Study population

An observational, analytical and prospective cohort study was performed in which 57 patients diagnosed with AGA were analysed. Having signed an informed consent form, a clinical history was compiled for each participant, and a general physical examination was performed. The inclusion criteria were patients diagnosed with AGA type III, based on the Ebling Classification for men. The exclusion criteria were patients with scarring alopecia, hypotrichosis, telogen effluvium, alopecia areata, and anagen effluvium. The mean age of the patients was 42.31 ±3.01 years.

Ethics and consent to participate

This study was carried out in accordance with the basic ethical principles – autonomy, beneficence, non-maleficence and distributive justice – and its development followed the rules of Good Clinical Practice, the principles set forth in the last Declaration of Helsinki (2013) and the Convention from Oviedo (1997). All pathological, clinical or personal data were anonymized and separated from any personal identifiers. The patient was duly informed, and informed consent of each subject was requested in writing.

Obtaining and processing the samples

Two tissue biopsies were taken from the same patient with a 6-mm punch: 1 from a control area in which there were healthy scalp follicles that were clinically and microscopically normal (generally occipital) (area C) and 1 from the area affected by alopecia (area A). These fragments were each placed in a sterile tube containing MEM (Minimum Essential Medium) with 1% antibiotic/antimycotic (both from Thermo Fisher Scientific, Waltham, MA, USA). The transfer and processing of the tissue samples were performed in all cases within 6 h after collection.

In the laboratory, the samples were processed in a class II Telstar AV 30/70 Müller 220 V 50 MHz laminar flow hood (Telstar SA Group, Terrassa, Spain), ensuring a sterile environment. The samples conserved in MEM were used for histological studies and cell isolation for in vitro studies. The samples were washed/hydrated several times with MEM without antibiotic to remove blood cells and were cut into fragments that were stored in F13 (60% ethanol, 20% methanol, 7% polyethylene glycol, and 13% distilled H₂O).

Cell isolation and culture

Under sterile conditions, the tissue segments were washed several times with MEM under sterile conditions and then opened longitudinally. After removing the adipose tissue, the sample was cut into small explants (1 mm²). Subsequently, the explants were digested in 0.1% type I collagenase (Worthington) in MEM (1 h at 37°C) under constant stirring. The enzyme reaction was stopped by adding the same volume of culture medium; then, the solution was centrifuged at 200 g for 7 min, and the medium was discarded. These explants were placed in 25 cm² Roux flasks (Nyclon-Intermed; Nunc A/S, Roskildo, Denmark) in which 0.5 ml of Amniomax complete medium (Gibco BRL, Life Technologies Carlsbad, CA, USA) was added to maintain moisture on the cultivation surface and to improve explant adherence. The culture flasks were then incubated vertically at 37°C in a 5% CO₂ environment in a culture oven for 2 h. Then, 2.5 ml of Amniomax was added to the flasks, and the flasks were incubated horizontally under the above conditions. Care was taken to avoid movements that could cause the explant to detach. The culture medium was carefully replaced twice a week, following established protocols for primary cultures [24].

Once the cells had grown to confluence, the cells were subcultured. This involved removing the medium and rinsing the cells 3 times in 2 ml of Hank’s balanced salt solution (Gibco BRL, Life Technologies), followed by the addition of 2 ml of trypsin-ethylenediaminetetraacetic acid solution at 1 : 250 (Gibco BRL, Life Technologies) and incubation at 37°C for 5 min. The enzymatic reaction was stopped by adding 4 ml of culture medium. The resulting cell suspension was centrifuged at 200 g for 7 min, and the cell pellet was resuspended in 9 ml of Amniomax medium. The cells in suspension were cultured at a density of 3 ml per 25 cm² Roux flask until confluence was obtained in an incubator with an atmosphere humidified with 5% CO² at 37°C. The cells were maintained for 90 days to observe their viability.

Histological studies

After fixation, the samples were dehydrated and embedded, as follows. The tissue was washed with distilled water for 20 min and then dehydrated by serial incubations in an ascending alcohol gradient, followed by standardized paraffin-embedding protocols for subsequent histopathological study [25]. After dehydration, paraffin blocks were made using moulds. Once the paraffin was solidified, an HM 350 S rotation microtome (Thermo Fisher Scientific, Massachusetts, USA) was used to obtain 5 µm-thick sections, which were spread in a hot water bath and collected on glass slides treated with a solution of 10% polylysine to facilitate the adhesion of the sections.

Protein expression studies

Protein expression analysis was performed by detecting antigen-antibody reactions (avidin-biotin complex) using chromogen peroxidase and phosphatase according to the following protocol. 1. The samples were washed 3 times with 1× phosphate-buffered saline (PBS) for 5 min each wash. 2. Non-specific binding sites were blocked with 3% bovine serum albumin (BSA) in PBS for 30 min at room temperature. 3. The samples were incubated with primary antibody diluted in 3% BSA and PBS overnight at 4°C (Cytokeratin 15: Abcam (AC-0018), dilution 1 : 400, 100% Triton 0.1% in PBS, 10 min, before incubation with blocking solution; Tropoelastin: Dr. Mecham (Washington University, EEUU), dilution: 1 : 750; p15/p16: Santa Cruz Biotechnology (sc-377412), dilution 1 : 50, 10 mM sodium citrate pH = 6 before incubation with blocking solution). 4. The samples were rinsed in PBS 3 times for 5 min each wash. 5. The samples were incubated with biotin-conjugated secondary antibody diluted in PBS for 1.5 h at room temperature (IgG (Rabbit) Sigma-Aldrich (B5283), dilution 1 : 1000; IgG (Mouse) Sigma-Aldrich (B0529), dilution 1 : 300). 6. The samples were washed with PBS 3 times for 5 min each wash. 7. To detect tropoelastin and p15/16, the samples were incubated with the avidin-peroxidase conjugate ExtrAvidin-Peroxidase (Sigma-Aldrich, St. Louis, MO, USA) for 60 min at room temperature. To detect CK15, the samples were incubated with an avidin-phosphatase conjugate (ExtrAvidin-Alkaline Phosphatase, Sigma-Aldrich, St. Louis, MO, USA) under the same conditions. 8. The samples were washed in PBS 3 times for 5 min each wash. 9. To visualize tropoelastin and p15/16, samples were incubated in the chromogenic substrate diaminobenzidine (Kit DAB, SK-4100) (Vector Laboratories, Burlingame, CA, USA). The chromogenic substrate was prepared immediately before exposure (5 ml of distilled water, 2 drops of buffer, 4 drops of DAB, and 2 drops of hydrogen peroxide). This technique results in brown staining. To visualize CK15, samples were incubated in alkaline chromogenic substrate for 15 min. 10. The samples were washed with distilled water 3 times for 5 min each wash to stop the reaction. 11. For nuclear staining, the samples were incubated in Carazzi’s haematoxylin for 5–15 min. 12. The samples were then rinsed in running water for 10 min and mounted in aqueous medium (Plasdone).

For all immunohistochemical studies, sections of the same tissue were used as a negative control, in which the incubation with primary antibody was substituted with incubation in blocking solution.

Statistical analysis

For each of the patients in the established groups, 5 sections and 10 fields per section were randomly selected. Patients were described as positive when the marked mean area in the analysed sample was greater than or equal to 5% of the total, following the anatomopathological protocol of Remmele and Schicketanz [26] and Cristóbal et al. [27]. The preparations were examined under a Zeiss Axiophot optical microscope (Carl Zeiss, Germany) equipped with an AxioCam HRc digital camera (Carl Zeiss, Germany). For the statistical analysis, GraphPad Prism 6.0 was used. The Mann-Whitney U test was applied. Data are expressed as the mean ± standard error of the mean (SEM). Significance was established at a value of p < 0.05 (*).

Cell viability

In vitro studies of fibroblasts from the scalp of patients showed differential behaviour with respect to their viability in cell culture conditions. It was observed that cells from areas with alopecia (A) showed a significantly higher percentage of dead cells after 90 days (17.00 ±1.09% C vs. 28.50 ±1.41% A, *p < 0.05). The cells from the control area (C) showed long-term cell viability characterized by the preservation of cell morphology and increased multilayer growth. These characteristics were not observed in area A, where cells grew in a monolayer (Figure 1).

Figure 1

Quantification of the percentage of dead cells in cell culture at 90 days in the control area (C) and in the area with alopecia (A). p < 0.05 (*). B, C – Representative images of cell morphology under in vitro conditions at 90 days in the control area (B) and in the area with alopecia (C)

Cytokeratin 15 expression

The study of CK15 protein expression by immunohistochemical techniques showed significant differences in relation to its expression in different parts of the HF and other skin appendages in areas C and A.

CK15 protein expression was significantly lower in the HF of area A (area A: 0.73 ±0.13; area C: 1.96 ±0.23; *p < 0.05) (Figure 2). For the hair bulb (HB), no differences in CK15 expression were observed between the 2 areas (0.41 ±0.04 C vs. 0.32 ±0.04 A). In the sebaceous gland (SG), CK15 expression was significantly lower in area A compared to area C (1.87 ±0.32 C vs. 0.31 ±0.09 A, *p < 0.05) (Figure 2).

Figure 2

Protein expression of cytokeratin 15 (CK15) in the hair follicle (HF), hair bulb (HB), sebaceous gland (SG) and epidermis (EP) in the control area (C) and in the area with alopecia (A). p < 0.05 (*). Protein expression images of CK15 in the HF and HB of the control area (A–C) and area with alopecia (E), as well as in the EP of area C (D) and A (F). Arrow: immunoprecipitated

The EP of the area near the outer root sheath was studied, and CK15 protein expression in area A was significantly lower (2.25 ±1.29 C vs. 1.29 ±0.11, *p < 0.05). The histological study revealed that the lower CK15 expression in the EP of area A appeared as small accumulations in the deeper epidermal region (Figure 2). No expression greater than 5% was observed in the dermis of either area and thus was considered negative expression.

Tropoelastin expression

Immunohistochemical techniques were used to study TE protein expression in the HF as well as other skin appendages. The histological analysis revealed differences in expression not only between areas C and A but also between different tissue structures (HF, HB, SG and EP). TE protein expression in the HF of area A, compared with that in the HF of area C, was significantly lower (0.96 ±0.16 C vs. 0.46 ±0.08 A, *p < 0.05) (Figure 3). For the HB, no significant differences were observed in terms of protein expression between the 2 study groups (0.27 ±0.09 C vs. 0.15 ±0.06 A). The mean TE expression in the HB was the lowest of all the structures studied. TE expression in the SG of area A was significantly lower (0.75 ±0.09 C vs. 0.18 ±0.05 A, *p < 0.05), as was TE expression in the EP of area A (1.27 ±0.16 C vs. 0.79 ±0.11 A, *p < 0.05) (Figure 3). TE expression was observed in the papillary dermis, with a mean of 0.81 ±0.12 in area C. Area A did not show expression greater than 5%; therefore, its expression was considered negative (Figure 3).

Figure 3

Protein expression of tropoelastin (TE) in the hair follicle (HF), hair bulb (HB), sebaceous gland (SG) and epidermis (EP) in the control area (C) and in the area with alopecia (A). p < 0.05 (*) Protein expression images of TE in the HF and HB of the control area (A, B) and in the affected area (D, E), as well as in the EP of area C (C) and A (F). Arrow: immunoprecipitated

Expression of cellular senescence markers

The study of cellular senescence marker protein expression was performed via the evaluation of p15/p16. p15/p16 protein expression in area A increased as a function of the different parts. The score was significantly higher in the HF of area A (0.854 ±0.095 C vs. 1.396 ±0.183 A, *p < 0.05); similarly, significantly higher p15/p16 protein expression was observed in HB (0.396 ±0.125 C vs. 1.115 ±0.143 A, *p < 0.05) and in EP (0.860 ±0.129 C vs. 1.292 ±0.137 A, *p < 0.05) (Figure 4). No increase in the p15/p16 cellular senescence marker expression was observed in the SG of these patients (0.979 ±0.048 C vs. 1.146 ±0.078 A).

Figure 4

Protein expression of senescence markers (p15/p16) in the hair follicle (HF), hair bulb (HB), sebaceous gland (SG) and epidermis (EP) in the control area (C) and in the area with alopecia (A). p < 0.05 (*) Protein expression images of p15/ p16 in the HF and HB of the control area (B) and in the affected area (C–E), as well as in the EP in area C (A) and A (F). Arrow: immunoprecipitated