: :fi $1 .\ . ‘r A}. ‘31.: ‘ . , 5?.139 in?“ . n. . V «murky; q , 7.5.4. :4: ,1 . x :1. x .6. jazz? 7.. : , - .15; . , ,5 ,.. fizi, 1 . 4 K “RUN?“ a . rrvsnwah; .. v! clA ‘ c1 . “Kw ”Wu. K. a... .. ‘ Drill. u... (3.25.... t a 20...}... Initial: 3.2.11. ahiizgtmmmuw . 9:...“ . Hahn.‘ 4... 4...}; a. :6 1G. 3:: hfi.‘ . . n 2.9.. an: In... sd‘.z|!‘§€. 1!. s I I . z 1" 1 1“ (053*. 3. . .. to. IV§£I1HWMH . ~ milgad.” Di: , 593.053? . '3‘ 11.3609"! 3‘3. its Ia'tafinzu n. . L.uz.:§...ti.. IASS-thnlh flffia‘rwa L: an... .s..i...1.9.!:flfi x“. .15.!!! “It!!! nit. E. , I 3. kid... . :i a... ‘igxhfiqgi I Ilsa? 1.1.1}. find}; it! . azbl‘ts! yl4ll. $1.... Ih.‘ Might-at .1- .5. .X‘} it u. {#91733. 12.... J?! as isy EA“. 2.3.2. glut}. . y 23!. . :0... n .4 . 9 i. . 96...”?! at; I; Liza... . .1 $3.3...”an 9;... £53.. j )3 3 LIBRARY Mlcahigan State n This is to certify that the dissertation entitled MECHANISMS OF ALL-TRANS RETINOIC ACID EFFECTS ON MURINE MYELOID DENDRITIC CELL ADHESION IN VITRO presented by Denise Elizabeth Lackey has been accepted towards fulfillment of the requirements for the PhD. degree in Human Nutrition Major Professor’s Signature 0 y //2 / 2 0/0 Date MSU is an Affinnative Action/Equal Opportunity Employer no-----.—.—._ _ PLACE IN RETURN Box to remove this checkout from your record. To AVOID FINES return on or before date due. MAY BE RECALLED with earlier due date if requested. DAIEDUE DAIEDUE DAIEDUE 5/08 K:IProj/Acc&Pres/ClRC/DateDue.indd MECHANISMS OF ALL-TRANS RETINOIC ACID EFFECTS ON MURINE MYELOID DENDRITIC CELL ADHESION IN VITRO By Denise Elizabeth Lackey A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Human Nutrition 2010 ABSTRACT MECHANISMS OF ALL-TRANS RETINOIC ACID EFFECTS ON MURINE MYELOID DENDRITIC CELL ADHESION IN VITRO By Denise Elizabeth Lackey Myeloid dendritic cells (DC) are professional antigen presenting cells (APC) that migrate to secondary lymphoid tissues where they activate naive T cells to proliferate and differentiate to T helper 2 cells. These cells help initiate an antibody mediated immune response. Vitamin A is known to be essential for normal immune function. All-trans retinoic acid (atRA), a bioactive metabolite of vitamin A, is necessary for DC development from myeloid progenitors in culture. An integral part of DC development in culture is the loss of adherence. We investigated the ability of atRA to modulate DC adhesion in culture. The percentage of developing DC that remained adherent was higher in the presence of a specific antagonist of retinoic acid receptor (RAR)-or compared to cells treated with atRA. AtRA treatment decreased DC surface expression of the adhesion molecule CD11a, but not the gene expression. AtRA treatment also dramatically increased gene and protein expression of the gelatinase pro-matrix metalloproteinase (pro-MMP)-9. However, gene expression and protein production of MMP-9’s primary biological inhibitor, tissue inhibitor of metalloproteinase (T lMP)-1, was unaffected by atRA treatment, altering the molar ratio of secreted pro-MMP-9:TIMP-1. Thus, atRA treatment resulted in a molecular excess of pro-MMP-9 to its primary inhibitor, indicating the possibility for increased MMP-9 enzymatic activity in atRA treated DC cultures. MMP-9 is a gelatinase essential for DC migration. We wanted to determine whether atRA treatment increases gelatinase activity in mouse myeloid DC cultures compared to cells cultured with RARa antagonist. Adding MMP-9 inhibitor significantly blocked DC gelatinase activity and increased adherence of DC in a dose dependent manner. AtRA-induced Mmp-Q expression in DC could be blocked by a transcriptional inhibitor. Hence, atRA up-regulation of Mmp-9 is primarily controlled at the transcriptional level. Since the Mmp-9 promoter contains no classic retinoic acid response element (RARE), we performed additional studies to determine how atRA regulates Mmp-9 transcription in DC. Electrophoretic mobility shift assays for the consensus Sp1, AP-1 and NF-KB binding sites located in the Mmp-9 promoter did not increase binding in response to atRA. Chromatin immunoprecipitation assays indicated that atRA increases recruitment of RARa and histone acetyltransferase p300 to, and histone H3 acetylation of, the Mmp-9 promoter. These data suggest that atRA regulates DC adhesion in vitno through MMP-9 gelatinase activity. AtRA increases Mmp-9 expression through a transcriptional mechanism involving enhancement of RARa promoter binding, recruitment of p300, and subsequent acetylation of histone H3, despite the absence of a consensus RARE. This dissertation is dedicated to my parents for their loving support and to my husband Tom for his love and daily encouragement. ACKNOWLEDGEMENTS I would like to thank my advisor, Dr. Kathleen Hoag, for her guidance throughout the process of completing my doctoral degree. She has expertly guided me through the planning and completion of my research, as well as supported me while I pursued professional development opportunities outside the laboratory. I would also like to thank my committee members, Drs. Dale Romsos, James Pestka, and Richard Schwartz, for helping me navigate degree and program requirements. Their technical suggestions have also been immensely helpful for testing my hypotheses. I would also like to acknowledge the members of Dr. Hoag’s laboratory, Shanna Ashley, M.S., David Duriancik, Ph.D., and Carmen Yu, for their help in processing samples for flow cytometry analysis and for contributing to a collegial atmosphere for pursuing our research questions. I would like to thank the National Institutes of Health as well as the Department of Food Science and Human Nutrition and the Biomedical Laboratory Diagnostic Program at Michigan State University for providing financial support for reagents and equipment. I would like to show my appreciation to the faculty and my fellow students in these programs for welcoming me into their community with openness. Finally, I would like to express my deepest gratitude to my friends and family, especially my husband, for their support and understanding as l pursued my degree. TABLE OF CONTENTS LIST OF TABLES ................................................................................................ viii LIST OF FIGURES .............................................................................................. ix LIST OF ABBREVIATIONS ................................................................................. xi CHAPTER 1 LITERATURE REVIEW ........................................................................................ 1 Introduction ........................................................................................................ 2 Vitamin A ........................................................................................................... 4 Vitamin A chemistry ....................................................................................... 4 Vitamin A discovery, digestion, storage, and transport .................................. 4 Vitamin A requirements and deficiency .......................................................... 7 Vitamin A signaling ...................................................................................... 11 Vitamin A biological functions ...................................................................... 13 Vitamin A and infection ................................................................................ 15 Vitamin A and innate immunity .................................................................... 18 Vitamin A and adaptive immunity ................................................................. 20 Vitamin A and gut immunity ......................................................................... 24 Dendritic cells .................................................................................................. 27 Matrix metalloproteinase-9 .............................................................................. 31 Matrix metalloproteinase structure and function .......................................... 31 MMP-9 and DC migration ............................................................................ 32 Regulation of MMP-9 expression ................................................................. 33 Vitamin A regulation of MMP-9 .................................................................... 37 Aims of the research ....................................................................................... 38 CHAPTER 2 RETINOIC ACID DECREASES ADHERENCE OF MURINE MYELOID DENDRITIC CELLS AND INCREASES PRODUCTION OF MATRIX METALLOPROTEINASE-9 ................................................................................. 47 Abstract ........................................................................................................... 48 Introduction ...................................................................................................... 49 Materials and Methods .................................................................................... 52 Results ............................................................................................................ 57 Discussion ....................................................................................................... 64 Acknowledgements ......................................................................................... 70 CHAPTER 3 VITAMIN A UP-REGULATES MATRIX METALLOPROTEINASE-9 ACTIVITY BY MURINE MYELOID DENDRITIC CELLS THROUGH A NON-CLASSICAL TRANSCRIPTIONAL MECHANISM ................................................................... 82 Abstract ........................................................................................................... 83 vi Introduction ...................................................................................................... 85 Materials and Methods .................................................................................... 88 Results ............................................................................................................ 96 Discussion ....................................................................................................... 99 Acknowledgements ....................................................................................... 106 CHAPTER 4 SUMMARY AND FUTURE DIRECTIONS ........................................................ 117 Summary ....................................................................................................... 118 Future directions ............................................................................................ 128 APPENDIX A .................................................................................................... 140 APPENDIX B .................................................................................................... 143 APPENDIX C .................................................................................................... 145 APPENDIX D .................................................................................................... 147 BIBLIOGRAPHY ............................................................................................... 153 vii LIST OF TABLES Table 1.1. Vitamin A requirements across the lifecycle ......................................... 9 Table 2.1. Induction of mouse matrix metalloproteinase mRNA relative to expression of receptor antagonist adherent DC in atRA treated adherent and CD11c+ floating mouse DC cultures ................................................................... 62 Table A1. Expression of all genes assayed on the Extracellular Matrix and Adhesion Molecules Gene Array after 48 h atRA rescue treatment or RARa antagonist treatment. ........................................................................................ 140 viii LIST OF FIGURES Figure 1.1. Vitamin A digestion and transport and storage ................................. 41 Figure 1.2. Classic vitamin A signaling mechanism ............................................ 43 Figure 1.3. Mouse Mmp-9 promoter with transcription factor binding sites noted ............................................................................................................................ 45 Figure 2.1. Mouse bone marrow cells cultured with RARa receptor antagonist develop an adherent phenotype .......................................................................... 71 Figure 2.2. AtRA rescue of mouse bone marrow-derived myeloid DC cultured with RARoc receptor antagonist reduces cell adherence ..................................... 73 Figure 2.3. The mouse bone marrow-derived floating myeloid DC population is enhanced by atRA rescue ................................................................................... 75 Figure 2.4. CD11a cell surface expression is reduced by atRA rescue treatment in mouse bone marrow-derived myeloid DC cultures .......................................... 78 Figure 2.5. AtRA rescue changes the gene expression and secretion of MMP-9 in mouse bone marrow-derived myeloid DC cultures ............................................. 80 Figure 3.1. DC gelatinase activity after 10 d culture with atRA treatment in the presence or absence of MMP-9 inhibitor or with RARa antagonist, AGN 194301. .......................................................................................................................... 1 07 Figure 3.2. Dendritic cell proportions and yields after 10 d culture in the presence or absence of MMP-9 inhibitor .......................................................................... 109 Figure 3.3. Adherent DC Mmp-9 expression after d 8 atRA rescue with or without transcriptional inhibitor ...................................................................................... 1 1 1 Figure 3.4. AtRA rescue of mouse bone marrow-derived adherent DC does not increase nuclear protein binding to common Mmp—9 regulatory sequences ..... 113 Figure 3.5. AtRA rescue increases RARa. association with and acetylation of histone H3 in the mouse Mmp-9 up-stream promoter region ............................ 115 Figure 4.1. Model of retinoic acid’s effect on dendritic cell development. ......... 135 Figure 4.2. Model of retinoic acid signaling to increase Mmp-9 production ....... 137 Figure B.1. AtRA rescue does not change leAM-1 secretion into cell culture medium in bone marrow-derived myeloid DC cultures ...................................... 143 Figure 0.1. Cell culture yields after 10 d culture in the presence or absence of MMP-9 inhibitor. ................................................................................................ 145 Figure 0.1. CD11a cell surface expression after 10 d culture in the presence or absence of MMP-9 inhibitor. ............................................................................. 150 LIST OF ABBREVIATIONS AGN: RARa antagonist AP: Activating protein Apo: Apolipoprotein APC: Antigen presenting cell atRA: all-trans retinoic acid AU: Arbitrary units BALT: Bronchial associated lymphoid tissue CBP: CREB-binding protein CCL: Chemokine C-C ligand CD: Charcoal-dextran filtered CH: Characterized ChlP: Chromatin immunoprecipitation COUP-TF: Chicken ovalbumin upstream promoter-transcription factor CRBP: Cellular retinol binding protein DC: Dendritic cell DMSO: Dimethyl sulfoxide DR: Direct repeat ECM: Extracellular matrix ELISA: Enzyme-linked immunosorbent assay EMSA: Electrophoretic mobility shift assay ERK: Extracellular signal-regulated kinase FBS: Fetal bovine serum xi FITC: Fluorescein isothioCyanate Foxp3: Forkhead box p3 GALT: Gut associated lymphoid tissue GAPDH: Glyceraldehyde-3-phosphate dehydrogenase GM-CSF: Granulocyte-macrophage colony-stimulating factor HAT: Histone acetyltransferase ICAM: Intercellular adhesion molecule ICE: Interleukin-1B converting enzyme IFN: Interferon Ii: Invariant chain lg: lmmunoglobulin IL: Interleukin IMDM: lscove’s modified Dulbecco’s medium ltgal: lntegrin (1L KLF: KriippeI-like factor LCAT: Lecithinzretinol acyltransferase LFA: Lymphocyte function-associated antigen LPS: Lipopolysaccharide MALT: Mucosal associated lymphoid tissue MCP: Monocyte chemoattractant protein MFI: Mean fluorescence intensity MHC: Major histocompatability complex MMP: Matrix metalloproteinase xii NCoA: Nuclear receptor coactivator NCoR: Nuclear receptor corepressor NF-KBI Nuclear factor kappa B NK: Natural killer PAF: Platelet activating factor PBS: Phosphate buffered saline P/CAF: p300/CBP associated factor PCR: Polymerase chain reaction PE: Phycoerythrin PEA3: Polyoma virus enhancer A-binding protein 3 PerCP: Peridinin chlorophyll protein PGE: Prostaglandin E PI3K: Phosphoinositide-B-kinase RA: Retinoic acid RAE: Retinol activity equivalents RAR: Retinoic acid receptor RARE: Retinoic acid response element RBP: Retinol binding protein RDA: Recommended dietary allowance RE: Retinyl ester Ral: Retinaldehyde Rol: Retinol RXR: Retinoid X receptor xiii RXRE: Retinoid X response element SMRT: Silencing mediator of RAR and thyroid hormone receptor SOCS: Suppressor of cytokine signaling Sp: Specificity protein STAT: Signal transducer and activator of transcription STRA: Stimulated by retinoic acid TGF: Transforming growth factor Th: T helper cell TIMP: Tissue inhibitors of matrix metalloproteinases TLR: Toll-like receptor TNF: Tumor necrosis factor Treg: Regulatory T cell TTR: Transthyretin UL: Upper limit xiv CHAPTER 1 LITERATURE REVIEW Introduction Vitamin A deficiency is a public health problem in developing countries, especially Africa, Southern Asia, and the Middle East. Deficiency mainly affects pre-school age children and pregnant women due to its importance in growth and development. It is estimated that 190 million pre-school aged children and 19.1 million pregnant women are deficient in vitamin A, with plasma retinol levels less than 0.7 pg/mL (WHO 2009). Vitamin A deficiency is associated with increased morbidity and mortality due to increased susceptibility to infection, especially those of the mucosal sites, including diarrheal and respiratory infections (Fawzi et al. 1993; Stephensen 2001; Villamor and Fawzi 2005). Vitamin A deficiency is known to compromise immune system function. Vitamin A deficiency impairs neutrophil and macrophage migration, phagocytosis, and bacterial killing ability (Twining et al. 1997; Stephensen 2001). At the same time, vitamin A deficiency leads to increased expansion of neutrophil precursors with impaired differentiation (Walkley et al. 2002). Vitamin A deficiency decreases natural killer cell number and function in rats, especially in aging animals (Dawson et al. 1999). Vitamin A deficiency also affects the adaptive immune system. Vitamin A deficiency decreases B cell production of the immunoglobulin (lg)G1, while addition of vitamin A sufficient T cells restores this production (Carman et al. 1989). IgG1 production requires a T helper (Th) 2 polarized response. Addition of all-trans retinoic acid (atRA), the bioactive metabolite of vitamin A to cell cultures enhances naive T cell development to Th2 through an antigen presenting cell (APC) intermediate (Hoag et al. 2002). Myeloid dendritic cells (DC), whose development from myeloid precursor cells requires retinoic acid (Hengesbach and Hoag 2004), are a likely cell type that helps polarize naive T cells to T helper 2 cells. Immature myeloid DC are resident in the peripheral tissues and capture, process, and present antigen. Maturation in response to antigen induces migration to lymphoid tissues where myeloid DC present antigen in their major histocompatibility complex (MHC) class I and II molecules to naive T cells. This results in polarization of the naive CD4+ T cell to Th2 cell, augmenting development of humoral immunity (Bastien and Rochette-Egly 2004; Hochrein and O'Keeffe 2008). DC migration through the extracellular matrix is dependent on the activity of the gelatinase matrix metalloproteinase (MMP)-9 (Ichiyasu et al. 2004). In the studies undertaken here, we attempted to determine how atRA regulates the myeloid DC adhesive phenotype and migration potential, with particular focus on retinoic acid-induced matrix metalloproteinase-9 expression by these cells. Vitamin A Vitamin A chemistry Retinoids, natural and synthetic molecules of the vitamin A family, consist of a B-ionone ring, a polyunsaturated isoprenoid chain, and a polar end group. In naturally occurring retinoids, the polar and group can be an alcohol (retinol), an aldehyde (retinal), or a carboxylic acid (retinoic acid). The various domains of the molecules in the vitamin A family make the compounds hydrophobic and susceptible to oxidation, photodegradation, and isomerization (Barua and Furr 1998; Noy 1999). The hydroxyl group on retinol can be esterified by long chain fatty acids and is more stable in this form. Both retinol and retinyl esters have a characteristic UV maximum of 325 nm (Ross 1999). Vitamin A disco very, digestion, storage, and transport Vitamin A, first identified in 1913 as a fat-soluble factor by Osborne and Mendel (1913) and McCollum and Davis (1913), is found in the form of retinyl esters in a variety of animal-derived foods such as egg yolk, fatty fish, dairy/cheese, and meat. In addition, pro-vitamin A B-carotene is obtained from orange and yellow fruits and vegetables and dark green leafy vegetables (Gibson 2007). Retinyl esters are cleaved in the lumen of the small intestine to retinol by pancreatic triglyceride lipase, cholesterol ester hydrolase, pancreatic lipase related proteins 1 and 2, as well as phospholipase B on the brush border membrane. Retinol is then absorbed by enterocytes. B-carotene is cleaved by B- carotene-15,15’-monooxygenase and converted to all-trans-retinol via enzymatic reduction from all-trans-retinal in the intestinal epithelium (von Lintig and Vogt 2004; Harrison 2005). Carotenoids can also be absorbed into the enterocyte intact and cleaved to all-trans-retinol inside the enterocyte (Vogel et al. 1999). Once inside the enterocytes, retinol, associated with cellular retinol binding protein (CRBP)-II, is re-esterified by lecithinzretinol acyltransferase (LRAT) and incorporated into nascent chylomicrons containing apolipoproteins (Apo) B43, E, and C (O'Byrne et al. 2005). These chylomicrons are absorbed by the lymphatics, enter the blood, and circulate through the body (Quadro et al. 2003). After triglycerides are broken down by lipoprotein lipase and the chylomicrons have lost ApoC, the resulting chylomicron remnants containing retinyl esters are taken up by the hepatocytes through recognition of ApoE (Vogel et al. 1999; Harrison 2005). The retinyl esters are hydrolyzed to retinol within the hepatocytes and cellular retinol binding protein (CRBP)-l aids in the transfer of retinol to the hepatic stellate cells. Retinol is again esterified within the hepatic stellate cells for storage purposes (Quadro et al. 2003). For retinol mobilization to the plasma, retinyl esters stored in the stellate cells are hydrolyzed to all-trans-retinol and packaged with retinol binding protein (RBP, synthesized by the hepatocytes) for secretion (Harrison 2005). RBP binds retinol in a hydrophobic pocket to solubilize it in the aqueous blood environment as well as to protect it from oxidation (Noy 1999). The physiological concentration of retinol in the plasma is 1-2 x 10’6 M (Ross and Hammerling 1994). In the circulation, the RBP-retinal complex further associates with transthyretin (TTR, synthesized in and secreted from the liver) to prevent retinol excretion by the kidney (van Bennekum et al. 2001). Retinyl esters present in chylomicrons can also serve as a source of retinol to non-hepatic tissues (Quadro et al. 2003, Figure 1.1). Alternatively, because of the importance of vitamin A in the maintenance of epithelial tissues, it has been shown that epithelial cells can locally store retinyl esters for periods where retinol is in high demand and RBP synthesis cannot occur quickly enough to supply a continuous plasma retinol concentration (Biesalski and Nohr 2004). Retinol is taken up by cells through the multi-transmembrane domain protein stimulated by retinoic acid (STRA)6, which binds directly to the RBP portion of the RBP-retinol complex (Kawaguchi et al. 2007). Natural STRA6 human polymorphisms associated with congenital defects such as mental retardation, congenital heart defects, lung hyperplasia, and intrauterine growth retardation, result from reduced retinol uptake by the target cells (Kawaguchi et al. 2008). Within the cell, retinol can be reversibly oxidized to retinaldehyde by retinol dehydrogenase, which is then further oxidized to retinoic acid by retinaldehyde dehydrogenase in an irreversible reaction. Retinoic acid can also be found in circulation after secretion by a cell that converted retinol to retinoic acid. The brain, liver, kidney, adipose tissue, seminal vesicle, and spleen can all obtain retinoic acid from the circulation (Kurlandsky et al. 1995). Retinoic acid is protonated in most biological (slightly acidic) membranes and can thus transverse the membranes for direct cellular uptake (Noy 1999). CYP26, a member of the cytochrome P450 family, contains a retinoic acid response element (RARE) sequence in its promoter region and is transcriptionally expressed in response to atRA signaling through the retinoic acid receptor (RAR)-retinoid X receptor (RXR) complex. CYP26 acts to oxidize atRA to less active polar metabolites 4-hydroxy-, 4-oxo-, and 18-hydroxy-RA for negative regulation of atRA signaling. The polar RA metabolites are then glucuronidated for elimination (Napoli 1999; Ross 2003). LRAT also acts to limit the pool of available retinol and is also up-regulated by atRA signaling. LRAT enzymatically converts retinol to retinyl esters. When the concentration of retinol bound CRBP is decreased with an increased concentration of apoCRBP, LRAT conversion of retinol to retinyl esters is inhibited to maintain sufficient cellular levels of retinol. Conversion of retinol to retinyl esters by LRAT functions to limit the extent of retinoic acid signaling in the cell by limiting the availability of retinol to be converted to retinoic acid (Ross 2003). Vitamin A requirements and deficiency The requirements for vitamin A vary by life cycle stage and sex and is expressed in retinol activity equivalents (RAE). One RAE is equivalent to 1 pg retinol or 12 pg B-carotene. The adequate intake of vitamin A for infants is 400- 500 RAE, based on the average daily intake from breast milk. During childhood, vitamin A requirements increase based on extrapolation from adult requirements with increased allowances for growth (IOM 2001). The recommended dietary allowance (RDA) of vitamin A climbs to 700 RAE and 900 RAE by 14 years of age for females and males, respectively, based on the concentration of retinol in the liver, the catabolic rate, the proportion of liver weight to total body weight, percent of liver vitamin A that is stored, and the ratio of liver vitamin A to total body vitamin A (McLaren 1994). During pregnancy and lactation the RDA increases to 770 RAE for fetal growth and 1300 RAE for secretion into milk, respectively (IOM 2001). Tolerable upper intake levels of vitamin A for adults occur at 3000 RAE but are as low as 600 RAE for infants and young children (Table 1.1). Table 1.1. Vitamin A requirements across the lifecycle. RDA, recommended dietary allowance; UL, upper limit. Age RDA1’2 UL1’2 Infants 0-6 mo 400 600 7-12 mo 500 600 Children 1-3 y 300 600 4-8 y 400 900 Males 9-13 y 600 1700 14-18 y 900 2800 >18 y 900 3000 Females 9-13 y 600 1700 14-18 y 700 2800 >18 y 700 3000 Pregnancy 518 y 750 2800 219 y 770 3000 Lactation 518 y 1200 2800 219 y 1300 3000 1Units in pg retinol activity equivalents (RAE)/d. 2Values are adapted from those published by the Institute of Medicine, the National Academies, 1998. It is estimated that 127 million pre-school aged children (between 6 months and 5 years of age) are vitamin A deficient. Of these, 4.4 million have xerophthalmia, the primary clinical manifestation of vitamin A deficiency presented as night blindness or Bitot’s spots (West 2002). Xerophthalmia can progress to corneal scarring and permanent blindness if vitamin A deficiency is not treated (Rastogi and Mathers 2002). At the same time, 19.8 million pregnant women have low vitamin A status, of whom 7.2 million show frank vitamin A deficiency (West 2002). Pre-school aged children and child-bearing women are g at risk for vitamin A deficiency due to the increased needs during periods of significant growth (WHO 1999). Aside from vision impairment, vitamin A deficiency is an important public health problem because it increases mortality and morbidity due to the increased susceptibility to infection (Villamor and Fawzi 2005). Vitamin A supplementation, either at periodic high doses or more frequent lower doses, can lead to decreases in general childhood mortality in developing countries (Fawzi et al. 1993; Villamor and Fawzi 2000). In a meta-analysis of the efficacy of vitamin A supplementation to prevent early childhood mortality in hospitalized measles patients and in the community-based setting, the authors found vitamin A supplementation in addition to standard treatment in the hospital setting was significantly protective against respiratory infection mortality, even in populations not commonly found to be vitamin A deficient, with greatest effect on children less than one year of age. In the community setting, both small frequent and large infrequent doses of vitamin A reduced childhood mortality, with a 30% 10 reduction in diarrhea-associated mortality and a smaller reduction in measles- associated mortality (Fawzi et al. 1993). Vitamin A deficiency leads to failure of the protective mucosal epithelial barriers. Because mucosal epithelium lines the gastrointestinal, respiratory, and urogenital tracts, it is not surprising that vitamin A deficiency has been associated with increased diarrheal disease and dysentery, urinary and reproductive tract infections, and severity of measles virus lung infections (West 2003; West 2004). Vitamin A signaling Retinoic acid, in the form of atRA, is the main bioactive metabolite of vitamin A used for cell signaling. In the traditional model of retinoic acid signaling, atRA enters the cell nucleus and binds to one of the three RAR isoforms, a, B, or y (Chambon 1996). Upon binding, the complex heterodimerizes with one of the three RXR isoforms, (1, B, or 7. After binding of the atRA ligand to the RAR/RXR heterodimer, the holoRAR/RXR complex binds to RARE in the promoter regions of retinoic acid responsive genes and associate with coregulatory proteins leading to promotion or inhibition of transcription and target gene expression (Bastien and Rochette-Egly 2004; Mark et al. 2006). The canonical RARE, called a direct repeat (DR)5, is the sequence PuGG/TT CA, followed by a direct repeat after five intervening nucleotides (Figure 1.2). The holoRAR/RXR complex is also able to bind to DR1 and DR2 RARE sites, though these are less common. Variation in the RARE structure allows for variable RAR/RXR binding characteristics,‘ and thus variable responses to atRA (Piedrafita and Pfahl 1999). RXRs are also able to homodimerize upon binding to the appropriate ligand, 11 such as 9-cis-RA and other synthetic RXR retinoids. The holoRXR/RXR dimer binds to the retinoid X response element (RXRE) and acts similarly to effect responsive genes (Piedrafita and Pfahl 1999). However, 9-cis-RA, though reported in animal tissues in earlier studies (Heyman et al. 1992), has not been confirmed to exist as a natural retinoid in vivo. In fact, it is possible the earlier studies that detected endogenous 9-cis-RA actually detected 9-cis-RA spontaneously isomerized from atRA (Wolf 2006). The RAR molecules bind atRA with strong affinity; Kd 0.2—0.7 x 10.9 M (Noy 1999). Binding of atRA to form a holoRAR/RXR complex initiates association with coactivator proteins such as CREB-binding protein (CBP)/p300, nuclear receptor coactivator (NCoA), and p300/CBP-associated factor (P/CAF), all which possess histone acetyl transferase (HAT) activity. Upon histone acetylation, transcriptional machinery is able to bind to the retinoid responsive gene for transcription initiation (Piedrafita and Pfahl 1999). Unliganded RAR binds DNA and associates with corepressors, including nuclear receptor corepressor (NCoR)-1 and NCoR-2/silencing mediator of RAR and thyroid hormone (SMRT) at the corepressor interaction domain of RAR to recruit histone deacetylases leading to chromatin condensation and transcriptional repression (Weston et al. 2003). At the same time, orphan receptors that bind RAREs as homodimers with high affinity, such as chicken ovalbumin upstream promoter-transcription factor (COUP-TF), can block binding of the holoRAR/RXR complex to the RARE and inhibit transcription of retinoic acid responsive genes (Piedrafita and Pfahl 1999). 12 The classical model of retinoic acid signaling explains direct transcriptional regulation by vitamin A for up to 132 genes. However, it is known that at least 532 genes are controlled by atRA, but not all are regulated through the traditional mechanism. Some are regulated by atRA via another transcription factor and some are transcriptionally regulated by atRA, but not through the classical mechanism (Balmer and Blomhoff 2002). The holoRAR-RXR complex has been shown to transrepress activating protein (AP)-1 binding in target promoters to down-regulate gene expression (Blomhoff and Blomhoff 2006). The holoRAR- RXR complex has also been shown to decrease nuclear factor-kappaB (NF-KB) promoter binding activity and vitamin A deficiency increases whole body NF-KB activity in mice (Austenaa et al. 2004; Blomhoff and Blomhoff 2006). The liganded RAR-RXR complex has been shown to interact with the transcription factor specificity protein 1 (Sp1) to augment expression of multiple genes, including urokinase plasminogen activator, transglutaminase, transforming growth factor (T GF)-B1, types I and II TGF-B receptors, and bone morphogenic protein-2, despite the lack of a traditional RARE site in the promoters (Suzuki et al. 1999; Shimada et al. 2001; Xu and Rogers 2007). Vitamin A biological functions Vitamin A is important in a large variety of biological processes, including vision, reproduction, body patterning in embryo development, immune function (Stephensen 2001), and, in the pro-vitamin A form of B-carotene, as an antioxidant. The role of vitamin A in vision is very well characterized, where 11- cis-retinaldehyde, with opsin, forms rhodopsin in the rod cells of the eye for 13 conversion to all-trans-retinaldehyde by photon stimulation during the vision cycle (Wolf 2001). In embryo development, RA signaling through gradient diffusion is required for bilateral symmetry and synchronous somite formation along the left/right body axis as well as control of the position of the traveling determination front along the anteroposterior axis (Campo-Paysaa et al. 2008). RA is synthesized from retinaldehyde in different tissues in a time— and tissue- dependent manner, controlled by expression of retinaldehyde dehydrogenase enzymes (Duester 2007). It is essential for development of the neural system, including neurite outgrowth, neuron patterning, and pathfinding (Clagett-Dame et al. 2006; Duester 2007), heart morphogenesis, late-stage asymmetry, looping, and vasculature formation (Zile 2004), and forelimb buds and eyes (Duester 2007; Campo-Paysaa et al. 2008). Vitamin A is also required for proper chondrocyte formation, hypertrophy, and differentiation in long bone growth plates, probably through induction of bone morphogenic protein expression (Adams et al. 2007). Expression of RARa is increased during mesenchymal cell differentiation to chondroprogenitor cell, while RARy is increased in the differentiation of the chondroprogenitor cell to the chondroblast. Expression of retinaldehyde dehydrogenase, necessary for the oxidation of retinaldehyde to RA for signaling through RAR, is increased in the mesenchymal cell for commitment to chondroprogenitor cell (Weston et al. 2003). Vitamin A deficiency decreases the number of mucus producing goblet cells in the mucosal epithelium (Stephensen 2001). Vitamin A’s importance in cellular differentiation, especially in cells with a high turnover rate such as at the 14 epithelium, results in keratinizing metaplasia and cellular separation and distortion even with mild vitamin A deficiency (Congdon and West 2002). It has been shown that intestinal cells cultured in vitro supplemented with retinol have decreased apoptosis and necrosis and increased transepithelial electrical resistence, cell proliferation, and migration in the presence of Clostn'dium difficile toxin A than do intestinal cells cultured with the toxin alone (Maciel et al. 2007). In vivo, it was also found that vitamin A deficient mice had significantly fewer goblet cells and upon infection with rotavirus, the deficient group showed complete destruction of the small intestine villi tips. This pathologic change to the villi tips was not found in uninfected deficient mice or infected vitamin A sufficient mice (Ahmed et al. 1990). Rats consuming a vitamin A deficient diet for seven weeks have increased amount of bacteria in the ileum and the colon with decreased amounts of Lactobacillus spp. and increased amounts of E. coli in the gut. The vitamin A deficient rats also have decreased levels of MUCZ and defensin 6 mRNA with increased levels of toll-like receptor (T LR)2 and TLR 5 mRNA in the small intestine and the colon. Decreased defensin 6 abundance can allow for bacterial overgrowth and penetration of the mucosal barrier (Amit- Romach et al. 2009). This dissertation will focus on the role of vitamin A in normal immune system myeloid dendritic cell function. Vitamin A and infection One very important role of vitamin A, in the form of atRA, is in the maintenance of the immune system and the protection against infection. Vitamin A was described as an anti-infective vitamin early. Rats that were maintained on 15 a vitamin A deficient diet were commonly found to have tongue abscesses, gastrointestinal tract infection, and kidney or bladder infection, with some examples of broncho—pneumonia (Green and Mellanby 1928). In a prospective study of children in Indonesia reported in 1983, children with Bitot’s spots or night blindness had six-fold and three-fold greater mortality rate, respectively, due mainly to diarrheal and respiratory disease over an 18 month period. In children with both symptoms, the mortality rate was increased over eight-fold (Sommer et al. 1983). Because vitamin A deficiency compromises the mucosal barriers of the eye, respiratory, and gastrointestinal tract (Stephensen 2001), the physical barriers that are the first defenses against infection, deficient individuals are more likely to suffer infections at these mucosal sites. One study has shown that children who had suffered from pneumonia, acute respiratory problems, and Mycobacten’um bovis BCG vaccine scarring (indicating an overactivated Th1 response) had lower plasma retinol levels than children who had not experienced these problems (Jason et al. 2002). In a study of Brazilian children, low retinol concentrations were correlated with increased urinary Iactulosezmannitol ratio from the lactulose—mannitol test, an indicator of intestinal permeability. The authors also measured total serum lgG in mice with varying serum concentrations of retinol due to RBP or 'I‘I'R deficiency. They found that the concentration of serum lgG decreased in mice with lower serum concentrations of retinol without a change in spleen, thymus or bone marrow populations of B cells, CD4+ T cells or CD8+ T cells. However, the authors did not determine 16 whether there was a difference in CD4+ Th cell polarization that could explain differences in serum lgG concentrations (Quadro et al. 2000). Vitamin A deficient children infected with diarrheal diseases have a decreased secretory lgA response (Stephensen 2001). Vitamin A status has been shown to be related to the severity of a measles infection, a respiratory virus, with high levels of vitamin A supplementation supporting a beneficial Th2 response and significantly decreasing mortality in children (Semba 1999). Secretory lgA response in the intestines is also required to respond to and clear infections causing diarrhea. When diarrheal disease is caused by Shigella infection, vitamin A supplemented children had shorter duration of disease, and also showed increased secretory lgA response (Villamor and Fawzi 2000). Vitamin A supplementation has also been shown to decrease the incidence of malaria in children between the ages of 1 and 3 years (Semba 1999), and increase the antibody response to the hepatitis B vaccine in infants (Newton et al. 2007). In a mouse model of allergic airway inflammation, a Th2 response dominated disease, vitamin A deficiency has also been linked with decreased severity of symptoms by decreasing serum levels of IgG1 and lgE and decreased levels of interleukin (lL)-4 and lL—5 in the bronchoalveolar lavage, while high dosage of dietary vitamin A increased serum lgE and pulmonary hyperresponsiveness (Schuster et al. 2008). In many cases, vitamin A supplementation does not show beneficial results on immune measures in children with vitamin A deficiency because they also have other nutrient deficiencies, particularly protein-energy malnutrition. In a 17 study of vitamin A deficient children between 2-6 years who are otherwise nutrient sufficient, researchers found that vitamin A deficient children (serum retinol <30 pg/dl) have lower levels of serum lgA and complement factor C3, with higher levels of lysozyme. High dose vitamin A supplementation for 3 months normalized levels of serum IgA, C3, and lysozyme (Lin et al. 2007). Similarly, in children of normal height between 1 and 5 years-old, vitamin A supplementation of 200,000 IU decreased the prevalence and intensity of reinfection with the gastrointestinal nematode Ascan's, but not in children with stunted growth, indicating that protein-energy malnutrition complicates the effectiveness of vitamin A supplementation (Payne et al. 2007). These studies indicate that vitamin A status is an important factor in the duration and overall outcdme during a pathogen infection in children and pregnant women. Understanding the mechanisms of vitamin A’s effects on cells of the immune system to prevent or fight infection is crucial to further refine supplementation schedules and create new treatments for children at risk for infection due to malnutrition. Vitamin A and innate immunity Vitamin A deficient rats have been shown to sequester retinol in the bone marrow, with concentrations four-fold higher than in the bone marrow of vitamin A sufficient rats (Twining et al. 1996). Deficiency also compromises the ability of macrophages and neutrophils to migrate to sites of infection, phagocytose, and kill bacteria (Twining et al. 1997; Stephensen 2001). However, neutrophils are found in significantly increased number in the spleen, peripheral blood, and bone marrow of vitamin A deficient rodents (Kuwata et al. 2000) and in vivo pan- 18 retinoic acid receptor (RAR) antagonism leads to neutrophil precursor cell expansion with impaired differentiation (Walkley et al. 2002). Treatment of human eosinophils with high concentrations (1 pM) of RA increases cell survival through down-regulation of the apoptotic enzyme caspase-3 in vitro (Ueki et al. 2008). CYP26 is an enzyme that oxidizes atRA for removal from the system. The Cyp26 gene is induced upon atRA treatment as a negative feedback mechanism. In rats treated with Iipopolysaccharide (LPS) after receiving an oral dose of atRA, liver expression of Cyp26 is abrogated compared to rats treated only with atRA. Treatment with only LPS has no effect on baseline Cyp26 expression. These results indicate that excess concentrations of atRA are important during inflammatory situations, so mechanisms exist to inhibit its oxidation (Zolfaghari et al. 2007). Even marginal vitamin A status has been shown to decrease natural killer (NK) cell number and function in rats, especially with increased age (Dawson et al. 1999). In adult men, whole body vitamin A stores are positively associated with peripheral blood NK cell and NKT cell number and monocyte oxidative burst intensity (Ahmad et al. 2009). Cells treated in vitro with atRA augment expression of the gene for CD1d, a non-classical antigen presenting molecule that can activate NKT cells through presentation of glycolipid antigens, through holoRARor/RXRa interaction with the RARE in the CD1d promoter region. It was also found that atRA treatment enhanced cultured spleen cell proliferation in response to the glycolipid antigen a—GalCer, indicating that CD1d up-regulated by atRA may increase glycolipid presentation to NKT cells (Chen and Ross 19 2007). These studies all indicate the importance of vitamin A in the regulation of the anti-viral and anti-tumor activities of NK and NKT cells. Vitamin A and adaptive immunity Vitamin A deficiency has profound effects on the adaptive immune response. Antibody mediated responses are one aspect of adaptive immunity affected by vitamin A deficiency. In a number of studies reviewed by Ross and Hammerling, vitamin A deficient rats make impaired IgM and lgG responses to primary and secondary vaccination with tetanus toxoid, but the secondary response can recover if the animals are repleted with retinol prior to secondary immunization (Ross and Hammerling 1994). Overall, vitamin A deficiency decreases antibody production in response to T cell-dependent antigens and type 2 T cell-independent antigens but has not shown an effect on antibody production in response to type 1 T-cell independent antigens (Ross and Hammerling 1994). For example, vitamin A deficiency decreases the specific IgM response to the protein antigen hemocyanin despite no change in B cell number in the spleen or lymph nodes and no difference in total IgM level in more deficient animals (Smith et al. 1987). While vitamin A deficiency decreases the frequency of B and T cells in the spleen, peripheral blood, and bone marrow due to neutrophil infiltration, the actual B and T cell yields in these tissues were not changed by deficiency (Kuwata et al. 2000). Vitamin A deficient mice also make an impaired IgG1 response after stimulation with hemocyanin antigen. This deficit is due to a defect in Th cells from vitamin A deficient mice. T cells from vitamin A deficient mice showed normal proliferation but decreased frequency. Addition of 20 T cells from vitamin A deficient mice to B cells from vitamin A sufficient mice resulted in diminished lgG1 production while addition of T cells from vitamin A sufficient mice to B cells from vitamin A deficient mice resulted in a normal response. The T cell defect was reversible upon addition of retinyl acetate to cultures, indicating that vitamin A is required for Th cell stimulation of B cells to produce an appropriate humoral response to antigen (Carman et al. 1989). After oral infection with rotavirus, vitamin A deficient mice show decreased serum levels of antibody specific to rotavirus (Ahmed et al. 1991). Vitamin A deficiency can also change the predominant antibody type produced during an infection. During influenza A infection in mice, vitamin A deficiency has been shown to decrease influenza A—specific salivary lgA production while increasing the total salivary lgA concentration and the influenza A specific serum lgGZa concentration (Stephensen et al. 1996). Vitamin A deficiency decreases the delayed type hypersensitivity response (a test used as a measure of Th1 function) to dinitrofluorobenzene in moderately deficient mice. However, results of the delayed type hypersensitivity response can be complicated by changes to multiple processes involved in the reaction and the decreased response seen here is likely due to impaired APC activity or development. Supplementation with vitamin A increases the delayed type hypersensitivity response to ovalbumin (Stephensen 2001), and also increases the serum antibody production in children after diphtheria and tetanus toxoid vaccination (Stephensen 2001). 21 Vitamin A signaling is known to prevent the development of a proinflammatory environment through the regulation of na'ive T cell development. In mice, vitamin A deficiency enhances the Th1 response to Tn'chinella spiralis infection through increased secretion of IF N-y, while suppressing the Th2 response through decreased production of lL-5 and lL-10, leading to an impaired humoral immune response (Cantorna et al. 1996). AtRA, the bioactive metabolite of vitamin A, has been shown to inhibit the synthesis of the Th1 cytokine IFN-y in Th1 cell culture and enhance the development of CD4+ T cells into Th2 cells through an APC intermediate (Hoag et al. 2002). Evidence exists that myeloid DC are the relevant APC whose function is modified by vitamin A (Hengesbach and Hoag 2004). The increased production of Th2 cytokines and decreased production of Th1 cytokines after stimulation with atRA mainly occurs through RARa signaling. Peripheral blood mononuclear cells treated with atRA decrease production of the Th1 differentiation cytokine lL-12p70 (Dawson et al. 2008). It has been found that supplementing mice with atRA after a DNA vaccination leads to Th2 polarization with a greater antibody response than mice not supplemented with atRA (Y u et al. 2005). Similarly, vitamin A deficiency has been found to lessen the symptoms of ovalbumin-induced allergic airway inflammation in mice while high-dose vitamin A supplementation increased the symptoms through modification of the Th1/Th2 balance. Vitamin A deficiency led to decreased serum lgE and IgG1 antibody, eosinophil, IL-4, and lL-5 concentrations in the bronchoalveolar lavage fluid, and decreased lung resistance after ovalbumin aerosol challenge. High-dose supplementation led to 22 the opposite responses, resulting in a response skewed toward the Th2 response (Schuster et al. 2008). AtRA also indirectly affects the inflammatory balance during na‘r've CD4+ T cell development through suppression of CD4+CD44+ memory T cells allowing TGF-B signaling to induce forkhead box protein 3 (Foxp3) expression (Hill et al. 2008) and decreases lL-10 and lL-17 expression in naive CD4+ T cells to produce regulatory T (Treg) cells (Maynard et al. 2009). Treatment of CD4+ T cells with TGF-B in the presence of RAR inhibitors increases lL-10 and lL-17 expression and decreases Foxp3 expression (Maynard et al. 2009). Conversely, it has also been found that vitamin A deficient mice show increased percentage of CD4+ memory T cells that produce lL-10 and decreased percentage of CD4+ memory T cells that produce the Th1 cytokines IF N-y and lL-2 from the spleen and draining lymph nodes 2-5 weeks post immunization. The researchers explain that their data are not consistent with the findings of other studies because the adjuvant used for immunization creates less Th1 bias at the time of immunization than other traditionally used adjuvants. They also explain that cells cultured for cytokine analysis were cultured in the presence of fetal bovine serum containing retinol. Overall, the researchers state these factors could lead to a response that is less biased to a Th1 response in the early stages (Stephensen et al. 2004). These studies indicate that while vitamin A deficiency traditionally leads to a Th1- biased response, the situation is very complex and can vary by severity of deficiency, type of immune challenge, and the tissues and/or cells under study. 23 B cell proliferation and differentiation are also influenced by atRA treatment. In one model of B cell differentiation to simulate activation by Th2 cells, stimulation of the B cell receptor and CD40, with lL-4 and atRA treatment increased the subset of large, less-proliferative, differentiated splenic B cells. These cells developed cell surface lgG1 expression through increased expression of activation-induced cytidine deaminase, Blimp-1, and cell surface CD138 and decreased expression of paired box gene-5 and gerrnline transcript genes. AtRA treatment also decreased the proportion of small, proliferative splenic B cells (Chen and Ross 2005; Chen and Ross 2007). These results suggest that atRA treatment increases the B cell humoral response upon activation with Th2 cells. In another experimental model, cultures of purified early B lineage cells were treated with slightly greater concentrations of atRA than physiologically normal. This treatment decreased total cell yield but increased yields of CD19+ B cells after exposure to atRA for at least 3 days and optimally for 5-7 days. It was found that RARa signaling increases gene expression of early B cell factor-1 and paired box gene-5, both required for B cell lymphopoiesis (Chen et al. 2008). Vitamin A and gut immunity A new line of research has recently emerged focusing on the role of vitamin A on gut immunity and mucosal tolerance through myeloid DCs. DC found in the gut associated lymphoid tissues (GALT) have been shown to produce atRA from retinol. Uptake of the atRA by nearby T cells leads to atRA 24 signaling through RAR in the T cells and increased expression of (x487 on the T cell surface for selective homing to the gut (lwata et al. 2004). Similarly, atRA produced by mesenteric lymph node DC leads to C025+ CD4+ Treg cell gut homing by increasing expression of (1487 for migration to mucosal tissues (Siewert et al. 2007). Stromal cells in mesenteric lymph nodes, but not the peripheral lymph nodes, recently have been found to produce atRA from retinol and induce CCR9 expression in T cells for homing to the gut (Hammerschmidt et al. 2008). Splenic DC have also been shown to produce atRA from retinol through induction of retinal dehydrogenase mRNA production upon signaling through TLR2. Increased atRA signaling induces expression of suppressor of cytokine signaling (SOCS)-3, resulting in decreased production of inflammatory cytokines and increased Treg cell differentiation (Manicassamy et al. 2009). AtRA produced by GALT-DC, along with IL-5, lL-6, and the GALT-DC themselves, are necessary to induce T cell-independent B cell lgA secretion in the gut. The authors also found that 0:487 and CCR9 expression for gut homing was dependent upon atRA producing GALT-DC. They further report that vitamin A deficiency in mice leads to loss of gut homing B cells and decreased lgA production in the gut due to decreased DC atRA production (Mora et al. 2006). DC treated with atRA also directly affects the gut environment. It has been found that human peripheral blood-derived DC treated with synthetic atRA produce decreased IL-12p70 concentration and decreased Th1 polarization function. Mice with induced colitis treated with synthetic atRA had decreased colitis induced IL- 25 12 in the intestinal lamina propria (Wada et al. 2009). TGF-B has been shown to promote the development of Treg cells when alone, but with interleukin (lL)-6, it promotes the development of the inflammatory, lL-17 secreting Th17 cells. It has been found that atRA can block the Th17 promoting effects of IL-6 through blockade of signal transducer and activator of transcription (STAT)-3 signaling which normally acts to up-regulates lL-17 and through RAR interaction with STAT5, augmenting Foxp3 transcription (Mucida et al. 2007; Mucida et al. 2009). Splenic cos+cozos+ DC are specialized DC that produce TGF-B to induce Foxp3+ Treg cell differentiation (Yamazaki et al. 2008). Macrophages present in the lamina propria induce development of Treg cells through IL-10, atRA, and TGF-B, while lamina propria DC induce the development of Th17 cells. The lamina propria macrophages can interact with the DC to prevent the development of Th17 cells and promote Treg cell development (Denning et al. 2007). 26 Dendritic cells DC are professional antigen presenting cells. They are the only APC able to activate naive T cells by presenting antigen and secreting specific cytokines to yield an adaptive immune response (lwasaki 2007). In the mouse, all DC are CD11c+, but subsets can be identified by patterns of other proteins expressed on their surface. These subsets are myeloid (CDBa' CD11b+), lymphoid (CD11b- + . + + - low CD8or ), plasmacytord (CD45RB Gr—1 ), and Langerhans DC (CD4 CDBa CD11b+ CD45RB- 00205”) which reside in the skin epidermis. The myeloid DC and lymphoid DC are both referred to as conventional DC (cDC) (Shortman and Liu 2002; Wilson and O'Neill 2003; Heath et al. 2004). Different subsets of DC can polarize T helper cells in the direction of either a Th1 or Th2 dominated response. Stimulation by lymphoid DC, partially through lL-12 production, influences na'i've Th0 cell polarization in the direction of Th1 differentiation, where lFN-y and lL-2 are produced by the activated Th1 cells at high levels. When lymphoid DC are activated, they also produce high levels of IL- 6 and TNF-or. These DC can efficiently cross-present exogenous antigen on cell- surface major histocompatibility complex (MHC) class I molecules for stimulation of CD8+ cytotoxic T cells and the adaptive immune response against viral infections. In contrast, stimulation by pathogen activated myeloid DC induces na'r‘ve Th0 cell polarization in the direction of Th2 response, possibly through secretion of monocyte chemoattractant protein (MCP)-1 and cell-surface OX40L, 27 leading to increased production of the Th2 associated cytokines lL-4 and lL-10. Myeloid DC are the most efficient stimulators of CD4+ Th cells through stimulation with antigen bound to MHC class II (Moser and Murphy 2000; de Jong et al. 2005; Hochrein and O'Keeffe 2008). Myeloid DC are derived from the bone marrow and reside in an immature stage in the peripheral tissues (Shortman and Liu 2002). Most immature DC in low/- the bone marrow are identified as CD11cint MHC-ll CD11b+ CD4' CD8 (Hochrein and O'Keeffe 2008). The DC precursors migrate from the bone marrow, via the blood, into the different lymphoid tissues (including the spleen, lymph nodes, and various MALT, depending on the chemokine signals expressed by these tissues) where they differentiate (lwasaki 2007). However, the origin of DC subtypes based on CD8a expression has come into question. Researchers have shown through donor studies that common myeloid progenitor cells from donor mice can give rise to CD80t+ (lymphoid) DC in the spleen and thymus of lethally irradiated mice (Traver et al. 2000). It has also been argued that both CDBa' (myeloid) and CD80:+ (lymphoid) DC can arise from a common lymphoid progenitor cell because both cell types could be found in the spleens of lethally irradiated mice after transfer of donor CD4low lymphoid precursor cells from the thymus (Martin et al. 2000). These researchers later discovered a CD11c+ MHC II' DC precursor population which constitutes 5% of blood mononuclear cells that could give rise to all DC subtypes including CD11b+ DC, 28 CD80t+ DC, and CD45RB+ plasmaCytoid DC. They found this precursor population was unrelated to myeloid and lymphoid lineages but carried early precursor markers (del Hoyo et al. 2002). Taken together, these studies question the traditional theories of DC ontogeny, where CD11b+ DC were thought to be always myeloid lineage and CD80t+ DC were thought to be always lymphoid Hneage. The general belief that immature DC are only generated in the bone marrow and then migrate into the tissues has also been called into question. A recent study has shown that DC resident in the spleen and the lymph nodes can be replenished by completing a limited number of cell divisions as well as by DC precursors from the bone marrow (Liu et al. 2007). The immature myeloid DC are able to effectively capture and process antigens and present them in their MHC classes I and II (Banchereau et al. 2000). Maturation of DC through stimulation by antigen or stress signals from self-tissues leads to DC migration to the T cell areas of the secondary lymphoid tissues, where they stimulate and activate both naive CD4+ and CDB+ T cells. The activation of T cells leads to adaptive immune responses and immune memory development (Banchereau and Steinman 1998; Steinman 1999). DC are activated through interaction of toll like receptor (I' LR) with antigen. In mice, all cDC express TLR 2, 4, and 6 which can recognize peptidoglycans, lipoproteins, zymosan from yeast, protozoan gchosylphosphatidylinositol, and LPS from Gram-negative bacteria. Only 29 lymphoid cDC express levels of the internal TLR 3 high enough to detect viral double-stranded RNA and respond with the production of cytokines and enhance cross-presentation. Plasmacytoid DC express high levels of the internal TLR 7, 8, and 9 for detection of intracellular viruses and produce very high levels of type | interferon in response to recognition (Hochrein and O'Keeffe 2008). Upon activation by antigen within peripheral tissues, mature DC must migrate to lymphoid tissues containing naive T cells to present antigen. It has recently been demonstrated that DC antigen presentation and cell motility are coupled through interaction of MHC class Il-associated invariant chain (Ii, CD74), involved in peptide loading, with the motor protein myosin ll. DC deficient in cathepsin S which cleaves the cytosolic tail of Ii in the MHC-II complex at the endolysosome show reduced migration, while DC deficient in li showed continuous motility compared to wild type DC which showed both motile and stagnant phases. Inhibition of myosin II in DC leads to motility similar to that of cathepsin S-deficient DC. In early stages of DC activation by LPS, there is greater association of li with myosin ll compared to untreated DC, leading to increased change in DC directionality and decreased DC velocity. LPS- stimulated DC deficient in li do not show changes in directionality or velocity compared to unstimulated wild type DC (Faure-Andre et al. 2008). 30 Matrix metalloproteinase-9 Matn'x metalloproteinase structure and function The matrix metalloproteinase (MMP) family is composed of twenty cell surface and soluble proteins that are dependent on Ca2+ and Zn2+. All the members have a pre-peptide and pro-domain containing a cysteine switch and the catalytic domain. The Zn2+ ion is located in the substrate-binding site found in the catalytic domain. The proteins are secreted as pro-enzymes and act to degrade molecules composing the extracellular matrix (ECM). The MMPs are important during embryonic implantation and development, tissue morphogenesis, including bone ossification and the formation of neural cell connections, as well as wound healing because they allow cell migration through the ECM and create space for cell proliferation (Vu and Werb 2000). The tissue inhibitors of metalloproteinases (T IMP) family of proteins is a group of endogenous inhibitors of the MMP family. The family consists of four members that are highly conserved. TIMP proteins form 1:1 molecular complexes with soluble MMPs to inhibit MMP function. TIMPs bind to the active site of MMPs through the conserved VIRAK sequence to inhibit MMP activity. The expression of TIMP-1 and -3 is inducible, while TIMP-2 expression is constitutive. TIMP-4 is expressed in a limited number of non-immune tissues (Chirco et al. 2006; Verstappen and Von den Hoff 2006). MMP-9 is secreted as a pro-enzyme that must be cleaved in order for full activity to be realized. MMP-2 (gelatinase A), MMP-3 (stromelysin 1), MMP-7 31 (matrilysin), and MMP-13 (collagenase—3) have the ability to activate latent pro- MMP—9 to active MMP-9 through enzymatic cleavage and removal of the cysteine sulfhydryl group in the pro-domain. MMP-3 is the most potent MMP activator of MMP-9. However, chemical activation of pro-MMP-9 is also possible and activation by neutrophil-produced hypochlorous acid is one of these mechanisms (Opdenakker et al. 2001; F ridman et al. 2003). MMP-9 acts to enzymatically degrade the ECM by cleaving gelatin, laminin, and types I and IV collagen. MMP- 9 has been associated with the immune system in general by aiding leukocytes in entering and leaving the blood and lymphatic vessels (Faveeuw et al. 2001; Opdenakker et al. 2001). For example, MMP-9 deficiency has been implicated in insufficient prevention of Escherichia coli induced peritonitis due to decreased immune cell recruitment (Renckens et al. 2006). MMP-9 has been shown to be crucial for DC migration. Blocking monoclonal antibody against MMP-9 inhibits the release of hematopoietic progenitor stem cells in response to lL-8 injection in rhesus monkeys (Pruijt et al. 1999). In MMP-9+ mice, recruitment of DC into the ainrvay lumen in response to allergen (Verrnaelen et al. 2003) and migration of cultured bone marrow DC through tracheal epithelial tight junctions is impaired by two-fold and four-fold, respectively, compared to wild type animals (Ichiyasu et al. 2004). MMP-9 and DC migration It has been reported that MMP-9 and TIMP expression in DC is regulated by maturation stage. Expression of TIMP-1, -2, and -3 is decreased when DC are activated by TNF-or or modified vaccinia virus Ankara, while MMP-9 activity is 32 increased under these conditions (Osman et al. 2002). The proinflammatory chemokine CC ligand (CCL)-5 is a potent inducer of both MMP-9 expression and secretion in immature DC, leading to increased DC migration through basement membranes (Chabot et al. 2006). Pro-MMP-9 can be found in association with cell surface associated a2(lV) chain of collagen IV, the hyaluronan receptor CD44, and CD11b (F ridman et al. 2003; Hu and Ivashkiv 2006). Recently, it has also been found that pro-MMP—9 can associate with integrin 8-1 (C029) on the cell surface and that thrombin can increase transcription of both genes in osteosarcoma cells to induce invasion of the ECM leading to increased cell proliferation and metastasis (Radjabi et al. 2008). It has been found that IF N-or activated DC produce high levels of cell surface associated MMP-9 that is necessary for the DC to migrate through the ECM in response to chemokine signaling (Hu and Ivashkiv 2006). TIMP expression impedes DC migration. PGEz-induced TIMP-1 has previously been shown to inhibit the migration of monocyte-derived DC, while incubation with monoclonal anti-TlMP-1 antibody restores the ability of DC to migrate (Baratelli et al. 2004). Finally, MMP-9 has been shown to cleave and activate lL-1B [independently of caspase—1/IL-1l3- converting enzyme (Schonbeck et al. 1998)]. TGF-B (Y u and Stamenkovic 2000), and plasminogen to angiostatin (Pozzi et al. 2000). Regulation of MMP-9 expression The upstream region of the Mmp-9 gene in the mouse contains four AP-1, one AP-2, one NF-KB, four polyoma virus enhancer A-binding protein (PEA)3, 33 and three Sp1 transcription factor binding sites within 2.8 kilobases for regulation of Mmp-9 expression. This promoter region also contains a TATA-Iike box transcription start site (Munaut et al. 1999). The mouse Mmp-9 promoter is shown in Figure 1.3. In comparison, the human Mmp-9 promoter region 500 bases upstream of the Mmp-9 gene contains two AP-1, one Sp1, and one TGF- 81 inhibitory element transcription factor binding sites along with a TATA-like box (Huhtala et al. 1991). Together, AP-1 and NF-KB binding to sites in the Mmp-9 promoter account for the majority of the studied regulation of Mmp-9 production, followed by Sp1 binding. In human monocytes, activation of extracellular signal- regulated kinase (ERK)1/2, through AP-1 binding, and NF-KB pathways are critical for the thrombin-induced production of MMP-9 (Chang et al. 2009). In the mouse epidermal cell line JB6 P+, TNFa induces MMP-9 production through the phosphoinositide 3-kinase (Pl3K)/Akt pathway, which activates the NF-icB pathway, and subsequent binding of AP-1 and NF-icB DNA binding sites (Hwang et al. 2009). Epidermal growth factor has also been shown to induce NF-KB nuclear translocation and c-jun N-tenninal kinase (JNK) signaling in prostate carcinoma cells (Kuo et al. 2009). Transcription factors can also work alone to regulate expression of Mmp- 9. In SKBR3 cells, it was found that epidermal growth factor signaling through the ERK pathway (an AP-1 DNA binding protein) induces Mmp-9 expression but signaling through the STAT3 pathway negatively regulates Mmp-9 expression (Kim et al. 2009). In human skin keratinocytes, dermal fibroblasts, and rat hepatic stellate cells, TNFa signals through p21-activated kinase-1, involving JNK, to 34 increase binding of the AP-1 element in the Mmp-9 promoter. NF-KB signaling does not appear to be important in this process (Zhou et al. 2009). However, AP- 1 binding in the Mmp-9 promoter does not consistently lead to increased Mmp-9 transcription. In the Raw 264.7 macrophage cell line, inhibition of JNK signaling, through the use of inhibitors or siRNA knock—down, induces Mmp-9 production (Lee et al. 2009). NF-icB is also an important transcription factor in the regulation of Mmp-9 production. MMP-9 activity is up-regulated in the neuroblastoma cell line SK-N- SH by p50 and p65 binding to the NF-icB site in the MMP-9 promoter, but not by Sp1 or AP-1 binding, in the spontaneous conversion from epithelial cell to neuroblast cell to allow for basement membrane invasion (Farina et al. 1999). By a similar mechanism in retinoic acid differentiation-resistant SK-N-BE neuroblastoma cells, 72 h of 1-10 pM atRA treatment induces p50/p65 NF-KB activation and binding to the Mmp-9 promoter with minimal activation of AP-1 and Sp1, resulting in augmented expression of MMP-9 and greater basement membrane invasive capacity (Farina et al. 2002). This finding is unexpected because atRA treatment has been found to decrease NF-KB binding to its consensus sequence in promoters in many previous studies. Signaling of relaxin through the NF-KB proteins p50 and p65 increases Mmp-9 production in the THP-1 human monocyte cell line within 30 minutes of treatment (Ho et al. 2007). The KISS-1 gene, found to suppress cancer cell metastasis, is associated with decreased levels of MMP-9 expression. KISS-1 expression reduces NF-tcB binding to the Mmp-9 promoter through increased levels of iKBO. resulting in 35 decreased translocation of p50/p65 to the nucleus. Thus, this indicates that NF- KB signaling is a major mechanism of MMP-9 expression during metastasis. KISS-1 expression has no effect on AP-1 or Sp1 binding (Yan et al. 2001). Signaling through binding of the Sp1 site in the Mmp-9 promoter has also been shown to regulate MMP-9 production. The anti-metastatic compound norcantharidin acts to down-regulate MMP-9 expression through inhibition of Sp1 DNA binding activity but not AP-1 or NF-KB DNA binding activity in colon cancer CT26 cell line, indicating that Sp1 binding is required to up-regulate Mmp—9 expression in these cells (Chen et al. 2008). However, it has also been shown that Sp1 binding to the Mmp-9 promoter can inhibit Mmp-9 expression. In human corneal epithelial cells, platelet-activating factor (PAF) induces Mmp-9 production that interferes with normal wound reepithelialization. PAF induces DNA binding for AP-1, NF-KB, and Sp1, but only AP-1 and NF-tcB consensus binding sequences are required for PAF to induce Mmp-9 transcription. Binding to Sp1 consensus sequence in the Mmp-9 promoter inhibited PAF induced MMP-9 production in this model (T aheri and Bazan 2007). Sp1 also inhibits Mmp—9 expression during hepatic injury. During the activation of hepatic stellate cells upon chronic liver injury, type IV collagen is broken down while cells secrete type III" collagen and MMP-9 and develop a myofibroblast phenotype. Activated hepatic stellate cells up-regulate the expression of AP-1 and NF-KB proteins and the binding of their consensus sequences in the Mmp-9 promoter was required for production of Mmp-9. Overexpression of Sp1/Sp3 proteins resulted in reduced Mmp-9 expression and removal of the Sp1 binding site from the Mmp—9 36 promoter increased promoter activity indicating that Sp1/Sp3 activity represses Mmp-9 transcription in these cells (Takahra et al. 2004). Collectively, these studies indicate that regulation of Mmp-9 transcription is complex and is dependent on cell type and multiple signal inputs. Vitamin A regulation of MMP-9 AtRA has been previously linked with negative regulation of Mmp-9 expression in diabetic skin cultures (Varani et al. 2002; Lateef et al. 2004), tumor cell invasion (Tsang and Crowe 2001; Andela and Rosier 2004), bronchoalveolar lavage cells (Frankenberger et al. 2001 ), and emphysema (Mao et al. 2003). In squamous cell carcinoma lines, atRA treatment displaces AP-1 coactivators from the Mmp-9 promoter, leading to decreased promoter activity (T sai et al. 2008). However, atRA has been shown to up-regulate Mmp-9 expression in other cell types. In mammary epithelial cells cultured on collagen, atRA induces the cells to form colonies containing lumen through a mechanism involving MMP-9 production (Montesano and Soulie 2002). In rat mammary tissue, Zaragoza et al. (2007) have shown through chromatin immunoprecipitation studies that atRA acts through RARa on the Mmp-9 promoter, in association with the coactivation molecule p300, to up-regulate gene expression during involution, despite the lack of a RARE site in the Mmp-9 promoter. The actual mechanism of RARtx interaction with the Mmp-9 promoter to positively regulate promoter activity is currently unknown. 37 Aims of the research The long-term goal of the research is to further elucidate the role of vitamin A in immune system function and the maintenance of optimal host immunity. The influence of vitamin A in the immune system is well studied in immune cell types such as neutrophils, macrophages, and lymphocytes, but current knowledge is limited in the area of DC biology. DC are especially important for bridging innate and adaptive immunity. When activated by antigen or microbial signals, these innate immune cells migrate from peripheral tissues to present antigen to the adaptive immune cells located in secondary lymphoid tissues. It is necessary to study the role of vitamin A’s control of DC adherence because vitamin A may be necessary for: 1. release of immature DC from the bone marrow, 2. DC migration to secondary lymphoid tissues, 3. synapse formation with lymphocytes necessary for lymphocyte activation. Preliminary studies investigated the phenomenon of increased DC adherence in vitro under vitamin A deficient conditions or in the presence of antagonist to the RARa receptor. It was found that adherent and floating DC previously cultured with RARa antagonist have decreased cell surface expression of integrin-aL (CD11a, ltgal) following rescue with physiologically relevant concentrations of atRA. AtRA rescue also increased levels of Mmp-9 expression but had no effect on gene expression of ltgal, as indicated by a gene array. These results were confirmed in a time course experiment by real-time 38 PCR and ELISA. The results from these studies were published and are presented in detail in Chapter 2 of this dissertation. Based on these preliminary findings, I have generated the following specific aims: Specific Aim 1: Determine whether vitamin A supported MMP-9 activity is essential for decreased dendritic cell adhesion. The working hypothesis is that murine bone marrow cells cultured in vitro under myeloid dendritic cell producing conditions in the presence of sufficient vitamin A lose their adhesive properties as they mature due to MMP-9 enzymatic activity. Further, it is hypothesized that loss or decrease of MMP-9 activity in the absence of vitamin A will increase dendritic cell adherence. Specific Aim 2: Identify the signaling mechanism by which retinoic acid increases MMP-9 expression in dendritic cells. The working hypothesis is that retinoic acid, through retinoic acid receptor (RAR)-a signaling, induces increased Mmp-9 promoter activity and gene expression. Further, it is hypothesized that RARa interacts with transcription factors with known binding sites on the Mmp-9 promoter. These research objectives are important because a critical function for DC is migration from the peripheral tissues to the spleen and lymph nodes to activate adaptive immune cells. I propose to elucidate vitamin A-dependent mechanisms of cellular adhesion modulation that are important to DC mobility. This basic research could have important implications for world health recommendations, especially in the area of vaccination. DC are vital to the ability to make a robust immune response to inoculations by delivering the vaccine to immune cells that 39 can develop protective immunity. The results from this research could potentially lend greater support to the role of vitamin A deficiency in infectious disease. In particular, understanding the exact mechanisms by which vitamin A modulates immune function could help to refine vitamin A supplementation guidelines by increasing recommended frequency of supplementation to ensure the maintenance of DC populations, and thus an optimal memory response to vaccinations and/or infectious insults. 40 Figure 1.1. Vitamin A digestion, transport, and storage. 15, 15’-O’ase, 15, 15’-monoxidase; B43, apolipoprotein B48; C, apolipoprotein C; Chm, chylomicron; CR, chylomicron remnant; CRBP, cellular retinol binding protein; E, apolipoprotein E; LRAT, lecithinzretinol acyltransferase; RBP, retinol binding protein; RE, retinyl ester; REH, retinyl ester hydrolases; Rol, retinol; SI, small intestine; ‘ITR, Transthyretin. 41 Lymph/Blood SI Lumen Enterocyte (RE ER... 1 REH Rol l 15,15‘-O‘ase (carotene A f Stellate Cell Rol Liver LRATj RE REHl k Rol Figure 1.1 42 Parenchymal Cell RIVEREH R:1:I @- Blood Figure 1.2. Classic vitamin A signaling mechanism. AtRA binds to form holoRAR/RXR which binds to RARE in a target gene promoter. A consensus DR5 RARE is depicted. AtRA, all-trans retinoic acid; Ral, retinal; RalDH, retinal dehydrogenase; RAR, retinoic acid receptor; RARE, retinoic acid response element; RBP, retinol binding protein; Rol, retinol; RoIDH, retinol dehydrogenase; RXR, retinoid X receptor. 43 Cytoplasm Nucleus AGTTCANNNNNAGTI'CA Figure 1.2 44 Figure 1.3. Mouse Mmp-9 promoter with transcription factor binding sites noted. AP, activating protein; NF-KB, nuclear factor kappa B; PEA3, polyoma virus enhancer A-binding protein 3; Sp1, specificity protein 1; TATA, TATA-like box. Bent arrow indicates transcription start site. Arrows beneath letters indicate primers used in PCR analysis of ChlP chromatin described in Chapter 3. From Munaut et al. 1999. 45 gctggagctcaccagtgagaagcatctaagagaagcttgggagaacacccagctctctctctccggc tcacaggtctgttcgttgggaagcacatgaaggtctgggcacacaggaggc -gaacag cttgctgaagacagatcaaggccctg'ctccaccatggtggcaggcgaggaggatggaaggccggg ggctgccggctgttggcaagactgtgccaaagctttcctgagtggagcagggcagggctggaggag gggaagggtccatgacgatctcacagctcgggag iaggaag §gtgtttchcccatccaggtcacc cc aaggcttagagccaagaccccagtctcctaatttccaatcacaaacctgacaccatcaactgaggt ctcgtgaacactgctgaaagtggtttttctgtgtttcgagagtctcattttatcctcagatcaatataggga caaaggcttgagcgacaaagggtctgtttttgttctttaaacagaag aggaaggatagtgctagcc tgagaaggatgaagcttctgcttgctcccacatgtgtgtgt {ccccccgccc cccaggc tcatctttc cttccccaaggagtcagcctgctggagctaggggtttgccccat Iggaattcccc Iaaatcctgcctc aaagagcctgctcccagaggccaggag aggaag c aagactctatcagg I gggcggj ggatgagaggatagaacctacagtgtggggatgggctccaggctgcactctggcca L--- OOOOOOOOOOOOO actgggctgatcagtcagggccgtcagacctagggctaggtgaatgccccatcctgcacaccctcctt ccctttcccacaaagtctgcagtttgcagaaactaaaccctgagttctgtggtttcctgtgggtctgggg gtcctgcctgacttggcaatgggggactgtgggcagggcataagggagggggtagtgtaaacacac acacacacacacacacacacacacacacacacacacacacacacacgc tgagtcag cataagc -——q ctggaggggag | gggcgg lggtcac-cgttttactgcctc Ittt-aaz-Iatctctgcaaag gcagcgttagcca aagctgcggtcctcaccatgagtccctggcagcccctgctcctg A l _... E ............. : ""_"'i—_——i . NF-KB—!E PEA3 5| Sp1 I ' TATA +1 I—.. " ' _ .............. .___.. _.__l Figure 1.3 46 CHAPTER 2 RETINOIC ACID DECREASES ADHERENCE OF MURINE MYELOID DENDRITIC CELLS AND INCREASES PRODUCTION OF MATRIX METALLOPROTEINASE-9 Lackey DE, Ashley SA, Davis AL, Hoag KA. Retinoic acid decreases adherence of murine myeloid dendritic cells and increases production of matrix metalloproteinase-9. J Nutr. 2008;138:1512-1519. Appendix A contains an expansion of data found in Table 2.1. 47 Abstract Myeloid dendritic cells (DC) are professional antigen presenting cells (APC) that migrate to secondary lymphoid tissues where they activate naive T cells to proliferate and differentiate to T helper 2 cells. These cells help initiate an antibody mediated immunie response. Vitamin A is known to be essential for normal immune function. We investigated the ability of all-trans retinoic acid (atRA), a bioactive metabolite of vitamin A, to modulate DC adhesion in culture. The percentage of developing DC that remained adherent was higher in the presence of a specific antagonist of retinoic acid receptor (RAR)-a compared to cells treated with atRA treatment from days 8-10 of culture (P <0.05). AtRA treatment on day 8 of the culture period decreased DC surface expression of the adhesion molecule CD11a (P <0.0001), but not the gene expression. Treatment with atRA also dramatically increased gene and protein expression of the gelatinase zymogen pro-matrix metalloproteinase (pro-MMP)-9 (P <0.05). However, gene expression and protein production of MMP-9’s primary biological inhibitor, tissue inhibitor of metalloproteinase (T lMP)-1, was unaffected by atRA treatment, altering the molar ratio of secreted pro-MMP-9leMP-1. Thus atRA treatment resulted in a molecular excess of pro-MMP-9 to its primary inhibitor (P <0.05), indicating the possibility for increased MMP-9 enzymatic activity in atRA treated DC cultures. These data suggest that atRA is essential to augment MMP- 9 expression in myeloid DC and can alter their surface expression of adhesion molecules. KEY WORDS: all-trans retinoic acid, dendritic cells, neutrophils, vitamin A 48 Introduction Vitamin A has long been known for its role in immunity. It is currently estimated that about 127 million pre-school aged children and 20 million pregnant women in developing countries are vitamin A deficient (West 2002) leading to increased risk of night blindness and mortality (Christian et al. 2001). Vitamin A deficiency compromises the mucosal barriers of the eye, respiratory, and gastrointestinal tract, the first defenses against infection (Stephensen 2001). Deficiency also compromises the ability of macrophages and neutrophils to migrate to sites of infection, phagocytose, and kill bacteria (Twining et al. 1997; Stephensen 2001), while increasing neutrophil precursor cell expansion (Kuwata et al. 2000; Walkley et al. 2002). Marginal vitamin A status has been shown to decrease natural killer cell number and function (Dawson et al. 1999). Vitamin A is also important for development of a memory response to antigens introduced in the form of infection or vaccination. Vitamin A supplementation increases the delayed type hypersensitivity response to ovalbumin [a measure of T helper (T h)1 function] in vitamin A deficient mice, and also increases the serum antibody production in children after diphtheria and tetanus toxoid vaccination (Stephensen 2001). In mice, vitamin A deficiency enhances the Th1 response to Tn'chl'nella spiralis infection through increased secretion of interferon (lFN)-y, while suppressing the Th2 humoral response through decreased production of interleukin (IL)-5 and lL—10 (Cantorna et al. 1994). All-trans retinoic acid (atRA), a bioactive metabolite of vitamin A, has been shown to inhibit the synthesis of the Th1 cytokine IFN-y in Th1 cell culture 49 (Cantorna et al. 1996) and enhance the development of CD4+ T cells into Th2 cells through an antigen presenting cell (APC) intermediate (Hoag et al. 2002). Evidence exists that myeloid dendritic cells (DC) are a relevant APC whose function is modified by vitamin A (Hengesbach and Hoag 2004). Myeloid DC are professional APCs derived from a common myeloid progenitor cell and in mice are characterized as CD11b+ CD11c+ CDBa' (Shortman and Liu 2002). There is evidence that a lineage marker negative CX3CR1+ CD117+ common progenitor cell in the mouse gives rise to both myeloid oc (com:hi CD113“) and macrophages (CD11b+Imid CD11cvafiab'e or CD11b' CD11c+lhi for alveolar macrophages), but not other mature myeloid cell types (Gonzalez-Juarrero et al. 2003; Fogg et al. 2006). Immature myeloid DC from the bone marrow reside in the peripheral tissues and are able to efficiently capture and process antigens and present them in the major histocompatibility class I and II molecules (Banchereau et al. 2000). DC maturation through activation by antigen or stress signals from self tissues leads to migration to the T cell areas of the lymph nodes and spleen where they stimulate and activate both cytotoxic and helper naive T cells. The activation of T cells leads to adaptive immune responses and memory (Banchereau and Steinman 1998; Steinman 1999). AtRA signals through the retinoic acid receptor (RAR) family, composed of a, 8, and y isotypes in the class II family of nuclear receptors (Chambon 1996). This family also includes the thyroid hormone receptor, vitamin D receptor, and 50 peroxisome proliferator-activated receptor (Chawla et al. 2001; Olefsky 2001). Upon atRA ligand binding, RAR heterodimerizes with a retinoid X receptor (RXR) family member, also consisting of 0L, [3, and y isotypes. The RAR/RXR dimer then binds to retinoic acid response elements within the promoter regions of retinoid responsive genes and associates with co-regulating proteins, eventually leading to the promotion or inhibition of transcription and target gene expression (Bastien and Rochette-Egly 2004; Mark et al. 2006). RARs are important in the regulation of hematopoiesis (Evans 2005). Deletion studies indicate that RARy is necessary for maintaining the hematopoietic stem cell population, while overexpression of RARa in bone marrow cells indicate RARa activation favors neutrophil development (Purton et al. 2006). We originally observed that treating BALB/cJ mouse bone marrow- derived myeloid DC cultures with the RARor-specific antagonist AGN 194301 throughout the culture period led to a decreased yield of floating DC, while a large number of DC remained adherent on the dish surface. AGN 194301 competes with atRA for binding on the ligand binding domain of RAR): and blocks atRA activity (Nagpal and Chandraratna 2000). Based on these initial observations, experiments were designed to assess the ability of atRA to rescue DC development after culture initiation with the RARa-specific antagonist. 51 Materials and Methods Materials Stock solutions of atRA (Sigma, St. Louis, M0) were prepared in dimethyl sulfoxide (Sigma) and stored at —70°C under an argon atmosphere in the dark. Recombinant murine granulocyte-macrophage colony—stimulating factor (GM- CSF) was purchased from PeproTech (Rocky Hill, NJ) and dissolved in molecular biology grade water. lscove’s modified Dulbecco’s medium (IMDM, BioWhittaker, Walkersville, MD) was supplemented with 10% (v/v) serum [characterized (CH) or charcoal/dextran treated (CD) F BS, HyClone, Logan, UT], 2 mmol/L L-glutamine, 100,000 U/L penicillin, 100 mglL streptomycin (BioWhittaker), and 10 pmol/L B-mercaptoethanol (Sigma) to make complete IMDM (clMDM). Animals Male BALB/cJ mice (Jackson Laboratories, Bar Harbor, ME) were maintained according to institutional guidelines set by University Laboratory Animal Resources undera protocol approved by Michigan State University’s Institutional Care and Use Committee. The animals were fed commercial solid pellets (22/5 Rodent Diet #8640, wat composition: 22.58% protein, 5.23% fat, 3.94% fiber, 7.06% ash, 51.15% nitrogen-free extract; metabolizable energy content, 16.00 kJ/g; Harlan Teklad, Madison, WI) and killed between 6 and 12 weeks of age by carbon dioxide asphyxiation. Cell Culture 52 Femurs and tibiae were collected from male BALB/cJ mice and kept on ice. Bones were sterilized in 70% ethanol for 2 min and transferred to RPMl-1640 medium (BioWhittaker) where the bone ends were clipped and the bone marrow was flushed by syringe. Cells were filtered and pelleted, then resuspended in ACK lysing buffer (BioWhittaker) to lyse red blood cells. Bone marrow cells were then pelleted and resuspended in RPMI-1640 medium. A portion of cells were mixed with trypan blue dye solution and counted to determine cell number and viability. Cells were cultured for DC development using a method based on that of lnaba et al. (1998) and modified by Lutz et al. (1999). For the atRA and receptor antagonist full-culture experiments, bone marrow cells were plated 2 x 106 per dish containing 10 mL clMDM supplemented with 10% CH-F BS and 20 pglL GM-CSF in the presence or absence of 1 or 10 nmol/L AGN 194301 (provided by Dr. R.A.S. Chandraratna, Vitae Pharmaceuticals, Irvine, CA), an RARa-specific antagonist, or 10 mL clMDM supplemented with 10% CD-FBS and 20 pg/L GM-CSF in the presence or absence of 1 nmoIIL atRA for 10 d, with fresh medium and treatment given on d 3, 6, and 8 of culture. For atRA rescue experiments, bone marrow cells were plated 2 x 106 per dish containing 10 mL clMDM supplemented with 10% CH- FBS and 20 pg/L GM-CSF with 10 nmol/L AGN 194301. Cells were given 10 mL fresh clMDM containing 20 pg/L GM-CSF and 10 nmol/L AGN 194301 on d 3 and 6 of culture. On d 8 of culture, all dishes had a complete medium replacement with clMDM containing 20 jig/L GM-CSF and either 10 nmol/L of 53 atRA or AGN 194301. Cells were harvested for RNA extraction and flow cytometry staining at various time points between d 8 and d 10 of culture. Flow Cytometric Analysis Floating cells were collected by rinsing culture dishes with staining buffer [1% FBS, 0.1% (w/v) sodium azide in PBS, pH 7.4] and pelleted by centrifugation. Adherent cells were removed with Accutase (Innovative Cell Technologies, Inc., San Diego, CA) according to manufacturer instructions, collected, and pelleted by centrifugation. The cells were washed with staining buffer and a portion of cells from each sample were mixed with trypan blue dye to determine cell viability and count. One million cells from each sample were incubated with purified anti- FcyRIl/lll monoclonal antibody from the 2.462 hybridoma to block non-specific binding. Each sample was incubated in a specific antibody cocktail to label a variety of individual cell surface molecules or to control for the isotype. Monoclonal antibodies used were phycoerythrin (PB-conjugated hamster anti- mouse CD11c, fluorescein isothiocyanate (FlTC)—conjugated rat anti-mouse Ly- 6C and Ly-6G (Gr-1), biotin conjugated rat anti-mouse CD11a, PE-conjugated hamster lgG1 )1, F lTC-conjugated rat lgGZD/K, biotin-conjugated rat lgGZa/K, and allophycocyanin-conjugated streptavidin (BD Biosciences Pharrningen, San Jose, CA). Cells were washed and fixed in 2% paraforrnaldehyde in PBS. Samples were run at the MSU flow cytometry facility using a BD FACS Vantage equipped with DNA digital system and analyzed using FCS Express v. 3.0 software (DeNovo Software, Ontario, CA). Real- Time PCR 54 Total RNA was isolated from adherent cells at 0, 4, 8, 12, and 24 h post atRA rescue during the in vitro culture period. Medium was removed from the culture dishes and the cells were incubated with TRlzoI reagent (InVitrogen, Carlsbad, CA) for cell lysis. RNA extraction was completed according to the manufacturer’s instructions. Chloroforrn was added to solubilize lipids in the samples and for phase separation. RNA was precipitated from the aqueous phase with isopropanol and washed with 75% ethanol. RNA pellets were air—dried, resuspended in molecular biology grade water (Sigma), and stored at -70°C. RNA concentration was measured at 260 and 280 nm using the CellQuant spectrophotometer (Amersham, Piscataway, NJ). Total RNA was reverse transcribed using the TaqMan Reverse Transcription System with MultiScribe reverse transcriptase and random hexamer primers (Applied Biosystems, Foster City, CA). Expression of cDNA relative to the ubiquitously expressed glyceraldehyde—3-phosphate dehydrogenase (Gapdh) housekeeping gene was analyzed using Assays-on- dTM Deman Gene Expression primer/probe sets (Applied Biosystems) for ltgal (Mm00801807_m1), Mmp-9 (Mm00600163_m1), and Timp-1 (Mm00441818_m1). PCR was performed on an ABI PRISM 7900HT Sequence Detection System (Applied Biosystems). Monitoring of PCR occurred in real-time by detection of FAM fluorescence on the 5’ end of the probe, quenched with a non-fluorescent quencher on the 3’ end of the probe. Relative cDNA expression was calculated from AACT values and comparison to a standard curve of multiple 10—fold dilutions using ABI 7700 software (Applied Biosystems). 55 Microarray Analysis Total RNA was isolated from adherent cells and CD11c+ purified floating cells on d 10 of culture. AtRA-treated floating cells were purified for CD11c+ cells by labeling with biotin—conjugated hamster anti-mouse CD1 1c, secondary incubation with BD Streptavidin Particles Plus-DM, and separated using the BD IMagnet (BD Biosciences Phamlingen). Medium was removed from the culture dishes and RNA extraction was completed as described above. Differential RNA expression was analyzed using the GEArray Q Series Mouse Extracellular Matrix and Adhesion Molecules Gene Array by SuperArray (Frederick, MD) by the manufacturer. ELISA Quantikine mouse ELISA kits for detection of secreted pro-MMP-9 and TIMP-1 from cell culture supernatants were purchased from R & D Systems (Minneapolis, MN). The manufacturer protocol was followed. Briefly, standard dilutions, internal controls, and sample dilutions were incubated in the appropriate pre-coated ELISA plate with assay diluent in technical duplicate. After washing, the wells were incubated with the appropriate conjugate. Finally, the wells were incubated with substrate and then the reaction was stopped with hydrochloric acid. The well absorbances were read on a plate reader at 450 nm, using a correction wavelength of 570 nm. SOFTmax Pro for Life Sciences software version 4.0 (Molecular Devices, Sunnyvale, CA) was used for concentration calculations from the standard curve. Statistical Analysis 56 Data were analyzed using lnStat software version 3.05 (GraphPad Software, Inc., San Diego, CA). Changes in percentages of CD11c+ and Gr-1+ cells in the atRA and receptor antagonist full culture experiments were determined using one-way ANOVA with a post-hoc Tukey-Kramer multiple comparisons test for the floating cells group and the KruskaI-Wallis nonparametric ANOVA with a post- hoc Dunn’s multiple comparisons test for the all cells group. Comparisons between two groups were made using Student’s unpaired ttest, except for ELISA experiments where the non-parametric Mann-Whitney test was employed. Data are presented as means :I: SEM and differences between groups of P <0.05 were considered significant. Results RARa antagonism increases adherence of CD11c+ DC Previous in vitro studies from our laboratory have shown that insufficient concentrations of vitamin A result in decreased DC development and increased neutrophil development (Hengesbach and Hoag 2004). However, supplementation of medium with physiological levels of atRA restored DC development (Hengesbach and Hoag 2004). We cultured mouse bone marrow cells in clMDM supplemented with GM-CSF and CH-FBS in the presence or absence of various dosages of the RARa-specific receptor antagonist AGN 194301, or CD-F BS (vitamin A depleted) in the presence or absence of atRA for 10 d. Floating cells were harvested alone, or floating and enzyme-detached adherent cells were pooled together for each sample. Cells were then dual 57 labeled for cell surface expression of Gr-1 and CD11c to identify neutrophils and DC, respectively. We found that the percentage of CD11c+ DC, when adherent and floating cells were combined, did not differ significantly between the treatment groups. However, when only floating cells were harvested, the percentage of CD11c+ DC was significantly reduced from 80% with the CH-FBS positive control to 45% with the addition of 10 nmol/L RARa antagonist treatment (P <0.05). At the same time, the percentage of CD11c+ DC was significantly increased from 10% with the CD-F BS negative control to 60% with the addition of 1 nmollL atRA (P <0.05, Figure 2.1A). When floating and adherent cells were pooled together, the percentage of Gr-1+ neutrophils increased slightly with 10 nmollL receptor antagonist treatment while the neutrophil population was reduced from 50% with CD-FBS supplementation to 20% with the 1 nmollL atRA treatment in the presence of CD-FBS supplemented IMDM, back to the levels of the CH-FBS control. However, when only floating cells were harvested, the percentage of Gr-1+ neutrophils increased significantly from 10% in the CH-FBS control to 50% with the 10 nmol/L receptor antagonist treatment (P <0.05), while the percentage decreased significantly from 70% with the CD-FBS control treatment to 20% with the 1 nmol/L atRA treatment in CD-F BS supplemented IMDM (P <0.05, Figure 2.13). Rescue of cultures with atRA increases the percentage of floating DC Based on the results indicating that RARa signaling is necessary for CD11c+ DC to become non-adherent cells, we next designed experiments to test 58 whether atRA can rescue DC development after culture with receptor antagonist. Primary mouse bone marrow cells were cultured in the presence of GM-CSF and the RARor-specific receptor antagonist AGN 194301 for eight days. The cells either remained in the presence of receptor antagonist or were rescued by replacement of receptor antagonist with 10 nmollL atRA on d 8. When cells were grown with the receptor antagonist, large cell clusters that were firmly adherent developed on the surface of the plastic dish. When the cells were rescued with atRA on d 8 of the culture period, the cells within the clusters began to release within 24-48 h (data not shown). On d 10, 48 h post-atRA rescue, the resultant adherent cell yield did not differ between the receptor antagonist treatment and the atRA rescue treatment. However, there was a significant increase (P <0.0001) in floating cell yield in atRA-rescued cultures compared to those that remained with the receptor antagonist (Figure 2.2A). The percentage of cells that remained adherent decreased by 10% with atRA treatment and the number of floating cells increased significantly (P <0.01) compared to cultures that received fresh medium and GM-CSF but remained in the receptor antagonist treatment (Figure 2.28). Upon harvest on d 10, cells were stained with Gr-1 to identify neutrophils and CD11c to identify myeloid DC. The CD11c+ DC predominated over Gr-1+ neutrophils in the adherent cells in both treatment groups (Figure 2.3A,B). However, atRA rescued cell cultures showed a significantly greater percentage (P <0.05) and yield (P <0.001) of CD11c+ floating DC compared to cells retained in culture with receptor antagonist (Figure 2.3A,C). Conversely, Gr-1+ neutrophils 59 predominated in the floating cell population for both treatment groups, and the percentage of Gr-1+ floating cells was not significantly affected by atRA rescue (Figure 2.38). However, total Gr-1+ neutrophils yield was significantly enhanced in the floating cell population when cells were cultured with atRA for the final 2 days of culture as opposed to remaining with receptor antagonist (Figure 2.30). 0011a cell surface expression is decreased by atRA rescue Based on the results showing that atRA rescue decreased adherence in DC cultures, we also examined the cell surface for expression of relevant cell adhesion molecules. On (I 10 of culture, both adherent and floating CD11c+ DC treated with atRA showed significantly decreased (P <0.0001) levels of the cell surface marker CD11a (integrin-aL) compared to cells that remained in receptor antagonist treatment for the full 10 d culture period, measured by mean fluorescence intensity (MF I, Figure 2.4A,B). Surface expression of the adhesion molecule and general myeloid cell marker CD11b was unaffected by atRA treatment (data not shown). Matrix metalloproteinase-9 mRNA expression and pro-MMP-9 secretion increases in adherent cells with atRA rescue A microarray study for mouse extracellular matrix and adhesion molecule gene expression using a GEArray Q Series array by SuperArray (Frederick, MD) was performed to further explore DC adhesion. We found that 48 h atRA rescue increases Mmp-9 expression 12-fold in adherent DC and 15-fold in CD11c+ floating cells compared to receptor antagonist treated adherent cells, indicating 60 that these differences in gene expression were not due solely to the adherent or floating phenotype (Table 2.1). Mmp-7 was the only other metalloproteinase that appeared to be retinoic acid-regulated. However, actual levels of expression were not above background (up to 0.5 arbitrary units (AU), a unit of signal intensity after interquartile normalization) for this gene, whereas atRA up- regulated Mmp-9 expression to 6 AU (Appendix A). Mmp-3, -8, -12, -13, and -19 were also highly expressed in these samples (from 8-20 AU, Appendix A), but their expression was unchanged by atRA treatment (Table 2.1). These results suggest that up-regulation of matrix proteases may play a role in the decrease in DC adhesion mediated by atRA. Complete normalized results of the gene array can be found in Appendix A. 61 Table 2.1. Induction of mouse matrix metalloproteinase mRNA relative to expression of receptor antagonist adherent DC in atRA treated adherent and CD110+ floating mouse DC cultures.1 mRNA Retinoic Acid Adherent2 Retinoic Acid Floating DC2 Mmp- 1a 2.47 1 .59 Mmp-2 1 .12 0.79 Mmp-3 1 .15 0.38 Mmp-73 4.44 3.58 Mmp-8 0.78 2.06 Mmp-9 12.65 15.49 Mmp-10 1 .15 1 .36 Mmp-1 1 0.63 0.67 Mmp-12 0.69 2.76 Mmp-13 0.76 0.59 Mmp-14 1 .27 0.94 Mmp-15 1 .63 1 .16 Mmp-16 0.95 0.38 Mmp-17 0.72 1 .94 Mmp-19 0.51 0.32 Mmp-20 0.79 1 .53 Mmp-23 1 .95 1 .37 Mmp-24 1 .30 0.43 1RNA extracted from adherent atRA-treated, adherent receptor antagonist- treated, or CD11c+ isolated floating DC. RNA was hybridized to SuperArray GEArray Q Series Mouse Extracellular Matrix and Adhesion Molecules Gene Array (SuperArray Biosciences). Signal intensity between treatments was normalized to the interquartile median value. 2Relative expression calculated as retinoic acid expression/receptor antagonist expression. 3Mmp-7 signal intensity is similar to blank sample intensity. 62 Quantitative real-time PCR was used to confirm the level of gene expression in two independent time course experiments. It was found that ltgal transcript abundance decreases over time in both the atRA and receptor antagonist treatments. Both treatments had the same level of ltgal transcript expression 24 h post-atRA rescue (data not shown). However, we found that atRA rescue treatment increased the expression of Mmp-9 transcript three fold within 4 h and continued to increase up to 14 fold within 24 h compared to the 0 h receptor antagonist treatment (Figure 2.5A). Levels of tissue inhibitor of metalloproteinase (Timp)-1 transcript, an endogenous inhibitor of MMP-9 activity, remained constant for both treatments throughout the time course (data not shown). We next used ELISA to determine whether MMP-9 and TIMP—1 protein production increased in conjunction with mRNA expression. Treatment with atRA was found to significantly increase the level of secreted pro-MMP-9 in conditioned medium on both d 9 and 10 of culture (24 and 48 h, respectively after the addition of fresh medium and rescue with atRA or continued treatment with receptor antagonist). Pro-MMP-9 protein increased significantly (P <0.05) to 26 pg/L compared to 4 [lg/L in the receptor antagonist group on day 10 (Figure 2.58). At the same time, levels of secreted TIMP-1 increased to 4 ngL in both receptor antagonist and atRA rescued treatment groups compared to the receptor antagonist treated group on d 8 (data not shown). The molar ratio of pro-MMP-9:TlMP-1 indicates that there was a 2:1 molar excess of pro-MMP-9 on day 10 of culture in the atRA treated group (48 hours after rescue), while those 63 cultures that retained treatment with the receptor antagonist showed a 4:1 molar excess of TIMP-1 (Figure 2.50). Discussion Vitamin A has been demonstrated to be important in the development of humoral immunity by polarizing na'l've CD4+ T cells to a Th2 response through an APC intermediate (Hoag et al. 2002). One of the vitamin A dependent APC has been identified as myeloid DC (Hengesbach and Hoag 2004). However, the mechanisms by which atRA modulates myeloid DC physiology have not yet been fully described. Therefore, we employed the RARa-specific receptor antagonist, AGN 194301, to investigate the effects of atRA on the development of cultured DC via signaling through the nuclear receptor RARa. We explored the ability of atRA to rescue DC development after culture initiation in the presence of the receptor antagonist. A number of new findings resulted from this approach. First, when primary bone marrow cells are cultured in medium depleted of retinol or in the presence of receptor antagonist in medium containing sufficient concentrations of retinol, a decreased percentage of DC and an increased percentage of neutrophils in the floating cell population results compared to cells cultured in clMDM supplemented with CH-FBS. In addition, atRA supplementation of CD-FBS containing medium increased the percentage of DC and decreased the percentage of neutrophils to the levels of the positive control treatment in the floating cell population. However, the percentage of DC and neutrophils in the total cell population does not differ between cultures receiving medium supplemented with CH-F BS in the presence or absence of receptor antagonist or cultures receiving medium supplemented with CD—F BS in the presence of atRA. These data indicate that inhibition of signaling through RARa blocks normal DC development and loss of adherence. Myeloid DC developed an adhesive phenotype when primary bone marrow cells were cultured in the presence of RARa-specific receptor antagonist, while rescuing these cultures with atRA reduced this phenotype and instead the DC showed a floating phenotype, as found when they are cultured exclusively in vitamin A sufficient medium. These results indicate that atRA acts to increase the percentage of DC that lose adherence from the plastic culture dish surface. However, the proportion of floating neutrophils did not decrease significantly upon atRA rescue compared to cells treated with the receptor antagonist and the actual neutrophil cell yield increased upon atRA rescue. It is possible that neutrophils are still able to develop from precursor cells in the presence of RARa-specific receptor antagonist while DC are not, thus atRA rescue over a limited 48 h period begins the process of DC development while neutrophils have an earlier advantage. We also found the level of the adhesion molecule CD11a was decreased on the cell surface of both adherent and floating cells in cultures where atRA rescue treatment was provided. CD11a dimerizes with CD18, integrin-[32. to form the cell-surface molecule lymphocyte function-associated antigen (LFA)-1 (Elangbam et al. 1997). We chose to focus on CD11a because it was the only cell surface molecule in a panel of adhesion molecules including CD24 (heat 65 stable antigen), CDS1 (platelet/endothelial cell adhesion molecule-1), CD43, CD80 (87-1), F4l80, and 33D1 that showed differential cell surface expression between receptor antagonist and atRA treated DC cultures (data not shown). Despite the strong reduction in CD11a protein detected on the cell surface of DC rescued with atRA, there is no difference in ltgal, the gene for CD11a, transcript expression between adherent cells treated with receptor antagonist and cells rescued with atRA over time. Pyszniak et al. (1994) have shown that soluble intercellular adhesion molecule (ICAM) can bind LFA-1 and compete for binding with antl-LFA-1 antibodies. There is evidence that MMP-9 can act to produce leAM-1 from cell surface ICAM-1. However, experiments in our lab show there is little variability in loam-1 mRNA expression over a 24 h period independent of RARa antagonist or atRA treatment. Likewise, we have not found a difference in leAM-1 production between the treatments at the different time points over a 48 h period (Appendix B). The maximal presence of leAM-1 in culture supernatant reached only 2.5 pglL for both culture conditions on d 10 of the culture period. Since leAM-1 was produced equally regardless of atRA treatment, it is unlikely that leAM-1 blocking LFA-1 could explain the differences in CD11a cell surface expression. Overall, our data indicate that atRA may regulate transcription of another molecule through RARa that affects CD11a surface expression. The protein MMP—9, which was up-regulated by atRA rescue, is a possible candidate, and has been shown to cleave a number of other cell surface proteins. These include dystroglycan, allowing for leukocyte entrance through the parenchymal basement 66 membrane of the blood-brain barrier (Agrawal et al. 2006), and galectin-3, which interacts with the extracellular matrix (Ochieng et al. 1998). MMP-9 has also been shown to proteolytically activate a number of cytokines that are secreted in latent forms, including lL-1B (Schonbeck et al. 1998) and transforming growth factor-8 (Yu and Stamenkovic 2000). AtRA rescue increases expression of the Mmp-9 transcript and the pro- MMP-9 protein, while leaving expression of TIMP-1 mRNA and protein, the primary endogenous inhibitor of MMP-9 (Vu and Werb 2000) unchanged. This results in a molecular excess of secreted pro-MMP-9 in atRA rescued cell cultures. Conversely, an excess of TIMP-1 is maintained in receptor antagonist treated cell cultures. MMP—9, also known as gelatinase-B, acts to degrade the extracellular matrix and is secreted as a pro-enzyme that must be cleaved for full activity to be realized (Opdenakker et al. 2001). It has been shown that use of monoclonal antibody against MMP-9 inhibits the release of hematopoietic progenitor stem cells from bone marrow in response to lL-8 injection in rhesus monkeys (Pruijt et al. 1999). MMP-9 has also been shown to be critical for DC migration (Randolph et al. 2005). In MMP-9+ mice, recruitment of bronchial associated lymphoid tissue DC into the airway lumen in response to allergen (Vermaelen et al. 2003) and migration of cultured bone marrow DC through tracheal epithelial tightjunctions is impaired (Ichiyasu et al. 2004). Prostaglandin E2-induced TIMP-1 has previously been shown to inhibit the migration of human monocyte-derived DC, while incubation with monoclonal anti-TlMP-1 antibody restores the ability of DC to migrate (Baratelli et al. 2004). 67 Retinoic acid has been previously linked with negative regulation of MMP- 9 levels in diabetic skin cultures (Varani et al. 2002; Lateef et al. 2004), tumor cell invasion (T sang and Crowe 2001; Andela and Rosier 2004), bronchoalveolar lavage cells (F rankenberger et al. 2001), and emphysema (Mao et al. 2003). Our data directly contradict these findings because we show that atRA rescue up- regulates the expression of the MMP-9 gene and secreted pro-enzyme in myeloid DC cultures compared to cultures treated with receptor antagonist. However, these previous studies all used pharmacological doses of atRA (21 pmollL for cell cultures), while our studies here used more relevant physiological concentrations. It is also possible that the various cell populations studied in the literature differ in their response to atRA treatment. Other cell and tissue types show MMP-9 responses to atRA similar to our results when treated with physiological concentrations. For example, a study completed by Montesano and Soulié (2002) shows that a low dose (100 pmollL) of atRA induces mammary epithelial cells cultured on collagen to form colonies containing lumen through a mechanism involving MMP-9 production. In rat mammary tissue, Zaragoza et al. (2007) showed through chromatin immunoprecipitation studies that atRA acts through RARa on the Mmp-9 promoter to up-regulate gene expression during involution, despite the lack of a RARE site. The SKNBE neuroblastoma cell line also responds to atRA by increasing production of MMP-9, allowing for production of outgrowing neurites and cell migration to produce a neuronal phenotype (Chambaut-Guerin et al. 2000). 68 Recently, Dannanin et al. (2007) have shown that immature DC cultured with pharmacological levels of supplemental atRA (1 pmollL) have increased migratory ability in vitro and when injected into tumors, compared to that of DC cultured in medium containing physiological concentrations of atRA, and that this migration is greatly diminished by inhibiting MMPs. Our findings that atRA rescue treatment up-regulated expression of MMP-9 and correspondingly decreased myeloid DC adhesion are in agreement with these findings. However, TIMP-1 did not decrease after atRA treatment in our studies, as was shown by Darrnanin et al. We have shown the progression from adherent DC to floating DC that is inhibited when RARa signaling is blocked can be rescued by atRA treatment, and this progression corresponds with up-regulation of MMP-9 and a decrease in cell surface expression of CD11a. Further studies are needed to verify whether MMP-9, which is up-regulated upon atRA rescue, is proteolytically active and the main effector for loss of DC adherence. These studies will involve the use of chemical inhibitors of MMP-9 and other relevant MMPs as well as functional studies that examine the amount of MMP-9 mediated proteolysis between the different DC culture treatments. DC need to migrate from infected and stressed peripheral sites to relevant lymphoid tissues to activate the adaptive immune response. This migration requires movement through extracellular matrix and changes in adhesion molecule expression (Randolph et al. 2005). Our results indicate that physiological concentrations of atRA may be necessary for optimal DC development and release from or migration through the extracellular matrix. 69 Acknowledgements We thank Dr. Louis King of the Michigan State University flow cytometry facility in the Research and Technology Support Facilities. 70 Figure 2.1. Mouse bone marrow cells [CD11c+ (A), Gr-1+ (B)] cultured with RARa receptor antagonist develop an adherent phenotype. Bone marrow cell cultures were generated in the presence of GM-CSF in medium supplemented with CH-FBS with or without receptor antagonist (AGN) or CD-FBS with or without atRA for 10 d. Floating cells only or all cells (floating plus adherent) were analyzed separately by flow cytometry. Values are means + SD, n=5 (floating cells) or 3 (all cells). Within a cell type, means without a common letter differ, P<0.05. 71 I: Floating Cells - All Cells 2 Y.Z 3 0 er #- o S Y D 0 e\° Cl-l-FBS CD-FBS Treatment CI-l-F BS CD—F BS Treatment Figure 2.1 72 Figure 2 with RAF I’El yields :iesence allure w Floating ( all coun 3856an CCIIBSpOI Figure 2.2. AtRA rescue of mouse bone marrow-derived myeloid DC cultured with RARa receptor antagonist reduces cell adherence [Absolute (A) and percent (8) yields of adherent and floating cells]. Bone marrow cultures were grown in the presence of GM-CSF and receptor antagonist for 8 (I then either remained in culture with 10 nmollL receptor antagonist or were rescued with 10 nmollL atRA. Floating cells and adherent cells were harvested separately from each treatment and counted on d 10. Results are from 2 independent experiments and presented as the mean + SD, n=10. Asterisks indicate different from the corresponding receptor antagonist treatment: **P<0.01, ***P<0.0001. 73 15': I: Receptor Antagonist «a; ‘ -Retinoic Acid 1m .- I X _ *** 5 10' u E a 2 s- o u o J 0:_1 Adherent Floating Total Cell Population 2100- 3 as o "' 75- I I6 . e\° .0 50- '13 t 25J *«Ir 8 ° i 0 , Adherent Floating Cell Population Figure 2.2 74 Figure 2.3. The mouse bone marrow-derived floating myeloid DC population is enhanced by atRA rescue [Percent (A, 8) and absolute (C, D) yields of CD11c+ (A, C) and Gr—1+ (8, D) adherent and floating cells]. Bone marrow cultures were grown in the presence of GM-CSF and receptor antagonist for 8 d, then either remained in culture with 10 anl/L receptor antagonist or were rescued with 10 nmol/L atRA. Cells were analyzed by flow cytometry on d 10. Results are representative of 2 independent experiments and presented as means + SD, n=6. Asterisks indicate different from the corresponding receptor antagonist treatment: *P<0.05, **P<0.01, or ***P<0.0001. 75 "/o CD11c" Cells % Gr-1+ Cells 100- \l 01 l 01 O l N 01 l 7:3:2' 100T \J ‘i‘ or ‘P h: ci" Adherent EZI Receptor Antagonist - Retinoic Acid FIE Floating Population O I Iii ;‘ Adherent Floating Population FQme23 76 Total CD11c+ Cells Total Gr-1“ Cells (n x106) NOD L i «b 0'! 1 A L fii ‘ E] Receptor Antagonist - Retinoic Acid m l O , Adherent Floating Population 5. int 4.. at l 2... 1- o F—‘Ii Adherent Floating Population Figure 2.3 continued 77 Figure 2.4. CD11a cell surface expression is reduced by atRA rescue treatment in mouse bone marrow-derived myeloid DC cultures [counts (A) and mean fluorescence intensity (8) in adherent and floating cells]. Cells were treated with receptor antagonist alone or rescued with atRA on d 8 of the 10-d culture. Cell surface expression for CD110 and CD11a (ltgal) was analyzed by flow cytometry after 10 d. In 8, results are representative of 2 independent experiments and presented as means + SD, n=6. ***Different from receptor antagonist treatment, P<0.0001. 78 Adherent Floating CD1 1 a I Isotype — Retinoic Acid * Receptor Antagonist 3000‘ [.2] Receptor Antagonist - Retinoic Acid ”'7 N o o ‘? *** CD11a MFI of c0116: Cells 8 8 Adherent Floating Cell Population Figure 2.4 79 Figure 2.5. AtRA rescue changes the gene expression and secretion of MMP-9 in mouse bone marrow-derived myeloid DC cultures. A, total RNA extracts made from pooled adherent cells on d 8 of culture at 0, 4, 8, 12, and 24 h after receptor antagonist treatment or atRA rescue were reverse transcribed to cDNA and Mmp-9 gene expression was assayed by quantitative real-time PCR in technical triplicates. Values were normalized to glyceraldehyde-3-phosphate dehydrogenase expression and are shown as fold of the 0 h receptor antagonist treatment. The results are representative of 2 independent experiments. 8, pro- MMP—9 in cell culture medium supernatants measured by ELISA. C, the pro- MMP-9:TlMP-1 molar ratios calculated from the concentration of each protein for each sample measured. The results are representative of 2 independent experiments and are means + SD, n=4. *Different from the receptor antagonist treatment on that day, P<0.05. 80 > Mmp-9 Fold mRNA Expression co pro-MMP-9 Concentration (p glL) O pro-MMP-9 :TIMP-1 Molar Ratio 0 Receptor Antagonist I Retinoic Acid. 15- ' 10‘ I 5-1 04> i O h-OI O O 0 Time (h) co C? E] Receptor Antagonist * - Retinoic Acid m c.’ _s O l ,H m . 8 9 10 Day of Culture 31 Zq I I'"—l 1O 9 Day of Culture Figure 2.5 81 CHAPTER 3 VITAMIN A UP-REGULATES MATRIX METALLOPROTEINASE-9 ACTIVITY BY MURINE MYELOID DENDRITIC CELLS THROUGH A NON-CLASSICAL TRANSCRIPTIONAL MECHANISM 82 Abstract Myeloid dendritic cells (DC) are specialized antigen presenting immune cells. Upon activation in peripheral tissues, DC migrate to lymph nodes to activate T lymphocytes. Matrix metalloproteinase (MMP)-9 is a gelatinase essential for DC migration. We have previously shown that all-trans retinoic acid (atRA), a bioactive metabolite of vitamin A, significantly augments DC MMP-9 mRNA and protein production. We investigated the mechanisms by which atRA increases MMP-9 activity in vitro. Mouse myeloid DC cultured with atRA demonstrated increased gelatinase activity compared to cells cultured with retinoic acid receptor (RAR)-a antagonist. Adding MMP-9 inhibitor significantly blocked DC gelatinase activity and increased adherence of DC in a dose dependent manner. AtRA-induced Mmp-9 expression in DC could be blocked by the transcriptional inihibitor actinomycin D. Since the Mmp-9 promoter contains no classic retinoic acid response element (RARE), we performed additional studies to determine how atRA regulates Mmp-9 transcription in DC. Electrophoretic mobility shift assays for the consensus Sp1, AP-1 and NF-KB binding sites located in the Mmp-9 promoter did not increase binding in response to atRA. Chromatin ’ immunoprecipitation assays indicated that atRA increases recruitment of RARa and histone acetyltransferase p300 to, and histone H3 acetylation of, the Mmp-9 promoter. These data suggest that atRA regulates DC adhesion in vitro through MMP-9 gelatinase activity. AtRA increases Mmp-9 expression through a transcriptional mechanism involving enhancement of RARa promoter binding, 83 recruitment of p300, and subsequent acetylation of histone H3, despite the absence of a consensus RARE. KEY WORDS: all-trans retinoic acid, dendritic cells, matrix metalloproteinase-9, vitamin A Introduction Worldwide, vitamin A deficiency affects an estimated 190 million pre- school aged children and 19.1 million pregnant women (WHO 2009) and increases morbidity and mortality due to increased susceptibility to infection (Villamor and Fawzi 2005). Vitamin A deficiency (VAD) has multiple effects on innate immunity: 1) mucosal epithelial integrity is impaired (Stephensen 2001); 2) the ability of macrophages and neutrophils to migrate to sites of infection, phagocytose and kill bacteria is abnormal (Twining et al. 1997; Stephensen 2001); 3) there are decreases in natural killer cell number and function in rats (Dawson et al. 1999); and 4) there are increased neutrophil numbers in the spleen, peripheral blood, and bone marrow in vivo (Kuwata et al. 2000). Vitamin A is also important in polarizing naive T helper (Th) cells to a Th2 phenotype. T cells from vitamin A deficient mice do not support lgG1 production by vitamin A sufficient B cells, which requires a Th2 response (Carman et al. 1989). VAD increases Th1 cytokine IF N-y secretion while impairing the Th2 response through diminished lL-5 and IL-10 production during Trichinella spiralis infection (Cantorna et al. 1996). In vitro, atRA decreases lFN-y production in Th1 cultures while enhancing Th2 development through an antigen presenting cell (APC) intermediate (Hoag et al. 2002). Vitamin A is a crucial nutrient in professional APC myeloid DC development from its precursors, as a vitamin A deficient environment in vitro permits granulocyte development instead (Hengesbach and Hoag 2004). 85 Myeloid DC are professional antigen presenting cells derived from hematopoietic stem cells. All mouse DC express the cell surface protein CD11c. Myeloid DC also possess CD11b but lack CD8a surface expression (Shortman and Liu 2002). Immature DC migrate from the bone marrow and reside in peripheral tissues where they can capture, process, and present antigen on their major histocompatibility complexes (MHC) classes I and Il (Banchereau et al. 2000). Upon maturation stimulated by antigen interaction with pattern recognition receptors, DC migrate to T cell areas of the lymph nodes and spleen where they present antigen to naive T cells (Banchereau and Steinman 1998; Steinman 1999; Hochrein and O'Keeffe 2008). Myeloid DC are the most efficient stimulators of T cells (Hochrein and O'Keeffe 2008). Matrix metalloproteinase (MMP)-9, a gelatinase secreted as a zymogen, degrades the extracellular matrix through cleavage of gelatin, collagen types I and IV, and laminin (Opdenakker et al. 2001). It is activated by removal of the pro-peptide by MMP-2, -3, -7, or -13, or through chemical cleavage (Opdenakker et al. 2001; Fridman et al. 2003). Experimental impairment of MMP-9 activity inhibits hematopoietic progenitor cell release from the bone marrow (Pruijt et al. 1999), impairs recruitment of DC to the bronchioalveolar associated lymphoid tissue in response to allergen (Vermaelen et al. 2003) and migration through epithelial tight junctions (Ichiyasu et al. 2004), and impairs migration of monocyte—derived DC in vitro (Baratelli et al. 2004). Mmp-9 expression is regulated through AP-1 and NF-tcB signaling pathways, including thrombin signaling in human monocytes (Chang et al. 2009), TNF-or in a mouse epidermal 86 cell line (Hwang et al. 2009), and epidermal growth factor in prostate carcinoma cells (Kuo et al. 2009). Mmp-9 expression is also often controlled through Sp1 signaling, including in the colon cancer CT26 cell line (Chen et al. 2009). All-trans retinoic acid (atRA), a bioactive metabolite of vitamin A, signals through the retinoic acid receptor (RAR) family consisting of three isoforms, a, B, and y (Chambon 1996). AtRA enters the nucleus and binds RAR which heterodimerizes with one of three retinoid X receptor (RXR) isoforms, creating a holo-RAR-RXR complex. In the classic mechanism of atRA signaling, the complex binds to a retinoic acid response element (RARE) in the gene promoter and associates with coregulatory proteins to affect target gene expression (Bastien and Rochette-Egly 2004; Mark et al. 2006). While this model explains the mechanism approximately 130 vitamin A-regulated genes, at least 400 genes controlled by atRA are not regulated through this mechanism as the promoters lack a consensus RARE (Balmer and Blomhoff 2002). Vitamin A increases MMP-9 expression and activity in the context of cell migration. AtRA induced-MMP-9 activity in mammary epithelial cells results in lumen morphogenesis (Montesano and Soulie 2002) and increases DC migration in vitro and in vivo (Dannanin et al. 2007). We have previously shown that atRA treatment results in loss of DC adherence in culture and augments expression and secretion of pro-MMP-9 through RARor signaling (Lackey et al. 2008). However, the promoter region of Mmp-9 lacks a RARE (Munaut et al. 1999). In the current studies, we sought to determine whether vitamin A-supported MMP-9 87 activity is essential for decreased dendritic cell adhesion in vitro and the mechanism by which atRA regulates Mmp-9 expression. Materials and Methods Materials AtRA (Sigma, St. Louis, MO), AGN 194301 (a gift from Vitae Pharmaceuticals), MMP-9 lnhibitorl (Calbiochem, San Diego, CA), and actinomycin D (Sigma) were dissolved in dimethyl sulfoxide (DMSO, Sigma) and stored at -80°C under an argon atmosphere in the dark. AGN 194301 was generously provided by Dr. Chandraratna and blocks atRA activity by competing for the ligand binding domain of RARa (Nagpal and Chandraratna 2000). Recombinant GM-CSF was purchased from PeproTech (Rocky Hill, NJ) and dissolved in molecular biology grade water. Iscove’s modified Dulbecco’s medium (IMDM, Lonza BioWhittaker, Walkersville, MD) was supplemented with 10% (v/v) of characterized (CH) fetal bovine serum (FBS, HyClone, Logan UT), 2 mmollL GlutaMax (Gibco lnvitrogen, Carlsbad, CA), 100 UIL penicillin, and 100 mgIL streptomycin (Lonza BioWhittaker) to prepare complete (c)lMDM. Animals Male C57BLI6 mice (Jackson Laboratories, Bar Harbor, ME) were maintained according to institutional guidelines set by University Laboratory Animal Resources undera protocol approved by Michigan State University’s Institutional Animal Care and Use Committee. The animals were fed commercial solid diet pellets (22/5 Rodent Diet #8640, Harlan Teklad, Madison, WI) and killed by carbon dioxide asphyxiation. 88 Bone Marrow Harvest Femurs and tibiae from mice were collected and kept on ice. Bones were surface-sterilized in 70% ethanol for 2 min and transferred to RPMl-1640 medium (Lonza BioWhittaker) where the bone marrow was flushed by syringe. Bone marrow cells were filtered and centrifuged, then resuspended in ACK lysing buffer (Lonza BioWhittaker) to remove red blood cells. Bone marrow cells were then centrifuged and resuspended in RPMI-1640 medium. A portion of cells were mixed with trypan blue dye solution and counted to determine cell number and viability. Cells were cultured for DC development using a method based on that of lnaba et al. (1998) and modified by Lutz et al. (1999). Cell Culture For gelatinase activity assays, bone marrow cells were plated at 2 x 104 cells per well containing 100 pL clMDM supplemented with 20 pg/L GM-CSF and 10 nmollL atRA or AGN 194301. For MMP-9 inhibitor studies, bone marrow cells were cultured at 2 x 106 cells per 100 mm dish containing 10 mL clMDM in the presence or absence of 10 nmol/L atRA, 50 nmollL, 500 nmollL, or 5 pmollL MMP-9 Inhibitor l, and equal DMSO concentration. Fresh medium and treatment were given on days 3, 6, and 8 of culture. For gelatinase activity assays, cultures treated with atRA either received 5 pmollL MMP-9 Inhibitor l or DMSO vehicle control on d 8. For MMP-9 inhibitor studies, cells were harvested on d 10 for flow cytometry staining. 89 For atRA rescue experiments, bone marrow cells were plated at 2 x 106 cells per 100 mm dish containing 10 mL clMDM supplemented with 20 jig/L GM- CSF and 10 nmol/L AGN 194301. Cells were given 10 mL fresh clMDM containing 20 pg/L GM-CSF and 10 nmol/L AGN 194301 on d 3 and 6 of culture. On d 8 of culture, all dishes had a complete medium replacement with clMDM containing 20 pg/L GM-CSF and either 10 nmollL of atRA or AGN 194301. Cells were harvested for RNA extraction, nuclear protein extraction, and chromatin immunoprecipitation at various time points on d 8 of culture. Gelatinase Assay Adherent cell and supernatant gelatinase activities were measured using EnzChek Gelatinase/Collagenase Assay Kit (Molecular Probes, lnvitrogen, Carlsbad, CA) according to the manufacturer‘s instructions. Specifically, adherent cells or. 100 pL supernatant were incubated with 100 pg/mL DQ gelatin and 1 X Reaction Buffer to a total volume of 200 pL for 24 h at room temperature. DQ Gelatin is heavily fluorescein-labeled fluorescently quenched gelatin. Digestion of the gelatin substrate releases fluorescent peptide products. MMP-9 Inhibitor l (5 pmollL ) was added to adherent cells or supernatant in 1 X Reaction Buffer for 1 h before addition of DO gelatin substrate. Gelatinase activity was measured in the laboratory of Dr. Michael Orth, Department of Animal Science, Michigan State University using SpectraMax Gemini EM Microplate Spectrofluorometer (Molecular Devices, Sunnyvale, CA) and SoftMax Pro software with excitation wavelength of 485 nm and emission wavelength of 538 nm. Sample fluorescence 90 was compared to Clostn'dium collagenase standards to measure units of gelatinase activity. Flow Cytometric Analysis Floating and adherent cells were harvested, counted, washed with staining buffer (1% FBS, 0.1% (wlv) sodium azide, in phosphate buffered saline, pH 7.4), and incubated with anti-Fclelllll monoclonal antibody from the 2.462 hybridoma to block non-specific binding. Each sample was then incubated with either an isotype control or a cell surface molecule-specific antibody cocktail, with 1 pg antibody used per 1 x 106 cells. Monoclonal antibodies used were PE- conjugated hamster anti-mouse CD11c, PerCP-Cy5.5-conjugated rat anti-mouse Ly-6G/Ly-6C (Gr-1), PE-conjugated hamster IgG1/}(, and PerCP-Cy5.5- conjugated rat lgGZDIK, (BD Biosciences Pharrningen, San Jose, CA). Samples were run at the MSU flow cytometry facility using a LSR II flow cytometer (BD Biosciences) and data analyzed using FCS Express v.3.0 software (DeNovo Software, Ontario, CA). Real- Time PCR Total RNA was isolated from adherent cells at 8 h post atRA rescue and actinomycin D treatment during the in vitro culture period. Medium was removed from the culture dishes and the cells were incubated with TRlzol reagent (InVitrogen, Carlsbad, CA) for cell lysis. RNA extraction was completed according to the manufacturer’s instructions. RNA pellets were air-dried, resuspended in molecular biology grade water (Sigma), and stored at —80°C. 91 RNA concentration was measured at 260 and 280 nm using the CellQuant spectrophotometer (Amersham, Piscataway, NJ). Total RNA was reverse transcribed using the TaqMan Reverse Transcription System with MultiScribe reverse transcriptase and random hexamer primers (Applied Biosystems, Foster City, CA). Expression of cDNA relative to the ubiquitously expressed glyceraldehyde-3-phosphate dehydrogenase (Gapdh) housekeeping gene was analyzed using Assays-on-DemandTM Gene Expression primer/probe sets (Applied Biosystems) for Mmp-9 (Mm00600163_m1). PCR was performed on an ABI 7300 Real Time PCR System (Applied Biosystems). Monitoring of PCR occurred in real-time by detection of FAM fluorescence on the 5’ end of the probe, quenched with a non-fluorescent quencher on the 3’ end of the probe. Relative cDNA expression was calculated from AACT values and comparison to a standard curve of multiple 10-fold dilutions using ABI 7300 System 808 software v.1 .3.1 (Applied Biosystems). Nuclear Extracts Adherent dendritic cells were harvested in triplicate at 0 h, 30 min, 1, 2, and 4 h post-rescue with atRA treatment. Cells were detached from the culture dish by incubation with Accutase (Innovative Cell Technologies, Inc., San Diego, CA) at 37°C for 10 min. Detached adherent DC were washed twice with ice cold PBS. Nuclear extracts were made using NE-PER Nuclear and Cytoplasmic Extraction Reagents (Pierce Biotechnology, ThennoScientific, Rockford, IL) supplemented with Halt Protease and Phosphatase Inhibitor (Pierce Biotechnology) at 1X concentration, according to the manufacturer’s instructions. 92 Protein concentration was determined by BioRad Protein Assay (BioRad, Hercules, CA). Electrophoretic Mobility Shift Assays (EMSA) The target sequences were synthesized with biotin at the 5’ position. Complementary oligonucleotides from transcription factor binding regions of the mouse Mmp-9 promoter were annealed in a buffer containing 10 mmollL Tris base, 1 mmoI/L EDTA, and 50 mmollL sodium chloride, pH 8.0 for 5 min at 95°C, then allowed to cool slowly to room temperature. Oligonucleotide sequences used were: Sp1 sequence 1 5’-GGA GGG GAG GGG CGG GGT CA—3’, Sp1 sequence 2 5’-GTG TGT CCC CCC GCC CCC CA—3’, NF-KB sequence 5’-'l'l'T GCC CCA TGG AAT TCC CCA AAT CCT GC-3’, AP-1 sequence 5’-ACA CGC TGA GTC AGC ATA AG-3’. EMSAs were performed using LightShift Chemiluminescent EMSA Kit (Pierce Biotechnology) according to the manufacturer’s instructions except that poly(dA-dT) (Sigma) was substituted for poly(dI—dC) to block nonspecific protein-DNA binding when Sp1 target oligonucleotides were used because of the high GC content. Briefly, nuclear extracts and biotin-labeled target DNA were incubated with binding buffer, 2.5% glycerol, 10 mmol/L magnesium chloride, 50 mmoI/L potassium chloride, and 50 mg/L poly(dl—dC) or poly(dA-dT), as appropriate, with or without 200-fold excess of unlabeled target DNA competitor for 1 h at room temperature. The reactions were loaded into 5% TBE Ready Gel pre-cast gel (ReadyGel, BioRad) and electrophoresed. DNA and protein were transferred to a Biodyne 8 nylon membrane (Pierce Biotechnology), UV cross-linked at 120 mJ/cmz, blocked, 93 incubated with streptavidin-linked horseradish peroxidase, washed, incubated with substrate, and exposed to Amersham Hyperfilm MP (GE Healthcare, Buckinghamshire, UK). Chromatin Immunoprecipitation (ChlP) Chromatin immunoprecipitation was performed using ChIP-IT Express Enzymatic Magnetic Chromatin Immunoprecipitation kit (Active Motif, Carlsbad, CA) according to the manufacturers’ instructions. Briefly, protein and DNA in the adherent DC were cross-linked by incubation in 1% formaldehyde in Dulbecco’s modified Eagle’s medium for 10 minutes, then the reaction was stopped by addition of glycine stop-fix solution for 5 minutes. The adherent DC were scraped in scraping solution (500 pmol/L phenylmethylsulfonyl fluoride (PMSF) in PBS). Cells were centrifuged and resuspended in lysis buffer supplemented with protease inhibitor cocktail and PMSF. Cells were lysed and nuclei released by dounce homogenization on ice. Nuclei were collected by centrifugation. Chromatin was sheared with 10,000 UIL Enzymatic Shearing Cocktail for 10 minutes at 37°C. The shearing was stopped by addition of 10 mmollL cold EDTA. Sheared chromatin was collected by centrifugation and 50 pL was incubated at 4°C for 4 h with protein G magnetic beads, protease inhibitor cocktail, ChIP buffer, and appropriate antibody for immunoprecipitation. Anti-Acetylated Histone H3 Lys 9/14 (sc—8655—R), anti-RARa (C-20, sc—551X), anti-p300 (N-15, sc-584X) (all from Santa Cruz Biotechnology, Santa Cruz, CA), or purified mouse IgG1K (BD Pharrningen), 3 pg each per immunoprecipitation reaction, was used. Beads were collected on a magnetic stand and washed in supplied ChIP buffers. 94 Chromatin was eluted and DNA/protein cross-links were reversed by incubation with reverse cross-linking buffer at 94°C, 15 min. The chromatin solution was then treated with 1 pg Proteinase K, 1 h at 37°C followed by treatment with Proteinase K Stop Solution. Chromatin was stored at —80°C for future PCR analysis. For input chromatin samples, chromatin was prepared for PCR analysis by addition of GNP buffer 2 and a final concentration of 0.1 mol/L sodium chloride and treated for DNA/protein reverse cross-linking and protein digestion. PCR analysis of immunoprecipitated chromatin Input and immunoprecipitated chromatin were analyzed by PCR with primer pairs corresponding to a 229 bp region of the murine Mmp-9 promoter, forward: 5’-GGT CTG GGC ACA CAG GAG GC-3’, reverse: 5’-GGG GTG ACC TGG ATG GGG CA-3’. Taq DNA polymerase, buffer, magnesium chloride (2.5 mmollL final concentration), and dNTP mixture (0.2 mmol/L final concentration) were from lnvitrogen. The thermocycling conditions were 94°C for 3 minutes for 1 cycle followed by 94°C for 45 s for denaturation, 60°C for 30 s for annealing, and 72°C for 90 s for extension for 36-38 cycles. PCR products were electrophoresed through a 3% agarose gel and stained with ethidium bromide. Gel pictures were taken using Gel Doc 1000 (BioRad). Statistical Analysis Data were analyzed using lnStat software version 3.05 (GraphPad Software, Inc., San Diego, CA). One-way ANOVA followed by Bonferroni’s multiple comparison post-test was used for gelatinase assays and MMP-9 inhibitor studies. Two-way ANOVA followed by Bonferroni post-test was used for 95 transcription inhibitor quantitative PCR study. A P <0.05 was considered significant. Each experiment was repeated a minimum of two times. Results AtRA rescue treatment increases MMP-9 activity in DC cultures DC cultures treated with atRA showed 20% greater adherent cell- associated gelatinase activity than cultures treated with AGN (Figure 3.1A). Conditioned medium from atRA treated cultures also showed significantly greater gelatinase activity compared to cultures only treated with AGN (0.3 UIL compared to 0 U/L, Figure 3.18), but this gelatinase activity was less than that associated with atRA treated adherent cells. Together, atRA treatment resulted in significantly greater total gelatinase activity in DC cultures than in DC cultures treated with AGN 194301 to inhibit RARa signaling (Figure 3.1 C). Addition of specific inhibitor of MMP-9 activity over the last 48 h of cell culture and during incubation with DQ Gelatin significantly decreased adherent cell—associated gelatinase activity compared to cells treated alone with atRA, but not supernatant-associated activity, resulting in a 30% reduction in total gelatinase activity (Figure 3.1A-C). Total DC culture gelatinase activity upon inhibition of MMP-9 activity was comparable to that of DC cultures treated with AGN 194301. MMP-9 inhibitor treatment increases DC adherence The percentage of floating and adherent DC (CD11c+ Gr-1' cells, analyzed by flow cytometry) cultured from bone marrow cells under DC- generating conditions did not differ greatly between treatments (Figure 3.2A). The total DC yields also did not differ significantly between treatments. However, 96 the floating cell yield decreased significantly by 35%. The adherent cell yield did not change significantly, as the MMP—9 inhibitor dose increased in culture (Figure 3.28). Because of this, the pattern of the percentage of total DC that were adherent increased significantly, from 32% to 49%, as the dose of MMP-9 inhibitor treatment was increased in culture (Figure 3.2C). The results were similar for the total live cell population in the cultures (Appendix C). These data indicate that MMP-9, up-regulated by atRA treatment, was important in the release of the adherent cells from the surface of the culture dish. However, MMP- 9 appeared to have little effect on the overall proportion of DC within the floating and adherent cell populations. AtRA transcriptionally regulates Mmp-9 expression in adherent DC Bone marrow-derived DC treated with RARa antagonist were rescued with 10 nmollL atRA or continued treatment as before. Actinomycin D (5 pg/mL) was added to some DC cultures 1 h before atRA rescue or RARa antagonist treatment and continued following rescue. RNA extracts were made 8 h post- rescue. It was found that atRA rescue of adherent DC significantly increased Mmp-9 abundance 6-fold compared to DC treated with RARa antagonist (P<0.0001). Mmp-9 expression in DC cultures treated with actinomycin D and atRA was significantly decreased from DC cultures rescued with atRA (P<0.0001), while not different from that of DC cultures treated with RARor antagonist only (Figure 3.3). These results indicated that atRA treatnent of DC up-regulated Mmp-9 expression through a transcriptional mechanism involving signaling through RARot. 97 AtRA extra Beca hypo trans we I him we a R use pll KE rel SE Ih 0? blr shc not mos AtRA rescue of adherent DC does not increase binding of nuclear protein extracts to common regulation sites in the Mmp-9 promoter Because the Mmp-9 promoter region does not contain a consensus RARE, we hypothesized that the atRA-RARa-RXR complex likely associated with another transcription factor that is known to up-regulate Mmp-9 expression. Specifically, we hypothesized that atRA treatment would increase nuclear protein extract binding to the Sp1 consensus sequence, as this positive interaction has been previously described in other atRA-induced gene expression in promoters lacking 3 RARE (Suzuki et al. 1999; Shimada et al. 2001; Xu and Rogers 2007). We used EMSA to determine whether atRA rescue increased binding of nuclear protein extract from adherent DC to consensus binding sequences for AP-1, NF- x8, and Sp1 families of transcription factors. We found that atRA rescue had no reproducible effect on nuclear protein extract binding to Sp1 consensus sequences at any time point tested (Sp1 4 h post-rescue, Figure 3.4A, 0 h, 30 min, 1 h, 2 h, data not shown). There was no nuclear protein extract binding to the AP-1 consensus sequence from any sample, indicating that the binding conditions may not have been the most favorable despite multiple attempts at optimization (data not shown). AtRA rescue resulted in decreased nuclear extract binding to the NF-KB consensus sequence as early as 30 min post-rescue compared to the RARor antagonist treatment (Figure 3.48, 2 and 4 h data not shown). These results indicate that the atRA-RARa-RXR signaling complex does not positively interact with the transcription factors that have been shown to be most responsible for increasing Mmp-9 expression in the past to increase binding 98 to net llllelal AtRA prom DC Chic RAF resc ass. ails p3l his all thl Ira SEC CUIIl Prev; to naked DNA. This does not exclude the possibility that the hoIo-RARor-RXR can interact with one of these transcription factors when binding chromatin. AtRA rescue increases association of RARa and p300 with the Mmp-9 promoter region and increases promoter histone acetylation in adherent DC Chromatin immunoprecipitation (ChIP) was used to determine the extent of RARa and cofactor association with the Mmp-9 promoter region after atRA rescue of DC treated with RARa antagonist for 8 d. It was found that RARa association with the promoter increased over time with maximal association 24 h after atRA rescue compared to RARa antagonist treatment. At the same time, p300 histone acetyltransferase association with the promoter and acetylation of histone H3 also increased after atRA rescue to a greater extent than in RARor antagonist treated DC (Figure 3.5). These results suggest that RARa associates with the promoter region of Mmp-9 and recruits p300 to acetylate histone H3, thus activating the promoter and resulting in the increased levels of Mmp-9 transcript we have previously shown after atRA rescue. Discussion Vitamin A has been demonstrated to induce expression of Mmp-9 and secretion of pro-MMP-9 by DC in conjunction with the release of DC from the culture dish surface (Lackey et al. 2008). However, it has not been shown previously in this culture system that the pro-MMP-9 released in response to 99 atRA signaling is active and that the absence of MMP-9 activity inhibits the release of DC from the culture dish surface. When primary bone marrow cells are cultured under DC generating conditions in the presence of 10 nmollL RARa antagonist AGN 194301, adherent cell- and conditioned culture medium-associated gelatinase activity is decreased compared to cells cultured in 10 nmollL atRA. Thus, DC require signaling through RARa to produce active gelatinase in vitro. Surprisingly, adherent cell-associated gelatinase activity was greater than that of conditioned culture medium- associated activity in atRA treated DC cultures. Inhibition of MMP-9 activity using 5 pmoI/L MMP-9 Inhibitor I results in decreased adherent cell-associated gelatinase activity in atRA treated DC cultures to levels found in RARa antagonist treated cultures. This indicates that MMP-9 is a relevant DC- associated gelatinase whose activity is increased by atRA signaling through RARa. However, there was baseline gelatinase activity in both the RARa antagonist and MMP-9 inhibitor treated cultures. In a previous analysis of adhesion molecule and extracellular matrix (ECM) gene expression, a number of other MMP molecules were expressed at high levels in DC, including the gelatinase MMP-2, though they were not regulated by atRA treatment (data not shown). While pro-MMP-9 is a secreted zymogen, many studies have shown that it can interact with other proteins on the cell surface. We have previously shown that atRA rescue from RARa antagonist treatment significantly increases production and secretion of pro-MMP-9 into the conditioned medium (Lackey st 100 al. 2008), but in the current experiments, the greatest gelatinase activity was associated with the cell surface of atRA treated DC. MMP-9 has been shown to associate with the cell surface proteins CD11b, integrin-[31 (C029), and the hyaluronan receptor CD44. The cell association can be useful to direct cell migration through the ECM or to increase activation of MMP-9 (Fridman et al. 2003; Radjabi et al. 2008). The design of the gelatinase assay experiments cannot resolve whether the cell-associated MMP-9 activity is due to MMP-9 association with the cell surface or the ECM. It is possible that active MMP-9 does associate with the DC surface itself as myeloid DC express high levels of CD11b (Shortman and Liu 2002). However, it is also possible that the active MMP-9 is within the DC-produced ECM in the dish where it actively digests that matrix. Pro-MMP-9 has a gelatin binding domain that allows it to associate with collagen a2(lV) chain in the ECM or associated with the cell surface with high affinity, inducing gelatinase activity without the proteolytic removal of the pro- peptide (Fridman et al. 2003). Flow cytometric analysis of DC using fluorescently labeled antibody against MMP-9 would resolve whether the MMP-9 is associated with proteins on the DC surface or with the ECM. Because MMP-9 was determined to be the primary source of gelatinase activity in DC cultures treated with atRA, the role of atRA-induced MMP-9 activity in releasing adherent bone marrow-derived DC was also examined. Inhibition of MMP-9 results in increased total cell and DC adhesion in vitro, compared to cultures treated only with atRA. These results indicate that MMP-9, up-regulated by atRA treatment, is important in the release of adherent DC from the culture 101 dish surface. MMP-9 does not regulate the proportion of bone marrow derived cells that develop into DC (Figure 3.2). This must be a different function of atRA, such as direct up-regulation of differentiation-related transcription factors. Because we were able to show that atRA increased MMP-9 activity and that this activity resulted in a decreased percentage of cells that were adherent in DC cultures in vitro, we next sought to determine the mechanism by which atRA signaling through RARa up-regulates Mmp-9 expression. While we have shown that atRA up-regulates Mmp-9 transcript abundance through a mechanism involving RARa (Lackey et al. 2008), the promoter region up-stream of the Mmp- 9 gene does not contain a classic consensus RARE. We aimed to determine the mechanism through which atRA acts to increase Mmp-9 expression in DC that results in developmental changes (adherent cell to floating cell) in our culture system. We first sought to establish whether the atRA-induced up-regulation of Mmp-9 expression in adherent DC is transcriptionally activated using the transcription inhibitor actinomycin D. It was found that atRA rescue increased Mmp-9 expression compared to RARa antagonist treatment, while actinomycin D treatment with atRA rescue did not increase Mmp-9 abundance. These results indicated that Mmp-9 expression in adherent DC is controlled by atRA through a transcriptional mechanism involving RARa. We hypothesized that due to that lack of a consensus RARE in the Mmp-9 promoter region, the holo-RARor-RXR complex interacts with another transcription factor to positively increase its binding, specifically Sp1. Nuclear protein extracts from primary bone marrow cells cultured under DC generating 102 conditions in the presence of the RARor antagonist AGN 194301 did not increase association with Sp1, AP-1, or NF-KB oligonucleotide probes upon treatment with 10 nmollL atRA. In fact, atRA treatment decreased protein binding to the NF-xB consensus sequences. It is likely that fresh GM-CSF supplied during medium change resulted in the initial increased nuclear protein binding to the NF-tcB consensus sequence that was inhibited by atRA treatment. GM-CSF, signaling through the GM-CSFRa and common cytokine receptor [3 chain, phosphorylates Jak2 which in turn activates tyrosine kinase. Tyrosine kinase activates Ras, which further activates Raf (mitogen-activated protein kinase/ERK kinase kinase, MEKK). MEKK phosphorylates and activates inhibitor of kappa B kinase (IKK)—|3, which phosphorylates inhibitor of kappa B (IKB) bound to NF-KB subunits to mark it for eventual proteosomal degradation. IKB phosphorylation releases NF-KB subunits to be phosphorylated and translocated to the nucleus. Tyrosine kinase also activates the phosphotidylinositol 3—kinase-Akt pathway, resulting in both IKKB and Ras activation (Hiraguri et al. 1997; Chang et al. 2003). Because there was no nuclear extract binding to the AP-1 consensus sequence from any sample, it is also possible that the binding conditions used were not optimal. In order to increase nuclear protein extract binding to the AP-1 consensus sequence, we increased the concentrations of glycerol, magnesium chloride, and potassium chloride as well as the time of binding from 20 min to 1 h. Because there was no increase in nuclear protein extract binding to these consensus sequences, it is not likely that atRA signaling through RARa works by positively interacting with one of the common families of transcription factors. 103 Chromatin was isolated from atRA-rescued DC cultures to determine which proteins associated with the Mmp-9 promoter following atRA treatment. Using ChIP, we found that both RARor and p300 associated with the Mmp-9 promoter following 24 h atRA treatment. P300 is a histone acetyltransferase that acetylates histone H3 at Iysines 14 and 18 and is a cofactor for RARa-mediated transcriptional regulation. It was also found that atRA rescue increased the acetylation of histone H3 at the lysine 9 and 14 postion as determined by immunoprecipitation. These results indicate that RARa interacted with the Mmp-9 promoter, despite the lack of a RARE within the promoter, and increased the potential for transcriptional activity due to the acetylation status of histone H3. However, it cannot be determined from these results whether RARa actually makes physical contact with the Mmp-9 promoter DNA directly or if it interacts with another protein that does. Altemately, the plasmid construct transfected contained 2200 base pairs of the Mmp-9 5’ untranslated region, but a RARE could further up-stream in the endogenous MMP-9 5’ untranslated region. In an experiment where we transfected DC with Mmp-9 promoter-firefly luciferase reporter plasmid construct, atRA rescue did not increase Mmp-9 promoter activity in adherent DC compared to RARa antagonist treated cultures (data not shown), indicating that the holo-RARa-RXR complex acts to increase Mmp-9 expression through an epigenetic mechanism because plasmid DNA is not histone-associated. The results from the ChlP experiments confirmed those of Zaragoza et al. (2007) who showed increased RARa and p300 associated with the Mmp-9 promoter region after vitamin A supplementation during mammary 104 involution in rat dams. We have previously found that atRA treatment increased Mmp-9 expression within 4 h of treatment (Lackey et al. 2008), but we did not see increased RARa and p300 association with the promoter after 4 h of atRA treatment. This indicates that RARa and p300 may associate with the Mmp-9 promoter for Iong-terrn atRA regulation of Mmp-9 expression. However, there may be an indirect mechanism for early atRA regulation of Mmp—9 expression, such as the displacement of NF-xB from the promoter. It would be interesting to also determine whether there is increased interaction of the histone methyltransferase MLL5 with, and hence increased methylation of, the Mmp-9 promoter region. It has been previously shown that MLL5, a histone lysine mono— and di-methyl transferase, acts as a co-regulator of RARa-induced transcription by methylating lysine 4 on histone H3 after it has been N-glycosylated (Fujiki et al. 2009). Methylated histone H3 (K4) and acetylated histone H3 also result in more permissive chromatin in a number of promoter regions, including that for Mmp-9, as well as increased association with MLL (Yan and Boyd 2006). We have shown MMP-9 activity, located mainly at the adherent DC surface or in the ECM, is responsible for the loss of DC adherence after atRA rescue through a transcriptional mechanism involving RARa. Further, we have shown that the transcriptional mechanism by which atRA up-regulates Mmp-9 expression for long-terrn regulation involves the interaction of RARa with the Mmp-9 promoter region along with the recruitment of the histone acetyl transferase p300 and increased acetylation of histone H3. AtRA treatment did not 105 increase transcription factor binding to AP-1, NF-KB, or Sp1 non-chromosomal DNA binding elements. However, these results do not necessarily preclude the possibility that the atRA-RARa-RXR complex could interact with one of these transcription factors to increase binding to the chromosomal DNA associated with histone proteins and specific to the Mmp-9 promoter. Additional studies are necessary to further elucidate the atRA-induced Mmp-9 expression mechanism in the mouse DC model. These studies will determine the mechanism of holo- RARa-RXR recognition of the Mmp-9 promoter region and with what other protein(s) the complex likely interacts. DC migration through the ECM is a critical aspect of their function, from release of immature DC from the bone marrow to the periphery to the migration of mature DC to secondary lymphoid tissues. Our findings suggest that vitamin A may play a large role in bridging innate and adaptive immunity through increased MMP-9 activity at the DC surface, allowing increased cell migration potential. Acknowledgements We thank Dr. Louis King of the Michigan State University flow cytometry facility in the Research and Technology Support Facilities. 106 Figur piese A. Ad lotal C‘JIIUI treatn detec diliell Figure 3.1. DC gelatinase activity after 10 d culture with atRA treatment in the presence or absence of MMP-9 inhibitor or with RARor antagonist, AGN 194301. A, Adherent cell and 8, conditioned cell culture medium gelatinase activity. C, Total gelatinase activity, calculated from adherent cell and conditioned cell culture medium activity. Results are presented as the mean + SD, n=5 for each treatment group, and show one representative experiment of two. (ND, not detectable). Treatment groups designated with different letters are significantly different with P <0.05 as determined by one-way ANOVA. 107 o. a _. t a T a N. 0...:ng or a .D _ a T .0 7 so i 9 as. _ see T _ .o b _ b fl — Vows .... a .e o ...r_. mu. m m ._. .... 9.. ._. o :5. 3.2.2 32.58 o o. o. o. :5. 3.2.3 82.33 .22. :8 222.3 :5. 3.53 8358 52.8: =8 8:23.28 Figure 3.1 108 Treatment Figure 3.2. Dendritic cell proportions and yields after 10 d culture in the presence or absence of MMP-9 inhibitor. A, Percent CD11c+ DC floating or adherent to culture dish. 8, Floating and adherent DC yields, calculated based on cell yield and proportion of cells that are DC. C, The percentage of total DC that are adherent, calculated as adherent DC yield divided by total DC yield multiplied by 100%. All cultures were treated with 10 nmollL atRA. Results are presented as the mean + SD, n=4 for each treatment group, and show one representative experiment of two. Treatment groups designated with different letters are significantly different with P <0.05 as determined by one-way ANOVA. 109 - Floating 100‘ l:lAdherent m=oo +0500 3 _ 0 1.2“... 5.. 29> :8 +350 4 5 C I «o xv ox es c e b. «as m be m Q t as a co m T e o \ so b.T so a 9% 6» so _ . _ 5 0 5 0 7 5 2 E30... U< so as... 2.8 .250 s Figure 3.2 110 Figure 3.3. Adherent DC Mmp-9 expression after d 8 atRA rescue with or without transcriptional inhibitor. Total RNA extracts made from adherent cells on d 8 of culture. Cultures were pre-treated with 5 pg/mL actinomycin D (ActD) 1 h before atRA rescue from RARa receptor treatment. The extracts were reverse transcribed to cDNA and Mmp-9 gene expression was assayed by quantitative real-time PCR in technical triplicates. Values were normalized to glyceraldehyde- 3-phosphate dehydrogenase expression and are shown as fold of the RARa antagonist treatment. The results are combined from 2 independent experiments and are presented as mean + SEM, n=4 for each treatment. Means without a common letter differ, P<0.05. Two-way ANOVA indicates significant interaction between atRA treatment and ActD treatment. 111 Mmp-9 Expression C O _ '3 7'5 [:1 Control 3]). g - ActD I.I.I _ Interaction P<0.0001 < 5.0 2 m E 2 2.5~ if a at —— a a g 00 L S ' RARa Antagonist atRA Treatment Figure 3.3 112 Figure 3.4. AtRA rescue of mouse bone marrow-derived adherent DC does not increase nuclear protein binding to common Mmp-9 regulatory sequences. Bone marrow cultures were grown in the presence of GM-CSF and RARa antagonist for 8 d, then either remained in culture with 10 nmollL RARa antagonist (AGN) or were rescued with 10 nmollL atRA. Nuclear protein was extracted 0 h, 30 min, 1, 2, and 4 h after RARa antagonist treatment or atRA rescue, then subjected to EMSA. Gel shifts using oligonucleotides for A, Sp1 (4 h); and 8, NF-KB (30 min and 1 h) consensus binding sequences matching those found in the mouse Mmp-9 promoter are shown. The results are representative of 2 independent experiments, n=3 for each treatment, although only 2 of 3 replicates are shown for NF-KB in panel 8. 113 Sp1 ID .0 9 n. AGN atRA Sp1 Shift Free Probe Unlabeled Probe NF-ch h 30 min 1 h a) .o 9 Q > 0 Z > CD 2 32 > (D Z I; NF-KB Shiftl Non-specific Binding — Free Probe Unlabeled Probe Figure 3.4 114 Figure 3.5. AtRA rescue increases RARa association with and acetylation of histone H3 in the mouse Mmp-9 up-stream promoter region. Bone marrow cultures were grown in the presence of GM-CSF and RARa antagonist for 8 (I, then either remained in culture with 10 nmol/L RARa antagonist or were rescued with 10 nmol/L atRA. Adherent DC were formaldehyde-fixed 0, 4, and 24 h after RARa antagonist treatment or atRA rescue, chromatin sheared, and immunoprecipitated against RARor, p300, acetylated histone H3, or an irrelevant lgG. Immunoprecipitated or diluted input chromatin was then amplified by PCR using primers for a 229 bp region of the murine Mmp-9 promoter located at -1087 bp to -859 bp from the transcription start site and products were separated by agarose electrophoresis. The results are representative of 2 independent experiments, n=2 for each treatment group. 115 Mmp-9 Promoter 0h 4h 24h Input Fbme35 116 CHAPTER 4 SUMMARY AND FUTURE DIRECTIONS 117 Summary Vitamin A deficiency has been known to affect morbidity and mortality since early in its study, including increased infection of the gastrointestinal, urogenital, and respiratory tracts in vitamin A deficient rats (Green and Mellanby 1928). Since the early studies showing an increased prevalence of mucosal site infections, the results have been substantiated by observations of increased respiratory infection (Jason et al. 2002) and intestinal permeability (Quadro et al. 2000) in children with low plasma retinol concentrations. It has been determined that the increased prevalence of mucosal infections is related to impaired production of secretory lgA in response to infection (Stephensen 2001). Vitamin A supplementation increased secretory lgA production in response to Shigella infection (Villamor and Fawzi 2000) and hepatitis B vaccine (Newton et al. 2007). In experimental models of vitamin A deficiency, T cells from vitamin A deficient mice resulted in decreased IgG1 production by B cells from vitamin A sufficient mice, revealing that vitamin A had an effect on the T cells which then led to impaired B cell antibody production (Carman et al. 1989). Production of both IgG1 and lgA isotypes required Th2 help. Vitamin A deficiency in mice has also been linked to decreased symptoms of allergic airway inflammation due to decreased serum lgE and IgG1 and decreased brochoalveolar lavage lL-4 and — 5, all responses that require Th2 support for production (Schuster et al. 2008). These observations have led to studies of vitamin A’s effect on naive T cell development. 118 In vitamin A deficient mice, the Th1 response to Trichinella spiralis infection was enhanced through secretion of IFN-y, which also suppressed the Th2 response by decreasing lL—5 and -10 production leading to an improved cell- mediated immune response and impaired humoral immunity (Cantorna et al. 1996). Addition of atRA to Th1 cell culture inhibited lFN-y production. AtRA also enhanced the development of naive T cells to Th2 cells through an antigen F presenting cell intermediate (Hoag et al. 2002). Myeloid DC are one type of antigen presenting cell that require atRA to develop from myeloid progenitor cells in culture. Without sufficient atRA, the progenitor cells developed into neutrophils [ instead (Hengesbach and Hoag 2004). These in vitro studies confirmed in vivo observations which showed increased neutrophil numbers in the spleen, peripheral blood, and bone marrow of vitamin A deficient mice (Kuwata et al. 2000). However, these expanded neutrophils had impaired migration, phagocytosis, and bacterial killing functions (Twining et al. 1997; Stephensen 2001) Myeloid DC, characterized as CD8a', CD11b+, and CD11c+ in mice, are derived from hematopoietic progenitor cells in the bone marrow (Shortman and Liu 2002). DC precursors migrate from the bone marrow, via the blood, into the different lymphoid tissues where they differentiate (lwasaki 2007). Immature myeloid DC resident in peripheral tissues capture, process, and present antigen in MHC class I and II molecules (Banchereau et al. 2000). DC mature in response to stimulation by antigen and stress signals through the pattern- recognition toll-like receptors (Hochrein and O'Keeffe 2008), leading to reduced 119 antigen capturing and induced migration to the T cell areas of secondary lymphoid tissues (Banchereau and Steinman 1998; Steinman 1999). There, DC activate naive T cells through antigen presentation along with the secretion of specific influential cytokines. Myeloid DC are known to polarize na'l've CD4+ T cells to Th2 cells through stimulation of the T cells with antigen-bound MHC class II and production of MCP-1 and OX40L (Moser and Murphy 2000; de Jong et al. 2005; Hochrein and O'Keeffe 2008). if MMP-9, a gelatinase that can degrade gelatin, types I and IV collagen, and laminin in the ECM, aids leukocyte migration into blood and lymphatic i vessels (Faveeuw et al. 2001; Opdenakker et al. 2001). DC migration has been found to be heavily reliant on MMP-9 activity. Inhibition of MMP-9 activity, through the use of monoclonal neutralizing antibody, the natural inhibitor TIMP-1, and MMP-9-,” mouse models, impaired the release of hematopoietic progenitor cells from the bone marrow (Pruijt et al. 1999), recruitment of bronchoalveolar associated lymphoid tissue DC to ainrvay lumen in response to allergen (Vermaelen et al. 2003), migration of cultured DC through tight epithelial junctions (Ichiyasu et al. 2004), and migration of cultured monocyte-derived DC (Baratelli et al. 2004). The chemokine CCL5, released under inflammatory conditions, induces a fold increase in pro-MMP-9 production by immature DC within 10 h to aid in migration through basement membranes (Chabot et al. 2006). MMP-9 production has been reported to be both positively and negatively regulated by atRA signaling. In tumor cell invasion (T sang and Crowe 2001; 120 Andela and Rosier 2004) and emphysema (Mao et al. 2003), atRA treatment decreased MMP-9 production. AtRA negatively regulated Mmp-9 production in squamous cell carcinoma lines through displacement of AP-1’s coactivators from the Mmp-9 promoter (Tsai et al. 2008). However, atRA also positively regulated Mmp-9 production in mammary epithelial cells (Montesano and Soulie 2002) and rat mammary tissue (Zaragoza et al. 2007) for lumen formation and involution, respectively, through recruitment of p300 to the Mmp—9 promoter, even though the Mmp-9 promoter lacks a RARE consensus sequence for the hoIo-RAR-RXR complex to recognize and bind. AtRA decreases DC adhesion We found that treatment of bone marrow cell cultures under DC-driving conditions with atRA in combination with CD-FBS supplemented medium led to increased production of CD11c+ DC and decreased yield of Gr—1+ neutrophils compared to cells in CD-F BS supplemented medium without atRA or in CH-FBS supplemented medium with the RARa-specific receptor antagonist AGN 194301. At the same time, use of the receptor antagonist or CD-F BS supplemented medium without atRA increased the percentage of DC that remain adherent after a 10 d culture period compared to cultures treated with atRA. Rescue of receptor antagonist-treated DC cultures with atRA on culture (I 8 resulted in increased overall floating cell yield with no change in adherent cell yield by d 10 of the culture period, resulting in a decreased percentage of cells that remain adherent compared to receptor antagonist treated cultures. The floating CD11c+ DC yield and proportion significantly increased by d 10 after atRA rescue compared to 121 receptor antagonist treated cultures. The percentage of floating Gr-1+ neutrophils did not differ between the two groups, but the yield increased after atRA rescue due to the overall floating cell yield increase. None of the adherent cell yields or DC percentage differed between the two treatments, though the adherent DC yield decreased insignificantly. These results indicate that atRA signaling through RARa is necessary for normal DC development and loss of adherence. Blocking this signal results in bone marrow cells developing primarily into neutrophils with the majority of DC remaining as adherent cells. Rescue of bone marrow cell cultures with atRA for 48 h after 8 d treatment with RARa antagonist is not long enough to yield a greater proportion of floating DC than floating neutrophils. Bone marrow cells cultured in the presence of GM-CSF, but with blocked signaling through RARa, develop into primary neutrophils, and the atRA rescue allows for their maintenance, while differentiation of myeloid precursors to DC only begins after addition of atRA to the culture medium. While examining factors that could account for changes in DC adherence after atRA rescue from receptor antagonist treatment with DC, it was found that atRA treatment decreased the magnitude of the cell adhesion molecule CD11a (integrin-ctL) expression on both the total cell population and the DC population of both adherent and floating cells. However, there was no difference in ltgal transcript expression between the two treatments over 48 h, indicating that atRA does not directly reduce ltgal expression at the transcription level. The decrease in CD11a cell-surface expression could not be explained by leAM-1 binding to CD11a and thus blocking the antibody binding to the relevant epitope, because 122 there was no difference in leAM-1 concentrations in conditioned supernatant between the two treatments. The use of an extracellular matrix and adhesion molecule-specific microarray identified Mmp-9 as a factor that was increased by atRA rescue of DC. There were a number of other extracellular matrix and adhesion molecule genes that were modulated either by atRA treatment or by developmental stage of the DC (adherent or floating, Appendix A). For example, the proteoglycanase A disintegrin-like and metalloprotease with thrombospondin type 1 motif (Adamts)-8 that, along with ADAMTS-1, -4, -5, -9, -15, and -20, cleaves aggrecan, versican, and brevican (Apte 2009) was down-regulated in both adherent and floating DC. The MMP modulator basigin (CD147) was upregulated in atRA-treated floating DC, but not in atRA-treated adherent DC. It has been reported that the presence of basigin on the DC surface indicates a later developmental stage (Woodhead et al. 1998). N-glycosylated basigin induces the expression of MMP-1, —2, -3, -9, -11. There is evidence that basigin, as a mediator of cell-cell interaction and an adhesion molecule, expression on antigen presenting cells is required for T cell activation (Iacono et al. 2007). It was confirmed that atRA rescue increased Mmp-9 production in adherent DC by 4 h with maximal production in 24 h post-rescue. AtRA rescue had no effect on the production of Timp-1, the chief natural inhibitor of MMP-9, in adherent DC. Similarly, atRA rescue increased the concentration of secreted pro- MMP-9 to a greater extent than the secretion of TIMP-1, resulting in a molecular excess of pro-MMP-9 and the potential for greater degradation of the ECM upon 123 atRA rescue compared to cultures treated with receptor antagonist. These results led to the hypothesis that the activity of MMP-9, whose production was increased in adherent DC by atRA rescue, is responsible for the decreased DC adherence due to atRA signaling through RARa. It was also hypothesized that this mechanism was partly due to atRA-induced MMP-9 activity resulting in decreased CD11a cell surface expression in atRA treated cells due to direct cleavage by MMP-9. AtRA increases MMP-9 activity It was important to determine whether the MMP-9 produced by DC upon atRA signaling through RARa was active. It was found that gelatinase activity was increased on the surface of adherent cells when treated with atRA compared to cells treated with receptor antagonist. When MMP-9 inhibitor was added to atRA treated DC cultures, gelatinase activity was reduced to that of receptor antagonist treated cultures. These results indicate atRA treatment increased gelatinase activity of DC cultures and that MMP-9 was a primary gelatinase responsible for atRA-induced gelatinase activity through RARot signaling. We also found that culturing bone marrow cells under DC-generating conditions in the presence of atRA and MMP-9 inhibitor increased the percent of all cells that remained adherent as well as the percent of CD11c+ DC that remained adherent. These results are summarized in Figure 4.1. The inhibitor did not affect the proportion of cells that were DC or the cell yield. The inhibitor also did not affect the cell surface expression of CD11a compared to cultures only treated with atRA. Thus, atRA-induced MMP-9 expression and activity was only responsible 124 for cell adherence, not progenitor cell differentiation into DC or cell proliferation. Unlike what was hypothesized, MMP-9 activity was not responsible for the atRA- induced decrease in CD11a cell surface expression. RARa interacts directly through the MMP-9 promoter It was important to determine the mechanism by which atRA regulated Mmp-9 expression in DC. Through the use of the transcription inhibitor actinomycin D, we found that atRA positively regulated Mmp-9 transcript abundance through a transcriptional mechanism. Because the Mmp-9 promoter does not contain a classic RARE, we measured binding of nuclear protein extracts to consensus oligonucleotide sequences in the Mmp-9 promoter using EMSA and found that neither Sp1, AP-1, nor NF-tcB showed increased binding to nuclear protein extracts after atRA treatment. However, the oligonucleotide sequences used in EMSA are not histone associated and the nuclear extract protein binding to the oligonucleotides only indicates global binding to these sequences in response to atRA, not Mmp-9 promoter specific binding. The binding also does not take place in the context of the cell environment. It is still possible that the holo-RARa-RXR complex interacts with one of these transcription factors in the early period after atRA treatment to increase binding to the chromosomal DNA associated with histone proteins and specific to the Mmp-9 promoter. While the EMSA results indicate that it is unlikely that the holo- RARa-RXR interacts with transcription factors in the Sp1, AP-1, or NF-icB families to positively increase binding to the Mmp-9 promoter to induce Mmp—9 transcription, this conclusion is not definitive. We did not do supershift assays to 125 determine whether there was increased binding of individual proteins within the nuclear protein extracts in the Sp1 family [Sp1, Sp3, or the transcriptional repressive Sp1-like Krilppel-like factor (KPF) members (Lomberk and Urrutia 2005)] after atRA rescue of adherent DC even without increased overall nuclear protein binding to the Sp1 consensus sequences. It is possible that with RARa antagonist treatment results in increased Sp1-like KPF binding to the Sp1 consensus sequences to block gene transcription, while atRA rescue results in increased Sp1 or Sp3 binding to the same sequences to up-regulate gene transcription. This state could result in a similar band shift as visualized by EMSA despite opposite biological effects in vivo. Through chromatin immunoprecipitation experiments, we found that 24 h atRA rescue increased RARa and histone acetyltransferase p300 association with the Mmp-9 promoter and increased acetylation of lysine 14 on the tail of histone H3, indicating a state permissive for increased gene expression. We have previously found that atRA treatment increased Mmp-9 expression within 4 h of treatment (Lackey et al. 2008), but we did not see increased RARa and p300 association with the promoter after 4 h of atRA treatment. These results indicate that the atRA-RARa-RXR signaling complex most likely interacts with an as yet unknown factor that recognizes the Mmp-9 promoter to increase its interaction with the promoter, recruit p300 to the promoter, and increase histone H3 acetylation and thus open the promoter for interaction with transcription machinery for increased gene expression for long-term expression regulation by atRA (Figure 4.2). However, for early atRA regulation, there may be an indirect 126 mechanism for Mmp—9 expression, such as the displacement of NF-KB from the promoter or increased binding of Sp1 to the promoter, which was only tested by EMSA and not by ChIP. 127 Future directions What other proteins interact with the holo-RARar—RXR complex at the Mmp-9 promoter region ? Our research indicated that holo—RARa-RXR interacted with the Mmp-9 promoter to recruit p300 and thus increased the acetylation of the Mmp-9 promoter and subsequent expression of Mmp-9. However, the RARa-RXR signaling complex cannot recognize and bind to the Mmp-9 promoter through currently recognized mechanisms due to a lack of a consensus RARE (Munaut et al. 1999). It is likely that the holo-RARa-RXR signaling complex interacts with another protein that can recognize and directly bind to the Mmp-9 promoter. While we could not demonstrate that atRA treatment of DC increased nuclear protein extract binding to the AP-1, NF-KB, or Sp1 consensus oligonucleotide sequences, it is still possible that in the environment of the cell and interaction with the histone associated Mmp-9 promoter, one of these transcription factors does interact with the hoIo-RARa-RXR complex. To determine the identity of the potential protein with which the holo-RARa-RXR complex interacts, it will be necessary to make nuclear protein extracts of adherent DC after treatment with atRA or RARor antagonist. The nuclear protein extracts must then be immunoprecipitated against RARa to enrich for proteins that associate with RARa in the nucleus of DC. The immunoprecipitate would be electrophoresed in two dimensions, first by isoelectric focusing, then by size separation. The separated proteins from each treatment of extract could then be compared for 128 differences in intensity and selected for further identification of the proteins from the atRA treatment that are more intense than those from the RARa antagonist treatment. The selected proteins would then be trypsin digested for preparation for tandem mass spectrometry to identify the possible proteins that interact with RARa in this system. Further experiments could test whether candidate proteins identified by mass spectrometry interact with the Mmp-9 promoter through the use of sequential chromatin immunoprecipitation (ChIP-ReChlP). First, fixed and sheared chromatin-protein complexes would be immunoprecipitated against RARa, then immunoprecipitated again using an antibody specific for the candidate protein. The complexes would then be reverse cross-linked followed by treatment with proteinase K, and the resulting chromatin would be amplified by PCR using primers designed to amplify the Mmp-9 promoter region. Our research has inferred an interaction between RARa and p300 in DC because they both show increased association with the assayed Mmp-9 promoter region after atRA treatment at 24 h. However, we have not shown that RARa and p300 interact in the same complex on the Mmp—9 promoter. ChlP- ReChIP could be used to determine whether these proteins interact in the same complex on the Mmp-9 promoter. DC cultures would be treated as before with 10 nmol/L RARa antagonist for 8 d followed by 24 h treatment with atRA or RARa antagonist. DC would be fixed and lysed, followed by chromatin shearing. Because p300 is more abundant 24 h after atRA treatment, it would be best to first immunoprecipitate the fixed chromatin-protein complexes using the antibody 129 against p300 followed by immunoprecipitation against RARa. The immunoprecipitated chromatin would then be amplified by PCR using the Mmp-9 promoter primers to visualize whether atRA rescue treatment increases binding of RARor and p300 in the same complex to the Mmp-9 promoter. Histone acetylation is not the only epigenetic change leading to increased chromatin accessibility to transcription machinery and increased gene expression. Even though methylation, such as at the K9 position on histone H3, is most frequently associated with gene silencing, methylation of histone H3 (K4) is associated with transcription elongation (Jenuwein and Allis 2001). It has been previously shown that the histone methyltransferase MLL5, a histone lysine mono- and di-methyl transferase, after it has been N-glycosylated, acts as a co- regulator of RARa-induced transcription by methylating lysine 4 on histone H3 (Fujiki et al. 2009). Methylation of histone H3 (K4) and acetylation of histone H3 also result in chromatin more permissive to transcriptional activity in a number of promoter regions, including that for Mmp-9, as well as increased association with MLL (Yan and Boyd 2006). Experiments using 24 h 10 nmol/L atRA rescue of DC after RARa antagonist followed by ChIP against MLL5 and lysine 4-methylated histone H3 followed by PCR amplification of the immunoprecipitated chromatin using Mmp-9 promoter primers would be useful in further determining the mechanism of atRA-induced Mmp-9 expression in DC. Does vitamin A deficiency impair DC migration and MMP-9 production in vivo? We have shown that MMP-9 produced by DC in response to atRA treatment is active in culture and contributes to a change in adherence. The next 130 step is to determine whether vitamin A status is linked to MMP-9 production and DC migration in vivo. To determine whether vitamin A deficiency has a systemic effect on MMP-9 production, the serum from vitamin A deficient mice could be tested for a significant decrease in pro-MMP-9 production by ELISA compared to the serum of vitamin A sufficient mice. If there is no difference in serum pro- MMP-9 between vitamin A deficient and sufficient groups, immature DC from the spleen could be purified and RNA extracted and reverse transcribed for quantitative PCR analysis of differences in Mmp-9 expression between the two diet groups. Based on the results determining whether MMP-9 production is systemically reduced during vitamin A deficiency in the mouse, either bone marrow-derived DC would be injected into vitamin A deficient or sufficient mice, or, if vitamin A deficiency does not result in systemically reduced MMP-9 production, bone marrow-derived DC treated either with atRA or RARot antagonist would be injected into vitamin A deficient mice. Mice would be sensitized to ovalbumin by intraperitoneal injection along with alum adjuvant, followed by exposure to 1% ovalbumin aerosol for 7 consecutive days two weeks later. F luorescentIy-labeled bone marrow—derived DC would be injected into vitamin A deficient or sufficient mice with allergic ainlvay inflammation. After the last day of aerosol exposure, bronchoalveolar lavage cells would be collected along with the lung and lymph nodes for processing into single cell suspensions for analysis by flow cytometry to determine the level of DC infiltration. It is expected that vitamin A deficiency, either of the animal or of the cultured DC, 131 would result in impaired DC migration to the site of ovalbumin-induced inflammation and to the lymph nodes. How are other extracellular matrix and adhesion molecules on DC afi’ected by atRA treatment? Many other genes for extracellular matrix and adhesion molecules were found to be regulated either by atRA or by floating versus adherent phenotype or developmental state. Two of these genes are AdamtsB and basigin. AdamtsB expression is likely down-regulated by atRA, while basigin expression seems to - be related to the later developmental state of the floating DC. As basigin expression is more closely related to the DC developmental state, it is likely that it may be up-regulated by atRA, but through an indirect mechanism related to its role in DC development. For both of these genes, it would be important to confirm gene expression level using reverse transcription and real-time PCR techniques over a time course to determine gene expression kinetics. It would next be important to confirm gene expression with protein expression. For ADAMTSB, protein expression could be confirmed by western blot using total protein extracts from adherent DC or ELISA using conditioned medium from DC cultures. For basigin, flow cytometric analysis of both floating and adherent cultured cells stained for CD11c, Gr—1, and CD147 (basigin) would be useful for determining the percentage of DC expressing basigin and the average expression level on the cell surface of both floating and adherent DC. How does atRA regulate CD11a surface expression? 132 AtRA treatment of DC resulted in decreased cell surface expression of the adhesion molecule CD11a (integrin-aL), but mRNA expression was not affected. We have also shown that inhibition of MMP-9 activity in the presence of atRA had no effect on the CD11a cell surface expression (Appendix D). The next step would be to determine how atRA regulates DC CD11a surface expression. The role of atRA treatment on CD11a protein expression could be explored by inhibiting both the rate of CD11a translation and the protein degradation machinery. In experiments where CD11a translation is inhibited, bone marrow cells would be grown in DC generating conditions in the presence of AGN 194301 as described in previous atRA rescue experiments. On day 8 of the culture period, cells would receive 20 mL fresh clMDM supplemented with 20 pg/L GM-CSF and either 10 nmollL AGN 194301 or 10 nmoVL atRA. Some culture plates in each treatment group would also be supplemented with the translational inhibitor cycloheximide on day 8. On day 10, the floating and adherent cells would be harvested separately, counted, and stained with fluorescently labeled monoclonal antibodies against Gr-1, CD11c, and CD11a for analysis by flow cytometry. A significant decrease in cell surface CD11a expression in cultures treated with both AGN and cycloheximide compared to cells only treated with AGN to a level of expression as seen in cultures rescued with atRA in the presence or absence of cycloheximide would indicate that atRA acts by decreasing the rate of translation from ltgal mRNA to protein. It would also have to be taken into consideration that atRA signaling through RARa could act to decrease CD11a protein half-life. If there were a 133 significant decrease in CD11a cell surface expression after atRA rescue and cycloheximide treatment compared to cells only rescued with atRA, the data would indicate that atRA signaling does not affect the translation rate of ltgal mRNA to protein but that the rate of degradation is greater in the presence of this signaling. In this case, it would be necessary to follow the experiment with studies using inhibitors of 208 and 26S subunits of the proteasome. Cells would be cultured in the same way as for the protein synthesis inhibitor studies, except the AdaAhX3L3VS proteasome inhibitor at 10 pmol/L dose would be used in the place of cycloheximide. If atRA acts to increase the degradation rate of CD11a, blocking degradation would lead to an increase in CD11a cell surface expression comparable to that of AGN treated cells. Overall, the results of these experiments would help to explain how atRA acts to decrease cell surface expression of CD11a on DC cultures in vitro. It is also possible that atRA treatment increased recycling of CD11a from the cell surface. To explore this possibility, bone marrow cells cultured under DC generating conditions and either treated with RARa antagonist or rescued with atRA could be harvested after 10 days and then incubated with fluorescently— labeled antibody against CD11a. Cellular localization could be determined by immunocytochemistry. If there was greater internally located CD11a after atRA rescue compared to RARa antagonist treated cells, these results would indicate that atRA did regulate cell surface CD11a expression through a mechanism based on recycling of the protein from the surface. 134 Figure 4.1. Model of retinoic acid’s effect on dendritic cell development. Retinoic acid receptor (RAR)-a antagonist treatment results in decreased DC yield and increased DC adhesion with mostly neutrophil development. Addition of retinoic acid to cultures increases MMP-9 production and activity, especially at the adherent DC surface, resulting in more floating DC. 135 o_o< eczema __oo ozeeoo E923 miss. __;oo_soz :8 2550 2V esom c . t, _ . pp o ® o new o me etc ® i0 .4 7 . ® e on. Q 4...... , )J or @ sew .... \\l/iIl\\////Iliit\t\\iII//lill\\\z 33:855.. om =oo 30‘ - Floating q a d _ _ O O O O O 5 4 3 2 1 E9853 can .9: 2.00 .30... he .x. Treatment Figure C1 146 APPENDIX D Introduction It has not been determined whether MMP-9 activity is responsible for the decrease in cell surface CD11a found in DC cultures in response to atRA signaling through RARa. Materials and Methods For the atRA full-culture experiments, bone marrow cells were plated 2 x 106 per well containing 10 mL clMDM supplemented with10% characterized (CH)-FBS or charcoal dextran-filtered (CD)-FBS, 2 mmollL GlutaMax, 100 UIL penicillin, and 100 mg/L streptomycin, 20 ng/mL GM-CSF and 10 nmollL atRA or AGN 194301 and in the presence or absence of 10 nmollL atRA, 50 nmollL, 500 nmollL, or 5 pmol/L MMP-9 Inhibitor I, and equal DMSO concentrations. Fresh medium and treatment were given on d 3, 6, and 8. Cells were harvested on d 10 for FACS staining. AGN 194301 was generously provided by Dr. Chandraratna and blocks atRA activity by competing for the ligand binding domain of RARa (Nagpal and Chandraratna 2000). Floating and adherent cells were harvested, counted, washed with staining buffer (1% FBS, 0.1% (wlv) sodium azide, in phosphate buffered saline, pH 7.4), and incubated with monoclonal antibody against FcyR-II/lll from the 2.4G2 hybridoma to block non-specific binding. Each sample was then incubated with either an isotype control or a cell surface molecule-specific antibody cocktail, with 1 pg antibody used per 1 x 106 cells. Each sample was then incubated with 147 0.25 pg PE-Cy7-conjugated streptavidin per 1 x 106 cells. Monoclonal antibodies used were PE-conjugated hamster anti-mouse CD11c, PerCP-Cy5.5-conjugated rat anti-mouse Ly-6G (Gr-1), biotin-conjugated rat anti-mouse CD11a, PE- conjugated hamster lgG1/),, PerCP-Cy5.5-conjugated rat lngb/K, and biotin- conjugated rat lnga/K, along with PE-Cy7-conjugated streptavidin (BD Biosciences Pharrningen, San Jose, CA). Samples were run at the MSU flow cytometry facility using a LSR ll flow cytometer (BD Biosciences) and analyzed using FCS Express v.3.0 software (DeNovo Software, Ontario, CA). Results CD11a cell surface expression is increased on both DC alone and on the total cell population, floating and adherent, when cells are grown in CD-FBS containing medium that is vitamin A deficient. It was hypothesized that MMP-9 can cleave CD11a from the cell surface and inhibition of MMP-9 activity would lead to elevated CD11a cell surface expression in atRA treated cultures similar to that found in AGN treated cultures (Lackey et al. 2008). However, the cell surface expression of CD11a remains unchanged by culture treatment with increasing doses of MMP-9 inhibitor compared to atRA treated cultures (Figure D.1A-8). Thus, it is unlikely that MMP-9 activity is the mode by which atRA down-regulates CD11a surface expression. Discussion Increased MMP-9 activity through atRA signaling does not regulate the cell surface expression of CD1 1a in bone marrow-derived DC cultures because 148 . 3"“. ' I"?! I."lm_\.‘l-V inhibition of MMP—9 activity did notchange the cell surface expression. Because MMP-9 inhibition did significantly increase the percentage of cells that develop an adherent phenotype, it is likely that the presence of CD1 1a on the cell surface is not significantly involved in the change of adherence seen in atRA treated cells compared RARa antagonist treated cells. It is also likely that atRA must decrease CD11a expression through another mechanism, such as decreased rate of translation or increased rate of degradation. AtRA has been known to increase the rate of protein degradation through increased ubiquitination. In human bronchial epithelial cells, atRA increases the rate of cyclin D1 degradation through polyubiquitination of lysine residues (Feng et al. 2007), while atRA increase degradation of insulin receptor substrate-1 through increased ubiquitination in atRA-sensitive breast cancer cell lines through activation of PKG-6 (del Rincon et al. 2004). 149 Figure 0.1. CD11a cell surface expression after 10 d culture in the presence or absence of MMP-9 inhibitor. A. Floating and adherent CD11a cell surface expression on all cells population. 8. Floating and adherent CD11a cell surface expression on CD11c+ Gr-1' DC. MF I, mean fluorescence intensity, n=4 for each treatment group. All treatment groups contained 10 nmollL atRA in characterized (CH)-FBS-containing medium except charcoal—dextran filtered (CD)-FBS group. Results are presented as the mean :t SD and show one representative experiment of two. Treatment groups designated with different letters are significantly different with P <0.05 as determined by one-way ANOVA. 150 9%.. x7 n .I h e a 3.” rod FA Y E x .D. a y XH b, a1 21 C Yflillll pl 0? mo mu 2 m8 x 5 as. 330 AHIIL a1 tees so y XH 0.1“. pl e. so y , x._H|IHlI « coo pl c. s so so ZLI 00% C 9% 6» co YH ill a. _ ox. oi n_o o c 2 1 €25 00 co Es. «2.00 Treatment Figure D1 151 BIBLIOGRAPHY 152 BIBLIOGRAPHY Adams SL, Cohen AJ, Lassova L. Integration of signaling pathways regulating chondrocyte differentiation during endochondral bone formation. J Cell Physiol. 2007;213:635-41. Agrawal S, Anderson P, Durbeej M, van Rooijen N, Ivars F, Opdenakker G, Sorokin LM. Dystroglycan is selectively cleaved at the parenchymal basement membrane at sites of leukocyte extravasation in experimental autoimmune encephalomyelitis. J Exp Med. 2006;203:1007-19. Ahmad SM, Haskell MJ, Raqib R, Stephensen CB. Markers of innate immune hi function are associated with vitamin A stores in men. J Nutr. 2009;139:377-85. ,_ Ahmed F, Jones DB, Jackson AA. The interaction of vitamin A deficiency and rotavirus infection in the mouse. Br J Nutr. 1990;63:363-73. Ahmed F, Jones DB, Jackson AA. Effect of vitamin A deficiency on the immune I response to epizootic diarrhoea of infant mice (EDIM) rotavirus infection in mice. -' Br J Nutr. 1991 ;65:475-85. Amit-Romach E, Uni Z, Cheled S, Berkovich Z, Reifen R. Bacterial population and innate immunity-related genes in rat gastrointestinal tract are altered by vitamin A-deficient diet. J Nutr Biochem. 2009;20:70—7. Andela V, Rosier R. The proteosome inhibitor MG132 attenuates retinoic acid receptor trans-activation and enhances trans-repression of nuclear factor kappaB. Potential relevance to chemo-preventive interventions with retinoids. Mol Cancer. 2004;3z8. Apte SS. A disintegrin-like and metalloprotease (reprolysin-type) with thrombospondin type 1 motif (ADAMTS) superfamily: functions and mechanisms. J Biol Chem. 2009;284:31493-7. Austenaa LM, Carlsen H, Ertesvag A, Alexander G, Blomhoff HK, Blomhoff R. Vitamin A status significantly alters nuclear factor-kappaB activity assessed by in vivo imaging. FASEB J. 2004;18:1255-7. Balmer JE, Blomhoff R. Gene expression regulation by retinoic acid. J Lipid Res. 2002;43:1773-808. Banchereau J, Briere F, Caux C, Davoust J, Lebecque S, Liu Y-J, Pulendran B, Palucka K. lmmunobiology of dendritic cells. Annu Rev Immunol. 2000;18:767- 81 1. Banchereau J, Steinman RM. Dendritic cells and the control of immunity. Nature. 1 998;392:245-52. 153 Baratelli FE, Heuze-Vourc'h N, Krysan K, Dohadwala M, Riedl K, Sharma S, Dubinett SM. Prostaglandin E2-dependent enhancement of tissue inhibitors of metalloproteinases-1 production limits dendritic cell migration through extracellular matrix. J Immunol. 2004;173:5458-66. Barua AB, Furr HC. Properties of retinoids. Structure, handling, and preparation. Mol Biotechnol. 1998;10:167-82. Bastien J, Rochette-Egly C. Nuclear retinoid receptors and the transcription of retinoid-target genes. Gene. 2004;328:1-16. Biesalski HK, Nohr D. New aspects in vitamin A metabolism: the role of retinyl esters as systemic and local sources for retinol in mucous epithelia. J Nutr. 2004;134:34538—78. Blomhoff R, Blomhoff HK. Overview of retinoid metabolism and function. J .Neurobiol. 2006;66:606-30. Campo-Paysaa F, Marletaz F, Laudet V, Schubert M. Retinoic acid signaling in development: tissue-specific functions and evolutionary origins. Genesis. 2008;46:640-56. Cantoma MT, Nashold FE, Chun TY, Hayes CE. Vitamin A down-regulation of IFN-gamma synthesis in cloned mouse Th1 lymphocytes depends on the CD28 costimulatory pathway. J Immunol. 1996;156:2674-9. Cantoma MT, Nashold FE, Hayes CE. In vitamin A deficiency multiple mechanisms establish a regulatory T helper cell imbalance with excess Th1 and insufficient Th2 function. J Immunol. 1994;152:1515-22. Carman JA, Smith SM, Hayes CE. Characterization of a helper T lymphocyte defect in vitamin A-deficient mice. J Immunol. 1989;142:388-93. Chabot V, Reverdiau P, Iochmann S, Rico A, Senecal D, Goupille C, Sizaret PY, Sensebe L. CCL5—enhanced human immature dendritic cell migration through the basement membrane in vitro depends on matrix metalloproteinase-9. J Leukoc Biol. 2006;79:767-78. Chambaut-Guerin A-M, Herigault S, Rouet-Benzineb P, Rouher C, Lafuma C. Induction of matrix metalloproteinase MMP-9 (92-kDa gelatinase) by retinoic acid in human neuroblastoma SKNBE cells. Relevance to neuronal differentiation. J Neurochem. 2000;74:508-17. Chambon P. A decade of molecular biology of retinoic acid receptors. FASEB J. 1996;10:940-54. 154 Chang CJ, Hsu LA, Ko YH, Chen PL, Chuang YT, Lin CY, Liao CH, Pang JH. Thrombin regulates matrix metalloproteinase-9 expression in human monocytes. Biochem Biophys Res Commun. 2009;385:241-6. Chang F, Steelman LS, Lee JT, Shelton JG, Navolanic PM, Blalock WL, Franklin RA, McCubrey JA. Signal transduction mediated by the Ras/RafIMEK/ERK pathway from cytokine receptors to transcription factors: potential targeting for therapeutic intervention. Leukemia. 2003;17:1263-93. Chawla A, Repa JJ, Evans RM, Mangelsdorf DJ. Nuclear receptors and lipid physiology: opening the x-files. Science. 2001;294:1866-70. Chen Q, Ross AC. Inaugural Article: Vitamin A and immune function: Retinoic acid modulates population dynamics in antigen receptor and CD38-stimulated splenic B cells. Proc Nat Acad Sci USA. 2005;102:14142-9. Chen Q, Ross AC. Retinoic acid promotes mouse splenic B cell surface lgG expression and maturation stimulated by CD40 and IL-4. Cell Immunol. 2007;249:37-45. Chen Q, Ross AC. Retinoic acid regulates CD1d gene expression at the transcriptional level in human and rodent monocytic cells. Exp Biol Med (Maywood). 2007;232:488-94. Chen X, Esplin BL, Garrett KP, Welner RS, Webb CF, Kincade PW. Retinoids accelerate B lineage lymphoid differentiation. J Immunol. 2008;180:138—45. Chen YJ, Chang WM, Liu YW, Lee CY, Jang YH, Kuo CD, Liao HF. A small- molecule metastasis inhibitor, norcantharidin, downregulates matrix metalloproteinase-9 expression by inhibiting Sp1 transcriptional activity in colorectal cancer cells. Chem Biol Interact. 2009;181:440-6. Chirco R, Liu XW, Jung KK, Kim HR. Novel functions of TIMPs in cell signaling. Cancer Metastasis Rev. 2006;25:99-113. Christian P, West KP, Jr., Khatry SK, LeClerq SC, Kimbrough-Pradhan E, Katz J, Shrestha SR. Maternal night blindness increases risk of mortality in the first 6 months of life among infants in Nepal. J Nutr. 2001;131:1510-2. CIagett-Dame M, McNeill EM, Muley PD. Role of all-trans retinoic acid in neurite outgrowth and axonal elongation. J Neurobiol. 2006;66:739-56. Congdon NG, West KP, Jr. Physiologic indicators of vitamin A status. J Nutr. 2002;132:28893—948. Darrnanin S, Chen J, Zhao S, Cui H, Shirkoohi R, Kubo N, Kuge Y, Tamaki N, Nakagawa K, et al. All-trans retinoic acid enhances murine dendritic cell 155 migration to draining lymph nodes via the balance of matrix metalloproteinases and their inhibitors. J Immunol. 2007;179:4616—25. Dawson HD, Collins G, Pyle R, Key M, Taub DD. The retinoic acid receptor- alpha mediates human T-cell activation and Th2 cytokine and chemokine production. BMC Immunol. 2008;9216—30. Dawson HD, Li N-Q, DeCicco KL, Nibert JA, Ross AC. Chronic marginal vitamin A status reduces natural killer cell number and function in aging Lewis rats. J Nutr. 1999;129:1510—7. de Jong EC, Smits HH, Kapsenberg ML. Dendritic cell-mediated T cell polarization. Springer Semin lmmunopathol. 2005;26:289-307. del Hoyo GM, Martin P, Vargas HH, Ruiz S, Arias CF, Ardavin C. Characterization of a common precursor population for dendritic cells. Nature. - 2002;415:1043-7. del Rincon SV, Guo Q, Morelli C, Shiu HY, Surrnacz E, Miller WH. Retinoic acid mediates degradation of IRS-1 by the ubiquitin-proteasome pathway, via a PKC- dependant mechanism. Oncogene. 2004;23:9269-79. Denning TL, Wang YC, Patel SR, Williams IR, Pulendran B. Lamina propria macrophages and dendritic cells differentially induce regulatory and interleukin 17-producing T cell responses. Nat Immunol. 2007;8z1086-94. Duester G. Retinoic acid regulation of the somitogenesis clock. Birth Defects Res C Embryo Today. 2007;81:84-92. Elangbam CS, Qualls CW, Jr., Dahlgren RR. Cell adhesion molecules—update. Vet Pathol. 1997;34:61-73. Evans T. Regulation of hematopoiesis by retinoid signaling. Exp Hematol. 2005;33:1055-61. Farina AR, Masciulli M-P, Tacconelli A, Cappabianca L, De Santis G, Gulino A, Mackay AR. All-trans-retinoic acid induces nuclear factor {kappa}B activation and matrix metalloproteinase-9 expression and enhances basement membrane invasivity of differentiation-resistant human SK-N-BE 9N neuroblastoma cells. Cell Growth Differ. 2002;13:343-54. Farina AR, Tacconelli A, Vacca A, Maroder M, Gulino A, Mackay AR. Transcriptional up-regulation of matrix metalloproteinase-9 expression during spontaneous epithelial to neurOblast phenotype conversion by SK—N-SH neuroblastoma cells, involved in enhanced invasivity, depends upon GT-box and nuclear factor kappaB elements. Cell Growth Differ. 1999;10:353—67. 156 Faure-Andre G, Vargas P, Yuseff Ml, Heuze M, Diaz J, Lankar D, Steri V, Manry J, Hugues S, et al. Regulation of dendritic cell migration by CD74, the MHC class ll-associated invariant chain. Science. 2008;322:1705-10. Faveeuw C, Preece G, Ager A. Transendothelial migration of lymphocytes across high endothelial venules into lymph nodes is affected by metalloproteinases. Blood. 2001;98:688-95. Fawzi WW, Chalmers TC, Herrera MG, Mosteller F. Vitamin A supplementation and child mortality. A meta-analysis. JAMA. 1993;269:898-903. Feng Q, Sekula D, Muller R, Freemantle SJ, Dmitrovsky E. Uncovering residues that regulate cyclin D1 proteasomal degradation. Oncogene. 2007;26:5098—106. Fogg DK, Sibon C, Miled C, Jung S, Aucouturier P, Littman DR, Cumano A, Geissmann F. A clonogenic bone marrow progenitor specific for macrophages and dendritic cells. Science. 2006;311:83-7. F rankenberger M, Hauck RW, Frankenberger B, Haussinger K, Maier KL, Heyder J, Ziegler-Heitbrock HW. All trans-retinoic acid selectively down-regulates matrix metalloproteinase-9 (MMP-9) and up-regulates tissue inhibitor of metalloproteinase-1 (TIMP-1) in human bronchoalveolar lavage cells. Mol Med. 2001;72263—70. Fridman R, Toth M, Chvyrkova I, Meroueh SO, Mobashery S. Cell surface association of matrix metalloproteinase-9 (gelatinase 8). Cancer Metastasis Rev. 2003;22:153-66. Fujiki R, Chikanishi T, Hashiba W, Ito H, Takada I, Roeder RG, Kitagawa H, Kato S. GlcNAcylation of a histone methyltransferase in retinoic-acid-induced granulopoiesis. Nature. 2009;459:455—9. Gibson RS. The role of diet- and host-related factors in nutrient bioavailability and thus in nutrient-based dietary requirement estimates. Food Nutr Bull. 2007;28:877-8100. Gonzalez-Juarrero M, Shim TS, Kipnis A, Junqueira-Kipnis AP, Orrne lM. Dynamics of macrophage cell populations during murine pulmonary tuberculosis. J Immunol. 2003;171:3128-35. Green HN, Mellanby E. Vitamin A as an anti-infective agent. Brit Med J. 1928;3537z691. Hammerschmidt SI, Ahrendt M, Bode U, Wahl B, Kremmer E, Forster R, Pabst O. Stromal mesenteric lymph node cells are essential for the generation of gut— homing T cells in vivo. J Exp Med. 2008;205:2483—90. 157 Harrison EH. Mechanisms of digestion and absorption of dietary vitamin A. Annu Rev Nutr. 2005;25:87-103. Heath WR, Belz GT, Behrens GM, Smith CM, Forehan SP, Parish IA, Davey GM, Wilson NS, Carbone FR, Villadangos JA. Cross-presentation, dendritic cell subsets, and the generation of immunity to cellular antigens. Immunol Rev. 2004;199:9-26. Hengesbach LM, Hoag KA. Physiological concentrations of retinoic acid favor myeloid dendritic cell development over granulocyte development in cultures of bone marrow cells from mice. J Nutr. 2004;134:2653—9. Heyman RA, Mangelsdorf DJ, Dyck JA, Stein RB, Eichele G, Evans RM, Thaller C. 9-cis retinoic acid is a high affinity ligand for the retinoid X receptor. Cell. 1992;68:397-406. Hill JA, Hall JA, Sun CM, Cai Q, Ghyselinck N, Chambon P, Belkaid Y, Mathis D, Benoist C. Retinoic acid enhances Foxp3 induction indirectly by relieving inhibition from CD4+CD44hi cells. Immunity. 2008;29:758-70. Hiraguri M, Miike S, Sano H, Kurasawa K, Saito Y, lwamoto l. Granulocyte- macrophage colony-stimulating factor and lL-5 activate mitogen-activated protein kinase through Jak2 kinase and phosphatidylinositol 3-kinase in human eosinophils. J Allergy Clin Immunol. 1997;100:845-51. Ho TY, Yan W, Bagnell CA. Relaxin-induced matrix metalloproteinase-9 expression is associated with activation of the NF-kappaB pathway in human THP-1 cells. J Leukoc Biol. 2007;81:1303-10. Hoag KA, Nashold FE, Goverman J, Hayes CE. Retinoic acid enhances the T helper 2 cell development that is essential for robust antibody responses through its action on antigen-presenting cells. J Nutr. 2002;132:3736-9. Hochrein H, O’Keeffe M. Dendritic cell subsets and toll-like receptors. Handb Exp Pharrnacol. 2008:153—79. Hu Y, Ivashkiv LB. Costimulation of chemokine receptor signaling by matrix metalloproteinase-9 mediates enhanced migration of IF N-alpha dendritic cells. J Immunol. 2006;176:6022-33. Huhtala P, Tuuttila A, Chow LT, Lohi J, Keski—Oja J, Tryggvason K. Complete structure of the human gene for 92-kDa type IV collagenase. Divergent regulation of expression for the 92- and 72-kilodalton enzyme genes in HT-1080 cells. J Biol Chem. 1991;266:16485-90. Hwang MK, Song NR, Kang NJ, Lee KW, Lee HJ. Activation of phosphatidylinositol 3-kinase is required for tumor necrosis factor-alpha-induced 158 upregulation of matrix metalloproteinase-9: its direct inhibition by quercetin. Int J Biochem Cell Biol. 2009, 41: 1592-600. Iacono KT, Brown AL, Greene Ml, Saouaf SJ. CD147 immunoglobulin superfamily receptor function and role in pathology. Exp Mol Pathol. 2007;83:283-95. Ichiyasu H, McCormack JM, McCarthy KM, Dombkowski D, Preffer F I, Schneeberger EE. Matrix metaIloproteinase-9-deficient dendritic cells have impaired migration through tracheal epithelial tight junctions. Am J Respir Cell Mol Biol. 2004;30:761-70. lnaba K, Swiggard WJ, Steinman RM, Romani N, Schuler G. Isolation of dendritic cells. In: Coligan JE, Kruisbeek AM, Margulies DH, Shevach EM, Strober W, editors. Current Protocols in Immunology. Hoboken, NJ: John Wiley & Sons; 1998. p. 3.7.1.-3.7.15 IOM. Dietary reference intakes for vitamin A, vitamin K, arsenic, boron, chromium, copper, iodine, iron, manganese, molybdenum, nickel. silicon, vanadium, and zinc. Washington, DC: Institute of Medicine; 2001. Iwasaki A. Mucosal dendritic cells. Annu Rev Immunol. 2007;25:381-418. Iwata M, Hirakiyama A, Eshima Y, Kagechika H, Kato C, Song S-Y. Retinoic acid imprints gut-homing specificity on T cells. Immunity. 2004;21:527-38. Jason J, Archibald LK, Nwanyanwu OC, Sowell AL, Buchanan l, Lamed J, Bell M, Kazembe PN, Dobbie H, Jarvis WR. Vitamin A levels and immunity in humans. Clin Diag Lab Immunol. 2002;92616-21. Jenuwein T, Allis CD. Translating the histone code. Science. 2001;293:1074-80. Kawaguchi R, Yu J, Honda J, Hu J, Whitelegge J, Ping P, Wiita P, Bok D, Sun H. A membrane receptor for retinol binding protein mediates cellular uptake of vitamin A. Science. 2007;315:820-5. Kawaguchi R, Yu J, Wiita P, Honda J, Sun H. An essential ligand-binding domain in the membrane receptor for retinoI-binding protein revealed by large-scale mutagenesis and a human polymorphism. J Biol Chem. 2008;283:15160—8. Kim S, Choi JH, Lim HI, Lee SK, Kim WW, Cho S, Kim JS, Kim JH, Choe JH, et al. EGF-induced MMP-9 expression is mediated by the JAK3/ERK pathway, but not by the JAK3/STAT-3 pathway in a SKBR3 breast cancer cell line. Cell Signal. 2009;21 :892-8. Kuo PT, Lin TP, Liu LC, Huang CH, Lin JK, Kao JY, Way TD. Penta-O-galloyl- beta-D-glucose suppresses prostate cancer bone metastasis by transcriptionally 159 repressing EGF-induced MMP-9 expression. J Agric Food Chem. 2009;57:3331- 9. Kurlandsky SB, Gamble MV, Ramakrishnan R, Blaner WS. Plasma delivery of retinoic acid to tissues in the rat. J Biol Chem. 1995;270:17850-7. Kuwata T, Wang IM, Tamura T, Ponnamperuma RM, Levine R, Holmes KL, Morse HC, Ill, De Luca LM, Ozato K. Vitamin A deficiency in mice causes a systemic expansion of myeloid cells. Blood. 2000;95:3349-56. Lackey DE, Ashley SL, Davis AL, Hoag KA. Retinoic acid decreases adherence of murine myeloid dendritic cells and increases production of matrix metalloproteinase-9. J Nutr. 2008;138:1512-9. Lateef H, Stevens MJ, Varani J. All-trans-retinoic acid suppresses matrix metalloproteinase activity and increases collagen synthesis in diabetic human skin in organ culture. Am J Pathol. 2004;165:167-74. Lee YS, Lan Tran HT, Van Ta Q. Regulation of expression of matrix metalloproteinase-9 by JNK in Raw 264.7 cells: presence of inhibitory factor(s) suppressing MMP-9 induction in serum and conditioned media. Exp Mol Med. 2009;41:259-68. Lin J, Song F, Yao P, Yang X, Li N, Sun S, Lei L, Liu L. Effect of vitamin A supplementation on immune function of well-nourished children suffering from vitamin A deficiency in China. Eur J Clin Nutr. 2007. Liu K, Waskow C, Liu X, Yao K, Hoh J, Nussenzweig M. Origin of dendritic cells in peripheral lymphoid organs of mice. Nat Immunol. 2007;82578-83. Lomberk G, Urrutia R. The family feud: turning off Sp1 by Sp1-like KLF proteins. Biochem J. 2005;392:1-11. Lutz MB, Kukutsch N, Ogilvie ALJ, Rossner S, Koch F, Romani N, Schuler G. An advanced culture method for generating large quantities of highly pure dendritic cells from mouse bone marrow. J Immunol Methods. 1999;223:77-92. Maciel AA, Oria RB, Braga-Neto MB, Braga AB, Carvalho EB, Lucena HB, Brito GA, Guerrant RL, Lima AA. Role of retinol in protecting epithelial cell damage induced by Clostridium difficile toxin A. Toxicon. 2007;50:1027-40. Manicassamy S, Ravindran R, Deng J, Oluoch H, Denning TL, Kasturi SP, Rosenthal KM, Evavold BD, Pulendran B. Toll-like receptor 2-dependent induction of vitamin A-metabolizing enzymes in dendritic cells promotes T regulatory responses and inhibits autoimmunity. Nat Med. 2009;15:401-9. Mao JT, Tashkin DP, Belloni PN, Baileyhealy l, Baratelli F, Roth MD. All-trans retinoic acid modulates the balance of matrix metalloproteinase-9 and tissue 160 inhibitor of metalloproteinase-1 in patients with emphysema. Chest. 2003;124:1724-32. Mark M, Ghyselinck NB, Chambon P. Function of retinoid nuclear receptors: lessons from genetic and pharmacological dissections of the retinoic acid signaling pathway during mouse embryogenesis. Annu Rev Phannacol Toxicol. 2006;46:451-80. Martin P, del Hoyo GM, Anjuere F, Ruiz SR, Arias CF, Marin AR, Ardavin C. Concept of lymphoid versus myeloid dendritic cell lineages revisited: both CD83Ipha(-) and CD8alpha(+) dendritic cells are generated from CD4(low) lymphoid-committed precursors. Blood. 2000;96:2511-9. Maynard CL, Hatton RD, Helms WS, Oliver JR, Stephensen CB, Weaver CT. Contrasting roles for all-trans retinoic acid in TGF-beta-mediated induction of Foxp3 and "10 genes in developing regulatory T cells. J Exp Med. 2009;206:343- 57. McCullum EV, Davis M. The necessity of certain lipins in the diet during growth. J Biol Chem. 1913:167-75. McLaren DS. Dietary vitamin A requirements. In: Blomhoff R, editor. Vitamin A in health and disease. New York: Marcel Dekker, Inc.; 1994. p. 665-70. Montesano R, Soulie P. Retinoids induce lumen morphogenesis in mammary epithelial cells. J Cell Sci. 2002;115:4419-31. Mora JR, Iwata M, Eksteen B, Song SY, Junt T, Senman B, Otipoby KL, Yokota A, Takeuchi H, et al. Generation of gut—homing lgA-secreting B cells by intestinal dendritic cells. Science. 2006;314:1 157-60. Moser M, Murphy KM. Dendritic cell regulation of TH1-TH2 development. Nat Immunol. 2000;1:199-205. Mucida D, Park Y, Cheroutre H. From the diet to the nucleus: vitamin A and TGF- beta join efforts at the mucosal interface of the intestine. Semin Immunol. 2009;21:14-21. Mucida D, Park Y, Kim G, Turovskaya 0, Scott I, Kronenberg M, Cheroutre H. Reciprocal TH17 and regulatory T cell differentiation mediated by retinoic acid. Science. 2007;317:256-60. Munaut C, Salonunni T, Kontusaari S, Reponen P, Morita T, Foidart JM, Tryggvason K. Murine matrix metalloproteinase 9 gene. 5'-upstream region contains cis-acting elements for expression in osteoclasts and migrating keratinocytes in transgenic mice. J Biol Chem. 1999;274:5588-96. 161 Nagpal S, Chandraratna RAS. Recent developments in receptor-selective retinoids. Curr Phann Des. 2000;6z919-31. Napoli JL. Interactions of retinoid binding proteins and enzymes in retinoid metabolism. Biochim Biophys Acta. 1999;14402139-62. Newton S, Owusu-Agyei S, Ampofo W, Zandoh C, Adjuik M, Adjei G, Tchum S, Filteau S, Kirkwood BR. Vitamin A supplementation enhances infants’ immune responses to hepatitis B vaccine but does not affect responses to Haemophilus influenzae type b vaccine. J Nutr. 2007;137:1272-7. Noy N. Physical-chemical properties and action of retinoids. In: Nau H, Blaner WS, editors. Retinoids: the biochemical and molecular basis of vitamin A and retinoid action. Heidelberg, Germany: Springer-Vedag; 1999. p. 1-29. O'Byme SM, Wongsiriroj N, Libien J, Vogel S, Goldberg IJ, Baehr W, Palczewski K, Blaner WS. Retinoid absorption and storage is impaired in mice lacking lecithinzretinol acyltransferase (LRAT). J Biol Chem. 2005;280:35647-57. Ochieng J, Green 8, Evans S, James O, Warfield P. Modulation of the biological functions of galectin-3 by matrix metalloproteinases. Biochim Biophys Acta. 1998;1379:97-106. Olefsky JM. Nuclear receptor minireview series. J Biol Chem. 2001;276:36863-4. Opdenakker G, Van den Steen PE, Van Damme J. Gelatinase B: a tuner and amplifier of immune functions. Trends Immunol. 2001;22:571-9. Osborne TB, Mendel L8. The relation of growth to the chemical constituents of the diet. J Biol Chem. 1913:311-26. Osman M, Tortorella M, Londei M, Quaratino S. Expression of matrix metalloproteinases and tissue inhibitors of metalloproteinases define the migratory characteristics of human monocyte-derived dendritic cells. Immunology. 2002;105:73-82. Payne LG, Koski KG, Ortega-Barria E, Scott ME. Benefit of vitamin A supplementation on ascaris reinfection is less evident in stunted children. J Nutr. 2007; 1 37: 1455-9. Piedrafita FJ, Pfahl M. Nuclear retinoid receptors and mechanisms of action. In: Nau H, Blaner WS, editors. Retinoids: the biochemical and molecular basis of vitamin A and retinoid action. Heidelberg, Germany: Springer-Verlag; 1999. p. 1 53-84. Pozzi A, Moberg PE, Miles LA, Wagner S, Soloway P, Gardner HA. Elevated matrix metalloprotease and angiostatin levels in integrin alpha 1 knockout mice cause reduced tumor vascularization. Proc Natl Acad Sci U S A. 2000;97:2202-7. 162 Pruijt JFM, Fibbe WE, Laterveer L, Pieters RA, Lindley IJD, Paemen L, Masure S, Willemze R, Opdenakker G. Prevention of interleukin-8-induced mobilization of hematopoietic progenitor cells in rhesus monkeys by inhibitory antibodies against the Metalloproteinase gelatinase B (MMP-9). Proc Nat Acad Sci USA. 1999;96:10863-8. Purton LE, Dworkin S, Olsen GH, Walkley CR, Fabb SA, Collins SJ, Chambon P. RARgamma is critical for maintaining a balance between hematopoietic stem cell self-renewal and differentiation. J Exp Med. 2006;203:1283-93. Pyszniak AM, Welder CA, Takei F. Cell surface distribution of high-avidity LFA-1 detected by soluble lCAM-1-coated microspheres. J Immunol. 1994;152:5241-9. Quadro L, Gamble MV, Vogel S, Lima AA, Piantedosi R, Moore SR, Colantuoni V, Gottesman ME, Guerrant RL, Blaner WS. Retinol and retinol-binding protein: gut integrity and circulating immunoglobulins. J Infect Dis. 2000;182:897-8102. Quadro L, Hamberger L, Colantuoni V, Gottesman ME, Blaner WS. Understanding the physiological role of retinoI-binding protein in vitamin A metabolism using transgenic and knockout mouse models. Mol Aspects Med. 2003;24:421-30. Radjabi AR, Sawada K, Jagadeeswaran S, Eichbichler A, Kenny HA, Montag A, Bruno K, Lengyel E. Thrombin induces tumor invasion through the induction and association of matrix metalloproteinase-9 and beta1-integrin on the cell surface. J Biol Chem. 2008;283:2822-34. Randolph GJ, Angeli V, Swartz MA. Dendritic-cell trafficking to lymph nodes through lymphatic vessels. Nat Rev Immunol. 2005;52617-28. Rastogi T, Mathers C. Global burden of vitamin A deficiency in the year 2000. Geneva: World Health Organization; 2002. Renckens R, Roelofs JJ, Florquin S, de Vos AF, Lijnen HR, van't Veer C, van der Poll T. Matrix metalloproteinase-9 deficiency impairs host defense against abdominal sepsis. J Immunol. 2006;176:3735-41. Ross AC. Vitamin A and retinoids. In: Shils ME, Olson JA, Shike M, Ross AC, editors. Modern nutrition in health and disease. 9th ed. New York: Lippincott VVlIliams & Wilkins; 1999. p. 305-27. Ross AC. Retinoid production and catabolism: role of diet in regulating retinol esterification and retinoic acid oxidation. J Nutr. 2003;133:2918-68. Ross AC, Hammerling UG. Retinoids and the immune system. In: Sporn MB, Roberts AB, Goodman DS, editors. The retinoids: biology, chemistry, and medicine. 2nd ed. New York: Raven Press, Ltd.; 1994. p. 521-43. 163 Schonbeck U, Mach F, Libby P. Generation of biologically active lL—1{beta} by matrix metalloproteinases: a novel caspase-1-independent pathway of IL-1{beta} processing. J Immunol. 1998;161:3340-6. Schuster GU, Kenyon NJ, Stephensen CB. Vitamin A deficiency decreases and high dietary vitamin A increases disease severity in the mouse model of asthma. J Immunol. 2008;180:1834-42. Semba RD. Vitamin A and immunity to viral, bacterial and protozoan infections. Proc Nutr Soc. 1999;58:719-27. Shimada J, Suzuki Y, Kim S-J, Wang P-C, Matsumura M, Kojima S. Transactivation via RAR/RXR-Sp1 interaction: characterization of binding between Sp1 and GC box motif. Mol Endocrinol. 2001;15:1677-92. Shortrnan K, Liu Y-J. Mouse and human dendritic cell subtypes. Nat Rev Immunol. 2002;2:151-61. Siewert C, Menning A, Dudda J, Siegmund K, Lauer U, Floess S, Campbell DJ, Hamann A, Huehn J. Induction of organ-selective CD4+ regulatory T cell homing. Eur J Immunol. 2007;37:978-89. Smith SM, Levy NS, Hayes CE. Impaired immunity in vitamin A-deficient mice. J Nutr. 1987;117:857-65. Sommer A, Tarwotjo I, Hussaini G, Susanto D. Increased mortality in children with mild vitamin A deficiency. Lancet. 1983;21585-8. Steinman RM. Dendritic cells. In: Paul WE, editor. Fundamental immunology. 4th ed. Philadelphia, PA: Lippincott-Raven; 1999. p. 547-73. Stephensen CB. Vitamin A, infection, and immune function. Annu Rev Nutr. 2001;21:167-92. Stephensen CB, Jiang X, F reytag T. Vitamin A deficiency increases the in vivo development of IL-10-positive Th2 cells and decreases development of Th1 cells in mice. J Nutr. 2004;134:2660-6. Stephensen CB, Moldoveanu Z, Gangopadhyay NN. Vitamin A deficiency diminishes the salivary immunoglobulin A response and enhances the serum immunoglobulin G response to influenza A virus infection in BALB/c mice. J Nutr. 1996;126:94-102. Suzuki Y, Shimada J, Shudo K, Matsumura M, Crippa MP, Kojima S. Physical interaction between retinoic acid receptor and Sp1: mechanism for induction of urokinase by retinoic acid. Blood. 1999;93:4264-76. 164 Taheri F, Bazan HE. Platelet-activating factor overturns the transcriptional repressor disposition of Sp1 in the expression of MMP-9 in human corneal epithelial cells. Invest Ophthalmol Vis Sci. 2007;48:1931-41. Takahra T, Smart DE, Oakley F, Mann DA. Induction of myofibroblast MMP-9 transcription in three-dimensional collagen I gel cultures: regulation by NF- kappaB, AP-1 and Sp1. Int J Biochem Cell Biol. 2004;36:353-63. Traver D, Akashi K, Manz M, Merad M, Miyamoto T, Engleman EG, Weissman IL. Development of CD8alpha-positive dendritic cells from a common myeloid progenitor. Science. 2000;290:2152-4. Tsai LN, Ku TK, Salib NK, Crowe DL. Extracellular signals regulate rapid coactivator recruitment at AP-1 sites by altered phosphorylation of both CREB binding protein and c-jun. Mol Cell Biol. 2008;28:4240—50. Tsang KJ, Crowe DL. Retinoic acid and extracellular matrix inhibition of matrix metalloproteinase 9 expression is mediated by the mitogen activated protein kinase pathway. Int J Oncol. 2001;18:369-74. Twining SS, Schulte DP, Wilson PM, Fish BL, Moulder JE. Retinol is sequestered in the bone marrow of vitamin A-deficient rats. J Nutr. 1996;126:1618-26. Twining SS, Schulte DP, Wilson PM, Fish BL, Moulder JE. Vitamin A deficiency alters rat neutrophil function. J Nutr. 1997;127:558-65. Ueki S, Mahemuti G, Oyamada H, Kato H, Kihara J, Tanabe M, Ito W, Chiba T, Takeda M, et al. Retinoic acids are potent inhibitors of spontaneous human eosinophil apoptosis. J Immunol. 2008;181 17689-98. van Bennekum AM, Wei S, Gamble MV, Vogel S, Piantedosi R, Gottesman M, Episkopou V, Blaner WS. Biochemical basis for depressed serum retinol levels in transthyretin-deficient mice. J Biol Chem. 2001;276:1107-13. Varani J, Perone P, Merfert MG, Moon SE, Larkin D, Stevens MJ. All-trans retinoic acid improves structure and function of diabetic rat skin in organ culture. Diabetes. 2002;51:3510-6. Verrnaelen KY, Cataldo D, Toumoy K, Tania M, Dhulst A, Louis R, Foidart J-M, Noel A, Pauwels R. Matrix metalloproteinase-9-mediated dendritic cell recruitment into the airways is a critical step in a mouse model of asthma. J Immunol. 2003;171:10116—1022. Verstappen J, Von den Hoff JW. Tissue inhibitors of metalloproteinases (T lMPs): their biological functions and involvement in oral disease. J Dent Res. 2006;85:1074-84. 165 Villamor E, Fawzi WW. Vitamin A supplementation: implications for morbidity and mortality in children. J Infect Dis. 2000;182:8122-833. Villamor E, Fawzi WW. Effects of vitamin A supplementation on immune responses and correlation with clinical outcomes. Clin Microbiol Rev. 2005;18:446-64. Vogel S, Gamble MV, Blaner W8. Biosynthesis, absorption, metabolism and transport of retinoids. In: Nau H, Blaner W8, editors. Retinoids: the biochemical and molecular basis of vitamin A and retinoid action. Heidelberg, Germany: Springer-Verlag; 1999. p. 31-95. von Lintig J, Vogt K. Vitamin A formation in animals: molecular identification and functional characterization of carotene cleaving enzymes. J Nutr. 2004;134:2518-68. Vu TH, Werb 2. Matrix metalloproteinases: effectors of development and normal physiology. Genes Dev. 2000;14:2123-33. Wada Y, Hisamatsu T, Kamada N, Okamoto S, Hibi T. Retinoic acid contributes to the induction of lL-12-hypoproducing dendritic cells. lnflamm Bowel Dis. 2009;15:1548-56. Walkley CR, Yuan Y-D, Chandraratna RAS. McArthur GA. Retinoic acid receptor antagonism in vivo expands the numbers of precursor cells during granulopoiesis. Leukemia. 2002;16:1763-72. West KP, Jr. Extent of vitamin A deficiency among preschool children and women of reproductive age. J Nutr. 2002;132:28578-668. West KP, Jr. Vitamin A deficiency disorders in children and women. Food Nutr Bull. 2003;24:878-890. West KP, Jr. Vitamin A deficiency as a preventable cause of maternal mortality in undernourished societies: plausibility and next steps. Int J Gynaecol Obstet. 2004;85 Suppl 11824-7. Weston AD, Blumberg B, Underhill TM. Active repression by unliganded retinoid receptors in development: less is sometimes more. J Cell Biol. 2003;161:223-8. Weston AD, Hoffman LM, Underhill TM. Revisiting the role of retinoid signaling in skeletal development. Birth Defects Res C Embryo Today. 2003;69:156-73. WHO. Distribution of vitamin A during national immunization days. Geneva: World Health Organization; 1999. 166 WHO. Global prevalence of vitamin A deficiency in populations at risk 1995- 2005: WHO Global Database on Vitamin A Deficiency. Geneva: World Health Organization; 2009. VWlson HL, O’Neill HC. Murine dendritic cell development: difficulties associated with subset analysis. Immunol Cell Biol. 2003;81:239-46. Wolf G. The discovery of the visual function of vitamin A. J Nutr. 2001;131:1647- 50. Wolf G. ls 9-cis-retinoic acid the endogenous ligand for the retinoic acid-X receptor? Nutr Rev. 2006;64:532-8. Woodhead VE, Binks MH, Chain BM, Katz DR. From sentinel to messenger: an extended phenotypic analysis of the monocyte to dendritic cell transition. Immunology. 1998;94:552-9. Xu J, Rogers MB. Modulation of bone morphogenetic protein (BMP) 2 gene expression by Sp1 transcription factors. Gene. 2007;392:221-9. Yamazaki 8, Dudziak D, Heidkamp GF, Fiorese C, Bonito AJ, Inaba K, Nussenzweig MC, Steinman RM. CD8+ CD205+ splenic dendritic cells are specialized to induce Foxp3+ regulatory T cells. J Immunol. 2008;181:6923-33. Yan C, Boyd DD. Histone H3 acetylation and H3 K4 methylation define distinct chromatin regions permissive for transgene expression. Mol Cell Biol. 2006;26:6357-71 . Yan C, Wang H, Boyd DD. KISS-1 represses 92-kDa type IV collagenase expression by down-regulating NF-kappa B binding to the promoter as a consequence of Ikappa Balpha -induced block of p65/p50 nuclear translocation. J Biol Chem. 2001;276:1164-72. Yu Q, Stamenkovic l. Cell surface-localized matrix metalloproteinase-9 proteolytically activates TGF-beta and promotes tumor invasion and angiogenesis. Genes Dev. 2000;14:163-76. Yu S, Xia M, Xu W, Chu Y, Wang Y, Xiong S. All-trans retinoic acid biases immune response induced by DNA vaccine in a Th2 direction. Vaccine. 2005;23:5160-7. Zaragoza R, Gimeno A, Miralles VJ, Garcia-Trevijano ER, Carmena R, Garcia C, Mata M, Puertes IR, Torres L, Vina JR. Retinoids induce MMP-9 expression through RARalpha during mammary gland remodeling. Am J Physiol Endocrinol Metab. 2007;292:E1140—E8. 167 Zhou L, Yan C, Gieling RG, Kida Y, Garner W, Li W, Han YP. Tumor necrosis factor-alpha induced expression of matrix metalloproteinase-9 through p21- activated kinase-1. BMC Immunol. 2009;10:15. Zile MH. Vitamin A requirement for early cardiovascular morphogenesis specification in the vertebrate embryo: insights from the avian embryo. Exp Biol Med (Maywood). 2004;229:598-606. Zolfaghari R, Cifelli CJ, Lieu 80, Chen Q, Li NQ, Ross AC. Lipopolysaccharide opposes the induction of CYP26A1 and CYP26B1 gene expression by retinoic acid in the rat liver in vivo. Am J Physiol Gastrointest Liver Physiol. 2007;292:G1029-G36. 168 oil 571 lllllllllllllllllllllllllllllll ll 3 6 o 3 0 3 9 2 1 3 llllllllllllllllll