Characterization of Pactolus: transcription and function

Pactolus is a transmembrane protein found mainly on maturing to mature neutrophils and is homologous to ? integrin subunits. The majority of Pactolus is sequestered inside the cell until the neutrophil is activated or undergoes apoptosis. The gene structure of pactolus was determined in order to create a Pactolus deficient mouse. At the start of this thesis, Pactolus function was unknown, and the Pactolus deficient mouse proved to be instrumental in elucidating Pactolus function. Pactolus was not detected to have any intracellular signaling or in vivo functions, but in an in vitro assay mixing wild type or Pactolus deficient neutrophils with macrophages, Pactolus function was discovered. Pactolus appears to specifically interact with a macrophage receptor to facilitate apoptotic neutrophil phagocytosis by the macrophage and to downregulate macrophage TNF-? production. In a second project, to elucidate the tissue specific transcriptional control of the pactolus gene, the pactolus promoter was analyzed utilizing EMSA protocols. These data suggest a role for PU.1 and NF-k;B in pactolus gene regulation.

THE CHARACTERIZATION OF PACTOLUS: TRANSCRIPTION AND FUNCTION by Rebecca Lynn Margraf A dissertation submitted to the faculty of The University of Utah in partial fulfillment of the requirements of the degree of Doctor of Philosophy in Experimental Pathology Department of Pathology The University of Utah May 2003 Copyright © Rebecca Lynn Margraf 2003 All Rights Reserved THE UNIVERSITY OF UTAH GRADUATE SCHOOL SUPERVISORY COMMITTEE APPROVAL of a dissertation submitted by Rebecca L. Margraf This dissertation has been read by each member he following supervisory committee and by majority vote has been found to be sati,sfictor . / /' J \ -2C - 02- ) ) - ~ OL ~ Robert S. FUJinami r Il-2~-02 II-Z-G-Oz- THE UNIVERSITY OF UTAH GRADUATE SCHOOL FINAL READING APPROVAL To the Graduate Council of the Uni versity of Utah: I have read the dissertation of Rebecca L. Margraf in its final foml and have found that (1) its format, citations, and bibliographic style are consistent and acceptable; (2) its illustrative materials including figures, tables, and charts are in place; and (3) the final manuscript is satisfactory to the supervisory committee and is ready for submission to The Graduate School. Date ( ( Approved for the Major Department Carl R. Kjeldsberg Chair Approved for the Graduate Council David S. Chapman Dean of The Graduate Scho I -~----- ... ABSTRACT Pactolus is a transmembrane protein found mainly on maturing to mature neutrophils and is homologous to B integrin subunits. The majority of Pactolus is sequestered inside the cell until the neutrophil is activated or undergoes apoptosis. The gene structure of pactolus was determined in order to create a Pactolus deficient mouse. At the start of this thesis, Pactolus function was unknown, and the Pactolus deficient mouse proved to be instrumental in elucidating Pactolus function. Pactolus was not detected to have any intracellular signaling or in vivo functions, but in an in vitro assay mixing wild type or Pactolus deficient neutrophils with macrophages, Pactolus function was discovered. Pactolus appears to specifically interact with a macrophage receptor to facilitate apoptotic neutrophil phagocytosis by the macrophage and to downregulate macrophage TNF -0 production. In a second proj ect, to elucidate the tissue specific transcriptional control of the pactolus gene, the pactolus promoter was analyzed utilizing EMSA protocols. These data suggest a role for PD.1 and NF-KB inpactolus gene regulation. TABLE OF CONTENTS ABSTRACT ....................................................................................... iv LIST OF FIGURES ............................................................................. viii ABBREVIATIONS ............................................................................... xi ACKNOWLEDGMENTS ...................................................................... xii Chapter 1. INTRODUCTION: PACTOLUS FUNCTION .................................. 1 Pactolus Discovery and Initial Characterization ............................ 2 Neutrophils: Development and Function .................................... 3 B2 Integrin Subunit: Function and the B2 Integrin Deficient Mouse ............................................................................. 6 Comparison of Pactolus to the B2 Integrin Subunit ........................ 11 Introduction to Thesis Work .................................................. 12 References ....................................................................... 15 2. GENOMIC ORGANIZATION, CHROMOSOMAL LOCALIZATION, AND TRANSCRIPTIONAL VARIENTS OF THE MURINE PACTOLUS GENE ................................................................................. 20 Abstract .......................................................................... 21 Introduction ..................................................................... 21 Materials and Methods ......................................................... 22 Results ........................................................................... 22 Discussion ........................................... , ........................... 25 References ....................................................................... 27 3. PACTOLUS DEFICIENT MICE DEMONSTRATE AN ALTERED TNF-a RESPONSE FOLLOWING NEUTROPHILIMACROPHAGE INTERACTIONS ................................................................... 28 Abstract .......................................................................... 29 Introduction ...................................................................... 29 Materials and Methods ......................................................... 30 Results ............................................................................ 36 Discussion ....................................................................... 53 Acknow ledgements ............................................................. 57 References ....................................................................... 58 4. PACTOLUS, A NOVEL PROTEIN INVOLVED IN THE CLEARANCE OF APOPTOTIC NEUTROPHILS ............................................... 62 Abstract .......................................................................... 63 Introduction ...................................................................... 63 Materials and Methods .......................................................... 66 Reslllts ............................................................................ 70 Discussion ....................................................................... 93 Acknowledgements ........................................................... 100 References ................................ ' ...................................... 100 5. CONCLUSION: PACTOLUS FUNCTION .................................... 104 Discussion ......................................................................... 105 References ...................................................................... 110 6. INTRODUCTION: PACTOLUSTRANSCRIPTION ....................... .113 Myeloid Specific Promoters ................................................. 114 PU.l Function .................. , ............................................... 114 PU.l and 132 Integrin Subunit Promoter Regulation ...................... 117 NF-KB Activation and Function ............................................. 118 Introduction to Thesis Work ................................................. 119 References ....................................................................... 121 7. IDENTIFICA TION OF PUTATIVE TRANSCRIPTIONAL CONTROL ELEMENTS OF THE MURINE PACTOLUS GENE ................................................................................ 125 Resttlts .......................................................................... 126 Methods .......................................................................... 160 References ...................................................................... 167 vi 8. CONCLUSION: PA eTOL US TRANSCRIPTION ............................ 170 Discussion ...................................................................... 171 References ...................................................................... 180 vii LIST OF FIGURES Figure Page 1.1 Intronlexon homology between the murine pactolus gene and the murine B2 integrin subunit gene ................................................. 7 2.1 Organization of the murine Pactolus gene .................................. 22 2.2 Sequence analysis of the Pactolus intronlexon borders ................... 23 2.3 Intronlexon homology between murine Pactolus and murine integrin subunit B2 .................... ..................................................... 24 2.4 Chromosomal localization of the murine Pactolus gene ................... 25 2.5 Identification of the Pactolus C transcript ................................... 26 2.6 Sequence and alignment of the Pactolus C transcript ...................... 26 2.7 Alignment and generation of the three alternatively spliced fonns of Pactolus .......................................................................... 27 3.1 Generation of a Pactolus null mouse ........................................ .37 3.2 Analysis of pac to lus expression in the deficient animals .................. 40 3.3 Distribution of bone marrow and blood leukocytes in Pactolus deficient animals ............................................................................ 44 3.4 Recruitment ofneutrophils in Pactolus deficient animals ................. 46 3.5 Infectious response by Pactolus deficient animals ......................... 50 3.6 Modulation ofTNF-a production by Pactolus deficient neutrophils ....................................................................... 54 4.1 Granule localization of intracellular Pactolus ............................... 72 4.2 Cell surface expression of Pac to Ius following neutrophil recruitment ....................................................................... 76 4.3 Translocation of Pac to Ius by macrophage supernatant and LTB4 ........ 81 4.4 Pertussis toxin blocks LTB4 induced Pactolus translocation ............. 83 4.5 Staurosporine induces Pactolus exocytosis .................................. 86 4.6 Pactolus deficient neutrophils demonstrate reduced uptake by macrophages ..................................................................... 94 7.1 Pactolus promoter sequence ................................................. 127 7.2 Multiple promoter alignment. ................................................ 129 7.3 RT-PCR for pactolus transcripts ............................................ 132 7.4 Nuclear extract concentration gradient EMSAs for two pactolus promoter fragments ........................................................... 134 7.5 The EMSA pattern with the Al pactolus promoter fragment mimics the F3 EMSA pattern .......................................................... 137 7.6 Nuclear extract concentration gradient EMSAs for the A 1 promoter fragment ........................................................................ 139 7.7 Oligo competition to discover the common Al (and F3) transcription factor binding site .............................................................. 142 7.8 The Al pactolus promoter fragment binds to transcription factor PU.l ............................................................................. 145 7.9 RT-PCR for transcription factors .......................................... 150 7.10 A 1 promoter fragment EMSA patterns using extract from cells that do not express pactolus ...................................................... 153 7.11 The binding site of the putative MMC blocking factor overlaps with the PU.l site ........................................................................ 156 7.12 Identification of the transcription factor binding site on the F5 pactolus promoter fragment ............... '" ........................................... 158 ix 7.13 F5 promoter fragment EMSA band supershifts with antibodies to the different NF-KB subunits .................................................... 161 8.1 Model of pactolus transcription .............................................. 177 x ABBREVIATIONS MIDAS, metal ion dependent adhesion site: LTB4, leukotriene B4: CFU, colony fonning units: TK, thymidine kinase: CTMC, connective tissue mast cell: MMC, mucosal mast cell: SCF, stem cell factor: DNFB, di-nitro-fluro-benzene: BMDM, bone marrow derived macrophages: MPO, myeloperoxidase: AP, alkaline phosphatase: MMP-9, gelatinase B: IMDEM, Iscove's Modified Dulbecco's Medium: PTXN, pertussis toxin: PKA, protein kinase A: PKG, protein kinase G: WT, wild type: ES, embryonic stem: PMA, phorbol myristral acetate: LF A-I, leukocyte function-associated antigen-I: Mac-I, macrophage antigen-I: PEBP2/CBF, polyoma enhancer binding protein 21 core binding factors: C/EBP, CCAA T 1 enhancer binding protein: M-CSFR , monocyte-colony stimulating factor receptor: GM-CSF, granulocyte macrophage-colony stimulating factor receptor: G-CSF, granulocyte -colony stimulating factor receptor: TBP, TAT A binding protein: ATRA, all trans retinoic acid: EMSA, electrophoretic mobility shift assay: NF­KB, nuclear factor-KB: transcription factor, TF. ACKNOWLEDGMENTS I'd like to thank my advisor Dr. John Weis, past and present members of the Weis labs, and members of the Pathology department for all the ideas, suggestions and critical evaluations of my work. This includes my committee: Dr. Gerald Spangrude, Dr. Ray Daynes, Dr. Tom McIntyre and Dr. Robert Fujinami. I'd also like to thank Kim Cash, the Molecular Biology Program (Tami Brunson, Barb Saffel), the Developmental Biology Training Grant (Dr. Carl Thummel, Debby Shill, Patty Lisieski) and the Weber Immunology Trust; all were instrumental to my graduate student career. I'd also like to thank my family - my husband: Craig Waddell, my parents: Robert and Nancy Margraf, my sister and her family: Jennifer, Scott, Skye and Vincent Vandeleest, my in-laws: Bruce and Kathy Waddell, and my brother-in-law: Clint Waddell. Also, a thank you to the friends I have made in Salt Lake City over the years. Thank you all for the support! CHAPTER 1 INTRODUCTION: PACTOLUS FUNCTION 2 Pactolus Discovery and Initial Characterization Pactolus was discovered during differential display between two types of mast cells derived from murine bone marrow}. Bone marrow cells grown in the presence of c­kit ligand (stem cell factor - SCF) developed into a connective tissue-like mast cell (CTMC) phenotype. These mast cells are generally found in the lung and in the serosa of body cavities2 . Bone marrow cells were cultured in interleukin 3 (IL-3) to derive the mucosal mast cell (MMC) phenotype. MMCs are found in the mucosa of the gastrointestinal tracr. CTMCs make pactolus transcripts, while the MMCs lose their ability to make pactolus after 2-3 weeks development in culture1 • Mast cells are involved in allergic reactions. They bind IgE antibody to their surface with their high affinity FCE Receptor. When antigen binds and cross-:-links the IgE surface antibody, the mast cells release many inflammatory mediators, proteases and histamine2 . This can lead to life threatening consequences, depending on the amount and location of mast cell activation. Although some characterization of Pactolus found in CTMCs have been done, the function of Pac to Ius on mast cells has been largely unexplored. This is due to no Pactolus cell surface expression on CTMCs and the discovery of Pac to Ius on another leukocyte important in host defense against pathogens: the neutrophil. Pactolus is expressed from early in neutrophil development, at the promyelocyte stage, to mature neutrophils3(unpublished data). Pactolus is homologous to I3 integrin subunits and is most identical (extra-cellular region is 68% identical) to the Il2 integrin subunit (CD 18)1. Like the Il2 integrin subunit, Pactolus has extra-cellular, transmembrane and cytoplasmic sequences. The pactolus transcript has two alternatively spliced fonns, the full-length fonn (Pac A) and a 3 truncated, presumed degraded form (Pac B). Most mice make a ratio of Pac A to Pac B, while C57BL/6 mice make only Pac A due to a point mutation in the Pac B acceptor site3 ,4. Alternative splicing is a method integrins can utilize to regulate signaling events or ligand binding5 ,6. Neutrophils: Development and Function The Pactolus protein is found on the cell surface of immature to mature neutrophils3 • Neutrophils are the first line of defense against foreign invaders. They accomplish this by phagocytizing and killing the pathogens. Neutrophils differentiate and mature in the bone marrow, then are released into the blood. In humans, neutrophils are plentiful in the blood and circulate for about 6 hours before undergoing apoptosis and elimination by Kupffer cells in the liver, unless they are recruited to an area of inflammation. Neutrophils bind to and migrate through the endothelial cell layer to enter an area of inflammation by a well-characterized sequence of events2 ,6-10. Resting, unactivated endothelium does not bind to leukocytes. After the endothelial cell layer is activated by inflammatory mediators [such as tumor necrosis factor-a (TNF -a), histamine, or lipopolysaccharide (LPS)], the cells express selectins and intercellular adhesion molecules (lCAMs), which are receptors for neutrophils and other leukocytes. The neutrophils can tether to the selectins, a transient interaction that causes the neutrophils to roll across the activated endothelium in the direction of the blood flow. Then the neutrophils are activated by the chemoattractants expressed by the endothelial cells and this initiates firm adhesion of the neutrophil to the endothelial cell via Ih integrins. At this point, the neutrophils can extravasate through the endothelial cell layer using integrins and CD311PECAM-l (platelet endothelial cell adhesion molecule-I). The maximal amount of neutrophil influx happens about 4-8 hours after infection2,11. 4 At the site of infection, neutrophils bind to pathogens using complement receptors, Fc receptors for antibodies, or specific bacterial sugar receptorslO. The neutrophils can kill pathogens by utilizing phagocytosis and respiratory burst and/or releasing inflammatory mediators, such as myeloperoxidase, proteases, cytokines or chemokines 10. The inflammatory mediators recruit additional leukocytes to the area of infection. After 24 hours, the neutrophils are replaced as the major leukocyte at the site of infection by macrophages2. Macrophages are important for downregulating inflammation, because the macrophages can phagocytize the neutrophils to help destroy their ingested pathogens or reduce the release of inflammatory mediators after neutrophil death10,12-14. Neutrophils that are not eliminated can become necrotic and eventually lyse, releasing their histotoxic compounds into the surrounding tissue, which leads to tissue damage and increased inflammation12,15. Macrophages recognize and bind to neutrophils during neutrophil activation (with CD31, ICAM-l), apoptosis (with CD31, avB3, scavenger receptors and phosphatidylserine receptor - PSR) and necrosis (with complement receptors, Fc receptors)13,16-20. Macrophages binding to neutrophils during activation can be important in sequestering the neutrophils to a certain location. For example, when ICAM on Kupffer cells binds to Mac-I (amB2) on neutrophils, it can sequester the neutrophils in the liver which was required for elimination of certain infections13 ,17. 5 Macrophages also need to recognize apoptotic and necrotic neutrophils in order to clear inflammation and to limit the release of potential tissue damaging reagents12,14. Necrotic neutrophils can cause macrophages to take on an activated, inflammatory cytokine profile (which can recruit more leukocytes), while macrophages phagocytizing apoptotic neutrophils become anti-inflammatory (to downmodulate inflammation)12,21-23. The macrophage has phagocytic receptors (PSR, uvfh, CD36) that can bind to apoptotic neutrophil ligands (exposed phosphatidylserine - PS, thrombospondin binding sites, modified lipids)16,19,24. PSR, UvB3, and CD36 signaling can specifically increase anti-inflammatory mediator production (IL-l 0, TGF -B, prostaglandin E2, and platelet activating factor - P AF) by the macrophages 14,22,25, which downmodulates the inflammatory cytokines (IL_l,TNF_u)14,16,22,26. Necrotic neutrophils are bound and phagocytized by other methods, the macrophage complement receptor and F c receptors, which causes an inflammatory macrophage cytokine profile15,21. Lysed neutrophils release annexin V that binds to exposed PS and blocks the PSRlPS interaction, which can specifically decrease signaling through the PSR (therefore increase inflammation). The lysed neutrophils also release proteases that can cleave the PSR receptor15,23. It is the mechanism of phagocytosis that mediates the macrophage response not something intrinsic to the apoptotic cell. Meagher et aI, demonstrated macrophage phagocytosis of apoptotic neutrophils can decrease the release of pro-inflammatory mediators, while the macrophages released more pro-inflammatory mediators after phagocytizing antibody and complement opsonized apoptotic neutrophils21 , Huynh et al. also compared macrophage response to apoptotic versus opsonized apoptotic cells in vivo and in vitro. The apoptotic cells elicited higher macrophage TGF-B secretion than the opsonized apoptotic cells. The increase in TGF-3 was due to PSR activation by the exposed PS on the apoptotic cellsl4 . So, macrophages phagocytizing apoptotic cells are anti-inflammatory, while phagocytosis of necrotic, lysed or opsonized cells causes the macrophage to be pro-inflammatory. fh Integrin Subunit: Function and the fh Integrin Deficient Mouse Pactolus is homologous to the 3 integrin subunit family of proteins and to 32 integrin subunit in particularl . The mRNA splice sites (Figure 1.1) and promoter regions of Pac to Ius and 32 are also similar, leading to the hypothesis that the pactolus gene is a duplication of the 32 integrin subunit gene4 • 6 Integrins are heterodimers consisting of a 3 subunit non-covalently pairing with an a subunit partner. There are many a subunits and 8 subunits, and they can pair in various combinations to form integrins. The integrins each have different (or overlapping) ligands and functions6 ,27. All four a subunits that associate with 82 (aL, am, ax and ad; or CDlla, b, c, d, respectively) and 32 itself have a MIDAS (metal ion gependant !!dhesion ~ite) domain. The MIDAS domain is a region of the protein that folds together to chelate a divalent cation and this domain is required for the integrin to bind to its ligand6,27-3o. Leukocyte function-associated antigen-l (LF A-I) is a combination of aL and 32 and binds members of the ICAM family_ LFA-l is expressed on all leukocytes and functions as an intercellular adhesion molecule for migration or for T cell co­stimulation2 ,31 ,32. Macrophage antigen-l (Mac-I, also called complement receptor 3 7 Figure 1.1. Intron/exon homology between the murine pactolus gene and the murine 81 integrin subunit gene. The coding sequences of Pactolus and B2 integrin are shown from exons 2 to 16. The full intronlexon map of the murine B2 integrin gene has not been published; exons are inferred from comparison with the intronlexonjunctions of the human B2 integrin gene33 ,34. Sequences have been adjusted to give the greatest degree of alignment. The gapped sequence of Pactolus in exon 6 represents the sequences within the MIDAS domain that are lacking in the Pactolus gene (compared with other members of the B integrin gene family). This is the specificity-determining loop (SDL). The conserved MIDAS motif (DXSXS) is highlighted by *. 8 Exon 2 Pac MLGQCTl1f-VLAGLLSlE~ <D18 MLGLRPSLLLALAGLFFLGS 3 Pac A.LllLCTKDNVST£QDCI RSGPS~AWCOKL <D18 AVSQECTKYKVSSCRDCIQSGPGCSWCQKL 4 Pac NFTGRGEPDSVRCDTPEOLLLKGCTSEYLVOPKSL~ESQE.!tKERDQROLSPRNVTVFLRP mIl NFTGPGEPDSLRCDTRAQLLLKGCPADOIMOPRSIANPEFOQRGQRKQLSPQKVTLYLRP '11'.'11'11' 5 Pac GQAATEI\Y.DEQ.B,TQDNSVDLYFl GLSG~ QGHlSNVQTLGSDLLKALNEI SRSGRI mIl GQAAAFNVTFRRAKGYPIOLYYL DLSYS LDDLNNVKKLGGDLLQALNEITESGRI Pac SOL 6 GF6SI:(NMI- - -- - ~ - - - - ----- - - - - - - - - - - - - -EQ.!!.I LKL TAOiSruR~LR.!Sru:.VSGK1.ATIKGQLDAV\lmI~L mIl GFG SFVDKTVLPFVNT HP EKLRN P CPNKEKACQPPFAFRHVL KL TDNSNQFQTEV G KQL I SGNLDA PE G GLOAI MQVAAC P 7 Pac GE I GWRNGTRFL VL VTDNDFHLA.KQ.KTLGT RQNTSDGRCHL DQ.GMYRSRGs,p ml8 EEl GWRNVTRLLVFATDDGFHFAGDGKLGAI LTPNDGRCHLEDNMYKRSNEF 8 Pac DYQSVV!ib1SKLA~NNI OPI FVYSRMVKrYE mIl DYPSVGQLAHKLSESNIQPIFAVTKKMVKrYE 9 Pac KLTTFIPKLTIGELSDDSSNVA~RNAYS mIl KLTEIIPKSAVGElSDDSSNVVQLIKNAYY 18 Pac KLSSI:(VlNHSTIISILmlliYCSNGTiNPGKPS6DCSGVQI HDQ mIl KLSSRVFLOHSTLPDTLKVTYDSFCSHGASSIGKSRGDCDGVQINNP 11 Pac VT FOV NIT ASECF RWf .F!.QA LG F M.!tSm,R:(L tLC ECQ~Q E !iQH HS L C6 GK GAl1E CG I CR mIl VTFQVKVMAS ECI QEQSFVI RALGFTDTVTVQVRPQCECHCROQS REOSL CGGKGVMECGI CR 12 Pac ~NSGYAGKNCECOTOGP ill,DLEGSCRKON SSII'ICSGLGDCI CGOC ECHT SOl PNKElYGQYCECD NV NCERYDGQVCGGP mIl CESGYl GKNCECQTQ GRSSQELERNCRKDNSSIV C SGLGDCI CGQCVCHTSDV P NKE I FGQY CECD NV NCERYNSQVCGGS 13 Pac ERGH£SCGRg:~RYSFVGSACOCRMSTSGCLNN.Br.1VECSGH[R~YCNRCLCDPGYQPPLCEKRtGYFHR£SEYV ml8 DRGSCNCGKCSCKPGYEGSACQCQRSTTGCLNARLVECSGRGHCQCNRCI CDEGYQPPI'ICEDCPSCGSHCRDNHT 14 Pac SCARCLKDNSAI KCRE£WHLLFSNTPFSNKTCMTERDSEGC'tITTYTLIQPD1SDI NSIYI KESL mIl SCAECLKFDKGPFEKNCSVQCAGI1TLQTIPLKKKPCKEKDSEGCWITYTLQQKDGRNIYNIHVEDSL 15 Pac Vg.EI S.t!lTI LliVl:(GVLLAVI FLLVYCMVY!..KGTQKAAKLP.B,KGGV mIl ECVKGPWVAAIVGGTVVGVVLIGVLLLVIWKALTHLTDLREYRRFEKEKL 16 Pac QiTLAQQPHFQEPHHVEPVWNQERQGTQ mIl KSQWNNDNPLFKSATTTVMNPKFAES 9 CR3) is a combination of am and 82. Mac-l binds many ligands to obtain these functions: adhesion, degranulation, inducer of oxidative burst, migration and/or phagocytosis2,35-37. Mac-l is found mainly on myeloid cells and its ligands are numerous: fibrinogen, LPS, complement component (iC3b), Factor X, ICAM-I, etc2,35. Neutrophil activators release additional Mac-l to the cell surface, which is important for Mac-l function at the site of infection (migration, phagocytosis)32. Similarly, the majority of Pac to Ius is sequestered inside the neutrophil, until neutrophil activation by inflammatory mediators (L TB4 or PMA) releases Pactolus to the cell surface3,38. Both LFA-l and (to a lesser extent) Mac-l are important for leukocyte firm adhesion to activated endothelial cells and subsequent recruitment to the site of infection2,31,32,36,37. LF A-I and Mac-l are the most abundant 82 integrins found on neutrophils. 82 also pairs with ax to form p 150,95 (also called complement receptor 4) and ad to create adhesion and phagocytic receptors2. This discussion will concentrate on the role of 82 integrin in neutrophil function, for that is the leukocyte that expresses Pactolus3. Humans deficient in 82 have leukocyte adhesion deficiency type-l (LAD 1) and their neutrophils cannot cross the endothelium, because they cannot firmly adhere to endothelial cells. This leads to frequent and severe infections6,36,39. Mice deficient in 82 40,41, a m 36,37, and UL31 ,32 have been created. 82 deficient mice have increased numbers ofneutrophils in the blood. Neutrophil recruitment to the skin was undetectable while neutrophil recruitment to the peritoneal cavity was normaI40,41. Scharffetter-Kochanek et al. proposed that there is a 82-independent method for neutrophil recruitment to the peritoneal cavity, while 82 is required for neutrophil recruitment to the skin40,4I. Neutrophils of human LAD 1 patients cannot be recruited to 10 the peritoneum, indicating there is no 32-independent pathway in humans39,42. Another hypothesis was proposed for this experimental result, that neutrophils are recruited to the peritoneal cavity because of the increase of neutrophils in the blood, but this theory does not explain the complete absence of neutrophils recruited to the skin43. LAD 1 patients and LF A-I deficient nlice also have blood neutrophilia, but demonstrate no peritoneal recruitment of neutrophils or a decrease in neutrophil recruitment (only at the 4 hour time point), respectively31,39,42. 32 deficient mice displayed a reduced ability to clear bacteria and did not recover from peritoneal bacterial infection as well as wild type mice41. The il2 deficient mice developed opportunistic skin infections due to the lack of recruited neutrophils41 . In contrast to the 32 deficient mice, neither LFA-l (OL) deficient mice nor Mac-l (urn) deficient mice develop spontaneous skin infections31 ,32,37. The LFA-l deficient mice have blood neutrophilia and demonstrated a 50% decrease only during early (4 hour) neutrophil recruitment to the peritoneal cavity31,32. The Mac-l deficient mice have an increase in neutrophils recruited to the peritoneal cavity. This was caused by delayed neutrophil apoptosis which results from the loss of Mac-l mediated phagocytosis. The reactive oxygen intermediates generated after phagocytosis normally induces neutrophil apoptosis37. Both LFA-l and Mac-l deficient mice show increased mortality with bacterial infections, due to less leukocyte migration and sepsis, respectivelyll. The Mac-l deficient neutrophils demonstrate reduced phagocytosis and respiratory burst37. 11 Comparison of Pactolus to the fh Integrin Subunit Pactolus is homologous to the 8 integrin subunit family of proteins and to 62 in particular1. Pactolus is hypothesized to be a duplication of the 62 gene due to the similarity in the genetic structure between the two genes, such as the same number of exons, conserved splice sites and conserved promoter regions (Figure 1.1 and Chapter 2)4. This duplication was after the evolutionary split between man and mouse, as we have not detected Pactolus in the human genome. The two genes map to different chromosomes in the mouse (pactolus - 16 and 82 - 10) that are syntenic to the same human chromosome, 21. While 82 integrin subunit is found on all leukocytes, Pactolus expression is more restricted to neutrophils and CTMCsl w 3 • Pactolus function was originally hypothesized to be integrin-like due to the significant homology in the extra-cellular region of 68% identityl. But there were several important differences that might indicate divergent functions or at least a divergent mechanism of action. Integrins are heterodimers with an a and 8 subunit and the subunits cannot be expressed on the surface without their partner6 ,44. We have found Pactolus on the neutrophil cell surface without a detectable alpha chainl. This may be due to the large deletion in ex on 6 of Pactolus (the specificity determining loop) with respect to all other 6 subunit sequences (Figure 1.1). When Takagi et al. deleted this sequence in various 6 subunits, some of them could now be expressed on the cell surface without their a partner4,44. Another important difference is in the MIDAS (metal ion gependant ~dhesion ~ite) domain that is found in a and 6 integrin subunits6 • The MIDAS domain is the area of the protein that folds together to bind a divalent metal ion which is required for ligand 12 binding27 ,28. The Pactolus MIDAS domain has two mutations, a point mutation in the conserved DXSXS sequence to GXSXS and a deletion in exon six (Figure 1.1). The original sequence of both mutations would be important in forming the integrin MIDAS structure and ligand bindingl ,4,27-3o. Pactolus probably binds to its ligand with a different mechanism than the integrins. The cytoplasmic tail of Pac to Ius does not contain the conserved actin cytoskeleton binding sequences of integrins or demonstrate signaling capacity such as autophosphorylation or calcium flux 3,5,45-47. B integrins contain a cysteine rich region that is internally disulfide bonded6 . About 90% of all cysteines are conserved between B2 and Pactolus, indicating that the overall structures of the two proteins may be similar even if their functions prove to be different4,44. Introduction to Thesis Work The first section of this thesis addresses the characterization of the Pactolus gene and protein and the discovery of Pactolus function. Chapter 2 characterizes the pactalus genetic structure 4. These data were utilized to compare B2 integrin gene structure to Pactolus (68% identical in the extra-cellular region)l. Similarities, such as conserved mRNA splice sites, conserved cysteines and a nearly identical sequence in a portion of the promoter region, indicates that Pactolus was a duplication of the B2 integrin subunit gene. Pactolus also may have B2 structure (due to conserved cysteines)44 and similar transcription regulation (Chapters 6-8). The mechanism of Pac to Ius function is proposed to be different than integrins, because Pactolus does not appear to have a partner and lacks actin cytoskeleton binding sequences. The mutations in the Pactolus MIDAS domain may indicate that Pactolus binds its ligand differently than integrins. Although, the mechanism of action is most likely divergent, Pactolus may still serve a role in cell adhesion, if it maintained functional homology with the integrins. 13 Our first attempts to discover Pactolus function by expressing Pactolus in other leukocytes was unsuccessful. Another cell type, COS cells, can express transfected Pactolus, but it does not have the extensive glycosylation that the mature form of Pactolus possesses. We believe this glycosylation may be important for Pactolus binding to its ligand3 • We did not have any Pactolus negative cell lines that could express the mature form of Pactolus after transfection for experimental functional comparisons. Also, we did not have a Pactolus blocking antibody or know the identity of the Pactolus ligand. So, the early work on Pactolus was done in primary neutrophils isolated from murine bone marrow or peritoneal wash after neutrophil recruitment. These initial characterizations of Pactolus lead to many questions as to its function. Pactolus was discovered to be on the cell surface and in granules of immature to mature neutrophils in the bone marrow3. Could Pactolus be required for neutrophil development within the bone marrow or have a role in the release of mature neutrophils into the periphery? The majority of Pactolus is found in the neutrophil granules and can be released by inflammatory mediators (LTB4, PMA)3,38. Mac-I, whose function is important at the site of infection, also can be released to the cell surface of neutrophils with activation32. Could Pactolus have a role in neutrophil function after activation (phagocytosis, oxidative burst) and/or during an immune response? 14 Thioglycollate elicited peritoneal neutrophils already have the majority of Pactolus on their cell surface, which does not increase with additional stimulation38 • Could Pactolus be required for neutrophil recruitment to the sites of infection? Previously, I discussed that the 62 integrin deficient mice have normal peritoneal neutrophil recruitment, while LAD 1 patients do not. This was hypothesized to be due to a murine specific 62-independent recruitment mechanism40,41. Could murine Pactolus be the 62-independent mechanism for neutrophil recruitment to the peritoneal cavity? Many of these questions of Pac to Ius function had to be answered in viva or needed a Pactolus negative neutrophil as an experimental control. One method utilized to address these questions of Pac to Ius function was to analyze a mouse that lacks Pactolus (Chapter 3t8 • The Pactolus null mouse was created by inserting the neomycin gene into exon 5, in the middle of the MIDAS domain-like sequences. The pactalus transcripts are not made past the neomycin insertion and the Pactolus protein cannot be detected in the pactalus (-1-) mouse. The Pactolus null mouse can develop mature neutrophils that are released into the blood and recruited to sites of inflammation. The overall immune function appears normal in the Pactolus null mouse, as phagocytotic function of the neutrophils and recovery from a bacterial infection was normal (unpublished data)48. We could not detect any signaling events after cross-linking Pactolus with antibody (Chapter 4)38, so our attention shifted to Pactolus as a ligand that might interact with another leukocyte popUlation to modulate the immune response. The Pactolus deficient mouse (discussed in Chapter 3) was instrumental in two discoveries that lead to a hypothesis of Pactolus function. In Chapter 4, we demonstrate that Pactolus may facilitate apoptotic neutrophil phagocytosis by macrophages38 • In 15 Chapter 3, the data indicate an interaction of neutrophils with macrophages through Pactolus could modulate macrophage production ofTNF-u48 . These new discoveries suggest that neutrophil Pactolus binds a macrophage receptor that can regulate TNF-u production and facilitate apoptotic neutrophil phagocytosis. References 1. Chen, Y., Garrison, S., Weis, 1. 1. & Weis, 1. H. Identification of pac to Ius, an integrin beta subunit-like cell-surface protein preferentially expressed by cells of the bone marrow. J Bioi Chern 273, 8711-8 (1998). 2. Abbas AK, L. A., Pober 1S. Cellular and Molecular Immunology (W.B. Saunders Company, 1997). 3. Garrison, S., Hojgaard, A., Patillo, D., Weis, 1. 1. & Weis, 1. H. Functional characterization of Pactolus, a beta-integrin-like protein preferentially expressed by neutrophils. J Bioi Chern 276, 35500-11 (2001). 4. Margraf, R. L., Chen, Y., Garrison, S., Weis, J. 1. & Weis, 1. H. Genomic organization, chromosomal localization, and transcriptional variants of the murine Pactolus gene. Mamm Genome 10,1075-81 (1999). 5. de Melker, A. A. & Sonnenberg, A. Integrins: alternative splicing as a mechanism to regulate ligand binding and integrin signaling events. Bioessays 21, 499-509 (1999). 6. Hynes, R. O. Integrins: versatility, modulation, and signaling in cell adhesion. Cell 69, 11-25 (1992). 7. Springer, T. A. Traffic signals for lymphocyte recirculation and leukocyte emigration: the multistep paradigm. Cell 76, 301-14 (1994). 8. Mul, F. P. et al. Sequential migration ofneutrophils across monolayers of endothelial and epithelial cells. J Leukoc Bioi 68, 529-37 (2000). 9. Siegelman, M. More than the sum of the parts: cooperation between leukocyte adhesion receptors during extravasation. J Clin Invest 107, 159-60 (2001). 10. Witko-Sarsat, V., Rieu, P., Descamps-Latscha, B., Lesavre, P. & Halbwachs­Mecarelli, L. Neutrophils: molecules, functions and pathophysiological aspects. Lab Invest 80, 617-53 (2000). 16 11. Prince, J. E. et al. The differential roles ofLFA-l and Mac-l in host defense against systemic infection with Streptococcus pneumoniae. J Immunol166, 7362- 9 (2001). 12. Savill, J. Apoptosis in resolution of inflammation. J Leukoc BioI 61, 375-80 (1997). 13. Gregory, S. H. et al. Complementary adhesion molecules promote neutrophil­Kupffer cell interaction and the elimination of bacteria taken up by the liver. J Immunol168, 308-15 (2002). 14. Huynh, M. L., Fadok, V. A. & Henson, P. M. Phosphatidylserine-dependent ingestion of apoptotic cells promotes TGF-betal secretion and the resolution of inflammation. J Clin Invest 109, 41-50 (2002). 15. Henson, P. M., Bratton, D. L. & Fadok, V. A. The phosphatidylserine receptor: a crucial molecular switch? Nat Rev Mol Cell BioI 2, 627-33 (2001). 16. Savill, J. & Fadok, V. Corpse clearance defines the meaning of cell death. Nature 407, 784-8 (2000). 17. Gregory, S. H. & Wing, E. J. Neutrophil-Kupffer cell interaction: a critical component of host defenses to systemic bacterial infections. J Leukoc BioI 72, 239-48 (2002). 18. Brown, S. et al. Apoptosis disables CD31-mediated cell detachment from phagocytes promoting binding and engulfment. Nature 418, 200-3 (2002). 19. Ren, Y., Silverstein, R. L., Allen, J. & Savill, J. CD36 gene transfer confers capacity for phagocytosis of cells undergoing apoptosis. J Exp Med 181, 1857-62 (1995). 20. Magnarin, M., Spessotto, P., Soranzo, M. R., Pontillo, A. & Zabucchi, G. Human neutrophils specifically interact with human monocyte-derived macrophage monolayers. Inflammation 24, 89-98 (2000). 21. Meagher, L. C., Savill, J. S., Baker, A., Fuller, R. W. & Haslett, C. Phagocytosis of apoptotic neutrophils does not induce macrophage release of thromboxane B2. J Leukoc BioI 52, 269-73 (1992). 22. Fadok, V. A. et al. Macrophages that have ingested apoptotic cells in vitro inhibit proinflammatory cytokine production through autocrine/paracrine mechanisms involving TGF-beta, PGE2, and PAF. J Clin Invest 101, 890-8 (1998). 17 23. Fadok, V. A., Bratton, D. L., Guthrie, L. & Henson, P. M. Differential effects of apoptotic versus lysed cells on macrophage production of cytokines: role of proteases. J Immunol166, 6847-54 (2001). 24. Fadok, V. A. et al. Different populations of macro phages use either the vitronectin receptor or the phosphatidylserine receptor to recognize and remove apoptotic cells. J Immunol149, 4029-35 (1992). 25. Fadok, V. A. et al. A receptor for phosphatidylserine-specific clearance of apoptotic cells. Nature 405, 85-90 (2000). 26. McDonald, P. P., Fadok, V. A., Bratton, D. & Henson, P. M. Transcriptional and translational regulation of inflammatory mediator production by endogenous TGF-beta in macrophages that have ingested apoptotic cells. J Immunol163, 6164-72 (1999). 27. Dickeson, S. K. & Santoro, S. A. Ligand recognition by the I domain-containing integrins. Cell Mol Life Sci 54, 556-66 (1998). 28. Yalamanchili, P., Lu, C., Oxvig, C. & Springer, T. A. Folding and function of I domain-deleted Mac-1 and lymphocyte function-associated antigen-I. J BioI Chem 275, 21877-82 (2000). 29. Bajt, M. L., Goodman, T. & McGuire, S. L. Beta 2 (CD18) mutations abolish ligand recognition by I domain integrins LFA-l (alpha L beta 2, CDllaJCDI8) and MAC-1 (alpha M beta 2, CD11b/CDI8). J BioI Chem 270, 94-8 (1995). 30. Goodman, T. G. & Bajt, M. L. Identifying the putative metal ion-dependent adhesion site in the beta2 (CDI8) subunit required for aiphaLbeta2 and aiphaMbeta2ligand interactions. J BioI Chem 271,23729-36 (1996). 31. Schmits, R. et al. LF A -I-deficient mice show normal CTL responses to virus but fail to reject inlmunogenic tumor. J Exp Med 183, 1415-26 (1996). 32. Ding, Z. M. et al. Relative contribution ofLFA-l and Mac-l to neutrophil adhesion and migration. J Immunol163, 5029-38 (1999). 33. Weitzman, J. B., Wells, C. E., Wright, A. H., Clark, P. A. & Law, S. K. The gene organisation of the human beta 2 integrin subunit (CDI8). FEBS Lett 294, 97-103 (1991). 34. Wilson, R. W. et al. Gene targeting yields a CDl8-mutant mouse for study of inflammation. J Immunol151, 1571-8 (1993). 35. Plow, E. F. & Zhang, L. A MAC-l attack: integrin functions directly challenged in knockout mice. J Clin Invest 99, 1145-6 (1997). 36. Lu, H. et al. LF A-I is sufficient in mediating neutrophil emigration in Mac-1- deficient mice. J Clin Invest 99, 1340-50 (1997). 18 37. Coxon, A. et al. A novel role for the beta 2 integrin CD 11 b/CD 18 in neutrophil apoptosis: a homeostatic mechanism in inflammation. Immunity 5, 653-66 (1996). 38. Garrison, S., Hojgaard, A., Margraf, R.L., Weis, J.J., Weis, J.H. Pactolus, a novel protein involved in the clearance of apoptotic neutrophils. (submitted). 39. Anderson, D. C. et al. Leukocyte LFA-1, OKM1, p150,95 deficiency syndrome: functional and biosynthetic studies of three kindreds. Fed Proc 44,2671-7 (1985). 40. Mizgerd, J. P. et al. Neutrophil emigration in the skin, lungs, and peritoneum: different requirements for CD 111CD 18 revealed by CD 18-deficient mice. J Exp Med 186, 1357-64 (1997). 41. Scharffetter-Kochanek, K. et al. Spontaneous skin ulceration and defective T cell function in CD18 null mice. J Exp Med 188, 119-31 (1998). 42. Etzioni, A., Doerschuk, C. M. & Harlan, J. M. Of man and mouse: leukocyte and endothelial adhesion molecule deficiencies. Blood 94, 3281-8 (1999). 43. Walzog, B., Scharffetter-Kochanek, K. & Gaehtgens, P. Impairment of neutrophil emigration in CD 18-null mice. Am J Physiol276, Gl125-30 (1999). 44. Takagi, J., DeBottis, D. P., Erickson, H. P. & Springer, T. A. The role of the specificity-determining loop of the integrin beta subunit I -like domain in autonomous expression, association with the alpha subunit, and ligand binding. Biochemistry 41, 4339-47 (2002). 45. Sampath, R., Gallagher, P. J. & Pavalko, F. M. Cytoskeletal interactions with the leukocyte integrin beta2 cytoplasmic tail. Activation-dependent regulation of associations with tatin and alpha-actinin. J BioI Chem 273, 33588-94 (1998). 46. Yan, S. R. & Berton, G. Antibody-induced engagement ofbeta2 integrins in human neutrophils causes a rapid redistribution of cytoskeletal proteins, Src­family tyrosine kinases, and p72syk that precedes de novo actin polymerization. J Leukoc BioI 64, 401-8 (1998). 47. Fagerholm, S., Morrice, N., Gahmberg, C. G. & Cohen, P. Phosphorylation of the cytoplasmic domain of the integrin CD 18 chain by protein kinase C isoforms in leukocytes. J BioI Chem 277, 1728-38 (2002). 48. Margraf, R. L., Hojgaard, A., Garrison, S., Weis, J.J., Weis, J.H. Pactolus deficient mice demonstrate an altered TNF-alpha response following neutrophiVmacrophage interactions. (submitted). 19 CHAPTER 2 GENOMIC ORGANIZATION, CHROMOSOMAL LOCALIZATION, AND TRANSCRIPTIONAL V ARIENTS OF THE MURINE PACTOLUS GENE Reprinted from Mammalian Genome. (1999) 10:1075 21 Mammalian Genome 10,1075-1081 (1999). Genomic organization, chromosomal localization, and transcriptional variants of the murine pactolus gene Rebecca L. Margraf, Yiyou Chen, Sean Garrison, Janis J. Weis, John H. \-Veis The Divisio.n of Cell Biology and Immunology, Department of Pathology, University of Utah School of Medicine, 50 N Medica] Drive, Salt Lake City, Utah 84132, USA Received: 13 April 19991 Accepted: 18 June 1999 Abstract. The Pactolus gene encodes a cell surface protein simi­lar to that of the ~ integrin subunit family. The murine Pactolus gene is comprised of 16 exons that encompass 24 kb of the ge­nome. The genomic organization of the Pactolus gene is very similar to that described for the human 132 integrin gene (and deduced for murine 132 integrin), including a separate exon con­taining only 5' untranslated sequences. The Pactolus gene was mapped to a terminal region of murine Chromosome (Chr) 16, distinct from the previously mapped site of the ~2 integrin gene on murine Chr 10. The Pactolus gene encodes three distinct tran­scripts via alternative splicing. The Pactolus A transcript encodes the full-length protein including transmembrane and cytoplasmic domains, while the Pactolus B transcript truncates translation be­fore reaching these membrane-anchoring sequences. A newly dis­covered form, Pactolus C, is found in neonatal samples (along with Pactolus A) and would also encode a prematurely terminated pro­tein. This form is derived from an alternative splicing event that skips part of exon II, deletes exon ] 2, and uses an alternative acceptor site upstream of exon 13. The formation of the Pactolus Band C forms is thus governed by a complex aJternative splicing mechanism that is affected by the developmental status of the animal. Introduction Pactolus was first defined as a differentially expressed gene prod­uct present in murine bone marrow mast cells derived in stem cell factor (SCP2) but lacking in the analogous cells derived in IL-3 (Chen et a1. 1998). Sequence analysis of the full-length Pactolus cDNA indicated it contained regions of high homology to mem­bers of the 13 integrin subunit family, particularly 132 integrin. Transcript analysis suggested that bone marrow possessed the highest level of Pac to Ius expression. Of interest was the finding that the Pactolus gene, in adult bone marrow, produced two dis­tinct transcripts. One of these transcripts (dubbed Pactolus A) would be expected to encode a protein very similar to that of the 13 integrins with C-terminal transmembrane and cytoplasmic do­mains. while the other transcript (dubbed Pactolus B) would en­code a truncated protein lacking these C-terminal domains. Biotin labeling of cell surface proteins followed by immunoprecipitation with an antibody specific for cytoplasmic domain residues con- Correspondence to: J.H. Weis firmed the expression of the full-length Pactolus protein on the surface of bone marrow cells (Chen et al. 1998). The importance of the integrin family in bone marrow devel­opment, inflammation, targeted cell homing, and general develop­ment is evident in the phenotype of natural or targeted mutations in such genes. For example, spontaneous mutations within the ~2 integrin give rise to a disease known as Lymphocyte Adhesion Deficiency in which the migration of neutrophils to sites of infec­tion is abrogated, resulting in chronic and life-threatening bacterial infections (Springer 1990). Mouse knockout studies on such chains as 137 and a4 have demonstrated an absence of lymphocytes within mucosal tissues (Wagner et al. 1996), or developmental defects in the heart and placenta (Yang et a1. ]995), respectively. RAG-l­deficient blastocysts reconstituted with a4-deficient ES cells give rise to viable animals that lack such morphological sites of immu­nity as the intestinal Peyer's patches (Arroyo et al. 1996). Thus, the famHy of integrin chains plays many roles in the functional devel­opment of the animal. The homology between Pactolus and the 132 integrin subunit (Wilson et aI. 1989) is evident for the first 631 N-terminal residues of the Pactolus protein (63% identity). However, two regions are quite dissimilar. First, the amino acids from residue 631 to the c-terminus of the Pactolus protein (residue 738) are quite distinct from any other 13 integrin subunit. Second, the Pactolus protein lacks residues presumably important for the functional folding of the MIDAS domain (metal ion dependent adhesion sequence). The MIDAS domain has been demonstrated to be critical for a and 13 integrin chain pairing, and subsequent ligand recognition (Bajt and Loftus 1994; Michishita et aJ. 1993; Rieu et al. 1994; Veda et al. 1994; Wardlaw et al. 1990). The Pactolus protein is altered at this site by losing the strict conservation of the DxSxS region (replac­ing it with GxSxS) and lacking approximately 28 amino acids within that region compared with the other 13 integrins. Thus, while Pac to Ius possesses many characteristic features of the 13 integrins, it is distinct enough such that the protein may be more appropri­ately referred to as 13 integrin-like instead of simply referring to it as a t3 integrin subunit. The similarity between Pactolus and the 13 integrins suggests that they may have evolved from a common precursor genetic element. To investigate this possibility, we cloned the murine Pac­tolus gene, compared its intronlexon structure with that known for the 132 inlegrin, and determined its chromosomal location. While the Pacto]us gene is most similar in structure to that of the 132 integrin subunit, it does not share a common chromosomal location with that gene (MacDonald et a1. 1991). In addition, further tran­script analysis of the Pactolus gene products has identified a third 22 1076 R.L. Margraf et at.: Pactolus gene struclure A. 1 Kb B. 100bp >- >- >- 859 826 803 --< ....c 900 828 >- 807 ....c 816 >- >- 842 837 --< 815 Fig. 1. Organization of the murine Pactolus gene. A. The structure of the murine Pactolus gene was determined by analysis of a series of genomic lambda phage contigs. Three representative overlapping phage are denoted as 5' Phage #1, #2, and 3' Phage #2. The distance from the first to the last exon is approximately 24 kb. Exons 15 and 16, denoted with the ., which indicates that their position was approximated from the Southern data. Pactolus transcript (dubbed Pactolus C) present in neonatal liver that is a1so predicted to encode a truncated form of the protein similar to the Pactolus B product. Materials and methods Genomic library analysis. Phage containing the Pactolus gene were isolated by screening a lambda dash (Stratagene) murine genomic library created with l29/sv liver DNA with two Pactolus cDNA probes. The probes (a 5' probe 859 and a 3' probe 814: see Fig. I) were made by PCR of Pactolus cDNA (Chen et al' 1998). Pactolus phage are named 5' Phage or 3 t Phage to indicate which cDNA probe was used to isolate these phage from the library. Fourteen independent clones were isolated: all were spe­cific for the same gene, Pactolus. Each phage contained about 15 kb of genomic DNA. Mapping and sequence of the Pactolus gene. Pactolus phage were grown in and purified from LE392 E. coli. The phage were ordered 5' to 3' along the Pactolus gene by dot blot analysis of the phage DNA, utiliz.ing four Pactolus cDNA probes (listed as 901,902,88], and 837). To locate the boundaries of the exons, I JLg of phage DNA was sequenced with Pactolus primers (10 pmol total). This was done for both intron boundaries of each exon. Exact intron sizes were determined where the sequences overlapped. Other introns were estimated by PCR between exons with the same primers as for sequencing. Placements of exons ]5 and 16 are esti­mates based on exon 16's inclusion of the terminal HindIlI site. Exons #2 and #4 were accurately placed on the restriction enz.yme (RE) map by restriction sites found in the sequences 5' and 3' of each exon, respectively. Sizes of introns # I, #4, and # 14 were estimated after exon placement was known. In addition, subcloning of specific Pactolus genomic fragments into plasmid vectors allowed for a confirmation of exon and restriction enzyme site placement. ....c 848 >- 814 ....c 804 Restriction sites are denoted: E, EcoRl; B, BamHI; G, BgII; H, Hindlll. B. The coding region (shaded area) and untranslated regions (while) of the full-length Pactolus cDNA are shown in relation to the positions of the exons. In addition, the position of probes described in Materials and meth­ods is shown, as are the relative locations of PCR primers used in the RT-PCR analysis of Pactolus transcripts (see Fig. 5). Transcript RT-PCR analysis. Neonatal mouse liver RNA and adult bone marrow RNA were analyzed for Pactolus transcripts with RT·PCR as previously published (Tan and Weis 1992). Primers designated for the PCR analysis are shown in Fig. I. FISH analysis. Three overlapping Pactolus phage (5' Phage # I, 5' Phage #2, and 3' Phage #2) were used as independent probes of a murine chro­mosomal spread. The Pactolus probes were labeled by nick translation with SpectrumOrange dUTP (Vysis, Inc.), and the chromosomes were visual­ized with DAPI. Oligonucleotides. Oligonucleotides used in this study are as follows (sequence written 5' to 3'): #803-CAAAGTGGACTTCCAGCGGAC, #804-GA TTCCACACTGGCTCCACATG, #807 -GTCCA GCTG­GCAAGCAAACTG. #8 I 4-GCTGTTGGACAACCTACACTC, #815· GCACCTGCAGATGCCACACTC, #816· TGCTTGCCAGCTGGAC­CACTG, #817-CCAGAGAGGCCCATCAGGAAG, #826· TTGCTGAATCCCAGGAAGAC, #828·GAGTCCGCAG· TCAGCTTCAG, #837·CGAGTGCGACAATGTCAACTG, #842- CATCACAGCTTCAGAGTGCT, #848· TAGT ACTCGGAGCAGC· GATGG, #859-GCTAT AAGGAGCTGAAGAATC, #88 J­AGGGAAGCTGCCGGAAGGAC, #900·CCGCTCGAGACAGCT· GCCTCTGGTCCCTC, #901 -CTGGGT ACCAGACAGAACAC, #903· TCGGCCACAGGAGCAGTGGC and #I005·ACTGTAGGC· GTICCTGATGAG. Results Isolation and characterization of the murine Pactolus gene. A mouse genomic library was screened in two successive hybridiza- 23 R.L. Margraf et at.: Pactolus gene structure 1077 A B 5' PACTOlUS INTRONlEXOH BOUNDARIES S' t'~l 60 1'+~~'''J.YiiiJtii_i;~~f- -750 ~TI\.l\'.IGATGITlGAJ\lGIm~ -690 tn::l\~TITCMGl'a:;AAAAI]N"rCAATr:J:r,A -630 'lU:Gll:::'J::I'IIXAA~TICATTI\GIl.CG~ i""001 61 EXON #2 {61 bpI -570 G'TC'rAO\Cl\!~~cro:::'IOGTC:cA'!13NJ.rT.l'D::lJX."TCAClCN:.;ITC'JC'AJ\(:'J:'(X;OC EXON 113 (89 bpI -450 GAAAC'IOC]\GlC:rtrK'TCAG\crCl3Aro3'~~ Ikar05 lkaros -390 G1\(£IG1G:lGICJICA~CC'IGI\.~ GATA-l 39l irtron4 - 1 £iKb" -330 ~~~O::;A:ro:rr1\O:rAT ~~~fllI.IlI1f1 ••• IIl ••• Ii~~~ -270 ~'IGJ:CTCCA~ ~ lkaros PHASE 1 -210 I'ICIG:CCIClOOICD\.TGIlG.l\A'lOCITCA~~ 1nIton4 392 EXOH *5 (171 bp) 562 In\ton 5 - an bp -150 CO:~3,lIJ;;;;q~~m:m::'PGl>£'OxrGll£!~'IljcNf'roI\-KGeiX.'I'C'CIffi3i!\GT'Af:?,1,ro"A E47 ClEBP PHAS€ 1 -90 a:ro:a\CA(~:KO\CrI(rr_CA AP-l AP-4 AP-' ELK.' in!ron5 563 EXON 116 (158 bp) 720 intron 6 -952 bp -30 TAI\(nro:mrr::roccACMO::xm::.x::.I'\G\GA G:'l'~~+30 inIr<n6 721 ., in!ron7 877 inlrnn8 973 1nIton11 1392 ... ecce i1t .... 13 1857 ·r ... CCTCAG eXON #7 (156 bp) EXON ,a (96 bpI EXON #9 (90 bp) EXON #10 (231 bpI EXON #13 (210 bp) EXON #14 (194 bpI EXON #15 (162bp) EXON #16 (451bp) 876 inlron 7 - 001 bp 1636 inlron 12 - 400 bp GAGT ... 2613 lions with cDNA sequences specific for the 51 and 3' regions of the coding sequence of Pactolus (Chen et aI. 1998). This screening identified ]4 total distinct phage, each possessing approximately 15 kb of genomic insert. Overlap analysis indicated all 14 were specific for Pactolus and did not represent any other related gene sequence. A subset of phage selected from the overlap analysis -1 1+ Fig. 2. Sequence analysis of the Pactolus intronlexon borders. A. The sequence of the 16exons ofmuriDe Pactolus are shown in order, 5' to 3'. The first exon is untranslated, while the second exon contains the initiating ATG, which is in bold and underlined. TIle exon sequences that are shown in the middle dark box represent only part of the sequence, with the exact number of base pairs within each exon noted in the parentheses. The numbering above the exons indicates their position within the cDNA sequence. Aanking introD sequences are shown in the lightly shaded boxes with the donor sites on the right (with the conserved GT sequence) and the acceptor sequences on the left (possessing the conserved [C/T]AG sequence)_ TIle sizes of the introns are shown on the right: they are either exact detenninations based upon sequence analysis or approximate val­ues (denoted with .) based on PCR. U indicates intron size approximated after localization of the exons. The phase of coding is denoted under the donor sites of the introns. Panel B. The first 750 bp 5' of the transcription start site for Pactolus is shown. The + I under the nuc]eotide G indicates the flTSt nucleotide of the mature mRNA. A number of consensus site sequences for transcription factors were located with Matlnspector (http://www.g.ffde/cgi-bin/ matsearch2.p/) (Quandt et aI. 1995). Only those siles with the highest degree of conservation are noted. were utilized for sequencing of the Pactolus gene. As shown in Fig. 1, panel A, together the three overlapping genomic phage encompass the entire genomic sequence of the Pactolus gene, which is approximately 24 kb in size. The Pactolus gene is com­posed of 16 distinct exons, as defined by both PCR probe hybrid­ization and sequence analysis. The various cDNA probes used in 24 1078 KL. Margraf et al.: Pactolus gene structure E'Qil 112 Pir MI GXl'! JPllTlGl SI ES IQ MlGPHSL 113 Pir N rocnsmv;:mm:ra'NPOCJlWPKI, IQ LL.l\LIIGLFFI.G~093~ #4 Pac NF'l!JRGEPrfMlOJ IK!TJ5EYI/W!fKSp'?nUSfRfPRQISPRN\lIVFIiBP IQ ~LLLKG:::P~YlRP #5 Pac CUATF'KVDffiBTOINj\!TILYI=lm SJSlIIEi!.<jNIIIJlJroquQlu1f]SR$RI IQ ~IlLTIl.M:!..SY~ 116 Pac GFGSIVWT---------------------------FWII!KI~BKQI/'Il$KIATPKWIDffiMJIlATq/ IQ ~PPFAFRHIlLKL~~ 117 Pac GEIGlIltrnEFJNI j\lIINIFHIllKrIm G'IRlI1l'SImJ!T 1JXi1VRS!J7EP IQ EEIGiRN'lIRLLVFA!n::tx~rnFJ>G:GKLGlrn .. :I'I;li[x~:m.~ 1/8 Pac tM:S\.M)I.1ISK! .lID)t!IQI>I'F'II\lPl)HM [l2 ~QPIFAVOO<MI1I\':NE 119 Pir KI.,TI'FTPKIrrtG& SUJ5S1:JIlNX.IENII.¥S IQ KL'IEIIPK.S1'.~ 1110 Pac KI.ssIV\lWHSTIPSII~ [l2~~lN'W '11 1112 1113 Pir ~~~C!1ffi\1QpprCBKBFQYFHIlCSSYX IQ~~~~ #14 Pir OCNO.Jq'NS1)I'I5rnF.J~~. &l~~ Fig. 3. Intronlexon homology between murine Pacto)us and murine integrin subunit 132. The coding sequences of Pacto)us and ~2 integrin are shown from exons 2 to ) 6. The fuJI intronlexon map of the murine 132 integrin gene has not been published; exons are inferred from comparison with the intronlexon junctions of the human f32 integrin gene (Weitzman et aL 1991; Wilson et aL 1993). Sequences have been adjusted to give the greatest degree of alignment. TIle gapped sequence of Pactolus in exon 6 represents the sequences within the MIDAS domain that are lacking in the Pactolus gene (compared with otber members of this gene family). #15 Pac vr;>.EISNITIJ.Icmy:N!.JAVIFl.JN"iQ:fJ'lI l!ImIW.Kf.Pf!!'(t;JN fQ El:'III'3:.~GVULVIW!O\l:lHL~ 1116 this and following analyses of Pactolus are shown in Fig. I, panel B. The specificity of the Pactolus cDNA probes. and thus the uniqueness of the phage clones, was confinned by probing mouse genomic DNA with the three Pactolus cDNA probes. All three of the probes identified genomic fragments similar in size to those predicted from the Pactolus genomic phage clones (data not shown). Of note, probe 881, which was derived from the Cys-rich region of the Pactolus sequence and has considerable sequence homology to the 112 integrin. also faintly detected fragments pre­sumed to be derived from the murine 112 integrin gene (data not shown). The coding regions within the genomic phage were determined by sequencing the phage DNA or plasmid subclones. As shown in Fig. 2, panel A, those sequences contained within the mature Pac­tolus transcript are derived from 16 exons of varying size. The first intron contains only 5' untranslated sequence. The initiating ATG is present within the second exon, which encodes the predicted signal sequence. The MIDAS-like domain region is included within exons 5 and 6. Exon IS encodes the transmembrane domain and a region of the cytoplasmic sequence. Exon 16 encodes the remaining cytoplasmic sequence as well as the full 3' untranslated region. The sizes of the exons and introns as well as the splicing phase are shown in Fig. 2, panel A. The 5' end of the Pactolus transcript maps to the first exon, which is located approximately 6.8 kb upstream of the second exon. The original Pactolus cDNA sequence contained the fun 5' untranslated region in that RACE analysis of the 5' region of Pactolus transcripts confirmed the guanine residue (GC­TATAA .... ) as the first base of the mature transcript (data not shown). As shown in Fig. 2, panel B, the putative promoter se­quence of the Pactolus gene upstream of the first exon is unusual in that it does not appear to possess a TAT A box to specify tran­scription initiation. However, at base -30 is aT AA sequence that may serve to help load the transcriptional apparatus. A number of possible transcriptional control sequences were found in this pu­tative promoter region. These include an ELK-I site at -46, two AP-I sites at -56 and -80, an AP-4 site at -58, a ClEBP site at -117, and an E47 site at -140. Additionally, a strong consensus site for NF-KB binding is present at -175, multiple putative Ikaros sites, and two GATA sequences. Most noteworthy of these latter sites is the GATA site at -560. The consensus sequence for the binding of GATA family members is (Aff)GATA(NG) (Evans and Felsenfeld 1989; Tsai et aI. 1989) which matches the Pactolus sequence ofTGATAG. While there are a number of GATA tran­scription family members, GATA-I might be suspected to playa central role in Pactolus expression. GATA-I is expressed only in hematopoietic derived cells including erythroid, megakaryocytic, mast cell and eosinophil lines, as well as in muItipotential pro­genitors (Orkin 1992). Therefore, the expression pattern of GA TA-I is consistent with the expression profile of Pactolus (Chen et aI. 1998, data not shown). Also of interest are the putative NF-KB and lkaros sites. NF-KB regulation of Pactolus expression would suggest its level of transcription can be induced with vari- R.L. Margraf et al.: Pactolus gene structure ous immunological or inflammatory stimuli (May and Ghosh 1998). Ikaros is a transcription factor preferentially expressed in lymphoid lineages (Kim et al. 1999; although it is also found in non-lymphoid cells such as mast cells: Y. Chen and J.H. Weis unpu blished data). Regulation via Ikaros would be consistent with the observed bone marrow· specific expression pattern seen for Pactolus. The Pactolus promoter region is similar to that of the 132 in­tegrin subunit in that the first intron of 132 integrin is also quite large, the first ex on does not encode protein sequences, and tran­scription initiation does not appear to make use of a TATA box sequence (Rosmarin et al. 1992; Weitzman et al. 1991). The simi­larity of the Pactolus gene to the 132 integrin gene is also striking with regards to ex on domain boundaries. When the splice sites between PactoIus and 132 integrin are aligned, the two utilize vir­tually identical junctions (Fig. 3). This conservation breaks down where there is little homology between Pactolus and 132 integrin, at exons 15 and 16. The gap shown in the coding region of exon 6 represents a sequence that is lacking in the MIDAS-like domain of Pac to Ius compared with the other 13 integrin subunits. Thus, the absence of a portion of the MIDAS region from the Pactolus transcript is not a result of alternative splicing, but instead repre­sents an apparent loss within the genomic sequence of Pactolus. Chromosomal location of the Pactolus gene. The high level of homology between Pactolus and the 132 integrin suggests they were derived from a common ancestor gene. Thus, it was of in­terest to determine whether the two genes were linked within the genome. FISH analysis was performed with three different Pacto­Ius phage clones, displayed in Fig. I, to probe mouse chromosome spreads (Fig. 4). All three clones mapped the Pactolus gene to the terminus of Chr 16, band C4. This is a different location from murine 132 integrin, which maps to murine Chr 10 (MacDonald et al. 1991). However, both the Pactolus and 132 integrin regions represent sequences syntenic to human Chr 21 (MacDonald et al. 1991). This raises the possibility that the two genes were linked during mammalian evolution but separated during rodent speciation. Identification of the Pactolus C transcript. As previously de­scribed, bone marrow is the tissue in the adult mouse with the highest expression level of Pactolus (Chen et al. 1998). Transcript analysis of mature marrow identified two distinct forms of Pacto­Ius which are derived from alternative splicing. One of these tran­scripts encodes a transmembrane form (full length) of Pactolus (Pactolus A), while the second encodes a smaller, truncated form (Pactolus B). The Pactolus B form appears to be unstable and to have a very short half life (S. Garrison and J.H. Weis, unpublished data). Since the fetal and neonatal liver is a site of hematopoietic cell development, we screened the liver for Pactolus transcripts. In comparison with adult liver, the neonatal liver did possess signifi­cant quantities of Pactolus transcripts (data not shown). We then screened neonatal liver cDNA samples, via RT-PCR, with a vari­ety of oligonucleotides (Fig. I B) which WOUld, in the entirety of the overlaps, include the entire Pactolus coding sequence. As shown in Fig. 5, virtually the same level of Pactolus transcription is evident in neonatal liver as adult marrow (based upon actin equivalence). While the majority of the bands between the samples are the same size (indicating identical transcript products), a unique band of 400 bp was evident in the neonatal liver sample with the oligo set 842 and 848. This band was excised from the gel, reamplified, cloned, and sequenced. As shown in Fig. 6, the se­quence from this insert matched that of exon II and exon 13, with an apparent loss of 37 bp from exon II and an insertion of 43bp 25 1079 A B Fig. 4. Chromosomal localization of the murine Pactolus gene. Fluores­cent in situ hybridization (FISH) was perfonned with Pactolus phage clones 5' Phage #1, #2, and 3' Phage #2, shown in Fig. lAo The two FISH panels are the same probed karyotype, the top panel is a DAP} stain, followed by a chromosomal banding pattern in the lower paneL The two red dots show the position of hybridization of the Pactolus probes. While only the data for 5' Phage #1 are shown, all three probes did hybridize to the tenninal region of murine Chr 16 at band C4 (arrows). 5' of exon 13. Exon 12 was not present in this sequence. This product, dubbed Pactolus C, would result in a protein with altered reading frame from the full length protein, thus creating a trun­cated product (Fig. 6). From the apparent instability of the Pactolus B product, the Pactolus C protein would also be expected to be short lived. Discussion The Pactolus gene is unusual in specifying three distinct tran­scripts, two of which would result in truncated products. As shown 1080 859 900 826 828 803 816 807 815 1384428 1 837 804 804 814 26 R.L. Margraf et aI.: Pactolus gene structure NA NANANANANANA MW 622 527 404 & . • .... 'l'l¥' 242/238 217 Exon 11 Exon 12 G F M D S V T V R V L P L eE C Q C Q E QrIT£J\'ItT..a.'I"ICJV3I'("'.&\GIVruill\ IIQ;Q; a.l\Om~ parA cn::TTCA-ro::;A~ Pace Q S Q H H 5 LeG G ~ GAM E C G I C R OCAGr.mr1t:~.,N!T!lQT~nr~]t'I'(plG pacl\ ~ Pact: eNS G Y A G K NeE COT 0 G P 5 5 Q GI\T~-.8P.AA¥'1ETr..Af7I!TI"hC-lQC!!.CIpYX!jlyT.Pa:"C1'G:i D LEG S C R K D N 5 S I M C 5 G L G D C leG Q C E C H T S DIP N K ElY G Q Y 8n:rrmnroczrmrr.mrrroc~~ C E C D N V NeE R Y D G Q v eGG P crrrr...l\C'j"Il'"!~K..8TA1IiAmxrMI 1\ !ljI IJ1.IH!I S PG C L QED £ 5 W Q P R L F S R K R P L L L Exon C G ReF CRY 5 F V GSA C Q C R M 5 T 13 GT<n:C'G81\LIIIIG!\MIIPCb,&!IQ>Jlili1# !!~ PacA PacC WPM L L 5 L Q L R G L 5 L P V P D V H S GeL N N R M VEe 5 G H G R eye N TK'NH'"ffi'IrII'...ycMCJI£r..A'I'f!'IDr~..A'l'trl'fCJID BKA ~~PacC F R L SEQ Q D G G V Q W P W 5 M L L • Fig. 6. Sequence and alignment of the Pactolus C transcript. The nucleo­tide sequence derived from the Pactolus C PCR fragment (PacC) shown in Fig. 5 is compared with the analogous sequence from the full-length Pac­tolus transcript (PacA :underlined) from exon II to exon 13. Amino acids encoded by the Pactalus A sequence are denoted above the PacA sequence, while the unique coding sequences of the PacC sequence are shown below the PacC sequence. Where there is coding unity between the PacA and Pac C sequence, only the PacA amino acids are shown. The Pactolus C se­quence splices out of ex on 11 early using an alternative donor site (con­served GD, skips exon 12, and utilizes an alternative acceptor site within intron 12. The alternative splicing changes the reading frame for the Pac­tolus C producl, resulting in a premature tennination of the protein, de­noted with • . in Fig, 7, panel A, the three Pactolus proteins predicted from the A, B, and C transcripts would all contain the same extracellular do­mains encoded in the first 10 exons. At exon II, the C form utilizes a distinct donor sequence not used by the A or B forms, Fig. S. Identification of the Pactolus C transcript. cDNA derived from RNA obtained from neonatal (5-day·old) liver (N) and adult bone marrow (A) was equalized for actin transcript content and then amplified via PCR with series of oligonucleotides positioned at intervals over the full coding sequence of Pactolus. The position of the oligonucleotides is shown in Fig. I, panel B. Molecular weight markers are shown on the left. The alternative spliced Pactolus C product is shown as a 400-bp band within the N lane of oligo's 8421848 (denoted by the box) and is denoted with the arrow on the right and an * by the specific band. and skips over exon 12 to an acceptor site in intron 12, 43 bp upstream of exon 13. The B fonn splices as the A fonn until ex on 13, where the B fonn utilizes a distinct acceptor site present 37 bp into exon 13 (Fig. 7, panel B). Both B and C transcripts produce products whose reading frames vary, owing to the alternative splic­ing of the primary transcripts, from each other as well as the native· Pactolus sequence. The alternative splice variants would be ex­pected to result in truncated proteins since both the B and C fonns generate stop codons within the exon 13 sequence. This analysis of the Pactolus gene has provided a variety of infonnation. We have been able to compare the structure of the Pactolus gene with that of the other reported /3 integrin genes. As previously reported, the nucleotide and deduced amino acid se­quence of Pactolus is closely related to that of /32 integrin (Chen et aL 1998). Both genes appear to have non-TATA box promoter regions and possess virtually identical exonlintron junctions. These data thus suggest that these two genes are derived from a common precursor sequence. In contrast, however, Pactolus and 132 integrin reside at two distinct chromosomal locations. It will be of interest to determine whether the human /32 integrin and Pactolus genes are linked, or whether their chromosomal separation occurred prior to rodent/primate differentiation. The organization of the Pactolus gene has also shed under­standing on the nature of the Pactolus splice variants. These vari­ants are not rare products: the ratio of the Pactolus B to the A fonn is about 10 to I in adult mouse marrow, while that of C to A in the neonatal liver sample is approximately I to 2. It is interesting to note that the production of such altered transcripts centers upon the recognition of splice sites within a confined region of the gene (i.e., exons II, 12, and 13). The structure of the B and C variants suggests the production of secreted, albeit unstable, proteins while the A form would be held within the membrane. Although these data help to explain how such variants are made, they do not address why they are produced. Experimentation to address that question is under way. Acknowledgments, This research was supported by National Institutes of Health (NIH) grants AI-42032 and AI-24158 (J,H. Weis), an award from the American Lung Association (J.H. Weis), NIH grants AI-3223 and AR-43521 (JJ. Weis), and funds from the Center of Excellence in Hema­tology, grant DK-49219, This work was also supported by the Huntsman Cancer Institute and the National Cancer Institute grant 5 P30 CA-42014. 27 R.L Margraf et al.: Pactolus gene structure 1081 A 100 bp Pactolus A Pactolus B Pactolus C B Pactolus B Pactolus A Fig. 7. Alignment and generation of the three alternatively spliced forms of Pactolus. Panel A. The cDNAs of the full-length (Pactolus A) and truncated Pactolus (Pactolus B and C) are shown divided into their 16 exons. The solid shaded area between the dashed lines is translated sequence, while the white area is untranslated. The extracellular (E-C), transmembrane (T), cytoplasmic (C), and the MIDAS-like (MIDAS) domains are shown below the cDNA (based on homology with ~2 integrin). In Pactolus B the sequence of exon 13 is partially lost owing to alternative splicing. Pactolus C is shown with a shortened exon 11 sequence, no exon 12, and an insertion of part of intron ] 2. Codon phases of selected introns are labeled within the white circles to show the alternative reading frames (compared with the full-length Pactolus A transcript) of the Pactolus B and C products. Forms B and C truncate early within exon 13 owing to their alternative reading frames, which possess multiple translation termination codons. All three fonns apparently utilize the same transcriptional start and stop sequences. Panel B. Diagram demonstrating how alternative splicing gives rise to the three forms of Pactolus transcripts. The only known alternatively spliced regions of Pactolus, exons It through 13, are shown. Exons are in bold, with the deletions shaded. The intronic sequence incorporated in the Pactolus C transcript is striped. Pactolus C References Anoyo AG, Yang JT, Rayburn H, Hynes RO (1996) Differential require­ments for alpha4 integrins during fetal and adult hematopoiesis. Cell 85, 997-1008 Bajt ML, Loftus JC (1994) Mutation of a ligand binding domain of beta 3 integrin. Integral role of oxygenated residues in alpha lib bela 3 (GPUb­IlIa) receptor function. J BioI Chern 269, 20913-20919 Chen Y, Gamson S, Weis 11, Weis JH (1998) Identification of pac to Ius, an integrin beta subunit-like cell-surface protein preferentially expressed by cells of the bone marrow. J BioI Chern 273, 8711-87 I 8 Evans T, Felsenfeld G (1989) The erythroid-specific transcription factor Bryf1: a new finger protein. Cell 58,877-885 Kim J, Sif S, Jones B. Jackson A, Koipally J et al. (1999) Ikaros DNA­binding proteins direct formation of chromatin remodeling complexes in lymphocytes. Immunity 10,345-355 MacDonald G, Chu ML, Cox DR (1991) Fine structure physical mapping of the region of mouse chromosome 10 homologous to human chromo­some 21. Genomics 11, 317-323 May MJ, Ghosh S (1998) Signal transduction through NF-kappa B. Im­munol Today 19,80-88 Michishita M, Videm V, Arnaout MA (1993) A novel divalent cation­binding site in the A domain of the beta 2 integrin CR3 (CDllblCDI8) is essential for ligand binding. Cell 72, 857-867 Orkin SH (1992) GATA-binding transcription factors in hematopoietic cells. Blood 80, 575-581 Quandt K, Frech K, Karas H, Win gender E, Werner T (1995) Matlnd and Matlnspector: new fast and versatile tools for detection of consensus matches in nucleotide sequence data. Nucleic Acids Res 23, 4878--4884 Rieu P, Ueda T, Haruta I, Sharma CP, Amaout MA (1994) The A-domain ofbeta 2 integrin CR3 (CDllb/CDI8) is a receptor for the hookwonn-derived neutrophil adhesion inhibitor NIP. J Cell BioI 127, 2081- 2091 Rosmarin AG, Levy R, Tenen DG (1992) Cloning and analysis of the CD18 promoter_ Blood 79, 2598-2604 Springer TA (1990) Adhesion receptors of the immune system. [Review] [165 refs]. Nature 346, 425-434 Tan SS, Weis JH (1992) Development of a sensitive reverse transcriptase PCR assay, RT-RPCR, utilizing rapid cycle times. Genome Res 2, 137-143 Tsai SF, Martin DI,ZonLl, D'Andrea AD, Wong GG et al. (1989) Cloning of eDNA for the major DNA-binding protein of the erythroid lineage through expression in mammalian cells. Nature 339, 446-451 Ueda T, Rieu P, Brayer J, Arnaout MA (1994) Identification of the comple­ment iC3b binding site in the beta 2 integrin CR3 (CD1Ib/CDl8). Proc Nat Acad Sci USA 91, 10680-10684 Wagner N. Lohler J, Kunkel EJ, Ley K, Leung E et al. (1996) Critical role for beta7 integrins in formation of the gut-associated lymphoid tissue. Nature 382, 366--370 Wardlaw AJ, Hibbs ML, Stacker SA, Springer TA (1990) Distinct muta­tions in two patients with leukocyte adhesion deficiency and their func­tional correlates. J Exp Med 172, 335-345 Weitzman 18, Wells CE, Wright AH, Clark PA, Law SK (1991) The gene organization of the human beta 2 integrin subunit (CDI8). FEBS Len 294,97-103 Wilson RW, O'Brien WE, Beaudet AL (1989) Nucleotide sequence of the eDNA from the mouse leukocyte adhesion protein CDI8. Nucleic Acids Res 17, 5397 Wilson RW, Ballantyne CM, Smith CW, Montgomery C, Bradley A et al. (1993) Gene targeting yields a CDt8-mutant mouse for study of inflam­mation. J Immunol lSI, 1571-1578 Yang JT, Rayburn H, Hynes RO (1995) Cell adhesion events mediated by aIpha 4 integrins are essential in placental and cardiac development. Development 121, 549-560 CHAPTER 3 P ACTOLUS DEFICIENT MICE DEMONSTRATE AN ALTERED TNF-a RESPONSE FOLLOWING NEUTROPHILIMACROPHAGE INTERACTIONS Submitted 2003 29 Abstract Pactolus is a 13 integrin-like subunit expressed by neutrophils. The majority of the Pactolus protein resides within the cell, requiring cellular activation for translocation to the surface. To define the function of Pac to Ius, a Pactolus null mouse was generated. Pactolus deficient animals mature appropriately although some progeny develop spontaneous skin lesions opportunistically infected with Staphylococcus aureus. Pactolus null mice possess normal numbers of neutrophils, display appropriate migration into sites of inflammation and combat introduced infections efficiently. When Pactolus deficient neutrophils were incubated with macrophages, an increased level ofTNF-a was released compared to the wild type cells. These data suggest neutrophils interact with macrophages via Pactolus to regulate TNF -a production. Introduction Pactolus is a protein predominantly expressed by maturing and mature mouse neutrophils 1 • Pactolus genetic structure and protein sequence is very similar to the 6 integrin subunits (particularly the 62 and 137 integrins)2,3; however, there are a number of significant differences. First, Pactolus does not appear to have the metal ion dependent adhesion site (MIDAS)3 domain required for integrinlligand recognition4 , suggesting Pactolus may recognize its ligand(s) by a different mechanism than integrins. Second, Pactolus has a deletion of the specificity determining loop_ This deficiency allows Pactolus to be stably expressed on the cell surface without requiring formation of a heterodimer with an a chain 1,2,5. Third, while the extracellular regions of the proteins show a high level of similarity, the transmembrane and cytoplasmic domains of Pactolus are quite distinct from the integrin subunit products6 ,7. So, Pactolus function( s) in the neutrophils could be distinct from the integrins despite sequence and gene structure similarities. The pactolus gene is transcribed in mouse neutrophils at all stages of maturity. From 1 to 10% of the total Pactolus protein is found on the cell surface with the rest sequestered within intracellular granulesl . Treatment of neutrophils with inflammatory signals such as that provided by PMA and leukotriene B4 (L TB4) induces the release of intracellular Pactolus to the cell surface1(unpublished data). 30 Although the structure and expression profiles of Pactolus provide clues as to its function in the biology of the animal, its role in the inflammatory innate immune response of th~ animal has only been conjectural. Therefore, we created a mouse strain deficient in Pactolus to define its role( s) in neutrophil functions. Elimination of the Pactolus product did not affect neutrophil development, neutrophil recruitment to inflammatory sites or the function of the neutrophil during infection. Instead we found that the loss of Pactolus influenced the macrophage response to neutrophils by altering TNF -a production. Materials and Methods Mice. C57BL/6 and 129/sv mice were from Jackson labs. Mice were older than 6 weeks for use in experiments. The mice were age and sex matched for each experiment. The mice were handled according to NIH protocols. Mice were anesthetized or sacrificed with isoflurane inhalation. Primers (sequence listed 5' to 3 '). Actin control for equivalent cDNA: primer #62 - GTAACAATGCCATGTTCAAT, primer #339 - CTCCATCGTGGGCCGCTCTAG. Southern probe PCR primers: primer #1047f - TCCCAAGCAGCTGCCCTTCTG, primer #860r - TTTGCTCCCTGGAGGCCTTG, primer #1008r - TGGCCATCATATCTCTCACAG, primer #1231 f - TGCAACTCTGGCTACGCTGG. Genotyping primers: primer #863f - CTCTGGCTCTGCGCAAGGCC, primer #1052f - CGCTCGATGTTCAGCCCAAGC, primer #1028r - CGATTCGGCCTGACCTGGAG. pactolus PCR primers: primer #1001f - CAAGGACAATGTTGAGCACCTG, #900r - CCGCTCGAGACAGCTGCCTCTGGTCCCTC, #803f - CAAAGTGGACTTCCAGCGGACCCAG, #828r - GAGTCCGCAGTCAGCTTCAG, #1010f - GCTTTGGATCCATTGTGAAC, #1004r - TAGGTTTTCACCATCCTTGAG. 31 Targeting construct. Two EcoRl fragments of the pactolus gene (8.4 Kb and 6.7 Kb) were subcloned into SK vector (Stratagene) from lambda phage containing pieces of the pactolus gene2 ,3. The Bg12 site of ex on 5 in the 8.4 Kb subclone was changed to a Cla1 site by cutting with Bgl2 and inserting a double-stranded oligo containing the CIa1 site. The KT3 Lox A plasmid that contained the neomycin gene under the control of the 32 Polymerase II promoter was cut with CIa!, the neomycin fragment was gel purified then inserted in the reverse orientation into exon 5's new CIa! site. The new plasmid containing 15.1 Kb of the pactolus gene interrupted by the neomycin gene was linearized by SaIl and Spe 1, then ligated into the TK 1-TK2C (thymidine kinase) vector Xho 1 and Xbal sites. The resulting plasmid was linearized with Notl for transfection into the ES cell line. Generation of the Pactolus deficient mouse. ES cell lines were transfected with the linear pactolus gene targeting construct then selected for G418 resistance and gancyclovir sensitivity. DNA was isolated from the ES cells and analyzed by Southern blot after digestion with EcoRI (5' probe) or Sstl (3' probe). The Southern probes were made by PCR of the either the lambda phage DNA (for the 5' probe - exon 1, primer #1047 & #860) or pactolus cDNA (for the 3' probe - exon 12, primer #1008 & #1231). An ES cell line containing one copy of the pactolus targeted allele, 2e#6, was injected into blastocysts and implanted in pseudo-pregnant female mice. The one male chimera offspring was mated to C57BL/6 mice. The Fl generation, heterozygous for the modified pactolus allele (+1-) was mated to generate the F2 population. The F2 mice homozygous for the wildtype pactolus allele (+1+) were mated to generate the wild-type mice used in this study, and the F2 mice homozygous for the disruptedPactolus allele (-1-) mice were mated to produce the Pactolus deficient mouse lineage. Tail DNA isolation and genotyping. Tail clips (1 cm) were added to 1 ml tail digestion buffer (0.1 M NaCI, 0.02 M EDTA, 1% SDS and 0.01 M Tris pH 8) with 0.2 mg proteinase K and incubated at 550C overnight. After the debris settled, 500 J.lL of supernatant was transferred to a new tube followed by phenol and chloroform extraction, 33 and ethanol precipitation. Nine ng of DNA was used per PCR reaction with the genotype primers (1 Jlg total/ 10 JlL reaction): #863, #1028, and #1052. (Primers 1,2, and 3 in figure 1 are primers #863, #1028, and #1052, respectively.) Genotyping PCR reactions used 680C annealing, 5 sec elongation and 28 cycles. Primers #863 f and # 1 028r made the exon 5 product, size 95 bp. Primers #1052fand #1028r made the neomycinlexon 5 product, size 130 bp. RNA isolation, cDNA synthesis and polymerase chain reaction (peR). Cells were isolated from murine bone marrow with PBS and total RNA was extracted8 • cDNA synthesis reactions were as described9 . The PCR reaction were prepared as described9 . The 95°C denaturing, 72°C elongation, 1 sec for annealing and denaturing times were kept constant, while the °c annealing, elongation time, number of reaction cycles, and amount of DNA (about 200 ng for cDNA and 9 ng for genomic DNA) per reaction varied with experiment. The actin PCR (product size 135 bp) required 60°C annealing, 5 sec elongation and 15 cycles, while the pactolus PCR required 60°C annealing, 4 sec elongation and 25 cycles. The pactolus PCR product with primers # 1001 f and #900r (which span exon 3 to 4, before the neomycin) was 223 bp, primers #803f and #828r (exon 4 to 6, over neomycin insertion) was 215 bp, primers #1010f and #1004r (exon 6 to 8, after the neomycin insertion) was 407 bp. 1m munoprecip ita tionlWestern. Bone marrow cells (2.5 X 107 cells per lane) were isolated with RPMI (Gibco), washed with PBS, erythrocytes were lysed with ACK (0.15 M NH4CL, 1.0 mM KHC03 , 0.1 mM EDTA), and the cells were washed again. The immunoprecipitation and western blot analysis was virtually identical as that previously described 1 • 34 Murine peripheral blood leukocyte isolation. Mice were anesthetized with isoflurane and blood was isolated into a heparinized syringe by cardiac puncture. Blood was collected and combined from 1,2 or 4 mice per experiment (n=5): for comparison between experiments the total cells were equalized to 2 mice. Heparin sodium salt (Sigma) was resuspended in 1 mL PBS, enough anti-coagulant for 5 mL of blood. Blood was mixed 1:1 with 20/0 Dextran solution (IX PBS, 10 mM Sodium azide and 20/0 Dextran T500 from Pharmacia Biotech) and incubated at 37°C for 30 minutes, which allows the erythrocytes to aggregate and settle out of solution. The upper phase was collected, brought to 2mL with PBS and centrifuged 1000 RPM for 5 minutes. The pellet was resuspended in ACK for 3 minutes, layered on 0.5 mL FCS and centrifuged at 1000 RPM for 5 minutes. The cells were resuspended, counted, and used for flow cytometry. Flow cytometry. Cells were stained, washed and analyzed using the Becton­Dickinson FACScan. Antibodies: anti-B220, -Grl, -CDI9, -c-kit, -F4-80, -CD4 and­CD8 (all from PharMingen). The neutrophil population was determined by positive staining cells for Grl and negative staining for F4/80. Sterile and infectious peritonitis. Wild type and Pactolus deficient mice were injected with 1 ml of 1 % Oyster glycogen or IX PBS into the peritoneal cavity. The peritoneal cells were isolated by cold PBS lavage after 5, 12, and 18 hour time points. Cells were analyzed by flow cytometry. Untreated mice and mice injected with IX PBS at different time points yielded the same results and are combined as "PBS" control. E.coli was grown in 4 ml sterile tryptic soy broth (Difco) overnight at 37°C in a candle jar, 1 ml was transferred to a new 4 ml tryptic soy broth tube and incubated for 4 more hours. The bacteria were washed with sterile 1 X PBS three times, then resuspended in 5 35 ml PBS. OD600 of 1 ml of bacteria culture was determined the approximate CFUs (an OD600 reading of 1.31 contained approximately 9X 1 08 CFU/ml). The E coli was diluted to 2Xl 07 CFU/ml and 500 JlI (IXI 07 CFUs) injected into the peritoneal cavity. Mice were monitored for survival over 10 days and were sacrificed if moribund (shaky, tipping over). The precise number ofCFUs injected (0.7-3 X 107 ) was determined by serial dilution of the bacterial culture and plating on blood agar plates to be incubated at 37°C overnight. Toxic dermatitis. Ear inflammation was initiated with 2% (vol/vol) DNFB (dinitrofluorobenzene - Sigma) in olive oil and analyzed as described previouslylo. Qualitative neutrophil recruitment was examined by microscopy. Cytokine assays. Supernatants were assayed for TNF-a content after 24hrs co­culture by sandwich enzyme-linked immunosorbent assay (ELISA) using paired antibodies. TNF-a levels were determined by using a biotinylated detection antibody and avidin-HRP. Values were obtained by comparison to data for recolnbinant TNF-a standards. Mouse macrophages. Murine macrophages were derived from femur and tibia bone marrow from C57BL/6 mice. Briefly, bone marrow cells were cultured in RPMI supplemented with L929-conditioned medium for 7 days at 37°C. Macrophages were recovered with ice-cold PBS and re-plated in 24-well culture dishes at a density of 2 x lOS/well in serum-free medium containing 1% Nutridoma (Roche, Indianapolis, IN). After overnight incubation at 37°C, nonadherent cells were removed, and 8x1 05 neutrophils in I ml serum-free medium, supplemented with 100ng/ml of purified LPSll 36 and 10 units/ml of INF -y (R&D Systems). After 24 hours incubation supernatants were harvested and TNF -u assayed by ELISA. Results Creation of a Pactolus deficient animal. A pactolus targeting construct was prepared for homologous disruption of the endogenous gene3 • The neomycin gene cassette was inserted into exon 5 of the gene (Figure 3.1a), the construct was then inserted into the TK-containing vector, linearized and electroporated into ES cells12 • The G418 resistant ES cell lines containing the targeted pactolus allele yielded a 12 Kb Sst 1 digest fragment compared to the wild type 12.7 Kb Sstl fragment with a 3' probe in a Southern blot (Figure 3.1b). The ES cells were also tested by EcoRl digest and a 5' probe to confirm appropriate recombination (data not shown). Approximately 50% of the ES cell lines analyzed contained the targeted allele. Targeted ES cell line, 2e#6 (Figure 3.1c), was injected into blastocysts and implanted in a C57BL/6 pseUdo-pregnant female. One male chimera was utilized to generate the germline transmission. F1 heterozygote animals (pactolus +/-) were mated to generate the wild type (pactolus +/+) mouse lineage and the homozygous targeted pactolus allele (pactolus -/-) mouse lineage used in this study. A PCR based assay was developed (Figure 3.1d) to easily discriminate between the various genotypes generated in such matings. Transmission rates for the first 95 offspring from such heterozygous matings were 30(+/+), 49(+/-) and 16(-/-). Litter sizes between wild type and pactolus null matings have been similar. Therefore, it appears that the pactolus null mutation does not appear to affect the viability or fertility of the mice. 37 Figure 3.1. Generation of a Pactolus null mouse. (a) Diagram of the pactolus gene3 . The neomycin cassette was inserted into Bgl2 site in exon 5. The two genomic phage subclones of the pactolus gene that were used to make the pactolus targeting construct are shown (8.4 and 6.7 Kb fragments). (b) The wild-type pactolus allele is aligned to the pactolus targeting construct and the expected recombination is shown (targeted allele). lfthe ES cell line DNA has recombined with the pactolus targeting construct, the size of the genomic DNA fragments after Sst 1 restriction enzyme digestion would decrease by 0.7 Kb. The probes used for Southern blot analysis of the ES cell lines are shown on the targeted allele. The 3' probe detected the Sst 1 digestion difference while the 5' probe detected the EcoRl digest difference. (c) Southern blot of the Sst! digest of the ES cell lines hybridized with the 3' probe. The first lane is an ES cell line that was recombination negative, the second lane was recombination positive and the third lane was mouse liver DNA control. The targeted ES cell line, 2e#6 (shown in 2nd lane), was used to make the pactolus null mouse. (d) Genomic DNA was isolated from mouse tails and the neomycin insertion in the targeted allele detected by PCR using the three primers shown. Primers 1 and 2 produced the exon 5 product with the wild-type allele and no product from the modified allele. Primers 2 and 3 produced the neomycinlexon5 (neo) product with the targeted allele DNA. An example ofPCR with DNA fronl a homozygouspactolus targeted mouse (-1-), a wild-type pactolus allele (+1+) and a heterozygous mouse (+1-) is shown. Primers 1,2, and 3 are primers #863, #1028, and #1052, respectively. a pacto/us Genomic Structure 1 Kb 6.7 Kb Subclone 8.4 Kb Subclone Bgl2 ~1--------i-lI-n ---lil-I----I --JIB----Ill----i-I--t--I I I I I 23 4 5 6 7 8 9 10 11 12 13 14 b Wild-Type pacto/us Allele E I Targeting Construct Targeted pacto/us Allele E if' c 12.7Kb 12 Kb ES cell Genomic -) IWT allele restricdon fragment - 12.7 Kb d E E I I I Exon 5 E I Targeted allele restriction fragment - 12 Kb ,E (-/-) (+/+) (+/-) Neo Exon5 Exon5 1... .....2. --- Exon 5 38 •• 15"16" .3. .......2 Exon 5 39 The homozygous pactolus targeted mice were analyzed for pactolus gene products (Figure 3.2a). Bone marrow mRNA isolated from wild type (+1+), heterozygous (+1-) and homozygous pactolus targeted (-1-) mice were utilized for RT-PCR. Three sets of pactolus primers were used: one set was before the neomycin insert (sequences from exon 3 to exon 4), another set flanked the neomycin insert (sequences from exon 5 to exon 6) and the third set was after the neomycin gene (sequences from exon 6 to exon 8). As controls for pactolus expression, mRNA from Pactolus negative mucosal mast cells (MMC .. derived with IL-3 from bone marrow) and Pactolus positive immature connective tissue-like mast cells (CTMC - derived with SCF) were utilized2 . Animals that possessed two targeted pactolus alleles clearly did not produce transcripts 3' of the neomycin insertion and would thus be unable to create a membr~ne bound fonn of the protein. Such animals were also analyzed for Pactolus protein production by imnlunoprecipitation Iwestern blot analysis (Figure 3.2b) and flow cytometry (not shown). The Pactolus protein was immunoprecipitated from bone marrow cells with anti-Pactolus monoclonal antibody and subsequently identified via western blot using a polyclonal anti-Pactolus sera (Bac Pac )1, The Pactolus protein can be found in two fonns, depending on the amount of glycosylation, in total bone marrow samples. These two fonns of Pactolus were detected in the wild type mice (+1+), while no Pactolus protein was detected in the pactolus (-I .. ) mice. Gross anatomical analysis of the Pactolus deficient animals (liver, bone marrow, lung, small intestine and spleen) did not show any difference between these animals and the wild type littennates (data not shown). On occasion, skin lesions were observed on the neck, ears and face of the homozygous deficient animals: similar lesions were not 40 Figure 3.2. Analysis of pactolus expression in the deficient animals. (a) RT-PCR for pactolus transcripts. Bone marrow mRNA was isolated from wild type (+1+), heterozygous (+1-), and two pac talus null (-1-) mice. RT-PCR was achieved using primers derived from the pactolus sequence before the neomycin insertion (from exon 3 to 4), over the neomycin gene (from exon 5 to 6), and after the neomycin insertion (from exon 6 to 8). The pactolus negative MMCs and the pactolus positive CTMCs are also shown as controls. Actin primers were used for equivalent loading. (b) Immunoprecipitation (IP)/western analysis for Pactolus. Bone marrow cells were isolated from wild type (+1+) andpactolus null (-1-) mice, lysed, and immunoprecipitated with either a Pactolus monoclonal antibody or an isotype control. The precipitated protein was detected by western blot with the Pactolus polyclonal rabbit anti-sera (post­immune sera). The two predominant forms of high molecular weight Pactolus (l30,OOOmr) and low molecular weight Pactolus (98,OOOmr) are shown. a or- (\J ~ ~ -'+ -I -". -I + + I I ......., ......., ......., ......., 0 0 2 2 2 5 Exon 3/4 Before Neo Exon 5/6 Over Neo Exon 618 AfterNeo actin b HMW Pactolus ----.... LMW Pactolus ----.... 41 Pactolus ~noclonal lsotype Ab IP Control IP ( -1-) (+1+) (- 1- )(+1+) 42 seen in heterozygote or wild type animals. Histological analysis of the affected skin showed a massive influx of neutrophils, with edema and some areas of necrosis and fibrosis. Staphylococcus aureus, an opportunistic member of normal skin flora was isolated from such lesions. Skin lesions were previously observed in mice lacking the 62 integrin subunitlO , and those lacking multiple selectin proteins 13 although these lesions were notable for the lack of infiltrating neutrophils. Neutrophil development and migration is not impaired with Pactolus deficiency. The analysis of the Pactolus deficient animals was then expanded, focusing upon possible defects in neutrophil development and function. Since Pactolus is expressed by maturing and mature neutrophils in the wild type bone marrow, and in myeloid cell lines (32Dc13 and MPRO)l (data not shown), we investigated whether the Pactolus deletion could inhibit neutrophil development or the release of mature neutrophils into the periphery. We isolated bone marrow cells from wild type (+/+) andpactolus null (-/-) mice and analyzed the leukocyte cell populations by flow cytometry (Figure 3.3a), utilizing antibodies for B cells, neutrophils and progenitor cells. The percentages of leukocyte populations were equivalent between such mice. Additionally, neutrophil populations were isolated from bone marrow cells by FACS, cytospun onto a slide, and Wright stained to analyze their morphology (data not shown). The Pactolus deficient animals possessed both immature and mature neutrophils in identical ratios to those found in wild type animals. Thus the production of bone marrow neutrophils is not influenced by the pactolus null mutation. This analysis was then extended by isolating peripheral blood leukocytes from wild type (+/+) and pactolus null (-/ -) mice, followed by their subset 43 analysis by flow cytometry (Figure 3.3b). Again, there was not a remarkable difference in blood leukocyte populations between the wild type and deficient animals. These data vary from those obtained from the 62 integrin deficient animals that demonstrated elevated neutrophils counts in the blood14 ,15. If Pactolus is not involved in marrow development/retention, could it be important for neutrophil migration into sites of inflammation? The occasional skin inflammation observed with the Pactolus null mice could have been due to delayed or absent neutrophil recruitment similar to that shown for the animal deficient in the 62 integrin sUbunitlO • Migration of neutrophils into the skin was evaluated by treating mouse ears with 2% di­nitro- fluro-benzene (DNFB) for 5 hours followed by histological analysis (Figure 3.4a)16, Neutrophils were detected in the ears of Pac to Ius deficient animals and demonstrated similar recruitment kinetics as was observed with treated wild-type ears. Recruitment of neutrophils into the peritoneal cavity of the Pactolus deficient animals was also evaluated. Interestingly, while the integrin subunit 62 deficient animal demonstrated undetectable neutrophil recruitment to the skin, the migration into the peritoneal cavity (and the lung) of such mice was normal, leading to the suggestion that a 62 independent adhesion mechanism is utilized for neutrophil migration into those sites10 ,14. Could Pactolus playa part in such a pathway? We chose to analyze the Pactolus deficient animals for both the kinetics of recruitment as well as determining total cell numbers. Neutrophils were recruited into the peritoneal cavity by the injection of 10 mg oyster glycogen 17, isolated, counted, and then analyzed by flow cytometry (Figure 3 .4b). We observed similar nUlnbers of neutrophils between wild type (+/+) and pactolus null 44 Figure 3.3. Distribution of bone marrow and blood leukocytes in Pactolus deficient animals. (a) Bone marrow cells were isolated from wild type [pac (+1+)] (n=5 mice) and Pactolus null mice [pac (-1-)] (n=4) and stained with antibodies for B cells (B220), neutrophils (Gr-l high), and leukocyte progenitors (c-kit) for flow cytometry. (b) Blood leukocytes were isolated from wild type (+1+) and Pactolus null mice (-1-) as described in Materials and Methods. Leukocytes were stained with antibodies for B cells (CD19), neutrophils (Gr-l high), monocytes (F4/80) and T cells (CD4+CD8) and detected by flow cytometry. Each circle represents a separate experiment using two mice each (n=5 experiments, while monocytes and T cells, n=3 and 2, respectively). Data are expressed as total cells (x 105 ), calculated by multiplying % cell population by total cells. 45 a 50 45 D pac ("1-) 40 • pac (+1+) 35 8. 30 ~ 'i 25 (,) 0~ 20 15 10 5 0 Bcalls Prog en itors b 14 0 12 0 10 - e IJ'!I Q ,.. 0 cs. 8 .'i! _6.75 -6.80 t) 6 ~ 0 I- 4 0 0 0 0 0 0 0 .. ·2.28 0 2 _ 2.1 0 0 ~1.68 0 0 -1.47 0 fo.81 ~1.03 e -S- 0.54 -@- 0.81 0 0 (-1-) (+1+) I (-f..) (+1+) (-1-) (+1+) I (-1-) (+1+) (-1-) (+1+) blood blood neutrophils neutrophils BeeUs Beells monocyte monocyte Teens Tcells total cells total cells Cell types 46 Figure 3.4. Recruitment of neutrophils in Pactolus deficient animals. (a) Toxic dermatitis. Wild type [pac (+1+)] and Pactolus deficient [pac (-1-)] mouse ears (n=5 mice each) were treated with 10 J.llof2% DNFB for 5 hours before sacrifice. Sample histology slides stained with hematoxylin and eosin are shown at two different magnifications. C - cartilage, E - epidermis, V - venule, N - neutrophils. (b) Sterile peritonitis. Pactolus deficient [pac (-1-)] and wild type [pac (+1+)] neutrophils were recruited to the peritoneal cavity with oyster glycogen. Leukocytes were isolated by peritoneal lavage at different time points: 5 (n=2), 12 (n=10) and 18 hour (n=6) and stained with antibodies for B cells (CDI9), neutrophils (Gr-l high), and macrophages (F4/80). Data is expressed as total neutrophils (Xl05 ), calculated by multiplying % cell population by total cells isolated. PBS is the average of no injection or PBS injection (n=4) for 12 hours, which yielded the same results. 47 A. pac (+1+) pac (-1-) 48 b 25 o pac ( .. / .. ) neutrophils • pac (+/+) neutrophils ..- LO ,0.. . 20 ->< .~ .r: C- -o 15 ..- :::s <J.) s:::: (.0.i- j 10 I- 5 0+----- PBS 5 12 18 Hours after oyster glycogen injection 49 (-1-) animals recruited at the different time points tested. These analyses were also carried out using thioglycollate or 30 mg oyster glycogen as the recruiting agent15 . Again, no difference was detected in neutrophil recruitment between the Pactolus deficient and wild type animals (data not shown). The recruited neutrophils were analyzed by flow cytometry for possible differences in expression of L-selectin, B2 integrin and CD 11 b (adhesion molecules that could compensate for the loss of Pactolus if this protein was important for neutrophil recruitment). Again no differences were observed (data not shown). These data, in total, suggest Pactolus is not required for migration of neutrophils into inflamed tissues, and that Pactolus is not part of the proposed B2-integrin independent pathway for peritoneal migrationl4 • Pactolus deficient animals do not show susceptibility to bacterial infections. The preceding data suggested that Pactolus is not essential for either neutrophil development or migration, but may be critical for the resistance to bacterial infection since the majority of the intracellular form of Pactolus is released to the cell surface after neutrophil activation] (unpublished data). This model was tested utilizing a peritonitis infection n10del with E.coli. Wild type, pactolus null, 129/sv and C57BL/6 mice were injected intraperitoneally with approximately lXl07 CFU E.coli. The pactolus null (-1-) mice fared the same as the wild type strain (+1+) and the two other mouse strains (Figure 3.5). This model was also tested in a similar experiment with Staphylococcus aureus (1 x 108 CFU per animal), which resulted in no apparent difference in resistance to the bacteria between the mouse types (data not shown). When mice are depleted of neutrophiIs by Gr-l antibodies during a S. aureus skin infection, they can develop poorly healed, crusted skin lesions 18. Our isolation of S. 50 Figure 3.5. Infectious response by Pactolus deficient animals. Pactolus wild-type [Pac (+I+)](n=46 mice), Pactolus null [Pac (-I-)](n=44), 129/sv (n=8) and C57BL/6 (n=9) mice were injected with approximately 1 X 107 CFU (0.7-3 X 107 exact CFU) E.coli into the peritoneal cavity. Approximate CFUs were determined by OD600 of the bacteria culture, while exact CFUs were determined by plating bacteria dilutions on blood agar plates. 51 120 -1]-, Pac (+/+) --*-. Pac (-/-) 100 --+-. 129/sv ~ C57BU6 80 ~ .~ 60 ::s en ~0 40 20 0 1 Days after infection 52 aureus from Pactolus deficient animal skin lesions suggested a similar sensitivity. However, using an established skin infection model with S. aureus, we saw no difference in the recovery of the Pactolus null and wild type mice (data not shown). The mechanism of formation of the anecdotal, spontaneous lesions sometimes found on the neck, ears and face of the Pactolus deficient animals, and from which S. aureus was isolated, is not known but is presumably distinct than simply the introduction of the bacterial strain (for example, spontaneous inflammation versus active infection). Pactolus deficient neutrophils enhance TNF-a release by macrophages. Macrophages follow neutrophils into the site of inflammation where they are instrumental in clearing both necrotic and apoptotic neutrophils. In a separate manuscript, we have shown that macrophages interact with neutrophils via Pactolus and an unknown macrophage receptor (Garrison et al.: submitted for publication). This Pactolus­mediated interaction appears to enhance phagocytosis of apoptotic neutrophils. Phagocytosis of necrotic cells is known to help macrophages take on an inflammatory phenotype while the uptake of apoptotic neutrophils cells is known to down regulate such a response 1 9. Therefore we sought to determine if the interaction of recruited neutrophils with macrophages could affect the macrophage cytokine profile, perhaps as an explanation for the regions of spontaneous inflammation seen in some of the Pactolus deficient animals. To test if the production of macrophage cytokines can be modulated through Pactolus binding, bone marrow derived macrophages were incubated with LPS and IFN-y for 24 hours in the presence ofneutrophils obtained from the 129/sv, C57BL/6 strains, wild type (+/+), and Pactolus deficient animals (-/ -). Supernatants were then analyzed for 53 the release of cytokines. As shown (Figure 3.6a), the level of TNF-a release was dramatically higher in cultures containing the Pactolus deficient neutrophils versus those from the wild type and inbred strains of animals. One of the principle cytokines associated with inflammatory macrophages is TNF-a. Similar analyses were done for IL-IO, TGF-B and IL-6: no change in the level of these cytokines was evident in these culture systems (data not shown), suggesting the Pactolus receptorlPactolus ligand interaction is specific for the modulation ofTNF-a. The TNF-a released in the co-cultures could be derived from two different sources: the neutrophil or the macrophage. To evaluate this concern, this co-culture experiment was performed in parallel with the analysis of the neutrophils themselves. As shown (Figure 3.6b), minimal TNF-a was released into the supernatant when macrophages were activated with LPS and IFN-y alone, or when the neutrophils isolated from the wild type or Pactolus deficient animals were incubated alone with LPS/IFN-y. But again elevated quantities ofTNF-a were released when the Pactolus deficient neutrophils were incubated with the activated macrophages compared with the TNF-a released from the macrophages incubated with the wild type neutrophils. These data suggest that the neutrophils were not the source of preformed TNF -a and that the released TNF-a was produced by the macrophages. Discussion This manuscript has documented that mice lacking Pactolus do not possess any obvious neutrophil developmental phenotype or sensitivity to infections. The immunological literature is rife with examples of the absence of an obvious phenotype 54 Figure 3.6. Modulation of TNF -a production by Pactolus deficient neutrophils. (a) Peritoneal neutrophils from 129/sv, C57BL/6, wild type (+/+), and Pactolus deficient (-/-) mice, were incubated 24 hours with bone marrow derived macrophages cultured with 100ng/ml purified LPS and 10 units/ml INF -y. After incubation, supernatants were immediately assayed for TNF-a content by ELISA using triplicate wells and determining the error by standard deviation. The control macrophage monolayer treated with only LPS and IFN-y is shown as LPS. (b) Peritoneal neutrophils from wild type littermate (-:1-/+), and Pactolus deficient (-1-) mice, were cultured as above with bone marrow derived macrophages (BMDM) or alone (neutrophils only). LPS represents control macrophages incubated with LPS and INF-y, without neutrophil addition. Data shown from both panels are derived from distinct experiments and are representative of a number of similar analyses. 55 a 1800 1600 1400 t$ uI. . 1200 Z l- 1000 E 800 """'- 0') c: 600 400 200 0 LPS 129/sv C57BL/6 (+/+) (-/~ ) b 1600 1400 1200 ts I u.. 1000 Z I- aoo E """'- 600 0') c: 400 200 0 LPS BMDM/(+I+) BMDM/(-I-) (+1+) (-1-) 56 with the deletion of a protein for which there are other redundant, functional analogues. Only when single cell behavior is analyzed, or when multiple null mutations are bred into a single animal, is a phenotype detected. The same has been shown for Pactolus in that the null animals appear normal, but the in vitro macrophage response to purified neutrophils lacking Pactolus is not. Our finding that the absence of Pactolus binding leads to an increased synthesis ofTNF-a by activated macrophages suggests a specific signaling pathway is affected. Obviously, one of the compelling questions that the data presented in this manuscript raises is what is ( are) the receptor on the surface of macrophages with which Pactolus interacts, and how does its ligation suppress the macrophage TNF-a response? The function of such an interaction must be intimately entwined with the neutrophil life cycle. Neutrophils have short life spans in the periphery of the animal. At the end of this time line, such cells still in circulation undergo apoptosis and are removed by the phagocytic macrophages of the liver. Neutrophils recruited to the tissues must also be disposed of in a similar fashion. Macrophages follow the neutrophils into a site of infection. They are required to phagocytize both necrotic cells/cell fragments, as well as the apoptotic neutrophils that have accumulated at such a site20 ,21. The removal of apoptotic neutrophils is an important step in the resolution of the inflammatory response22 . Savill and Fadok describe "eat me" flags that apoptotic cells display to facilitate phagocytosis23 . Such flags include exposure of phosphatidylserine usually maintained on the inner membrane, and changes in sugar content on the cell surface that are recognized 57 by phagocytic lectins. A number of mouse neutrophilligandlmacrophage receptor pairs have been described, to varying degrees of specificity, that can facilitate neutrophil removat24 - 33 • Pactolus possesses a peptide core with N-linked sugar modifications that are extensively decorated with sialic acid residues. Additionally, the bulk of the Pactolus protein is held within the neutrophil and is not translocated to the cell surface until the cell is activated by soluble inflammatory mediators or upon apoptosis. Thus Pactolus could serve as an "eat me" flag facilitating the uptake of apoptotic neutrophils by macrophages. Our unpublished data (Garrison, et al; submitted) indicate that the phagocytosis of apoptotic neutrophils by macrophages is decreased in the absence of Pac to Ius. The data in this manuscript suggest that during neutrophil/macrophage interaction, that the ligation of Pactolus to its putative receptor on the surface of macrophages limits the production and release ofTNF-a by the macrophage. A number of studies have demonstrated that the uptake of apoptotic cells by macrophages does confer upon the macrophage an anti­inflammatory phenotype34 ,35, part of which is the limited release ofTNF-a. We hypothesize that macrophages engulfing apoptotic neutrophils would respond with an inflammatory agenda (with regards to TNF-a release) except for the inhibitory effect of Pactolus binding. Deducing this Pactolus-dependent pathway that controls such a macrophage response will be a major focus of future experimentation. Acknowledgements The authors would like to thank the faculty and staff of the University of Utah core facilities, especially that of the Homologous Knockout and Transgenic facility, for 58 their assistance in this work. We would also like to thank all of the members of the Weis' laboratories for their robust discussions and dissection of the work. References 1. Garrison, S., Hojgaard, A., Patillo, D., Weis, J. J. & Weis, J. H. Functional characterization of pactolus, a {beta} Integrin like protein preferentially expressed by neutrophils. J Bioi Chem 18, 18 (2001). 2. Chen, Y., Garrison, S., Weis, J. J. & Weis, J. H. Identification of pac to Ius, an integrin beta subunit-like cell-surface protein preferentially expressed by cells of the bone marrow. Journal of Biological Chemistry 273, 8711-8 (1998). 3. Margraf, R. L., Chen, Y., Garrison, S., Weis, J. 1. & Weis, 1. H. Genomic organization, chromosomal localization, and transcriptional variants of the murine Pactolus gene. Mamm Genome 10, 1075-81 (1999). 4. Humphries, M. J. Integrin activation: the link between ligand binding and signal transduction. [Review] [59 refs]. Current Opinion in Cell Biology 8, 632-40 (1996). 5. Takagi, J., DeBottis, D. P., Erickson, H. P. & Springer, T. A. The role of the specificity-determining loop of the integrin beta subunit I-like domain in autonomous expression, association with the alpha subunit, and ligand binding. Biochemistry 41, 4339-47 (2002). 6. Vuori, K. Integrin signaling: tyrosine phosphorylation events in focal adhesions. J Membr Bioi 165, 191-9 (1998). 7. Hauck, C. R., Hsia, D. A. & Schlaepfer, D. D. The focal adhesion kinase--a regulator of cell migration and invasion. IUBMB Life 53, 115-9 (2002). 8. Chirgwin, J. M., Przybyla, A. E., MacDonald, R. J. & Rutter, W. J. Isolation of biologically active ribonucleic acid from sources enriched in ribonuclease. Biochemistry 18, 5294-9 (1979). 9. Tan, S. S. & Weis, J. H. Development of a sensitive reverse transcriptase PCR assay, RT-RPCR, utilizing rapid cycle times. Genome Research 2, 137-43 (1992). 10. Scharffetter-Kochanek, K. et al. Spontaneous skin ulceration and defective T cell function in CD18 null mice. J Exp Med 188, 119-31 (1998). 59 11. Hirschfeld, M., Ma, Y., Weis, J. H., Vogel, S. N. & Weis, J. J. Cutting edge: repurification of lipopolysaccharide eliminates signaling through both human and murine toll-like receptor 2. J Immunol165, 618-22 (2000). 12. Bunting, M., Bernstein, K. E., Greer, J. M., Capecchi, M. R. & Thomas, K. R. Targeting genes for self-excision in the germ line. Genes & Development 13, 1524-8 (1999). 13. Frenette, P. S., Mayadas, T. N., Rayburn, H., Hynes, R. O. & Wagner, D. D. Susceptibility to infection and altered hematopoiesis in mice deficient in both P­and E-selectins. Cell 84, 563-74 (1996). 14. Mizgerd, J. P. et al. Neutrophil emigration in the skin, lungs, and peritoneum: different requirements for CD ll1CD 18 revealed by CD 18-deficient mice. J Exp Med 186, 1357-64 (1997). 15. Walzog, B., Scharffetter-Kochanek, K. & Gaehtgens, P. Impairment of neutrophil emigration in CDl8-null mice. Am J Physiol276, G1125-30 (1999). 16. Goebeler, M., Gutwald, J., Roth, J., Meinardus-Hager, G. & Sorg, C. Expression of intercellular adhesion molecule-l in murine allergic contact dermatitis. Int Arch Allergy Appl Immunol93, 294-9 (1990). 17. Yoshikawa, T., Takano, H., Yoshida, N. & Kondo, M. Simultaneous intraperitoneal administration of OK-432 and serum enhances superoxide generation from migrated polymorphonuclear leukocytes, with special emphasis on the role of complements. Immunopharmacol Immunotoxicol17, 265-82 (1995). 18. MoIne, L., Verdrengh, M. & Tarkowski, A. Role of neutrophil leukocytes in cutaneous infection caused by Staphylococcus aureus.lnfect Immun 68,6162-7 (2000). 19. Haslett, C. Resolution of acute inflammation and the role of apoptosis in the tissue fate of granulocytes. Clin Sci (Lond) 83, 639-48 (1992). 20. Newman, S. L., Henson, J. E. & Henson, P. M. Phagocytosis of senescent neutrophils by human monocyte-derived macrophages and rabbit inflammatory macrophages. J Exp Med 156, 430-42 (1982). 21. Savill, J., Fadok, V., Henson, P. & Haslett, C. Phagocyte recognition of cells undergoing apoptosis. Immunol Today 14, 131-6 (1993). 22. Malech, H. L. & Gallin, J. 1. Current concepts: immunology. Neutrophils in human diseases. N EnglJ Med317, 687-94 (1987). 60 23. Savill, J. & Fadok, V. Corpse clearance defines the meaning of cell death. Nature 407, 784-8 (2000). 24. Fadok, V. A., Warner, M. L., Bratton, D. L. & Henson, P. M. CD36 is required for phagocytosis of apoptotic cells by human macrophages that use either a phosphatidylserine receptor or the vitronectin receptor (alpha v beta 3). J Immunol 161, 6250-7 (1998). 25. Savill, 1., Hogg, N., Ren, Y. & Haslett, C. Thrombospondin cooperates with CD36 and the vitronectin receptor in macrophage recognition ofneutrophils undergoing apoptosis. J Clin Invest 90, 1513-22 (1992). 26. Fadok, V. A. et al. Different populations of macrophages use either the vitronectin receptor or the phosphatidylserine receptor to recognize and remove apoptotic cells. J Immunol149, 4029-35 (1992). 27. Ashman, R. F., Peckham, D., Alhasan, S. & Stunz, L. L. Membrane unpacking and the rapid disposal of apoptotic cells. Immunol Lett 48, 159-66 (1995). 28. Fadok, V. A. et al. The ability to recognize phosphatidylserine on apoptotic cells is an inducible function in murine bone marrow-derived macrophages. Chest 103, 102S (1993). 29. Fadok, V. A. et al. Exposure of phosphat idyl serine on the surface of apoptotic lymphocytes triggers specific recognition and removal by macrophages. J Immunol148, 2207-16 (1992). 30. Fadok, V. A. et al. Particle digestibility is required for induction of the phosphatidylserine recognition mechanism used by murine macrophages to phagocytose apoptotic cells. J Immunol151, 4274-85 (1993). 31. Ramprasad, M. P. et al. The 94- to 97 -kDa mouse macrophage membrane protein that recognizes oxidized low density lipoprotein and phosphatidylserine-rich liposomes is identical to macrosialin, the mouse homologue of human CD68. Proc Natl Acad Sci USA 92, 9580-4 (1995). 32. Sambrano, G. R. & Steinberg, D. Recognition of oxidatively damaged and apoptotic cells by an oxidized low density lipoprotein receptor on mouse peritoneal macrophages: role of membrane phosphatidylserine. Proc Natl Acad Sci USA 92, 1396-400 (1995). 33. Ramprasad, M. P., Terpstra, V., Kondratenko, N., Quehenberger, O. & Steinberg, D. Cell surface expression of mouse macrosialin and human CD68 and their role as macrophage receptors for oxidized low density lipoprotein. Proc Natl Acad Sci USA 93, 14833-8 (1996). 61 34. Fadok, V. A. et al. Macrophages that have ingested apoptotic cells in vitro inhibit pro inflammatory cytokine production through autocrine/paracrine mechanisms involving TGF-beta, PGE2, and PAF. J Clin Invest 101, 890-8 (1998). 35. Meagher, L. C., Savill, J. S., Baker, A., Fuller, R. W. & Haslett, C. Phagocytosis of apoptotic neutrophils does not induce macrophage release of thromboxane B2. J Leukoc Biol 52, 269-73 (1992). CHAPTER 4 PACTOLUS, A NOVEL PROTEIN INVOLVED IN THE CLEARANCE OF APOPTOTIC NEUTROPHILS Submitted 2003 63 Abstract Pactolus is a cell surface protein expressed by murine neutrophils. Pactolus is similar to the 13 integrins except it lacks a functional MIDAS domain and is expressed without an a chain partner. The majority of the Pactolus protein is held within the cell in granules as a highly glycosylated form and is released to the surface following inflammatory activation, or ligation of Pac to Ius on the cell surface. Neutrophil activation studies suggest that Pactolus does not serve as an activating or phagocytic receptor for the neutrophil but instead is expressed on the surface as a ligand for a macrophage receptor. Neutrophils lacking Pactolus demonstrate a loss of efficient phagocytosis by macrophages suggesting Pactolus serves as a phagocytic ligand for macrophage uptake of apoptotic neutrophils. Introduction Pactolus is a 13-integrin-like protein found on mature and immature murine neutrophilsl . It exhibits close homology, in the extracellular region, to the 32 and 37 integrin sUbunits2 . However, unlike the 32 integrin subunit, Pactolus is missing the metal ion-dependent adhesion site (MIDAS) domain that is important for ligand binding and subunit interactions3 - 7 . All integrins are heterodimers consisting of a and 3 subunits. An a chain partner for Pactolus has not been found suggesting Pactolus is expressed as a monomer on the cell surface. The predicted size of the full length Pactolus peptide is 81,OOOmr; however, extensive glycosyl modifications provide for a final mature protein of approximately 130,OOOmrl . 64 The Pactolus gene produces two distinct transcripts, one that predicts a truncated form and a second that produces the transmembrane protein. The truncated form appears to be rapidly degraded upon translation (if indeed it is translated) while the transmembrane form has a long half life (greater than 12 hours). The neutrophil is the major, if not exclusive, mature cell that transcribes the Pactolus gene and produces the protein. Two distinct alleles of Pactolus exist: that of the C57BL/6 animal specify production for only the transmembrane form while those ofBALB/c, C3H1HeJ and 129/sv produce both the full length and truncated transcriptsl . Stimulation of neutrophils with PMA or Pactolus anti-sera can induce an immediate increase of Pactolus surface expression independent of protein synthesis. This suggests that Pactolus is stored in an intracellular compartment and, upon stimulation, is exocytosed to the surface of the neutrophil. This type of response is similar to that of other neutrophil activation receptors such as human CR1 and human/mouse Mac-I, both of which serve as neutrophil phagocytic receptors8 - lO . N eutrophils playa critical role during an inflammatory response since they are the first major cell type recruited to a site of infection (for reviewll ). When an infection occurs, neutrophils migrate from the blood to the tissue site. This process involves activation of both the endothelial cell layer as well as the circulating neutrophils, relying upon the selectins and integrins (and their respective ligands) for extravasations12 . The selectins are instrumental for the tethering and rolling of the neutrophils on the endothelial cells while the integrins are involved in tight adherence and extravasations into the tissue matrix13 • The neutrophils then follow a gradient of cytokines, chemokines and bacterial/viral products to the site of infection. When neutrophils arrive at the focus 65 of infection, they are in a primed state and can follow several options to clear the infection. They can degranulate, releasing proteases, cytokines and superoxides, creating an environment inhospitable for the infective agent. The primed neutrophil can also phagocytose the pathogen, usually making use of complement and antibody deposited on the pathogen surface as opsonins for the relevant neutrophil receptors14. Many such recruited and activated neutrophils are in excess and consequently undergo apoptosis and are removed from the site via phagocytosis by macrophages15. This cleanup step can make use of a number of receptor/ligand interactions between the neutrophil and macrophage (some well characterized, some not)16-19 removing both apoptotic and necrotic cells2o• Our previous analyses of Pactolus suggested it would function as an activating receptor for neutrophils, perhaps aiding in cell migration and/or phagocytosis of pathogens. It was with this model in mind that we analyzed the expression of Pactolus by migrating neutrophils and the cellular responses to Pactolus ligation. Surprisingly we found that Pactolus did not appear to be involved in any neutrophil function including migration, crosslinking activation or phagocytosis. Thus the opposite perspective was investigated: that Pactolus serves as a binding ligand to enh