The peak wavelength of the UV LED is 293 nm. Because of studies have shown that in all spectra, 290~300 nm UV wavelengths are most effective for vitamin D production in the equivalent model of human skin or human skin^[25-26. The LED irradiation array was constructed according to the experimental desired irradiation area (30 cm²) for the experimental rats. The aluminum substrate was used to dissipate heat, a rapid pressing tool and a defined position block were installed for fixing, the upper seal and the 120 mm× 25 mm gap pressure plate were installed, and the exhaust fans were installed at both ends to keep the temperature constant [Figure 1(a)]. The reflectance of the box was 1.3, and the power density of the UV radiation was 0.8 mW· cm^-2, delivered for a duration of 1 000 s, twice a week, for a total of 12 weeks. The energy density was selected according to the rats' minimum sub erythema dose for rats. The irradiation parameters was are listed in Table 1. The LED irradiation group received the UV LED irradiation treatment, while the sham operation group did not. After 12 weeks of irradiation, the rats in all groups were euthanized, and samples (right femur, liver and skin) were harvested.

Table 1

Table 1

Table 1 Irradiation parameters Parameter Wavelength/nm 293± 10 Power density/(mW· cm-2) 0.8 Application time/s 1 000 Energy density/(J· cm-2) 0.8 Number of applications Twice a week Energy deposited per cm2/J 0.8 Distance of irradiation to skin/cm 1

Table 1 Irradiation parameters

60 healthy 3-month-old female SD rats (180~200 g) were purchased from the Experimental Animal Center of the General Hospital of the Chinese People's Liberation Army. The room temperature of the animal feeding environment was constant [(20± 2) ℃], the relative humidity was 40%, and the ventilation was adequate. Rats were housed with water ad libitum, in a 12:12 hour light-dark cycle with protection from UVB exposure until the end of the study. According to the principle of random grouping, the 60 osteoporotic rats were randomly divided into unovariectomized (n=24) and ovariectomized (n=36). The unovariectomized rats were then divided into the sham operation group (Sham, n=12) and the detection model group (Sham, n=12). The ovariectomized rats were divided into either the LED irradiation group (LED, n=12), the no treatment group (OVX, n=12), or the detection model group (OVX, n=12). Except for the sham operation group, all groups underwent a bilateral ovariectomy and were fed with no vitamin D supplementation, 0.50% calcium and 10.00% total fat feed. The sham operation group only had a small amount of adipose tissue removed from the side of the ovaries, did not undergo an ovariectomy, and was fed with standard feed following the operation. During the operation, hair was sheared from the back to form a 2 cm× 10 cm area with hair removal, which was disinfected with iodophor and anesthetized with an intraperitoneal injection of 2% pentobarbital sodium. After that, a longitudinal incision was made along the midline of the back, about 2 cm long. After the skin was cut, the pink ovaries and tightly connected uterine horns located at both sides' back and lower part of the kidneys could be seen. The uterine horn was ligated and cut off, the ovaries were removed, and the back fascia, muscles and skin were sutured^[24]. After 3 months, the bone microarchitecture and 25-hydroxyvitamin levels in the rats in the model construction test group were analyzed to confirm the successful generation of an osteoporosis rat model.

Blood samples were obtained from the caudal vein of each rat at 0, 4, 8 and 12 weeks after irradiation. Blood was collected in centrifuge tubes and centrifuged at 3 000 r· min^-1 for 15 min. After static placement, the separated plasma supernatant was harvested and stored at -20 ℃ for later analysis. Serum PTH, serum P1NP, serum 25(OH)D₃, serum bone TRAP and serum ALP levels were measured using ELISA kits following the manufacturer's protocol (Enzyme-linked Biotechnology, Shanghai, China). Serum calcium and phosphorus concentrations were determined using an automatic biochemical analyzer (General Hospital of Chinese People's Liberation Army). Reference range: Serum calcium (2.0~3.2 nmol· L^-1), serum phosphorus (0.8~2.8 nmol· L^-1). Among the compounds analyzed, 25(OH)D₃ levels were analyzed at 0, 4, 8 and 12 weeks post-irradiation, and all other metabolic indexes were detected at 0 weeks and 12 weeks post-irradiation. All analyses were performed in duplicate.

Following euthanasia, rat livers were fixed with 4% paraformaldehyde, and total RNA was extracted using the Trizol method (Invitrogen, California, USA). After determining the RNA concentration, cDNA was synthesized via a reverse transcription kit (Promega, Wisconsin, USA) according to the reverse transcription system and conditions. The relative mRNA level in the expressed samples was detected using real-time fluorescence quantitative PCR. The primers for CYP27A1 were designed according to Going, Alexandrova^[27].

At the end of all experiments, the rats were euthanized and their backs were irradiated; 0.2 g of skin tissue was harvested and soaked in 4% paraformaldehyde solution prepared in advance. Each sample was photographed at 10 randomly selected locations, and the histological changes were analyzed^{[28, 29]}. The rat skin tissue was embedded into wax blocks for immunohistochemical (IHC) analysis, and the primary antibodies used for IHC were an anti-p53 antibody (Abcam, Cambridge, United Kingdom).

The data are presented as the average ± standard deviation. SPSS v24.0 software (IBM, Armonk, NY, USA) was used for statistical analysis. Statistical significance was determined using one-way ANOVA, and the differences between mean values were determined using the minimum significance (LSD) method. p< 0.05 was considered statistically significant.

A deep UV (DUV) LED irradiation array is fabricated in this work. The array consists of 30 AlGaN-based LED chips. Each with a dimension of 3.5 mm× 3.5 mm. The DUV LED module has a peak wavelength of (293± 5) nm with an FWHM of 10 nm. Turn on voltage and optical power of the DUV LED are about 5.5 V and 2 mW at 20 mA, respectively. As illustrated in Fig.1, the LED chips are placed on aluminum-based heatsink with a thickness of 2 mm by eutectic bonding to obtain the UV irradiation array. In addition, the current injection level of the array was adjusted to obtain a UV irradiation power density of 0.8 mW· cm^-2 at a distance of 1 cm.

Fig.1

	Figure Option ViewDownloadNew Window
	Fig.1 (a) Schematic of rat irradiation devise with LED module, (b) Relative spectral irradiance of prepared LED modules

In order to determine the full width and half maximum wavelengths of the LED irradiation device, we used an SPD-200 UV radiometer (Sensing, Hangzhou, China) to measure the spectrum of each LED module. This way, could determine where the peak is 293 nm, and the full width at half maximum is 10 nm (2b), which met our requirements for this study. In order to characterize our rat osteoporosis model, we analyzed the level of 25-hydroxyvitamin and the bone microarchitecture of rats before and 3 months after bilateral ovariectomy. 25(OH)D₃ is present at a high concentration in the blood and has a long half-life and is stable nature, so it is a reliable index to reflect the vitamin D levels in the body. As shown in Figure 2. The 25(OH)D₃ level in OVX was significantly lower than that of Sham (p< 0.05). The BV/TV, Tb.N and Tb.Th in OVX significantly decreased, while BS/BV and Tb.Sp significantly increased when compared with the Sham (p< 0.05). These results indicated that we had successfully established a rat osteoporosis model.

Fig.2

	Figure Option ViewDownloadNew Window
	Fig.2 We had successfully established a rat osteoporosis model (a): 25(OH)D3; (b): Bone surface area/bone volume; (c): Trabecular thickness; (d): Trabecular separation; (e): Bone volume/total volume; (f): Trabecular number; * p< 0.05

Serum 25(OH)D₃ is a reliable index reflective of the vitamin D levels in the body. As shown in [Figure 3(a)], 4 weeks into the UV LED irradiation treatments, the serum 25(OH)D₃ level in the LED irradiation group increased significantly when compared to control rats (p< 0.05, all groups). With continued irradiation treatments through 8 and 12 weeks, the serum 25(OH)D₃ level continued to increase and was maintained at a high level (p< 0.05, all groups). However, rats in the OVX6 not exposed to UV LED irradiation had a low level of serum 25(OH)D₃ at 4, 8, and 12 weeks, while the Sham6 group maintained a normal level of serum 25(OH)D₃ at 4, 8 and 12 weeks.

Fig.3

	Figure Option ViewDownloadNew Window
	Fig.3 Effect of irradiation on the serum 25(OH)D3 and expression of vitamin D 25-hydroxylase (a): 25(OH)D3; (b): Relative 25-hydroxylase mRNA; * p< 0.05

Serum TRAP is secreted by osteoclasts and is often used as a bone resorption marker. Serum PTH is an important hormone that regulates calcium and phosphorus metabolism in the human body. PTH levels are often observed when the bone resorption rate increases^[30]. As shown in Fig.4(a, b), serum PTH and serum TRAP levels decreased significantly following UV irradiation in the LED irradiation group compared to the unirradiated OVX group (p< 0.05). No significant changes were observed in the Sham group's serum PTH and serum TRAP levels before and after the experiment. Phosphorus and calcium are basic substances of bone mineralization and participates in the whole process of bone remodeling. As shown in Figure 4(c) and (f), no significant differences were observed in serum phosphorus and serum calcium between the LED irradiation group, Sham group and OVX group. Serum bone ALP and serum P1NP are markers reflecting bone formation and osteoblast activity^[31]. As shown in Figure 4(d) and (e), the serum bone ALP and serum P1NP levels in the LED irradiation group were significantly higher than those of the unirradiated OVX group (p< 0.05). There were no significant changes in serum bone ALP and serum P1NP levels between the OVX group and the Sham group before and after the experiment.

Fig.4

	Figure Option ViewDownloadNew Window
	Fig.4 (a) Parathyroid hormone; (b) Tartrate-resistant acid phosphatase; (c) Serum calcium; (d) Serum phosphorus; (e) Alkaline phosphatase; (f) Procollagen 1N-telopeptide; * p< 0.05

Vitamin D 25-hydroxylase can convert vitamin D into 25(OH)D₃. In order to detect changes indicative of alterations in 25(OH)D₃ metabolism, we studied the mRNA expression level of vitamin D 25-hydroxylase by RT-PCR. Relative expression levels in Sham and LED irradiation groups are expressed with reference to in the OVX group as 1.0. As shown in Figure 3(b), the mRNA level of vitamin D 25-hydroxylase in the LED irradiation group was significantly higher than that in the OVX group. The results showed that vitamin D converted from 7-dehydrocholesterol in UV-irradiated skin is likely sent to the liver, where elevated levels of vitamin D25-hydroxylase could, in turn, convert a large amount of vitamin D to 25(OH)D₃, which would explain the significant increase in serum 25(OH)D₃ observed in the rats.

Phospho-p53 are proteins whose levels are indicators of the incidence of DNA damage^[32]. We therefore, performed IHC analysis of these DNA damage markers in UV LED irradiated tissues from the osteoporosis rats and quantified the cumulative optical density for each marker. The results showed no significant difference in cumulative optical density for either DNA damage marker when comparing the UV LED irradiation group to controls (Figure 5), indicating DNA damage was not induced as a result of treatment.

Fig.5

	Figure Option ViewDownloadNew Window
	Fig.5 The effects of UV LED irradiation on skin health (a)— (c): IHC with anti-Phospho-p53 antibody for dermis and epidermis (original magnification 400× , bars indicate 100 μm); (d): Cumulative optical density of Phospho-p53 protein detected by IHC