Propionibacterium acnes and Acne

Acne is a multifactoral disease which is restricted to man and usually presents during and immediately after puberty [1, 2]. The disease is localized to skin regions such as the face, back and chest, with a high number of sebaceous follicles [1, 2] and the condition has been associated with a high sebum excretion rate (SER) [1, 2]. The disease in its various forms and severity is treatable utilizing a range of topical and systemic drugs [1]. The use of cis-retinoic acid to successfully manage severe cases of acne has greatly affected research into the cause of acne and the mechanisms which lead to non-inflamed and inflamed lesions. However, important questions remain unanswered: what causes the variety of the distribution of lesions in patients? what factors are involved in the initiation of inflammation? is Propionibacterium acnes involved in the initiation process? and what mechanisms are involved in the natural regression of the disease in the majority of patients when they reach their mid-twenties?This communication proposes hypotheses, which place P. acnes as an important contributing factor in acne, and it is intended that the hypotheses should provoke increased experimental activity to answer the questions previously listed.An assumption will be made that before puberty the sebaceous follicles are normal and because the follicle wall has a functional barrier the lumen has a low free water content. The microflora of the skin surface, and probably the follicles, is low in population density [3]. During puberty the SER increases [4], but there is a distribution of the SER in the total population of the sebaceous follicles. This is believed to be due to a variation in end-organ response [1]. The work of Stewart et al. [5] predicts that in a follicle with a high SER the system may become linoleic acid deficient, which subsequently will affect the barrier function of the follicle wall. The next assumption is that there will be an increased flux of water from the dermis into the lumen of the follicle, which would promote colonization of the follicle and/or an increase in the microbial content [6]. The location of microbial cells within the duct is also important to future events. The nearer to the surface colonization occurs, the fewer problems will ensue because the follicle wall is thicker.If colonization occurs on the follicle wall where the number of keratinized layers is low, then there is the increased possibility that pro-inflammatory cytokines will be released from the keratinocytes and these messengers will diffuse to the dermis to initiate inflammation by a conventional mechanism [7]. The two possible general non-exclusive routes are shown in figure 1.The possible mechanisms for keratinocyte damage leading to cytokine release and further perturbation of follicle wall barrier function are, first, the production of porphyrin [8, 9] by P. acnes in the presence of increasing oxygen tensions because of changes in the follicle wall barrier function caused by linoleic acid deficiency. The subsequent interaction of molecular oxygen with released porphyrin will produce toxic reduced oxygen species and free radicals, which will damage the adjacent keratinocytes. Secondly, it is feasible that P. acnes has a quorum-sensing mechanism, whereby it up-regulates extracellular enzyme production by an autocrine signal process which is triggered at high bacterial cell density. Quorum sensing has been described in other bacteria [10]. Extracellular enzymes such as lipase, protease, hyaluronate lyase and neuraminidase [11] have the potential to both affect keratinocyte integrity and barrier function of the follicular wall. Thirdly, up-regulation of cytokine production could be indirectly affected by the production of heat shock proteins by P. acnes in response to nutritional stress. The initial influx of water into the follicle would result in a higher nutrient concentration per bacterial cell which would diminish as the bacterial cell population rises. Accompanying this change would be a resultant local environmental pH shift, predicted to move toward the acidic side by microbial metabolism. These impositions would invoke the normal bacterial stress response [12] with the subsequent release of heat shock proteins from the bacteria which would be involved in the increased production of cytokines [13].The older literature indicates that the initial cellular infiltrate to the acne lesion is neutrophilic [14]. More recent evidence clearly indicates that in very early inflamed lesions the infiltrate is lymphocytic [15, 16]. It is suggested that over time the lymphocytic infiltrate becomes more specific to the variety of bacterial antigenic material with subsequent non-specific neutrophil attack due to the bacterial chemo-attractants produced and the normal long-term response associated with tissue damage and repair [17]. The duration of the lesion and its problematic progression to papule, pustule or cyst will be dependent on the bacterial load in the follicle and whether major follicular wall damage occurs allowing expansion of the highly persistent P. acnes number in the follicle and very local dermal tissue. Another major determinant will be the individual’s inherent capacity for wound repair.Normal wound repair mechanisms have been used to explain the healing of a lesion. Relevant to the final regression of acne in an individual is whether a specific follicle can cycle through the process a number of times before it is structurally changed by wound repair mechanisms such that the initiation process is no longer possible. Aldana et al. [18] have evidence that it is normal for cyclic behaviour of the general state of a follicle to occur with a possibility for abnormal consequences at a particular phase. It is not unthinkable that a round of initiation, inflammation and repair (fig. 2) might alter the follicle so that the variance in the behaviour of the now ‘normal’ follicle is restricted. The duration of the disease in a patient would be dependent, in this hypothesis, on the number of potential acne-prone follicles and the number of cycles each one could endure without alteration to inhibit future cycles.The hypotheses put forward are at the present time dependent on both strong evidence and assumptions. These assumptions require vigorous and critical investigations and all are dependent on experimental systems analysing the activities of individual normal follicles and lesions. It is disappointing to observe that our present technologies are in the most part inadequate to address these fundamental problems.The authors wish to thank Prof. John Strauss for his contribution to research into acne which has led in part to their enthusiasm to pursue research in this topic.