*3.5. Modeling of D3 Production with Ageing*

As shown in Figure 3, a linear regression model using data from both cohorts across the age spectrum (21 to 69 years) was created with age as the independent variable and log D3 production (change from baseline to peak) as the dependent variable (*p* = 0.023). Age accounted for 20 percent of the variance in D3 production (*r2* = 0.206). The regression model (Figure 4) further demonstrated that for every decade of life, there is a 13 percent decrease in mean D3 production.

**Figure 3.** Vitamin D3 Production and Age Model. The linear regression model with age as the independent variable and log D3 production was constructed. Age accounted for 20% of the variance in D3 production (*r2* = 0.206).

**Figure 4.** Vitamin D3 Production Age Continuum Modeling. The simple linear regression model with decade as the independent variable and log D3 production was used to demonstrate the 13% decrease in D3 production per decade of life. D3 production at age 70 years is approximately half that produced at age 20. The graph demonstrates that D3 production is possible even in the later decades of life.

#### **4. Discussion**

The primary aim of the current study was to characterize the response of a single bout of sun exposure on cutaneous vitamin D synthesis in a cohort of younger and older adults, and determine if the response in older adults differed from that of a younger cohort. Overall, we found that 30 min of sun exposure (15-min to the arms, torso and legs on both the front and backsides of the body while lying in the supine and prone positions) significantly increased serum vitamin D3 by an average of 9.8 nmol/L with no significant differences in cohorts. The peak response in D3 concentration to a single bout of "sensible sun exposure," however, was about 1/5th the response observed following delivery of 1 MED in a photobiology laboratory [42]. A model created from continuous data on peak D3 concentration in response to sun exposure found evidence that D3 production in adults declines with ageing (by ~13% per decade) but is still possible at 120 years of age. To our knowledge, this is the first study to evaluate the effect of a single bout of sensible sun exposure on vitamin D3 concentration using natural sunlight exposure of a carefully monitored duration as the sole UVB source, which is important for understanding the implication of "sensible sun exposure" guidelines in older as well as younger individuals.

Previous calculations from studies performed in a photobiology laboratory suggest that the equivalent 10,000–25,000 IUs of vitamin D can be synthesized from UVB irradiation in individuals wearing a bikini in peak July sun [43] after exposure of 1 MED, suggesting that sun exposure is the most significant source of vitamin D. While sufficient exposure to the UVB radiation from sunlight is important for cutaneous vitamin D synthesis, too much sun exposure increases the risk for photo ageing and skin cancer [33]. Guidelines for "sensible sun exposure," which are thought to promote vitamin D synthesis at minimal risk of excess exposure, were established from studies conducted within a photobiology laboratory. Only a few studies have investigated and compared (rather than modeled) the efficacy of solar and artificial UVB radiation on cutaneous synthesis [44] or estimated the outdoor exposure time necessary to achieve a serum 25(OH)D concentration equivalent to a specific oral dose (e.g., 1000 IU) of supplemental vitamin D according to season, location and skin type [45].

The most accepted guidelines established by Holick et al. [2] indicate that exposing the arms, hands and face to one-third to one-half of a MED, which is about five minutes at noon in Boston for individuals with skin type II, with a frequency of two to three times per week during the spring, summer or fall, is more than adequate to achieve sufficient vitamin D status [46]. More specific guidelines are difficult to make because of the multiple variables involved in cutaneous production of vitamin D, such as season, latitude, cloud cover, skin pigmentation and body surface area exposed. Updated guidelines suggest exposure of the arms, legs and torso (when possible) to sunlight for approximately 25% to 50% of the time it would take to develop a mild sunburn (e.g., 1 MED) for this same frequency (two–three times/week), and exemplify that if 30 min of noontime sun would cause a mild sunburn, than 10 to 15 min of exposure (followed by sun protection) should be sufficient for adequate vitamin D synthesis [30]. The present study suggests that a single session of solar exposure to both the front and back-sides of the body at the outer limits of the "sensible sun exposure" guidelines was sufficient to promote D3 synthesis in most individuals aged 20 to 69 years who self-identified with skin types II and III, but also that some individuals may be non-responsive to a single exposure. This is of interest because UVB exposure achieved in a natural environment is likely to be more variable than that of a photobiology laboratory, where the dose of UVB delivered can be controlled and directly quantified. For example, one study at a latitude of 56◦ N concluded that artificial UVB exposure to the hands and face was at least eight times as effective at increasing 25(OH)D synthesis than solar UVR from early spring exposure under natural conditions [44]. Another estimated from spectral characteristics that sunbeds were ~25–30% more efficient in producing pre-vitamin D than mid-June sun at 59◦ N latitude [34]. Despite the variability of natural sunlight, achieving vitamin D from "sensible sun exposure" is inexpensive and freely available.

Results of the current study provide important data to help address the influence of ageing on vitamin D synthesis and the appropriate sensible sun exposure guidelines for older individuals. Previous research has demonstrated decreased cutaneous D3 production in older adults. A classic ex vivo study of MacLaughlin and Holick found that skin samples obtained from older adults aged 77 to 82 had significantly less 7-DHC concentration compared to skin samples from young adults [12]. A two-fold decrease in pre-vitamin D3 synthesis in skin samples from the older adults compared to those of an 8- and 18-year-old was also observed [12]. Another study in the photobiology laboratory found that older adults ages 62–80 years with type III skin produced three times less D3 than young adults age 20–30 years with this same skin type following simulated whole-body sunlight exposure of 32 mJ/cm<sup>2</sup> [17]. To our knowledge, however, previous research has not firmly established a specific age at which a reduction in cutaneous D3 occurs.

In the current study, both older and younger individuals experienced significant positive increases in circulating D3 following sun exposure, with only a trend for a different pattern of response between cohorts. The older cohort tended to experience a peak in circulating D3 at 48 h in contrast to the younger cohort, who experienced peak D3 concentration nearly equally at 24 h or 48 h. While studies in the photobiology laboratory indicated that D3 concentrations typically peak within 24 to 48 h following UVB exposure [42], the later peak time in the majority of the older cohort (nine out of 10 participants) could be explained by an age-related decline in the surface area between the dermis and epidermis that ultimately affects nutrient exchange [47,48]. Reduction in basal cell growth in keratinocytes is a major consequence of epidermal thinning [47,48], which has the potential to influence D3 production.

Perhaps more importantly, our regression modeling conducted throughout the adult age continuum (ages 20–69) showed that alterations in cutaneous vitamin D3 production declines throughout adulthood. Our regression modeling (Figures 3 and 4) demonstrated that for every decade of life, there is a 13 percent decrease in D3 production (1.3% per year); by the seventh decade, vitamin D3 production is approximately half that at age 20. The model, however, demonstrates that even at 120 years of age, vitamin D3 production is still possible. This is in support of thinking that " ... skin has a great capacity to make vitamin D even in the elderly" [2]. This also highlights that there is

no specific age at which cutaneous synthesis suddenly stops but rather that it declines slowly over the years.

While not an original intent of the study, the current study also provides interesting results concerning "non-responders" that constituted approximately 17 percent of both cohorts. These non-responders did not have an increase in circulating vitamin D3 concentration following a full 30-min of exposure. Curiously, however, there were little differences between these five individuals and the rest of the group other than their baseline serum D3 concentration, which were on average higher than the 25 responders (34 vs. 11 nmol/L) despite reporting both limited sun exposure and no supplemental vitamin D for at least 3 months prior to study initiation. One individual in the older cohort was also the largest participant, with Class II obesity (BMI > 35.2 kg/m2; body fat = 43.4%). These data combined with our regression models (best subset method) suggest that baseline vitamin D status may influence cutaneous vitamin D synthesis or at least its appearance in circulation. Biochemically it is well-recognized that vitamin D synthesis is regulated through control mechanisms that prohibit vitamin D intoxication through sun exposure by converting excess pre-vitamin D3 to biologically inert molecules including tachysterol and lumisterol [33]. Baseline 25(OH)D concentration, which averaged 33.1 ± 3.0 ng/mL and were generally sufficient, could also have significantly impacted cutaneous vitamin D synthesis [32] even though D3 was more directly influential in the modeling. Other possible explanations include genetic variation [49] or the misclassification of skin type [45] in the non-responders. For example, genetic variants in close proximity genes involved in cholesterol synthesis, hydroxylation and vitamin D transport have been linked to the elevated risk of vitamin D insufficiency in individuals of European descent, and could be present in some of our non-responders [49]. Alternately, misclassification of individuals with skin types IV and V may have resulted in decreased D3 synthesis because time of exposure was not sufficient in these skin types [32].

Additionally, while D3 increased significantly in response to the sun exposure session, 25(OH)D did not change from baseline to 72 h post exposure in the younger and older cohorts or from baseline to 168 h in the older cohort who had data collected at this additional time point. The observed phenomenon may be a result of a combination of factors including the single exposure, sample timing and participant baseline 25(OH)D concentration and/or adiposity [13]. Although vitamin D3 is known to quickly rise within 24–48 h after artificial UVB exposure, 25(OH)D is thought to gradually rise and peak 7 to 14 days post exposure [42]. The kinetics of 25(OH)D synthesis, appearance in circulation and sequestration in adipose tissue following sun exposure, however, has not been fully elucidated and may require several repeated exposures under natural conditions [44], perhaps to achieve a certain peak D3 concentration. The relatively good status of our participants also may have also contributed. Bogh et al., for example, reported an inverse relationship between baseline 25(OH)D and subsequent increases in 25(OH)D concentration after UVB exposure [50]. Additionally, as previously stated, our participant's 25(OH)D levels were quite good, and under these conditions, further elevations in circulating 25(OH)D levels become refractory due to enzyme inhibition [51]. Thus, a rise of only of 9.8 nmol vitamin D3 is simply not enough to drive product levels higher. Furthermore, adiposity in general was relatively high in some of our participants and could have diminished the appearance of 25(OH)D in circulation. This may be related to adipose tissue sequestration of D3 and 25(OH)D following synthesis [52,53].

While to our knowledge this is the first study to evaluate the effect of a single bout of sensible sun exposure on vitamin D3 concentration using sunlight exposure for a carefully monitored duration, the study is not without limitations. Limitations include the relatively small sample size, variable characteristics between the older and younger cohorts (the older cohort had more women), inclusion of participants with optimal baseline serum 25(OH)D, inability to study subjects all on the same exposure day and time span between measurements in the older and younger cohort. The later points, however, are an inherent reality of "sensible sun exposure" guidelines. Our exclusionary criteria limited the number of qualified participants—particularly in the older cohort—combined with the unpredictable number of warm, cloudless days and a limited sample size. The dependence

on accurate self-reporting and knowledge of dietary vitamin D intake, supplement usage and sun exposure may have also allowed for inclusion of participants with optimal serum 25(OH)D that may have accounted for several non-responders. Additionally, more frequent sampling of D3 over the first 72 h (e.g., every 6 to 12 h), a longer follow-up period for 25(OH)D sampling, serial exposures (i.e., three consecutive days) and assessment of age-related thickness differences in the epidermis and/or dermis could be employed in future studies to fully capture the relationship between peak D3, serum 25(OHD concentration [42] and ageing to confirm a rise in 25(OH)D relative to the rise in vitamin D3 following sun exposure. Employment of non-invasive procedures including optical coherence tomography, two-photon microscopy [10] or confocal laser scanning microscopy [11] would be particularly important to better understand the effect of age-related epidermal thickness changes in relation to vitamin D3 production following controlled sunlight exposure.

#### **5. Conclusions**

A single 30-min bout of sun exposure during late spring at close to solar noon was sufficient to observe an increase in vitamin D3 concentration within 24 to 48 h in a cohort of younger and older adults as has been previously observed in the photobiology laboratory with a measured dose of UVB radiation. The response, however, was ~25 to 30% of that observed with delivery of 1 MED in the photobiology laboratory. This study is one of the first to evaluate and practically apply UVB exposure from natural sunlight to examine the effect of subcutaneous vitamin D3 synthesis in younger and older adults. Regression modeling of the appearance of D3 in circulation suggested that age accounted for approximately 20 percent of the variance in D3 production from baseline to peak and revealed a 13% decrease in D3 production with every decade of life. Increases in serum 25(OH)D, however, were not observed at 3 days post-exposure in our younger cohort or at 7 days postexposure in our older cohort. Additional research is needed to better formulate sensible sun exposure guidelines that optimize vitamin D status but avert skin damage, which can lead to skin cancers and other sun associated problems [54]. Additional research is needed to determine the influence of age on timing of peak D3 concentration, investigate the kinetics of in vivo D3 adipose tissue sequestration to explain the relationship between cutaneous D3 production and 25(OH)D response and better understand the specific limitations of ageing (which potentially include reduced dermal and epidermal thickness) on cutaneous vitamin D production.

**Author Contributions:** Conceptualization, D.E.L.-M. and B.M.A.; methodology, J.R.C., L.M.C., P.J.W. and D.E.L.-M.; software, P.J.W.; validation, P.J.W., B.W.H. and D.E.L.-M.; formal analysis, J.R.C., L.C., P.J.W., K.G.G. and D.E.L.-M.; investigation, J.R.C., L.M.C., P.J.W. and D.E.L.-M.; resources, P.J.W., B.W.H. and D.E.L.-M.; data curation, J.R.C., L.M.C., K.G.G. and D.E.L.-M.; writing—original draft preparation, J.R.C., L.M.C. and D.E.L.-M.; writing—review and editing, B.W.H., B.M.A. and J.F.K.; visualization, P.J.W., J.F.K. and D.E.L.-M.; supervision, D.E.L.-M.; project administration, J.R.C., L.M.C. and D.E.L.-M.; funding acquisition, J.R.C., L.M.C. and D.E.L.-M. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded in part graduate student grants from the Rocky Mountain Chapter of the American College of Sports Medicine (L.M.C.) and the University of Wyoming Department of Family and Consumer Sciences (L.M.C. and J.R.C.).

**Acknowledgments:** The authors would like to thank all participants for their time and cooperation. We would also like to thank Katelyn Zavala, RN, Susan Lescznske, RN, Megan Pankey, RN and Rebecca Munn, RN for assistance with blood drawing, Margaretha Troutman and Kelley Fischer for preparing lunches and Sarah Rich for assistance with data entry.

**Data Availability:** Data available upon reasonable request.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
