Inhibition of Propionibacterium acnes by Bacteriocin-Like Inhibitory Substances (BLIS) Produced by Streptococcus Salivarius

Abstract

We report the in vitro inhibition of Propionibacterium acnes (P. acnes) by a bacteriocin-like inhibitory substance (BLIS-like substance) produced by Streptococcus salivarius (S. salivarius). Bacteriocins are proteinaceous substances produced by bacteria that are capable of inhibiting the growth of similar bacterial strains. Unlike classical antibiotics, they have a relatively narrow spectrum of killing activity, resulting in a reduction in the intensity of selection for resistance. These findings suggest that BLIS may potentially be used for its anti-P. acnes activity in the treatment of acne.

Introduction

Individuals with acne vulgaris are often treated chronically with antibiotics.^1,2 While effective in treating acne, antibiotics can have adverse effects. Antibiotic use can lead to the development of resistance in Propionibacterium acnes (P. acnes),^3,4 resulting in therapeutic failure and transference of resistance to other bacteria in the skin⁵ and gastrointestinal flora of close contacts.⁶ Individuals with acne that have been treated with antibiotics are about 3 times as likely to carry group A streptococcus (GAS)—a potentially pathogenic bacteria in oropharynx—as those who are not on antibiotics or those without acne.⁷ Furthermore, we have shown that the odds of developing an upper respiratory tract infection among acne patients taking antibiotics is 2.23 times greater than those not on antibiotics.⁸

All of the antibiotics commonly used to treat acne, including tetracycline, doxycycline, minocycline, and erythromycin, alter the normal bacterial flora in the throat that have an important regulatory effect on GAS.^7,9-16 Importantly, some of these normal flora produce substances that impede the ability of more pathogenic organisms to colonize, invade, and infect^16-18. Many of these compounds are part of a class of proteinaceous bactericidal substances called lantibiotics or bacteriocins.^12,19-22 One type of lantibiotic, called bacteriocin-like inhibitory substances (BLIS), is produced by Streptococcus salivarius (S. salivarius), a facultative anaerobic α-hemolytic streptococcus from the viridans group that is part of the normal throat flora.^11,20,22 S. salivarius produces at least a dozen separate types of BLIS. While other microorganisms may also inhibit GAS, S. salivarius—through BLIS production—is generally believed to be the main inhibitor of GAS in the oropharynx.^20,22

We hypothesized that BLIS might also be effective against P. acnes, which contributes to the pathogenesis of acne vulgaris.²³ In this study, we set out to recover S. salivarius strains from the oropharynx of a group of acne patients and controls, and measure their ability to produce BLIS capable of inhibiting growth of P. acnes and GAS.

Methods

A total of 106 male and female patients were recruited to participate in this cross-sectional study. This convenience sample was comprised of young, healthy individuals treated in the Department of Dermatology at the Hospital of the University of Pennsylvania. Individuals were excluded if they had taken an antibiotic for anything other than acne in the preceding 3 months. This study was approved by the Institutional Review Board of the University of Pennsylvania.

The dorsum of the tongue of each participant was swabbed and the specimen collected using a BBL Culture Swab Collection and Transport System (Difco, Detroit, Michigan). Cultures were diluted and directly inoculated on mitis-salivarius agar (Difco, Detroit, Michigan) using a sterile technique. Plates were incubated upside down at 37°C in 5% carbon dioxide for 24 hours. Colonies were screened by morphology consistent with S. salivarius (large, soft, fluffy colonies due to levan production from the sucrose in the agar). This is the standard method used to screen for S. salivarius.^24,25

BLIS production was evaluated using a modified deferred antagonism assay.²⁵ Each S. salivarius isolate was considered a producer strain. The isolates were grown aerobically in Bacto Tryptic Soy Broth, Soybean-Casein Digest Medium (BD Diagnostics, Sparks, MD) and a cotton swab was charged with growth and used to inoculate a diametric streak across Columbia Blood Agar. Each plate was then incubated for 18 hours at 35°C in 5% CO₂, and the resulting macroscopically visible growth removed with the edge of a glass slide. The plate was then inverted over chloroform vapors for 30 minutes, and subsequently exposed to air for 15 minutes. Three strains of group-A streptococci (ATCC 12344, 19615, and 12384) were used as indicator strains and streaked at right angles to the test streak. The plate was incubated anaerobically (BD BBL GasPak, Sparks, Maryland) for 24 hours at 35°C. Bacterial inhibition was considered significant if the zone of inhibition of the indicator strain was as least twice the width of the original test streak. This clear zone of inhibition served as evidence that a clinical isolate of S. salivarius was able to produce BLIS, and that this BLIS was effective in inhibiting the growth of a strain of GAS. Isolates that inhibited any or all of the GAS strains were considered BLIS producers.

The same deferred antagonism assay was repeated for P. acnes. This procedure was identical to the one described above except that P. acnes ATCC 6919 was used as the indicator strain, and grown on Columbia blood agar under anaerobic conditions for 72 h hours at 35°C. Isolates were picked and grown anaerobically in soy broth before being streaked at right angles to the S. salivarius test streaks.

We identified those human clinical isolates of S. salivarius that produced BLIS using both GAS standard reference strains (ATCC 19615, 12344, andl2384) and P. acnes standard reference strain ATCC 6919 as indicator strains.

Results

Of the 106 participants recruited (mean age 28.4 [SD 8.1], 76 [71.7%] female), cultures from 38 (35.8%) grew S. salivarius (Table 1). Of these 38 isolates, 33 (86.8%) survived freezing and thawing and were available for assays of P. acnes and GAS inhibition. Of the 33 surviving isolates, 11 (33.3%) inhibited P. acnes growth and 13 (39.4%) inhibited GAS growth. Five of these strains produced a form of BLIS that was able to inhibit both GAS and P. acnes, while the remaining strains showed activity against one or the other indicator strain (either P. acnes or GAS).

Table 1.

BLIS-producing capabilities of S. salivarius strains based on acne diagnosis.

Characteristic	Group I:Acne	Group II: No Acne
Total	77	29
S. salivarius strains producing P. acnes-inhibiting BLIS/total S. salivarius strains	8/18 (44%)	3/15 (20%)

Discussion

There is a rich history of research related to BLIS-producing organisms, which has primarily focused on the role of these substances in prophylaxis for, and treatment of, halitosis,²⁷ strep pharyngitis,²⁸ and otitis media.²⁹ This study demonstrates that S. salivarius is capable of producing BLIS active against 2 bacteria, P. acnes and GAS. While BLIS's activity against GAS was already known, our findings of activity against P. acnes—a contributor to the pathogenesis of acne vulgaris—are novel. These findings are important in that they suggest that BLIS may potentially be used for its anti-P. acnes activity in the treatment of acne.

Limitations of the current study include the use of only one P. acnes indicator strain, which might have resulted in an underestimate of the percentage of S. salivarius strains considered capable of anti-P. acnes activity. However, it is important to note that BLIS activity against P. acnes would be clinically meaningful only in the case that a particular acne patient's strain of P. acnes is sensitive to the BLIS substance(s) that that patient is exposed to. No difference was noted in the prevalence of S. salivarius strains demonstrating anti-P. acnes activity in those with or without a diagnosis of acne (Table 1). S. salivarius, however, resides in the oropharynx and thus would not be expected to play a large role in P. acnes regulation of the facial skin. Consequently, the topical administration of either of these bacteria or a purified form of the BLIS-like substance they produce merits study.

The rapid rise of multiresistant bacterial pathogens has made it necessary to identify alternative methods of combating infection³¹ as well as inflammatory conditions. BLIS could potentially serve as a natural alternative to antibiotics. Bacteriocins have a relatively narrow spectrum of killing activity, resulting in a reduction in the intensity of selection for resistance. We have identified strains of S. salivarius capable of producing BLIS against both GAS and P. acnes. To the best of our knowledge this is the first time that the capability of S. salivarius to produce a substance inhibitory to P. acnes has been reported. This substance shows promise as a potential future treatment for infections caused by GAS, such as pharyngitis and impetigo, as well as for acne.

The present study provides information on the BLIS-producing capabilities of S. salivarius. Further research should focus on determining whether BLIS has in vivo effects on P. acnes, whether P. acnes develops resistance to these effects, and whether there may be a role for a topical formulation of BLIS, or BLIS-producing bacteria, in the treatment of acne vulgaris.

Disclosures

WPB and DJM hold a provisional patent for the use of bacteriocin-like inhibitory substance (BLIS) for the treatment of acne. This manuscript was supported by an Agency for Healthcare Research and Quality (AHRQ) Centers for Education and Research on Therapeutics cooperative agreement (grant HS10399), NIH grant K24-AR02212, and an American Dermatology Association summer student research award for WPB.