acne; P. acnes has T-cell mitogenic activity.

BACKGROUND: Circumstantial evidence suggests that Propionibacterium acnes has a role in the inflammation of acne. This could be effected by antigenic or superantigenic or mitogenic reactions. OBJECTIVES: The purpose of this investigation was to determine whether P. acnes had only antigenic activity or additional superantigenic and mitogenic activity. METHODS: A lymphocyte transformation assay was used to detect responses to a mixture of eight P. acnes whole cell isolates, and their supernatant culture fluids. In order to determine the nature of T-cell reactions to P. acnes cells a mouse-antihuman major histocompatibility complex class II monoclonal antibody was used in the lymphocyte transformation assay to inhibit the antigenic stimulation of lymphocytes. An analysis of the T-cell receptor (TCR) variable region beta (BV) repertoire was undertaken using flow cytometry of the unstimulated and stimulated cells. RESULTS: Peripheral blood mononuclear cells (PBMNC) from adults with no history of acne responded strongly to stationary growth phase cells of P. acnes, less strongly to cells in the exponential growth phase. No response was detected to supernatant culture fluids. PBMNC from five cord blood samples (CBMNC) responded maximally after 3 and 7 days of incubation with stationary growth phase cells of P. acnes. The reaction of CBMNC to P. acnes cells was not suppressed completely by the blocking antibody. The analysis of the TCRBV repertoire indicated that P. acnes induced no deletion or over-representation of certain BV element-bearing T cells. The TCRBV analysis was repeated after preincubation with the blocking antibody. Deletion of T cells bearing certain BV components occurred and there was no over-representation of T cells carrying certain BV components. CONCLUSIONS: Two mechanisms of lymphocyte activation by P. acnes cells are proposed, antigen and mitogen driven. These results are consistent with the histological evidence of inflammation in acne lesions.

Production of the mycelial phase of Malassezia in vitro.

To study the pathogenicity of Malassezia, the agent of pityriasis versicolor, it is necessary to obtain the mycelial form in vitro. A range of different components and conditions were tested to induce yeast cells of the organism to produce mycelia in vitro using different culture media. A mycelial culture medium was developed that consisted of bacteriological peptone, glucose, yeast extract, ox bile, glycerol, glycerol monostearate, Tween 80, squalene, glycine, potassium nitrate, sodium chloride, ferrous sulphate and magnesium sulphate with or without agar. The liquid and solid medium had a pH of 5.6 and the temperature of incubation was 30 degrees C. Cultures were incubated in air. This medium was able to induce some strains of Malassezia to produce up to 40% mycelium in vitro. In total, 33 different strains of Malassezia obtained from the skin of the healthy individuals and patients with pityriasis versicolor were tested for mycelium production. The strains of Malassezia capable of producing mycelium in vitro all possessed the serovar A antigen.

Heat shock proteins and inflammatory acne vulgaris: molecular cloning, overexpression and purification of a propionibacterium acnes GroEL and DnaK homologue.

Propionibacterium acnes is associated with inflammatory acne. The genes encoding two putative mediators of inflammation, the heat shock proteins GroEL and DnaK, were cloned from this organism and sequenced. groEL and dnaK encode proteins of 56.8 and 66.4 kDa, respectively, which show a high degree of homology (>75% similarity) to the GroEL and DnaK proteins of mycobacteria and streptomycetes. The promoter regions of both genes contain inverted repeat sequences believed to be involved in the transcriptional regulation of heat shock genes. Recombinant P. acnes GroEL and DnaK were overexpressed in Escherichia coli with C-terminal histidine tags. The recombinant proteins were purified from E. coli by metal affinity chromatography. These proteins will now be used in immunological investigations to determine their role in inflammatory acne.

The effects of lipid extraction on the immunomodulatory activity of Malassezia species in vitro.

Malassezia spp. are believed to be the causative agents of pityriasis versicolor and are strongly implicated in seborrhoeic dermatitis. The yeast also forms part of the normal human cutaneous microflora. We have previously shown that when Malassezia yeast cells are incubated with human peripheral blood mononuclear cells (PBMCs), they are capable of reducing the levels of pro-inflammatory cytokines produced. In order to test the hypothesis that this immunoevasive phenomenon may be related to the unusually high level of lipid in the Malassezia yeast cell wall, we have compared the immunomodulatory capacity of normal and lipid-depleted yeast cells. Stationary phase yeast cells of Malassezia sympodialis, M. globosa and M. restricta were treated with chloroform/methanol to extract the surface lipids. The lipid-depleted and non-depleted yeast cells were then co-cultured with human PBMCs from three different human donors at a ratio of 20 yeasts per leukocyte for 24 h. The levels of interleukin (IL)-1beta, IL-6 and tumour necrosis factor (TNF)-alpha were then determined by enzyme-linked immunosorbent assay (ELISA). The results demonstrated that extraction of lipid reversed the yeast cell capacity to reduce the levels of pro-inflammatory cytokines. The levels of IL-1beta, IL-6 and TNF-alpha produced in response to lipid-extracted Malassezia of all three species were either no different from or significantly greater (P < 0.05; ANOVA) than the constitutive control levels. These results suggest that the lipid microfibrillar layer of Malassezia may prevent the yeast cells from inducing inflammation and provide an explanation for the normal commensal status of the organism on human skin. The hypothesis that the lipid layer is absent or altered in seborrhoeic dermatitis may provide an explanation for the inflammatory nature of this dermatosis.

The cutaneous microflora of adolescent, persistent and late-onset acne patients does not differ.

The cutaneous microbiology and antibody status to Propionibacterium acnes of patients with persistent (males, n = 32; females, n = 33) and late-onset (females, n = 25) acne were compared with individuals with adolescent acne (males, n = 22; females, n = 18) and normal control volunteers (persistent acne: males, n = 26; females, n = 30; late-onset: females, n = 20). Males had significantly higher grades of acne compared with females (P < 0.05). The microflora consisted in the main of propionibacteria, staphylococci and Malassezia; other bacteria represented less than 0.01% of the total microflora. At all sites for all samples there were significantly more propionibacteria than staphylococci or Malassezia (P < 0.05). There were significantly higher (P < 0.05) numbers of microorganisms in follicular casts from patients compared with their control volunteers for female facial skin and male back skin. Twenty-six papules and 48 normal follicles were analysed. A bimodal distribution of microbial colonization was noted, with about 90% of normal follicles and about 10% of acne follicles having no detectable viable microorganisms. Anti-P. acnes IgG antibody titres were measured using a secondary fluorescein isothiocyanate antibody technique, and no significant differences in titre were found between any groups of patients (P > 0.05). Correlation analysis showed no association between the population densities of P. acnes and anti-P. acnes IgG titres. There were no differences in the microbiology of skin of adolescent acne patients, persistent acne patients or late-onset acne patients which could account for these various forms of acne.

Shift-down in growth rate rather than high cell density induces toxic shock syndrome toxin-1 gene expression in Staphylococcus aureus.

A luciferase-based reporter system for the expression of the toxic shock syndrome toxin-1 gene (tst) of Staphylococcus aureus FRI 1187 was used in continuous culture to determine whether high cell density on transient shift-down or shift-up of specific growth rate (mu) induced expression of tst. Little expression occurred at steady state at a low dilution rate (D) and in a transient period of increasing mu. However, a rapid and approximately 130-fold increase in expression occurred during a transient shift-down of mu. These findings suggest reduction of mu is a key element in the control of tst expression.

The effects of Malassezia on pro-inflammatory cytokine production by human peripheral blood mononuclear cells in vitro.

Malassezia spp., the causative agents of pityriasis versicolor, are members of the normal human cutaneous microflora. Utilizing a combination of both enzyme-linked immunosorbent assay (ELISA) and bioassay, we have investigated the ability of both formalin-preserved and viable Malassezia (serovars A, B and C) to modulate pro-inflammatory cytokine (IL-6, IL-1beta and TNF-alpha) release by human peripheral blood mononuclear cells (PBMNC) in vitro, over a 48-h co-incubation period. The results demonstrated that formalin-preserved Malassezia (serovars A, B and C) at mid-exponential phase were generally able to induce a pro-inflammatory cytokine response at a yeast cell to PBMNC ratio of 1:1. In addition, the results consistently demonstrated that at a yeast cell to PBMNC ratio of 20:1, formalin-preserved Malassezia, irrespective of serovar, growth phase or PBMNC donor, were capable of significantly (P<0.05) decreasing the release of both immunochemical IL-6 and IL-1beta plus bioactive IL-1beta and TNF-alpha below that of unstimulated culture medium control values. This was apparent following 24- and 48-h co-incubation times, where maximal cytokine production was detected after 24 h. Similar results were obtained for the effect of viable Malassezia on pro-inflammatory cytokine release by PBMNC. Our results suggest that a possible inhibitory component, present perhaps within the cell wall of Malassezia, was responsible for this depressive effect on pro-inflammatory cytokine production.

Investigation of the mechanism of action of 2% fusidic acid lotion in the treatment of acne vulgaris.

We describe the results of a single-centre, double-blind, vehicle-controlled, parallel group study on the quantitative effects of 2% fusidic acid lotion (Fucidin lotion) in facial acne vulgaris. The trial was completed by 52 patients aged 15-25 years with mild to moderate acne who had been randomized to either Fucidin Lotion (n = 25) or its base (n = 27). Primary outcome measures included colony counts of Propionibacterium acnes and micrococcaceae and measurements of skin surface lipid free fatty acids and sebum excretion rate. Clinical assessment was based on the acne grade, count of inflamed and non-inflamed lesions and evidence of a primary irritant dermatitis. There was a variable but gradual reduction in lesion counts with the maximum improvement at 12 weeks for inflamed lesions, where the reduction was 19.9% for fusidic acid and 24.7% for the placebo. The non-inflamed lesions decreased by 10.8% in the fusidic acid group and increased by 15.9% in the placebo group; this difference was not statistically significant. Although the fusidic acid reduced the micrococcaceae count by 1 log cycle, inferring adequate compliance, there was no reduction in the counts of P. acnes, surface free fatty acids or sebum excretion rate. This study has failed to explain the mechanism of action of topical fusidic acid.

The effects of acne treatment with a combination of benzoyl peroxide and erythromycin on skin carriage of erythromycin-resistant propionibacteria.

Concomitant application of 5% w/w benzoyl peroxide and 3% w/w erythromycin has previously been shown to prevent the overgrowth, on the skin of acne patients, of erythromycin-resistant coagulase-negative staphylococci, which occurs when the antibiotic is used alone. Two in vivo studies were carried out to assess the ability of the same therapeutic combination to inhibit the growth of pre-existing erythromycin-resistant propionibacteria and to prevent the selection of resistant strains during treatment. A double-blind clinical trial in 37 patients with mild to moderate acne vulgaris showed that the combination brought about a > 3 log10 c.f.u. reduction in total propionibacterial numbers/cm2 after 6 weeks therapy (P < 0.001, Wilcoxon's matched pairs) and also significantly reduced the number of erythromycin-resistant propionibacteria (P < 0.05). In contrast, erythromycin alone reduced the total propionibacterial count by < 1.5 log10 c.f.u./cm2 after 6 weeks (P < 0.05) and did not affect the number of erythromycin-resistant strains. The combined formulation was significantly more effective at reducing total propionibacterial numbers at 6 (P < 0.01, Mann-Whitney) and 12 weeks (P < 0.05) than erythromycin alone, although, after 12 weeks, the anti-propionibacterial efficacy of both preparations was less marked. Five patients on combination therapy, and five treated with erythromycin alone, acquired erythromycin-resistant strains de novo at week 6 or week 12. In an open study in 21 acne patients, who each carried > 10(3) c.f.u. erythromycin-resistant propionibacteria/cm2 skin pretreatment, the combination of erythromycin and benzoyl peroxide reduced the total propionibacterial count by > 2.5 log10 and the number of erythromycin-resistant strains by a similar amount (P < 0.001, Wilcoxon). This was accompanied by highly significant reductions in acne grade and lesion counts (P < 0.001). These data suggest that the combination of 5% w/w benzoyl peroxide and 3% w/w erythromycin has greater in vivo anti-propionibacterial activity than 3% w/w erythromycin alone, and brings about significant clinical improvement in acne patients with high numbers of erythromycin-resistant propionibacterial strains pretreatment.

Cloning, nucleotide sequence determination and expression of the Staphylococcus aureus hyaluronate lyase gene.

The Staphylococcus aureus 8325-4 hyaluronate lyase gene (hysA) was identified after detecting hyaluronate lyase activity expressed by phages from a genomic library. The hysA open reading frame, capable of encoding a protein of 91 980 Da, was identified by Tn5 mutagenesis and nucleotide sequencing. HysA shares 35 and 36% amino acid sequence identity with group B streptococcal hyaluronate lyase and pneumococcal hyaluronidase, respectively. A 94-kDa protein was expressed in Escherichia coli minicells, a result consistent with the coding capacity of hysA. Identification of the S. aureus 8325-4 hyaluronate lyase gene will allow the regulation of this putative virulence determinant to be studied.