Study at a glance
Can a device containing liver and kidney cells model vitamin D activation?
Researchers built a small laboratory device containing human liver and kidney cells in connected chambers. Liquid containing vitamin D3 flowed through the device for 24 hours. This allowed them to model the stages in which the liver and kidneys change vitamin D3 into other forms. They tested the liquid leaving the device for vitamin D compounds and for effects on a separate set of cells.
Which form of vitamin D was detected in the device?
Chemical testing identified 25-hydroxyvitamin D3, a form produced along the way as vitamin D becomes active. This directly showed that the device could carry out that conversion. Cells in the flowing system also showed higher activity of genes for vitamin D-processing enzymes than cells grown without that flow.
How did liquid from the device affect other laboratory cells?
Liquid from the device containing liver and kidney cells produced changes linked to laboratory-grown leukaemia cells becoming more mature. Most measured responses were stronger when the kidney-cell chamber was included than in a liver-only arrangement. These results were consistent with further vitamin D processing. The leukaemia cells were used to detect biological activity; this was not a cancer-treatment trial.
Did the researchers detect the final active form of vitamin D?
The final active form, 1,25-dihydroxyvitamin D3, was not detected in the collected liquid. Its amount may have been too low to detect, but that possibility does not prove it was present. The cell responses did not identify every substance responsible. The study therefore did not chemically confirm the complete activation sequence, establish how quickly it occurs in people, or demonstrate health benefits.
Read the full article below ↓This evidence review examines a primary in-vitro organ-on-chip study that reconstructed sequential hepatic and renal vitamin D3 metabolism using interconnected human cell models. HepG2 cells were used to represent the hepatic stage and renal proximal tubule epithelial cells (RPTEC) to represent the renal stage. The study directly demonstrated conversion of vitamin D3 to 25-hydroxyvitamin D3 [25(OH)D3] in the microfluidic system and investigated whether passage through the combined liver: kidney model produced biological activity consistent with further activation. The central evidential boundary is important: 25(OH)D3 was detected directly by LC-MS/MS, whereas 1,25-dihydroxyvitamin D3 [1,25(OH)2D3] was not detected analytically in the chip eluate. Evidence for the final activation step was therefore functional and indirect rather than direct chemical detection.
Study classification and identity
Study type: Primary in-vitro experimental organ-on-chip study.
Population or experimental system: A multi-compartment microfluidic model containing human HepG2 liver cells and human renal proximal tubule epithelial cells (RPTEC), with HL-60 cells subsequently used as a functional response system for the collected chip eluate.
Parent study or cohort: Not applicable.
Ancillary analysis: Not applicable.
Exposure or intervention: Vitamin D3-containing medium passed through microfluidic cell-culture chambers designed to model sequential hepatic and renal metabolism.
Comparison or grouping structure: The investigators compared microfluidic and static culture conditions, vitamin D3-treated and control chip eluates, liver-only/liver: liver configurations with liver: kidney configurations, and responses to reference vitamin D3 metabolites.
Primary endpoint category: Formation of vitamin D3 metabolites and biological activity of metabolites or other factors present in chip eluate.
Study purpose and research question
The researchers asked whether a multi-compartment microfluidic system could reproduce important stages of human vitamin D3 metabolism outside the body. Their model was designed around the established physiological sequence in which vitamin D3 undergoes hydroxylation associated with hepatic metabolism to form 25(OH)D3, followed by renal bioactivation associated with conversion of 25(OH)D3 towards 1,25(OH)2D3.
For this evidence review, the relevant question is deliberately narrower than the complete purpose of the paper: what does the experiment directly demonstrate about sequential hepatic and renal activation of vitamin D3?
The study is therefore a valid partial fit for this evidence slot. It experimentally models liver-to-kidney metabolic sequencing and directly verifies formation of 25(OH)D3, but it does not directly chemically verify every molecular step of the complete activation sequence. In particular, 1,25(OH)2D3 was not detected in the chip eluate by LC-MS/MS. The evidence supporting further activation comes instead from biological responses to the eluate and from comparisons involving the renal compartment.
Study overview
The investigators used interconnected microfluidic chambers to create a simplified liver: kidney model. HepG2 cells occupied the hepatic component and RPTEC cells the renal component. The chambers were connected by microfluidic flow so that material leaving one cellular environment could pass into the next.
This arrangement matters because vitamin D activation is not a single chemical event. The experiment was constructed to model sequential metabolism: vitamin D3 enters the hepatic component, metabolites produced there pass downstream, and the renal component can then act upon material arriving from the hepatic stage.
The researchers also examined whether microfluidic cultivation altered cellular characteristics relevant to vitamin D metabolism. Expression of the vitamin D-metabolising enzymes CYP2R1, CYP27B1 and CYP24A1 was higher under on-chip cultivation than under the corresponding static culture conditions.
Vitamin D3-containing medium was introduced into the chip, and the resulting eluate was examined both chemically and functionally. Chemical analysis was used to identify vitamin D metabolites. Separately, collected eluate was applied to HL-60 acute myeloid leukaemia cells, where expression of differentiation markers was assessed as a functional readout of biological activity.
Study design and measurement approach
The two-compartment microfluidic platform physically separated the hepatic and renal cell models while connecting them through controlled fluid flow. This allowed the investigators to examine consecutive cellular processing rather than simply mixing the two cell types in one conventional culture.
Vitamin D3 was supplied to the system at a final concentration of 20 µM and passed through the chip at 20 µL per hour for 24 hours. Eluted medium was then collected for downstream analysis.
The researchers used LC-MS/MS to investigate the presence of vitamin D3 and selected metabolites, including 25(OH)D3 and 1,25(OH)2D3. Identification of 25(OH)D3 was supported by comparison of the chromatographic and mass-spectrometric signal with reference standards.
The biological component of the experiment used HL-60 cells treated with chip eluate for 24 hours. Differentiation-related markers including CD11b, CD14, osteopontin and parvalbumin were examined by RT-qPCR, while CD11b and CD14 were also assessed at protein level using flow cytometry.
The study therefore contains two importantly different kinds of evidence: direct analytical evidence identifying a metabolite in the eluate and functional evidence showing that the eluate produced biological responses. These evidence types must not be treated as interchangeable.
For the enzyme framework underlying these metabolic steps, see Vitamin D Enzyme Pathways (CYP27B1 / CYP24A1).
What the study directly measured
The investigators directly measured expression of vitamin D-metabolising CYP enzymes in the cellular models, vitamin D metabolites recovered from chip eluate by LC-MS/MS, and differentiation-marker responses in HL-60 cells exposed to that eluate.
LC-MS/MS directly detected 25(OH)D3 in eluate from the microfluidic systems. The mass spectra of the detected peaks were compared with reference substances and supported identification of the metabolite as 25(OH)D3.
The reported concentration of 25(OH)D3 was approximately 1.48 µM in eluate from the liver: liver chip and 1.26 µM in eluate from the liver: liver: kidney: kidney configuration used for the analytical comparison. This provided direct evidence that the supplied vitamin D3 had undergone metabolic 25-hydroxylation within the experimental system.
The study also directly measured the response of HL-60 cells to collected eluate. Vitamin D3-containing liver: kidney chip eluate increased expression of the investigated differentiation markers compared with control conditions. The researchers additionally compared liver-only and liver: kidney configurations and found stronger induction for most of the measured markers when the renal cell component was present.
What the study did not directly measure
The most important negative analytical finding is that 1,25(OH)2D3 was not detected in the chip eluate by LC-MS/MS. The authors considered possible explanations including concentrations below the analytical detection limit, low extraction yield, or the relatively short half-life of the metabolite.
Accordingly, the study did not directly chemically demonstrate recovered 1,25(OH)2D3 in the eluate. Similarity between the biological activity of the eluate and responses associated with activated vitamin D metabolites cannot be converted into direct chemical detection.
The experiment also did not measure hepatic and renal vitamin D metabolism in intact human participants. It did not quantify circulating vitamin D metabolite dynamics, endocrine regulation by calcium, phosphate, parathyroid hormone or FGF23, or clinical outcomes resulting from vitamin D activation.
What cannot be inferred from this study
This experiment cannot establish that the concentrations, rates or relative contributions observed in the chip reproduce those occurring in an intact human liver and kidney. HepG2 and RPTEC cultures are experimental models, and the microfluidic environment is a reconstruction of selected physiological relationships rather than an intact endocrine system.
The study also cannot establish from its LC-MS/MS results that the renal chamber definitely produced a measurable quantity of 1,25(OH)2D3. The functional findings are consistent with further metabolic activation, but they do not identify the complete molecular composition responsible for the biological response.
Nor can the HL-60 differentiation findings be interpreted as evidence that vitamin D3 treatment produces anti-cancer effects in people. HL-60 cells served here as an experimental functional readout. This study was not a cancer-treatment trial and measured no clinical cancer outcomes.
Statistical and analytical approach
The study combined chemical analysis with gene-expression and protein-expression comparisons. Vitamin D metabolites in chip eluate were investigated using liquid chromatography coupled to tandem mass spectrometry, with standards used to establish chromatographic separation and relevant fragmentation patterns.
For the cellular response experiments, RT-qPCR was used to assess differentiation-marker mRNA and flow cytometry was used for selected cell-surface proteins. Statistical comparisons described in the study used an unpaired two-tailed Student’s t-test, with significance thresholds reported at p ≤ 0.05, p ≤ 0.01 and p ≤ 0.001.
Some experimental components were based on small numbers of biological replicates, while other reported comparisons incorporated more than five independent experiments. The strength of any particular finding therefore depends on the relevant experimental component rather than on a single uniform sample structure across the paper.
Results
The clearest chemically verified result was conversion of supplied vitamin D3 to 25(OH)D3. LC-MS/MS identified 25(OH)D3 in microfluidic eluate, providing direct evidence that the experimental system could perform an important metabolic step associated with hepatic vitamin D activation.
On-chip cultivation was also associated with increased expression of CYP2R1, CYP27B1 and CYP24A1 compared with static cultivation. These findings showed that the microfluidic environment altered expression of enzymes relevant to vitamin D metabolism and supported the use of the platform as a metabolically active experimental model.
Functionally, eluate collected from vitamin D3-fed liver: kidney chips induced HL-60 differentiation markers more strongly than control eluate and more strongly than vitamin D3 itself under the comparison conditions. Protein-level analysis of CD11b and CD14 also showed greater expression following treatment with vitamin D3-containing chip eluate than in the reported control comparisons.
When the researchers compared a hepatic configuration with a liver: kidney configuration, most of the investigated differentiation markers were more strongly induced by eluate from the liver: kidney system. This supports a functional contribution from consecutive metabolism involving the renal cell compartment.
However, the analytical and functional results did not completely converge. Although 25(OH)D3 was directly detected, 1,25(OH)2D3 was not. The biological activity of the eluate was consistent with the presence of activated vitamin D metabolism, but the experiment could not chemically attribute that entire response to detected 1,25(OH)2D3.
The authors investigated one possible explanation for the enhanced activity of the chip eluate: increased albumin might have prolonged metabolite availability. Adding bovine serum albumin to 1,25(OH)2D3 treatment did not reproduce a significant change in the measured target-gene response. The authors therefore considered other explanations, including additional vitamin D metabolites or factors secreted by cells within the microfluidic system.
Measurement considerations and study limitations
The most consequential measurement limitation is the failure to detect 1,25(OH)2D3 directly in the chip eluate. This prevents the functional response from being treated as direct analytical proof of the final renal activation product.
The authors specifically recognised that failure of detection did not necessarily demonstrate absence. Low concentration, extraction efficiency and analytical detection limits could all contribute. But the opposite is equally important: a plausible reason for non-detection is not evidence that the compound was present. The chemically verified and functionally inferred components must therefore remain separate.
The experimental models impose additional limitations. HepG2 cells are a hepatic cell-line model and RPTEC cells represent renal proximal tubular epithelium; neither reproduces the full structural, endocrine and metabolic environment of an intact human organ. Microfluidic flow may improve selected cellular functions relative to static culture while still remaining an engineered approximation.
The HL-60 assay is likewise a functional experimental system rather than a direct measurement of vitamin D activation in a person. Its differentiation response indicates biological activity in the collected eluate but does not uniquely identify the molecule or molecules responsible for that activity.
Generalisability and population context
The study is best generalised as proof that selected sequential features of vitamin D3 metabolism can be reconstructed in a connected human-cell microfluidic model. It demonstrates metabolic formation of 25(OH)D3 and provides functional evidence consistent with additional processing when hepatic and renal cell compartments operate consecutively.
It should not be generalised into quantitative claims about normal human vitamin D metabolism. The concentrations used, engineered flow conditions, cell types and experimental endpoints differ fundamentally from the integrated physiology of an intact person.
The experiment therefore contributes mechanistic model evidence rather than population evidence. It does not establish how activation differs by age, kidney function, liver function, vitamin D status, diet, genetics, disease state or other sources of human variability.
Mechanistic inference boundary
The study directly supports a hepatic-like metabolic step because 25(OH)D3 was chemically detected after vitamin D3 passed through the cellular microfluidic system. It also shows that adding a renal cellular stage altered downstream biological activity in a manner consistent with consecutive metabolism.
The final conversion to 1,25(OH)2D3 occupies a different evidential category. CYP27B1 expression and the biological response of the eluate are compatible with renal bioactivation, but 1,25(OH)2D3 itself was not detected in the eluate by LC-MS/MS. The paper therefore provides indirect functional support for complete activation rather than direct analytical verification of the final metabolite.
This distinction prevents a biologically plausible sequence from being reported as though every stage had been chemically observed. The study is valuable precisely because it combines direct and functional evidence, but those evidence types retain different strengths.
For the wider cellular metabolism framework, see Vitamin D Metabolism at the Cellular Level.
Interpretation within vitamin D physiology
The study provides an experimental illustration of why vitamin D activation is best understood as a sequence rather than as a property of vitamin D3 alone. Vitamin D3 entered a connected cellular system, 25(OH)D3 emerged as a directly detected metabolic product, and the downstream liver: kidney eluate produced biological activity that differed from the parent vitamin under the experimental conditions.
That makes the paper particularly relevant to hepatic and renal activation because the architecture of the experiment deliberately separates and reconnects cellular models representing those two metabolic territories. The model allows consecutive processing to be examined in a way that static single-cell cultures cannot reproduce as directly.
The study nevertheless remains a model of selected components of the pathway. It does not reproduce the complete endocrine regulation of vitamin D activation in vivo, and it does not directly demonstrate every intermediate or final product implied by the physiological sequence.
For the parent physiology framework, see Vitamin D Activation.
For the broader question of how metabolism differs between tissues, see Tissue-Specific Vitamin D Metabolism.
Evidence boundaries and final interpretation
The strongest defensible interpretation is that Theobald and colleagues reconstructed important features of sequential vitamin D3 metabolism in a multi-compartment human-cell microfluidic system. The study directly demonstrated formation of 25(OH)D3 and showed biological effects from liver: kidney chip eluate consistent with additional metabolic activation.
The principal boundary is equally important: the experiment did not directly detect 1,25(OH)2D3 in that eluate. It therefore cannot be described as direct chemical demonstration of the complete vitamin D3-to-25(OH)D3-to-1,25(OH)2D3 sequence. The final stage is supported indirectly through functional evidence and the behaviour of the connected experimental system.
This is a valid partial fit for the hepatic-and-renal-activation evidence slot because the paper experimentally addresses consecutive liver-like and kidney-like metabolism. Its legitimate contribution is stronger than a purely conceptual description of the pathway but narrower than direct in-vivo proof of the complete human activation sequence.
For a concise explanation of this study, see Hepatic and Renal Activation of Vitamin D3: Study Summary.
Primary source and publication details
Theobald J, Abu el Maaty MA, Kusterer N, Wetterauer B, Wink M, et al.
In vitro metabolic activation of vitamin D3 by using a multi-compartment microfluidic liver-kidney organ on chip platform.
Scientific Reports. 2019;9:4616.
PMID: 30874583
PMCID: PMC6420623
DOI: 10.1038/s41598-019-40851-9
View the verified primary source
Related reading
Clinical characteristics and risk factors for mortality in COVID-19 inpatients in Birjand, Iran: a single-center retrospective study
This study aimed to evaluate the clinical characteristics in COVID-19 inpatients including (survivor and non-survivor); thus helping clinicians give the right treatment and assess prognosis and guide the treatment.
Original paper ↗Read more →Determinants of severity among hospitalised COVID-19 patients: Hospital-based case-control study, India, 2020
Risk factors for the development of severe COVID-19 disease and death have been widely reported across several studies.
Original paper ↗Read more →Effects of Decreased Vitamin D and Accumulated Uremic Toxin on Human CYP3A4 Activity in Patients with End-Stage Renal Disease
The purpose of this study is to elucidate the reason for the decrease in hepatic clearance in patients with end-stage renal disease.
Original paper ↗Read more →Genomic actions of 1,25-dihydroxyvitamin D3 on insulin receptor gene expression, insulin receptor number and insulin activity in the kidney, liver and adipose tissue of streptozotocin-induced diabetic rats
The candidate VDREs identified may respond to 1,25D3 via activation of the vitamin D receptor, although this remains to be investigated.
Original paper ↗Read more →High ferritin is associated with liver and bone marrow iron accumulation: Effects of 1-year deferoxamine treatment in hemodialysis-associated iron overload
Iron (Fe) supplementation is a critical component of anemia therapy for patients with chronic kidney disease (CKD).
Original paper ↗Read more →In vitro metabolic activation of vitamin D3 by using a multi-compartment microfluidic liver-kidney organ on chip platform
Organ-on-chip platforms provide models that allow the representation of human physiological processes in cell-based miniaturized systems.
Original paper ↗Read more →Integrated multi-omics analysis reveals that SMARCAL1 serves as an immune related prognostic marker and promotes liver cancer growth
In our study, we examined the expression patterns, clinical features, and survival analysis from The Cancer Genome Atlas and GEO datasets.
Original paper ↗Read more →Interaction Between Liver Metabolism and Gut Short-Chain Fatty Acids via Liver-Gut Axis Affects Body Weight in Lambs
Thus, we examined the jejunal microbiota, fermentation parameters, digestive enzymes, morphology, and liver metabolic profiles in different growth development lambs to investigate the liver: gut axis’s role in their development.
Original paper ↗Read more →Long-term renal outcome in methylmalonic acidemia in adolescents and adults
Our objective was to precisely address the clinical and biological characteristics of long-term CKD in MMA adolescent and adult patients.
Original paper ↗Read more →Metabolic dysfunction-associated steatotic liver disease and bone mineral density in type 2 diabetes mellitus: insights from a single-center cross-sectional study in North India
This study aimed to evaluate the association between MASLD and BMD at different skeletal sites among adults with type 2 diabetes mellitus in North India.
Original paper ↗Read more →MiR-125b promotes migration and invasion by targeting the vitamin D receptor in renal cell carcinoma
To investigate the expression of miR-125b and vitamin D receptor (VDR) in renal cell carcinoma (RCC) and assess the biological function of miR-125b in RCC.
Original paper ↗Read more →Potential Adverse Effects of Urinary Bisphenol A Concentrations on Renal Function, Gestation, and Newborn Characteristics in Healthy Pregnant Women
BPA was detected in 78% of urine samples, compared with 10% for BPF and 5% for BPS, making it the predominant bisphenol.
Original paper ↗Read more →Regulation of HHLA2 expression in kidney cancer and myeloid cells
HHLA2 expression in A498 and 786-O ccRCC cell lines was examined in vitro and in subcutaneous tumor xenografts in NSG mice.
Original paper ↗Read more →Risk factors of primary and recurrent fractures in postmenopausal osteoporotic Chinese patients: A retrospective analysis study
Compared to the unfractured group, the fractured group was older and had lower height and bone mineral density (all P.
Original paper ↗Read more →Serum clinical laboratory tests and risk of incident dementia: a prospective cohort study of 407,190 individuals
Among 407,190 participants from the UK Biobank (median follow-up of 9.19 years), we investigated the linear and nonlinear effects of 30 laboratory measures on the risk of all-cause dementia using Cox models and restricted cubic spline models.
Original paper ↗Read more →Teriparatide treatment of osteoporosis in solid organ transplant recipients: a single-center experience
Bone mineral density (BMD), incident fractures, biochemical parameters, renal function, and adverse events were evaluated during treatment and follow-up.
Original paper ↗Read more →The beneficial effects of simultaneous supplementation of Lactobacillus reuteri and calcium fluoride nanoparticles on ovariectomy-induced osteoporosis
The present study aimed to investigate the simultaneous effects of calcium fluoride nanoparticles (CaF2 NPs) and lactobacillus reuteri ATCC PTA 6475 (L.
Original paper ↗Read more →The effect of fructose exposure on amino acid metabolism among Chinese community residents and its possible multi-omics mechanisms
The consumption of fructose has increased dramaticly during the last few decades, inducing a great increase in the risk of intrahepatic lipid accumulation, hypertriglyceridemia, hyperuricemia and cancer.
Original paper ↗Read more →The Vitamin D status is associated with serum C-reactive protein and adhesion molecules in patients with renal cell carcinoma
This study investigated the association of vitamin D status with serum C-reactive protein (CRP) and adhesion molecules among RCC patients.
Original paper ↗Read more →Vitamin D receptor suppresses proliferation and metastasis in renal cell carcinoma cell lines via regulating the expression of the epithelial Ca2+ channel TRPV5
The aim of this study was to elucidate whether VDR could regulate the expression of TRPV5 and affect proliferation and metastasis in RCC.
Original paper ↗Read more →Vitamin D Status and Potential Therapeutic Options in Critically Ill Patients: A Narrative Review of the Clinical Evidence
Several researchers have investigated the impact of vitamin D supplementation showing its feasibility, safety, and effectiveness, although conflicting results have put into question its real benefit in critically ill patients.
Original paper ↗Read more →Original sources
In vitro metabolic activation of vitamin D3 by using a multi-compartment microfluidic liver-kidney organ on chip platform
Organ-on-chip platforms provide models that allow the representation of human physiological processes in cell-based miniaturized systems.
Publication details
View the verified primary source