Spatial detection and consequences of nonrenal calcitriol production as assessed by targeted mass spectrometry imaging

The immune benefits of vitamin D3 supplementation beyond calcium and phosphate maintenance are highly clinically debated. Kidney expression of CYP27B1 is the source of endocrine, circulating 1,25(OH)2D3 (active form of vitamin D) that maintains serum calcium and phosphate. 1,25(OH)2D3 may also be made by the CYP27B1 enzyme in nonrenal cells, like immune cells, in a process driven by cellular availability of 25(OH)D3 and inflammation. Due to the endocrine nature of 1,25(OH)2D3 in circulation, it is difficult to discern between these 2 sources. We recently created a regulatory deletion model of Cyp27b1 (M1/M21-DIKO) where mice have normal inflammatory-regulated Cyp27b1 expression in nonrenal tissues (unlike global Cyp27b1-KO) but no expression within the kidney. Here, utilizing on-tissue chemical derivatization and matrix assisted laser desorption ionization-mass spectrometry imaging (MALDI-MSI), we investigated the distribution of 1,25(OH)2D3 and 25(OH)D3 in the kidney, liver, spleen, and thymus. MALDI-MSI demonstrated increased 1,25(OH)2D3 in nonrenal tissues such as the spleen after vitamin D3 supplementation in M1/M21-DIKO mice. Additionally, from this, we found increased Il4 and decreased Tnfa in the spleen after vitamin D3 supplementation. Taken together, these data demonstrate nonrenal production of 1,25(OH)2D3 in vivo and provide a consequence of vitamin D3 supplementation and nonrenal 1,25(OH)2D3 production in cytokine changes.

Sample extraction plasma: 125 µl of plasma sample was placed in a standard 2 ml glass screw top autosampler vial and the vial capped using a magnetically transportable PolyMag™ cap (GERSTEL, Germany).The sample is then placed on the vial tray of the MultiPurpose Sampler (MPS).50 µL of internal standard solution (d6-25OH-D3, 100ng/ml)) and 50µL (1d6-1α,25(OH)2D3, 100 pg/ml) were added to the sample leading to a final concentration of 25 ng/ml) and 25 pg/ml respectively, followed by 200 µL of a 0.2 M zinc sulphate solution to enhance the sensitivity of the assay.Following this, 500 µl of methanol is added to the vial to precipitate the proteins.The vial was then moved using magnetic transport to the CF-100 centrifuge whereby the contents were thoroughly vortexed for 1 minute to assist in the protein precipitation.The vial was then centrifuged at 3000 rpm for 1 minute to separate the proteins from the supernatant.A 10 mg C18 Vitamin D ITSP SPE cartridge was solvated with 100 µl of methanol and then equilibrated with 100 µL of HPLC grade water.500 µL of the supernatant is then loaded onto the SPE cartridge, before the cartridge was washed with 100 µL of 70 % methanol in water.The cartridge was then dried with 250 µL of air.Analytes are eluted with one 250 µL aliquot of methanol into a 500 µl high recovery vial.The polarity of the final solution is then adjusted by the addition of 100 µL of HPLC grade water, to improve the peak shape of the analytes.Samples were then divided in two aliquots of 150µL.One aliquot was transferred to a 200µL insert (2ml amber vial) for LC-MS/MS analysis.The other 150µL was dried under gently stream of Nitrogen at RT and then subjected to a one-step derivatization using Ampiflex® Diene as reagent to improve the ionization efficiency.To the dried residue, 30µL of Ampiflex solution was added as per manufacture specification, vortexed 30 and incubated for 60 min at RT.Then, sodium Borohydride solution (10mM) (10µL) was then added to reduce the complexity of the Chromatogram by reducing the enamine double bond formation of the Ampiflex derivative.Finally, MiliQ water (30µL) was added to the sample, vortexed for 15 s, and then transferred into 200µL HPLC inserts (2ml amber vials) for further LC-MS/MS.

Sample extraction for tissue homogenate:
Tissue sections (~50 mg) were homogenized by adding 700µL of hexane/acetone mixture (1:1, v/v) and using ultrasonication for 1min, amplitude 80% and interval of 0.5-0.9s(UP50H, MMTG, US) in a 1.2 mL Eppendorf and then added 50µL and 100µL of 25(OH)D3 (100ng/ml) and 1α,25(OH)2D3 (100pg/ml) respectively.Samples were kept in ice bath during homogenization.300µLof MiliQ water was added to the mixture, vortexed for 30s, ultrasonicated for 10 min and centrifuged for 10 min at 15,000 RPM.The supernatant (~600µL) was dried down by evaporation using a gently steam of nitrogen at RT.Samples were reconstituted in (80:20 (v/v)) water: methanol and transferred to the MPS (200µL) and then followed sample preparation steps are per plasma analysis previously described in Experimental Procedures except for the addition of ISTDs.

Acquisition and validation parameters:
Duplicate injections were made of each sample preparation and analysis reposted as the average (ng/ml) for plasma and (ng/g) for tissues for 25(OH)D3 and (pg/ml) and (pg/g) for plasma and tissue respectively for 1α,25(OH)2D3.For each analyte (native and derivatized) calibration curves were plotted with peak area ratios of vit D metabolites to the respective ISTD versus a range of analyte concentration in the corresponding units.For both analytes, two peak were observed (corresponded to their diastereomers) and summed areas were used for quantitation.
For 25(OH)D3, dynamic range (triplicate injection) was from 2-100 ng/ml (plasma) equivalent to 5-250 ng/g (tissue homogenate) with a lower Limit of Quantitation (LLOQ) of 2ng/ml (plasma) and 5 ng/g (tissues).The calibration curve for plasma was used for quantitation of tissue homogenate factoring a dilution factor of 1.5 in the calculations.Three replicates for each concentration level (2,16,38 and 100 ng/ml) were processed.The average percentage of Coefficient of Variation (CV%) was 9.5 for calibrators and 8-13% for QCs with a mass accuracy of between 77 (at LLOQ) -106% for both calibration and QCs (Fig. S5).
For 1α,25(OH)2D3, dynamic range (triplicate injection) was form 2-100pg/ml (plasma) equivalent to 5-250 ng/g (tissue homogenate) with a lower Limit of Quantitation (LLOQ) of 2pg/ml (plasma) and 5 pg/g (tissues).The calibration curve for plasma was used for quantitation of tissue homogenate factoring a dilution factor of 2.5 in the calculations.Three replicates for each concentration level (2, 20, 50 and 100 ng/ml) were processed.The average percentage of Coefficient of Variation (CV%) was 5.7 for calibrators and 10-17% for QCs with a mass accuracy of between 89 -104% for both calibration and QCs (Fig. S6).
Accuracy and precision replicated samples at two concentration levels were analyzed in separate runs to determine intra-day precision and accuracy.Intra-day accuracy and precision were calculated by processing six replicates at two concentration levels.Precision of the assay was estimated by the CV% for each concentration level.Accuracy was represented as recovery% from the nominal concentration as per Table S4.S9 which is adapted from the aforementioned protocol.For a given model, approximately 2 g (∼250 mg/layer × eight layers) of the concordant Vit D metabolites free (Golden West Biological, CA, USA) bulk tissue was homogenized without additional solvent using the FastPrep 24 bead homogenizer (MP Biomedicals) and stainless-steel lysing matrix (MP Biomedicals).The bulk homogenate was then aliquoted into four homogenizing vials using a positive displacement pipette.After determining the mass of the tissue homogenate in each vial, an appropriate amount of the standard was spiked into each homogenate to yield the desired final tissue concentration for (5-250 ng/g) for 25OHD3 and (5-250 pg/g) for 1α,25(OH)2D3.The volume of standard spiked into each homogenate was maintained below approximately 3% (w/w) to minimize the impact on the native tissue density.A mold was prepared from a 3 ml syringe by drawing back the plunger and removing the luered end.Enough of the blank tissue homogenate was added to the mold to sufficiently coat the plunger of the syringe (∼250 μL).The mold was immersed into dry ice cooled isopropanol to freeze the homogenate layer without submerging the open end of the syringe.Once completely frozen, the tissue plug was removed from the mold by carefully depressing the syringe plunger.A longitudinal (vertical) crosssection of the tissue plug sampling all concentration levels was then obtained by cryosectioning as previously described for target tissue sections.Sections were collected at the same thickness as the tissue to be quantified and thaw-mounted to the same substrate (indium tin oxide (ITO) coated microscope slide) as the sample to be quantified so that both exhibited to the same sample preparation conditions.
Histological Staining: Tissue sections (kidney and spleen) were stained using hematoxylin (0.02 g/L in ethanol) (Sigma-Aldrich, Dorset, UK) and eosin (0.003g/L in water + 1% Na2CO3) dyes as follows: fixed in 20°C acetone (Honeywell, UK) for 10 min and air-dried.Sections were rehydrated in 70 % v/v ethanol (EtOH) (≥99.8 %, Honeywell, Arlington, UK) (2 min) and tap water (5 min).They were then immersed in hematoxylin dye (6 min), rinsed in water (2 min) and placed in a solution of 10 % v/v acetic acid (Sigma Aldrich, Dorset, UK) in 95 % EtOH (1 min).To enhance the efficacy of the hematoxylin stain, slides were rinsed in water for 15 min in a bluing step.Sections were transferred to eosin dye for 15 s and dipped in water 2-3 times rapidly or until streaking stopped.Stained sections were dehydrated in increasing concentrations of EtOH (50 % -100 % v/v) for 2 min each and cleared by two changes of xylene (reagent grade, Fisher Scientific, Loughborough, UK) also at 2 min each.Histological mount (Histamount National Diagnostics, Atlanta, US) was applied before sections were completely dry and a glass coverslip was applied.A 600-dpi image was taken of tissues on slides using a scanner (Epson Perfection V330, software version3.9.2.5 EN, Seiko Epson Corporation, Nagano, Japan).

Supplemental References
1. Higashi, T., Shimada, K., andToyo'oka, T. (2010) Advances in determination of vitamin D related Figure S1 Figure S3 flow rate was set to 0.35 ml/mi throughout the chromatographic run and the column temperature was maintained at 40 o C. A divert valve was used to minimize the contamination of the MS system and it was set to waste in the time interval of 0-5 min and 11-15 min.Mass Spectrometry analysis (Qtrap Ready 5500+ Mass Spectrometer (MS), SCIEX, Framingham, US) was caried out in positive ion mode using the TurboIonSpray Electrospray ionization (ESI) for detection of 1α,25(OH)2D3 and the corresponding ISTD and Atmospheric pressure Chemical Ionization (APCI) for detection of 25(OH)D3 and the corresponding ISTD.Multiple Reaction Monitoring (MRM) was used as acquisition mode.MS source parameters were set as follows: CAD: (Arb) 6, CUR (Arb): 40, I Spray Voltage (V): 5000, GSI (Arb): 35, GSII (Arb): 40, TEM ( o C): 450 and corona current (APCI): (5µA).For MRM transitions, please see Table
of Standards and Quality Control Samples: For 1α,25(OH)2D3: Stock solution of both 1α,25(OH)2D3 and d6-1α,25(OH)2D3 (as internal standard (ISTD)) (100µg/ml, in methanol) were used to preparate working solutions, standards calibration curves and quality controls (QCs).1α,25(OH)2D3 stock solution was further diluted with surrogate biomatrix (Golden Mass Spect Gold® Human Serum, Ultra-Low Vitamin D, Lipid Free, Cell Culture Collective INC, CA, US) to provide calibration standards in the range of 5-100pg/ml and an ISTD concentration of 100 pg/ml.Quality control samples were independently prepared in the surrogate biomatrix at four different concentrations (2,20,50,100 pg/m)l corresponding to (LLOQ, QCL, QCM, QCH).For 25(OH)D3, commercially available lyophilized four-point calibration standards and QC material (2, 16, 38 and 100ng/ml) corresponding to (LLOQ, QCL, QCM and QCH) respectively were prepared according to manufacture specifications (Chromsystem, München, GmBH).Stock solution of d6-25(OH)D3 at 100ug/ml in methanol was prepared and used as ISTD for quantitation.Quality surrogate calibration standards were prepared freshly daily from the working solutions.All stock and working solution were stored at -80 o C until analysis.

Table S3 : Multiple Reaction Monitoring Table
Not applicable, CE: Collision energy, DP: Delustering potential, CXP: Collision cell exit potential and EP: Exit potential.