Novel UV‐B Phototherapy With a Light‐Emitting Diode Device Prevents Atherosclerosis by Augmenting Regulatory T‐Cell Responses in Mice

Abstract

Background

Ultraviolet B (UV‐B) irradiation is an effective treatment for human cutaneous disorders and was shown to reduce experimental atherosclerosis by attenuating immunoinflammatory responses. The aim of this study was to clarify the effect of specific wavelengths of UV‐B on atherosclerosis and the underlying mechanisms focusing on immunoinflammatory responses.

Methods and Results

Based on light‐emitting diode technology, we developed novel devices that can emit 282 nm UV‐B, which we do not receive from natural sunlight, 301 nm UV‐B, and clinically available 312 nm UV‐B. We irradiated 6‐week‐old male atherosclerosis‐prone Apoe ^−/− (apolipoprotein E‐deficient) mice with specific wavelengths of UV‐B and evaluated atherosclerosis and immunoinflammatory responses by performing histological analysis, flow cytometry, biochemical assays, and liquid chromatography/mass spectrometry‐based lipidomics. Irradiation of 282 nm UV‐B but not 301 or 312 nm UV‐B significantly reduced the development of aortic root atherosclerotic plaques and plaque inflammation. This atheroprotection was associated with specifically augmented immune responses of anti‐inflammatory CD4⁺ Foxp3 (forkhead box P3)⁺ regulatory T cells in lymphoid tissues, whereas responses of other immune cells were not substantially affected. Analysis of various lipid mediators revealed that 282 nm UV‐B markedly increased the ratio of proresolving to proinflammatory lipid mediators in the skin.

Conclusions

We demonstrated that 282 nm UV‐B irradiation effectively reduces aortic inflammation and the development of atherosclerosis by systemically augmenting regulatory T‐cell responses and modulating the balance between proresolving and proinflammatory lipid mediators in the skin. Our findings indicate that a novel 282 nm UV‐B phototherapy could be an attractive approach to treat atherosclerosis.

Nonstandard Abbreviations and Acronyms

CCR4

C‐C chemokine receptor 4

Foxp3

forkhead box P3

LC

liquid chromatography

LED

light‐emitting diode

LN

lymph node

MS

mass spectrometry

Teff

effector T cell

Th

T helper type

Treg

regulatory T cell

Clinical Perspective.

What Is New?

• By taking advantage of light‐emitting diode technology, we developed novel devices that emit 282, 301, or 312 nm ultraviolet B (UV‐B) and demonstrated that irradiation of 282 nm but not 301 or 312 nm UV‐B effectively reduces the development of aortic root atherosclerotic plaques and plaque inflammation in hypercholesterolemic apolipoprotein E‐deficient mice.

• The atheroprotective effect of 282 nm UV‐B was associated with specific augmentation of anti‐inflammatory regulatory T‐cell responses, minor effects on other immune responses, increased plasma vitamin D levels, and favorably modulated lipid mediator balance in the skin.

What Are the Clinical Implications?

• Considering that we cannot receive shorter 282 nm UV‐B from natural sunlight, a novel 282 nm UV‐B phototherapy could be an attractive immunomodulatory approach to treat atherosclerosis.

Coronary artery disease (CAD) and stroke arise from atherosclerosis and are major causes of mortality worldwide. Despite state‐of‐the‐art intensive treatment in patients at high risk of atherosclerotic disease, residual risk of the disease, which involves chronic inflammation of the arterial wall, ¹ should not be ignored. Notably, clinical evidence indicates the possibility that anti‐inflammatory therapies could be approaches for the prevention of CAD in patients with the past disease history. ² , ³ Immune responses via innate and adaptive immunity are critically responsible for vascular inflammation and the development of atherosclerotic disease. ⁴ A recent study using single‐cell proteomic and transcriptomic analyses demonstrated that T cells appear to be the major cellular components in human carotid artery plaques and display activated and differentiated phenotype, ⁵ indicating the critical role of activated T cells in atherosclerotic plaque development. Importantly, a recent clinical trial investigated CAD events in patients with cancer treated with immune checkpoint inhibitors and demonstrated that T‐cell activation leads to an increased risk of CAD, confirming the involvement of T‐cell–mediated immune responses in the development of human atherosclerotic disease. ⁶ Experimental studies have shown that effector T cells (Teffs) promote atherosclerosis by producing proinflammatory cytokines, ⁷ whereas Foxp3 (forkhead box P3)‐expressing regulatory T cells (Tregs), indispensable for mediating immunological tolerance and controlling excessive immune responses, prevent atherosclerosis through anti‐inflammatory actions ⁸ or lipid metabolism modulation. ⁹ A number of approaches using antibodies, ¹⁰ , ¹¹ , ¹² cytokines, ¹³ , ¹⁴ , ¹⁵ and vaccination with atherosclerosis‐related antigens ¹⁶ are effective for controlling atherosclerotic vascular disease by shifting the Teff/Treg balance toward Treg responses. However, clinical therapies to intervene in the T‐cell balance are unavailable so far, because of concerns about the safety and efficacy of these treatment strategies.

It is widely accepted that sunlight exposure is necessary for our health. Sunlight includes ultraviolet light that is categorized into 3 wavelengths such as UV‐A (320–400 nm), UV‐B (280–320 nm), and UV‐C (100–280 nm). Because it is considered that the atmosphere and the ozone layer block UV‐B shorter than 285 nm and UV‐C, we receive UV‐A and longer wavelengths of UV‐B from natural sunlight in daily life. Among these wavelengths of UV light, UV‐B has attracted much attention, because it is indispensable for the cutaneous synthesis of vitamin D and maintenance of the immune system. ¹⁷ Notably, it has been reported that UV‐B exerts anti‐inflammatory and immunosuppressive actions via production of anti‐inflammatory IL (interleukin)‐10 and vitamin D, activation of skin tolerogenic dendritic cells, and induction of Tregs. ¹⁷ Based on the unique nature of UV‐B irradiation, UV‐B‐based phototherapy is clinically used for the treatment of immunoinflammatory cutaneous disorders including psoriasis, atopic dermatitis, and cutaneous T‐cell lymphoma, despite its well‐known adverse effects such as premature skin aging and potential risk of skin cancer. ¹⁸ We recently reported that broadband UV‐B (a continuous spectrum from 280 to 320 nm with a peak around 313 nm) irradiation shifts the Teff/Treg balance toward Treg responses and attenuates the development of atherosclerosis and angiotensin II‐induced abdominal aortic aneurysm in hypercholesterolemic Apoe ^−/− (apolipoprotein E‐deficient) mice, providing an attractive approach for the treatment of atherosclerotic vascular disease. ¹⁹ , ²⁰ Considering that within the UV‐B spectrum, narrow‐band UV‐B (a narrow peak around 311 nm) and excimer light (308 nm) therapy are known to be effective for treating psoriasis, ¹⁸ different wavelengths of UV‐B may have divergent effects on immune responses and disease state. In the context of atherosclerosis, there might also be some specific UV‐B wavelengths that can effectively limit atherosclerotic lesion formation. In consideration of the deleterious effects due to chronic excessive UV‐B exposure described above, irradiation doses used for treatment should be minimized for its clinical application. Therefore, identification of appropriate UV‐B irradiation conditions including wavelengths and irradiation doses for the prevention of atherosclerosis could open an avenue for the clinical application of UV‐B therapy.

In this study, by taking advantage of light‐emitting diode (LED) technology that allows us to irradiate specific wavelengths of UV‐B, we developed novel devices that emit 3 specific wavelengths of UV‐B including 282 nm UV‐B, which we do not receive from natural sunlight, and clinically available 312 nm UV‐B. We investigated the effect of specific wavelengths of UV‐B on atherosclerosis and the underlying mechanisms in hypercholesterolemic Apoe ^−/− mice focusing on immunoinflammatory responses. We for the first time demonstrated that 282 nm UV‐B irradiation effectively limits atherosclerosis by systemically augmenting Treg responses and modulating the balance between proresolving and proinflammatory lipid mediators in the skin. We also found that clinically relevant 312 nm UV‐B has a potentially anti‐atherogenic action. Our data indicate that the effect of UV‐B on atherosclerosis differs depending on the wavelengths. Because we cannot receive shorter 282 nm UV‐B from natural sunlight, this UV‐B irradiation could be an attractive immunomodulatory approach to treat CAD.

Methods

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Animals

All mice were on a C57BL/6 background and fed a standard chow diet (CLEA, Tokyo, Japan). In total, 281 mice were used in this study. Apoe ^−/− mice were previously described. ¹¹ We housed mice in a cage for each treatment group in a specific pathogen‐free animal facility at Kobe Pharmaceutical University. All animal experiments were approved and registered by the Animal Care Committee of Kobe Pharmaceutical University (permit numbers: 2018–001, 2019–006, 2020–045, 2021–034, 2022–006, 2023–036) and conform to the National Institutes of Health Guide for the Care and Use of Laboratory Animals and the Animal Research: Reporting of In Vivo Experiments guidelines.

UV‐B Irradiation

LED lamps (Nikkiso Co., Ltd., Japan) that can emit 282, 301, or 312 nm wavelength of UV‐B were newly developed and used in this study (a narrow peak around 282, 301, or 312 nm for each LED lamp). The irradiance of 282, 301, or 312 nm UV‐B was 5.04, 4.20, 2.52 J/m² per second at a distance of 14 cm, respectively. The mice were placed 14 cm below the bank of lamps and were irradiated for indicated weeks after shaving their backs in the animal facility, whereas nonirradiated mice were not subjected to these procedures. Instead of the 282 nm wavelength of UV‐B, 284 nm wavelength of UV‐B (a narrow peak around 284) was also used in some experiments. We confirmed that the immunological actions of 284 nm UV‐B are similar to those of 282 nm UV‐B. The irradiance of 284 nm UV‐B was 5.88 J/m²/second at a distance of 14 cm. The UV‐B irradiation dose used in all experiments was 2 kJ/m². Engineering specialists (Nikkiso Co., Ltd.) carefully measured the wavelengths of the UV‐B irradiation devices and we confirmed the exact value of each UV‐B wavelength (Figure S1). Randomization and allocation concealment were performed. Littermate mice were equally allocated to each treatment group. During experiments, animal/cage location was not performed. Investigators were not blinded to treatment allocation. Criteria for exclusion was defined as serious burns, but during at least 3 observations per week we did not find such symptoms and no UV‐B‐irradiated mice were excluded. Experimental procedures other than UV‐B irradiation were performed in our laboratory rooms.

Assessment of Biochemical Parameters

After overnight fasting, blood was collected by the cardiac puncture under anesthesia by intraperitoneal injection of medetomidine hydrochloride (0.3 mg/kg), midazolam (4 mg/kg), and butorphanol tartrate (5 mg/kg) (all WAKO, Osaka, Japan). Plasma was obtained through centrifugation and stored at −80 °C until measurement. Concentrations of plasma total cholesterol, high‐density lipoprotein cholesterol, and triglyceride were determined enzymatically using an automated chemistry analyzer (SRL, Tokyo, Japan), and 1,25‐dihydroxyvitamin D and 25‐hydroxyvitamin D were analyzed by radioimmunoassay (SRL).

Atherosclerotic Lesion Assessment

To examine the effect of specific wavelengths of UV‐B on atherosclerosis, we irradiated 6‐week‐old Apoe ^−/− mice with each wavelength of UV‐B once weekly for 14 weeks and analyzed atherosclerotic lesions in the aortic root and thoracoabdominal aorta at 20 weeks of age. Mice were anesthetized as described and the aorta was perfused with saline. The aorta was dissected from the middle of the left ventricle to the bifurcation of the iliac artery. For aortic root lesion analysis, the samples were cut in the ascending aorta, and the proximal samples containing the aortic sinus were embedded in OCT compounds (Tissue‐Tek; Sakura Finetek, Tokyo, Japan). Five consecutive sections (10 μm thickness), spanning 600 μm of the aortic sinus, were collected from each mouse and stained with hematoxylin–eosin. Some sections were stained with Oil Red O (WAKO) for representative photomicrographs of the aortic sinus atherosclerotic lesions. Stained sections were digitally captured using a fluorescence microscope (BZ‐X810; KEYENCE, Osaka, Japan). For quantitative analysis of atherosclerosis, the total lesion area of 5 separate sections from each mouse was obtained with the use of the ImageJ (National Institutes of Health) as previously described. ¹⁹ For en face lesion analysis, the aorta was excised from the proximal ascending aorta to the common iliac artery bifurcation and fixed in 10% buffered formalin. After the adventitial tissue was carefully removed, the aorta was opened longitudinally, pinned on a black wax surface, stained with Oil Red O, and captured digitally with a digital camera. The percentage of stained lesion area per total area of the aorta was determined by ImageJ as described previously. ¹⁹

Histological Analysis of Atherosclerotic Lesions

Immunohistochemistry was performed on 4% paraformaldehyde‐fixed cryosections (10 μm) of mouse aortic roots using antibodies to identify macrophages (MOMA‐2, 1:400; BMA Biomedicals) and T cells (CD4, 1:100; BD Biosciences), followed by detection with biotinylated secondary antibodies and streptavidin‐horseradish peroxidase. Staining with Masson's trichrome was used to delineate the fibrous area. Stained sections were digitally captured using a fluorescence microscope (BZ‐810; KEYENCE), and the percentage of the stained area (the stained area per total atherosclerotic lesion area) was calculated. Quantification of CD4⁺ T cells was done by counting positively stained cells, which was divided by total plaque area.

Flow Cytometry

For flow cytometric analysis of lymphoid tissues, skin‐draining lymph node (LN) cells and splenocytes were isolated and stained in PBS containing 2% fetal calf serum. We used axillary and inguinal LNs as skin‐draining LNs. Flow cytometric analysis was performed by FACSAria III (BD Biosciences) using FlowJo software version 10.8.1 (Tree Star). The antibodies used were described in Table S1. Gating strategy of flow cytometric analysis was shown in Figure S2. Intracellular staining of Foxp3 was performed using the Foxp3 staining buffer set (Thermo Fisher Scientific) according to the manufacturer's instructions. All staining procedure was performed after blocking Fc receptor with anti‐CD16/CD32. Surface stainings were performed according to standard procedures at a density of 5–10×10⁵ cells per 50 μL, and volumes were scaled up accordingly.

Cytokine Assay

In all cell culture experiments, we used RPMI 1640 medium (Sigma) supplemented with 10% fetal calf serum, 50 μmol/L 2β‐mercaptoethanol, and antibiotics. For the culture of splenocytes, whole isolated cells were cultured in 24‐well plates in RPMI medium at a concentration of 4×10⁶ cells/mL and stimulated with 2 μg/mL concanavalin A (Sigma). Culture supernatants were collected at 72 hours and analyzed by ELISA for IL‐4, IL‐10, IL‐17, and IFN (interferon)‐γ using paired antibodies specific for corresponding cytokines according to the manufacturer's instructions (R&D Systems).

Treg Suppression Assay

For analysis of in vitro suppressive function of Tregs, CD4⁺CD25⁺ Tregs and CD4⁺CD25⁻ T cells were purified from the skin‐draining LNs and spleen of UV‐B‐irradiated or nonirradiated mice using a CD4⁺CD25⁺ Regulatory T Cell Isolation Kit (Miltenyi Biotec) and CD4 (L3T4) microbeads (Miltenyi Biotec) according to the manufacturer's instructions. The purity of each population was >95% by flow cytometric analysis. Purified CD4⁺CD25⁺ Tregs from UV‐B‐irradiated or nonirradiated mice were cocultured with responder CD4⁺CD25⁻ T cells (2.5×10⁴ cells) from other nonirradiated mice labeled with carboxyfluorescein diacetate succinimidyl ester (Thermo Fisher Scientific) at the indicated ratios in the presence of soluble anti‐CD3 antibody (0.5 μg/mL, clone 145‐2C11; BD Biosciences) and splenic antigen‐presenting cells treated with mitomycin C (WAKO) in 96‐well round‐bottomed plates. The cocultured cells were maintained at 37 °C with 5% CO₂ for 3 days. Proliferation of CD4⁺CD25⁻ T cells labeled with carboxyfluorescein diacetate succinimidyl ester was analyzed by flow cytometry.

Intracellular Cytokine Staining

Splenocytes were stimulated with 20 ng/mL phorbol 12‐myristate 13‐acetate (Sigma) and 1 mmol/L ionomycin (Sigma) for 5 hours in the presence of Brefeldin A (Thermo Fisher Scientific). After staining for surface antigens, intracellular cytokine staining was performed using an intracellular cytokine staining kit (BD Biosciences) and anticytokine antibodies according to the manufacturer's instructions.

Quantitative Reverse Transcription‐Polymerase Chain Reaction Analysis

Total RNA was extracted from the aorta after perfusion with RNA (Life Technologies) using TRIzol reagent (Life Technologies). After isolation of CD4⁺CD25⁺ Tregs and CD4⁺CD25⁻ T cells using a CD4⁺CD25⁺ Regulatory T Cell Isolation Kit and CD4 (L3T4) microbeads (Miltenyi Biotec), total RNA was extracted using a RNeasy mini kit (Qiagen). For reverse transcription (RT), a PrimeScript RT reagent Kit (Takara) was used. Quantitative polymerase chain reaction (PCR) was performed using a TB Green Premix Ex Taq (Takara) and a StepOnePlus Real‐Time PCR System (Thermo Fisher Scientific) according to the manufacturer's protocol. The primers used are described in Table S2. Amplification reactions were performed in duplicate and fluorescence curves were analyzed with included software. β‐actin was used as an endogenous control reference. The expression levels of the target genes were normalized so that the mean values in nonirradiated mice were set to 1.

Liquid Chromatography/Mass Spectrometry/Mass Spectrometry‐Based Lipidomics

Various lipid mediators in the skin were analyzed by liquid chromatography/mass spectrometry (MS)/MS‐based lipidomics as previously described. ²¹ Deuterated internal standards d₄‐leukotriene B₄, d₈‐5‐HETE, d₄‐prostaglandin E₂, and d₅‐resolvin D2, representing each chromatographic region of identified lipid mediators, were added to the samples (500 pg each) to facilitate quantification. The samples were extracted by solid‐phase extraction on C18 columns and were subjected to liquid chromatography–MS/MS. The system consisted of a Q‐Trap 6500 (Sciex) equipped with a Shimadzu LC‐30AD HPLC system. A ZORBAX Eclipse Plus C18 column (100 mm×4.6 mm, 3.5 μm; Agilent Technologies) was used with a methanol/water/acetic acid gradient of 55:45:0.01 to 98:2:0.01 (v/v/v) at a 0.4 mL/min flow rate. For monitoring and quantifying the levels of targeted lipid mediators, the multiple reaction monitoring method was developed with signature ion pairs Q1 (parent ion)/Q3 (characteristic fragment ion) for each molecule. Identification was conducted with published criteria using the liquid chromatography retention time, specific fragmentation patterns, and at least 6 diagnostic fragmentation ions. Quantification was carried out on the basis of the peak area of the multiple reaction monitoring chromatograph, and the linear calibration curves were obtained with authentic standards for each compound.

Statistical Analysis

Normality was assessed by Shapiro–Wilk normality test. Two‐tailed Student's t test or Mann–Whitney U‐test was used to detect significant differences between 2 groups where appropriate. One‐way ANOVA followed by Dunnett's post hoc test, 2‐way ANOVA followed by Tukey's multiple comparisons test, or Kruskal–Wallis test followed by Dunn's post hoc test was performed for multiple groups where appropriate. A value of P<0.05 was considered statistically significant. Data were expressed as mean±SD. No data were excluded from the analysis. Investigators were not blinded to the data analysis. For statistical analysis, GraphPad Prism version 7.0 (GraphPad Software Inc.) was used.

Results

Effect of Specific Wavelengths of UV‐B on the Development of Atherosclerosis and Plaque Inflammation

No adverse effects including skin cancer or burns were observed during the course of experiment. Interestingly, 282 nm UV‐B‐irradiated mice showed a marked decrease in atherosclerotic lesions in the aortic root compared with nonirradiated mice (Figure 1A). Although irradiation of other wavelengths of UV‐B did not affect atherosclerosis significantly, there was a tendency toward reduction in atherosclerotic plaques in 312 nm UV‐B‐irradiated mice (Figure 1A). Means±SD of mean atherosclerotic lesion area in the aortic sinus was 2.38±0.79×10⁵ μm² in control nonirradiated mice, 1.83±0.54×10⁵ μm² in 282 nm UV‐B‐irradiated mice (P=0.030, versus controls), 2.05±0.75×10⁵ μm² in 301 nm UV‐B‐irradiated mice (P=0.260, versus controls), and 1.95±0.58×10⁵ μm² in 312 nm UV‐B‐irradiated mice (P=0.104, versus controls), as shown in Figure 1A. We also performed en face analysis of thoracoabdominal aortas, showing no difference in the aortic plaque burden between UV‐B‐irradiated and nonirradiated mice (Figure 1B). Means±SD of mean atherosclerotic lesion area in the aorta was 5.57±1.95% in control nonirradiated mice, 5.17±1.56% in 282 nm UV‐B‐irradiated mice (P>0.999, versus controls), 5.95±2.32% in 301 nm UV‐B‐irradiated mice (P>0.999, versus controls), and 6.02±3.05% in 312 nm UV‐B‐irradiated mice (P>0.999, versus controls), as shown in Figure 1B. There were no significant differences in body weight and plasma lipid profile between UV‐B‐irradiated and nonirradiated mice (Table). Because UV‐B irradiation is known to play a critical role in the synthesis of immunosuppressive vitamin D, we analyzed plasma levels of biologically active 1,25‐dihydroxyvitamin D and circulating storage form 25‐hydroxyvitamin D. Plasma levels of 25‐hydroxyvitamin D were significantly increased in 282 (P=0.041) or 312 (P=0.048) nm UV‐B‐irradiated mice, whereas plasma 1,25‐dihydroxyvitamin D levels were unaltered in all wavelengths of UV‐B‐irradiated mice, suggesting that the effect of UV‐B irradiation on plasma vitamin D levels may vary depending on the wavelengths (Table).

Figure 1. The effect of specific wavelengths of ultraviolet B (UV‐B) on the development of atherosclerosis and plaque inflammation.

Six‐week‐old male Apoe ^−/− (apolipoprotein E‐deficient) mice were irradiated with 3 specific wavelengths of UV‐B (282, 301, and 312 nm) at 2 kJ/m² once weekly for 14 weeks and euthanized at 20 weeks of age. Atherosclerotic lesions were then assessed. Nonirradiated male Apoe ^−/− mice served as controls. A, Representative photomicrographs of Oil Red O staining and quantitative analysis of atherosclerotic lesion area in the aortic sinus of UV‐B‐irradiated (n=20 for 282 nm UV‐B, n=23 for 301 nm UV‐B, n=22 for 312 nm UV‐B) or nonirradiated mice (n=20). B, Representative photomicrographs of Oil Red O staining and quantitative analysis of atherosclerotic lesion area in the aorta of UV‐B‐irradiated (n=21 for 282 nm UV‐B, n=17 for 301 nm UV‐B, n=20 for 312 nm UV‐B) or nonirradiated mice (n=22). C through E, Representative sections and quantitative analyses of MOMA‐2⁺ macrophages (C), CD4⁺ T cells (D), and collagen (E) in the aortic sinus. Arrowheads indicate the CD4⁺ T cells; n=10 per group. Samples of the aortic root and thoracoabdominal aorta were collected from 6 experiments (n=3–4 per group in each experiment). Black bars represent 50, 200, or 500 μm as described. Data points represent individual animals. Horizontal bars represent means. Error bars indicate SD. *P<0.05; **P<0.01; 1‐way ANOVA followed by Dunnett's post hoc test (A and E) or Kruskal–Wallis test followed by Dunn's post hoc test (C and D).

Table .

Body Weight and Plasma Lipid Profile and Vitamin D Levels in 20‐Week‐Old Ultraviolet B‐Irradiated or Nonirradiated Male Apoe ^−/− Mice

	Nonirradiated	UV‐B 282 nm	UV‐B 301 nm	UV‐B 312 nm
Body weight, g	32.16±1.60 (n=16)	31.63±2.00 (n=15)	31.53±1.80 (n=15)	31.87±1.20 (n=16)
Total cholesterol, mg/dL	524.9±96.51 (n=10)	563.4±134.0 (n=10)	576.6±108.4 (n=10)	547.3±97.66 (n=10)
High‐density lipoprotein‐cholesterol, mg/dL	18.30±4.57 (n=10)	19.50±5.34 (n=10)	19.20±4.29 (n=10)	18.90±4.70 (n=10)
Triglycerides, mg/dL	91.00±28.95 (n=10)	75.50±25.39 (n=10)	78.70±34.78 (n=10)	81.70±18.66 (n=10)
25‐OH vitamin D, ng/mL	54.00±14.44 (n=5)	76.60±7.23* (n=5)	65.00±11.85 (n=5)	74.00±13.62* (n=5)
1,25‐(OH)2 vitamin D, pg/mL	251.0±69.56 (n=5)	259.4±62.59 (n=5)	300.6±50.78 (n=5)	272.8±62.10 (n=5)

All data are expressed as the mean±SD. Apoe ^−/− indicates apolipoprotein E‐deficient; and UV‐B. ultraviolet B.

^* P<0.05; 1‐way ANOVA followed by Dunnett's post hoc test.

To unveil the effect of each wavelength of UV‐B irradiation on atherosclerotic plaque component, we performed immunohistochemical studies of atherosclerotic lesions in the aortic sinus. Notably, the atherosclerotic lesions of 282 nm UV‐B‐irradiated mice exhibited significantly decreased accumulation of macrophages by 20% (P=0.002) and CD4⁺ T cells by 32% (P=0.020) compared with those of nonirradiated mice, whereas there was no difference in the accumulation of these inflammatory cells in the lesions of 301 or 312 nm UV‐B‐irradiated mice (Figure 1C and 1D). The proportion of collagen content in the aortic sinus plaques was significantly increased only in 282 nm UV‐B‐irradiated mice compared with nonirradiated mice (P=0.001; Figure 1E). To reveal the effect of each UV‐B irradiation on aortic immunoinflammatory responses, we examined mRNA expression of proinflammatory cytokines and transcription factors specific for Tregs or helper T‐cell subsets in atherosclerotic aortas by quantitative RT‐PCR. Although a trend toward decrease in aortic IFN‐γ mRNA expression was observed in 282 nm UV‐B‐irradiated mice (P=0.066, versus controls), there was no significant difference in the expression of other genes in the aortas between each wavelength of UV‐B‐irradiated and nonirradiated mice (Figure S3).

Effect of Specific Wavelengths of UV‐B on T‐Cell Populations in Peripheral Lymphoid Tissues

Next, we investigated the effect of each wavelength of UV‐B on immune responses focusing on CD4⁺ T cell populations in peripheral lymphoid tissues. We irradiated Apoe ^−/− mice with each wavelength of UV‐B once weekly for 6 weeks and analyzed the populations of CD4⁺Foxp3⁺ Tregs and CD4⁺CD44^highCD62L^low effector memory T cells in the skin‐draining LNs and spleen by flow cytometry. Strikingly, 282 nm UV‐B irradiation significantly increased the frequency (P<0.001) and number (P=0.008) of CD4⁺Foxp3⁺ Tregs in the skin‐draining LNs of Apoe ^−/− mice (Figure 2A). In addition, the frequency of CD4⁺Foxp3⁺ Tregs was also increased in the spleen of 282 nm UV‐B‐irradiated mice (P=0.019; Figure 2A). On the other hand, 301 or 312 nm UV‐B had no major effect on the frequency and number of CD4⁺Foxp3⁺ Tregs in peripheral lymphoid tissues (Figure 2A). There was no change in the frequency of CD4⁺CD44^highCD62L^low effector memory T cells in the peripheral lymphoid tissues of 282 or 301 nm UV‐B‐irradiated mice, although their number was modestly increased in the skin‐draining LNs of 282 nm UV‐B‐irradiated mice (P=0.047; Figure 2B). Interestingly, the frequency (P=0.036) and number (P=0.001) of CD4⁺CD44^highCD62L^low effector memory T cells were significantly decreased in the spleen of 312 nm UV‐B‐irradiated mice (Figure 2B).

Figure 2. The effect of specific wavelengths of ultraviolet B (UV‐B) on T‐cell populations in peripheral lymphoid tissues.

Male Apoe ^−/− (apolipoprotein E‐deficient) mice were irradiated with 3 specific wavelengths of UV‐B (282, 301, and 312 nm) at 2 kJ/m² once weekly for 6 weeks. Nonirradiated male Apoe ^−/− mice served as controls. Four days after the last UV‐B irradiation, lymphoid cells from skin‐draining lymph nodes (LNs) and spleen were prepared. A and B, Representative results of CD4⁺ Foxp3 (forkhead box P3)⁺ regulatory T cells (Tregs) (A) and CD4⁺CD44^highCD62L^low effector memory T cells (B) in the skin‐draining LNs assessed by flow cytometry. The graphs represent proportions and total numbers of CD4⁺Foxp3⁺ Tregs (A) and CD4⁺CD44^highCD62L^low effector memory T cells (B) in the skin‐draining LNs and spleen; n=12 per group from 3 experiments (n=4 per group in each experiment). C and D, Expression levels of cytotoxic T lymphocyte‐associated antigen‐4 (CTLA‐4) and CD103 were analyzed gating on CD4⁺Foxp3⁺ Tregs in the skin‐draining LNs and spleen (C) and CD4⁺Foxp3⁻ non‐Tregs in the skin‐draining LNs and spleen (D). Histograms show mean fluorescence intensity (MFI); n=8 per group from 2 experiments (n=4 per group in each experiment). E, Purified CD4⁺CD25⁺ Tregs from the skin‐draining LNs and spleen of UV‐B‐irradiated (n=2) or nonirradiated mice (n=2) were cocultured with carboxyfluorescein diacetate succinimidyl ester (CFSE)‐labeled responder CD4⁺CD25⁻ T cells from other nonirradiated mice (n=2) at the indicated ratios, and their suppressive function was assessed by evaluating proliferation of CFSE‐labeled T cells (mean±SD of triplicate wells). Data points represent individual animals (A through D). Horizontal bars represent means. Error bars indicate SD. *P<0.05; **P<0.01; 1‐way ANOVA followed by Dunnett's post hoc test: (A top left and bottom right, B bottom left and right, C, D, and E) or Kruskal–Wallis test followed by Dunn's post hoc test (A top right and bottom left and B top right).

To determine the effect of each wavelength of UV‐B on the activation and function of CD4⁺ T cells, we analyzed the expression of activation‐ or function‐associated molecules in CD4⁺Foxp3⁺ Tregs in peripheral lymphoid tissues by flow cytometry. The expression of CD103 was markedly upregulated in CD4⁺Foxp3⁺ Tregs in the skin‐draining LNs (P=0.013) and spleen (P<0.001) of 282 nm UV‐B‐irradiated mice (Figure 2C). In these mice, the expression levels of cytotoxic T lymphocyte‐associated antigen‐4 (CTLA‐4) were also increased in CD4⁺Foxp3⁺ Tregs in the skin‐draining LNs (P=0.003) and tended to be increased in splenic CD4⁺Foxp3⁺ Tregs (P=0.092, versus controls; Figure 2C). There were no major changes in the expression of these molecules in CD4⁺Foxp3⁺ Tregs in the skin‐draining LNs and spleen of 301 or 312 nm UV‐B‐irradiated mice (Figure 2C). We also examined the mRNA expression of function‐associated molecules (Foxp3, Il10, and Tgfb) in magnetic‐activated cell sorting‐separated CD4⁺CD25⁺ Tregs by quantitative RT‐PCR. However, there was no major difference in the mRNA expression of these molecules between each wavelength of UV‐B‐irradiated and nonirradiated mice (Figure S4A).

We further analyzed the expression of activation‐ or function‐associated molecules in CD4⁺Foxp3⁻ non‐Tregs in the peripheral lymphoid tissues by flow cytometry. The expression of CD103 was downregulated in the skin‐draining LNs (P=0.002) and spleen (P=0.031) of 312 nm UV‐B‐irradiated mice, whereas the expression of these molecules was not altered in 282 or 301 nm UV‐B‐irradiated mice (Figure 2D). The expression of CTLA‐4 in CD4⁺Foxp3⁻ non‐Tregs was not changed by each wavelength of UV‐B irradiation (Figure 2D). Quantitative RT‐PCR analysis revealed no difference in the mRNA expression of T helper type 1 (Th1)‐related Tbx21 or Th2‐related Gata3 in magnetic‐activated cell sorting‐separated CD4⁺CD25⁻ non‐Tregs between each wavelength of UV‐B‐irradiated and nonirradiated mice (Figure S4B). The mRNA expression of Th17‐related Rorc was undetected.

We performed an in vitro suppression assay to define the suppressive function of Tregs, and interestingly, found that 282 nm but not 301 or 312 nm UV‐B irradiation augmented the suppressive function (Figure 2E).

Collectively, these data indicate that 282 nm UV‐B irradiation resulted in a marked increase in CD4⁺Foxp3⁺ Tregs with high suppressive capacity, which contributed to the reduction of atherosclerosis. Although 312 nm UV‐B irradiation had no significant effect on frequency and activation of CD4⁺Foxp3⁺ Tregs and atherosclerosis, this wavelength of UV‐B reduced the number and activation of CD4⁺Foxp3⁻ non‐Tregs in peripheral lymphoid tissues, implying the potential to mitigate inflammatory immune responses.

Effect of Specific Wavelengths of UV‐B on T‐Cell Immune Responses and Other Immune Cell Populations in Peripheral Lymphoid Tissues

To determine whether each wavelength of UV‐B irradiation affects T‐cell immune responses, we irradiated Apoe ^−/− mice with each wavelength of UV‐B once weekly for 6 weeks and analyzed cytokine production from splenic T cells by ELISA and intracellular cytokine staining. There were no significant differences in the production of proatherogenic Th1‐related cytokine IFN‐γ, anti‐inflammatory cytokine IL‐10, and Th17‐related cytokine IL‐17 from splenic T cells stimulated with concanavalin A in vitro between each wavelength of UV‐B‐irradiated and nonirradiated mice (Figure 3A). The production of Th2‐related cytokine IL‐4 from concanavalin A‐stimulated splenic T cells was undetectable in all mice. The proportions of IFN‐γ‐producing Th1 cells, IL‐4‐producing Th2 cells, IL‐10‐producing CD4⁺ T cells, and IL‐17‐producing Th17 cells in spleen were not different between each wavelength of UV‐B‐irradiated and nonirradiated mice (Figure 3B).

Figure 3. The effect of specific wavelengths of ultraviolet B (UV‐B) on T‐cell immune responses and other immune cell populations in peripheral lymphoid tissues.

(DCs), and the expression of CD80 and CD86 on CD11c⁺MHC‐II⁺ DCs were determined by flow cytometry; n=5 to 6 per group. Data points represent individual animals. Horizontal bars represent means. Error bars indicate SD. *P<0.05; **P<0.01; 1‐way ANOVA followed by Dunnett's post hoc test. MFI indicates mean fluorescence intensity.

We also analyzed various immune cells in spleen by flow cytometry. All wavelengths of UV‐B irradiation had minor effect on the proportions of CD8⁺ T cells, B220⁺ B cells, Ly6G⁺ neutrophils, CD11b⁺Ly6C^high monocyte, natural killer cells, natural killer T cells, and CD11c⁺ major histocompatibility complex‐II⁺ dendritic cells, and the expression of CD80 and CD86 on dendritic cells (Figure 3C).

Effect of Specific Wavelengths of UV‐B on the Expression of Various Chemokine Receptors on T Cells in Peripheral Lymphoid Tissues

Teffs or Tregs may migrate to the aorta and promote or attenuate lesional inflammation, respectively. Chemokine system is known to be crucial for the recruitment of various immune cells including T cells to atherosclerotic lesions. ²² To elucidate the effect of each wavelength of UV‐B irradiation on migratory capacity of CD4⁺ T cells, we analyzed the expression of several major chemokine receptors. We isolated CD4⁺CD25⁺ Tregs and CD4⁺CD25⁻ non‐Tregs from the skin‐draining LNs of Apoe ^−/− mice irradiated with each wavelength of UV‐B once weekly for 6 weeks or nonirradiated mice and analyzed the mRNA expression of chemokine receptors by quantitative RT‐PCR. Interestingly, upregulated mRNA expression of CCR4 (C‐C chemokine receptor 4), which was reported to play a critical role in the recruitment of T cells to the inflamed tissues of inflammatory bowel disease and allergic lung disease, ²³ was observed in Tregs from 282 nm (P=0.034) but not 301 or 312 nm UV‐B‐irradiated mice, whereas the expression of other chemokine receptors (Ccr5, Ccr6, Ccr7, and Ccr8) was not altered in any wavelengths of UV‐B (Figure 4A). On the other hand, there were no significant differences in the mRNA expression of major chemokine receptors (Ccr4, Ccr5, Ccr6, Ccr7, Ccr8, and Cxcr3) in non‐Tregs between each wavelength of UV‐B‐irradiated and nonirradiated mice (Figure 4B). In line with the upregulated CCR4 expression at the mRNA levels, flow cytometric analysis confirmed that CD4⁺Foxp3⁺ Tregs in the skin‐draining LNs and spleen of 282 nm UV‐B‐irradiated mice expressed higher levels of CCR4 than those from nonirradiated mice (skin‐draining LNs: P=0.002, spleen: P=0.013; Figure 4C). We further characterized the CCR4^high Tregs in detail by analyzing the expression of activation‐ or function‐associated molecules CTLA‐4 and CD103 in CD4⁺Foxp3⁺ Tregs in the skin‐draining LNs and spleen. The expression of CTLA‐4 and CD103 was markedly higher in CCR4^highCD4⁺Foxp3⁺ Tregs than that in CCR4^lowCD4⁺Foxp3⁺ Tregs in nonirradiated mice (skin‐draining LNs CTLA‐4: P<0.001, skin‐draining LNs CD103: P<0.001, spleen CTLA‐4: P<0.001, spleen CD103: P=0.001; Figure 4D and 4E), indicating that CCR4 expression could be used as a marker to define activated Tregs with high suppressive function and migratory capacity. Notably, the expression levels of these activation‐ or function‐associated molecules in CCR4^highCD4⁺Foxp3⁺ Tregs were dramatically increased by 282 nm UV‐B irradiation (Figure 4D and 4E).

Figure 4. The effect of specific wavelengths of ultraviolet B (UV‐B) on the expression of various chemokine receptors on T cells in peripheral lymphoid tissues.

Male Apoe ^−/− (apolipoprotein E‐deficient) mice were irradiated with 3 specific wavelengths of UV‐B (282, 301, and 312 nm) at 2 kJ/m² once weekly for 6 weeks. Nonirradiated male Apoe ^−/− mice served as controls. Four days after the last UV‐B irradiation, lymphoid cells from the skin‐draining lymph nodes (LNs) and spleen were prepared. A and B, Expression of mRNAs for chemokine receptors (Ccr4, Ccr5, Ccr6, Ccr7, and Ccr8) in CD4⁺CD25⁺ regulatory T cells (Tregs) (A) and for chemokine receptors (Ccr4, Ccr5, Ccr6, Ccr7, Ccr8, and Cxcr3) in CD4⁺CD25⁻ non‐Tregs (B) in the skin‐draining LNs was quantified by quantitative real‐time reverse transcription polymerase chain reaction and normalized to β‐actin. Fold change relative to control group is shown. CD4⁺CD25⁺ Tregs or CD4⁺CD25⁻ non‐Tregs were purified from pooled skin‐draining LNs of 3 to 4 mice in each group; n=5 samples per group. Data points represent individual pooled samples. C, Representative results of CCR4 (C‐C chemokine receptor 4) expression on CD4⁺ Foxp3 (forkhead box P3)⁺ Tregs in the skin‐draining LNs analyzed by flow cytometry. Expression levels of CCR4 were analyzed gating on CD4⁺Foxp3⁺ Tregs in the skin‐draining LNs and spleen. D and E, Expression levels of cytotoxic T lymphocyte‐associated antigen‐4 (CTLA‐4) and CD103 were analyzed gating on CCR4^low or CCR4^high CD4⁺Foxp3⁺ Tregs in the skin‐draining LNs (D) and spleen (E); n=5 per group. Data points represent individual animals (C through E). Horizontal bars represent means. Error bars indicate SD. *P<0.05; **P<0.01; 1‐way ANOVA followed by Dunnett's post hoc test (A), 2‐tailed Student's t test (C), and 2‐way ANOVA followed by Tukey's multiple comparisons test (D and E). MFI indicates mean fluorescence intensity.

These results indicate that 282 nm UV‐B irradiation not only expands activated Tregs but also may enhance their migratory function without substantially affecting Teff responses, which may contribute to reduced inflammation and lesion development possibly by promoting Treg migration to the atherosclerotic lesions.

Effect of 282 nm UV‐B on the Production of Lipid Mediators in the Skin

An imbalance between proresolving and proinflammatory lipid mediators is linked to the pathogenesis of chronic inflammatory diseases including atherosclerotic disease. ²⁴ Although arachidonic acid‐derived lipid mediator prostaglandin E₂ is known to be produced in the skin by UV‐B irradiation, ¹⁷ its effect on other lipid mediators remains largely unknown. To determine the effect of 282 nm UV‐B irradiation on the balance of proresolving and proinflammatory lipid mediators, we irradiated Apoe ^−/− mice with 282 nm UV‐B once weekly for 6 weeks and examined the production of various lipid mediators in the skin by liquid chromatography/MS/MS‐based lipidomics. The amount of arachidonic acid in the skin was not affected by 282 nm UV‐B irradiation (Figure 5A). Production of proinflammatory lipid mediators (prostaglandin D₂, prostaglandin E₂, and prostaglandin F_2a) derived from arachidonic acid was markedly decreased (prostaglandin D₂: P=0.034, prostaglandin E₂: P=0.045, prostaglandin F_2a: P=0.008), whereas leukotriene B₄ derived from arachidonic acid was significantly increased in the skin of 282 nm UV‐B‐irradiated mice (P=0.003) compared with nonirradiated mice (Figure 5A). On the other hand, production of a proresolving lipid mediator resolvin E4 derived from eicosapentaenoic acid was significantly increased in the skin of 282 nm UV‐B‐irradiated mice compared with nonirradiated mice (P=0.014), while the amount of eicosapentaenoic acid in the skin was not altered by 282 nm UV‐B irradiation (Figure 5B). In addition, similar results were obtained for the amount of docosahexaenoic acid and the production of its derived proresolving lipid mediators (resolvin D1, resolvin D2, resolvin D4, resolvin D5, protectin D1, maresin 1, and maresin 2) (Figure 5C). To reveal whether these favorable changes occur systemically, we also examined lipid mediators in the peripheral blood. Although some of lipid mediators derived from arachidonic acid, eicosapentaenoic acid, and docosahexaenoic acid were detected in the plasma of both 282 nm UV‐B‐irradiated and nonirradiated mice, important lipid mediators with anti‐inflammatory properties described previously were below the detectable levels (data not shown). Nevertheless, we speculate that the increased ratio of proresolving to proinflammatory lipid mediators in the skin might be involved in the reduction of atherosclerosis.

Figure 5. The effect of 282 nm ultraviolet B (UV‐B) on the production of lipid mediators in the skin.

Male Apoe ^−/− (apolipoprotein E‐deficient) mice were irradiated with 282 nm UV‐B at 2 kJ/m² once weekly for 6 weeks. Nonirradiated male Apoe ^−/− mice served as controls. Four days after the last UV‐B irradiation, the back skin of the mice was isolated. The production of various lipid mediators in the skin was analyzed by liquid chromatography/mass spectrometry/mass spectrometry‐based lipidomics. Quantification of arachidonic acid (AA) and its derived lipid mediators (A), eicosapentaenoic acid (EPA) and its derived lipid mediators (B), and docosahexaenoic acid (DHA) and its derived lipid mediators (C) in the skin; n=5 per group. Data points represent individual animals. Horizontal bars represent means. Error bars indicate SD. *P<0.05; **P<0.01; Mann–Whitney U‐test: PGF_2a (A) and LTB₄, 5‐HEPE and 18‐HEPE (B), and resolvin D1 and protectin D1 (C); 2‐tailed Student's t test: PGE₂, prostaglandin D2, 15deoxy‐d12, 14 PGJ₂, 12‐HETE, 15‐HETE, 5, 15‐diHETE, and 5‐HETE (A), resolvin E4, 12‐HEPE, and 15‐HEPE (B), and maresin 2, 14‐HDHA, 17‐HDHA, resolvin D5, maresin 1, 4‐HDHA, and 7‐HDHA (C). LTB₄ indicates leukotriene B₄; ND, not detected; PGD₂, prostaglandin D₂; PGE₂, prostaglandin E₂; PGF_2a, prostaglandin F_2a; and TXB₂, thromboxane B₂.

Discussion

Although solid evidence indicates that immunoinflammatory responses in the arterial wall are critically responsible for the initiation and progression of atherosclerotic disease, ⁴ effective clinical therapies to directly intervene in inflammatory reactions are unavailable so far. In the present study, we developed novel UV‐B devices using LED technology and for the first time demonstrated that 282 nm UV‐B irradiation effectively reduces atherosclerosis and plaque inflammation in hypercholesterolemic Apoe ^−/− mice, associated with specific augmentation of anti‐inflammatory Treg responses, minor effects on other immune responses, increased plasma vitamin D levels, and favorably modulated lipid mediator balance in the skin. On the other hand, we found that 301 nm UV‐B irradiation neither reduces atherosclerosis nor induces these anti‐inflammatory changes. We also showed that irradiation of clinically relevant 312 nm UV‐B modestly reduces the proportion and number of effector memory T cells and tends to limit atherosclerotic lesion formation. Thus, we found that each specific wavelength of UV‐B has a unique role in the regulation of T‐cell immune responses and atherosclerosis. Considering that we cannot receive shorter 282 nm UV‐B from natural sunlight, this UV‐B treatment could be a unique therapeutic approach to atherosclerosis.

Accumulating experimental and clinical evidence indicates that T‐cell–mediated immune responses substantially contribute to the development and prevention of atherosclerosis depending on the cell types. ⁷ Teffs including IFN‐γ‐producing Th1 cells exacerbate atherosclerosis, whereas CD4⁺Foxp3⁺ Tregs play an antiatherogenic role through suppressing Teff‐mediated inflammation. ²⁵ Here, we found that the proportion and activation of Tregs in the skin‐draining LNs and spleen were specifically increased and their suppressive capacity was also enhanced following 282 nm UV‐B irradiation at 2 kJ/m² in hypercholesterolemic mice, and a minor effect on Teff immune responses was observed. In our previous study using broadband UV‐B (280–320 nm), UV‐B irradiation at the same 2 kJ/m² induced Treg expansion and activation only in the skin‐draining LNs, whereas higher 5 kJ/m² UV‐B irradiation systemically expanded Tregs with potent suppressive function, ¹⁹ indicating that our 282 nm UV‐B irradiation appears to be more effective in augmenting systemic Treg responses including its proportion and function. Our present data suggest that 282 nm UV‐B irradiation at 2 kJ/m² limited atherosclerotic lesion development and accumulation of macrophages and CD4⁺ T cells in the plaques, whereas the same dose of broadband UV‐B irradiation did not affect CD4⁺ T‐cell accumulation in the plaques despite reduced atherosclerotic lesion formation and macrophage accumulation. Thus, our data suggest that treatment with 282 nm UV‐B could specifically augment Treg responses and efficiently regulate aortic inflammatory immune responses and atherosclerosis, though it is possible that other wavelengths of UV‐B might also be effective. In consideration of the decreased number and downregulated expression of activation markers in non‐Tregs from 312 nm UV‐B‐irradiated mice, we suppose that the anti‐inflammatory action of broadband UV‐B may include various mechanisms derived from multiple wavelengths of UV‐B, which could cooperatively contribute to the prevention of atherosclerosis.

Chemokine system has a crucial role in T‐cell migration to the atherosclerotic lesions. ²² Notably, CCR4 was shown to mediate Treg recruitment to the inflammatory sites and play a protective role in the development of inflammatory or allergic disease. ²⁶ , ²⁷ Intriguingly, we found that CCR4 was the only chemokine receptor whose mRNA expression was upregulated on Tregs following 282 nm UV‐B irradiation. Given that skin resident Tregs are known to highly express CCR4, ²⁸ our finding of the upregulation of CCR4 expression on Tregs in peripheral lymphoid tissues may reflect their promoted migration to these sites following 282 nm UV‐B irradiation. Notably, CCR4^high Tregs highly expressed CTLA‐4 and CD103 showing activated phenotype, which was more pronounced following 282 nm UV‐B irradiation. Given the critical role of CTLA‐4 expressed by Tregs in maintaining their suppressive function under physiological ²⁹ or atherosclerotic conditions, ³⁰ our finding of the upregulated CTLA‐4 expression in CD4⁺Foxp3⁺ Tregs including CCR4^highCD4⁺Foxp3⁺ Tregs would imply their enhanced suppressive capacity, which was confirmed by an in vitro suppression assay. These findings suggest that 282 nm UV‐B irradiation could expand Tregs with potent suppressive function and high migratory capacity. Although the frequency of CD4⁺Foxp3⁺ Tregs in para‐aortic LNs and the mRNA expression of their specific marker Foxp3 in atherosclerotic aorta were not altered in 282 nm UV‐B‐irradiated mice (Figure S3 and data not shown), we speculate that 282 nm UV‐B‐expanded Tregs would migrate to the atherosclerotic lesions and efficiently suppress aortic inflammation and lesion development.

Phototherapy and photochemotherapy with UV are clinically established treatments for inflammatory skin diseases including psoriasis and atopic dermatitis, ¹⁸ although excessive UV‐B exposure could potentially increase the risk of skin cancer. Within the UV‐B spectrum, narrow‐band UV‐B (a narrow peak around 311 nm) is used most often for treating psoriasis, ¹⁸ indicating that the efficacy of UV‐B therapy may vary depending on its wavelengths. Therefore, identification of specific wavelength of UV‐B that can effectively prevent atherosclerosis could lead to the development of a novel approach for the treatment of CAD with less adverse effects. In this study, using LED technology that allowed us to irradiate specific wavelengths of UV‐B, we demonstrated that among 3 wavelengths of UV‐B, only 282 nm UV‐B irradiation reduces the development of atherosclerosis and infiltration of inflammatory cells in the lesions, providing first evidence that the preventive effect of UV‐B on atherosclerosis differs depending on the wavelength. This atheroprotection seems to be derived from systemic induction of atheroprotective Tregs with potent suppressive and migratory functions and increased ratio of proresolving to proinflammatory lipid mediators in the skin but not from systemic immunosuppression. Thus, we identified a specific wavelength of UV‐B that effectively limits atherosclerosis and the underlying mechanisms and suppose that this treatment could avoid unwanted general immunosuppression derived from dysregulated innate or adaptive immune responses. As we do not receive 282 nm UV‐B from natural sunlight due to protection by the atmosphere and the ozone layer, biological effects of this UV‐B remained unclear. This is the first report to identify the biological effects of 282 nm UV‐B focusing on T‐cell immune responses. To translate our findings to clinical settings, further investigation to examine more detailed biological effects and safety of this UV‐B will be required.

As described here, narrow‐band UV‐B irradiation is an effective treatment for human psoriasis. However, the mechanisms for this preventive effect have not been completely elucidated. One possible mechanism may be systemic expansion of Tregs, which is supported by previous reports showing that in patients with skin autoimmune disease, bath‐psoralen UV‐A or narrow‐band UV‐B irradiation increased peripheral CD4⁺CD25⁺Foxp3⁺ Tregs in association with improved disease state. ³¹ , ³² An interesting finding of our study is that UV‐B irradiation with similar 312 nm wavelength tended to inhibit atherosclerotic plaque formation in the aortic root, although Treg proportion in the peripheral lymphoid tissues was not altered by this UV‐B irradiation. Because the UV‐B dose used in this study is relatively lower than that used in clinical situations, ³³ it will be important to investigate the effect of higher dose of 312 nm UV‐B irradiation on Treg proportion and the development of atherosclerosis.

Vitamin D is known to have an immunosuppressive and anti‐inflammatory action. ³⁴ An epidemiological study showed the association between vitamin D deficiency and increased cardiovascular events and mortality, ³⁵ although no prior studies have provided evidence that supplementation of vitamin D reduces the risk of atherosclerotic cardiovascular events in humans. In atherosclerotic Apoe ^−/− mice, we previously found that oral supplementation of active form of vitamin D₃ (calcitriol) prevented atherosclerotic lesion formation, associated with the induction of immunosuppressive tolerogenic DCs and CD4⁺Foxp3⁺ Tregs in the peripheral lymphoid tissues and atherosclerotic lesions, ³⁶ providing direct evidence for the anti‐atherogenic role of vitamin D at least in mice. These reports collectively indicate a possible role of vitamin D in Treg expansion and atheroprotection. Accordingly, we speculate that increased plasma vitamin D levels may partly be involved in 282 nm UV‐B‐dependent specific induction of CD4⁺Foxp3⁺ Tregs and prevention of atherosclerosis development. On the other hand, the proportion of CD4⁺Foxp3⁺ Tregs and atherosclerotic lesion size were not altered in 312 nm UV‐B‐irradiated mice despite a significant increase in plasma vitamin D levels, implying a minor antiatherogenic action of vitamin D alone.

Regarding possible mechanisms for 282 nm UV‐B‐dependent modulation of Treg number and function, other mechanisms would also be involved. External environment stimulates the immune system in the skin, where Langerhans cells, a major subset of epidermal antigen presenting cells, ³⁷ are known to play an important role in regulating excessive immune responses and maintaining immune tolerance. UV‐B exposure causes skin inflammatory reactions and DNA damage and was reported to upregulate the expression of receptor activator of NF‐κB (nuclear factor kappa B) ligand in epidermal keratinocytes, which is critical for the maintenance of Tregs in the periphery by modulating Langerhans cell function. ³⁸ In our previous study using Langerhans cell‐depleted Apoe ^−/− mice, we demonstrated that epidermal Langerhans cells play an indispensable role in broadband UV‐B‐dependent expansion of Tregs and prevention of atherosclerotic plaque development. ¹⁹ Although we have not examined the role of Langerhans cells in 282 nm UV‐B‐dependent modulation of Treg responses, we speculate that a similar mechanism may operate in the present study. It will be of interest to elucidate the detailed mechanisms for 282 nm UV‐B‐dependent Treg changes.

Proresolving lipid mediators have been shown to dampen inflammation and contribute to the stabilization of atherosclerotic plaques. ²⁴ Although UV‐B irradiation was reported to affect production of prostaglandin E₂ in the skin, its effect on the balance between proresolving and proinflammatory lipid mediators remains largely unknown. Therefore, in addition to the effect on anti‐inflammatory Treg responses, we searched for further mechanisms for 282 nm UV‐B‐dependent atheroprotection by focusing on the production of lipid mediators. Strikingly, we found that 282 nm UV‐B irradiation remarkably increased the ratio of proresolving to proinflammatory lipid mediators in the skin. Notably, among various proresolving lipid mediators that were significantly increased following 282 nm UV‐B irradiation, docosahexaenoic acid‐derived resolvin D1, resolvin D2, and maresin 1 were reported to increase Treg differentiation and expression of its suppressive function‐associated molecules ³⁹ and play an antiatherogenic role. ²⁴ , ⁴⁰ Although in our study these lipid mediators were undetectable in the plasma of 282 nm UV‐B‐irradiated and nonirradiated mice, it is possible that similar favorable changes in lipid mediators might also occur systemically and in atherosclerotic lesions following 282 nm UV‐B irradiation, which could provide another possible mechanism for 282 nm UV‐B‐dependent augmentation of Treg responses and regulation of intraplaque infiltration of inflammatory cells and plaque formation.

The present study has some limitations. We found that 282 nm UV‐B irradiation significantly reduced the development of aortic root atherosclerotic plaques and accumulation of inflammatory cells, whereas atherosclerotic lesion size and mRNA expression of inflammatory markers in thoracoabdominal aortas were not affected by this treatment. These inconsistent results may be derived from the limited efficacy of this therapy or differences in methodologies to evaluate atherosclerotic plaque size and portion of the aorta analyzed. Although our 282 nm UV‐B therapy appears to be safe and effective in reducing the development of atherosclerosis in hypercholesterolemic mice, this phototherapy may have a minor atheroprotective effect or cause adverse responses in humans. In the context of application in clinical settings, extensive clinical investigation will be required.

Conclusions

In summary, we demonstrated that 282 nm UV‐B irradiation effectively prevents aortic inflammation and the development of atherosclerosis by augmenting anti‐inflammatory immune responses. We identified unrecognized roles of specific wavelengths of UV‐B in controlling T‐cell immune responses and atherosclerosis. Our data indicate that a novel 282 nm UV‐B phototherapy, which cannot be replaced by natural sunlight exposure, could be an attractive immunomodulatory approach to treat CAD, though careful evaluation is required to confirm the safety of this treatment.

Sources of Funding

This work was supported by Japan Society for the Promotion of Science (JSPS) KAKENHI Grant Numbers 18K08088 (N.S.) and 21K08042 (N.S.) and research grants from Pfizer Japan Inc. (Y.R.), Astellas Pharma Inc. (Y.R.), Novartis Pharma K.K. (N.S.), Takeda Science Foundation (N.S.), Suzuken Memorial Foundation (N.S.), and Senshin Medical Research Foundation (N.S.).

Disclosures

None.

Supporting information

Tables S1–S2

Figures S1–S4