Human cathelicidin antimicrobial peptide (CAMP) gene is a direct target of the vitamin D receptor and is strongly up-regulated in myeloid cells by 1,25-dihydroxyvitamin D3
- *Department of Medicine, Division of Hematology/Oncology, Cedars-Sinai Medical Center, David Geffen School of Medicine at UCLA, Los Angeles, California, USA; and
- †The Granulocyte Research Laboratory, Department of Hematology, Rigshospitalet, University of Copenhagen, Copenhagen, Denmark
- ↵Correspondence: 1 Correspondence: Division of Hematology/Oncology, Cedars-Sinai Medical Center, Davis Bldg. 5019, 8700 Beverly Blvd., Los Angeles, CA 90048, USA. E-mail: gombarta{at}csmc.edu
Abstract
The innate immune system of mammals provides a rapid response to repel assaults from numerous infectious agents including bacteria, viruses, fungi, and parasites. A major component of this system is a diverse combination of cationic antimicrobial peptides that include the α- and β-defensins and cathelicidins. In this study, we show that 1,25-dihydroxyvitamin D3 and three of its analogs induced expression of the human cathelicidin antimicrobial peptide (CAMP) gene. This induction was observed in acute myeloid leukemia (AML), immortalized keratinocyte, and colon cancer cell lines, as well as normal human bone marrow (BM) -derived macrophages and fresh BM cells from two normal individuals and one AML patient. The induction occurred via a consensus vitamin D response element (VDRE) in the CAMP promoter that was bound by the vitamin D receptor (VDR). Induction of CAMP in murine cells was not observed and expression of CAMP mRNA in murine VDR-deficient bone marrow was similar to wild-type levels. Comparison of mammalian genomes revealed evolutionary conservation of the VDRE in a short interspersed nuclear element or SINE in the CAMP promoter of primates that was absent in the mouse, rat, and canine genomes. Our findings reveal a novel activity of 1,25-dihydroxyvitamin D3 and the VDR in regulation of primate innate immunity.—Gombart, A. F., Borregaard, N., Koeffler, H. P. Human cathelicidin antimicrobial peptide (CAMP) gene is a direct target of the vitamin D receptor and is strongly up-regulated in myeloid cells by 1,25-dihydroxyvitamin D3.
A major concern for public health in both developed and developing countries is the alarming increase of antibiotic resistance in bacteria (1)⇓. Drug-resistant bacteria such as Pseudomonas aeruginosa and Staphylococcus aureus pose serious problems for immunocompromised persons and are major sources of life-threatening nosocomial infections. In 2000, nearly 660,000 cases of sepsis developed in the United States. This resulted in an in-hospital mortality rate of nearly 18% (2)⇓. Among survivors of sepsis, an increased risk of death and decreased quality of life occurred after discharge from the hospital (3⇓, 4)⇓.
This impending crisis has spurred the search for new therapeutic agents to combat antibiotic resistance. One potential solution lies within a system all animals are “born with,” the innate immune system responsible for keeping us healthy (5)⇓. It provides animals the capacity to repel assaults quickly from numerous infectious agents including bacteria, viruses, fungi, and parasites (6⇓7⇓8⇓9⇓10⇓11)⇓. Diverse combinations of cationic antimicrobial peptides (AMPs) including α- and β-defensins and cathelicidins comprise a major component of this defense in mammals. Because bacteria have difficulty developing resistance against AMPs and are quickly killed by them, this class of antimicrobial agents is being commercially developed as a source of peptide antibiotics (1⇓, 12⇓, 13)⇓. The majority of the pharmaceutical effort has concentrated on the development of topically applied agents (13)⇓. The expense and difficulty of preparing large amounts of peptide and the uncertainty in systemic use of these peptides have slowed their development beyond topical treatments.
One AMP that shows promise is the human cathelicidin antimicrobial peptide (CAMP), also known as hCAP18/LL-37/FALL-39. It is the only known human cathelicidin. The cathelicidins are a family of proteins consisting of a C-terminal cationic AMP domain that is activated by cleavage from the N-terminal cathelin portion of the propeptide. The majority of the CAMP propeptide is stored in secondary or specific granules of neutrophils from which it can be released at sites of microbial infection (14)⇓. In addition to neutrophils, various white blood cell populations express hCAP18. These include natural killer cells, γδT cells, B cells, monocytes (15)⇓, and mast cells (16)⇓. CAMP/hCAP18 is secreted into the blood and significant levels are found in the plasma (17)⇓.
CAMP is synthesized and secreted in significant amounts by those tissues that are exposed to environmental microbes. This includes the squamous epithelia of the mouth, tongue, esophagus, lungs, intestine, cervix, and vagina (18⇓, 19)⇓. In addition, it is produced by salivary and sweat glands (20⇓, 20)⇓, epididymis, testis (21)⇓, and mammary glands (22⇓23⇓24)⇓. Expression in these tissues results in secretion of the polypeptide in wounds (25)⇓, sweat (26)⇓, airway surface fluids (19)⇓, seminal plasma (27)⇓, and milk (22⇓, 23)⇓. CAMP/hCAP18 possesses several important activities including bactericidal, anti-sepsis, chemoattraction, and promotion of angiogenesis and wound healing. The possibility of extrinsically manipulating endogenous expression of CAMP for systemic and localized therapeutic benefit is very attractive.
Since their discovery more than a decade ago, the majority of expression studies have been focused on the detection of cathelicidins in various tissues; however, the transcriptional mechanisms that regulate cathelicidin gene expression have not been adequately elucidated. Understanding the signaling pathways and the downstream transcription factors that regulate CAMP gene expression in a tissue-specific manner is crucial for designing approaches for therapeutic manipulation of endogenous gene expression. Because AMPs serve a role in host defense and may act as mediators of other biological processes, their expression is tightly regulated.
The experimental focus of this study was to identify extracellular signals and the downstream transcription factors that activate transcription of the CAMP gene, with the ultimate goal of extrinsically manipulating its endogenous expression for systemic and localized therapeutic benefit. We provide evidence that the CAMP gene is a direct target of the transcription factor vitamin D receptor (VDR) that mediates the strong up-regulation of CAMP in response to treatment of cells with 1,25-dihydroxyvitamin D3 [1,25(OH)2D3 or vitamin D3] and its analogs. Induction of the endogenous CAMP by these relatively safe (FDA approved) compounds may provide important novel therapeutic uses from promotion of wound healing to protection against bacteremia and sepsis after surgery, chemotherapy, or severe burns.
MATERIALS AND METHODS
Tissue culture and reporter assays
The human myeloid leukemia cell lines U937, NB4, HL60, and ML1 were cultured in RPMI1640 (Invitrogen, Carlsbad, CA, USA) containing 10% fetal calf serum (FCS; Omega Scientific, Inc., Tarzana, CA, USA). Human bone marrow cells isolated from either two normal or one acute myeloid leukemia patient were cultured in RPMI1640 containing 10% FCS for short-term experiments. Bone marrow (BM) -derived macrophages (Mφ) were obtained by culturing normal human bone marrow (NHBM) cells in RPMI1640 containing 10% FCS, 200 ng/mL GM-CSF, and 5% WeHi-3B conditioned medium (source of IL-3) for 14 days. The bone marrow samples were obtained from patients after informed consent was given. Approval for the collection of these samples was obtained from the Cedars-Sinai Medical Center Institutional Review Board. The immortalized keratinocyte cell line HaCat (a kind gift from Dr. Norbert Fusenig, Heidelberg, Germany) and colon cancer cell line HT29 were cultured in DMEM containing 10% FCS. All media were supplemented with antibiotics (100 units penicillin/streptomycin; Invitrogen). Cells were treated with various concentrations and durations of 1,25(OH)2D3, a vitamin D3 analog, or vehicle (ethanol). The 1,25(OH)2D3 and compound I (1,25R,26-(OH)3-22-ene-D3) were synthesized and generously provided by Dr. Milan Uskokovic at Hoffmann-LaRoche, Inc. (Nutley, NJ, USA). Analogs KH1060 (20-epi-22oxa-24a,26a,27a-tri-homo-1,25(OH)2D3) and EB1089 (1,25-dihydroxy-22,24-diene, 24,26,27-trihomo) were synthesized by Leo Pharmaceutical Products (Ballerup, Denmark) and generously provided by Dr. Lise Binderup. U937 cells were treated for 24 h with vehicle (ethanol), LPS (1 μg/mL), 12-O-tetradecanoylphorbol 13-acetate (TPA, 10 ng/mL), TNF-α (1 ng/mL), INF-α (10 ng/mL), IFNγ (50 ng/mL), IL-2 (2.5 ng/mL), IL-6 (10 ng/mL), GM-CSF (1 ng/mL), G-CSF (60 ng/mL), estradiol (1×10−8 M), dihydrotestosterone (DHT, 1×10−8 M), or all-trans retinoic acid (ATRA, 5×10−7 M). Cyclohexamide (Sigma, St. Louis, MO, USA) was used at 20 μg/mL and the absence of protein synthesis was determined by measuring 35S-methionine incorporation. Cyclohexaminde was added 30 min before the vehicle or 1,25(OH)2D3. Actinomycin D (Sigma) was used at 10 μg/mL and added at the same time as vehicle or 1,25(OH)2D3.
Murine 32Dcl3 cells (a generous gift from Alan Friedman, Johns-Hopkins, Baltimore, MD, USA) were cultured in IMDM (Invitrogen) supplemented with 10% FCS and 10% Wehi3B-conditioned medium. Cells were treated with 1,25-dihydroxyvitamin D3 or ethanol for 0, 24, and 48 h and total RNA was harvested. The 1,25(OH)2D3 and compound I (both 0.05 μg/mouse) were administered to beige/nude/x-linked (bnx) nu/nu nude mice every 2 days for 6 wk. The bone marrow cells were flushed from the femurs and total RNA was isolated. Bone marrow cells were flushed from femurs of VDR-deficient mice or wild-type littermates (28)⇓. Red blood cells were lysed and cells were plated in IMDM supplemented with 10% FCS. Cells were treated with 1,25(OH)2D3 or ethanol for 24 h and total RNA was harvested. BM-derived macrophages were obtained from VDR-deficient and wild-type murine femurs as described previously (29)⇓. Cells were treated with ethanol or 1,25(OH)2D3 for 0, 24, and 48 h and total RNA was harvested.
U937 cells were electroporated using a BTX T820 (Genetronics Biomedical, Ltd., San Diego, CA, USA). The settings were low voltage, 200 V, 10 ms, 1 pulse in 250 μL of cells at 2 × 107 cells/mL in a 4 mm cuvette. A total of 20 μg plasmid was used per transfection. After transfection, cells were treated with 1,25(OH)2D3 or vehicle at the concentration and times indicated in the figure legends. Cell lysates were prepared and luciferase activities determined using the dual luciferase assay system as described by the manufacturer (Promega, Madison, WI, USA). Transfection efficiency was normalized to the renilla luciferase expression vector phTKRL (Promega).
Recombinant plasmids
Primers 5′-CCGACGCGTCATACTGAGTCTCACTCTGTTACC-3′ and 5′-CCGCTCGAGGGTCCCCATGTCTGCCTC-3′ were used to amplify the human CAMP promoter (nucleotides –693 to +14) from human genomic DNA (30)⇓. This fragment was subcloned into the firefly luciferase reporter plasmid pXP2 (31)⇓ and called pXP2-CAMP-Luc. Subsequently deletion mutants pXP2-CAMP(ΔSmaI)-Luc and pXP2-CAMP(ΔHindIII)-Luc were generated by restriction enzyme digestion, fill-in, and religation of the purified linear plasmid. Constructs were verified by nucleotide sequencing.
Analysis of RNA and protein expression
Total RNA was prepared using Trizol Reagent (Invitrogen), electrophoresed through a formaldehyde-containing 1% agarose gel, and transferred to a positively charged nylon membrane (Hybond N+) for Northern analysis (Amersham Pharmacia Biotech, Piscataway, NJ, USA). The blots were sequentially probed with 32P-labeled DNA probes (Strip-EZ™, Ambion, Inc., Austin, TX, USA) specific for the CAMP, CDllb, and β-actin mRNAs.
For quantitative real-time PCR (QRT-PCR), total RNA was prepared, treated with DnaseI (Invitrogen), and cDNAs were synthesized by reverse transcription using Superscript II reverse transcriptase as described by the manufacturer (Invitrogen). The cDNAs were then analyzed by QRT-PCR using a fluorescent probe (Applied Biosystems, Foster City, CA, USA) against CAMP (5′-6fam-ACCCCAGGCCCACGATGGAT-tamra-3′) or 18S (32)⇓ at a final concentration of 200 nM per reaction. Primers against CAMP (forward, 5′-GCTAACCTCTACCGCCTCCT-3′ and reverse, 5′-GGTCACTGTCCCCATACACC-3′) or 18S (32)⇓ were used at 600 nM per reaction. PCR was performed using HotMaster™ Taq polymerase (Eppendorf AG, Hamburg, Germany) on an iCycler PCR machine equipped with an optical module (Bio-Rad Laboratories, Hercules, CA, USA). The protocol was 95°C, 1 min followed by 45 cycles of 95°C, 15 s and 60°C, 1 min, during which time data were collected. Standard curves were generated by PCR using serial dilutions of known quantities of CAMP or 18S cDNA and were included on each plate to quantify the ng of CAMP or ng of 18S cDNA in each sample. PCR was performed in triplicate for each sample.
Primers against murine CAMP/CRAMP (forward, 5′-GCAGTTCCAGAGGGACGTC-3′ and reverse, 5′-GTTCCTTGAAGGCACATTGC-3′) were used at 200 nM per reaction. PCR was performed using SYBR green (Molecular Probes, Eugene, OR, USA) as described previously (32)⇓. The protocol was 95°C, 1 min followed by 45 cycles of 95°C, 15 s; 60°C, 30 s; and 65°C, 1 min, during which time data were collected. The relative fold change between samples was determined using data normalized for 18S expression. Samples were analyzed in triplicate.
Western blot and immunfluorescent microscopy analyses were performed essentially as described previously (33)⇓. The total cell lysates were electrophoresed through 20% polyacrylamide-SDS gels. The hCAP18 antibody was used at 2.0 μg/mL for Western blot analysis and 4.0 μg/mL for immunofluorescent (IF) microscopy (14)⇓. The anti-GAPDH monoclonal antibody was used at a 1:10,000 dilution (Research Diagnostics, Inc., Flanders, NJ, USA).
Chromatin immunoprecipitation (ChIP) assays
The ChIP assays were performed essentially as described by the manufacturer (Upstate, Inc., Chalottesville, VA, USA). Briefly, ∼1 × 107 cells were incubated with vehicle or 1,25-dihydroxyvitamin D3 (1×10−7 M for 4 h). Protein/DNA complexes were cross-linked in 1% formaldehyde for 10 min. The reaction was terminated with the addition of glycine to 0.125 M final concentration. The cells were washed in ice-cold PBS containing PMSF (10 μg/mL), resuspended in 1 mL of SDS-lysis buffer containing protease inhibitors, and incubated on ice for 10 min. The lysates were sonicated 3×, 10 s at 30% output to shear the DNA. The sonicated lysate was pelleted at 13K rpm for 10 min at 4°C. Supernatant (0.2 mL) was mixed with 1.8 mL of dilution buffer and precleared with protein A-agarose for 1 h on ice. Antibody (2 μg) against VDR (mixed SC-1008 [1 μg] and SC-1009 [1 μg], Santa Cruz Biotechnology, Santa Cruz, CA, USA), C/EBPε (2 μL) (34)⇓, preimmune serum, or no antibody was added and the samples incubated overnight at 4°C. A slurry of ssDNA/protein A agarose was added and the mixture was incubated with rocking overnight at 4°C. The agarose/antibody/protein/DNA complex was pelleted and washed in low salt (1×), high salt (1×), LiCl (1×), and TE (2×). The complex was removed from the protein A-agarose in elution buffer (2×500 μL); cross-links were reversed in 100 mM NaCl at 65°C for 4 h, proteinase K treated, phenol/chloroform extracted, and ethanol precipitated. The promoter fragment was detected by PCR using primers against the CAMP promoter (forward, 5′-ACCGTGCCCTGCCTCATTC-3′ and reverse, 5′-TGGTCCCCATGTCTGCCTC-3′). The 430 bp fragment was cloned and sequenced to verify that the CAMP promoter was amplified. QRT-PCR was performed using SYBR Green (Molecular Probes, Eugene, OR, USA) essentially as described (32)⇓.
RESULTS
Induction of CAMP gene expression by 1,25(OH)2D3
In an initial screen to identify extracellular signals that might induce CAMP gene expression, we treated the myeloid leukemia cell line U937 with various inflammatory factors (LPS, TPA, TNF-α, INF-α, and INF-γ), cytokines and growth factors (IL-2, IL-6, GM-CSF and G-CSF) and seco-steroid hormones (DHT, estradiol, ATRA and 1,25(OH)2D3) (Fig. 1⇓ A and data not shown). As determined by QRT-PCR and Northern blot analyses, only 1,25(OH)2D3 induced CAMP expression significantly (Fig. 1A, B⇓ ). The induction was also observed in HL60 (Fig. 1B, C⇓ ) and NB4 (Fig. 1C⇓ ). Induction of the CAMP gene occurred by day 1 in the U937 and HL60 cell lines, but was stronger in the U937 cell line (Fig. 1B⇓ ). A time course of from 1 to 24 h in U937 indicated that CAMP induction began between 1 and 3 h after addition of 1,25(OH)2D3 and prior to induction of the differentiation marker CDllb at 12 h (Fig. 1B⇓ ). CAMP induction continued throughout the 5 days of treatment and was dose responsive (Fig. 1B, C⇓ ). Each of the chemically synthesized 1,25(OH)2D3 analogs—KH1060, EB1089, and compound I—strongly induced CAMP gene expression (Fig. 1D⇓ ). Levels of induction were similar to those observed for 1,25(OH)2D3. No induction was observed with ATRA (5×10−7 M, 1–5 days) in U937, HL60, and NB4 (Fig. 1A⇓ and data not shown). This is consistent with the inability of human myeloid leukemia cell lines to express significant levels of mRNAs for secondary granule genes even when induced to undergo granulocytic differentiation by ATRA (35)⇓.
Induction of CAMP mRNA expression by 1,25[OH]2D3. A) U937 cells were treated with vehicle (–, ethanol) or the indicated compounds for 24 h as described in Materials and Methods. Expression levels of CAMP were determined by QRT-PCR. Standard curves with known amounts of CAMP or 18S cDNA were included to measure the starting quantity of CAMP (ng) and 18S (ng) cDNA in each sample. Graphs depict the ratio of CAMP/18S in each sample (±sd). PCR was performed in triplicate for each sample. B) U937 or HL60 cells were treated with vehicle (0) or 1 × 10–7 M 1,25[OH]2D3 for 1, 3, or 5 days (upper panel) or for 1, 3, 6, 12, or 24 h (lower panel). Total RNA was analyzed by Northern blot using probes against CAMP, CDllb, or β-actin. C) U937, HL60, and NB4 cells were treated with either vehicle or decreasing molar concentrations of 1,25[OH]2D3 for 24 h. Total RNA was subjected to Northern blot analysis as described for panel B. D) U937 cells were treated with vehicle (0), 1,25[OH]2D3 (VitD3), or one of its analogs at 1 × 10–7 M for 12 or 24 h. Total RNA was subjected to Northern blot analysis with probes against CAMP or β-actin. E) U937 cells were treated with vehicle (0) or 1 × 10–7 M 1,25[OH]2D3 for 1, 2, 4, or 6 h in the absence (–ActD) or presence (+ActD) of actinomycin D (10 μg/mL). Expression levels of CAMP were determined by QRT-PCR and normalized to 18S. F) U937 cells were treated with vehicle (0) or 1,25[OH]2D3 for 0, 6, and 9 h in the absence (–) or presence (+) of cyclohexamide (CHX, 20 μg/mL). Total RNA was subjected to Northern blot analysis as described in panel D. G) U937 cells were treated with vehicle (0) or 1 × 10–7 M 1,25[OH]2D3 for 12 or 24 h. The cDNAs from total RNA were analyzed by RT-PCR using primers against myeloperoxidase (MPO), α-defensin (HNP-3), matrix metalloprotease 8 (MMP8), lactoferrin (LTF), CAMP, β-actin, and 18S. Amplification for all genes was 35 cycles except CAMP (30 cycles), β-actin (25 cycles), and 18S (10 cycles). A negative control (c, ddH2O) and a positive control (normal bone marrow RNA, BM) were included.
The induction of CAMP was blocked by actinomycin D, indicating it occurred at the level of transcription (Fig. 1E⇓ ). The data suggested that the human CAMP gene was a direct transcriptional target of the VDR. The steroid hormone receptor family members are generally present in the cytosol or bound to the DNA in an inactive state and require activation by binding ligand (36)⇓. Upon binding to ligand, they immediately translocate to the nucleus and bind vitamin D response elements (VDREs) in target genes and induce gene expression. The model predicts that ongoing protein synthesis is not required for this process to occur. To test this, we treated U937 cells with 1,25(OH)2D3 in the presence or absence of cyclohexamide (CHX) to block protein synthesis. Induction of CAMP gene expression occurred in the absence of ongoing protein synthesis (presence of CHX) (Fig. 1F⇓ ). CHX did not induce CAMP gene expression (data not shown). These data further support the hypothesis that the CAMP gene is a direct target of the VDR and not activated by secondary events such as the synthesis of other transcription factors that are induced by VDR.
To determine the specificity of CAMP induction by 1,25(OH)2D3, we tested other neutrophil primary [MPO (myeloperoxidase) and HNP3 (α-defensin)] and secondary [MMP8 (matrix metalloproteinase 8) and LTF (lactoferrin)] granule genes for induction. We did not observe induction of these genes after 24 h of treatment, whereas CAMP was significantly up-regulated (Fig. 1G⇓ ). The data demonstrate that the 1,25(OH)2D3 induction of neutrophil granule genes is restricted primarily to CAMP.
Induction of CAMP is independent of monocytic differentiation
Vitamin D3 promotes macrophage-like differentiation of U937 and HL-60 (37)⇓. To determine whether differentiation was responsible for the induction of the CAMP gene in AML cell lines, we treated HL-60 and U937 cells with 1,25(OH)2D3 or TPA. Both compounds promoted macrophage-like differentiation of these cells as demonstrated by the induction of the differentiation marker CDllb, but induction of CAMP mRNA was observed only with 1,25(OH)2D3, not TPA (Fig. 2⇓ A). The data suggested that induction of differentiation was not sufficient for CAMP expression. Furthermore, 1,25(OH)2D3 induced expression of CAMP was observed in the AML cell line NB4, which does not undergo macrophage differentiation when treated with 1,25(OH)2D3 (Fig. 1C⇓ ). Finally, to demonstrate that CAMP induction occurs in the absence of differentiation, we treated sub-lines derived from HL-60, which are unable to differentiate in response to vitamin D3 (HL60R) or ATRA (HL60Δ404) with 1,25(OH)2D3 and found that CAMP was induced in the absence or presence of differentiation (Fig. 2B⇓ , HL60R vs. HL60Δ404, respectively).
CAMP mRNA induction occurs in the absence of monocytic differentiation in patient samples and in nonmyeloid cells. A) The myeloid leukemia cell lines HL60 and U937 were treated with vehicle (0), 1,25[OH]2D3 (1×10–7 M), or TPA (5 ng/mL) for 1, 3, or 5 days. Total RNA was extracted and analyzed by Northern blot. B) The HL60 sub-lines HL60R (pan-resistant) and HL60Δ404 (ATRA-resistant) were treated with vehicle (–) or 1,25[OH]2D3 (+, 1×10–7 M) for 24 h. Northern blot analysis was performed. C) Bone marrow cells from normal human patients (NHBM, upper panel) were cultured in RPMI1640 + 10% FCS with vehicle (0) or 1,25[OH]2D3 for 72 or 120 h. BM-derived Mφ were treated for 24 h with vehicle (0) or an increasing concentration of 1,25[OH]2D3. AML BM cells were treated for 24 h with vehicle (0) or 1,25[OH]2D3 (lower panel). Duration of treatment and the concentrations of 1,25[OH]2D3 are indicated along the bottom of the graphs. Total RNA was prepared and cDNAs were analyzed using QRT-PCR. The fold change within each set is indicated inside the bar. As a positive control for induction, U937 cells were treated with vehicle (0) or 1,25[OH]2D3 for 12 and 24 h. D) 1,25[OH]2D3 induces CAMP expression in keratinocyte (HaCat) and colon cancer (Ht-29) cell lines. Cells were treated with vehicle (–) or 1,25[OH]2D3 (+, 1×10–7 M) for 24 h. Total RNA was prepared; cDNAs were synthesized and analyzed by QRT-PCR.
Induction of the CAMP gene occurs in bone marrow cells from normal humans and a patient suffering from acute myeloid leukemia
To determine whether CAMP induction by vitamin D3 occurs in hematopoietic cells other than leukemia cell lines, we treated total bone marrow (BM) cells and BM-derived macrophages (BM Mφ) from two normal individuals and BM cells from one AML patient with 1,25(OH)2D3 in vitro. Total RNA was harvested, cDNAs synthesized, and the quantity of CAMP mRNA expression was determined by QRT-PCR using a Taqman probe assay (Fig. 2C⇓ ). As demonstrated previously, a strong induction of CAMP was observed for U937 treated with 1,25(OH)2D3 (Fig. 2C⇓ , upper panel). Similarly strong induction of CAMP was observed in two normal human bone marrow cell samples and in BM Mφ (Fig. 2C⇓ , upper panel). The AML cells had a high baseline level of CAMP expression, which was induced further in a dose-responsive manner by 6- and 11-fold (Fig. 2C⇓ , lower panel). These data demonstrate that 1,25(OH)2D3 can markedly enhance the expression level of CAMP mRNA in normal and diseased human BM cells and that the induction is not a cell line phenomenon.
The induction of CAMP by 1,25(OH)2D3 was not limited to myeloid cells. We observed induction of CAMP mRNA in the keratinocyte cell line HaCat and the colon cancer cell line HT-29 by QRT-PCR (Fig. 2D⇓ ). Induction was not as robust as that observed in the myeloid cells.
To determine whether the induction of CAMP mRNA expression resulted in an increase of CAMP (hCAP18) protein expression, Western blot and immunofluorescent microscopy analyses were performed on U937 cells treated with 1,25(OH)2D3 (Fig. 3⇓ A, B). At 18 h and 36 h post-treatment, increased levels of hCAP18 were observed compared with untreated cells (Fig. 3A, B⇓ ). An ELISA performed on the medium from U937 cells treated for 24 h with ethanol or 1,25(OH)2D3 showed that CAMP was secreted into the medium (Fig. 3C⇓ ).
Induction of CAMP protein hCAP18 in U937 treated with vitamin D3. Cells were treated with either vehicle (–) or 1,25[OH]2D3 (+, 1×10–7 M) for 18 and 36 h. A) Cytospins of cells treated for 36 h were prepared and IF for hCAP18 was performed. Photographs were taken at 200× magnification. Examples of strongly positive cells are indicated by black arrows B) Total cell lysates were analyzed by Western blot for hCAP18 expression. The position of hCAP18 is indicated by arrow at the right (upper panel). The positions of the molecular weight markers are indicated at the left. Subsequent probing of the same blot for GAPDH demonstrated equivalent loading of protein in each lane (lower panel). (C) Levels of hCAP18 in the medium of U937 cells treated with vehicle or 1,25[OH]2D3 were determined by ELISA.
Identification of a functional VDRE in the human CAMP promoter
We hypothesized the existence of a VDRE in the CAMP promoter to explain the strong induction of CAMP mRNA expression by exposure to vitamin D3. A search of the upstream region revealed a classical DR3-type VDRE (38)⇓ at –615 bp from the transcriptional start site (Fig. 4⇓ A) (30)⇓. PCR was used to amplify the human CAMP promoter from nucleotides –693 to +14 (30)⇓. This fragment was subcloned into the firefly luciferase reporter plasmid pXP2 and called pXP2-CAMP-Luc (Fig. 4A⇓ ). Subsequently, deletion mutants pXP2-CAMP(ΔSmaI)-Luc and pXP2-CAMP(ΔHindIII)-Luc were generated by restriction enzyme digestion using the SmaI and HindIII sites, respectively (Fig. 4A⇓ ).
Identification of a functional VDRE in the human CAMP promoter. A) Sequence of the human CAMP promoter (–693 to +14) (30)⇓ as it was cloned into the firefly luciferase reporter vector pXP2. Restriction enzyme sites are indicated across the top of the sequence and transcription factor binding sites are indicated across the top and bottom. These include CCAATT displacement protein (CDP), STAT3, C/EBP, PU.1, and VDR. The sequences of the primers used for chromatin IP are underlined (line with filled circles at each end). Two additional constructs were generated by deleting from the 5′-end with Smal and with HindIII. The shaded box indicates the position of a repetitive element (SINE) in the promoter (schematic diagram). B) U937 cells were transfected (twice in duplicate) with the CAMP promoter-firefly luciferase reporter constructs and a renilla expression vector, phTKRL. Each transfection was treated with vehicle (–) or 1,25[OH]2D3 (+, 1×10−7 M) for 18 h. Dual luciferase assays were performed and firefly luciferase activity was normalized to renilla luciferase activity. The untreated and treated conditions for each construct were compared. CAMP-Luc (pXP2-CAMP-Luc); ΔSmaI [pXP2-CAMP(ΔSmaI)-Luc], and ΔHindIII [pXP2-CAMP(ΔHindIII)-Luc]. C) ∼ 1 × 107 cells were incubated in the absence (–) or presence (+) of 1,25[OH]2D3 at 1 × 10−7 M for 4 h. ChIP assays were performed and the promoter was detected by conventional (upper panel) PCR (reverse image of ethidium bromide stained gel; 30 cycles) and QRT-PCR (lower panel). The relative amount of CAMP promoter gDNA was determined in each sample by SYBR Green QRT-PCR. Differences (fold change) were normalized to the preimmune (Pre, average of untreated and treated) and indicated by the number within each bar. The positions of the DNA markers are indicated in base pairs (bp) at the left of the panel and the size of the expected promoter product is indicated at the right. Anti-ε, rabbit anti-C/EBPε antiserum.
The CAMP promoter constructs were transfected into U937 cells that were subsequently treated with vehicle or 1,25(OH)2D3. After 18 h treatment, cell lysates were prepared and dual luciferase assays were performed. In the absence of 1,25(OH)2D3, luciferase activity for all reporter constructs, including the empty parental vector, was similarly low (Fig. 4B⇓ ). This is consistent with the very low levels of endogenous CAMP mRNA expression in untreated U937. Upon treatment, the full-length promoter construct pXP2-CAMP-Luc was consistently activated 2- to 2.5-fold (Fig. 4B⇓ ). The deletion mutants pXP2-CAMP(ΔSmaI)-Luc and pXP2-CAMP(ΔHindIII)-Luc were not activated. pXP2-CAMP(ΔSmaI)-Luc still possesses the VDRE; however, the SmaI site used for the generation of the construct is immediately adjacent to the VDRE (Fig. 4A⇓ ), suggesting that a single or several nucleotides located 5′ to the VDRE is required for the response. These data demonstrate that this VDRE is required for activation of the CAMP promoter by vitamin D3.
VDR binds to the CAMP promoter in cells
To determine whether VDR complexes were actually binding to the CAMP promoter, we performed ChIP assays on chromatin prepared from U937 cells treated with vehicle or 1,25(OH)2D3 for 4 h (Fig. 4C⇓ ). Because the VDRE is located in a repetitive DNA element or short interspersed nuclear element (SINE), it was difficult to design primers for PCR that specifically amplified that region of the CAMP promoter (Fig. 4A⇓ , shaded boxes). Therefore, we designed primers to the nonrepetitive region near the transcriptional start site that specifically amplifies the CAMP promoter (Fig. 4A⇓ ). The chromatin was sheared to an average size of ∼1 kb and immunoprecipitated with antibodies (Ab) specific for the VDR and C/EBPε proteins. C/EBPε activates CAMP gene expression (39)⇓ and was included as a positive control. For negative controls chromatin was immunoprecipitated with protein A-Sepharose (No Ab) or preimmune serum (Pre). The samples were amplified by conventional PCR and visualized by ethidium bromide staining (Fig. 4C⇓ , upper panel) or QRT-PCR (Fig. 4C⇓ , lower panel). Extremely low background levels were detected in the negative controls (Fig. 4C⇓ , No Ab or Pre). A significant level of the promoter was immunoprecipitated by anti-VDR Ab (22-fold above background) without 1,25(OH)2D3 treatment, and this increased by > 2-fold (48-fold above background) with treatment (Fig. 4C⇓ , lower panel). Binding of C/EBPε to the promoter was similar under both conditions (76- and 89-fold), demonstrating that vitamin D3 treatment is not increasing the amount of C/EBPε binding to the promoter (Fig. 4C⇓ ). These results indicated that VDR is binding to the CAMP promoter in both a ligand-dependent and -independent manner, consistent with current models of steroid-hormone gene regulation.
Induction of CAMP by vitamin D3 is not evolutionarily conserved
To elucidate further the role of the VDR in regulating CAMP gene expression, we examined the expression of the murine CAMP/CRAMP gene in RNA from untreated bone marrow cells from a VDR-deficient mouse and its wild-type littermate (Fig. 5⇓ A, left panel). Bone marrow RNAs from C/EBPε-deficient and wild-type mice were included as controls (Fig. 5A⇓ , left panel). As expected, the C/EBPε-deficient bone marrow lacked expression of CRAMP (40)⇓. In contrast, CRAMP was expressed in the VDR-deficient cells at a level comparable to the wild-type littermate. Furthermore, intraperitoneal treatment of BNX mice with 1,25(OH)2D3 or vitamin D3 compound I over 6 wk did not significantly alter CRAMP expression in bone marrow compared with a vehicle-treated mouse (Fig. 5A⇓ , middle panel). We did not observe induction of CRAMP in murine cell lines 32Dcl3 (Fig. 5⇓ , right panel), NIH3T3 and Wehi3B (data not shown). Finally, CRAMP induction was not observed in C/EBPε-deficient or wild-type bone marrow cells cultured in vitro with 1,25(OH)2D3 (Fig. 5B⇓ ) or in BM Mφ from VDR-deficient or wild-type mice (Fig. 5C⇓ ). Indeed, an ∼ 2-fold decrease was observed by 24 h post-treatment (Fig. 5B, C⇓ ) and 5-fold by 48 h (Fig. 5C⇓ ).
Vitamin D3 induction of CAMP is not conserved in the murine system. A) Total RNA from bone marrow cells flushed from the femurs of C/EBPε, VDR wild-type (WT), or knockout (KO) mice were analyzed for murine CAMP (CRAMP) expression by Northern blot (left panel). Total RNA of BM cells from BNX mice treated for 6 wk with vehicle (–), 1,25[OH]2D3 (D3), or vitamin D3 analog compound I were examined for CRAMP expression (middle panel). The murine myeloid cell line 32Dcl3 was treated with vehicle (0) or 1,25[OH]2D3 at 1 × 10−7 M for 24 and 48 h. Total RNA was analyzed by Northern blot for CRAMP expression (right panel). β-Actin levels were used to demonstrate even loading of the samples. B) The BM cells from C/EBPε WT or KO mice were cultured with vehicle (–) or 1,25[OH]2D3 for 24 h. Relative levels of CRAMP were determined by QRT-PCR. (C) BM Mφ from VDR WT or KO mice were treated with vehicle (–) or 1,25[OH]2D3 for 24 or 48 h. Relative levels for CRAMP were determined by QRT-PCR. D) Screen shot (Human May 2004 Assembly) of the human CAMP genomic region (chr3: 48, 237, 952-48, 423, 990; UCSC Genome Browser; http://genome.ucsc.edu) (64⇓, 65)⇓. Conservation of the human genome {International Human Genome Sequencing Consortium) compared with the chimpanzee (Chimpanzee Genome Sequencing Consortium), dog (The Broad Institute and Agencourt Bioscience), rat (Rat Genome Sequencing Consortium) and mouse (Mouse Genome Sequencing Consortium) genomes are depicted by histograms and the Alignment Net. The positions of SINEs and LINEs are indicated. The location of the VDRE within the SINE is indicated by the arrow.
We compared genomes from human, chimpanzee, rat, dog, and mouse to determine conservation of the promoter region for each CAMP gene (Fig. 5D⇓ ). Though significant homology was observed, a gap was identified at –409 bp upstream from the start site of transcription in the human promoter. This was a due to a SINE conserved only in the human and chimpanzee genomes and absent in the others (Fig. 5D⇓ ). The VDRE is located in this SINE. Thus, the mouse gene lacks a VDRE. This is consistent with the observed absence of CRAMP induction by vitamin D3.
DISCUSSION
In this study, we showed that 1,25-dihydroxyvitamin D3 and three of its analogs induced expression of the antimicrobial peptide gene CAMP in AML, immortalized keratinocyte, and colon cancer cell lines, as well as bone marrow-derived macrophages and fresh bone marrow cells from normal individuals and one AML patient. The induction of antimicrobial protein genes by vitamin D3 in myeloid cell lines was restricted primarily to CAMP.
Activation of the CAMP gene occurred via a consensus VDRE in the promoter that is bound by VDR. The VDR is expressed in a wide range of tissues; potentially, CAMP can be induced in all of these tissues. A recent report published during the preparation and submission of this manuscript reported findings consistent with those reported here (41)⇓. They observed induction by 1,25(OH)2D3 in purified monocytes, neutrophils, and cell lines from lung as well as head and neck squamous cell carcinomas (41)⇓. This study expands on these observations by demonstrating that not only does 1,25(OH)2D3 induce CAMP gene expression, but so do analogs of vitamin D3. We showed that induction occurs in the cells of the bone marrow. Moreover, we discovered that the induction of CAMP by vitamin D3 does not occur in mice. In fact, it appears that this mechanism is conserved in primates (humans and chimpanzees) and not in other mammals as suggested by the absence of the VDRE in the murine, rat, and canine CAMP promoters. The VDRE is present in a SINE element of the Alu-Sx subfamily. These elements can retrotranspose from a progenitor element to other locations in the genome during evolution. It would appear that this event occurred in a primate progenitor. Whether this element is conserved in all primates (other than humans and chimpanzees), as well as in New World and Old World monkeys, has not been determined.
The biological importance of this regulation is intriguing. Unfortunately, the murine model for VDR-deficiency will not prove useful. Perhaps examination of vitamin D-resistant rickets patient samples for CAMP expression may elucidate the importance of vitamin D3 regulation of CAMP. It is interesting that these patients suffer from frequent dental abscesses (42)⇓. Decreased expression of CAMP/hCAP18 protein may contribute to this as it does in Kostmann syndrome patients who lack hCAP18 (43)⇓. Finally, CAMP/hCAP18 has immunomodulating properties ascribed to it. Whether or not it mediates some of those immunosuppressive properties of vitamin D3 or acts to counter them needs clarification.
While these observations further expand the role of vitamin D3 in immunomodulation in humans (44⇓, 45)⇓, they also indicate that the use of vitamin D3 and its analogs provides a method to manipulate extrinsically the expression of CAMP. This may provide additional avenues for using relatively safe compounds in the treatment of human disease and injury. Enhancing the expression of CAMP expression could prove advantageous. Protective effects of CAMP overexpression in respiratory epithelia were observed in a cystic fibrosis model (46)⇓. The systemic expression of CAMP/hCAP18 in mice improved survival rates after intravenous injection of lipopolysaccharide (LPS) (47)⇓. LPS is a component of the bacterial cell wall of gram-negative bacteria such as Escherichia coli or P. aeruginosa. Massive gram-negative bacterial infection can result in septic shock due to the large amounts of LPS present in the blood. Thus, hCAP18 may not only aid in clearance of bacterial infection, but may protect against the sepsis. This protection probably derives from the ability of CAMP to bind to LPS and neutralize it (48⇓49⇓50⇓51)⇓. The hCAP18 peptide has been shown to inhibit LPS-induced cellular responses such as release of TNF-α, tissue factor, and nitric oxide, protecting mice and pigs from septic shock (48⇓, 52)⇓. In vitro, hCAP18 inhibits macrophage activation by LPS and other bacterial components (51)⇓. If endogenous hCAP18 levels can be increased by extrinsic manipulation, then the potential exists to treat conditions that are susceptible to the development of sepsis. Boosting CAMP/hCAP18 levels potentially could protect against this condition after surgery and speed wound healing.
Like VDR expression, CAMP expression is widespread. It is important for barrier defenses in the skin. Mice deficient in CAMP are much more susceptible to skin infection than wild-type mice (53)⇓. CAMP expression is up-regulated during cutaneous infection, injury, or inflammation (psoriasis) of the skin (54⇓55⇓56)⇓. Decreased levels of hCAP18 in the skin of individuals with atopic dermatitis (AD) correlates with their increased susceptibility to skin infection compared with those with psoriasis (55)⇓. Vitamin D3 and its analogs have proven safe and effective in the treatment of psoriasis. It remains to be determined whether CAMP induction occurs with the topical application of vitamin D3. If so, treatment of CAMP-deficient AD with vitamin D3 may prove beneficial also.
Increasing CAMP expression by vitamin D3 treatment may prove beneficial in other instances. CAMP is up-regulated in gastric inflammation caused by Heliobacter pylori infection (57)⇓ and infection of cultured epithelial cells with Salmonella and entero-invasive E. coli modestly induced CAMP mRNA expression (58)⇓. In contrast, infection by Shigella spp. was reported to down-regulate CAMP mRNA expression in the colon (59)⇓. Chronic oral bacterial infections occur in Kostmann syndrome patients who suffer from a severe chronic neutropenia. These patients lack expression of hCAP18 in their saliva, plasma, and neutrophils (43)⇓. Patients suffering from specific granule deficiency lack expression of defensins and hCAP18 and suffer severe, recurrent bacterial infections (60)⇓. Up-regulating CAMP/hCAP18 expression in these conditions could prove therapeutically beneficial.
The induction of CAMP expression by cytokines and growth factors has been reported in a number of tissues; but 1,25(OH)2D3 and its analogs are strikingly potent in myeloid cells. The induction was less striking in the HaCat and HT-29 cell lines, but combining vitamin D3 treatment with other compounds known to activate CAMP expression may increase expression. Treatment of cultured keratinocytes or composite keratinocyte grafts with LPS or IL-1α induced CAMP expression (61)⇓; on the other hand, TNF-α, Il-4, Il-6, IL-8, IL-10, and INF-γ did not. The growth factor insulin-like growth factor-1 that is important in wound healing was found to induce both the CAMP mRNA and protein in primary human keratinocytes; TGFα and proinflammatory cytokines IL-1β, IL-6, and TNF-α were not (62)⇓. In epithelial cells of the colon, hCAP18 expression is restricted to differentiated cells in the human colon and ileum (58⇓, 63)⇓. Consistent with this, hCAP18 expression was induced by differentiation of colon epithelial cell lines and by short chain fatty acids independent of differentiation, but not by proinflammatory mediators including IL-1α, IL-6, TNF-α, INF-γ, LPS, or PMA (58⇓, 63)⇓. Combining those cytokines or growth factors with vitamin D3 offers the possibility of obtaining synergistic activation of the CAMP gene. Such synergy was reported for LPS and vitamin D3 in neutrophils (41)⇓. Synergistic activation of the CAMP gene could prove useful in treating skin grafts for burn patients or in boosting immunity to opportunistic infections in chemotherapy patients.
Acknowledgments
We thank Ook Kim, Dih-Yih Chen, and Jonathan Frank for technical assistance and James O’Kelly for insightful discussions and helpful suggestions. We are grateful to Drs. Lise Binderup, Milan Uskokovic, Norbert Fusenig, and Alan Friedman for providing reagents. This work was supported by NIH grant CA26038-20, the Cindy and Allan Horn Foundation, Parker Hughes Trust, the King Harbor Yacht Club Tom Collier Memorial Fund, and the Inger Fund. H.P.K. holds the Mark Goodson endowed chair for Cancer Research and is a member of the Jonsson Cancer Center.
- Received October 25, 2004.
- Accepted March 2, 2005.
