Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2021 Sep 10.
Published in final edited form as: J Steroid Biochem Mol Biol. 2020 Apr 30;201:105688. doi: 10.1016/j.jsbmb.2020.105688

VITAMIN D REGULATION OF HAS2, HYALURONAN SYNTHESIS AND METABOLISM IN TRIPLE NEGATIVE BREAST CANCER CELLS.

CJ Narvaez 1, D Grebenc 2, S Balinth 1,#, JE Welsh 1
PMCID: PMC8432753  NIHMSID: NIHMS1736990  PMID: 32360595

Abstract

The vitamin D receptor (VDR) and its ligand 1,25(OH)2D3 (1,25D) exert anti-tumor effects, but considerable heterogeneity has been reported in different model systems. In general, cell lines derived from aggressive tumor subtypes such as Triple Negative Breast Cancer (TNBC) express low levels of VDR and are less sensitive to 1,25D than those derived from more differentiated tumor types. We have previously reported that 1,25D inhibits hyaluronic acid synthase 2 (HAS2) expression and hyaluronic acid (HA) synthesis in murine TNBC cells. Here we confirmed the inhibitory effect of 1,25D on HA synthesis in human Hs578T cells representative of the mesenchymal/stem-like (MSL) subtype of TNBC. Because HA synthesis requires the production of hexoses for incorporation into HA, we predicted that the high HA production characteristic of Hs578T cells would require sustained metabolic changes through the hexosamine biosynthetic pathway (HBP). We thus examined metabolic gene expression in Hs578T cell variants sorted for High (HAHigh) and Low (HALow) HA production, and the ability of 1,25D to reverse these adaptive changes. HAHigh populations exhibited elevated HA production, smaller size, increased proliferation and higher motility than HALow populations. Despite their more aggressive phenotype, HAHigh populations retained expression of VDR protein at levels comparable to that of parental Hs578T cells and HALow subclones. Treatment with 1,25D decreased production of HA in both HAHigh and HALow populations. We also found that multiple metabolic enzymes were aberrantly expressed in HAHigh cells, especially those involved in glutamine and glucose metabolism. Notably, Glutaminase (GLS), a known oncogene for breast cancer, was strongly upregulated in HAHigh vs. HALow cells and its expression was significantly reduced by 1,25D (100 nM, 24 h). Consistent with this finding, Seahorse extracellular flux analysis indicated that respiration in HAHigh cells was significantly more dependent on exogenous glutamine than HALow cells, however, acute 1,25D exposure did not alter metabolic flux. In contrast to GLS, the glutamate transporter SLC1A7 was significantly reduced in HAHigh cells compared to HALow cells and its expression was enhanced by 1,25D. These findings support the concept that 1,25D can reverse the metabolic gene expression changes associated with HA production in cancer cells with aggressive phenotypes.

Keywords: Vitamin D, Hyaluronan, HAS2, metabolism, TNBC

1. Introduction

Triple negative breast cancers (TNBC) which lack estrogen and progesterone receptors and do not exhibit amplification of HER2 represent the most lethal subtype of breast cancer with few treatment options. New therapeutic targets for these aggressive tumors are critically needed. The Cancer Genome Atlas (TCGA) datasets indicate that greater than 95% of TNBC express the vitamin D receptor (VDR)(1). The VDR ligand 1,25-Dihydroxyvitamin D3 (1,25D) and other VDR agonists induce growth arrest and apoptosis and inhibit invasion in TNBC (2–4). Through genomic profiling of TNBC cells derived from wild-type (WT) and VDR knockout (VDRKO) mice, HAS2 was identified as a 1,25D repressed gene (2). The HAS2 gene is overexpressed in a high percentage of breast tumors, especially in basal-like TNBC (5). HAS2 encodes the enzyme hyaluronan synthase 2 which synthesizes the polysaccharide hyaluronic acid (HA), a major component of the extracellular microenvironment. Tightly regulated biosynthesis and degradation of HA is important for normal tissue homeostasis (6) and overproduction of HA and its cell surface receptor CD44 has been associated with tumor cell proliferation, invasion and metastasis (5). Patients whose breast tumors express high levels of HAS2 and HA have more aggressive tumors and poor survival (7).

HA synthesis is dependent upon the supply of substrates, UDP-N-aceytylglucosamine (UDP-GlcNAc) and UDP-glucuronic acid (UDP-GlcUA) which are generated from the hexosamine biosynthetic pathway (HBP). By upregulating glucose uptake and its flux into biosynthetic pathways, cancer cells increase their supply of UDP-hexoses (8). Elevated GFPT (glutamine-fructose-6-phosphate transaminase), the rate limiting enzyme of the HBP, is found in many cancers and contributes to increased UDP-GlcNAc (9). GFPT catalyzes the conversion of fructose-6-P to glucosamine-6-P in the presence of glutamine. Not only is glutamine required for HBP, but it’s an important source of glutamate which can enter the tricarboxylic acid (TCA) cycle for energy production (9). We have shown that 1,25D alters cellular handling of both glutamine and glutamate in non-transformed mammary epithelial cells, but its effect on metabolism in breast cancer cells have not been characterized (10).

To study the effects of 1,25D on HA production and metabolism in TNBC, we utilized an established human cell line (Hs578T) representative of the mesenchymal/stem-like (MSL) subtype of TNBC. Hs578T cultures are composed of morphologically distinct sub-populations with differential invasive potential (11). We hypothesized that Hs578T sub-populations might exhibit differences in HA synthesis and/or metabolism that confer unique phenotypes. To test this hypothesis, we sorted parental Hs578T cells into HAHigh and HALow populations based on cell surface HA abundance. The HAHigh subclone displayed a more aggressive phenotype (enhanced proliferation and motility, increased HA production and altered expression of multiple enzymes involved in glucose and glutamine metabolism) than the HALow subclone. However, both subclones retained VDR expression and exhibited decreased proliferation, reduced HAS2 expression and lower HA production in response to 1,25D treatment. In the HAHigh subclone, the effects of 1,25D were independent of changes in mitochondrial respiration but were associated with changes in metabolic gene expression.

2. Materials and methods

2.1. Cell Culture

Hs578T cells were purchased from the American Type Culture Collection (Rockville, MD) and grown in Dulbecco’s Modified Eagle Medium with high glucose (DMEM; Sigma-Aldrich, St. Louis, MO) supplemented with 10% fetal bovine serum (FBS; Atlanta Biologicals, Norcross, GA), 15 mM HEPES, and 10 µg/mL insulin (Sigma). Hs578T cells were routinely passaged every 4–5 days at a density of 4,000 cells/cm2 with media changes every 2 or 3 days, and maintained in a humidified 5% CO2 incubator at 37°C.

2.2. Cell Sorting and Flow Cytometry

Sorting of Hs578T cells into high and low HA producing populations was accomplished on a BD FACSAria II cell sorter (BD Biosciences, San Jose, CA) using the 488-nm laser to detect FITC signal, and equipped with BD FACSDiva software (v6.1.3; BD Biosciences). Parental Hs578T cells were plated in four T-150 flasks at a density of 4,000 cells/cm2 and grown for 96 h. The cells were trypsinized and pooled. Thereafter, all steps were performed at 4°C. The pooled cells were blocked with PBS/4% BSA, incubated for 30 min with biotinylated HA binding protein (bHABP; MilliporeSigma, Burlington, MA) diluted 1:30 in blocking buffer, and fluorescently labeled for 15 min with Alexa Fluor 488 streptavidin (Invitrogen Molecular Probes, Eugene, OR) diluted 1:800 in blocking buffer. PBS/0.2% BSA was used for washes to remove unbound bHABP and streptavidin. Cells were passed through an 85 µM nozzle tip and sorted into HAHigh (15% brightest) and HALow (15% dimmest) populations based on bHABP fluorescence which labeled cell-associated HA. The sorted cells were expanded in culture and re-sorted up to six times (12). Monitoring of bHABP fluorescence in sorted populations was by flow cytometry. Briefly, cells were plated in 100-mm dishes, harvested and stained as described above, and analyzed on a BD LSR II Flow Cytometer (BD Biosciences). 10,000 events gated on single cell populations were collected and post-acquisition analysis was performed using FlowJo 7.6.5 (FlowJo, LLC, Ashland, OR).

2.3. Confocal Microscopy

Hs578T subclones were plated in Lab-Tek II CC2 4-well chamber slides (Krackeler Scientific, Albany, NY) with a density of 5,000 cells/cm2 and grown for 96 h. The cells were fixed in 1% formaldehyde in PBS for 15 min and blocked overnight in PBS/1% BSA containing 0.02% sodium azide. The slides were then incubated with bHABP, diluted 1:30 in blocking buffer, for 30 min at room temperature in a humidified chamber. Slides were washed three times with PBS, followed by 30 min incubation with Alexa Fluor 488 streptavidin diluted 1:400 in blocking buffer. Slides were washed three times and coverslips were applied with Prolong Gold antifade mountant with DAPI (Invitrogen). Images were taken with a 40× oil immersion objective on a Leica DMI6000 microscope with TCS SP5 confocal laser scanner using Leica Application Suite AF version 2.6.0.7266 software (Leica Microsystems, Buffalo Grove, IL).

2.4. Particle Exclusion Assay

Hs578T subclones were plated in Lab-Tek II 4-well chambered coverglass #1.5 borosilicate (Krackeler) with a density of 500 cells/cm2 and grown for 48 h. The cells were rinsed with prewarmed Hank’s Balanced salt solution (HBSS; Sigma), fixed for 10 min in 1% formaldehyde in HBSS at room temperature, washed three times with HBSS, and incubated for 15 min with Hoechst 33258 (1mg/mL; Sigma) diluted 1:150 in PBS/0.2% BSA. The cells were washed three times with PBS/0.2% BSA, and 7.5 × 108 fixed sheep red blood cells (Fitzgerald Industries International, Acton, MA) were allowed to settle for 20 min at 37°C. Images were taken with a 20× objective using confocal microscope (Leica).

2.5. MUSE Count and Viability Assay

For growth assay, Hs578T subclones were plated in 6-well plates at a density of 2,000 cells/cm2, treated the next day with 100 nM 1,25D or vehicle control and grown for up to 144 h. At 72 h, media containing treatments were replaced. The cells were harvested by trypsinization, pooled with medium plus washes and pelleted by centrifugation. The cells were resuspended in complete media diluted 1:10 with PBS (1% FBS/PBS). For determination of count and viability, an aliquot of the cell suspension was diluted 1:10 with MUSE Count & Viability reagent (EMD Millipore, Billerica, MA), which differentially stained viable and non-viable cells based on their permeability to two DNA binding dyes, and incubated for 5 min in the dark at room temperature according to manufacturer’s protocol.

The multiparametric fluorescent detection of individual cells was performed by MUSE Cell Analyzer (EMD Millipore), a microcapillary flow cytometer equipped with a 532-nm laser, a forward scatter and two fluorescence (YLW 576/26, RED 680/30) detectors. Data were captured and analyzed using MUSE 1.4 analysis software.

2.6. Immunoblotting

Hs578T subclones were plated in 100-mm dishes at a density of 6,000 cells/cm2 and treated the next day with 100 nM 1,25D or vehicle control for 72 h. Whole cell lysates were prepared by washing the cell monolayer two times with ice-cold PBS and scraping into Cell Extraction Buffer (Invitrogen) supplemented with 1 mM PMSF and Halt protease inhibitor cocktail (ThermoFisher Scientific, Waltham, MA) according to manufacturer’s protocol. Lysates were solubilized in Laemmli sample buffer, separated by SDS-PAGE, transferred to PVDF, and immunoblotted with VDR (D6) mouse monoclonal antibody (Santa Cruz Biotechnology, Santa Cruz, CA), HAS2 rabbit polyclonal antibody (AW5442; Abgent, San Diego, CA), or GFPT1 rabbit polyclonal antibody (GFAT1, #3818; Cell Signaling Technology, Danvers, MA). Specific antibody binding was detected by horseradish peroxidase-conjugated secondary antibodies (GE Healthcare Bio-Sciences, Pittsburgh, PA) exposed to SuperSignal West Dura ECL substrate (ThermoFisher Scientific) on myECL CCD digital imaging system (ThermoFisher Scientific). After imaging, blots were stripped with Restore Western Blot Stripping Buffer (ThermoFisher Scientific) and reprobed with GAPDH mouse monoclonal antibody (Bio-Rad Antibodies, Hercules, CA).

Non-nuclear membrane (NNM) fractions isolated by sequential centrifugation were used to detect GLS (13). Briefly, Hs578T subclones were plated in 150-mm dishes at a density of 3,000 cells /cm2 and treated the next day with 100 nM 1,25D or vehicle control for 96 h. The cells were harvested by trypsinization, washed and pelleted by centrifugation. The cell pellets were resuspended in Digitonin Buffer (25 ug/mL digitonin, 150 mM NaCl, 50 mM HEPES pH 7.4), incubated for 10 min at 4°C and then centrifuged at 2,000 × g to pellet the cells. The supernatant is enriched in cytosolic proteins. The pellet was resuspended in NP40 Buffer (1% NP40, 150 mM NaCl, 50 mM HEPES pH 7.4), incubated for 30 min at 4°C and then centrifuged at 7,000 × g to pellet nuclei and cell debris. This supernatant is enriched in NNM proteins such as ER, golgi and mitochondrial proteins. The NNM fractions were solubilized in Laemmli sample buffer, separated by SDS-PAGE, transferred to PVDF, immunoblotted with GLS rabbit monoclonal antibody (AP8809B; Abgent) and developed as described above. ATP synthase mouse monoclonal antibody (ATP5A; BD Biosciences) was used as loading control.

2.7. HA ELISA

Hs578T subclones were plated in 6-well plates at a density of 2,000 cells/cm2, treated the next day with 100 nM 1,25D or vehicle control and grown for up to 96 h. Media were collected and stored @−20°C. Nunc Maxisorp 96-well Immunoplate (ThermoFisher Scientific) was coated with HABP (non-biotinylated; Millipore) solution overnight at 4°C. Next day, the plate was washed with PBS containing 0.1% Tween 20 (PBST) and blocked with PBST containing 2% BSA for 1 h at 37°C. After another series of washes, samples and HA standards (10 to 320 ng/mL; R&D Systems, Minneapolis, MN) were added and incubated for 1 h at 37°C. The samples and standards were removed after another series of washes, and bHABP was added to the plate and incubated for 1 h at 37°C. Unbound bHABP was removed after a series of washes, and the secondary streptavidin-HRP (R&D Systems) was added to plate, and incubated for 1 h at 37°C. Unbound secondary was removed after a series of washes, and TMB peroxidase substrate (SureBlue TMB; SeaCare Life Sciences, Milford, MA) was added to develop blue color for about 20 min. The reaction was stopped with 1 M HCl, and the absorbance was measured at 450 nm with a background correction at 590 nm using Victor3 V microtiter plate reader (PerkinElmer, Waltham, MA).

2.8. Gene Expression Analysis

Hs578T subclones were plated in quadruplicate on 100-mm plates at a density of 6,000 cells/cm2. Two days after plating, the cells were treated for 24 h with 100 nM 1,25D or vehicle control. Total RNA was isolated using RNeasy Plus Mini kits (Qiagen, Valencia, CA). Three independent cDNA stocks were prepared from each sample, using multiscribe reverse transcriptase and random hexamer primers (Applied Biosystems, Foster City, CA), pooled, and each sample was analyzed in quadruplicate using PowerUp SYBR Green Master Mix (Applied Biosystems) on a QuantStudio 12K Flex Real-Time PCR System (Applied Biosystems). Primer sequences were designed using Primer-BLAST (National Library of Medicine) and DNA melting curve of predicted PCR products were verified using uMeltSM v2.02 (www.dna.utah.edu/umelt/um.php;(14)). Primers were ordered from Integrated DNA Technologies (Coralville, IA). Data were calculated using the ∆∆Ct method, normalized against 18S, and expressed relative to values from Hs578T parental cells, which were set at 1.

2.9. Mitochondrial Respiration

For the Agilent Seahorse (Santa Clara, CA) Mito Stress Test, Hs578T subclones were plated at a density of 6,000 cells per well in XFp cell culture microplates and incubated at 37°C in a 5% CO2 incubator. The next day, cells were treated for 24 h with 100 nM 1,25D or vehicle in complete media with or without 2 mM glutamine. The Mito Stress Test measures key parameters of mitochondrial function by directly measuring the oxygen consumption rate (OCR) of live cells in real time. To perform the assay, cells were switched to buffer-free XF assay media (Agilent Seahorse) containing 1 mM pyruvate, 25 mM glucose, and with or without 2 mM glutamine one hour prior to assay. OCR was measured on the XFp analyzer before and after serial injections of 1 µM oligomycin, 1 µM carbonyl cyanide 4-(trifluoromethoxy) phenylhydrazone (FCCP), and a mix of 0.5 µM each of rotenone and antimycin A according to manufacturer’s instructions. The OCR data were normalized to DNA concentration in each well as assessed by Nanodrop (ThermoFisher Scientific). Cell Mito Stress Test data represent the Mean ± standard deviation (SD) of three independent biological replicates.

2.10. Statistics

Data are expressed as Mean ± SD and statistical analysis for most assays was by one-way ANOVA followed by Bonferroni multiple comparison as a post hoc test. For Seahorse data, mean values were analyzed using Newman-Keuls multiple comparison post hoc test. For time course proliferation data, statistical evaluation was by unpaired t-test. Statistical analyses were performed using GraphPad Prism 4 software (San Diego, CA). Differences were considered significant at p<0.05 in all experiments.

3. Results

3.1. Generation of high and low HA producing Hs578T subclones

In previous studies we observed that Hs578T cultures exhibit morphological heterogeneity, with small fibroblastic cells surrounding islands of larger epithelial-like cells (Fig 1A, Top). To determine whether these two sub-populations exhibited differences in HA metabolism or VDR signaling, parental Hs578T cells were sorted into High (HAHigh) and Low (HALow) HA populations with biotinylated hyaluronic acid binding protein (bHABP) which binds to cell-associated HA (Fig 1A, Bottom). The sorted populations were expanded in culture and re-sorted six times to generate subclones with distinct HA profiles. At high density (Fig 1B, Top Panels), sorted cells showed a clear distinction in morphology, with HAHigh cells demonstrating fibroblastic features whereas HALow cells were flat and large. Thus, sorting Hs578T cells for HA production divided the two morphological types of cells evident in the parental strain. Flow cytometry data following sort #6 confirmed the separation of HAHigh and HALow cell lines with respect to cell-associated HA (Fig. 1B, Bottom Panels). There was very little overlap in HABP signal between the two populations, and no overlap in their peaks.

Fig 1. HA expression in sorted Hs578T subclones.

Fig 1.

Open in a new tab

A. Parental Hs578T were sorted with biotinylated hyaluronic acid binding protein (bHABP) into HAHigh and HALow populations. The sorted cells were expanded in culture and re-sorted up to six times. Top: Phase images of parental Hs578T. Bottom: Flow cytometry data of bHABP binding to cell-associated HA detected by Alexa Fluor 488 streptavidin of parental Hs578T highlighting the HAHigh and HALow sorted regions. Isotype control, dashed line. B. Top Panels: Phase images of HAHigh and HALow subclones. Bottom Panels: Flow cytometry data of bHABP binding to cell-associated HA of HAHigh and HALow subpopulations after 6 rounds of cell sorting. Isotype control, dashed line. C. Validation of the effectiveness of the sort into high and low HA subpopulations. Left: Confocal imaging of fluorescently labeled bHABP binding to cell-associated HA on Hs578T subclones. Right: Phase images and visualization of pericellular HA coat by exclusion of sedimenting fixed erythrocytes on Hs578T subclones.

Immunofluorescence and particle exclusion assays were performed to verify the differences in morphology and HA production in the two cell lines. Confocal imaging with bHABP shows more intense surface HA staining in the HAHigh population than the HALow population (Fig.1 C, Left). To specifically visualize the pericellular matrix (i.e., cell coat), which is mostly comprised of HA anchored to the cell surface via HA-binding receptors and HA synthases (15), we conducted particle exclusion assays. This technique utilizes sedimenting fixed erythrocytes (which are repelled by HA and inherently fluorescent) to outline the edges of the pericellular matrix. Consistent with flow cytometry and confocal imaging demonstrating extensive cell surface bHABP in the HAHigh population, these cells have a more extensive pericellular coat than the HALow population (Fig. 1C, Right). Phase contrast imaging of the low-density cultures used for the particle exclusion assay also highlights the smaller overall size and pronounced surface protrusions of the HAHigh cells compared to the HALow cells.

3.2. Vitamin D signaling pathway in Hs578T subclones

Once the HAHigh and HALow subpopulations were established, we examined the integrity of the VDR signaling pathway. A time course was initially performed to compare the proliferation rates of Hs578T subclones in the presence and absence of 1,25D. Under control conditions, HAHigh grow much faster than HALow cells, displaying a 50% increase in cell number after 144 h in culture (Fig. 2A). Treatment with 100 nM 1,25D significantly decreased cell number in HALow cells only after 144 h whereas HAHigh cells displayed a significant (20–25%) decrease in cell number within 96 h (Fig. 2B). These data suggest that the rapidly proliferating HAHigh cells may be more sensitive to the effects of 1,25D. However, VDR protein expression on western blots was comparable and 1,25D exposure (100 nM, 72 h) upregulated VDR abundance equally in the two subclones (Fig. 2C). Despite comparable VDR protein expression in cell lysates from the subclones, qPCR analysis demonstrated higher VDR gene expression in HAHigh cells both in the presence and absence of 1,25D (Fig 2D). The discrepancy in VDR expression at the mRNA and protein levels may be related to differences in cell density or time in culture which could affect VDR abundance. CYP24, a mitochondrial P450 enzyme that initiates the degradation of 1,25D, is tightly controlled by 1,25D/VDR. CYP24 gene expression was significantly upregulated by 1,25D in both cell lines, but this induction was 9-fold greater in HAHigh cells (Fig 2D). Overall, these results indicate that the VDR signaling pathway is intact in both subclones, but HAHigh cells display higher sensitivity to 1,25D. Differences in 1,25D responsiveness (CYP24A1 induction or growth arrest) may be related to the unique properties of each cell line. For example, HA activation of CD44 triggers multiple receptor tyrosine kinases (ERBB2, ErbB3, EGFR, IGF1R-β, PDGFR-β, and c-MET) and VDR may be a downstream phosphorylation target of these. HAhigh cells also exhibit a higher proliferation rate which likely renders them more sensitive to growth inhibitory agents.

Fig 2. Effects of 1,25D on cell number, VDR and CYP24 gene expression in Hs578T subclones.

Fig 2.

Open in a new tab

Cells were plated at a density of 2,000 cells/cm2 in 6-well plates, treated the next day with ethanol (E) vehicle or 100 nM 1,25D (D3), harvested and counted every 24 h up to 144 h by using the MUSE Cell Analyzer as described in Materials and methods. A. Time Course. Comparison of proliferation rates between the two cell lines in the absence of 1,25D. Data represent Mean ± SD of three replicates and are representative of at least three independent experiments. B. Cell number after treatment with 1,25D. Data represent Mean ± SD of three replicates and are representative of at least three independent experiments. C. VDR immunoblot. Cell lysates were collected after 72 h treatment with vehicle (EtOH) or 100 nM 1,25D (D3) as described in Materials and methods. Representative blot of at least three independent experiments. GAPDH is loading control. D. VDR and CYP24 gene expression. Cells were treated for 24 h with vehicle (EtOH) or 100 nM 1,25D, RNA isolated, cDNA prepared, and qPCR performed as described in Materials and methods. Data represent Mean ± SD of four independent biological replicates. Bars annotated with different letters are significantly different (p<0.05) as assessed by one-way ANOVA. *p<0.05, HAHigh vs HALow (A); 1,25D vs ethanol (B) as evaluated by unpaired t-test.

3.3. Effects of 1,25D on HAS2 expression and HA production in Hs578T subclones

The expression of HAS2, the major HA synthesizing enzyme in Hs578T cells, was examined to assess its contribution to the differential production of HA in the Hs578T subclones. Relative to HALow cells, gene and protein expression of HAS2 was significantly higher in HAHigh cells, confirming that increased production of HA in these cells is associated with elevated expression of HAS2 (Fig 3A, 3B). Of the remaining HA synthesizing enzymes, HAS1 was not detected in parental Hs578T cells or either subclone, and HAS3 expression was comparable (data not shown). In previous studies, we discovered through genomic profiling that HAS2 is a 1,25D repressed gene in murine TNBC cells (2). As shown in Fig 3, 1,25D significantly decreased HAS2 gene and protein expression in the HAHigh subclone (Fig 3A, 3B) confirming that 1,25D also represses HAS2 in human TNBC cells. We also measured secreted HA in conditioned media by ELISA and found that HAHigh cells produced 65% higher amounts of soluble HA compared to HALow cells. In addition, a significant decrease in secreted HA was detected in both subclones after prolonged 1,25D treatment (Fig 3C).

Fig 3. 1,25D downregulates HAS2 and HA secretion in Hs578T subclones.

Fig 3.

Open in a new tab

Cells were plated, treated, harvested, and RNA isolated (24 h) or cell lysates (72 h) collected as described in Fig 2. A. HAS2 gene expression. Data represent Mean ± SD of four biological replicates. B. HAS2 protein expression. Representative blot of at least three independent experiments. GAPDH is loading control. C. Secreted HA. Cells were plated at a density of 2,000 cells/cm2 in 6-well plates, treated the next day with vehicle control or 100 nM 1,25D for 96 h, media collected, and HA ELISA performed as described in Materials and methods. Data represent Mean ± SD of three replicates and are representative of at least three independent experiments. Bars annotated with different letters are significantly different (p<0.05) as assessed by one-way ANOVA (A,C).

3.4. Effects of 1,25D on HBP and glutamine metabolism in Hs578T subclones

Flux of glucose and glutamine through the hexosamine biosynthetic pathway (HBP) is essential to provide UDP-hexosamines to HAS2 for synthesis of HA. Because 1,25D has been shown to alter glucose and glutamine handling in breast cells, we assessed expression of relevant metabolic genes in our model system. GFPT (glutamine-fructose-6-phoshate transaminase) is the rate limiting enzyme of HBP that controls the flux of glucose into the pathway via catalyzing the formation of glucosamine-6-P from fructose-6-P and glutamine. GFPT has two isomers, GFPT1 is ubiquitously expressed, and GFPT2 is normally expressed in heart, nervous and reproductive systems, and can also be found in tumors outside of the brain (16). Paradoxically, GFPT1 protein expression was more highly expressed in HALow cells compared to HAHigh cells but GFPT1 gene expression was significantly higher in the HAHigh subclone (Fig 4A, 4B). Thus, regulation and activity of GFPT1 may differ between the two subclones. Also shown in Fig 4B, GFPT2 is more highly expressed in the HALow than the HAHigh subclone but its expression in both subclones is significantly lower than that of the parental Hs578T cells (4-fold and 11-fold lower, respectively). There was no effect of acute treatment with 1,25D on expression of either GFPT isoform (Fig 4A, 4B).

Fig 4. Differential expression of GFPT, and genes involved in the synthesis and hydrolysis of glutamine in Hs578T subclones.

Fig 4.

Open in a new tab

Cells were plated, treated, harvested, and RNA isolated (24 h) or cell lysates (72 h) collected as described in Fig 2. A. GFPT1 protein expression. Representative blot of at least three independent experiments. GAPDH is loading control. B. GFPT1 and GFPT2 gene expression. C. GLS protein expression. NNM were isolated from cells after 96 h treatment with vehicle (EtOH) or 100 nM 1,25D (D3) as described in Materials and methods. Representative blot of at least three independent experiments. ATP5A is loading control. *, non-specific. D. GLUL and GLS gene expression. Data represent Mean ± SD of four independent biological replicates. Bars annotated with different letters are significantly different (p<0.05) as assessed by one-way ANOVA (B, D).

As noted above, glutamine is an essential substrate for GFPT in the formation of UDP-hexosamines for HA synthesis. Glutamine is also an important source of glutamate which can enter the TCA cycle for energy production. GLUL (glutamate-ammonia ligase) is a glutamine synthetase that catalyzes the synthesis of glutamine from glutamate and ammonia in an ATP-dependent reaction. Consequently, GLUL expression has been shown to render breast cancer cells glutamine independent. We previously observed that 1,25D downregulates GLUL expression and promotes glutamine dependency in breast epithelial cells (10). Despite high HA production, GLUL expression was undetectable in HAHigh cells (Fig 4D). Furthermore, protein and RNA expression of GLS (glutaminase), the enzyme that catalyzes the hydrolysis of glutamine to glutamate and ammonia, were upregulated in HAHigh cells. These data indicate limited glutamine synthesis coupled with increased glutamine catabolism in HAHigh cells (Fig 4C, D), which implies their metabolism would be highly dependent on exogenous glutamine. Although 1,25D did not alter expression of GLUL or GLS in HALow cells, it downregulated GLS in HAHigh cells potentially limiting catabolism of glutamine for energy production (Fig 4C, D).

3.5. Effects of 1,25D on Glutamate/Glutamine and Glucose transporters in Hs578T subclones

We also assessed whether genes associated with cellular uptake of HBP substrates were altered in HALow vs. HAHigh cells or were regulated by 1,25D (Fig 5). The glutamine and essential amino acid transporter SLC7A5 (typically associated with poor prognosis in breast cancer), was significantly upregulated in HAHigh cells relative to HALow cells. The high affinity glutamine transporter SLC1A5 was similarly expressed in the two subclones (data not shown). In contrast, the glutamate transporters, SLC1A3 and SLC1A7, were more highly expressed in HALow cells (Fig 5A). The low expression of SLC1A7, but not SLC1A3, was partially restored by 1,25D in HAHigh subclone. 1,25D did not alter aberrant expression of SLC7A5 in HAHigh cells (Fig 5A). Other glutamate transporters, SLC1A1 and SLC7A11, are similarly expressed in both subclones (data not shown). HAHigh and HALow cells displayed differential expression of glutamine and glutamate transporters, with HALow demonstrating increased overall expression of glutamate transporters and HAHigh showing increased glutamine transporter expression.

Fig 5. Differential expression of Glutamine/Glutamate and Glucose transporters in Hs578T subclones.

Fig 5.

Open in a new tab

Cells were plated, treated, harvested, RNA isolated (24 h) and qPCR performed as described in Fig 2. A. Gene expression of glutamine transporter (SLC7A5) and glutamate transporters (SLC1A3, SLC1A7). B. Gene expression of glucose transporters (GLUT1, GLUT3, GLUT4). Data represent Mean ± SD of four independent biological replicates. Bars annotated with different letters are significantly different (p<0.05) as assessed by one-way ANOVA (A, B).

Of the glucose transporters, the most significant difference was detected in GLUT3 expression, which was elevated in HALow cells but undetectable in HAHigh cells. GLUT1 was moderately upregulated in HAHigh cells and GLUT4 was comparable in the two cell lines. Overall, these data suggest that the subclones selected for differential HA production may utilize different mechanisms for uptake and metabolism of glutamine and glucose, and that 1,25D may selectively alter expression of the genes underlying these mechanisms (17, 18).

3.6. Evaluation of mitochondrial respiration in Hs578T subclones

To assess whether populations selected for differential HA production exhibited altered metabolism, we conducted real time metabolic flux assays on a Seahorse XFp (Agilent). The Mito Stress Test measures key parameters of mitochondrial function (Fig 6A) by directly measuring the oxygen consumption rate (OCR) of live cells before and after serial injections of oligomycin (ATP synthase inhibitor), FCCP (uncoupler of oxygen consumption and ATP production), and a mix of rotenone and antimycin A (inhibitor of complex I and III, respectively).

Fig 6. Effect of glutamine and 1,25D on mitochondrial respiration in live cells.

Fig 6.

Open in a new tab

HAHigh or HALow cells were plated in Agilent Seahorse XFp cell culture microplates in the presence or absence of 2 mM glutamine. The XFp flux analyzer measured temporal changes in oxygen consumption rate (OCR) of live cells as described in Materials and methods. A. Cell Mito Stress Test OCR profile. B. OCR of HALow and HAHigh cells in the presence or absence of 2 mM glutamine. Top: Data represents time course of OCR before and after serial injections of oligomycin, FCCP and rotenone/antimycin A. Bottom: Calculated parameters of mitochondrial respiration: ATP production and % spare capacity. C. Cell Mito Stress Test of HAHigh cells pretreated for 24 h with 100 nM 1,25D or vehicle (EtOH) in the presence or absence of 2 mM glutamine. Data represents time course of OCR before and after serial injections of drugs and calculated parameters of mitochondrial respiration: basal respiration, maximal respiration, ATP production, proton leak, and % spare capacity. Data represent Mean ± SD of at least three independent biological replicates (B, C). Bars annotated with different letters are significantly different (p<0.05) as assessed by one-way ANOVA.

Mito Stress Tests were conducted with HALow and HAHigh cells in the presence or absence of 2 mM glutamine. As shown in Fig 6B, OCR profiles were similar in HAHigh and HALow cells cultured under control conditions (i.e., media containing both glucose and glutamine), with basal and maximal OCR of approximately 40 and 50 respectively. In the presence of glutamine, HAHigh cells exhibited much greater ATP production which in the absence of glutamine was reduced to levels comparable to HALow cells (Fig 6B, Bottom). Moreover, spare capacity was significantly reduced in the absence of glutamine in HAHigh compared to HALow cells (Fig 6B, Bottom). These data are consistent with the changes in glutamine metabolic enzymes (particularly GLUL) in HAHigh cells which suggested this population would demonstrate dependence on exogenous glutamine.

We also conducted Cell Mito Stress Tests of HAHigh cells pretreated for 24 h with 100 nM 1,25D in the presence and absence of glutamine. As noted above, all key parameters of mitochondrial function were significantly reduced in the absence of glutamine. However, 1,25D did not affect mitochondrial function or the response to glutamine deprivation in HAHigh cells (Fig 6C).

4. Discussion

The major conclusion from this study is that VDR signaling is retained in basal-like TNBC cells, and that 1,25D regulates multiple facets of the aggressive phenotype in these cells. This conclusion is based on the assessment of proliferation, gene expression and production of HA, a known extracellular mediator of tumor aggressiveness. Using human Hs578T breast cancer cells, we confirmed previous studies in murine TNBC cells (2) that 1,25D negatively regulates HAS2, the major HA synthesizing enzyme. We extended our studies to demonstrate that 1,25D reduces both secreted and cell-associated HA content in Hs578T cells. In previous studies we utilized cells derived from VDR null mice to demonstrate that the VDR was necessary and sufficient for down regulation of HAS2 by 1,25D (2). Thus, both tumor VDR and vitamin D status are key factors in regulation of tumor HA signaling. Since HA and HAS2 have consistently been associated with the stem cell phenotype, tumor aggressiveness and poor survival in breast cancer (8, 19–21) these results support the concept that VDR signaling is tumor suppressive, even in advanced breast cancer.

Our data is consistent with published reports that 1,25D and structurally related VDR agonists inhibit TGF-β1-induced HAS2 expression and HA synthesis in cultured human fibroblasts (22). Furthermore, oral vitamin D reduced serum HA in normal rats, and topical 1,25D reduced HA accumulation in UV-treated hairless mouse skin (23, 24). In a clinical trial of hepatitis patients, Sabry et al found that vitamin D supplementation (15,000 IU/week for 48 weeks) significantly reduced serum HA at all three time points studied (12, 24 and 48 weeks) (25). Collectively, these data suggest that vitamin D comprehensively inhibits the HA pathway both in vitro and in vivo, supporting the translational relevance of our work.

To further investigate the links between VDR, HA and tumor phenotype, we sorted Hs578T cells based on cell surface HA and generated two subclones (HAHigh and HALow) with differential HA production. We show that elevated HA production in HAHigh cells correlates with elevated HAS2 gene/protein expression and is associated with increased proliferation, altered metabolic gene expression and enhanced glutamine dependence. Importantly, HAHigh cells retain an intact VDR signaling pathway and display higher sensitivity to 1,25D than HALow cells. 1,25D induces CYP24A1, inhibits proliferation, selectively alters metabolic gene expression and reduces HAS2/HA secretion in HAHigh cells. Although some of these effects were also observed in cultures with lower HA production, the magnitude of changes induced by 1,25D was typically less in HALow cells. The underlying basis for altered 1,25D responsiveness in these two subclones with similar VDR expression is unclear but may relate to inherent differences in signaling pathways that impinge on VDR activity. Various hormones, growth factors and oncogenes alter VDR expression and/or activity (26–29), thus differences in HA triggered signaling pathways in these two cell lines may contribute to altered VDR activity. In support of this suggestion, genomic profiling of HAHigh and HALow cells identified 358 differentially expressed genes between these two cell lines (data not shown). Gene ontology analysis highlighted 23 significantly enriched Molecular Function terms and over 700 enriched Biological Processes in this dataset (Supp Tables 1 and 2). As expected, several of the enriched functional categories are directly related to the HA pathway (glycosaminoglycan binding, heparin binding, cell adhesion molecule binding, acetylgalactosaminyltransferase activity, extracellular matrix binding), while others include lipase activity, calcium binding, and binding to multiple receptors. We also used Clariom assays of 1,25D treated cells (100 nM/24 h) to identify potential VDR targets. Surprisingly, 1,25D had limited effects on the global genome profile in these two cell lines, with only 11 and 5 genes significantly altered by 1,25D in HAHigh and HALow cells respectively (Supp Table 3). Other than CYP24A1, there was no overlap in the list of 1,25D regulated genes in the two cell lines. Further studies will be necessary to confirm the differential gene regulation by 1,25D in the HAHigh and HALow cells and explore the underlying basis for distinct VDR targets.

Focused analysis of specific genes involved in metabolism identified GLS as a potential VDR target. GLS encodes a glutaminase enzyme that is emerging as a therapeutic target for advanced breast cancers due to its correlation with poor prognosis (30, 31). GLS was elevated in HAHigh relative to HALow cells, and 1,25D significantly reduced its expression only in HAHigh cells. We also found that the glutamate transporter SLC1A7 was reduced in HAHigh relative to HALow cells and that 1,25D significantly increased its expression in HAHigh cells. Overexpression of SLC7A5 is associated with increased tumor size, high nuclear grade, and estrogen and progesterone receptor-negativity (32). Other genes involved in glutamine metabolism (GLUL) and transport (SLC7A5, SLC1A3) were unaltered by 1,25D, although several were profoundly different in HAHigh and HALow cells.

Since glutamine is essential for HBP flux which generates hexoses for HA production, altered metabolic gene expression in HAHigh cells was not surprising. However, we expected that GFPT1 and/or GFPT2, the genes that encode the HBP rate limiting enzyme GFPT, would be elevated in HAHigh cells. Dysregulation of the HBP pathway is common in tumorigenesis (24). GFPT1/2 is highly expressed in many cancers and can predict poor prognosis in breast (33), pancreatic (34), and colon cancer (35). However, loss of GFPT1 expression predicts poor prognosis in gastric cancer (36) indicating tumor specific contributions by this pathway. Surprisingly, we detected only a modest increase in GFPT1/2 gene expression, and no consistent change in GFPT1 protein expression, in HAHigh cells relative to HALow cells. Furthermore, neither GFPT1 nor GFPT2 expression was altered by 1,25D in either cell line. Thus, changes in GFPT did not correlate with the observed alterations in HA production in this model system.

Because glutamine is critical for HA production, but minimal changes in GFPT expression were detected, we compared glutamine dependency in HAHigh and HALow cells. In general, TNBC cells are more sensitive to glutamine deprivation than other breast cancer molecular subtypes because of low expression of GLUL, and hence, insufficient endogenous glutamine synthesis for survival (32, 37, 38). In this study, the more aggressive HAHigh cells had undetectable expression of GLUL and displayed greater glutamine dependence (detected as reduction in mitochondrial respiratory capacity in the absence of glutamine) compared to HALow cells. However, despite effects of 1,25D on metabolic gene expression in HAHigh cells, 1,25D did not alter basal mitochondrial function or the response to glutamine deprivation in this subclone.

Although we have focused on HA biosynthesis in this study, increased flux through the HBP also provides substrates for protein glycosylation. The stability and activity of HAS2 is dynamically regulated by O-GlcNAc by increasing half-life of the enzyme and trafficking to the plasma membrane where it is active. Loss of O-GlcNAc leads to rapid HAS2 degradation in proteasomes (39). Further studies to assess O-GlcNAc in relation to HAS2 expression in HAHigh and HALow cells treated with 1,25D would be of interest.

In summary, Hs578T cells, a model of mesenchymal/stem like TNBC, retain VDR and sensitivity to 1,25D. Cells selected for high cell surface HA exhibit HAS2 overexpression and increased HA secretion in association with glutamine dependency and high rates of proliferation and migration. 1,25D suppresses proliferation, HAS2 gene expression and HA production in cells selected for high HA production via mechanisms that do not involve changes in GFPT or mitochondrial respiration.

Supplementary Material

Supp Data

5. Acknowledgements

We would like to thank Steve Lotz for his technical expertise on sorting the parental Hs578T into HAHigh and HALow subclones.

6. Funding Source Declarations

This work was supported by the National Institutes of Health RO1CA194500 to JW.

7. References

  • 1.Cancer Genome Action Network: Comprehensive molecular portraits of human breast tumours. Nature, 490(7418), 61–70 (2012) doi: 10.1038/nature11412 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.LaPorta E and Welsh J.: Modeling vitamin D actions in triple negative/basal-like breast cancer. Journal of Steroid Biochemistry and Molecular Biology, 144, 65–73 (2013) doi: 10.1016/j.jsbmb.2013.10.022 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.So JY, Smolarek AK, Salerno DM, Maehr H, Uskokovic M, Liu F and Suh N.: Targeting CD44-STAT3 signaling by Gemini vitamin D analog leads to inhibition of invasion in basal-like breast cancer. PloS one, 8(1), e54020–e54020 (2013) doi: 10.1371/journal.pone.0054020 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Flanagan L, Packman K, Juba B, O’Neill S, Tenniswood M and Welsh J.: Efficacy of Vitamin D compounds to modulate estrogen receptor negative breast cancer growth and invasion. J Steroid Biochem Mol Biol, 84(2–3), 181–92 (2003) doi: 10.1016/s0960-0760(03)00028-1 [DOI] [PubMed] [Google Scholar]
  • 5.Udabage L, Brownlee GR, Nilsson SK and Brown TJ: The over-expression of HAS2, Hyal-2 and CD44 is implicated in the invasiveness of breast cancer. Experimental Cell Research, 310(1), 205–217 (2005) doi: 10.1016/j.yexcr.2005.07.026 [DOI] [PubMed] [Google Scholar]
  • 6.Heldin P, Basu K, Olofsson B, Porsch H, Kozlova I and Kahata K.: Deregulation of hyaluronan synthesis, degradation and binding promotes breast cancer. The Journal of Biochemistry, 154(5), 395–408 (2013) doi: 10.1093/jb/mvt085 [DOI] [PubMed] [Google Scholar]
  • 7.Okuda H, Kobayashi A, Xia B, Watabe M, Pai SK, Hirota S, Xing F, Liu W, Pandey PR, Fukuda K, Modur V, Ghosh A, Wilber A and Watabe K.: Hyaluronan synthase HAS2 promotes tumor progression in bone by stimulating the interaction of breast cancer stem-like cells with macrophages and stromal cells. Cancer Res, 72(2), 537–47 (2012) doi: 10.1158/0008-5472.Can-11-1678 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Chanmee T, Ontong P, Izumikawa T, Higashide M, Mochizuki N, Chokchaitaweesuk C, Khansai M, Nakajima K, Kakizaki I, Kongtawelert P, Taniguchi N and Itano N.: Hyaluronan Production Regulates Metabolic and Cancer Stem-like Properties of Breast Cancer Cells via Hexosamine Biosynthetic Pathway-coupled HIF-1 Signaling. J Biol Chem, 291(46), 24105–24120 (2016) doi: 10.1074/jbc.M116.751263 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Akella NM, Ciraku L and Reginato MJ: Fueling the fire: emerging role of the hexosamine biosynthetic pathway in cancer. Bmc Biology, 17 (2019) doi: 10.1186/s12915-019-0671-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Beaudin S and Welsh J.: 1,25-Dihydroxyvitamin D Regulation of Glutamine Synthetase and Glutamine Metabolism in Human Mammary Epithelial Cells. Endocrinology, 158(12), 4174–4188 (2017) doi: 10.1210/en.2017-00238 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Hughes L, Malone C, Chumsri S, Burger AM and McDonnell S.: Characterisation of breast cancer cell lines and establishment of a novel isogenic subclone to study migration, invasion and tumourigenicity. Clin Exp Metastasis, 25(5), 549–57 (2008) doi: 10.1007/s10585-008-9169-z [DOI] [PubMed] [Google Scholar]
  • 12.Tamada Y, Takeuchi H, Suzuki N, Aoki D and Irimura T.: Cell surface expression of hyaluronan on human ovarian cancer cells inversely correlates with their adhesion to peritoneal mesothelial cells. Tumour Biol, 33(4), 1215–22 (2012) doi: 10.1007/s13277-012-0369-4 [DOI] [PubMed] [Google Scholar]
  • 13.Holden P and Horton WA: Crude subcellular fractionation of cultured mammalian cell lines. BMC Res Notes, 2, 243 (2009) doi: 10.1186/1756-0500-2-243 [doi] [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Dwight Z, Palais R and Wittwer CT: uMELT: prediction of high-resolution melting curves and dynamic melting profiles of PCR products in a rich web application. Bioinformatics, 27(7), 1019–1020 (2011) doi: 10.1093/bioinformatics/btr065 [DOI] [PubMed] [Google Scholar]
  • 15.McLane LT, Chang P, Granqvist A, Boehm H, Kramer A, Scrimgeour J and Curtis JE: Spatial Organization and Mechanical Properties of the Pericellular Matrix on Chondrocytes. Biophysical Journal, 104(5), 986–996 (2013) doi: 10.1016/j.bpj.2013.01.028 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.de Queiroz RM, Oliveira I. A. j., Piva B, Bouchuid Catão F, da Costa Rodrigues B, da Costa Pascoal A, Diaz B. L. o., Todeschini AR, Caarls MB and Dias WB: Hexosamine Biosynthetic Pathway and Glycosylation Regulate Cell Migration in Melanoma Cells. Frontiers in Oncology, 9(116) (2019) doi: 10.3389/fonc.2019.00116 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Zheng W, Tayyari F, Gowda GA, Raftery D, McLamore ES, Shi J, Porterfield DM, Donkin SS, Bequette B and Teegarden D.: 1,25-dihydroxyvitamin D regulation of glucose metabolism in Harvey-ras transformed MCF10A human breast epithelial cells. J Steroid Biochem Mol Biol, 138, 81–9 (2013) doi: 10.1016/j.jsbmb.2013.03.012 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Zhou X, Zheng W, Nagana Gowda GA, Raftery D, Donkin SS, Bequette B and Teegarden D.: 1,25-Dihydroxyvitamin D inhibits glutamine metabolism in Harvey-ras transformed MCF10A human breast epithelial cell. J Steroid Biochem Mol Biol, 163, 147–56 (2016) doi: 10.1016/j.jsbmb.2016.04.022 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Li YJ, Li LL, Brown TJ and Heldin P.: Silencing of hyaluronan synthase 2 suppresses the malignant phenotype of invasive breast cancer cells. International Journal of Cancer, 120(12), 2557–2567 (2007) doi: 10.1002/ijc.22550 [DOI] [PubMed] [Google Scholar]
  • 20.Chanmee T, Ontong P, Mochizuki N, Kongtawelert P, Konno K and Itano N.: Excessive hyaluronan production promotes acquisition of cancer stem cell signatures through the coordinated regulation of Twist and the transforming growth factor beta (TGF-beta)-Snail signaling axis. J Biol Chem, 289(38), 26038–56 (2014) doi: 10.1074/jbc.M114.564120 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Li P, Xiang T, Li H, Li Q, Yang B, Huang J, Zhang X, Shi Y, Tan J and Ren G.: Hyaluronan synthase 2 overexpression is correlated with the tumorigenesis and metastasis of human breast cancer. Int J Clin Exp Pathol, 8(10), 12101–14 (2015) [PMC free article] [PubMed] [Google Scholar]
  • 22.Slominski A, Janjetovic Z, Tuckey RC, Nguyen MN, Bhattacharya KG, Wang J, Li W, Jiao Y, Gu W, Brown M and Postlethwaite AE: 20S-hydroxyvitamin D3, noncalcemic product of CYP11A1 action on vitamin D3, exhibits potent antifibrogenic activity in vivo. J Clin Endocrinol Metab, 98(2), E298–303 (2013) doi: 10.1210/jc.2012-3074 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Koshiishi I, Mitani H, Sumita T and Imanari T.: 1,25-Dihydroxyvitamin D(3) prevents the conversion of adipose tissue into fibrous tissue in skin exposed to chronic UV irradiation. Toxicol Appl Pharmacol, 173(2), 99–104 (2001) doi: 10.1006/taap.2001.9178 [DOI] [PubMed] [Google Scholar]
  • 24.Huang GR, Wei SJ, Huang YQ, Xing W, Wang LY and Liang LL: Mechanism of combined use of vitamin D and puerarin in anti-hepatic fibrosis by regulating the Wnt/beta-catenin signalling pathway. World J Gastroenterol, 24(36), 4178–4185 (2018) doi: 10.3748/wjg.v24.i36.4178 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Sabry D, Al-Ghussein MA, Hamdy G, Abul-Fotouh A, Motawi T, El Kazaz AY, Eldemery A and Shaker M.: Effect of vitamin D therapy on interleukin-6, visfatin, and hyaluronic acid levels in chronic hepatitis C Egyptian patients. Ther Clin Risk Manag, 11, 279–88 (2015) doi: 10.2147/tcrm.S66763 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Arriagada G, Paredes R, Olate J, van Wijnen A, Lian JB, Stein GS, Stein JL, Onate S and Montecino M.: Phosphorylation at serine 208 of the 1alpha,25-dihydroxy Vitamin D3 receptor modulates the interaction with transcriptional coactivators. J Steroid Biochem Mol Biol, 103(3–5), 425–9 (2007) doi: 10.1016/j.jsbmb.2006.12.021 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Jiang Y and Fleet JC: Phorbol esters enhance 1alpha,25-dihydroxyvitamin D3-regulated 25-hydroxyvitamin D-24-hydroxylase (CYP24A1) gene expression through ERK-mediated phosphorylation of specific protein 3 (Sp3) in Caco-2 cells. Mol Cell Endocrinol, 361(1–2), 31–9 (2012) doi: 10.1016/j.mce.2012.03.008 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Lee SM, Meyer MB, Benkusky NA, O’Brien CA and Pike JW: The impact of VDR expression and regulation in vivo. J Steroid Biochem Mol Biol, 177, 36–45 (2018) doi: 10.1016/j.jsbmb.2017.06.002 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Ting HJ, Yasmin-Karim S, Yan SJ, Hsu JW, Lin TH, Zeng W, Messing J, Sheu TJ, Bao BY, Li WX, Messing E and Lee YF: A positive feedback signaling loop between ATM and the vitamin D receptor is critical for cancer chemoprevention by vitamin D. Cancer Res, 72(4), 958–68 (2012) doi: 10.1158/0008-5472.CAN-11-0042 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Lampa M, Arlt H, He T, Ospina B, Reeves J, Zhang B, Murtie J, Deng G, Barberis C, Hoffmann D, Cheng H, Pollard J, Winter C, Richon V, Garcia-Escheverria C, Adrian F, Wiederschain D and Srinivasan L.: Glutaminase is essential for the growth of triple-negative breast cancer cells with a deregulated glutamine metabolism pathway and its suppression synergizes with mTOR inhibition. PloS one, 12(9), e0185092 (2017) doi: 10.1371/journal.pone.0185092 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Saha SK, Islam SMR, Abdullah-Al-Wadud M, Islam S, Ali F and Park KS: Multiomics Analysis Reveals that GLS and GLS2 Differentially Modulate the Clinical Outcomes of Cancer. J Clin Med, 8(3) (2019) doi: 10.3390/jcm8030355 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Cha YJ, Kim E-S and Koo JS: Amino Acid Transporters and Glutamine Metabolism in Breast Cancer. International journal of molecular sciences, 19(3), 907 (2018) doi: 10.3390/ijms19030907 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Dong T, Kang X, Liu Z, Zhao S, Ma W, Xuan Q, Liu H, Wang Z and Zhang Q.: Altered glycometabolism affects both clinical features and prognosis of triple-negative and neoadjuvant chemotherapy-treated breast cancer. Tumour Biol, 37(6), 8159–68 (2016) doi: 10.1007/s13277-015-4729-8 [DOI] [PubMed] [Google Scholar]
  • 34.Yang C, Peng P, Li L, Shao M, Zhao J, Wang L, Duan F, Song S, Wu H, Zhang J, Zhao R, Jia D, Zhang M, Wu W, Li C, Rong Y, Zhang L, Ruan Y and Gu J.: High expression of GFAT1 predicts poor prognosis in patients with pancreatic cancer. Sci Rep, 6, 39044 (2016) doi: 10.1038/srep39044 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Vasconcelos-Dos-Santos A, Loponte HF, Mantuano NR, Oliveira IA, de Paula IF, Teixeira LK, de-Freitas-Junior JC, Gondim KC, Heise N, Mohana-Borges R, Morgado-Diaz JA, Dias WB and Todeschini AR: Hyperglycemia exacerbates colon cancer malignancy through hexosamine biosynthetic pathway. Oncogenesis, 6(3), e306 (2017) doi: 10.1038/oncsis.2017.2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Duan F, Jia D, Zhao J, Wu W, Min L, Song S, Wu H, Wang L, Wang H, Ruan Y and Gu J.: Loss of GFAT1 promotes epithelial-to-mesenchymal transition and predicts unfavorable prognosis in gastric cancer. Oncotarget, 7(25), 38427–38439 (2016) doi: 10.18632/oncotarget.9538 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Kim S, Kim DH, Jung WH and Koo JS: Expression of glutamine metabolism-related proteins according to molecular subtype of breast cancer. Endocr Relat Cancer, 20(3), 339–48 (2013) doi: 10.1530/ERC-12-0398 [DOI] [PubMed] [Google Scholar]
  • 38.Kung HN, Marks JR and Chi JT: Glutamine synthetase is a genetic determinant of cell type-specific glutamine independence in breast epithelia. PLoS Genet, 7(8), e1002229 (2011) doi: 10.1371/journal.pgen.1002229 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Vigetti D, Deleonibus S, Moretto P, Karousou E, Viola M, Bartolini B, Hascall VC, Tammi M, De Luca G and Passi A.: Role of UDP-N-acetylglucosamine (GlcNAc) and O-GlcNAcylation of hyaluronan synthase 2 in the control of chondroitin sulfate and hyaluronan synthesis. J Biol Chem, 287(42), 35544–55 (2012) doi: 10.1074/jbc.M112.402347 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supp Data
NIHMS1736990-supplement-Supp_Data.docx (28.7KB, docx)

RESOURCES