Vitamin D deficiency (VitD-) is commonly encountered in patients with rheumatoid arthritis (RA) (and other autoimmune diseases). Some studies show VitD levels impact expression of RA, but the mechanism(s) have not been well explored and have received scant attention in rodent models of RA. Interestingly, like VitD levels, in studies looking at only plasma gelsolin (pGSN), it was observed that pGSN levels also inversely correlate with C-reactive protein (CRP) in patients with RA. Gelsolin (GSN) has been shown to be protective in the TNF? transgenic (tg) mouse model of RA. It is known that sphingosine 1-phosphate (S1P) (acting through G protein- coupled receptors on T cells, B cells, monocytes/macrophages, and dendritic cells) affect immune and inflammatory responses and has been found to be elevated in RA synovial fluid but not osteoarthritis synovial fluid. Furthermore, pGSN binds to S1P, thereby regulating its ability to engage S1P receptors. We found humans, in addition to converting vitamin D3 (VitD3) to 25(OH)D3, also generate a noncalcemic analog, 20(OH)D3, with a serum concentration 1/20th of that of 25(OH)D3. In the type II collagen-induced arthritis (CIA), we found surprisingly that 20(OH)D3 treatment elevates pGSN levels and increases S1P levels in splenocyte cultures from these 20(OH)D3-treated mice. We discovered that supplementing VitD in humans elevates pGSN in sera and that 25(OH)D levels in normals and in RA and OA correlate fairly well with pGSN. Also, we found both 20(OH)D3 and 1,25(OH)2D3 increased sphingosine kinase (Sphk)2 mRNA in mouse splenocytes and increased GSN mRNA in cultured mouse myoblasts and pGSN production by explant cultures of mouse skeletal muscle. Knocking down GSN or Sphk2 worsens arthritis in mouse models of RA. Both GSN and Sphk2 regulate S1P levels in plasma or lymphoid tissue, which is important in immune function, including lymphocyte trafficking, Th1/Th2/Th17 cytokine expression, and control of FoxP3 regulatory T cells (Tregs). No studies have evaluated levels of 25(OH)D, S1P, and pGSN in the same RA population. We find 25(OH)D serum levels correlate fairly well with pGSN levels in RA patients, although concordance is not 100%. We further hypothesize that RA patients who have both high 25(OH)D and pGSN will have less RA disease activity score, while those with low pGSN and low 25(OH)D will have higher RA disease activity scores. Our overarching hypothesis is that there are interactions amongst VitD, GSN, and CD4+ T cell Sphk2, CD4+ T cell S1P receptor subtypes, and/or S1P that may be important in the modulation of autoimmunity and inflammatory arthritis. Whether administering VitD to humans with VitD deficiency/insufficiency will raise pGSN levels and modulate S1P, is unknown, and is the topic of Specific Aim 3. The three highly translational Specific Aims of this proposal are the following:
Specific Aim 1 A: Assess the association of serum levels of 25(OH)D, pGSN, and S1P separately, in patients with RA with measures of RA disease activity.
Specific Aim 1 B: Assess whether a clinical prediction model or models containing serum 25(OH)D, pGSN, and/or S1P levels provide more reliable prediction of the selected markers or measurements of RA disease activity than any single predictor alone.
Specific Aim 2 : To test the hypothesis that VitD replacement therapy in humans with OA or RA with VitD- will elevate levels of pGSN, reduce levels of S1P in plasma, and change lymphocyte expression of Sphk2.
Specific Aim 3 : To test the hypothesis that high dose VitD replacement therapy in patients with OA or RA and VitD- will reduce Th1, Th9, and Th17 effector T cells, increase Th2 and CD4+Tregs in peripheral blood, and whether there is interaction with existing plasma levels of pGSN and S1P. We also found that serum 25(OH)D levels correlate with pGSN in patients with RA and (in a small pilot study) that VitD supplementation in humans [healthy volunteers (HV) and patients with RA] raised 25(OH)D levels from insufficiency to normal range and was also associated with increases in pGSN. This suggests that (in both mice and humans) pGSN is partially regulated by VitD.

Public Health Relevance

About 40% of Veterans are deficient in vitamin D (VitD), with increased risk for poor outcomes of rheumatoid arthritis (RA), cancer, heart disease, stroke, lung diseases, liver disease, Alzheimer's disease, and other diseases. We found VitD compounds suppress arthritis in mice with RA-like arthritis and affect metabolism of two other molecules [plasma-gelsolin (pGSN) and sphingosine 1-phosphate (S1P)] that may help suppress arthritis. VitD status [level of 25(OH)D] does not always correlate with arthritic activity in RA. We propose to determine whether some combination of serum levels of 25(OH)D, pGSN and/or S1P will better reflect RA disease activity and be more precise in assessing activity of RA. If we show high dose VitD treatment raises levels of pGSN, this could be a means to elevate pGSN levels-which may improve survival in Veterans with acute trauma, major surgery, chronic renal dialysis, burns, pneumonia, Lyme encephalitis, malaria, and febrile seizures-conditions in which low pGSN is associated with increased mortality.

National Institute of Health (NIH)
Veterans Affairs (VA)
Non-HHS Research Projects (I01)
Project #
Application #
Study Section
Immunology A (IMMA)
Project Start
Project End
Budget Start
Budget End
Support Year
Fiscal Year
Total Cost
Indirect Cost
Memphis VA Medical Center
United States
Zip Code
Slominski, Andrzej T; Kim, Tae-Kang; Hobrath, Judith V et al. (2017) Characterization of a new pathway that activates lumisterol in vivo to biologically active hydroxylumisterols. Sci Rep 7:11434