This research has identified an ethnic-specific association between 25(OH)D and serum hepcidin concentrations, specifically in South Asian and NZ European premenopausal females.

Within this cohort of participants, NZ Europeans who tended to have lower (≤ 3.5 nM) hepcidin concentration had higher (80 nmol·L−1) 25(OH)D concentration. Such results would be in alignment with research from vitamin D supplement studies, demonstrating that one week following a single oral dose of vitamin D3 in healthy adults (n = 28), a 73% decrease in hepcidin was observed [19]. Similarly, Bacchetta et al. [10] observed a decrease in hepcidin concentration by 34%, 24 h following a single oral dose of vitamin D2 (100,000 IU). Previous research has provided evidence to suggest that the reduction in serum hepcidin levels is likely due to the direct binding of 1α,25(OH)2D3 to the vitamin D receptor present on the HAMP promoter gene, decreasing HAMP mRNA expression [10]. The results in the NZ Europeans would suggest that in the absence of vitamin D supplementation, adequate vitamin D levels concentrations are likely to be associated with low levels of hepcidin. However, since this study did not measure HAMP gene expression, it is likely that basal iron status is still the primary determinant of variations in serum hepcidin [5], but that adequate vitamin D levels may support this appropriate homeostatic hepcidin activity. Future research may be required to verify these findings, particularly in the absence of vitamin D supplementation.

Conversely, South Asians in this study who tended to have higher (> 3.5) hepcidin concentrations had higher (38 nmol L−1) 25(OH)D concentrations. This finding contradicts the association observed in the NZ Europeans and results reported previously in other cohorts (Caucasian, African American, and Korean) [10, 13]. To our knowledge, this association has not been reported in any of the previously published literature. Within our cohort, the South Asians had significantly higher BF% (38.34%) and BMI (25.88 kg/m2) compared to NZ Europeans (27.49% and 22.91 kg/m2, respectively). It is possible that the higher BF% could be a contributing factor to the increased IL-6 concentration observed in South Asians. Interleukin-6 is known to be released by adipose tissue and is directly correlated with adiposity [8], in addition to its role as an upregulator of HAMP gene transcription. It is worth noting that body composition has been demonstrated to be associated with ethnicity [29, 30]. For the same BMI, Asians are frequently reported to have higher BF% compared to Caucasians, this is particularly prominent in South Asians, Malay, and Chinese [30]. Furthermore, it has been observed that South Asian females have a higher fat to lean mass ratio compared to Australian Aboriginal, Chinese, and European females [29]. Specifically in NZ, Asian Indians were identified to have 8% higher BF% compared to Europeans [31]. Potential factors that could be contributing to the varying body composition in ethnic groups include genetics, environmental, and intra-uterine development [30]. Therefore, it is possible that the higher BF% and IL-6 levels may be contributing factors to the increased hepcidin levels and the positive association with 25(OH)D levels observed in the South Asian participants within this study.

There are additional factors that may have contributed to the contradictory relationship between hepcidin and 25(OH)D in South Asian premenopausal females within the study. Firstly, it was noted that the geometric mean of 25(OH)D in the cohort of South Asians was lower compared to NZ Europeans and ‘other’ ethnic groups. Only 33% of South Asians were identified as having adequate vitamin D levels (≥ 50 nmol·L−1). In comparison, 95% of NZ Europeans had adequate 25(OH)D concentration with a geometric mean of 75 nmol L−1. In NZ, South Asians have been identified as a high-risk group for vitamin D deficiency [32]. In a cohort of 228 South Asian females, the median 25(OH)D concentration was 28 nmol L−1, with only 16% of the cohort presenting with sufficient (≥ 50 nmol·L−1) 25(OH)D concentrations [32]. Previous research studies have demonstrated that adequate levels of vitamin D are likely to have anti-inflammatory actions through the downregulation of T helper 1 cells, subsequently reducing the production of pro-inflammatory cytokines, including IL-6 [33]. Furthermore, in vitro research has demonstrated a dose-dependent relationship between 1,25(OH)2D3, reducing pre-hepcidin cytokines, IL-6, and IL-1β [34]. Therefore, there is the possibility that the lower mean 25(OH)D concentration observed in South Asians may not be sufficient to exert a suppressive effect on IL-6 concentrations or the HAMP transcription, and as a result, hepcidin levels may have remained elevated in this ethnic group within this study. However, this proposition requires further investigation with research that will determine if there is a dose-dependent effect of 25(OH)D on IL-6 and hepcidin expression in different ethnic cohorts.

Within this study, no association was identified between 25(OH)D and SF concentration in any ethnic group within this research. Such results would conflict with those noted in Caucasian female athletes and African Americans, where individuals who presented as vitamin D deficient (< 75 nmol L−1) were noted as being at a greater risk of iron deficiency [11]. Conversely, the association between vitamin D levels and iron deficiency risk was not noted for the Caucasians within a mixed-gender cohort [12], a result that is similar to the lack of association between vitamin D and SF noted in this current study. The inconsistencies in the association between vitamin D and iron deficiency status are not limited to cross-sectional research studies but have also been noted in vitamin D supplement trials [12]. Due to the inconsistencies in research design, gender, and ethnicity, future research is required to verify if the association between vitamin D and iron deficiency risk is associated with SF or hepcidin.

Finally, the hepcidin cut-off for stratification used in the study (> 3.5 or ≤ 3.5 nM) was based on the median hepcidin concentration of our cohort. However, this threshold has been suggested by recent research as a possible determinant of iron depletion, with previous results demonstrating increased iron absorption rates in females when hepcidin concentrations fell below ≤ 3.09 nM [35]. Within this previous research, SF of 51.1 µg·L−1 corresponded to hepcidin levels of 3.09 nM and was subsequently suggested as an initial threshold required for iron deficiency detection in young premenopausal females [35]. However, within the current study, SF means for all ethnic groups was below 51.1 µg·L−1, and therefore, all participants may have been at risk of iron depletion and in time iron deficiency, regardless of vitamin D concentrations. Within the NZ Europeans, reduced hepcidin concentration observed in those with higher 25(OH)D concentrations may suggest adequate homeostatic response of hepcidin to the individual’s current iron status. However, in South Asians, despite the reduced SF concentration, hepcidin remained elevated. This may suggest additional confounding factors (e.g. BF%, IL-6, dietary intake) could strongly influence hepcidin concentration when iron stores are reduced, thus impacting iron homeostasis in this cohort.

This study is the first cross-sectional study in healthy premenopausal females that includes measures of serum hepcidin when investigating the associations between vitamin D and iron status. Due to the complementary and then conflicting observation of 25(OH)D and hepcidin concentration in NZ European and South Asians, respectively, further research is needed to increase knowledge on the dose-dependent effect of vitamin D on hepcidin concentrations and iron regulation in these populations. Consideration for this future research is the assessment of vitamin D binding protein, which is known to be ethnically diverse and may contribute to the interaction between vitamin D and hepcidin that was observed in the current study [36]. In addition, researchers may consider conducting research in larger samples of each of the ethnic groups identified in the current study, as we acknowledge that the current power calculation of the study was not ethnic group specific. Finally, to improve the results presented here, researchers may seek to collect data in defined seasons and in age-matched groups to reduce any influence this may have on the iron and vitamin D results.