Vitamin D is increasingly recognised as an important factor in regulating immunity and respiratory health, and vitamin D deficiency is associated with several immune-mediated airway diseases, including asthma and chronic obstructive pulmonary disease (COPD) (1–3). This is attributed to its diverse immuno-modulatory functions including the capacity to promote anti-microbial pathways, suppress inflammatory responses while enhancing immuno-modulatory functions such as anti-inflammatory IL-10 production (4, 5).

Alpha-1 antitrypsin (AAT) plays a critical role in protecting lung parenchymal tissue from direct elastinolytic degradation and pro-inflammatory effects of serine proteases, notably neutrophil elastase (6). It is predominantly synthesized in the liver by hepatocytes and is secreted into the plasma as the most abundant circulating antiprotease. The metastable structure of native AAT and other members of the serine protease inhibitor (serpin) superfamily provides the potential for stabilizing conformational change (7). This is utilized in the functional mechanism of protease inhibition, but also renders AAT vulnerable to point mutations, as seen in the Glu342Lys (Z) variant. Within hepatocytes, that are the major source of systemic AAT, this triggers misfolding in the endoplasmic reticulum, aberrant conformational change and stabilizing intermolecular linkage to form AAT polymers (8). The results are a deficiency of circulating functional AAT, and retention of misfolded and polymerized AAT within hepatocytes associated with pro-inflammatory and pro-fibrotic gain-of-function (9). This combination predisposes Z variant homozygotes (denoted Pi^*ZZ) to severe and early-onset COPD and liver disease (hepatitis, cirrhosis, and hepatocellular carcinoma) (10). The lung disease seems predominantly driven by loss-of-function mechanisms though extracellular polymers encourage neutrophil chemotaxis and activity and are found in the circulation and lung (11).

AAT displays multiple immuno-regulatory functions including inhibition of neutrophil chemotaxis (12), induction of IL-1Ra by macrophages (13) and the development of tolerogenic dendritic cells and their production of IL-10 (14, 15). We have recently shown the requirement of AAT in the 1,25(OH)₂D3-mediated induction of IL-10 in CD4+ T cells (16). In addition to immunomodulatory effects mediated by antiprotease function, immunomodulatory effects of AAT that are independent of antiprotease function are now recognized (17). The absence of these effects may further exacerbate the pro-inflammatory state in lungs of individuals with AAT deficiency.

Cells within the lungs e.g., alveolar epithelial cells, monocytes and neutrophils are capable of secreting AAT to supplement liver-derived AAT in the airways (18–20). We have demonstrated that the active form of vitamin D, 1,25(OH)₂D3, upregulates expression of SERPINA1, the gene encoding the serine protease inhibitor AAT, and AAT protein secretion by human primary CD4+ T cells (16). Conversely this was not seen for monocytes. Moreover, vitamin D reduces secretion of matrix metalloproteinase (MMP-)9 that degrades AAT (21, 22). These data suggest that vitamin D availability may boost AAT levels within the airway microenvironment and so represent an additional mechanism by which vitamin D protects the airways.

Our published data demonstrate that 1,25(OH)₂D3 alone does not upregulate AAT in human CD8+ T cells (16). Given the proposed role of CD8+ T cells in chronic airway conditions such as COPD (23), we investigated whether a co-factor that acts in concert with 1,25(OH)₂D3 may induce AAT synthesis by human CD8+ T cells. TGF-β was investigated, since previous studies reported the capacity of hepatoma cell lines and human lung epithelial cell lines to produce AAT in response to TGF-β (24, 25). Furthermore, TGF-β and vitamin D are reported to act co-operatively. Independent studies report, for example, that TGF-β increases gene and protein expression for VDR (26) and the enzyme CYP27B1, which converts precursor vitamin D, 25(OH)D, to the active moiety, 1,25(OH)₂D3 (27, 28). Furthermore, 1,25(OH)₂D3 in combination with TGF-β increases the frequency of CD4+ Foxp3+ Treg cells (29). These data provide the rationale for the present study to investigate whether 1,25(OH)₂D3 and TGF-β interact with each other for enhanced effects in human CD8+ T cells.

A delicate balance between proteases and anti-proteases contributes to homeostasis in the lungs and several mediators, including vitamin D and AAT, contribute to this. This study highlights the following findings: (i) CD8+ T cells, in addition to CD4+ T cells, represent a cellular source for 1,25(OH)₂D3-induced AAT in the airway microenvironment, (ii) TGF-β acts as a co-factor in this vitamin D controlled regulatory axis and (iii) this axis is defective in PiZZ patients with severe AAT deficiency. Based on this and our previous findings (16) we propose that T cell-derived AAT is likely to function as an intermediate for immunomodulatory roles of vitamin D in the airway (1, 5). The significant correlation between circulating levels of vitamin D with airway AAT levels in a pediatric cohort further supports a functional relationship between these two mediators in vivo. Our in vitro studies indicate this will be mediated by 1,25(OH)₂D3 whilst implicating at least one further mediator upstream of the AAT response that we have not identified in our work to date. This work therefore provides an important starting point for further work to elucidate clear cascades of regulation and so better understand the relationship between vitamin D and AAT.

Hepatocytes are recognised as the major source of circulating AAT secreting 1–2 g/d into the circulation, however there is also good evidence for local synthesis, including in the airways (42, 43). The lower, but still significant levels of AAT secreted by T cells suggest that T cell derived AAT may act locally and have immunomodulatory functions, which have been reviewed elsewhere (44). In brief, AAT acts on structural and innate cells to increase anti-inflammatory mediators such as IL-1Ra, IL-10, to inhibit pro-inflammatory cytokine release, inhibit neutrophil chemotaxis and degranulation, and promote tolerogenic dendritic cells. In the adaptive arm, AAT increases the frequency of Foxp3 Treg cells and T cell synthesis of anti-inflammatory IL-10, whilst inhibiting pro-inflammatory mediators as well as B cell proliferation and autoantibody production. Many of these data arise from animal models where AAT is reported to control autoimmune disease, help to prevent transplant rejection and elastase-induced emphysema (45–48). Importantly, human intervention studies that build on these data are now emerging [reviewed by Song (49)]. Two recent examples report concomitant clinical and parallel immune readouts. AAT suppressed experimental graft vs. host disease (GVHD) by downmodulating inflammation, and AAT infusion was subsequently studied for treatment of steroid-resistant acute GVHD in humans. Magenau et al. reported a durable clinical response in refractory acute GVHD and that AAT administration appeared safe, with low rates of infection and with increasing ratios of T regulatory to effector T cells (50). Campos et al. (51) used double dose AAT infusion in AAT-deficient individuals to restore circulating AAT levels within the normal range, as well as to reduce serine protease activity and numerous inflammatory mediators in bronchoalveolar lavage. Additionally, findings from gene therapy studies demonstrate that low level secretion of AAT is sufficient to induce Treg responses in the local microenvironment, favouring toleration of the associated adeno-associated virus capsid (52).

Many of the immunoregulatory functions ascribed to AAT are also regulated by vitamin D. Observational studies and clinical trials indicate that vitamin D incrementally improves airway health in established chronic respiratory disease, especially in individuals who are profoundly vitamin D deficient (2, 53, 54). The capacity of vitamin D to induce antimicrobial functions, to enhance clinical and immune responses to corticosteroids, to promote epithelial barrier integrity as well as to augment multiple pathways linked to peripheral tolerance are all proposed to underpin these clinical effects (1, 5, 55). It seems probable that vitamin D and AAT demonstrate both common and independent immunoregulatory activities that contribute to airway health, however the interplay between these mediators requires further investigation.

We propose a potential role for vitamin D induced AAT synthesis by human T cells which is discussed in the context of local immune regulation in the airways. Circulating levels of 25-hydroxyvitamin D, the accepted measure of vitamin D status, correlated with AAT levels in the airways in a pediatric asthma cohort and matched non-asthmatic controls provides some support for this concept. However, it is highly unlikely that T cell-derived AAT alone accounts for this correlation. In our earlier publication we found no evidence that 1,25(OH)₂D3 increased SERPINA1 mRNA and/or AAT protein levels in monocytes, respiratory epithelial cells or primary hepatocytes (16). However, the capacity of vitamin D to control other pathways that regulate protease:antiprotease balance may also be relevant. This could include the previously reported downregulation by vitamin D of MMP-9 (21) which is known to inactivate AAT (56), that may further enhance the correlation between vitamin D and AAT levels in the lung.

The ChIP-Atlas database suggests there are no VDR binding sites within SERPINA1 implying that indirect and/or non-binding effects of vitamin D may explain the association reported here. The current study does not define the mechanisms by which this may occur. However, links between circulating measures of vitamin D status and AAT levels reported here, in a study of type 2 diabetes patients (40), and suggested by a study in COVID-19 patients (57) support such an association.

Vitamin D has been shown to upregulate TGF-β levels both here and in independent studies, including in T cells and bronchial epithelial cells (32, 58). On the other hand, TGF-β has been shown to enhance VDR, with data showing that TGF-β1 increases VDR expression via SMAD3/4 in human granulosa-lutein cells (26). We speculate that the capacity of TGF-β to both independently increase AAT synthesis in hepatoma and bronchial epithelial cell lines (24, 25) as well as its capacity to enhance the response to vitamin D through increased expression of VDR and RXRα, as described here in T cells, might also contribute to the observed association between airway AAT and circulating vitamin D. Similar effects of TGF-β on VDR gene and protein expression have been reported elsewhere (26). These studies support co-operation between TGF-β1 and vitamin D to drive target gene expression. The interaction of vitamin D and TGF-β to induce an immunomodulatory response via increased local levels of AAT may contribute to the anti-inflammatory repertoire within the pleiotropic potential of TGF-β. Whether TGF-β acts to enhance the response to vitamin D by other cell types and/or other functions of interest remains to be investigated, but the latter at least is suggested by other studies in human T cells (29). As far as we are aware, whether vitamin D supplementation regulates local AAT levels in healthy or patient populations is still unknown but these data are awaited with interest.

PiZZ patients' hepatocytes synthesise normal levels of AAT mRNA and nascent polypeptide, but are unable to secrete signficant amounts of AAT into the circulation (15% of the normal circulating levels) and reduced levels are also evident in the airways (59). We previously demonstrated a loss of correlation of SERPINA1 mRNA levels with an immunomodulatory readout, IL-10 gene expression, in PiZZ patient CD4+ T cells compared with those from controls (16). We attributed this to defective secretion of Z variant AAT protein from these cells. In hepatocytes the sequestration of Z AAT into retained polymeric species is related to the high levels of translation of the aggregation-prone variant. Our data presented here, despite the limitation of comparatively low patient numbers, confirm significantly impaired secretion of AAT is also observed in PiZZ T cells, that generally express far lower levels of AAT protein than the hepatocytes, relative to those from healthy donors. This could not be overcome by vitamin D in the presence or absence of TGF-β.

AAT deficiency such as that associated with the PiZZ mutation, is linked to an inherited increase in the risk of COPD and severe emphysema. This finding led to the paradigm that an imbalance between antiproteases and neutrophil-derived proteases contributes to the pathogenesis of emphysema (10). However, aberrant immune activation in AAT deficiency linked to lung disease also exists, with increasing support for a parallel role of anti-inflammatory properties of AAT that are lost in AAT deficiency. For example Baraldo et al. describe adaptive immune activation in T and B lymphocytes together with a marked increase in lymphoid follicles as a prominent feature in AAT deficiency (60). Similarly AAT-deficient individuals demonstrate increased circulating levels of IL-17A (61) and increased pro-inflammatory mediator profiles in the airways (44, 51).

Bronchial epithelial cells line the airways and are amongst the first cells to be activated by exogenous stimuli such as microbes, particulate matter air pollution, allergens, and other inflammatory stimuli. They are thought to play a pivotal role in controlling immune responses in the airways through actions on dendritic cells and other cells (62). Taken together our data support the responsiveness of lung T cell and airway epithelial cells to 1,25(OH)₂D3 as drivers of the correlation of systemic 25(OH)D3 with lung AAT levels observed in vivo. HBECs express CYP27B1 and therefore are able to convert precursor vitamin D to the bioactive 1,25(OH)₂D3 (38). Our finding in HBECs further supports this by demonstrating increased bioactive TGF-β synthesis. We therefore propose a model of T cell-epithelial cell interactions in which bronchial epithelial cells may generate bioactive 1,25(OH)₂D3 and TGF-β the latter potentially supplemented by bioactive TGF-β production in CD4+ T cells, to promote anti-inflammatory functions in T cells (Figure 6). We previously proposed a T cell-epithelial cell interaction whereby 1,25(OH)₂D3 enhances the production of soluble ST2, the decoy receptor that antagonizes IL-33 signaling in HBECs (33). These findings together suggest a physiological supportive network between epithelial and T cells of potential benefit in the control of chronic airway disease.

For simplicity the model proposed here highlights HBECs as a likely source of TGF-β. However, a wide range of cells in the airways may also contribute including fibroblasts, smooth muscle cells and inflammatory cells. Defining the most relevant celluar source of TGF-β, and this may not reflect a single cellular source, would likely require a conditional animal knockout model to address this. The precise cellular sources of TGF-β in the microenvironment are not central to the model. TGF-β (in all isoforms) is synthesized as an inactive precursor (a latency-associated protein bound to bioactive TGF-β) that is secreted into the matrix in complex with latent TGF-β binding protein. Activation e.g., mediated by integrins, stretch, and/or proteolysis is then required for the generation of the bioactive moiety (41).

Several limitations of our findings raise questions for future research. For example, whilst we demonstrate the capacity of vitamin D to increase AAT in human T cells, the lack of an obvious vitamin D response element within SERPINA1 suggests that intermediates in this pathway and the regulation by which this effect occurs remain to be clarified. It will also be useful to define the molecular pathways involved in local generation of the active vitamin D metabolite, 1,25(OH)₂D3, and TGF-β required for AAT induction. Whilst a correlation between circulating 25-hydroxyvitamin D and airway AAT levels in a pediatric cohort exists, it seems unlikely that this is due solely to vitamin D induction of T cell-derived AAT and the contribution of other vitamin D regulated pathways e.g., downregulation of MMP-9, remain to be fully defined. Furthermore, these data would be complemented by evidence that vitamin D supplementation of vitamin D deficient individuals increases AAT levels in vivo. Finally, defining complementary, distinct and/or overlapping anti-inflammatory effects of vitamin D and AAT in patients will facilitate mechanistic and future clinical studies.

In summary, our study demonstrates that vitamin D increases AAT synthesis in human T cells, via a TGF-dependent mechanism, a pathway that is impaired in individuals with a genetic mutation (PiZZ) that results in decreased levels of AAT secretion. We propose that signals from adjacent airway cells, support vitamin D-mediated induction of AAT in situ in human lung. Some immunoregulatory properties of AAT and vitamin D overlap such as the induction of IL-10 and increase in Foxp3-Treg frequency, leading us to propose that the role of T cell-derived AAT is most likely to be immunomodulatory and may represent an intermediate of some immunodulatory functions of vitamin D. Notably, both oral vitamin D supplementation and AAT infusion are under investigation for their capacity to inhibit inflammation in a range of human immune-mediated pathologies. For AAT these data are still limited and only now emerging (49), whilst for vitamin D these data are complicated by enormous variability in part related to variable study designs (1, 4). Nevertheless, both appear to demonstrate good safety profiles and are well-tolerated, although differences in the cost of a common oral vitamin D supplement available over the counter vs. GMP-grade manufacture of AAT, are likely to be large. Comparison of biological and immune readouts of ongoing trials may provide insight into the most pertinent, overlapping and distinct effects of the two mediators. Sole AAT augmentation therapy is currently being studied, but the potential therapeutic implication of using vitamin D supplementation as an adjuvant/combination therapy to overcome any limitations faced by sole AAT therapy may be of future interest and precedent exists for the use of vitamin D as an adjunct treatment (31, 53).

The original contributions presented in the study are included in the article/supplementary material, further inquiries can be directed to the corresponding authors.

The studies involving human participants were reviewed and approved by Healthy donors (Human peripheral blood cell laboratory studies to investigate the role of immune and inflammatory pathways in respiratory disease by the local research ethics committee, REC reference: 14/LO/1699) were recruited and provided full written informed consent with full Research Ethics Committee approval. COPD patients without AAT deficiency (Chest clinic, Guy's Hospital, 14/LO/1699), and PiZZ-AAT deficient patients with or without COPD (Targeting dysfunctional mechanisms in alpha-1 antitrypsin deficiency, The Royal Free Alpha-1 clinic, REC reference: 13/LO/1085) were recruited and provided full written informed consent with full Research Ethics Committee approval. The study in a pediatric asthma cohort and control children undergoing diagnostic bronchoscopy for non-asthma related purposes was approved by the Royal Brompton and Harefield Research Ethics Committee (09/H07008/48) which has been previously described (14). Informed consent was obtained from parents and age-appropriate assent from children. Serum levels of 25-hydroxyvitamin D were measured as previously described (14). Written informed consent to participate in this study was provided by the participants' legal guardian/next of kin.

CH, BG, and Y-HC designed the study. BG recruited and screened participants. Y-HC performed functional experiments, supported by LR, SD, PP, and CC and supervised by CH and BG. SD, AG, and AB were responsible for studies in the pediatric cohort. CC, Y-HC, and CH wrote the manuscript, which all authors revised and approved. All authors contributed to the article and approved the submitted version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.