Bioavailability of vitamin D biofortified pork meat: results of an acute human crossover study in healthy adults

Abstract Vitamin D intakes are concerningly low. Food-based strategies are urgently warranted to increase vitamin D intakes and subsequently improve 25-hydroxyvitamin D (25(OH)D) concentrations. This acute randomised three-way crossover study investigated the efficacy of vitamin D biofortified pork derived from pigs exposed to UVB light to increase serum 25(OH)D3 concentrations, compared to a dose-matched vitamin D3 supplement and control pork in adults (n = 14). Blood samples were obtained at baseline and then 1.5, 3, 6, 9 and 24 h postprandially. There was a significant effect of time (p < 0.01) and a significant treatment*time interaction (p < 0.05). UV pork and supplement significantly increased within-group serum 25(OH)D3 concentrations over timepoints (p < 0.05) (max. change 0.9 nmol/L (2.2%) UV pork, 1.5 nmol/L (3.5%) supplement, 0.7 nmol/L (1.9%) control). Vitamin D biofortified pork modestly increased 25(OH)D3 concentrations and produced a similar response pattern as a dose-matched vitamin D supplement, but biofortification protocols should be further optimised to ensure differentiation from standard pork.


Introduction
Achieving an optimal vitamin D status is of critical public health importance. Vitamin D is required for good musculoskeletal health (Institute of Medicine 2011) and a range of putative non-skeletal health effects, most notably immune function (Kheiri et al. 2018;Manson et al. 2019;Murdaca et al. 2019;Scragg 2019;Martens et al. 2020;Pittas et al. 2020;Sluyter et al. 2020). Vitamin D inadequacy (< 50 nmol/L; Holick et al. 2011) is a global concern, with reported prevalence as 24, 37 and 40% in the United States, Canada and Europe, respectively (Sarafin et al. 2015;Cashman et al. 2016;Schleicher et al. 2016).
Whilst vitamin D can be synthesised following skin exposure to ultraviolet-B (UVB) light (290-315 nm) (Holick 2007), owing to public health concerns regarding sun safety, in combination with modern indoor lifestyles, it is evident that endogenous synthesis may not be an effective means of maintaining an adequate vitamin D status across the year (Fink et al. 2019). Thus, dietary sources remain an important component to achieve optimal circulating 25(OH)D concentration and yet, relatively few foods contain high levels of vitamin D. Within the UK population, key dietary contributors to vitamin D intake include meat (25%), eggs (22%), fish (17%) and cereal products (16%) (Public Health England 2020). It can therefore be challenging to achieve a sufficient vitamin D status from food sources alone. Even when accounting for supplement usage, vitamin D intake in UK adults is approximately half of the current recommended nutrient intake (RNI) (Public Health England 2020). There is clear mounting evidence that populations are unable to meet vitamin D requirements through current levels in food, and additional food-based strategies are urgently warranted to ensure an increase in consumer vitamin D intakes and status (Cashman 2020;Buttriss et al. 2022). Thus, current strategies are failing to improve vitamin D status to ensure sufficient 25hydroxyvitamin D (25(OH)D) concentrations [a reliable biomarker of vitamin D status] are achieved. A recent simulation study within a Dutch population suggested it is not possible to achieve adequate vitamin D intake by consuming natural sources without implications on energy intake and carbon emissions from farm-to-fork (Bruins and L etinois 2021). Therefore, it appears sensical to enhance the vitamin D content in commonly consumed foods through other means. Moreover, there is a demand to push innovation in the food industry to benefit consumers and at-risk subgroups (Cashman and Kiely 2016).
Biofortification ('bio-addition' or 'bio-enrichment'), particularly of animal-derived foods, represents a potential food-based strategy to reduce rates of hypovitaminosis D (Buttris et al. 2022). Whilst a myriad of on-farm studies demonstrate that UV exposure and/or supplementing feed increases vitamin D concentrations in pork meat, there is no evidence from human randomised controlled trials (RCTs) to determine efficacy (Neill et al. 2023a). Vitamin D biofortification RCTs are currently limited to eggs, fish, bread baked with UV-treated yeast and mushrooms as the food vehicle (Neill et al. 2023a). Except for bread baked with UV-treated yeast (Itkonen et al. 2016), in general, biofortified foods improved vitamin D status. Pork meat is widely consumed globally and naturally contains both vitamin D 3 and the 25(OH)D 3 metabolite; the latter of which is potentially five times more potent (Cashman et al. 2017;Guo et al. 2018;Quesada-Gomez and Bouillon 2018;Graeff-Armas et al. 2019;Jakobsen et al. 2019).
Thus, the aim of the current study was to investigate the efficacy of vitamin D biofortified pork to increase 25(OH)D 3 concentrations, compared to a dose-matched vitamin D 3 supplement and control pork in adults.

Preparation of vitamin D UV-biofortified pork
Full details of on-farm methodology are reported by Neill et al. (2023b). Twenty male and female pigs were randomly allocated to daily UV or non-UV (control) treatments for 9 weeks prior to slaughter, and both groups were offered feed containing the maximum EU-regulated vitamin D 3 (2000 IU/kg; (EFSA) European Food Safety Authority 2005). Vitamin D-biofortified pork derived from pigs exposed to UVB light (hereafter referred to as 'UV pork') and control pork were processed into loin before being cooked and minced. All cooked loins (n ¼ 7 UV, n ¼ 9 control) were analysed individually by liquid chromatography-tandem mass spectrometry (LC-MS/MS) and an equal subsample of loins (n ¼ 2 UV, n ¼ 2 control) were then selected for the present RCT based on the highest and lowest vitamin D 3 results, from each treatment group, respectively (Table 1). A summary of these full results is presented elsewhere (Neill et al. 2023b). The portion size of the biofortified UV pork (340 g cooked weight) was calculated to dose-match the vitamin D 3 supplement (5 mg). The portion size of the control pork was then matched to that of the UV pork (1.82 mg vitamin D 3 /340g portion). Researchers were blinded to the two types of pork for the RCT.

LC-MS analysis of pork used in human RCT
Concentrations of vitamin D metabolites (vitamin D 3 , vitamin D 2 , and their respective hydroxylated forms, 25(OH)D 3 and 25(OH)D 2 ) in cooked minced pork loin were measured at Agri-Food and Biosciences Institute (Belfast, Northern Ireland). One loin from each pig was analysed in duplicate but both loins were used in the RCT. Extraction of the four main vitamin D forms was adapted from previously described methods (Ding et al. 2010;Trenerry et al. 2011;Strobel et al. 2013) using liquid chromatography-tandem mass Table 1. Vitamin D concentrations in UV and control pork loin from 9-week on-farm intervention.

Subjects
Fourteen apparently healthy Caucasian male (n ¼ 5) and female (n ¼ 9) free-living adults (23 ± 6 y) were recruited from Ulster University and the surrounding area between December 2019 and February 2020. The study exclusion criteria are presented in

Study protocol
This study was a randomised-controlled, double-blind three-way crossover postprandial study (n ¼ 14). Due to the COVID-19 pandemic and aligned with governmental advice to avoid all unnecessary social contact and travel, research studies involving direct human contact were suspended at Ulster University on 17 March 2020 which affected the final visit for three participants (scheduled for 26 March 2020). A washout period of 2 weeks between the three intervention exposures, previously implemented by similar studies (Guo et al. 2017;Wagner et al. 2019), was selected due to half-life of vitamin D (24 h) and 25(OH)D (3 weeks) elimination from circulation (Hollis and Wagner 2013). Participants were randomly assigned to receive either UV pork, control pork or a vitamin D supplement during their first visit, followed by a 2-week washout period ( Figure 1). When participants returned for their second and third visits, they crossed over to receive a different treatment on each occasion. The random order assignment of treatments to participants, conducted by the clinical trials manager using MINIM software (Evans et al. 2013), was stratified by sex, age group and body mass index (BMI, kg/m 2 ) with an allocation ratio of 1:1:1.
Following fasted baseline (0 h) blood sample, the study coordinator administered the pork or supplement as a study breakfast. Pork was mixed with a 150 g Dolmio tomato and basil sauce (Mars Ltd.) and heated in a microwave to an internal temperature of 63 0 C. Pork was weighed before serving and any Table 2. Inclusion exclusion criteria.

Inclusion Criteria
Exclusion criteria Free-living, apparently healthy Caucasian adults Aged 18-65 years at recruitment Body mass index (BMI) !18.5 and <25 kg/m 2 If consuming vitamin D supplements, willing to discontinue 4 weeks prior and for duration of study Non-smokers Non-Caucasian adults Adults <18 or >65 years at recruitment Taking vitamin D supplement and not willing to discontinue vitamin D supplementation for 4 weeks prior to and for duration of study Current smokers Pregnant/lactating females Use of tanning facilities or winter vacation planned during the intervention period to a location expected to increase cutaneous synthesis Severe medical illness Medications which interfere with vitamin D metabolism, e.g. steroid medications (e.g. prednisone), weight loss drug orlistat (e.g. Xenical and Alli), cholesterol-lowering drug cholestyramine (e.g. Questran, LoCholest and Prevalite), seizure drugs Phenobarbital and Dilantin, anti-tuberculosis, statins or thiazide diuretics Intestinal malabsorption syndrome Excessive alcohol use (>14 units/ week) leftovers were weighed and recorded. Participants were asked to consume the pork as quickly as possible. Compliance (%) was calculated using the amount of pork left unconsumed, if any, by each participant, i.e. [weight of pork consumed/weight of pork provided] x 100. Participants receiving the supplement were provided with an identical test breakfast including two slices (80 g) of toasted white bread with 40 g strawberry jam. Full details on how study breakfasts were packaged, prepared and presented to participants are detailed in Supplemental Material. All participants received the same lunch which was eaten in the clinic dining area. All meals were controlled for vitamin D content and bottled water was provided. Five serial blood samples were obtained over at 1.5, 3, 6, 9 and 24 h postprandially. Participants were asked to maintain their normal diet and lifestyle for the duration of the study; however, a list of vitamin D-rich foods (e.g. oily fish, liver, egg yolk and fortified products) to avoid throughout the 24-hour period was provided

Physical measurements and dietary assessments
Height and weight were measured before the baseline blood sample to determine BMI (kg/m 2 ). Standing height (m) was measured at the first appointment only to the nearest 0.5 cm using a calibrated stadiometer (SECA, Model 220, Germany). Body weight (kg) was measured at all study appointments, recorded without footwear or heavy clothing and measured to the nearest 0.1 kg using portable scales (Seca; Brosch Direct Ltd, Peterborough, UK). During the last appointment, all participants completed a validated food frequency questionnaire (Weir et al. 2016) to assess habitual vitamin D intake.

Blood collection and 25-hydroxyvitamin D analysis
Whole blood samples were collected into labelled 4 ml serum tubes (Greiner Bio-One Ltd) and allowed to clot for 30 min at room temperature. Samples were centrifuged (3500 rpm; 15 min; 4 C) and the supernatant apportioned to labelled 1.5 ml aliquots. Aliquots of serum were stored in monitored À80 C freezers at Ulster University according to Human Tissue Act (HTA) standards until analysis.
Serum samples were analysed for vitamin D metabolites (25(OH)D 2 and 25(OH)D 3 , ng/ml) at Ulster University with high-performance liquid chromatography-tandem mass spectrometry (HPLC-MS/MS) according to a method adapted from Aronov et al. 2008. A full description of the analytical method is provided in Supplemental Material. In brief, intermediate internal standard stock solution (50 ml, 100 ng/ml), quality controls (200 ml x 3) and 500 ml precipitation reagent (acetonitrile with 0.1% v/v formic acid) were added to the serum sample (200 ml), vortexed mix at full power for 30 sec and centrifuged (7558 g; 5 min). 600 ml supernatant was then mixed with 700 ml methyl tert-butyl ether (MTBE), vortexed for 30 sec at full power and centrifuged (7558 g; 5 min). 1 ml of the resulting supernatant was then evaporated to dryness under nitrogen and reconstituted in 100 ml PTAD (5 mg PTAD in 10 ml acetonitrile) before being transferred to a 200 ml HPLC insert with 50 ml distilled water.
A Shimadzu Nexera UHPLC system (Kyoto, Japan) using a polar C18 column (Phenomenex, Torrance, US) in a gradient mode from 50% acetonitrile with 0.1% v/v formic acid to 100% acetonitrile achieving a run time of 15 min was used to perform the chromatographic separation. Mass spectrometry analysis was performed using an API 4000 mass spectrometer (ABSCIEX, MA, UK) employing a turbo ion spray source in positive ion mode. Multiple reaction monitoring (MRM) was used as an acquisition method. Quantitation was performed by internal standard ratios using a linear regression model.
Results were populated as ng/ml and thus multiplied by 2.5 to convert to nmol/L. Both 25(OH)D 3 and 25(OH)D 2 were extracted, however the majority of vitamin D 2 results were below the lower limit of quantification (LLOQ) and thus excluded from further analyses. Dynamic range for 25(OH)D 3 and 25(OH)D 2 was 2.5-200 ng/mL and the LLOQ of the method was 5 ng/mL and 2.5 ng/mL for 25(OH)D 2 and 25(OH)D 3 , respectively.

Statistical analysis
Statistical analyses were performed using the Statistical Package for the Social Sciences (SPSS) (IBM SPSS Statistics for Windows, version 25.0; IBM Corp.), with significance set at p < 0.05 throughout. Normality of data was assessed by the Shapiro-Wilk test and natural logarithm transformation was conducted when necessary to normalise data (or approach normality). Formal power calculations could not be performed owing to a lack of previous studies using biofortified pork, however, based on comparable crossover bioavailability studies within the area (Guo et al. 2017;Wagner et al. 2019), a sample size of 14 was selected. Vitamin D sufficiency, insufficiency and clinical deficiency were categorised as >50, 30-50 and <30 nmol/L, respectively (Holick et al. 2011;Institute of Medicine 2011). Aligned with the Consolidated Standards for Reporting Trials guidelines, missing data were subjected to intention-to-treat (ITT) analysis (Moher et al. 2010). Therefore, all participants (n ¼ 14) randomised at baseline were included in statistical analyses. Serum 25(OH)D 3 concentrations were analysed as ITT to account for the two participants who dropped out after the first and second visit and three participants who could not attend the third visit due to the COVID-19 national lockdown. One-way ANOVA and Tukey post hoc tests were performed to ascertain between-group differences in 25(OH)D 3 concentrations. To assess within-group differences and the effect of treatment, time, and treatment Ã time interactions, the time course from baseline (0 h) to 24 h were analysed by two-factor repeated measures ANOVA. Postprandial summary measures, including area under the curve (AUC), incremental AUC (iAUC), maximum concentration (Cmax), time to reach Cmax (Tmax) and increment from baseline to maximal concentrations [iCmax (iCmax ¼ Cmaxfasting value)] were conducted. A double-blinded protocol was upheld until statistical analyses were completed.

Vitamin D concentration of pork used in human RCT
Means ± standard deviation (SD) vitamin D 3 , 25(OH)D 3 , and total vitamin D activity of UV and control pork are presented in Table 1. Pork loins containing the highest vitamin D 3 concentration from the UV group and lowest vitamin D 3 concentration in the control group were selected for consumption in the human RCT. Full onfarm results are presented elsewhere (Neill et al. 2023b). The vitamin D 3 :25(OH)D 3 ratio for UV and control pork samples was 6.45 and 3.48, respectively. Mean compliance of pork consumption was 98% (range 86-100%). Mean time taken to consume pork serving was 28 ± 22 min.

Characteristics of subjects in the RCT
Of the 14 adults who started the study, nine completed the intervention. By endpoint, 11 participants had consumed UV pork, 13 consumed control pork and 11 received the vitamin D supplement (Figure 2). Baseline characteristics are presented in Table 3.
All study visits were completed from January to March. At baseline visit 1, vitamin D sufficiency (>50 nmol/L), insufficiency (30-50 nmol/L) and the clinical deficiency (<30 nmol/L) were evident in 21, 43 and 36% of the participants, respectively. The mean 25(OH)D 3 concentration in the total group shows insufficient vitamin D status at the start of the study in (39.7 ± 20.5 nmol/L) (Table 3). Fasting baseline (0 h) serum 25(OH)D 3 concentrations at visits 1, 2 and 3 were not significantly different (p > 0.05). BMI did not differ with the exception of groups A and B during their third visit (i.e. when consuming supplement and UV pork, respectively; p < 0.05). Participants (n ¼ 9) had a mean ± SD dietary vitamin D intake of 4.81 ± 3.65 mg/day.

Effects of intervention
Following the consumption of the three treatments, and when applying ITT analysis, there was a significant effect of time (p ¼ 0.003) and a significant treatment Ã time interaction (p ¼ 0.037) (Table 4). Treatment was non-significant (p ¼ 0.145). Per protocol analysis only showed significant time effects (F (5, 40) ¼ 3.43; p ¼ 0.046). The following results hereafter refer to ITT analysis. Both UV pork and supplement resulted in within-group differences in serum 25(OH)D 3 concentrations over time (p < 0.05), with the highest 25(OH)D 3 concentrations (Cmax) observed at 9 h timepoint for both (Table 4). Post-hoc tests revealed a similar pattern of change over time within the UV pork and supplement group. After a slight decrease from baseline to 1.5 h timepoint (only significant in the supplement group), 25(OH)D 3 concentrations peaked significantly at 9 h, returning to baseline values by the 24 h timepoint (Table 4). There was no difference between treatment groups at any timepoint (p > 0.05), including baseline (0 h). The mean AUC was markedly higher for both UV pork and supplement, compared to control pork (supplement (95% CI): 1027 nmol/L (628-1427); UV pork: 987 nmol/L (611-1362); control pork: 894 nmol/L (519-1270), however, this did not reach statistical significance (p > 0.05). Pairwise comparisons for iAUC, Cmax and iCmax showed no difference between treatments in ITT or per protocol analyses. The Tmax was 9 h for both UV pork and supplement, producing Cmax of 42.4 and 44.3 nmol/L, respectively. For control pork, the Tmax for serum 25(OH)D 3 concentration was 3 h (Cmax 38.4 nmol/L). Similar results were apparent when participants were stratified based on sufficient (!50 nmol/L) or insufficient (<50 nmol/L) vitamin D status at baseline. The magnitude of change for vitamin D from baseline to Cmax was 0.9 nmol/L, 1.5 nmol/L and 0.7 nmol/L for UV pork, supplement and control pork, respectively. Relative change was reported as 2.17% for UV pork, 3.50% for supplement and 1.86% for control pork.

Discussion
Despite many on-farm studies highlighting the feasibility and effectiveness of increasing vitamin D in pigs by feed supplementation and/or UV radiation (Neill et al. 2023a), to our knowledge, this RCT is the first to investigate the postprandial effect of vitamin D-biofortified pork on 25(OH)D 3 concentrations in human participants.
In our three-way, acute crossover study in healthy adults during winter in Northern Ireland, biofortified pork derived from pigs exposed to UVB light showed a similar response pattern as a dose-matched vitamin D 3 supplement. A significant between-group effect was observed in time points of both UV pork and supplement, reaching maximum concentration at 9 h and suggesting the food matrix did not impact absorption compared to supplemental vitamin D 3 . Traditional exogenous fortification long-term studies have similarly demonstrated equal bioavailability between vitamin D supplements and enriched cheese (Wagner et al. 2008), bread (Natri et al. 2006) and orange juice (Biancuzzo et al. 2010). However, in the current study, neither UV pork nor the supplement resulted in significantly different 25(OH)D 3 concentrations at any time point compared to control pork, which peaked earlier (3 h) and did not show withingroup change over time. Of note, the control pork still provided vitamin D 3 owing to its presence in animal feed.
Significant changes to 25(OH)D 3 concentrations from 0-9 h suggest pork meat may be an effective vehicle for vitamin D biofortification due to its liberation from the food matrix. The Cmax observed in vitamin D treatment groups in the present study (42.4-44.3 nmol/L) is in accordance with results from a fortified dairy drink 24-hour postprandial study (34.6-40.2 nmol/L) providing 20 mg vitamin D 3 or 25(OH)D 3 (Guo et al. 2017). However, peak plasma 25(OH)D 3 was observed later at 24 h (Tmax) and mean baseline status was $8nmol/L lower. Alternatively, an acute 48-hour supplementation study providing 500 mg vitamin D 3 produced 58.5-118.25 nmol/L Cmax with Tmax vitamin D 3 at 10 h, despite a substantially higher baseline status (Wagner et al. 2019). Whilst there are some similarities between the Cmax and Tmax observed in the present work and higher dose trials (Guo et al. 2017;Wagner et al. 2019), direct comparison proves difficult owing to differing single-dose, vitamin D vehicle, metabolite analysed and baseline status of the cohorts included.
Our biofortified pork contained elevated concentrations of vitamin D 3 and the arguably more potent metabolite, 25(OH)D 3 , unlike the supplement which contained only the parental form of vitamin D. An older study (Van den Berg 1997) estimated the relative bioavailability of pig meat as approximately 60% compared to a supplement, however this related to vitamin D 2 , rather than vitamin D 3 ; the latter of which would be more prominent in animal sources. In addition, pork naturally contains fat, aiding vitamin D absorption, whereas the gummy supplement (containing glucose syrup) was served with a low fat, carbohydrate and sugar-based breakfast meal. Previous research is inconsistent regarding the effect response based on differing fat content in vitamin D enriched food and beverages (Tangpricha et al. 2003;Wagner et al. 2008;McCourt et al. 2022). It could be argued that the type of vitamin D supplement used (e.g. capsule, gummies, oral spray or drops) may have impacted bioavailability and thus its comparison with UV pork (Wagner et al. 2019;Grammatikopoulou et al. 2020). Nevertheless, it is more likely if any difference did exist, this would be minimal owing to the low 5 mg serving.
Despite these considerations, we found that the food matrix did not improve absorption or have a noticeable difference on serum 25(OH)D 3 concentrations compared to the matched vitamin D Table 4. Postprandial serum 25-hydroxyvitamin D 3 (25(OH)D 3 ) concentrations (nmol/L) in healthy adults after consuming either biofortified pork, matched-dose vitamin D 3 supplement or control pork (n ¼ 14). Data presented as mean ± SD. Treatment: p ¼ 0.145, Time: p ¼ 0.003, Treatment Ã Time: p ¼ 0.037; two-factor repeated measure ANOVA.
Values not sharing a common superscript letter in rows (a, b, c) are significantly different (p < 0.05) between timepoints; within-group repeated measures ANOVA with pairwise comparisons (Tukey post-hoc tests). No significant difference between treatments at any timepoint; one-way ANOVA. Control pork from pigs which received no UV exposure. SD, standard deviation; hr, hour; UV, ultraviolet; n, number of participants.
supplement. This may be explained by the modest dose of vitamin D 3 offered (5 mg/portion) or the acute 24-hour study design as longer-term exposure may be required to confirm such a hypothesis. Of note, the 5 mg vitamin D concentration provided in the present study is the specified EU nutrient reference  Research Council 2006) for varying life stages and therefore, shows efficacy within a longer-term duration, as confirmed in an earlier RCT (Cashman et al. 2008). Similar outcomes may be observed if UV pork was consumed in the longer-term as part of habitual diets, however this would need to be confirmed in larger, chronic human RCTs. Whilst magnitude of response did not differ between treatments in the present study, should vitamin D-biofortified pork show significant difference to status from longer studies, it nevertheless would need to be considered alongside the (bio)fortification of a variety of foods, to account for dietary diversity and ensure potential benefits of biofortification reach the majority of the UK population. Previous longer-term RCTs in biofortified eggs, fish and mushrooms have demonstrated improved human circulating 25(OH)D 3 , 25(OH)D 2 and total 25(OH)D responses following longer-term consumption (Pinto et al. 2020;Neill et al. 2023a). If future research in biofortified pork identified significant status response and was subsequently integrated within a mandatory policy, a lower vitamin D concentration, such as that presented in the current work, may be deemed more effective when enriching a range of sources to avoid risk of overdose (Hirvonen et al. 2007). Unlike fortification and supplemental studies, vitamin D biofortification is inherently restricted by the level of natural UV-induced enhancement biologically possible, in addition to maximum vitamin D restrictions within European Union animal feed regulations ((EFSA) European Food Safety Authority 2005). Inter and intra-variability are to be expected from on-farm vitamin D biofortification trials and such data will naturally only be available at the end of intervention periods post-slaughter whereby further biofortification increases are no longer possible for that group of animals. Conversely, this may offer appeal as, similar to humans, pigs are capable of self-regulating cutaneous synthesis by a tightly controlled feedback system to prevent toxicity whereby pre-vitamin D 3 is converted to biologically inert tachysterol and lumisterol following UVB exposure (MacLaughlin et al. 1982).
Participant vitamin D dietary intakes in the current study were suboptimal (4.81 ± 3.65 mg/day), approximately half of 10 mg/day currently recommended ((SACN) Scientific Advisory Committee on Nutrition 2016). Whilst low, this result is considerably higher than the mean vitamin D intakes from food alone amongst UK adults in the National Diet and Nutrition Survey (NDNS) (2.9 ± 2.3 mg/day; Public Health England 2020). This may be partly attributed to the different dietary assessment methodologies, where the FFQ used in the current study is more likely to capture habitual long-term intakes compared to the shorter-term 4-day food diary used in NDNS (Weir et al. 2016).
Meat hosts a wealth of important macro and micronutrients such as protein, B-vitamins, zinc and iron (Williams 2007;Marangoni et al. 2015), and even without biofortification strategies, remains the major contributor (30%) to adult vitamin D intakes in the UK (Public Health England 2018). Nevertheless, it would be amiss to fail to mention environmental and public health messaging encouraging a reduction in red and processed meat consumption (Godfray et al. 2010;(SACN) Scientific Advisory Committee on Nutrition 2010; Springmann et al. 2018;Willett et al. 2019). Aligned with guidance to limit excessive meat consumption, in the future biofortification may reassure consumers that despite fewer eating occasions and/or smaller portions, biofortified pork meat can still provide significantly to vitamin D recommended nutrient intake (RNI) levels (10 mg/day; (SACN) Scientific Advisory Committee on Nutrition 2016). Furthermore, as supplemental uptake is generally low (17%) amongst UK populations (Public Health England 2020), biofortified foods proven to increase vitamin D status within human RCTs, may offer an effective alternative to those averse to supplementation or foods rich in vitamin D such as oily fish, liver and egg yolk, as consumers can maintain their preferred dietary habits whilst still benefiting from additional vitamin D intakes. Importantly, sensory analysis should be investigated to confirm speculated consumer acceptability and any negative change in sensory properties.

Strengths and limitations
The strength of our study lies in the novelty of data presented as, to the best of the author's knowledge, human response to any vitamin D-biofortified meat, let alone pork meat specifically, has not been investigated to date. The crossover study design affords robust comparability, with participants acting as their own control, minimising the risk of confounding factors and providing greater biological homogeneity. In addition, despite the animal feed providing the maximum permitted vitamin D for pig diets in Europe (2000 IU/kg), the study was still able to induce an increase with UV exposure and is translatable in its industrial application. Considering laboratory analysis, LC-MS was used which is currently referred to as the 'gold standard' method for analysing vitamin D concentrations owing to its recognised flexibility, sensitivity, and specificity (El-Khoury et al. 2011).
Nevertheless, some potential limitations should be noted. The authors acknowledge the modest dose of vitamin D 3 provided (5 mg), hindered by the substantial pork serving size, and the sufficient baseline status of some participants. Owing to a lack of consensus regarding the most appropriate circulating 25(OH)D concentration, in addition to ethical and practical considerations, the current study did not measure participants' 25(OH)D concentration as part of screening (Pittas et al. 2020). Instead, eligibility criteria were set in such a way to target low vitamin D consumers, who may have been more at-risk of insufficient vitamin D status. Some participants (n ¼ 3), however, presented with sufficient baseline vitamin D status (> 50 nmol/l; 68.5 ± 12.1 nmol/L) which may have affected the 25(OH)D 3 concentration increase observed over 24 h. In future work, measuring circulating 25(OH)D 3 concentrations during screening may avoid recruitment of nutritional replete participants and ensure vitamin D interventions target those who would naturally benefit most from such interventions. Nevertheless, an adequate vitamin D status does not necessarily equate to an optimal vitamin D status, and debate surrounding differing deficiency cut-off values adds a lack of clarity to this important consideration (Cashman 2020).
The present study provided a supranatural portion of pork which is not realistic in real-life scenarios and caused some participants to struggle to consume the pork serving in a timely manner. Increasing pigs' UV exposure to consequently increase vitamin D concentration in pork could ensure a more psychologically appealing portion with a greater vitamin D single dose, which should be considered by future biofortification studies. Lastly, results from this work are limited in application to healthy pork-consuming Caucasian populations, yet such food-based strategies are paramount for at-risk groups such as ethnic minority populations and those housebound who herald greater risk of deficiency (Guo et al. 2019). Whilst widespread vitamin D-biofortified pork production may improve vitamin D intakes at a population level (Neill et al. 2021), undoubtedly, the greatest impact from biofortification will be observed by its implementation across a range of animal and plant-based foods to account for dietary preference and should be explored.

Conclusion
Despite concern for low vitamin D status in UK populations, deficiency will remain a major public health issue unless effective dietary strategies are established to address limitations in the current food supply. The present study provides robust and novel data suggesting vitamin D-biofortified pork meat increases modest 25(OH)D 3 concentrations postprandially. Commoditybased biofortification may provide an innovative and viable additional food-based approach to suboptimal vitamin D status, which should be considered as part of a wider mandatory (bio)fortification policy. Further investigation is warranted to realise the full potential of biofortification strategies in the future of pork meat production and implications on longer-term vitamin D status in consumers.

Funding
This work was funded as part of a Department for the Economy (DfE) Co-operative Awards in Science and Technology (CAST) PhD studentship, supported by Devenish Nutrition Limited. Data availability statement Due to the nature of this research, participants of this study did not agree for their data to be shared publicly, so supporting data is not available.