SEARCH

SEARCH BY CITATION

Keywords:

  • innate immunity;
  • tuberculosis;
  • macrophage;
  • vitamin D;
  • CYP27b1;
  • CYP24;
  • cathelicidin;
  • hsc70

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS AND CONCLUSIONS
  5. Acknowledgements
  6. REFERENCES

Tissue availability of the active vitamin D metabolite, 1,25-dihydroxyvitamin D [1,25(OH)2D] is dependent on expression of the activating enzyme 1α-hydroxylase (CYP27b1) and its catabolic counterpart 24-hydroxylase (CYP24). The activity of these two enzymes is in turn controlled by factors including affinity of the serum vitamin D–binding protein (DBP) for 25-hydroxyvitamin D [25(OH)D]; the availability of enzyme cofactors; and the relative amount of hydroxylase gene product expressed. In recent years, it has become clear that directed trafficking of substrate and enzyme is also a pivotal component of the regulated process of hormone synthesis by both renal and extrarenal tissues expressing the CYP27b1 and CYP24 genes. Extracellular regulatory trafficking events are defined by the quantity of substrate 25(OH)D entering the circulatory pool. Entry into some target cells in vivo, such as the macrophage and proximal renal tubular epithelial cells, requires 25(OH)D binding to serum DBP, followed by recognition, internalization, and intracellular release. The “released” intracellular substrate is moved to specific intracellular destinations (i.e., the mitochondrial CYP enzymes and the vitamin D receptor [VDR]) by the hsc70 family of chaperone proteins. Synthesis of 1,25(OH)2D is also regulated by CYP24 and its metabolically inactive splice variant CYP24-SV. Finally, initiation of transcription of 1,25(OH)2D-regulated genes, such as the CYP24, requires movement of the CYP27b1 product, 1,25(OH)2D, to the VDR in the same cell for intracrine action or export to another cell for paracrine action. In either case, the 1,25(OH)2D ligand is required for the VDR to heterodimerize with the retinoid x receptor and compete away the dominant-negative acting, heterogeneous nuclear ribonucleoprotein (hnRNP)-related, vitamin D response element–binding proteins that inhibit hormone-directed transactivation of genes. In this review, we use vitamin D–directed events in the human innate immune response to Mycobacterium tuberculosis as a physiologically relevant model system in which to highlight the importance of these intracellular traffic patterns.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS AND CONCLUSIONS
  5. Acknowledgements
  6. REFERENCES

If one uses the macrophage, the target cell that this laboratory has focused on for many years, the traditional view holds that control of vitamin D metabolism and action is achieved by the amount of free prohormone 25(OH)D to which the cell is exposed; the relative abundance of enzymes that synthesize (CYP27b1) and catabolize (CYP24) active 1,25-dihydroxyvitamin D [1,25(OH)2D]; the abundance of endogenous vitamin D receptor (VDR) to which the ligand 1,25(OH)2D can bind; and the ability of the liganded VDR to engage other transfactors, such as the retinoid x receptor, as well as co-activators and repressors associated with the regulation of hormone-directed gene transcription (Fig. 1). In the following review, we describe how these mechanisms can be integrated to control an important feature of extrarenal vitamin D function, namely the induction of macrophage innate immunity through synthesis of the antimicrobial cathelicidin LL37. Based on these studies, we postulate that control of 1,25(OH)2D synthesis and action is much more complicated than initially envisioned and involves novel mechanisms that mediate the uptake, intracellular transport, substrate availability, and DNA targeting of liganded VDR. Work presented here will describe why macrophage vitamin D synthetic and metabolic events are of relevance to human physiology; discuss the consequences of alternative splicing of CYP24 on these events; outline the role of the dominant-negative–acting transfactors in the hnRNP family to control 1,25(OH)2D-mediated transcription; and describe what we are beginning to learn about the entry and trafficking of vitamin D metabolites in target cells.

thumbnail image

Figure FIG. 1.. Control of the endocytic internalization, metabolism, and action of 25-hydroxylated vitamin D metabolites in the human macrophage. (Left) Uptake of DBP-bound 25(OH)D (inset); IDBP (hsc70)-directed transport of 25(OH)D to the mitochondrial CYP27b1-hydroxylase and CYP24-hydroxylase; and interaction of 1,25(OH)2D with the VDR, promoting dimerization with the RXR and conferring transactivational potential on the receptor complex. (Right) Consequences of alternative splicing of the CYP24 gene with production of a translated CYP24 splice variant (CYP24SV) lacking an amino-terminal mitochondrial targeting sequence. The end result is the expression of cytoplasmically localized “decoy” that binds 25(OH)D and 1,25(OH)2D, disallowing their entrance into the intermitochondrial membrane space and precluding their 1- and 24-hydroxylation.

Download figure to PowerPoint

RESULTS AND CONCLUSIONS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS AND CONCLUSIONS
  5. Acknowledgements
  6. REFERENCES

Vitamin D and human immune responses

The schematic representation in Fig. 2 describes mammalian immune response as recently reviewed.[1] The initiating step is a breech in the epithelial barrier between the bacterial-laden outside environment and the relatively sterile internal environment of the host. Once the bacterial products, so called pathogen-associated molecular patterns or PAMPs, gain access to the host interior, they are recognized by a subclass of pattern recognition receptors embedded in the plasma membrane of macrophages and dendritic cells. Among these are the toll-like receptors (TLRs), which recognize an array of widely varying kinds of antigens, including peptides, lipids, and nucleic acids.[2] Liganded TLRs recruit MyD88 adaptor proteins, which initiate a number of intracellular signaling pathways, many of which terminate in the transaction of NF-κB. The end result is the induction of innate immunity in the form of antigen phagocytosis and destruction, followed by the initiation of cell (T lymphocyte) and/or antibody (B lymphocyte)-directed adaptive immune responses. In some cases, tissue injury in the form of septic shock or persistent inflammation may occur as a consequence of TLR signaling, depending on the control mechanisms that modulate innate and adaptive immunity. Data from our group suggest that 1,25(OH)2D, made by and acting in activated macrophages and dendritic cells, is a crucial component of regulation of human immunity by promoting microbial killing while, at the same time, keeping a check on the vigor of the adaptive immune response to presented antigen.[3] Moreover, the ability of 1,25(OH)2D to inhibit NF-κB signaling[4] and suppress macrophage TLR expression[5] suggests that it may also play a key role as an autocrine feedback regulator of macrophage responses.

thumbnail image

Figure FIG. 2.. Vitamin D and immune responses at barrier sites. Initiation of the human macrophage responses to bacterially derived pathogen-associated molecular patterns (PAMPs) interacting with toll-like receptors (TLRs). The TLR family is shown alongside their proposed PAMP ligands. Activation of TLR2/1 and TLR4 induces CYP27b1 and increases local capacity for synthesis of 1,25-vitamine D, which can activate innate immunity through enhanced transcription of antimicrobial peptides or suppress inflammation through feedback inhibition of macrophage NF-κB signaling and through modulation of dendritic cell–T lymphocyte interaction.

Download figure to PowerPoint

Macrophages, mycobacteria, and vitamin D

For some time now it has been known that macrophages stimulated by live Mycobacterium tuberculosis, either in vivo in the pleural effusion fluid of patients with active tuberculosis or when removed to an in vitro environment and incubated with substrate 25(OH)D3, will make substantial amounts of 1,25(OH)2D3.[6] It is also known that IFN-γ concentrated in the pleural effusion fluid of such patients greatly accelerates this synthetic reaction to the point that patients can become hypercalcemic or hypercalciuric from macrophage overproduction of the hormone.[7] The question has always been whether the production of the hormone locally is an immunologically relevant event in terms of mycobacterial killing. Now, armed with knowledge of the role of TLRs 1 and 2 as transducers of the mycobacterial immunity in human macrophages (Fig. 1), our laboratory has been able to fully define the role of 1,25(OH)2D as an important component of normal innate immune responses.[8]

Initial data showed that it was TLR2/1-primed human macrophages that killed the bacterium, whereas M. tuberculosis persisted and even grew inside TLR2/1-primed dendritic cells. We sought to define which genes were involved in the macrophage M. tuberculosis killing reaction, using DNA microarray analysis of macrophages and dendritic cells before and after stimulation of the cells with a M. tuberculosis ligand (PAMP) known to interact with the TLR2/1. Results showed that the VDR was among the most strongly upregulated genes in TLR2/1-activated macrophages, but it was subsequently determined that expression of CPY27b1 was also induced, raising the possibility of an autocrine action for 1,25(OH)2D in these cells. A candidate gene for this is the defensin-like cathelicidin gene, known to possess a VDR element (VDRE) enhancer in its promoter.[9, 10] Dose–response experiments showed that 1,25(OH)2D3 induced cathelicidin mRNA and protein expression in human macrophages but not dendritic cells. Furthermore, when infected with a fluorescently labeled surrogate for M. tuberculosis, Bacille Calmette Guérin (BCG)-green fluorescent protein (GFP), the organism and cathelicidin were observed to be co-localized at the membrane of the monocyte.

In summary, these data suggested that flooding the outside of the cell with 1,25(OH)2D3 promotes transactivation of the cathelicidin gene, thus augmenting endogenous LL37 production and killing of ingested M. tuberculosis. Crucially, the requirement for exogenous 1,25(OH)2D3 was obviated if the macrophages were cultured in human serum versus FCS. Further analysis showed that this was caused by greatly varying levels of substrate 25(OH)D in these different sera. Specifically, 25(OH)D levels were much higher in human serum (75 nM) than FCS (18 nM), so that when used at the 10% level in macrophage cultures, the cells would be exposed to either 7.5 or 1.8 nM, respectively, in human serum and FCS. This suggested that variations in the circulating concentration of 25(OH)D, rather than 1,25(OH)2D, played a key role in defining macrophage function. Additional studies using 25(OH)D3 added to FCS confirmed the dose-dependent induction of cathelicidin expression in macrophages but only in cells in which there was TLR2/1 induction of CYP27b1. In other words, the mechanism was dependent on autocrine activation of 25(OH)D to 1,25(OH)2D. The contrast in effect between FCS and human serum also indicated that intracrine activation of relatively small amounts of 25(OH)D to 1,25(OH)2D could be as effective as exposure to relatively high extracellular concentrations of 1,25(OH)2D in stimulating antimicrobial activity in macrophages. This led us to study the mechanisms that underpin the differential handling, metabolism, and action of these vitamin D metabolites.

Novel means of control of CYP24 expression and action

A consistent observation from studies of human macrophages treated with either exogenous 1,25(OH)2D3 or 25(OH)D3 is that, in addition to induction cathelicidin expression, there is potent upregulation of CYP24 mRNA levels. If the conventional view holds true that this is a negative feedback mechanism resulting in the expression of an enzyme that is catabolic to 1,25(OH)2D, how is the effect of 1,25(OH)2D on innate immunity maintained? A possible explanation arose from recent studies in which we showed that, although both exogenous 1,25(OH)2D3 and 25(OH)D3 are able to induce mRNA for CYP24 in macrophages, this does not result in a concomitant increase in 24-hydroxylase enzyme activity.[11] One of the reasons for this discrepancy in protein and functional enzyme is the presence in these cells of a translatable splice variant of CYP24. Figure 3 provides a schematic representation of how this splice variant is generated. The 5' end of the wildtype CYP24, containing the mitochondrial targeting sequence, is deleted by alternative splicing of intron 2. An in-frame, alternative start site of translation is created encoding a protein that is missing its amino-terminal mitochondrial targeting domain but retaining both its sterol and heme binding domains. Based on these observations, we predicted that the CYP24-splice variant (CYP24-SV) has the potential to act as a cytoplasmic “decoy” for conventional mitochondrial CYP24 substrates such as 25(OH)D and 1,25(OH)2D (Fig. 1, right panel). Subsequent studies showed that, when overexpressed in macrophage-like cells, CYP24-SV is a much more efficient attenuator of 1,25(OH)2D synthesis than wildtype CYP24. In contrast, antisense inhibition of CYP24-SV enhanced 1,25(OH)2D production. These data suggested that, like its wildtype counterpart, CYP24-SV acts to control local levels of 1,25(OH)2D. However, what distinguishes the splice variant is its ability to do this through interaction with 25(OH)D rather than 1,25(OH)2D as is classically observed with wildtype CYP24. Indeed, based on the structures of the CYP3A4 and CYP2C8 proteins, preliminary 3D molecular modeling of CYP24 and CYP24-SV suggests that loss of exons 1 and 2 not only eliminates the mitochondrial targeting region of CYP24 but also affects the substrate binding pocket of the enzyme (data not shown). We are currently studying the extent to which this may result in preferential binding of 25(OH)D in the CYP24-SV.

thumbnail image

Figure FIG. 3.. Regulation of intracellular 1,25(OH)2D levels by CYP24 and CYP24-SV. Schematic describing alternative splicing of the CYP24 gene to yield an amino-terminally truncated translation product lacking a mitochondria targeting sequence but retaining both a sterol substrate and heme binding domain.

Download figure to PowerPoint

Novel mechanisms for control of 1,25(OH)2D-induced transactivation

Both CYP24 and its splice variant counterpart are potently stimulated by 1,25(OH)2D at a transcriptional level. We have recently shown that an important determinant of the transcriptional potential of liganded VDR is the dominant-negative–acting vitamin D response element binding protein (VDRE-BP; Fig. 4), now recognized to be the human hnRPC1/C2 protein.[12] When overexpressed in cells, hnRNPC1/C2 inhibits VDRE-mediated transcription, whereas siRNA suppression of hnRNPC1/C2 increases transcription. Although hnRNPC1/C2 is expressed in all normal cells, inappropriately high levels of the protein have been found in a patient who presented with symptoms of hereditary vitamin D–resistant rickets (HVDRR).[13] Chromatin immunoprecipitation (ChIP) analysis of cells from normal 1,25(OH)2D-responsive subjects has shown that, in the absence of added 1,25(OH)2D3, the CYP24 VDRE is normally occupied by hnRNPC1/C2, but this is displaced within 15 min of exposure to 1,25(OH)2D3.[12] The displacement of hnRNPC1/C2 was accompanied by reciprocal binding of liganded VDR to the VDRE. In contrast, ChIP analysis of cells from the patient with HVDRR who overexpressed hnRNPC1/C2 showed that, in both the absence and presence of 1,25(OH)2D3, VDR and hnRNP co-occupied the VDRE for as long as 45 min. Such dysregulation of normal hnRNPC1/C2 interaction with VDRE seems to be an entirely novel cause of hormone resistance. However, it is also apparent that hnRNPC1/C2 plays a pivotal role in the temporal “signature: of the VDR–VDRE interaction required to initiate normal 1,25(OH)2D-mediated transcription; future studies will be aimed at defining the role of this mechanism in controlling vitamin D function in diverse tissues and in different physiological/pathophysiological scenarios.

thumbnail image

Figure FIG. 4.. Dominant-negative regulation of 1,25(OH)2D (1,25D)-VDR–directed gene expression by the VDRE-BP. Shown is the conversion from a dominant-positive–acting VDR-RXR interaction with the VDRE in the proximal promoter of the CYP24 gene to the dominant-negative–acting state by competitive displacement of the VDR-RXR from the VDRE by the VDRE-BP.

Download figure to PowerPoint

Uptake and intracellular trafficking of vitamin D metabolites

For vitamin D metabolites to be further metabolized or act in target cells, they must traverse the plasma membrane of that target cell. The work of Nykjaer et al.[14] indicates that in vivo endocytic uptake of the vitamin D metabolite–serum vitamin D–binding protein (DBP) complex, in some but not all potential target cells, is mediated by megalin and cubulin, members of the LDL-like receptor protein family. To study these events in vitro, we developed an Alexa-green-fluorescent DBP that has an affinity and capacity for 25(OH)D similar to that of unlabeled human DBP. Endocytic uptake of fluorescent DBP was observed in megalin-positive BN16 rat yolk-sac cells and in human macrophages. These observations suggest that, in addition to the presumed nonspecific passive diffusion of 25(OH)D into cells such as macrophages, there may also be specific facilitated uptake of the vitamin D metabolite. We are currently assessing whether this mechanism provides an additional level of control for the local generation of active 1,25(OH)2D.

One possibility is that internalized DBP carrying its cargo of 25(OH)D interacts specifically with the plasma membrane anchored megalin/cubulin complex, which, in turn, associates with the cytoplasmically located intracellular DBP (IDBP; Fig. 1, inset, left panel), which we now know to be the constitutively expressed human heat shock protein-70 (hsc70).[15] Hsc70 enhances 1,25(OH)2D synthesis and action by promoting the intracellular uptake of 25-hydroxylated vitamin D metabolites.[16] Furthermore, when overexpressed in kidney cells, hsc70 enhances synthesis of 1,25(OH)2D and also stimulates 1,25(OH)2D-mediated VDRE promoter luciferase activity. Our working hypothesis states that hsc70 acts as an ATP-dependent intracellular chaperone that helps to move substrate 25(OH)D to mitochondrial CYP27b1-hydroxlase and the subsequent product of that catalytic reaction, 1,25(OH)2D, to the VDR to promote transactivation.

Clinical relevance of diminished 25(OH)D uptake by the human macrophage

The mechanisms involved in the uptake, intracellular trafficking, and gene promoter interactions of vitamin D metabolites are still far from clear. In contrast, it is now evident that it is the extracellular concentration of 25(OH)D, and not 1,25(OH)2D, that is the physiologically relevant, extracellular signal directing 1,25(OH)2D action in cells such as macrophages.[8] This was shown in experiments using vitamin D–deficient sera from blacks. As expected, the measured 25(OH)D levels in the pigmented black subjects were significantly lower than those of lightly pigmented white subjects. Compared with vitamin D–sufficient serum, when these sera were used in vitro to condition macrophages after TLR2/1 induction of CYP27b1, the vitamin D–insufficient serum was much less capable of supporting cathelicidin gene expression, an event that could be rescued by normalizing the extracellular 25(OH)D levels by the addition of exogenous 25(OH)D3. Thus, exposure of the human monocyte–macrophage to M. tuberculosis elicits the expression of both the VDR and CYP27b1 genes. If extracellular concentrations of 25(OH)D are adequate, than substrate is endocytically internalized bound to the serum DBP (Fig. 1, inset, left panel). Once inside the cell, vesicular acidification disrupts DBP-25(OH)D binding, making the substrate available to intracellular, heat shock protein–related chaperones that shuttle 25(OH)D to the mitochondrial CYP27b1 and its product 1,25(OH)2D to the VDR for transactivation of the cathelicidin gene with resultant antimicrobial actions on ingested M. tuberculosis. From a clinical standpoint, it is hypothesized that the diminished availability of substrate 25(OH)D to the macrophage CYP27b1 compromises the host's response to M. tuberculosis. In this way, we have proposed that vitamin D deficiency is a pivotal determinant of immune responses to M. tuberculosis and a crucial consideration in managing the worldwide epidemic of tuberculosis.

Acknowledgements

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS AND CONCLUSIONS
  5. Acknowledgements
  6. REFERENCES

We thank Gloria Kiel for help in preparation of this manuscript. This work was supported by NIH Grants DK033139, AR037399, AR050626, DK055843, DK058891, and RR00425.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS AND CONCLUSIONS
  5. Acknowledgements
  6. REFERENCES
  • 1
    Karin M, Lawrence T, Nizet V 2006 Innate immunity gone awry: Linking microbial infections to chronic inflammation and cancer. Cell 124: 823835.
  • 2
    Modlin RL, Cheng G 2004 From plankton to pathogen recognition. Nat Med 10: 11731174.
  • 3
    Adams JS 2006 Vitamin D as a defensin. J Musculoskelet Neuron Interact. 6: 344346.
  • 4
    Sun J, Kong J, Duan Y, Szeto FL, Liao A, Madara JL, Li YC 2006 Increased NF-kappaB activity in fibroblasts lacking the vitamin D receptor. Am J Physiol Endocrinol Metab 29: E315E322.
  • 5
    Sadeghi K, Wessner B, Laggner U, Ploder M, Tamandl D, Friedl J, Zugel U, Steinmeyer A, Pollak A, Roth E, Boltz-Nitulescu G, Spittler A 2006 Vitamin D3 down-regulates monocyte TLR expression and triggers hyporesponsiveness to pathogen-associated molecular patterns. Eur J Immunol 36: 361370.
  • 6
    Barnes PH, Modlin RL, Bikle DD, Adams JS 1989 Transpleural gradient of 1,25-dihydroxyvitamin D in tuberculous pleuritis. J Clin Invest 83: 15271532.
  • 7
    Adams JS, Modlin RL, Diz MM, Barnes PH 1989 Potentiation of the macrophage 25-hydroxyvitamin D-1-hydroxylation reaction by human tuberculous pleural effusion fluid. J Clin Endocrinol Metab 69: 457460.
  • 8
    Liu PT, Stenger S, Li H, Wenzel L, Tan BH, Krutzik SR, Ochoa MT, Schauber J, Wu K, Meinken C, Kamen DL, Wagner M, Bals R, Steinmeyer A, Zugel U, Gallo RL, Eisenberg D, Hewison M, Hollis BW, Adams JS, Bloom BR, Modlin RL 2006 Toll-like receptor triggering of a vitamin D–mediated human antimicrobial response. Science 24: 17701773.
  • 9
    White JH 2004 Profiling 1,25-dihydroxyvitamin D3-regulated gene expression by microarray analysis. J Steroid Biochem Mol Biol 89: 239244.
  • 10
    Gombart AF, Borregaard N, Koeffler HP 2005 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. FASEB J 19: 10671077.
  • 11
    Ren SY, Nguyen L, Adams JS, Hewison M 2005 Alternative splicing of the vitamin D-24-hydroxylase. J Biol Chem 280: 20602061.
  • 12
    Chen H, Hewison M, Adams JS 2006 Functional characterization of heterogeneous nuclear ribonuclear protein C1/C2 in vitamin D resistance: A novel response element binding protein. J Biol Chem 281: 3911439120.
  • 13
    Chen H, Hewison M, Hu B, Adams JS 2003 Heterogenous nuclear ribonucleoprotein (hnRNP) binding to hormone response elements: A cause of vitamin D resistance. Proc Natl Acad Sci USA 100: 61096114.
  • 14
    Nykjaer A, Dragun D, Walther D, Vorum H, Jacobsen C, Herz J, Melsen F, Christensen EI, Willnow TE 1999 An endocytic pathway essential for renal uptake and activation of the steroid 25(OH) vitamin D3. Cell 96: 507515.
  • 15
    Wu S, Ren S, Chen H, Chun RF, Gacad MA, Adams JS 2000 Intracellular vitamin D binding proteins: Novel facilitators of vitamin D-directed transactivation. Molec Endocrinol 14: 13871397.
  • 16
    Adams JS, Chen H, Chun R, Gacad MA, Encinas C, Ren S, Nguyen L, Wu S, Hewison M, Barsony J 2004 Response element binding proteins and intracellular vitamin D binding proteins: Novel regulators of vitamin D trafficking, action and metabolism. J Steroid Biochem Mol Biol 89-90: 461465.