Is it time to switch on or off the green light for ultraviolet-induced red fluorescence as a surrogate marker for Propionibacterium acnes in vivo?

Abstract

Despite over 100 years of history, the role of Propionibacterium acnes in acne vulgaris is still controversial. Numerous studies have been published providing arguments for or against the role of P. acnes. Various sampling and detection techniques have been used to support the claims. P. acnes has been sampled by skin swab, scrub, cyanoacrylate gel biopsy, tape stripping, direct lesion extraction and surgical skin biopsy with subsequent culture, 16S rDNA analysis and direct visualization by immunofluorescence and in situ hybridization.1 in a recently published issue of the journal, Xu et al.2 use ultraviolet-induced red fluorescence as a surrogate method to address the old question on acne vulgaris–P. acnes association and claim that P. acnes may be absent or present in lower density in comedones thus, eventually, challenging the role of P. acnes as the aetiology of acne vulgaris. Powerful claims, but is evidence robust enough or does it exist at all? These questions eventually lead to the more fundamental question: How to link or refute the link between a ubiquitous resident of human skin (P. acnes) and the age-dependent skin disease (acne vulgaris)? In medicine, it is imperative to investigate the whole organ where pathology is suspected. The same rule applies to a microscopic organ such as the hair follicle or a major organ such as the lung. Acne vulgaris is a disease of hair follicle where P. acnes may spread down the whole organ. Therefore, the first step in any study addressing a possible role of P. acnes in acne vulgaris is to ensure that the whole hair follicle is studied. Xu et al. used corneum plugs extracted from comedones as the object in their study. However, the corneum plug can hardly be considered a representative of the whole hair follicle as a varying and unpredictable amount of material is obtained. In an elegant experiment, Plewig showed that the skin biopsy taken directly after extraction shows abundant remaining material in the hair follicles.3 Extracted plugs will inevitably vary in size and weight which must be taken into consideration as well. The ability of P. acnes to produce coproporphyrins was demonstrated already in 1960, long before the bacterial genome was deciphered.4 The ability of coproporphyrins to produce red fluorescence upon stimulation with ultraviolet light and the presence of these substances in hair follicles further led to suggestion to use this phenomenon as a non-invasive parameter to monitoring P. acnes. Moreover, several studies have claimed to establish a correlation between fluorescence and P. acnes density. However, the evidence provided by these studies is not without controversy. McGinley et al.5 reported correlation between follicular fluorescence (originating in hair follicles) and P. acnes counts sampled by scrub (originating from skin surface). Therefore, two unrelated anatomical sites were compared. In another study, Borelli et al.6 analysed three lesions and reported an unusual, for such few observations, high correlation between coproporphyrin III and P. acnes counts. Other studies failed to establish any link between P. acnes and facial fluorescence and attributed the latter phenomenon to sheer sebum levels.7 P. acnes are reported to occur as biofilms in hair follicles,8 and the biofilm matrix can effectively block ultraviolet light.9 Whether the presence of P. acnes biofilms may be accounted for the lack of fluorescence in acne lesions is not known. Red fluorescence of the skin is a non-specific phenomenon and is observed in psoriatic plagues and can, at least partly, be due to refraction and reflection of light by keratinocytes. Sloughed keratinocytes are a known content of acne plaques and can, therefore, not be excluded as possible contributors to red fluorescence seen in acne lesions. In their report, Xu et al. claim to show the absence or low density of P. acnes by a surrogate method with unknown specificity and sensitivity without unequivocally showing P. acnes itself. This is imperative to persuade a critical reader. In conclusion, while the sheer idea of using non-invasive markers to establish the presence and monitor the levels of P. acnes in vivo is very attractive, a long road for ultraviolet-induced red fluorescence lies ahead and it remains to be seen whether the green light awaits at the end of the road.