Introduction

Human skin provides a habitat for various microorganisms that stably maintain communities through commensal relationships. The symbiotic relationships between the skin and the microbiome produce complex protective barriers against external environmental factors. Diverse metabolites produced by skin microbiome provide favorable efficacy to human skin [1]. In this study, we collected skin-microbiome samples from two different age groups of females. One group was in their 20 s, and the other was in their 40 s. Analysis of 16S rRNA gene sequences showed a newly found bacterium that shares 93.4% homology with the genus Sporichthya, indicating the discovery of a novel genus. We isolated a novel bacterial stain, Epidermidibacterium keratini, called EPI-7T [2], from skin samples. Additionally, the younger skin appeared to have high proportion of EPI-7T, while the older skin had no EPI-7T but rather other types of bacteria. Skin probiotic strain EPI-7T stained gram-positive, was aerobic and heterotrophic, and consisted of non-motile, non-spore-forming, rod-shaped cells (Fig. 1). Good growth was obtained on R2A agar but not on NA, ISP 2, or TSA.

Fig. 1
figure 1

Morphology of strain EPI-7T viewed by SEM

Full size image

Skin microorganisms produce various metabolites and influence skin cells directly or indirectly [3]. Among these microorganisms, the most common are of the genus Staphylococcus, which has been reported to be involved in Toll-like receptor (TLR) signaling in inflammation and wound regeneration [4, 5]. Indeed, several species in the genus Staphylococcus have been linked to various inflammatory diseases of the skin; however, no study has linked these organisms to aging. In addition, no report has described the relationships between aging and skin microorganisms. Therefore, in the current study, we analyzed the distribution of EPI-7T by age and investigated mechanisms related to aging.

Experimental

Materials

Epidermidibacterium keratini EPI-7 in R2A medium was provided by COSMAX R&I Center, Seongnam, Republic of Korea, in May 2018. A voucher specimen (KHU-NPCL-201805) has been deposited at the Laboratory of Natural Products Chemistry, Kyung Hee University.

General experimental procedures

Instruments and chemicals used in this study were prepared based on the previously described methods [6].

Extraction and isolation

EPI-7T culture solutions grown in R2A medium (50 L) were centrifuged, filtrated, and evaporated under reduced pressure. The concentrates were extracted in 80% aqueous MeOH (500 mL × 3). The combined concentrates (144 g) were poured into H2O (2.2 L) and successively extracted with EtOAc (2.2 L × 3) and n-BuOH (2.0 L × 3). Each layer was concentrated under reduced pressure to obtain EtOAc (EPE, 10.7 g), n-BuOH (EPB, 42.8 g), and H2O (EPH, 90.5 g) fractions. Fraction EPE (10.7 g) was applied to silica gel (SiO2) column chromatography (c.c.) (Φ 5.5 × 35 cm) and eluted with EtOAc-n-BuOH–H2O (20:3:1 → 7:3:1, 2 L of both) with monitoring by TLC to provide 12 fractions (EPE-1 to EPE-12). Fraction EPE-3 [168.0 mg, elution volume/total volume (Ve/Vt) 0.200–0.250] was subjected to octadecyl SiO2 (ODS) c.c. [Φ 2 × 10 cm, MeOH–H2O (2:1), 400 mL] to yield four fractions (EPE-3-1 to EPE-3-4) and a purified compound 1 [EPE-3-1, 20.5 mg, Ve/Vt 0.000–0.225, TLC (SiO2) Rf 0.70, EtOAc-n-BuOH–H2O (15:3:1), TLC (ODS) Rf 0.88, MeOH–H2O (2:1)].

1,1′-Biuracil (1) White amorphous powder; IR (KBr) νmax 1715, 1674, 1418 cm−1; low resolution positive ESI/MS m/z 113 [M/2 + 1]+, 223 [M + 1]+, 267 [M + 2Na-H]+; High resolution positive ESI/MS m/z 223.0459 [M + 1]+ (Calcd. for C8H7N4O4 223.0467); 1H- and 13C-NMR (600 and 150 MHz, DMSO-d6, δH and δC), see Table 1.

Table 1 1H- and 13C-NMR data of 1,1′-biuracil (1) and uracil (2) (600 MHz, DMSO-d6, δH; 150 MHz, DMSO-d6, δC)
Full size table

Cell culture and treatment

The human fibroblast cell line (Hs68) was purchased from the American Type Culture Collection (Manassas, VA, USA). The cells were cultured in Dulbecco’s modified Eagle’s medium supplemented with 1% Antibiotic Antimycotic Solution (DMEM; HyClone Laboratories, Inc., Logan, UT, USA) and 10% fetal bovine serum at 37 °C in an atmosphere of 5% CO2. For UV irradiation and treatment, Hs68 cells were seeded at 80% confluence into 6-well plates and incubated in an atmosphere of 5% CO2 at 37 °C. After 24 h, the cells were washed once with phosphate-buffered saline (PBS) and placed in fresh PBS. Next, 12 mJ/cm2 of UVB (wavelength 290–320 nm, maximum peak 311 nm) was applied in the presence of crosslinker (UVP; Upland, CA, USA), and then EPI-7 (0.1– 1%) or 1,1′-biuracil (0.1–10 ppm) was administered into the cells through serum-free medium for 24 h.

Cell viability test

Hs68 cells were seeded in 48-well plates, incubated for 24 h, and then treated with the indicated concentrations of EPI-7 or 1,1′-biuracil for another 24 h. After removing the culture medium, 3-(4, 5-dimethylthiazol-2-yl)-2, 5-diphenyltetrazolium bromide (MTT) solution was added to each well and incubated for 4 h. After termination of the reaction, the medium was discarded, and dimethylsulfoxide (DMSO) was added to dissolve the formazan crystals. Optical density (OD) values were measured at 570 nm using a microplate reader and normalized to that of the control.

RNA isolation and real-time PCR

Total RNA was isolated from cells using TRIzol reagent according to the manufacturer’s instruction (TaKaRa, Shiga, Japan). cDNA was synthesized from 1 µg of total RNA using Reverse Transcription Premix (Elpis-biotech, Daejeon, Korea) under the following reaction conditions: 45 °C for 45 min and 95 °C for 5 min. Gene expression signals were quantified with real-time PCR, and the data were analyzed using StepOne PlusTM system software (Applied Biosystems, Foster City, CA, USA). Real-time PCR amplification reactions were performed using SYBR Green PCR Master Mix with premixed ROX (Applied Biosystems, Foster City, CA, USA). The following primer pairs (Bioneer, Daejeon, Korea) were used in the reactions performed in an ABI 7300 following the manufacturer’s protocol: β-actin (F: 5′-GGCCATCTCTTGCTCGAAGT-3′ and R: 5′-GACACCTTCAACACCCCAGC-3′), type I procollagen (F: 5′-CTCGAGGTGGACACCACCCT-3′ and R: 5′-CAGCTGGATGGCCACATCGG-3′), fibrillin (F: 5′-AATGTCAGACGAAGCCAGGG-3′ and R: 5′-GATTTGGTGACGGGGTTCCT-3′), MMP-1 (F: 5′-CGAATTTGCCGACAGAGATGA-3′ and R: 5′-GTCCCTGAACAGCCCAGTACTT-3′), MMP-3 (R: 5′-ATTCCATGGAGCCAGGCTTTC-3′ and R: 5′-CATTTGGGTCAAACTCCAACTGTG-3′). The reaction conditions were as follows: initiation at 50 °C for 2 min and 95 °C for 10 min, followed by 40 cycles of 95 °C for 10 s and 60 °C for 1 min. The expression of β-actin was used as an internal control.

Results and discussion

Concentrated R2A medium including E. keratini EPI-7T were successively partitioned into EtOAc, n-BuOH, and aqueous fractions. Repeated SiO2 and ODS c.c. of the n-BuOH fraction yielded one new uracil derivative.

Compound 1, white amorphous powder, showed an IR absorbance band of carboxyl (1715, 1674 cm−1), aromatic (1674 cm−1), and amide group (1418 cm−1). IR spectrum of it was very similar to uracil with the exception for the absence of an amide moiety [7]. Its molecular weight and molecular formula were respectively 222 Da and C8H6N4O4 from the molecular ion peak m/z 223.0459 [M + 1]+ (Calcd. for C8H7N4O4 223.0467) in positive HR ESI/MS. The four carbon signals at δC 164.3 (s, C-4,4′), 151.5 (s, C-2,2′), 142.2 (d, C-6,6′), and 100.2 (d, C-5,5′) in the 13C-NMR spectrum suggested that compound 1 was very similar to uracil (2), one of the pyrimidine bases of nucleic acids [8, 9]. The carbon signals of two amide carbonyls [δC 164.3 (C-4,4′) and δC 151.5 (C-2,2′)], a nitrogenated olefine methine [δC 142.2], and an olefine methine [δC 100.2] were observed. The above mentioned 13C-NMR and ESI/MS data suggested compound 1 to be a dimer of uracil (2, molecular weight 112 Da) with a symmetrical structure. The 1H-NMR spectrum (600 MHz, DMSO-d6, δH) showed amine (δH 10.96, 2H, br. s, H-NH-3,3′) and aromatic (δH 7.38, 2H, d, J = 7.2 Hz, H-6,6′; δH 5.40, 2H, br. d, J = 7.2 Hz, H-5,5′) proton signals. As shown in Fig. 2, the uracil (2) showed amine proton signals at two different chemical shifts in the 1H-NMR spectrum, while compound 1 had only one chemical shift, indicating the uracil dimer was linked through an N–N linkage. J values (7.2 Hz) of the nitrogenated aromatic proton signals (H-6,6′) confirmed the proton signals to show only one 3J coupling in the heterocyclic structure, which also indicated an N–N linkage between NH-1 and NH-1′. As shown in Fig. 2, the nitrogenated aromatic proton signals (H-6,6′) in uracil (2) were split as dd through two 3J couplings with H-5 and NH-1. Also, the gHMBC spectrum (Additional file 1: Fig. S5) showed correlations between the nitrogenated aromatic proton signal (H-6,6′) and two amide carbonyl carbons [(C-4,4′) and (C-2,2′)] as well the olefine methine carbon (C-5,5′) and between the aromatic proton (H-5,5′) and the amide carbonyl carbon (C-4,4′) as well the nitrogenated olefine methine carbon (C-6,6′). Based on these findings, compound 1 was identified as 1,1′-biuracil, which was revealed to be a new compound. A similar pyrimidine compound, 5,5′-biuracil, has been reported as a synthetic compound [10].

Fig. 2
figure 2

1H-NMR spectrum of 1,1′-biuracil (1) and uracil (2) (600 MHz, DMSO-d6)

Full size image

To determine the effects of EPI-7T culture solutions and 1,1′-biuracil, we first examined the effects of EPI-7T culture solutions and 1,1′-biuracil on cell viability of Hs68 human fibroblasts using MTT assay (Fig. 3), which showed no cytotoxicity at concentrations less than 1% and 10 ppm, respectively.

Fig. 3
figure 3

Cell viability of EPI-7T culture solutions and 1,1′-biuracil in Hs68 human fibroblasts

Full size image

Next, we evaluated the expression levels of MMP-1 and the skin aging-associated factors type I procollagen, fibrillin, in UV-irradiated fibroblasts [11, 12]. Hs68 human fibroblasts were irradiated with UVB (12 mJ/cm2) and then treated with EPI-7T culture solutions ranging from 0.1 to 1% or 1,1′-biuracil ranging from 0.1 to 10 ppm for 24 h. The mRNA expression levels were measured using RT-qPCR. As shown in Fig. 4, EPI-7T culture solutions increased type I procollagen and fibrillin mRNA expression, which were suppressed by UV irradiation. In addition, the mRNA expression level of MMP-1, the major collagen-degrading proteinase, was significantly reduced by EPI-7T culture solutions. These results indicated that EPI-7T culture solutions exerted skin anti-aging effects.

Fig. 4
figure 4

The effect of EPI-7T culture solutions on the regulation of anti-aging associated mRNA expressions in UV-irradiated fibroblasts

Full size image

Furthermore, we measured the mRNA expression levels of type I procollagen, fibrillin, and MMP-1 after treatment with 1,1′-biuracil in UV-irradiated Hs68 fibroblasts to clarify the anti-aging effects of 1,1′-biuracil derived from EPI-7T culture solutions. Application of 1,1′-biuracil did not regulate type I procollagen or MMP-1 mRNA expression (data not shown), whereas it significantly increased fibrillin mRNA expression and reduced that of MMP-3, the fibrillin-degrading proteinase [13], in the UV-irradiation condition (Fig. 5). Taken together, these results suggest that 1,1′-biuracil is a key molecule in EPI-7T culture solutions, exerting protective effects against UV-irradiated skin aging.

Fig. 5
figure 5

Anti-aging functions of 1,1′-biuracil in UV-irradiated fibroblasts

Full size image