Skip to main content
PMC home page
Translational Oncology logoLink to Translational Oncology
. 2020 Sep 18;13(12):100869. doi: 10.1016/j.tranon.2020.100869

A novel vitamin D gene therapy for acute myeloid leukemia

Yi Xu a,b, Kimberly Payne c, Linh Hoang Gia Pham b, Park Eunwoo b, Jeffrey Xiao a,b, David Chi b, Justin Lyu b, Rosalia Campion a, Samiksha Wasnik b, Il Seok Jeong a, Xiaolei Tang b,e, David Jeston Baylink b, Chien Shing Chen a, Mark Reeves a, Mojtaba Akhtari a, Saied Mirshahidi f, Guido Marcucci d, Huynh Cao a,⁎
PMCID: PMC7509076  PMID: 32956997

Abstract

Current treatment approaches for older adult patients with acute myeloid leukemia (AML) are often toxic and lack efficacy. Active vitamin D3 (1,25(OH)2D3) has been shown to induce myeloid blast differentiation but at concentrations that have resulted in unacceptable, off-target hypercalcemia in clinical trials. In our study, we found that the combination of 1,25(OH)2D3 and the hypomethylating agent (HMA) 5-Azacytidine (AZA) enhanced cytotoxicity and differentiation, and inhibited proliferation of several AML cell lines (MOLM-14, HL60) and primary AML patient samples. This observation was corroborated by our RNA sequence analysis data in which VDR, CD14, and BAX expression were increased, and FLT-3, PIM1 and Bcl-2 expression were decreased. To address the hypercalcemia issue, we genetically engineered MOLM-14 cells to constantly express CYP27B1 (the VD3 activating enzyme, 1-α-hydroxylase-25(OH)D3) through lentiviral transduction procedures. Subsequently, we used these cells as vehicles to deliver the CYP27B1 enzyme to the bone marrow of AML mice. We observed that AML mice with CYP27B1 treatment had longer overall survival compared to no treatment and displayed no significant change in calcium level.

Introduction

Acute myeloid leukemia (AML) is a bone marrow (BM) malignancy in which immature hematopoietic myeloid precursors have lost their ability to differentiate and are locked in a highly proliferative state [1,2]. Standard treatment for AML consists of induction chemotherapy with a combination of anthracyclines and cytarabine. An important factor in AML treatment is the patient's performance status (PS). Many elderly patients cannot tolerate intensive regimens because of co-morbidities and poor PS [3,4]. As a result, outcomes for this group, who actually constitute the majority of AML, are extremely poor [5]. Recently, with the introduction of hypomethylating agents (HMAs) and venetoclax, we observed longer progression-free survival for this population; however, most patients will eventually succumb to the disease and toxicity remains a concern [6]. Further research on less toxic and more effective therapeutic modalities for older AML patients is necessary [7].

Vitamin D is a group of fat-soluble secosteroids responsible for essential roles in bone metabolism and calcium homeostasis [8]. An equally important but less known function of active 1,25(OH)2D3 is its autocrine/paracrine effect on local tissues (i.e. skin, placenta, colon, pancreas, and macrophage) through extra-renal expression of CYP27B1. The CYP27B1 gene encodes a member of the cytochrome P450 superfamily of enzymes, which hydroxylases 25-hydroxyvitamin D3 at the 1-alpha position, resulting in active 1,25(OH)2D3. Ogunkolade et al. found that the expression of CYP27B1 mRNA in healthy colon tissue surrounding cancerous tissue in patients with colorectal cancer was significantly lower compared to colon tissue of healthy patients, suggesting that the tumor could have secreted endocrine/paracrine factors that could affect the enzyme level [9]. We have limited knowledge of CYP27B1 expression in the BM of AML patients.

There is considerable interest in using 1,25(OH)2D3 to treat cancer due to its ability to inhibit proliferation and induce differentiation in many types of cancer cells (i.e. colon, breast, prostate). As a secosteroid hormone, 1,25(OH)2D3 binds to the nuclear vitamin D receptor (VDR), which heterodimerizes with the retinoid X receptor (RXR) and attaches to the promoter regions of target genes [[10], [11], [12], [13], [14]]. It has been reported that 1,25(OH)2D3 induces differentiation of monoblast cells through interactions with major molecular pathways such as MAPKs, JAK/STAT, and PI3K/Akt [[15], [16], [17], [18], [19]]. 1,25(OH)2D3 stimulated monocytic differentiation was associated with increased ERK and JNK activity and was augmented by p38 MAPK inhibition [20]. Prior studies demonstrated that increased PI3K/Akt activity promotes neutrophil and monocyte development and is critical to induction of monocytic differentiation and protection from apoptosis [15,17,18]. In addition to the two aforementioned pathways, 1,25(OH)2D3 also causes phospholipase A2-mediated release of arachidonic acid from leukemic cells, which in turn increases protein kinase C activity, leading to monocytic differentiation [21,22]. However, the concentrations of 1,25(OH)2D3 required to induce differentiation in vitro are typically in the range of 100–1000 nanomolar (nM) [23], and such a serum level would result in serious hypercalcemia in humans (typical concentration around 0.1 nM). Thus, many clinical trials failed because they could not achieve therapeutic levels of 1,25(OH)2D3 without systemic hypercalcemia [[23], [24], [25]].

There is a mutual interaction between the 1,25(OH)2D3 system and epigenetic mechanisms. 1,25(OH)2D3 has been reported to be able to alter methylation of DNA in the promoters of some genes. Conversely, 1,25(OH)2D3 insensitivity is related to methylation of the VDR promoter, which impairs 1,25(OH)2D3 regulation of tumor suppressor genes [[26], [27], [28]]. Epigenetic modification drugs have the potential to reverse 1,25(OH)2D3 insensitivity. VDR reactivation can be induced by HDAC inhibitors in combination with 1,25(OH)2D3, which has been shown to upregulate a unique group of suppressed gene targets in control of proliferation and induction of apoptosis [29,30]. Other study also indicates a synergistic role of DNA demethylation in 1,25(OH)2D3 metabolism and enhancement of 1,25(OH)2D3 efficiency for monocytic differentiation by a DNA hypomethylating agent [31]. VDR interacts with chromatin modifiers and remodelers directly or indirectly to fine-tune gene expression, whereas VDR and 1,25(OH)2D3 target genes can be silenced by DNA methylation or histone modification [[26], [27], [28],32,33]. Because both 5-Azacytidine (AZA), a hypomethylating agent approved by the FDA for palliative use in AML patients and 1,25(OH)2D3 demonstrate the ability to affect gene expression and induce leukemic differentiation, it is possible that their combined treatment might result in enhanced anti-leukemic effects. Here, we study the anti-leukemia effects of combining 1,25(OH)2D3 and AZA on AML cells in vitro, and ex vivo. Our studies also address the issue of systemic hypercalcemia through the CYP27B1 gene therapy. We use of vehicle cells to carry the CYP27B1 gene to the BM to produce local, high concentration 1,25(OH)2D3, thus limiting systemic hypercalcemia.

Materials and methods

The list of reagents including manufacturers and catalogues of antibodies and kits are found in the Supplementary data (Table S1). Detailed information on cell lines, primary leukemic samples, mice and IACUC approved protocols can be found in Supplementary materials and methods.

Combination of 1,25(OH)2D3 and AZA in vitro and ex vivo

1 × 106 cells/ml AML primary cells and cell lines were plated into 12-well plates for 2 days in the presence of different concentrations of 1,25(OH)2D3, or AZA, or combination of 1,25(OH)2D3 and AZA. The cells were then harvested and examined for the expression of CD14 or phenotypic blasts by flow cytometry. Each experiment was replicated a minimum of three times for reproducibility.

Preparation of plasmid constructs and lenti-viruses

The lenti-CYP-GFP plasmid was used in this study, which carried the cytochrome P450 family 27 subfamily B member 1 (CYP27B1) gene for encoding the 25-hydroxyvitamin D 1α-hydroxylase (1α-hydroxylase) (Fig. S3). The detailed protocol for generating lentivirus can be found in Supplementary materials and methods.

RNA-sequencing and data processing

AML blast cell samples pretreated for 48 h with 1,25(OH)2D3, or AZA, or combination of 1,25(OH)2D3 and AZA were collected and sent to BGI Hong Kong for RNA preparation and sequencing. RNA-seq libraries were prepared by BGI Hong Kong. Expression data was compared between both samples by the analysis of individual selected genes for differential expression. Plots represent differentially expressed genes in fragments per kilobase of exon per million fragments mapped (FPKM).

Establishment of xenograft mouse model and combination treatment

To generate the MOLM-14/NRG xenograft AML mouse model, on day −15, NRG mice were injected with intra-peritoneal (IP) AZA (10 mg/kg), and 5 × 105 cherry-MOLM-14 cells were injected on day −14 intravenously (IV) through tail vein. AZA was used to deplete BM cells in vivo. Some mice were sacrificed at day 0 for histological analysis to confirm the establishment of the AML model. Treatments were given on day −1 and day 0 based on study group (see specific description in Fig. 5A). The mortality and body weights of experimental mice were recorded daily.

Fig. 5.

Fig. 5

Open in a new tab

Enhanced cellular differentiation and survival of mice engrafted with CYP27B1-MOLM14 without hypercalcemia. A The survival curve of engrafted NRG mice is shown. On day −1, NRG mice were intraperitoneally injected with AZA. Then mice were separated into four groups to receive intravenously one of the following MOLM-14 engraftments: 1) Control (injection of 5 × 105 GFP-MOLM-14 on day 0), 2) low CYP27B1 arm (injection of 5 × 105 CYP27B1-GFP-MOLM-14 (CLG-MOLM-14) on day 0), 3) middle CYP27B1 arm (injection of 2 × 106 CLG-MOLM-14 on day 0), 4) high CYP27B1 arm (injection of 4 × 106 CLG-MOLM-14 on day 0). B The femur BM cells of experimental mice were isolated and evaluated by FACS. Representative FACS plots show expression of CD14. C Immunohistolochemistry was performed to display co-localization of CD14 expression and GFP+ MOLM-14 cells inside the BM of experimental animals on day 14. D Calcium levels measured on day 7 and 14 were reported as percentage changes compared to baseline values for each group. Where applicable, data are means ± SEM from each group and were analyzed by Student t-test. *p < 0.05; N = 5.

Statistical analysis

Statistical significance was assessed by ANOVA or by independent student “t” test for comparison between two groups. All values were presented as mean ± SEM. In comparing the survival time of the mice, all times were measured from the day of BM transplant, estimated by the method of Kaplan and Meier. Results were considered significant when the p value was <0.05.

For original data, please contact hcao@llu.edu.

Results

Enhanced in vitro anti-leukemic effects of 1,25(OH)2D3 in combination with AZA on AML cell lines

We tested the combination of AZA and 1,25(OH)2D3 for anti-leukemia effects on MOLM-14 cells, a monoblast cell line harboring a FLT3 internal tandem duplication mutation (FLT3-ITD). We treated MOLM-14 cells with different concentrations of 1,25(OH)2D3 and AZA alone and in combination for 48 h. Cytotoxic effects were determined by flow cytometry (FACS) to detect CD14 and viability dye (Fig. 1A, see Fig. 1B for gating strategy). Combination therapy significantly reduced the percentage of viable blasts (Viable/CD14-cells) compared to 1,25(OH)2D3 alone and AZA alone (25.7% vs. 60.5% and vs. 69.0%, respectively) (Fig. 1A, C). Similar results were obtained with the HL60 AML cell line; however, combination therapy was not effective for the THP-1 AML cell line (Figs. S1, S2). Further FACS analysis showed that 1,25(OH)2D3 and AZA in combination significantly reduced the blast populations through increased cell death (62.4% combination, vs. 17.4% 1,25(OH)2D3 alone, and 30.9% AZA alone).

Fig. 1.

Fig. 1

Open in a new tab

Combined treatment with 1,25(OH)2D3 and AZA significantly reduces human AML blasts in vitro. Human AML patient-derived cell lines MOLM-14 cells and MOLM-14-TKIR were cultured with 1,25(OH)2D3 and AZA alone and in combination, or left untreated (control). 48 h later, cells were harvested and analyzed by FACS for expression of CD14 and viability dye. A Representative FACS plots of MOLM-14 cells show the viable blast population (CD14– viability dye–) under the indicated conditions. B FACS gating strategy for MOLM-14 cells. C Percentage of blasts cells within the MOLM-14 cells under indicated conditions. D, E MOLM-14 cells were collected after 48 h of treatment and analyzed by FACS for the expression of Ki67. Representative FACS plots show the viable Ki67+ blasts under indicated conditions. F, G MOLM-14 TKIR cells were collected after 48 h of treatment and analyzed by FACS for the expression of Ki67. Representative FACS plots show the viable Ki67+ blasts under indicated conditions. Where applicable, data are means ± SEM from each group and were analyzed by Student t-test. *p < 0.05; **p < 0.01; N = 6.

In addition to enhanced cytotoxicity, combination therapy also significantly reduced the proliferation of viable MOLM-14 blasts, as indicated by Ki-67 staining (see Fig. S3 for gating strategy), when compared to no treatment, 1,25(OH)2D3 alone and AZA alone (25.2% vs. 73.6%, 55.5% and 49.0%, respectively) (Fig. 1D, E). Similar results were obtained (Fig. 1F, G), with MOLM-14-TKIR, an acute myeloblastic leukemia cell line resistant to midostaurin (see Supplementary materials and methods for generation of this cell line). Collectively, our data suggest that the combination of 1,25(OH)2D3 and AZA is effective in enhancing cell death and inhibiting the proliferation of AML blasts in vitro.

Effects of combined treatment with 1,25(OH)2D3/AZA on primary AML and CD34+ hematopoietic stem cells (HSCs)

Next, we tested the combination therapy on primary AML blasts. Blasts were treated with different concentrations of 1,25(OH)2D3, AZA, and their combination for 48 h. FACS analyses demonstrated that combination therapy consistently reduced blasts more than either single therapy alone in patient A (Fig. 2A and B). Data from 7 primary leukemia patient samples and their mean reduction in blast counts are shown in Table 1 and Fig. 2D. Combination therapy was effective against patient samples (obtained from BM marrow aspirates) with different mutations, particularly the FLT-3 mutation (more than 3-fold reduction in blast counts, AZA vs. combination therapy, p < 0.05) (Fig. 2D, Table 1). Importantly, when using this combination to treat patient samples (initial blast population negative for CD34), we noticed that while the therapy suppressed the blast population, it did not affect the CD34+ population (Fig. 2C). It would be interesting in future experiments to determine whether this population is actually CD34 + HSCs or leukemic stem cells.

Fig. 2.

Fig. 2

Open in a new tab

Combined treatment with 1,25(OH)2D3 and AZA significantly reduces primary AML blasts ex vivo. A Human AML patient BM cells were cultured with 1,25(OH)2D3 and AZA as single agents and in combination. 48 h later, the cells were harvested and analyzed by FACS for expressions of CD117 and CD13 among viable cells. Representative FACS plots show viable blasts (CD13+CD117+ viability dye–) under the indicated conditions. B Percentages of CD13+ CD117+ AML blast cells under indicated conditions. C Percentages of CD34+ HSCs under indicated conditions (N = 3). D Therapeutic effect of 1,25(OH)2D3, AZA and their combination on AML patient samples after 48 h. Where applicable, data are means ± SEM from each group and were analyzed by Student t-test. *p < 0.05.

Table 1.

Primary blast percentage after AML treatment.

Demographic AML subtype
Cytogenetic/FISH		 Molecular marker Initial blast count (%) Blast count (%) after 1,25-VD3 (80 nM) Blast count (%) after 5-Azacytidine (5 μM) Blast count (%) after combination
30y F M1, normal +NPM1, +FLT3-ITD 14% 6.88% 14% 2.44%
47y M M4, normal +NPM1 36% 2.56% 19.8% 0.91%
50y F M5, normal +NPM1, +FLT3-ITD 19.3% 3.85% 21.8% 2.35%
32y F M4, inv.(16) Normal 71.2% 17.1% 28.2% 11.0%
53y F M2, t(8;21)(q22;q22); RUNX1-ETO +FLT3-TKD 20.8% 8.44% 1.3% 0.8%
A.2568 +FLT3-ITD 69.6% 58.7% 23.6% 19%
A.2431 +FLT3-ITD 69% 61.7% 44.9% 35.2%
Open in a new tab

Mechanism of action of combined treatment with 1,25(OH)2D3 and AZA

To understand the mechanism of 1,25(OH)2D3 based treatment for AML, in context of combination with AZA, we performed the RNA-Seq transcriptome profiling of different AML cells lines, which were pre-treated with either 1,25(OH)2D3 alone, AZA alone or their combination in vitro for 48 h (Fig. 3). Here, we report the RNA-Seq results from MOLM-14, MOLM-14-TKIR and THP-1.

Fig. 3.

Fig. 3

Open in a new tab

RNA-seq analysis of AML cells treated with combination therapy and single agents. A–F Selected RNA seq results were segregated into 3 categories, genes primarily acting on: 1) differentiation, 2) proliferation and 3) cell death. G, MOLM-14 cells were cultured with 1,25(OH)2D3 or AZA or a combination of 1,25(OH)2D3 and AZA for 48 h then harvested and analyzed G by RT-qPCR for expression of human VDR and H By Western blot for protein expression of human VDR (N = 3). Where applicable, data are means ± SEM from each group and were analyzed by Student t-test. *p < 0.05.

THP-1 had the lowest baseline expression of human VDR with FPKM (detailed descriptions in methods) approximately 3 times lower than MOLM-14 and MOLM-14-TKIR. In addition, THP-1's VDR expression did not change significantly after treatment with 1,25(OH)2D3 or combined 1,25(OH)2D3 and AZA. This was consistent with our FACS assay, showing that THP-1 had minimal therapeutic response (CD14+ differentiation) in response to 1,25(OH)2D3 or the combination of 1,25(OH)2D3 and AZA treatment in vitro (Fig. S1).

The combination of 1,25(OH)2D3 and AZA was most effective in up-regulating baseline VDR expression in both MOLM-14 cells (100% vs. 45% for AZA treatment alone) and MOLM-14-TKIR cells (87% vs. 24% for AZA treatment alone) (Fig. 3A). This result was confirmed by quantitative PCR (qPCR) experiments (2.4 fold increase, combination vs. no treatment, p < 0.05) and Western blot assays (6.6 fold increase, combination vs. no treatment, p < 0.05) (Fig. 3G, H).

RNA-seq data were consistent with our FACS assays, confirming that the combination of 1,25(OH)2D3 and AZA has enhanced efficacy against MOLM-14 and MOLM-14-TKIR cells.

We identified 3 major transcriptome changes that could explain the mechanisms of enhanced anti-leukemic activity of combination therapy: 1) Differentiation of blasts into mature monocytes with CD14+ expression (200 fold up-regulation for MOLM-14 cells, and even higher in MOLM-14-TKIR cells, Fig. 3B); 2) suppression of blasts proliferation by inhibiting FLT-3 expression (45% reduction in 1,25(OH)2D3, 134% reduction in the combination of 1,25(OH)2D3 and AZA for MOLM-14 cells, Fig. 3C) and a reduction in downstream PIM1 expression (up to 689% with the combination of 1,25(OH)2D3 and AZA vs. no treatment for MOLM-14 cells, Fig. 3C, D); 3) enhancing apoptosis of blasts by down-regulation of BCL-2 (484% reduction in the combination of 1,25(OH)2D3 and AZA for MOLM-14 cells, Fig. 3E) and up-regulation of BAX (56% in the combination of 1,25(OH)2D3 and AZA for MOLM-14 cells, Fig. 3F).

Collectively, our in vitro, ex vivo and RNA-seq data indicate that combining 1,25(OH)2D3 administration with AZA reduces the percentage of AML blast cells, through proliferation suppression, enhanced differentiation and cell death. It is important to recognize that 1,25(OH)2D3 based treatment will likely not be effective in blast cells with low baseline expression of VDR, as demonstrated in THP-1 cells.

Deficient expression of the CYP27B1 enzyme in the BM of MDS/AML patients

Given that CYP27B1 can be downregulated in tissue surrounding tumors [9], we asked whether CYP27B1 expression is also affected in BM of AML patients. We collected BM aspirates from both non-AML/MDS patients, and AML/MDS patients. Western blot data showed that CYP27B1 protein levels were significantly lower in AML/MDS patients' BM, as compared to controls (>2 fold, p < 0.05) (Fig. 4A). Next, we addressed whether suppressed expression of CYP27B1 is corrected after AML treatment. We found that induction chemotherapy treatment augmented the expression of CYP27B1 in day 30 BM when the patient achieved remission (Fig. 4B).

Fig. 4.

Fig. 4

Open in a new tab

Endogenous and ectopically expressed CYP27B1 enzyme in Human AML BM. A Protein levels of CYP27B1 in human BM samples were determined by Western blot analyses. Levels from non-AML/MDS patients (left), and AML/MDS patients (right) are shown (N = 5 in AML/MDS patients, and N = 8 in control patients). Data are means ± SEM from each group and were analyzed by Student t-test. *p < 0.05 vs. control (normal). B CYP27B1 protein levels in AML patient BM before and after chemotherapy induction. C Schematic of experimental design. On day −1, NRG mice were intraperitoneally injected with AZA. On day 0, 1 × 106 CYP27B1-MOLM-14 cells were transplanted through tail vein injection into the experimental group. In the control group, NRG mice received 1 × 106 GFP-MOLM-14 on day 0. On day 14, mice were euthanized for histological analyses. The CYP27B1-GFP lenti-viral vector and virus used in this study and stable expression in the CYP27B1-MOLM-14 cell line are shown in Fig. S3. D Immunohistology was performed to confirm homing of GFP+ engrafts inside the femur BM (upper image). Double staining shows co-localization of CYP27B1 expression (red color) and GFP+ engrafts (lower images). Where applicable, data are means ± SEM from each group and were analyzed by Student t-test. *p < 0.05; (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

CYP27B1-MOLM14 cells show stable transgene expression and function in vitro and in vivo

FACS was performed to confirm the generation of a CYP27B1-MOLM-14 cell line by analyzing the GFP expression, which displayed 98.8% GFP+ in the CYP27B1-MOLM-14 cell line comparing to the non-GFP+ naïve MOLM-14 (Fig. S4). We confirmed the function of CYP27B1 in generating 1,25(OH)2D3 in vitro by measuring 1,25(OH)2D3 level after adding 25(OH)D3 substrate into cell cultures with CYP27B1-GFP-MOLM-14 for three days (Table S2). We determined that CYP27B1-transfected cells generated more 1,25(OH)2D3 than GFP-MOLM-14 by adding 50 nM of 25(OH)D3 into cell cultures of MOLM-14-CYP27B1 and GFP-MOLM-14 (Fig. S5).

The first question is where these vehicle cells will localize after engraftment and the second question is whether they will continuously express the CYP27B1 transgene in vivo. To address these questions, we transplanted MOLM-14 cells engineered to express CYP27B1, and followed these engrafted cells in vivo. For this AML xenograft model, a dose of AZA was injected IP into NRG mice on day 0 to ablate the BM. On day 1, half a million of either GFP-MOLM-14 cells or CYP27B1-GFP-MOLM-14 cells were injected into mice via tail vein. GFP-labeled MOLM-14 cells served as the vector control group. On day 14, mice were euthanized and histology was performed on femur BM (Fig. 4C). Under fluorescent microscopy, femur BM was occupied with GFP positive MOLM-14 cells, which were also stained with CYP27B1 antibodies (Fig. 4D). Immunohistochemistry showed that many co-localized cells stained positively for both green (GFP) and red (CYP) in the CYP27B1-GFP-MOLM-14 group. As expected, GFP-MOLM-14 control cells on the bottom panels only stained for green (GFP). These data demonstrated that CYP27B1-MOLM-14 cells are functional in localizing to the BM and in persistently expressing 1α-hydroxylase to generate active Vitamin D.

In vivo effects of CYP27B1

Above we established proof of principle for delivery of CYP27B1 via transduced MOLM-14. Next, we determined whether expressing the transgene CYP27B1 in MOLM-14 blasts improves the survival of MOLM-14 xenograft mice. In our first survival preclinical model, we performed IP injections on 4 groups of mice with AZA on day 0 to condition the BM and improve engraftment, and on day 1 they were treated as follows: 1) 5 × 105 GFP-MOLM-14 cells, 2) 5 × 105 CYP27B1-GFP-MOLM-14 cells, 3) 2 × 106 CYP27B1-GFP-MOLM-14 cells, and 4) 4 × 106 CYP27B1-GFP-MOLM-14 cells. We observed mice for physiologic changes and their median overall survival was compared between the four treatment groups using Kaplan Meier curve. BM cells harvested from freshly sacrificed mice were analyzed by FACS. Of AML xenografts in group 1, 90% experienced lower body paralyses, and significant weight loss, in contrast to the healthy CYP27B1-treated group 2 (Fig. S6 and videos in Fig. S7). Median survival for group 2 was significantly prolonged compared to group 1 (29.4 days vs. 24.75 days, p < 0.01) (Fig. 5A). While injecting more engineered MOLM-14 cells would lead to increase local expression of CYP27B1, it would certainly increase the leukemic burden in the BM and eventually would lead to early demise of AML mice as observed in group 3 and 4.

Next, we examined the fate of transplanted AML blasts inside the mouse BM. FACS showed that cells collected from the BM of group 2 at day 14 after transplantation had increased expression of CD14 compared to group 1 (7.44% vs. 0.92%, p < 0.05) (Fig. 5B). IHC demonstrated that some transplanted GFP+ blasts had started to express CD14 (red color, Fig. 5C).

To assess for hypercalcemia, we obtained peripheral blood from each group of mice at baseline, day 7 and day 14, and measured calcium levels using the Calcium Colorimetric Assay Kit. No significant changes were observed in calcium levels in any of the groups at day 7 or 14 (Fig. 5D).

Discussion

Here we provide evidences for the use of gene therapy to deliver 1,25(OH)2D3 therapy while overcoming systemic hypercalcemia. We have demonstrated that although a high concentration of 1,25(OH)2D3was effective at inducing MOLM-14 and HL-60 to undergo differentiation, cells with low baseline VDR (THP-1) did not respond to 1,25(OH)2D3. Mutations leading to AML are heterogeneous, thus this could possibly explain why 1,25(OH)2D3 is effective against some AML cell lines and patient samples, but not others.

Low expression of VDR mRNA before and after treatment with 1,25(OH)2D3correlated with a lack of downstream expression of genes that control proliferation and cell death. This could possibly be a potential reason for mixed responses to 1,25(OH)2D3treatment for AML in many clinical trials due to the varying expression of baseline VDR of leukemic blasts [12,34,35]. Interestingly, with combination treatment of 1,25(OH)2D3and AZA we noticed a significant decrease in mRNA expression of the FLT-3 and PIM-1 genes in both MOLM-14 and MOLM14-TKIR. Constitutively active FLT3 signaling up-regulates PIM-1 expression via the STAT5 pathway resulting in phosphorylation of BAD protein. Silencing PIM-1 sensitizes resistant cells to FLT3 inhibitors [36]. This has important clinical implications because treating resistant FLT-3 cells with 1,25(OH)2D3could potentially resensitize them to TKI. Another important implication is that 1,25(OH)2D3can potentially be combined with commercially approved FLT-3 and Bcl-2 inhibitor for synergistic effects in AML treatment. The use of 1,25(OH)2D3should be selective and will likely not work for AML cells with low baseline expression of VDR as demonstrated above. Our finding is consistent with recent publication from Paubelle et al. that VDR is important for myeloid progenitor differentiation and is a prognostic factor in AML. Further, mice with deficient VDR have increased numbers of hematopoietic and leukemia stem cells [37].

This report is the first to identify a significant decrease in CYP27B1 protein levels in the BM aspirate of MDS/AML patients vs. non-MDS/AML patients and to show that it increases in the BM aspirate in an AML patient after 7 + 3 induction chemotherapy.

As mentioned before, supra-physiologic treatment with 1,25(OH)2D3results in systemic hypercalcemia. Gene therapy allows us to bypass this limitation without compromising treatment efficacy. Further, CYP27B1 gene therapy also serves to replenish the depleted 1-alpha-hydroxylase in the bone marrow of AML mice. We demonstrated that treatment for gene therapy experimental arms were not limited by systemic hypercalcemia, but were limited by the number of engineered AML cells we can inject due to concern for increased blast burden. Another important factor to be considered in future experiments is 25(OH)D3 substrate, which can be depleted with repeated, interval gene therapy treatment. Future experiments will involve a more clinically suitable vehicle cell such as HSCs. Our lab has successfully generated CYP27B1-GFP CD34 + HSC cells that constantly expressed CYP27B1 gene ex vivo (data not shown). Without concern for hypercalcemia or increase in AML burden by using the appropriate vehicle cells, we can repeatedly inject engineered HSCs on a regular interval. Oral supplement with 25(OH)D3 would also be considered. We envision a clinical scenario, where an older patient will receive a dose of autologous, engineered HSCs after several doses of palliative chemotherapy to reduce blast burdens.

In summary, our data provides strong in vitro and ex vivo evidence to support combination therapy of 1,25(OH)2D3 and AZA for AML treatment. For future experiments, we will utilize a more suitable vehicle such as CD34 + HSC and add AZA, a hypomethylating agent, for potential synergistic effect.

The following are the supplementary data related to this article.

Supplementary material

mmc1.docx (429KB, docx)

Supplementary figure. Original film images of Western blot.

A: Film image for the VDR expression for Fig. 3H.

B: Film image for the CYP27B1 expression for Fig. 4A.

C: Film image for the CYP27B1 expression for Fig. 4B.

mmc2.pptx (406.5KB, pptx)
Supplemental Fig. 7

Videos showing improved functional activities in CYP27B1-MOLM14 injected NRG mice on Day21.

Download video file (3MB, flv)
Download video file (2.9MB, flv)

CRediT authorship contribution statement

Huynh Cao: conceptualization, methodology, supervision, writing – review and editing; Yi Xu: methodology, validation, investigation, formal analysis, writing – original draft; Linh Pham: investigation, writing – original draft; Park Eunwoo: investigation, writing – original manuscript; Jeffrey Xiao – investigation; David Chi: investigation; Justin Lyu: investigation; Rosalia Campion: formal analysis, investigation; Samiksha Wasnik: investigation, resources; Il Seok Jeong: software; Xiaolei Tang: conceptualization; David Jeston Baylink: conceptualization, supervision, funding acquisition; Chien Shing Chen: conceptualization, supervision; Mark Reeves: supervision, writing – review and editing; Kimberly Payne: supervision, writing – original draft; Mojtaba Akhtari: writing – review and editing; Saied Mirshahidi: resources; Guido Marcucci: supervision, writing – review and editing.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

This work was supported by the American Society of Clinical Oncology Young Investigator Award and Loma Linda University Research Innovation Grant (to HC), and California Institute for Regenerative Medicine (to DB, HC). We would like to acknowledge the Loma Linda University Cancer Center Biospecimen Laboratory for providing patients' bone marrow aspirates.

References

  • 1.Estey E., Dohner H. Acute myeloid leukaemia. Lancet. 2006;368(9550):1894–1907. doi: 10.1016/S0140-6736(06)69780-8. [DOI] [PubMed] [Google Scholar]
  • 2.Tenen D.G. Disruption of differentiation in human cancer: AML shows the way. Nat. Rev. Cancer. 2003;3(2):89–101. doi: 10.1038/nrc989. [DOI] [PubMed] [Google Scholar]
  • 3.Rots N.Y. Induced differentiation of U937 cells by 1,25-dihydroxyvitamin D3 involves cell cycle arrest in G1 that is preceded by a transient proliferative burst and an increase in cyclin expression. Blood. 1999;93(8):2721–2729. [PubMed] [Google Scholar]
  • 4.Devesa S.S. Cancer surveillance series: changing geographic patterns of lung cancer mortality in the United States, 1950 through 1994. J. Natl. Cancer Inst. 1999;91(12):1040–1050. doi: 10.1093/jnci/91.12.1040. [DOI] [PubMed] [Google Scholar]
  • 5.Dohner H., Weisdorf D.J., Bloomfield C.D. Acute myeloid leukemia. N. Engl. J. Med. 2015;373(12):1136–1152. doi: 10.1056/NEJMra1406184. [DOI] [PubMed] [Google Scholar]
  • 6.DiNardo C.D. Venetoclax combined with decitabine or azacitidine in treatment-naive, elderly patients with acute myeloid leukemia. Blood. 2019;133(1):7–17. doi: 10.1182/blood-2018-08-868752. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Holohan C. Cancer drug resistance: an evolving paradigm. Nat. Rev. Cancer. 2013;13(10):714–726. doi: 10.1038/nrc3599. [DOI] [PubMed] [Google Scholar]
  • 8.Holick M.F. Sunlight and vitamin D for bone health and prevention of autoimmune diseases, cancers, and cardiovascular disease. Am. J. Clin. Nutr. 2004;80(6 Suppl):1678S–1688S. doi: 10.1093/ajcn/80.6.1678S. [DOI] [PubMed] [Google Scholar]
  • 9.Ogunkolade B.W. Expression of 25-hydroxyvitamin D-1-alpha-hydroxylase mRNA in individuals with colorectal cancer. Lancet. 2002;359(9320):1831–1832. doi: 10.1016/s0140-6736(02)08680-4. [DOI] [PubMed] [Google Scholar]
  • 10.Arbour N.C., Prahl J.M., DeLuca H.F. Stabilization of the vitamin D receptor in rat osteosarcoma cells through the action of 1,25-dihydroxyvitamin D3. Mol. Endocrinol. 1993;7(10):1307–1312. doi: 10.1210/mend.7.10.8264662. [DOI] [PubMed] [Google Scholar]
  • 11.Racz A., Barsony J. Hormone-dependent translocation of vitamin D receptors is linked to transactivation. J. Biol. Chem. 1999;274(27):19352–19360. doi: 10.1074/jbc.274.27.19352. [DOI] [PubMed] [Google Scholar]
  • 12.Marchwicka A. Perspectives of differentiation therapies of acute myeloid leukemia: the search for the molecular basis of patients' variable responses to 1,25-dihydroxyvitamin d and vitamin d analogs. Front. Oncol. 2014;4:125. doi: 10.3389/fonc.2014.00125. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Kahlen J.P., Carlberg C. Identification of a vitamin D receptor homodimer-type response element in the rat calcitriol 24-hydroxylase gene promoter. Biochem. Biophys. Res. Commun. 1994;202(3):1366–1372. doi: 10.1006/bbrc.1994.2081. [DOI] [PubMed] [Google Scholar]
  • 14.Vaisanen S. Spatio-temporal activation of chromatin on the human CYP24 gene promoter in the presence of 1alpha,25-Dihydroxyvitamin D3. J. Mol. Biol. 2005;350(1):65–77. doi: 10.1016/j.jmb.2005.04.057. [DOI] [PubMed] [Google Scholar]
  • 15.Hmama Z. 1alpha,25-dihydroxyvitamin D(3)-induced myeloid cell differentiation is regulated by a vitamin D receptor-phosphatidylinositol 3-kinase signaling complex. J. Exp. Med. 1999;190(11):1583–1594. doi: 10.1084/jem.190.11.1583. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Marcinkowska E., Kutner A. Side-chain modified vitamin D analogs require activation of both PI 3-K and erk1,2 signal transduction pathways to induce differentiation of human promyelocytic leukemia cells. Acta Biochim. Pol. 2002;49(2):393–406. [PubMed] [Google Scholar]
  • 17.Hughes P.J. The vitamin D receptor-mediated activation of phosphatidylinositol 3-kinase (PI3Kalpha) plays a role in the 1alpha,25-dihydroxyvitamin D3-stimulated increase in steroid sulphatase activity in myeloid leukaemic cell lines. J. Cell. Biochem. 2008;103(5):1551–1572. doi: 10.1002/jcb.21545. [DOI] [PubMed] [Google Scholar]
  • 18.Zhang Y., Zhang J., Studzinski G.P. AKT pathway is activated by 1, 25-dihydroxyvitamin D3 and participates in its anti-apoptotic effect and cell cycle control in differentiating HL60 cells. Cell Cycle. 2006;5(4):447–451. doi: 10.4161/cc.5.4.2467. [DOI] [PubMed] [Google Scholar]
  • 19.Liu M. Transcriptional activation of the Cdk inhibitor p21 by vitamin D3 leads to the induced differentiation of the myelomonocytic cell line U937. Genes Dev. 1996;10(2):142–153. doi: 10.1101/gad.10.2.142. [DOI] [PubMed] [Google Scholar]
  • 20.Hughes P.J. Vitamin D3-driven signals for myeloid cell differentiation—implications for differentiation therapy. Leuk. Res. 2010;34(5):553–565. doi: 10.1016/j.leukres.2009.09.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Kang H.K. Up-regulation of phospholipase Cgamma1 and phospholipase D during the differentiation of human monocytes to dendritic cells. Int. Immunopharmacol. 2004;4(7):911–920. doi: 10.1016/j.intimp.2004.04.001. [DOI] [PubMed] [Google Scholar]
  • 22.Lopez-Lluch G. Inhibition of COX activity by NSAIDs or ascorbate increases cAMP levels and enhances differentiation in 1alpha,25-dihydroxyvitamin D3-induced HL-60 cells. Arch. Biochem. Biophys. 2005;436(1):32–39. doi: 10.1016/j.abb.2004.12.031. [DOI] [PubMed] [Google Scholar]
  • 23.Pakkala S. Vitamin D3 analogs: effect on leukemic clonal growth and differentiation, and on serum calcium levels. Leuk. Res. 1995;19(1):65–72. doi: 10.1016/0145-2126(94)00065-i. [DOI] [PubMed] [Google Scholar]
  • 24.Petrich A. Phase II study of doxercalciferol for the treatment of myelodysplastic syndrome. Leuk. Lymphoma. 2008;49(1):57–61. doi: 10.1080/10428190701713648. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Yoshida Y. A randomized study of alfacalcidol in the refractory myelodysplastic anaemias. A Japanese cooperative study. Int. J. Clin. Pharmacol. Res. 1993;13(1):21–27. [PubMed] [Google Scholar]
  • 26.Doig C.L. Recruitment of NCOR1 to VDR target genes is enhanced in prostate cancer cells and associates with altered DNA methylation patterns. Carcinogenesis. 2013;34(2):248–256. doi: 10.1093/carcin/bgs331. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Marik R. DNA methylation-related vitamin D receptor insensitivity in breast cancer. Cancer Biol. Ther. 2010;10(1):44–53. doi: 10.4161/cbt.10.1.11994. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Thorne J.L. Epigenetic control of a VDR-governed feed-forward loop that regulates p21(waf1/cip1) expression and function in non-malignant prostate cells. Nucleic Acids Res. 2011;39(6):2045–2056. doi: 10.1093/nar/gkq875. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Khanim F.L. Altered SMRT levels disrupt vitamin D3 receptor signalling in prostate cancer cells. Oncogene. 2004;23(40):6712–6725. doi: 10.1038/sj.onc.1207772. [DOI] [PubMed] [Google Scholar]
  • 30.Rashid S.F. Synergistic growth inhibition of prostate cancer cells by 1 alpha,25 Dihydroxyvitamin D(3) and its 19-nor-hexafluoride analogs in combination with either sodium butyrate or trichostatin A. Oncogene. 2001;20(15):1860–1872. doi: 10.1038/sj.onc.1204269. [DOI] [PubMed] [Google Scholar]
  • 31.Koschmieder S. Decitabine and vitamin D3 differentially affect hematopoietic transcription factors to induce monocytic differentiation. Int. J. Oncol. 2007;30(2):349–355. [PubMed] [Google Scholar]
  • 32.Cao H. Application of vitamin D and vitamin D analogs in acute myelogenous leukemia. Exp. Hematol. 2017;50:1–12. doi: 10.1016/j.exphem.2017.01.007. [DOI] [PubMed] [Google Scholar]
  • 33.Fetahu I.S., Hobaus J., Kallay E. Vitamin D and the epigenome. Front. Physiol. 2014;5:164. doi: 10.3389/fphys.2014.00164. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Kim M. Application of vitamin D and derivatives in hematological malignancies. Cancer Lett. 2012;319(1):8–22. doi: 10.1016/j.canlet.2011.10.026. [DOI] [PubMed] [Google Scholar]
  • 35.Zhou J.Y. 1,25-Dihydroxy-16-ene-23-yne-vitamin D3 prolongs survival time of leukemic mice. Proc. Natl. Acad. Sci. U. S. A. 1990;87(10):3929–3932. doi: 10.1073/pnas.87.10.3929. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Zhou J., Chng W.J. Resistance to FLT3 inhibitors in acute myeloid leukemia: molecular mechanisms and resensitizing strategies. World J. Clin. Oncol. 2018;9(5):90–97. doi: 10.5306/wjco.v9.i5.90. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Paubelle E. Vitamin D receptor controls cell stemness in acute myeloid leukemia and in normal bone marrow. Cell Rep. 2020;30(3):739–754. doi: 10.1016/j.celrep.2019.12.055. (e4) [DOI] [PubMed] [Google Scholar]

Associated Data

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

Supplementary Materials

Supplementary material

mmc1.docx (429KB, docx)

Supplementary figure. Original film images of Western blot.

A: Film image for the VDR expression for Fig. 3H.

B: Film image for the CYP27B1 expression for Fig. 4A.

C: Film image for the CYP27B1 expression for Fig. 4B.

mmc2.pptx (406.5KB, pptx)
Supplemental Fig. 7

Videos showing improved functional activities in CYP27B1-MOLM14 injected NRG mice on Day21.

Download video file (3MB, flv)
Download video file (2.9MB, flv)

Articles from Translational Oncology are provided here courtesy of Neoplasia Press

RESOURCES