The Microbiological Signature of Folliculitis Decalvans.

New Data Suggests Folliculitis Decalvans is Likely More Complex than a Story of Staphylococcus Aureus

Authors of a new study set out to determine if the follicular microbiota (bacteria and fungi) residing in FD-affected hair follicles are different than normal healthy hairs and if an aberrant immune response was present in the pathogenesis of FD.

To do so, the authors conducted a study of ten patients affected by FD. Scalp biopsies were taken from affected and healthy scalp to identify the follicular microbiome. To identify the follicular microbiome, a technique known as “next-generation sequencing” of the V3-V4 region of the 16S rRNA gene (in the case of bacteria) and the ITS-1 region (in the case of fungi) was carried out.

To determine how patient’s immune system responds to bacteria, peripheral blood mononuclear cells (PBMCs) were obtained, and their cytokine production was quantified after incubation with various stimuli and compared with healthy controls.

Cytokine production by PBMCs were determined at baseline and after stimulation with immune stimulatory agents phorbol myristate acetate (PMA) and ionomicine and after incubation with pathogen-associated molecular patterns (PAMPs) obtained from boiled bacterial suspensions including laboratory strains of Staphylococcus aureus (ATCC25923), Staphylococcus epidermidis (ATCC14990) and wild bacteria isolated from patients biopsies including Staphylococcus aureus, Staphylococcus epidermidis, Staphylococcus capitis, Staphylococcus hominis, Staphylococcus lugdunensis, Klebsiella pneumoniae and Klebsiella aerogenes.

The authors measured the cytokine levels of interleukin-8 (IL-8), interleukin-1β (IL-1β), interleukin-6 (IL-6), tumor necrosis factor α (TNF-α), interleukin-12p70 (IL-12p70) and interleukin 10 (IL-10).

Results.

 A total of 10 patients (8 men and 2 women) with a mean age of 37.6 years (range 29-48 years) were recruited.

BACTERIA AND FUNGI IN FD

There was a significant difference regarding bacteria comparing follicular microbiota of healthy and FD-affected hairs.

The most frequent bacteria found in follicular biopsies from FD were Staphyloccocus spp., Cutibacterium sp., Bacteroides sp., Dysgonomas sp., Alistipes sp., Soleaferrea sp., Fusobacterium sp., Desulfovibrio sp., Klebsiella sp. and non-specified Enterobacteriacea.

Staphylococcus spp. represented 25.9% of the total count of follicular bacterial microbiota in FD-affected biopsies but only 6.6% of the total bacteria in healthy follicles so Staphylococcus was overrepresented in FD.

Interestingly, the cultured bacteria had the typical antimicrobial susceptibility pattern, and did not show any of the known acquired resistance mechanisms. The most prevalent fungi in FD biopsies were (>10 % of the global composition) were Malasezzia restricta, Malasezzia globosa and Malasezzia sympodialis.

The total numbers of bacteria did not differ in FD but rather the types of bacteria. The fungi were not different between the FD and healthy controls. Specifically there was no statistically significant difference was found regarding α-diversity in neither bacteria  nor fungi  comparing follicular microbiota of healthy and FD-affected hair follicles. The β-diversity analysis showed statistically significant differences regarding bacteria  but not fungi (p=0.52)

When the researchers looked closer at the types of bacteria, they found an enriched presence of the Firmicutes phylum, especially bacteria from the Lachnospiraceae family (Tyzerella sp., Coprococcus comes, Blautia obeum and Agathobacter sp.) and the Ruminococcaceae family. Additionally, Bacteriodales vadin HA21 also had an increased presence in FD-affected follicles. In healthy follicles, the Actinobacteria phylum, especially the Actinobacteria class, Corynebacteriales order and Corynebacteriaceae family including Lawsonella clevelandensis, were significantly overrepresented.

Certain bacteria in FD such as Ruminococcaceae, Agathobacter sp., Tyzzerella sp. and Bacteriodales vadin HA21 family were significantly associated with disease status .

CYTOKINE PRODUCTION

The authors point out that some aspects of the immune system are quite normal in immune cells from patients with FD but some are quite abnormal – particularly those involving responses to bacteria.

The authors first looked at cytokine expression after stimulation with phorbol myristate acetate (PMA) and ionomycin which are known stimulators of immune cells. Cytokine expression  of TNF-α, IL-1β, IL-6, IL-8, IL-10 and IL-12p70 in peripheral blood mononuclear cells (PBMCs) showed no statically significant differences between FD patients and healthy controls regarding their baseline levels or after stimulation with phorbol myristate acetate (PMA) and ionomycin.

Next, the authors looked at cytokine expression after stimulation with various bacterial antigens. Here, differences were found.  IL-10 levels were significantly decreased in FD patients after incubation with the laboratory strain of S. aureus (p=0.03), wild S. aureus (p=0.02), S. epidermidis (p=0.006) and S. capitis (p=0.02). TNF-α levels were significantly decreased in FD patients after incubation with the laboratory S. aureus strain (p=0.006), S. epidermidis (p=0.003) and S. capitis (p=0.02). IL-6 levels were significantly decreased in FD patients after incubation with S. epidermidis and S. capitis (p=0.02).  The authors propose that this was evidence for ‘immune exhaustion.”

Comments and Discussion

This is a really important paper for us all to be aware of.

For years, we have come to believe that folliculitis decalvans is a disease related to Staphylococcus aureus.  This paper reminds us that this is probably not quite how it is and the connection to Staphylococcus aureus is probably not so simple.

As the authors point out, most studies in the past few decades have just relied among taking samples of the pustules or scalp of FD and growing them in standard conditions. These sorts of standard techniques have significant limitations to detect certain types of bacteria including certain anaerobic bacteria that lie deeper in the skin. The authors remind us that  complete follicular microbiome of FD has not been well studied. 

This study shows us several important points about the microbiome in FD. Here are some key points:

1. There was no statistically significant differences regarding total bacterial and fungi microbiota comparing pathological and healthy hair follicles (ie alpha diversity). So it’s not that FD appears “overgrown” with bacteria but rather the types of bacteria that reside in the follicle from patients with FD seem different.

2. There was a difference in beta diversity so it seems that the pathological hair follicles have a specific bacterial microbiota compared to healthy follicles. It’s not clear if this is the reason for the disease or they are merely a bystander but the bacteria are different.   There is somethings special about the changes in the skin or hair follicle that allow staphylococcus to grow. We don’t understand all that yet.

3. There was no difference in fungi β-diversity and therefore, fungi are probably not associated with the pathogenesis of folliculitis decalvans.

4. in addition to Staphylococcus aureus, the authors showed that various bacteria from the Firmicutes phylum were associated with FD follicles. So there appears to be a dysbiosis. Firmicutes bacteria are predominant in human gut, in the oral cavity and perianal skin. Why they are in FD follicles is not clear.

5. Certain members of the Firmicutes phylum seem relevant.  Specifically, the Lachnospiraceae family bacteria (mainly Tyzerella sp., Coprococcus comes, Blautia obeum and Agathobacter sp.) and the Ruminococcaceae family were identified as bacteriological biomarkers of the disease.

6. Purified monocytes from FD patients secrete cytokines like TNF-α, IL-1β, IL-6, IL-8, IL-10 and IL-normally in response to traditional stimulants like PMA and ionomycin. However, IL-10, IL-6 and TNF-α production was significantly decreased in FD patients after incubation of peripheral monocytes with S. aureus, S. epidermis and S. capitis. The authors proposed that this suggest that the capacity of from FD patients’ PBMCs to release cytokines may be impaired due to a chronic exposure to exogenous stimuli.  It may be that chronic stimulation of the immune system leads to some kind of ‘immune exhaustion’.

Moreno-Arrones et al. 2021 FD Study

This 2023 study complements the Spanish team’s 2021 study. There, the authors studied 5 patients with FD and 5 with lichenoid folliculitis decalvans (also called lichen planopilaris folliculitis decalvans phenotypic spectrum).   Half of the patients in the study had a typical pattern of FD (eg, pustules or yellowish purulent exudate), and 5 patients presented FD associated with lichenoid features ( ie LPPFDPS)

Statistically significant increased levels of S aureus were found in patients with a typical FD pattern compared with patients with FDLPPPS. Levels of S aureus were greater than 20% of the total follicular bacterial composition in typical FD patients but  less than 20% in FDLPPPS patients. All in all the authors found that  in typical neutrophilic FD, Staphylococcus bacteria were the best microbiologic biomarker to help distinguish FD from FDLPPS.

In their 2021 study,  the S. aureus bacteria from patients with FD did not show resistance patterns indicating that they are not more virulent. They may grow either because of local immune dysfunction or altered epidermal barrier function or a combination of both.

The authors showed that certain bacteria were more likely to be found in classic neutrophilic FD and certain other bacteria were more likely to be found in LPPFD. Therefore, the authors propose that there is a “signature” of organisms that can be used to denote classic FD and a different signature to denote LPPFDPS

Why occur does this occur?

The authors state that all the factors are not clear why this all occurs. However there may be altered skin or hair follicle related barrier function in FD or some kind of local immune dysfunction. Normally, the epidermal barrier would prevent the colonization of anaerobic bacteria, but if this barrier were altered dysbiosis could occur due to the colonization of opportunistic bacteria. If there was a  local immune dysfunction, dysbiosis could occur. Both mechanisms could be relevant to FD

The authors here propose that the dysbiosis could help local bacteria like Staphylococcus proliferate in the skin of FD but it’s unlikely that a single bacteria is solely responsible for the development of the disease.  The authors suggest that the impact of S. aureus on the development of FD may not be as significant as previously thought as its role may be purely opportunistic.

The Biphasic presentation of FD

From their 2021 and 2023 studies, the authors propose a “biphasic” presentation of FD. In stage 1 or the “neutrophilic stage” of the disease, an progressive increase in colonization of S. aureus (caused by a hypothetical follicular structural alteration or immune dysfunction) would provoke a neutrophil mediated inflammatory response, that in turn damages the hair follicle. As a result of this damage there could be a collapse of the follicular immuneprivilege with resulting exposure of follicular antigens that  triggers a lichenoid response against the hair follicle. In this stage 2 or “lichenoid stage” of the disease, there is a mixed inflammatory (aerobic and anaerobic bacteria), but without significant predominance of S. aureus.

There is also some kind of immune exhaustion that happens as well. The immune cells are less able to respond to bacterial antigens.  This impaired immune response in turn may complicate the clearance of pathologic bacteria leading to a chronic condition.

This may have important impact on treatment. In stage 1 “neutrophilic stage” treatments might include anti-staphylococcal agents. Doxycycline, clindamycin, rifampin are well known agents used in FD. In stage 2 of the FDLPPPS stage, immunosuppressive treatments might receive the priority over antibiotics to limit hair destruction and continued autoimmune response. Treatments at this stage include topical steroids, steroid injections, doxycycline, hydroxychloroquine, tofacitinib and tacroliumus.

REFERENCE

Moreno-Arrones OM et al. Folliculitis decalvans has a heterogeneous microbiological signature and impaired immunological response. Dermatology. 2023 Jan 30.

Moreno-Arrones et al. Folliculitis decalvans microbiologic signature is specific for disease clinical phenotype. J Am Acad Dermatol. 2021 Nov;85(5):1355-1357.