CHARACTERIZATION  OF  EQUINE  ALVEOLAR  MACROPHAGE  PHENOTYPES  IN  RECURRENT   AIRWAY  OBSTRUCTION   By   Margaret  Eilidh  Wilson   A  DISSERTATION   Submitted  to   Michigan  State  University   in  partial  fulfillment  of  the  requirements   for  the  degree  of   Comparative  Medicine  and  Integrative  Biology  –  Doctor  of  Philosophy   2014   ABSTRACT   CHARACTERIZATION  OF  EQUINE  ALVEOLAR  MACROPHAGE  PHENOTYPES  IN  RECURRENT   AIRWAY  OBSTRUCTION     By   Margaret  Eilidh  Wilson   A  central  feature  of  recurrent  airway  obstruction  (RAO)  is  an  enhanced  sensitivity  to  hay   dust  (HD)  compared  to  control  horses,  but  the  cellular  and  molecular  mechanisms   accounting  for  this  differential  airway  sensitivity  are  unknown.  The  microbial  components   of  HD  can  stimulate  pathogen  recognition  receptors  (PRRs),  which  are  abundant  on   alveolar  macrophages  (AM).  Thus  inhaled  HD  would  be  expected  to  evoke  an  inflammatory   response  from  AM,  however,  work  in  other  species  indicates  that  the  nature  of  the   response  depends  on  the  macrophage  phenotype.  Two  major  polarized  phenotypes,   classified  by  their  distinct  gene  expression  and  functions,  have  been  described.  The   classically-­‐activated  M1  exerts  a  robust  pro-­‐inflammatory  response  and  the  alternatively-­‐ activated  M2  produces  anti-­‐inflammatory  cytokines,  and  little  in  the  way  of  pro-­‐ inflammatory  cytokines.  Thus,  our  overarching  hypothesis  was  that  the  HD   hypersensitivity  observed  in  RAO  may  be  mediated  by  the  presence  of  pro-­‐inflammatory   (M1)  or  the  absence  of  immune-­‐regulatory  (M2)  phenotypes.     There  are  limited  data  characterizing  equine  M1  and  M2  phenotypic  markers,  thus   to  address  this,  equine  AM  were  cultured  in  IFNγ+LPS  or  IL-­‐4  generating  equine  M1  and  M2   phenotypes  respectively,  and  the  gene  expression  of  predicted  surrogate  M1/M2  genes  and   the  effect  of  polarization  on  the  response  to  pro-­‐inflammatory  agonists  was  evaluated.   Equine  M1s  were  characterized  by  increased  expression  of  pro-­‐inflammatory  genes  (TNFα,   IL1β,  IL-­‐12p40,  IL-­‐8,  CD80),  regulatory  IL-­‐10  and  a  potentiated  pro-­‐inflammatory  gene   expression  response  when  stimulated.  Equine  M2s  were  characterized  by  elevated   scavenger  receptor  CD206,  low  expression  of  M1  associated  genes  and  potently  suppressed   pro-­‐inflammatory  cytokines  and  IL-­‐10  when  stimulated.  Further,  this  study  determined   that  canonical  murine  macrophage  markers  iNOS  and  arginase  were  not  M1  or  M2   associated  in  the  horse.         Next  the  expression  of  equine  M1  and  M2  transcriptional  markers  and  the  cytokine   response  to  HD  components  was  assessed  in  AM  from  RAO-­‐susceptible  and  control  horses   prior  to  (at  baseline)  and  following  exposure  to  hay.  These  results  determined  that   different  AM  phenotypes  exist  in  RAO-­‐susceptible  and  control  horses  at  baseline:  RAO-­‐ susceptible  horses  had  greater  gene  expression  of  IL-­‐10  and  enhanced  responsiveness  to   LPS  stimulation  suggesting  an  M2-­‐like,  immune  regulatory  phenotype  that  maintains  LPS   responsiveness.  Further,  exposure  to  hay  induced  different  phenotypes  in  both  groups:  AM   from  control  horses  expressed  elevated  IL-­‐1β,  IL-­‐8,  IL-­‐10,  CD206,  and  developed  enhanced   responsiveness  to  LPS  indicating  a  mixed  M1/M2  phenotype.  In  RAO-­‐susceptible  horses,   hay  exposure  only  increased  the  expression  of  CD206,  an  M2  marker.     In  summary,  these  data  provide  the  first  characterization  of  the  transcriptional   signature  of  equine  AM  M1  and  M2  phenotypes  that  will  assist  in  studying  their  role  in   equine  pulmonary  disease.  Furthermore,  these  data  demonstrate  that  divergent  AM  exist  in   RAO-­‐susceptible  and  control  horses,  indicating  that  the  AM  plays  a  role  in  RAO-­‐ immunopathology. ACKNOWLEDGMENTS   I  would  like  to  thank  my  mentor  Dr.  Ed  Robinson  for  his  advice,  guidance  and  good  humour   during  my  PhD  training.  I  have  learnt  so  much  from  all  of  my  fellow  Pulmonary  Lab   members  that  will  stand  me  in  good  stead.  Thank  you.   iv   TABLE  OF  CONTENTS   LIST  OF  TABLES  ............................................................................................................................................  vii LIST  OF  FIGURES  ...........................................................................................................................................  viii KEY  TO  ABBREVIATIONS  ..........................................................................................................................  ix   Chapter  1.  Literature  Review  ...................................................................................................................  1   Overview  ...........................................................................................................................................................  1   Section  1.  Recurrent  Airway  Obstruction  ..........................................................................................  3   The  Impact  of  RAO  .........................................................................................................................  3   The  Inflammatory  Response  of  RAO  ......................................................................................  4   The  Role  of  an  Antigen-­‐Specific  Response  in  RAO  ...........................................................  5   The  Non-­‐Specific  Response  ........................................................................................................  9   The  Alveolar  Macrophage  in  RAO  ...........................................................................................  13   Section  2.  The  Biology  of  Alveolar  Macrophages  ............................................................................  16   Origins  of  Alveolar  Macrophages  ............................................................................................  16   The  Role  of  the  Alveolar  Macrophage  in  the  Healthy  Lung  ..........................................  17   M1  and  M2  Phenotypes  ...............................................................................................................  20   Utilizing  Gene  Expression  Signatures  to  Identify  Macrophage  Phenotypes  ........  26   Activation  Pathways  of  M1  and  M2  Macrophages  ...........................................................  28   The  Inflammatory  Response  of  M1/M2  to  Subsequent  PAMP  Stimulation  ..........  30   Macrophage  Phenotypes  and  Airway  Disease  ...................................................................  30   Section  3.  Conclusion  and  Hypotheses  ................................................................................................  33   Chapter  2.  Polarized  Equine  Alveolar  Macrophages  Have  a  Species  Specific  Gene   Expression  Profile            ...................................................................................................................................  36   Abstract  ..............................................................................................................................................  36   Introduction  ......................................................................................................................................  37   Materials  and  Methods  ................................................................................................................  39                          Animals  .......................................................................................................................................  39    Collection  of  Hay  Dust  ..........................................................................................................  39    Isolation  of  Alveolar  Macrophages  ................................................................................  40    Cell  Culture  ................................................................................................................................  41      RNA  Extraction,  Reverse  Transcription  and  Quantitative  Real  Time  PCR  ....  41    Data  Analysis  ............................................................................................................................  44                              Results  ................................................................................................................................................  44        Gene  Expression  of  Polarized  Phenotypes    ..................................................................  44                        PAMP  Stimulation  of  Polarized  Alveolar  Macrophage  Phenotypes  ..................  47            Inflammatory  Cytokines  ..................................................................................................  47   Regulatory  Cytokines  .........................................................................................................  49   Surface  Receptors  ................................................................................................................  50   Discussion  ..........................................................................................................................................  51   v   Chapter  3.  RAO-­‐Susceptible  and  Control  Horses  Possess  Different  Alveolar   Macrophage  Phenotypes  ...........................................................................................................................  56   Abstract  ..............................................................................................................................................  56   Introduction  ......................................................................................................................................  57   Materials  and  Methods  ................................................................................................................  59                            Animals  ......................................................................................................................................  59                      Study  Design  .............................................................................................................................  59                          Pulmonary  Function  Tests  .................................................................................................  60                          Isolation  of  Alveolar  Macrophages  ................................................................................  60                        Cell  Culture  ................................................................................................................................  62                        RNA  Extraction,  Reverse  Transcription  and  Quantitative  Real  Time  PCR  ....  62   Data  Analysis  .............................................................................................................................  63   Results  .................................................................................................................................................  64   BALF  Cytology  and  Evaluation  of  Pulmonary  Function    .........................................  64                    Freshly  Harvested  Alveolar  Macrophages  ......................................................................  66            Phenotype  of  Plated  Alveolar  Macrophages  .................................................................  69   Discussion  ..........................................................................................................................................  76   Chapter  4.  Concluding  Discussion  .........................................................................................................  82   APPENDICES  ...................................................................................................................................................  85   Appendix  1.  Selection  of  Reference  Genes  for  Quantitative  Real  Time  PCR  .........  86   Appendix  2.  Characterization  of  β-­‐Glucan  Receptor  Isoforms  in  the  Horse  .........  90   BIBLIOGRAPHY  ..............................................................................................................................................  101   vi   LIST  OF  TABLES   Table  1.  Species  Specific  Differences  in  M1/M2  Polarized  Phenotypes  ...............................  27   Table  2.  Life  Technologies  Assay  ID  for  Proprietary  (A)  and  Custom        Designed  (B)  Taqman  Gene  Assays  .....................................................................................  43   Table  3.  Candidate  Genes  for  M1  (A)  and  M2  (B)  Macrophage  Phenotypes  .......................  46   Table  4.  Effect  of  Natural  Challenge  on  BALF  Cytology  and  Pulmonary  Function  ...........  65   Table  5.  Gene  Expression  in  Freshly  Harvested  Alveolar  Macrophages        in  RAO-­‐Susceptible  Horses  Relative  to  Control  Horses  at  Pasture  and        Following  Natural  Challenge  ..................................................................................................  67   Table  6.  The  Effect  of  Natural  Challenge  on  Gene  Expression  of  Freshly  Harvested        Alveolar  Macrophages  in  RAO-­‐Susceptible  and  Control  Horses  ............................  68   Table  7.  Panel  of  Candidate  Reference  Genes  ...................................................................................  87   Table  8.  Descriptive  Statistics  of  Candidate  Reference  Genes  Based  on  the  Cross    Threshold  Point  (CT)  .................................................................................................................  89   Table  9.  BestKeeper  Analysis  Showing  Correlation  Coefficient  (r)  and  P-­‐Value  (p)        From  Candidate  Reference  Genes  ........................................................................................  89   Table  10.  Location  of  Primers  Used  to  Identify  β-­‐Glucan  Splice  Isoforms  ...........................  98   Table  11.  Tissue  Specific  Expression  of  β-­‐Glucan  Receptor  Splice  Isoforms  ......................  98   vii   LIST  OF  FIGURES   Figure  1.  The  Effect  of  Polarization  on  PAMP  Stimulation  of  Inflammatory    Cytokines  .......................................................................................................................................  48       Figure  2.  The  Effect  of  Polarization  on  PAMP  Stimulation  of  Regulatory  Cytokines  .......  49   Figure  3.  The  Effect  of  Polarization  on  PAMP  Stimulation  of  Surface  Receptors  ..............  50   Figure  4.  Natural  Challenge  Study  Design  ..........................................................................................  60   Figure  5.  Alveolar  Macrophages  From  RAO-­‐Susceptible  and  Control  Horses  Differ  in   Gene  Expression  of  Pro-­‐Inflammatory,  Regulatory  and  Surface  Receptors  at   Baseline  and  Following  Natural  Challenge  .....................................................................  71   Figure  6.  Alveolar  Macrophages  From  RAO-­‐Susceptible  Horses  Exhibit  Enhanced  LPS    Responsiveness  at  Baseline  but  Not  After  Natural  Challenge  ................................  72   Figure  7.  Natural  Challenge  Differentially  Alters  the  Phenotype  of  Alveolar  Macrophages          From  RAO-­‐Susceptible  and  Control  Horses  ...................................................................  75   Figure  8.  Expression  of  β-­‐Glucan  Receptor  Splice  Isoforms  Using  RT-­‐PCR  ........................  94   Figure  9.  Equine  β-­‐Glucan  Receptor  Splice  Isoforms  Determined  by            Cloning  and  Sequencing    ........................................................................................................  96   Figure  10.  Expression  Profile  of  β-­‐Glucan  Receptor  Splice  Isoforms  in  Alveolar              Macrophages  From  RAO-­‐Susceptible  and  Control  Horses  .....................................  99   viii   KEY  TO  ABBREVIATIONS   ARG  1/2   arginase  isoform  1/2   AMcase   acidic  mammalian  chitinase   αVβ6   integrin  (alpha  v  beta  6)   BAL   bronchoalveolar  lavage   BALF   bronchoalveolar  lavage  fluid   βGR   β-­‐glucan  receptor   BGUS   β-­‐glucuronidase   BL   baseline   B2M   β-­‐2  microglobulin   CCL2   chemokine  (c-­‐c)  motif  ligand  2     c/EBP   CCAAT/enhancer-­‐binding  protein   CD-­‐   cluster  of  differentiation   cDNA   complementary  deoxyribonucleic  acid   Cdyn   dynamic  compliance   CHIT-­‐1   chitotriosidase   CHI3L3   chitinase  3-­‐like  3   CHI3L4   chitinase  3-­‐like  4   CLEC7A   C-­‐type  lectin  domain  family  7,  member  A   CT   cross  threshold  point   CTLD   C-­‐type  lectin-­‐like  carbohydrate  recognition  domain   CR   cysteine  rich  domain   ix   CXCL2   chemokine  (c-­‐x)  motif  ligand  2     ΔCT   delta  cross  threshold   ΔPplmax   delta  trans-­‐pulmonary  pressure  (maximum)   EDTA   ethylenediaminetetraacetic  acid   ELF1α   elongation  factor  1α     FIZZ1   found  in  inflammatory  zone  protein  1   FN   fibronectin  domain   GAPDH   glyceraldehyde-­‐3-­‐phosphate  dehydrogenase   HD   hay  dust     HPRT   hypoxanthine  phosphoribosyltransferase   IFNγ   interferon-­‐gamma   IgE   immunoglobulin  type  E   IL-­‐   interleukin   IL-­‐4Rα     interleuking-­‐4  receptor  alpha   iNOS   inducible  nitric  oxide  synthase   IRF-­‐   interferon  regulatory  factor  -­‐   ISRE   interferon  sequence  response  elements   JAK   janus  kinase   LPS   lipopolysaccharide   MACS   magnetic-­‐activated  cell  sorting   MMP-­‐   matrix  metalloproteinase   MIP2   macrophage  inflammatory  protein  2   M1   classically  activated  macrophage   x   M2     alternatively  activated  macrophage   MyD88   myeloid  differentiation  primary  response  gene  (88)   NFκB     nuclear  factor-­‐κB   PAMP     pathogen  associated  molecular  pattern   Pep   peptidoglycan     PI3K     phosphoinositide-­‐binding  protein   PPRγ     peroxisome  proliferator  activated  receptor  gamma   PRR     pathogen  recognition  receptor   qRT-­‐PCR   quantitative  real  time  polymerase  chain  reaction   RAO     Recurrent  airway  obstruction   RELMα/β   resistin  like  molecule  alpha/beta   RIN   RNA  integrity  number     RL   pulmonary  resistance   RNA     ribonucleic  acid   RPMI     Roswell  Park  Memorial  Institute   RT-­‐PCR   reverse  transcription  polymerase  chain  reaction   SD   standard  deviation   SDHA     succinate  dehydrogenase  complex   SOCS-­‐     suppressor  of  cytokine  signaling-­‐   STAT     signal  transducer  and  activator  of  transcription   Th-­‐     T-­‐  helper  lymphocyte   TLR     toll-­‐like  receptor   TNFα     tumor  necrosis  factor-­‐  alpha   xi   TGFβ       UCSC       Zym       transforming  growth  factor  beta   University  of  California  Santa  Cruz   zymosan xii   Chapter  1.   Literature  Review   Overview   A  fundamental  characteristic  of  equine  recurrent  airway  obstruction  (RAO)  is  an   exaggerated  inflammatory  response  to  hay  dust  (HD).  However,  the  exact  mechanisms  that   elicit  this  uncontrolled  inflammatory  response  remain  uncertain.  Hay  dust  contains  a   medley  of  microbial  components  containing  conserved  pathogen  associated  molecular   patterns  (PAMPs)  that  can  stimulate  innate  immune  cells  through  pathogen  recognition   receptors  (PRRs).  Alveolar  macrophages  are  equipped  with  abundant  PRRs  and  inhaled  HD   would  be  expected  to  evoke  an  inflammatory  response.  However,  it  is  clear  from  work  in   other  species  that  the  nature  of  the  response  depends  on  the  macrophage  phenotype,   which  is  itself  governed  by  the  microenvironment  in  which  the  macrophage  resides.  Two   main  phenotypes,  classified  by  their  distinct  gene  expression  and  functions  have  been   described.  The  M1  phenotype  is  particularly  sensitive  to  PRR  stimulation  and  exerts  a   robust  pro-­‐inflammatory  response.  In  contrast  the  M2  phenotype  produces  anti-­‐ inflammatory  cytokines,  little  in  the  way  of  pro-­‐inflammatory  cytokines,  and  can  function   in  tissue  repair.     An  imbalance  of  macrophage  phenotypes  can  be  associated  with  chronic  inflammatory   disease,  however,  it  is  unknown  if  an  imbalance  of  macrophage  phenotypes  contributes  to   the  enhanced  sensitivity  to  HD  that  is  observed  in  RAO.    This  lead  me  to  generate  the   overarching  hypothesis  that  the  enhanced  sensitivity  to  HD  observed  in  RAO-­‐susceptible   horses  could  be  mediated  by  the  presence  of  pro-­‐inflammatory  (M1)  phenotypes  or  the   absence  of  anti-­‐inflammatory  (M2)  phenotypes.  To  address  this  overarching  hypothesis  I   1   first  characterized  the  gene  expression  patterns  of  in-­‐vitro  induced  M1  and  M2  equine   alveolar  macrophage  phenotypes.  Then,  to  investigate  the  role  of  M1/M2  phenotypes  in   RAO  immune-­‐pathology  I  next  evaluated  the  M1/M2  expression  patterns  in  alveolar   macrophages  from  RAO-­‐susceptible  horses  during  disease  remission  and  after  HD   exposure  (disease  exacerbation),  and  compared  the  alveolar  macrophage  gene  expression   patterns  with  that  in  control  horses.     2   Section  1.  Recurrent  Airway  Obstruction   The  following  section  provides  background  detailing  the  inflammatory  response  that   occurs  during  RAO  exacerbation  and  discusses  the  proposed  mechanisms  that  mediate  this   hypersensitivity  response.  Finally  the  role  of  the  alveolar  macrophage  in  RAO  is  discussed.       The  Impact  of  RAO   Recurrent  airway  obstruction  is  a  prevalent,  chronic,  inflammatory  lung  disease  that   affects  adult  horses  and  is  characterized  as  a  hypersensitivity  to  inhaled  HD.  Following   inhalation  of  HD,  RAO-­‐susceptible  horses  develop  clinical  disease  (airway  neutrophilic   inflammation  and  bronchoconstriction)  which  culminates  in  varying  degrees  of  respiratory   distress  and  impaired  performance.1  These  clinical  signs  resolve  when  the  stimulus  (HD)  is   removed  and  susceptible  horses  return  to  a  subclinical  (remission)  state.     The  prevalence  of  RAO  may  vary  by  region,2  but  in  Great  Britain  it  has  been  estimated  to   affect  a  substantial  14%  of  the  general  horse  population.3  There  is  no  cure  for  RAO  and   susceptible  horses  require  life  long  management,  which  fundamentally  requires   permanently  eliminating  exposure  to  hay.4,5  However,  indoor  housing  and  feeding  hay  are   common  equine  husbandry  practices  and  owner  compliance  to  implement  and  maintain   environmental  modifications  is  poor6  resulting  in  recurrent  episodes  of  clinical  disease.7   Although  the  monetary  cost  of  the  disease  to  the  equine  industry  is  not  known,  RAO   presents  a  significant  welfare  issue  as  affected  horses  suffer  reduced  quality  of  life  due  to   recurrent  bouts  of  respiratory  distress  and  chronic  coughing.  Further,  the  resulting   impaired  athletic  performance  results  in  early  athletic  retirement  putting  horses  at   increased  risk  for  surrender  or  premature  euthanasia.2  Critically,  despite  the  prevalence  and     3   impact  of  RAO  there  remains  a  gap  in  our  understanding  of  the  immunologic  mechanisms   that  underlie  this  disease  and  if  novel  therapeutic  targets  are  to  be  sought  we  must  first   understand  the  basic  disease  mechanisms.   The  Inflammatory  Response  of  RAO   Inhalation  of  HD  triggers  an  influx  in  non-­‐septic  neutrophils  and  increased  mucus   secretion  into  the  airways.  This  inflammatory  response  is  accompanied  by   bronchoconstriction  and  a  non-­‐specific  airway  hyperresponsiveness.  Consequently,   affected  horses  have  increased  coughing  and  visibly  labored  breathing  pattern  (expiratory   dyspnea).  Pulmonary  inflammation  and  dysfunction  gradually  resolve  with  avoidance  of   HD  and  during  remission  bronchoalveolar  lavage  (BAL)  cytology  and  pulmonary  function   may  be  indistinguishable  from  normal  horses.     Challenge  with  HD  in  RAO  induces  neutrophil  movement  into  the  airways  within  4   hours.8  These  neutrophils  are  primed9  and  produce  oxidative  products10  which  can   promote  tissue  damage  and  oxidative  stress.11  Further,  neutrophils  can  produce  pro-­‐ inflammatory  cytokines  TNFα,  IL-­‐1β  and  potent  neutrophilic  chemokines  IL-­‐8  and  CXCL2   (MIP-­‐2),  that  may  further  amplify  inflammation.12   A  number  of  chemokines  (IL-­‐8,  IL-­‐17  and  CXCL2)  are  elevated  during  chronic   RAO.13–16  Increased  expression  of  IL-­‐8  and  CXCL2  are  detected  in  BAL  cells  and  respiratory   epithelium.13–15,17,18  The  cellular  source  of  IL-­‐17  is  unclear  and  it  may  originate  from  Th17   lymphocytes  cells  or  neutrophils.13  Importantly,  elevations  of  IL-­‐8  and  IL-­‐17  ensue  after  the   initial  neutrophil  influx  into  the  airway13,18  and  the  exact  mediators/cytokines  that   stimulate  the  early  recruitment  of  neutrophils  following  HD  exposure  remain  unclear.   4   However,  bronchial  epithelial  cell  cultures  from  RAO-­‐susceptible  horses  display  enhanced   up-­‐regulation  of  CXCL2  expression  following  in-­‐vitro  HD  stimulation,  which  may  account   for  the  early  influx  of  neutrophils,  however,  this  has  yet  to  be  verified  in-­‐vivo.19   Exposure  to  hay  (RAO-­‐exacerbation)  induces  increased  expression  of  IL-­‐1β  and  TNFα   13,15,20  IL-­‐1018  and  TGFβ16  but  not  IL-­‐615,16,20  in  BAL  cells.  The  exact  source  of  these   cytokines  is  not  well  established,  as  BAL  is  comprised  of  a  mixture  of  macrophages,   lymphocytes  and  neutrophils.  A  number  of  other  inflammatory  mediators  are  elevated   during  RAO-­‐exacerbation  including  matrix  metalloproteinase-­‐9  (MMP-­‐),21  leukotriene-­‐ B4,22  and  arachidonic  acid  metabolites  15-­‐hydroxyeicosatetraenoic23  and  thromboxane.24   Importantly,  the  role  of  interferon-­‐gamma  (IFNγ)  or  IL-­‐4  and  IL-­‐13,  (Th1  and  Th2   cytokines  respectively),  in  the  pathogenesis  of  RAO  is  unclear  as  there  are  mixed  findings   among  studies  (discussed  below).     The  Role  of  an  Antigen-­‐Specific  Response  in  RAO   Recurrent  airway  obstruction-­‐susceptible  horses  are  hypersensitive  to  HD   compared  to  unaffected  horses.25  Over  the  last  30  years,  RAO  research  has  focused  on   identifying  specific  causal  agents  to  which  RAO-­‐susceptible  horses  are  hypersensitive  and   identifying  mechanisms  that  mediate  this  hypersensitivity.  The  association  of  disease   exacerbation  and  moldy  hay26,27  are  suggestive  that  RAO  could  be  a  hypersensitivity  to   specific  fungal  or  actinomycete  allergens,  as  these  pathogens  are  abundant  in  poorly  cured   (moldy)  hay.  Thus,  it  has  been  proposed  that,  akin  to  atopic  asthma,  a  type  1   hypersensitivity  response  to  specific  fungal  elements  underlies  the  RAO  immunopathology,   5   i.e  RAO  is  an  allergen  specific  response.  However,  certain  features  of  the  RAO  response  differ   from  the  typical  atopic  response.     Allergen  provocation  in  human  asthmatics  induces  a  biphasic  response;  an  acute   phase/immediate  response  and  a  late  phase/delayed  response,  which  are  precipitated  by   different  immune  mechanisms.  Atopic  asthma  is  characterized  by  elevated  levels  of   allergen-­‐specific  IgE  antibodies  that  bind  to  high  affinity  receptors  on  mast  cells  and   basophils  within  the  airway.  Inhalation  of  allergen  crosslinks  IgE  and  activates   degranulation  of  mast  cells  and  basophils  releasing  a  variety  of  inflammatory  mediators   including  histamine,  leukotrienes,  prostaglandins,  platelet  activating  factor  and  cytokines   (e.g.  IL-­‐4).  These  inflammatory  mediators  contribute  to  the  immediate  allergic   inflammatory  response  characterized  by  bronchoconstriction,  vasodilation  and  airway   edema,  a  response  that  occurs  within  minutes  of  allergen  exposure.     Allergen  also  activates  allergen-­‐specific  CD4-­‐Th2  cells  that  are  critical  in   orchestrating  the  delayed  allergic  response  that  develops  within  hours  after  exposure.   Allergen  activated  CD4-­‐Th2  cells  proliferate  and  express  effector  cytokines  IL-­‐5,  IL-­‐13  and   IL-­‐4  that  contribute  to  pathology;  IL-­‐5  stimulates  airway  eosinophilia,  IL-­‐13  promotes   mucus  hypersecretion  and  airway  hyperresponsiveness  and  IL-­‐4  promotes  IgE  isotype   class  switching  and  promotes  further  Th2  cell  differentiation.     The  diagnosis  of  allergic  asthma  in  humans  is  principally  dependent  on  detection  of   systemically  elevated  allergen-­‐specific  IgE  (i.e  elevated  sera  antibody)  or  a  positive   intradermal  skin  test  indicative  of  allergen  specific  sensitized  mast  cells.  A  number  of   studies  have  failed  to  detect  elevated  allergen-­‐specific  IgE  (serum)  in  RAO-­‐affected   horses28–31  and  immediate  skin  reactions  (wheal  formation)  during  intradermal  skin   6   testing  is  not  associated  with  the  RAO-­‐phenotype.30,32  Taken  together,  this  suggests  that   systemic  production  of  IgE  and  mast  cell  sensitization  does  not  occur  in  RAO.  In  contrast,   two  early  studies  demonstrated  elevated  allergen-­‐specific  IgE  (to  Micropolyspora  faeni  and   Aspergillus  fumigatus)  in  BALF28,29  which  could  suggest  that  local  production  of  IgE  could   contribute  to  RAO.  However,  these  studies  utilized  in-­‐vitro  allergen  assays  with  low   specificity  for  equine  allergen-­‐specific  IgE  and  specific-­‐allergen  epitopes,  which  may  have   resulted  in  an  increased  rate  of  false  positives.33  The  performance  of  in-­‐vitro  allergen   assays  has  been  improved  by  using  pure  allergen  extracts  and  monoclonal  antibodies34  but   the  measurement  of  allergen-­‐specific  IgE  in  BALF  of  RAO  horses  has  not  been  repeated.   However,  RAO-­‐affected  horses  do  not  possess  more  IgE  positive  cells  in  BALF35  or  lung   tissue  samples36  and  following  exposure  to  organic  dust,  RAO  horses  do  not  develop   immediate  changes  in  pulmonary  function.37  Taken  together  these  data  suggest  that  unlike   allergic  asthma,  allergen-­‐specific  mast  cell  sensitization  and  degranulation  are  not  features   of  RAO  and  exacerbations  of  RAO  develop  independent  of  IgE-­‐mediated  mechanisms.   Following  exposure  to  HD,  inflammation  and  pulmonary  dysfunction  develop  within   hours;  however,  it  remains  unclear  if  CD4-­‐Th2  cells  drive  this  delayed  pulmonary  response   in  RAO  (as  described  above).  Again,  the  horse  exhibits  some  distinct  differences  from   typical  atopic  response.  Primarily,  RAO  exacerbation  is  associated  with  neutrophil   recruitment  (and  not  eosinophils)  into  the  airway  lumen.8  Further,  no  studies  have   detected  underlying  Th2  cell  polarization  during  remission  and  Th2  cytokines  following   allergen  exposure  are  not  consistently  detected.  Two  studies  demonstrated  that  increased   IL-­‐4  and  IL-­‐5  expression  develops  in  RAO-­‐susceptible  horses  within  the  first  24  hours  of   hay  exposure.38,39  Further,  horses  with  chronic  inflammation  (≥9  days  hay  exposure)  also     7   exhibited  elevated  IL-­‐4/IL-­‐5  and  reduced  IFNγ  expression38,39  suggestive  of  a  polarized  Th2   cytokine  profile  similar  to  human  allergic  airway  disease.40,41  In  contrast,  short  exposure   (24-­‐48  hours)  to  hay  had  no  effect  on  IFNγ,  IL-­‐4  or  IL-­‐5  in  BAL  cells  or  isolated  BAL   lymphocytes.42–44  Therefore,  it  is  unclear  if  polarized  allergen-­‐specific  Th2  cells  orchestrate   the  initial  neutrophilic  influx.  Further,  chronic  exacerbations  may  be  characterized  by   elevated  IFNγ  expression  (Th1  polarized)  16,20,42  elevated  IL-­‐13  and  IFNγ  expression  (mixed   Th1/Th2)45  or  reduced  Th2  cytokine,  IL-­‐13.42,46  Thus  the  role  of  Th2  and  Th1  cytokines  in   disease  pathogenesis  remains  unclear.     It  is  likely  that,  similar  to  asthma,  sub-­‐phenotypes  of  RAO  exist  with  heterogeneous   immune  mechanisms  that  could  account  for  these  varied  results.47–49  Indeed,  one  group  of   investigators  demonstrated  that  the  RAO-­‐affected  horses  from  two  different  lineages  show   associations  with  two  different  chromosome  locations  (genetic  heterogeneity).  This   suggests  that  between  different  families,  different  genes/immune  mechanisms  mediate  the   disease  but  ultimately  produce  an  identical  clinical  phenotype  (neutrophilic  inflammation,   coughing  and  bronchoconstriction).48,50,51  This  genetic  heterogeneity  may  have  significant   impact  on  the  study  of  the  disease  as  Lanz  et  al52  demonstrated  that  peripheral  blood   mononuclear  cells  from  the  two  horse  families  respond  differently  to  in-­‐vitro  stimulation   with  HD.         Furthermore,  it  remains  to  be  determined  if  the  lymphocytic  bronchiolitis  that   develops  during  RAO  is  composed  of  polarized  Th2  lymphocytes  (as  apposed  to  the  luminal   lymphocytes).  It  has  been  suggested  that  RAO  is  similar  to  “intrinsic  asthma”53  which   differs  from  atopic  asthma  (extrinsic)  due  to  the  absence  of  peripheral  eosinophilia  or   allergen-­‐specific  IgE.  Affected  patients  do  not  show  sensitivity  to  commonly  tested   8   allergens  (i.e.  negative  skin  prick  test)  thus  the  causative  agent  remains  unspecified.   However,  it  has  been  suggested  that  the  allergen-­‐specific-­‐IgE  and  -­‐CD4-­‐Th2  cells  remain   localized  within  the  lung  as  bronchial  biopsies  from  extrinsic  and  atopic  asthmatics  show   similar  bronchial  mucosal  cellular  infiltrate54  which  would  indicate  that  the  mechanism  of   intrinsic  asthma  resembles  extrinsic  asthma.   Taken  together,  an  allergen-­‐specific  IgE-­‐mediated  mechanism  is  not  a  component  of  RAO   immunopathology  and  evidence  for  allergen-­‐specific  Th2  lymphocytes  as  mediators  of  the  HD   hypersensitivity  is  conflicting.  Thus,  it  remains  unclear  if  an  allergen  specific   immunopathology  is  responsible  for  the  enhanced  sensitivity  to  HD.       The  Non-­‐Specific  Response         Clinical  signs  of  RAO  can  be  induced  by  exposure  to  individual  allergens  that  are   present  in  HD  (e.g.  Aspergillus  fumigatus,  and  Micropolyspora  faenia  now  renamed  as   Saccharopolyspora  rectivirgula),26,27,55  but  these  experimental  inhalational  studies  do  not   completely  reproduce  the  severity  of  disease  and  development  of  pulmonary  dysfunction  is   inconsistent.  While  this  could  be  attributed  to  the  experimental  designs  (inadequate   quantity  of  antigen  delivered,  inadequate  frequency  of  exposures  or  failure  to  identify  and   deliver  the  correct  antigen)  it  is  also  suggestive  that  naturally  induced  RAO  exacerbations   require  activation  of  additional  immune  mechanisms.  Indeed,  both  the  adaptive  and  the   innate  immune  system  are  activated  in  atopic  asthma  as  protein  allergens  (e.g  HD  mite  and   cat  dander)  are  associated  with  non-­‐allergenic  ligands  such  as  fungi  and  bacteria  which   engage  the  innate  immune  system.56  HD  contains  a  mixture  of  plant  stems,  pollens,  many   different  species  of  fungi,  mite  feces  and  exoskeleton  and  bacteria.57,58  Thus,  although     9   molds  may  act  as  allergens,  the  HD  mixture  is  abundant  with  PAMPs  that  can  stimulate  the   PRRs  of  the  innate  immune  system.     The  bacterial  ligand  lipopolysaccharide  (LPS)  is  present  in  HD  and  stimulates  the   innate  immune  system  through  ligation  of  Toll-­‐like  receptor  4  (in  cooperation  with  CD14   and  MD-­‐2).  LPS  on  its  own  (at  concentrations  equivalent  to  those  present  in  HD)  are   insufficient  to  induce  an  inflammatory  response.  However,  the  presence  of  LPS  on  fungal   extracts21,59  and  HD  particles60  enhances  their  inflammatory  potential  as  depletion  of  LPS   significantly  reduces  neutrophilic  inflammation  and  the  activity  of  MMP-­‐9  in  RAO-­‐ susceptible  horses.     In  a  series  of  papers,  Pirie  et  al61,62  determined  that  both  LPS  and  HD  induce   pulmonary  neutrophilia  in  a  dose  dependent  fashion.  However,  in  comparison  to  control   horses,  RAO-­‐susceptible  horses  develop  robust  pulmonary  neutrophilia  at  lower   concentrations  of  these  agonists.  Yet,  delivery  of  very  high  agonist  concentrations  can   induce  an  equivalent  neutrophilic  response  in  control  horses.  Thus,  RAO-­‐affected  and   control  horses  have  a  difference  in  sensitivity  but  both  groups  can  respond  in  a  similar   manner  (using  pulmonary  neutrophilia  as  an  end  point)  suggesting  that  a  shared  non-­‐ allergen-­‐specific  mechanism  may  play  an  important  role.   Indeed,  a  number  of  studies  suggest  that  exposure  to  hay  can  induce  similar   inflammatory  responses  in  control  and  RAO-­‐susceptible  horses,  but  that  the  responses   often  differ  in  magnitude  or  duration.  As  in  RAO-­‐susceptible  horses,  HD  induces  a  lower   degree  of  pulmonary  neutrophilia63,64  associated  with  elevated  IL-­‐8  expression65  in  control   horses.  Further,  initial  exposure  induces  a  transient  cough  rather  than  a  chronic  cough.63   Interestingly,  RAO-­‐susceptible  and  control  horses  exposed  to  5  days  of  moldy  hay  showed   10   similar  degrees  of  bronchiolitis  and  histopathological  scores,  further  suggesting  a  common   response  to  hay  exposure.66  Further,  in  comparison  to  controls,  RAO-­‐susceptible  horses,   exhibit  similar  but  greater  increases  in  systemic  acute  phase  proteins.67  Also,  HD  induces   similar  increases  of  IL-­‐8  and  TNFα  expression  in  peripheral  blood  neutrophils  from  RAO-­‐ susceptible  and  control  horses.  However,  following  natural  challenge,  IL-­‐4  induced  IL-­‐8  is   enhanced  in  RAO-­‐affected  horses  compared  to  controls.68     Overall,  this  suggests  that  HD  is  a  noxious  stimulant  with  the  capacity  to  induce  a   non-­‐specific  inflammatory  response  in  all  horses.  However,  compared  to  control  horses,   the  inflammatory  response  that  occurs  in  RAO-­‐susceptible  horses  is  unregulated  and   greater  in  magnitude,  suggesting  that  differential  sensitivity  could  occur  at  the  level  of  the   innate  immune  system.  The  mechanisms  that  underlie  the  increased  sensitivity  to  HD  or   LPS  are  unknown.  With  regard  to  LPS  responsiveness,  there  is  no  evidence  that  RAO-­‐ susceptible  horses  (in  remission)  express  greater  levels  of  TLR-­‐4  in  BALF  cells13  or   respiratory  epithelium,69  although  elevations  are  associated  with  exposure  to  hay.13,16,69   Asthmatics  may  have  higher  levels  of  airway  CD14  which  may  contribute  to  LPS   sensitivity70  but  expression  of  CD14  has  not  been  investigated  in  the  horse  airways.   Furthermore,  polymorphisms  of  TLR4  have  also  been  associated  with  asthma  and  may  play   a  role  in  LPS  sensitivity.71  Within  the  healthy  horse  population,  inter-­‐horse  LPS   responsiveness  varies  greatly.  Variability  is  not  attributed  to  TLR4  polymorphisms,72   however,  it  remains  to  be  determined  if  RAO-­‐susceptible  horses  possess  TLR4   polymorphisms  that  enhance  LPS  responsiveness.     It  is  also  unknown  if  RAO-­‐susceptible  horses  exhibit  increased  sensitivity  to  other   PAMPs,  such  as  β-­‐glucan  and  peptidoglycan,  that  are  present  in  HD.  However,  the  potency     11   of  HD  (assessed  by  degree  of  neutrophilia  and  active  MMP9  levels)  is  correlated  with  the   content  of  the  fungal  cell  wall  component  β-­‐glucan.25,73  β-­‐glucan  stimulates  the  PRR  the  β-­‐ glucan  receptor  and  in  a  model  of  Aspergillus  fumigatus  allergic  disease,  stimulation  of  the   β-­‐glucan  receptor  contributes  to  development  of  airway  hypersensitivity,  airway   neutrophilia,  pro-­‐inflammatory  and  pro-­‐allergic  cytokines  indicating  that  this  receptor   plays  an  important  role  in  fungal  allergy.74  Despite  its  potential  importance  there  have   been  no  studies  evaluating  the  expression  of  this  receptor  in  RAO-­‐susceptible  horses  and  it   is  unknown  if  RAO-­‐susceptible  horses  would  also  display  hypersensitivity  to  β-­‐glucan   exposure.     In  addition  to  LPS,  which  originates  from  gram-­‐negative  bacteria,  HD  (and  other   agricultural  organic  dusts)  contain  gram-­‐positive  bacteria.75,76  Exposure  of  mice  to  a  swine-­‐ barn-­‐dust  induces  significant  neutrophilic  pulmonary  inflammation  and  the  presence  of   high  quantities  of  peptidoglycan  (a  component  of  gram-­‐positive  bacterial  cell  walls)  plays  a   significant  role.  Peptidoglycan  stimulate  TLR-­‐2,  and  TLR-­‐2  knock  out  mice  display  a   significantly  attenuated  inflammatory  response  on  exposure  to  swine  dust  extract.77  The   response  to  exposure  of  TLR-­‐2  agonists  (such  as  peptidoglycan)  has  not  been  investigated   in  RAO-­‐affected  horses.  However,  investigations  of  the  response  to  Sacchropolyspora   rectivirgula  inhalation  (formerly  Micropolyspora  faeni,  a  thermophilic,  gram  positive,   actinomycete  prevalent  in  moldy  hay)  suggest  that  RAO-­‐susceptible  horses  have  increased   sensitivity  to  this  agent:  RAO-­‐susceptible  horses  developed  a  greater  influx  of  airway   neutrophils  compared  to  control  horses.27  Further,  in  a  separate  study,  RAO-­‐susceptible   and  control  horses  both  developed  similar  degrees  of  airway  neutrophils  but  only  RAO   horses  developed  pulmonary  dysfunction.78  Sacchropolyspora  rectivirgula  is  a  known   12   allergen  in  people  and  is  associated  with  hypersensitivity  pneumonitis  (HP)  an  allergic   alveolitis  mediated  by  allergen-­‐specific  Th1  cells.79  However,  certain  clinical  and  pathologic   features  of  HP  differ  from  RAO:  HP  is  associated  with  systemic  signs  of  fever  and  malaise   and  histopathology  demonstrates  alveolitis,  lymphocytic  interstitial  infiltration  and   fibrosis,  and  the  presence  of  non-­‐necrotizing  granulomas.80  It  is  unknown  if  the   peptidoglycan  content  of  Sacchropolyspora  rectivirgula  could  contribute  to  this  apparent   sensitivity  via  engagement  of  the  innate  immune  system.  Previous  studies  have   demonstrated  that  RAO  is  not  associated  with  increased  elevations  in  TLR-­‐2  expression  of   bronchial  epithelial  cells,65,81  but  expression  of  TLR-­‐2  in  alveolar  macrophages  has  not  been   investigated.     The  current  evidence  indicates  that  HD  can  incite  an  inflammatory  response  in  both   non-­‐RAO-­‐susceptible  horses  and  RAO-­‐susceptible  horses  suggesting  activation  of  common   innate  pathways.  However,  in  contrast  to  control  horses  that  regulate  the  inflammatory   response,  RAO-­‐susceptible  horses  develop  uncontrolled  inflammation.  The  exact  mechanisms   through  which  HD  elicits  this  uncontrolled  inflammatory  reaction  remains  uncertain  but  it  is   possible  the  differential  sensitivity  lies  at  the  level  of  the  innate  immune  system  such  as  the   alveolar  macrophage.         The  Alveolar  Macrophage  in  RAO     There  are  little  data  focusing  on  the  contribution  of  the  alveolar  macrophage  in  RAO   pathogenesis.  In  RAO-­‐susceptible  horses,  HD  exposure  is  associated  with  an  influx  of   neutrophils,  but,  in  contrast,  the  absolute  numbers  of  macrophages  do  not  change43,59,60  or   are  significantly  decreased27,38,62,82,83  and  the  macrophage  cytology  percentage  is  frequently     13   reduced.35,38,42,84  In  control  horses,  exposure  to  HD  generally  does  not  alter  macrophage   absolute  number.27,38,42,62  The  reduced  macrophage  numbers  could  be  explained  by   increased  apoptosis  or  migration  from  the  airways.  However,  there  is  little  difference  in  the   proportion  of  apoptotic  alveolar  macrophages  (<3%  difference  between  groups)  between   control  and  RAO-­‐susceptible  horses  following  exposure  to  hay,85  thus  it  is  unlikely  that  this   contributes  to  the  reductions  in  macrophage  numbers.  It  is  also  possible  that  low   macrophage  yield  could  be  a  consequence  of  reduced  lavage  recovery  due  to   bronchoconstriction  and  mucus  plugging  in  the  diseased  lung86  and  may  not  accurately   reflect  changes  in  macrophage  numbers  in  the  RAO-­‐affected  lung.  It  is  also  possible  that   reduced  numbers  of  alveolar  macrophages  could  delay  resolution  of  inflammation  as  the   alveolar  macrophage  clears  airway  neutrophils  via  phagocytosis.9   It  is  unclear  if  the  alveolar  macrophage  contributes  to  the  inflammatory  response   during  RAO-­‐exacerbation.  Following  acute  exposure  to  HD,  alveolar  macrophages  from   RAO  affected  horses  express  higher  levels  of  pro-­‐inflammatory  cytokines  (TNF-­‐α,  IL-­‐1β  and   IL-­‐8)  compared  to  normal  horses87  suggesting  that  a  pro-­‐inflammatory  macrophage   phenotype  exists  in  RAO-­‐susceptible  horses.  However,  in  contrast,  Joubert  et  al88  found  no   difference  in  the  pro-­‐inflammatory  response  of  alveolar  macrophages  from  RAO-­‐affected   and  normal  horses  exposed  to  natural  challenge.     A  potential  mechanism  for  the  enhanced  sensitivity  and  uncontrolled  inflammation   in  RAO  is  the  presence  of  divergent  alveolar  macrophage  phenotypes.  Macrophages  are   positioned  at  the  host  environment  interface,  and  possess  a  vast  array  of  receptors   enabling  them  to  respond  to  pathogens.  Importantly,  they  can  develop  into  distinct   functional  phenotypes  that  alter  their  sensitivity  to  stimulation.  Macrophages  can  polarize   14   into  pro-­‐inflammatory  or  anti-­‐inflammatory/regulatory  phenotypes  and  2  basic   macrophage  phenotypes  are  described  with  counter  active  functions.  The  M1  phenotype  is   considered  pro-­‐inflammatory,  and  functions  to  augment  microbiocidal  effector  functions   required  for  successful  host  defense  against  pathogens.  These  functions  include  enhanced   respiratory  burst  capacity,  antigen  presentation  and  production  of  pro-­‐inflammatory   cytokines  in  response  to  subsequent  microbial  stimulation.  In  contrast,  M2  macrophages   are  considered  anti-­‐inflammatory,  have  poor  microbiocidal  mechanisms  and  oppose  and   regulate  inflammation  and  promote  tissue  repair  or  specialize  in  host  defense  against   parasites.  Thus,  divergent  alveolar  macrophage  phenotypes  could  contribute  to  the  enhanced   sensitivity  and  uncontrolled  inflammation  observed  in  RAO.  This  formed  the  basis  for  my   overarching  hypothesis,  that  the  enhanced  sensitivity  to  hay  dust  observed  in  RAO-­‐susceptible   horses  may  be  mediated  by  the  presence  of  pro-­‐inflammatory  M1  phenotypes  or  the  absence   of  anti-­‐inflammatory  M2  phenotypes.     15   Section  2.  The  Biology  of  Alveolar  Macrophages     The  following  section  provides  background  detailing  the  critical  importance  of  alveolar   macrophages  within  the  healthy  lung,  their  functional  plasticity  in  response  to  changing   environments  and  development  of  M1  and  M2  phenotypes.  Further,  this  section  discusses   alveolar  macrophage  phenotypes  in  chronic  airway  disease  with  the  focus  on  asthma.     Origins  of  Alveolar  Macrophages   Until  recently  the  classical  theory  was  that  alveolar  macrophages  (and  other  tissue   macrophages)  are  primarily  derived  from  circulating  monocytes  produced  by  common   myeloid  progenitor  cells  in  the  bone  marrow.  Further,  interstitial  macrophages  could  serve   as  an  intermediary  step  between  blood  monocytes  and  alveolar  macrophages.89     However,  it  is  now  acknowledged  that  resident  alveolar  macrophages  are  derived  from   embryonic  yolk  sac  progenitor  cells  that  are  maintained  through  out  life  by  local  self-­‐ renewal.90  Alveolar  macrophages  are  long-­‐lived  cells  (at  least  4  months)  and  maintainance   of  alveolar  macrophage  cell  numbers  under  steady  state  conditions  can  be  accomplished  by   local  proliferation  of  the  embryonically  derived  tissue  alveolar  macrophages  rather  than   constitutive  replenishment  by  infiltrating  monocytes.91,92  However,  during  inflammation,   peripheral  monocytes  are  recruited  to  the  lung,  attracted  by  the  release  of  chemotactic   stimuli  (CC-­‐Chemokine  ligand  2  [CCL2]).93  These  infiltrating  monocytes  (phenotypically   characterized  as  “inflammatory”  monocytes)  then  differentiate  into  macrophages   (infiltrating  macrophages).  It  is  unclear  if  these  infiltrating  macrophages  then  persist  long   term  to  become  resident  macrophages.  It  has  been  demonstrated  that  weeks  after  an   inflammatory  insult,  resident  alveolar  macrophages  are  once  again  predominately  derived   16   from  the  embryonically  derived  resident  population,  indicating  local  proliferation  and  re-­‐ establishment  of  numbers.92  Thus  airway  macrophages  may  be  derived  from  blood  or  local   proliferation,  but  regardless  of  their  origin,  once  in  the  airway,  exposure  to  the  distinctive   pulmonary  environment  will  impact  their  phenotype.     The  Role  of  the  Alveolar  Macrophage  in  the  Healthy  Lung   Tissue  macrophages  exist  in  every  organ  and  have  an  integral  role  in  maintaining   tissue  homeostasis.  However,  within  each  organ,  resident  macrophages  are  highly  adapted   to  perform  the  unique  functions  that  are  required  for  each  specific  environment.94  Thus,   the  environment  induces  functional  heterogeneity.       The  lungs  are  continuously  exposed  to  aeroallergens  and  PAMPs  and  it  is  a  constant   challenge  for  the  pulmonary  immune  system  to  differentiate  between  innocuous  and   harmful  antigens  and  respond  appropriately.  Unnecessary  activation  of  the  immune  system   will  result  in  costly  damage  to  the  delicate  lung  architecture,  but  on  the  other  hand  a   sluggish  response  to  inhaled  pathogens  could  result  in  severe  microbial  infections.  Thus   the  pulmonary  immune  system  must  monitor  the  challenge  material  carefully  and  regulate   an  appropriate  response.     In  the  healthy  human  lung,  the  alveolar  macrophage  is  the  most  abundant  leukocyte   lining  the  airways  and  alveoli  and  constitutes  90%  of  the  cells  of  the  bronchoalveolar   lavage.95  By  contrast,  in  the  healthy  equine  lung,  the  macrophage  remains  a  dominant  cell   type  but  constitutes  between  40-­‐70%  of  the  cells  of  the  BAL,  with  the  remainder  of  cells   being  mostly  lymphocytes  (30-­‐60%)  with  a  few  neutrophils  (<5%).96  In  the  healthy  lung,   the  resident  alveolar  macrophage  plays  a  key  role  in  maintaining  pulmonary  homeostasis   17   by  performing  a  janitorial  role  engaged  in  phagocytosis  of  aeroallergens,  cellular  debris   such  as  apoptotic  or  necrotic  cells,  and  pulmonary  surfactant.97  However,  the  alveolar   macrophage  provides  first  line  defense  against  inhaled  pathogens  and  is  equipped  with  a   vast  array  of  PRRs  that  recognize  PAMPs.  Stimulation  of  these  PRRs  can  incite  an   inflammatory  response.     However,  alveolar  macrophages  also  express  a  variety  of  receptors  that  block  the   inflammatory  response  when  engaged  (negative  regulators).  The  unique  environment  of   the  healthy  lung  expresses  ligands  that  engage  these  negative  regulators  maintaining  the   resident  alveolar  macrophage  in  a  quiescent  state  suppressing  their  inflammatory  potential   by  blocking  inflammatory  pathways.  Cross  talk  between  bronchiolar  and  alveolar  epithelial   cells  down-­‐regulates  the  alveolar  macrophage.  The  regulatory  protein  CD200  expressed  by   respiratory  epithelial  cells  and  apoptotic  cells  binds  to  the  alveolar  macrophages  CD200   receptor  (CD200R),  inhibiting  pro-­‐inflammatory  signaling  pathways.98  Further,  TGFβ  is   produced  in  healthy  lungs  but  is  secreted  in  an  inactive  (latent)  form.  Bronchial  and   alveolar  epithelium  express  the  αVβ6  integrin  which  activates  latent  TGFβ,99  and  αVβ6   tethered  TGFβ  further  suppresses  the  induction  of  inflammatory  cytokines  in  the  alveolar   macrophage.   Other  soluble  factors  that  are  present  in  high  quantities  in  the  lung  contribute  to   basal  suppression  of  the  alveolar  macrophage.  Surfactant  proteins  (SPA  and  SPD)  bind  to   the  alveolar  macrophage  receptor  SIRPα  (signal  inhibitory  regulatory  protein-­‐α)  and   inhibit  NFKB  activation.100  IL-­‐10  is  constituently  expressed  by  alveolar  epithelial  cells101   and  binding  to  the  IL-­‐10  receptor  of  alveolar  macrophages  reduces  inflammatory  cytokine   production.    Further,  prostaglandin-­‐E,  produced  by  bronchial  epithelium,  suppresses   18   microbicidal  activity.102  Gap-­‐junction  channels  allow  direct  communication  between  a   population  of  alveolar  macrophages  and  alveolar  epithelium  so  that  waves  of  calcium  can   provide  an  immunosuppressive  signal  to  the  alveolar  macrophage.103  Thus,  in  many  ways,   the  unique  lung  environment  exerts  a  regulatory  influence  on  the  alveolar  function   maintaining  a  basal  quiescent  state.  Furthermore,  adoptive  transfer  of  other  tissue   macrophages  into  the  lung  allows  them  to  “take  on”  these  alveolar  macrophage   characteristics.104   Resting  alveolar  macrophages  also  actively  suppress  neighboring  dendritic  cells  and   T-­‐cells  further  contributing  to  the  tonic  immune  suppression  in  the  healthy  lung.97   However,  the  need  to  suppress  the  inflammatory  potential  of  the  alveolar  macrophage   must  be  balanced  with  the  need  to  protect  the  lung  from  inhaled  pathogens  and  stimulation   with  PAMPs  can  shift  the  balance  from  the  immunosuppressive  state  to  a  pro-­‐inflammatory   response.  Through  the  previously  mentioned  mechanisms  the  lung  establishes  an  elevated   threshold,  which  must  first  be  overcome  to  ensure  that  a  pro-­‐inflammatory  response  is   only  reached  when  absolutely  necessary.     Thus  the  alveolar  macrophage  has  a  dual  function  in  the  lung:  to  maintain  homeostasis  and   to  initiate  host  defense.  The  macrophage  is  well  equipped  to  respond  to  a  wide  variety  of   presented  pathogens  but  this  pro-­‐inflammatory  potential  must  be  curtailed  to  ensure  that   delicate  lung  tissue  is  not  unnecessarily  damaged.  This  is  possible  because  the  alveolar   macrophage  possesses  a  plasticity  that  allows  it  to  respond  to  signals  in  the  environment  and   adapt  appropriately.           19   M1  and  M2  Phenotypes     Macrophages  are  equipped  with  a  vast  array  of  receptors  that  impart  extensive   monitoring  capabilities  enabling  macrophages  to  detect  a  variety  of  alterations  in  their   environment  including  exogenous  pathogenic  molecules,  endogenous  alarmins  released   during  tissue  damage,  and  locally  released  cytokines.  The  macrophages  integrate  this   information  and  respond  accordingly.  Thus  alveolar  macrophages  are  sentinel  cells  that   play  a  critical  role  in  initiating  inflammatory  reactions  and  becoming  effector  cells  -­‐   protecting  the  host  from  pathogenic  challenge.  However,  uncontrolled  inflammation  is   detrimental  and  macrophages  are  also  key  in  regulating  and  resolving  the  inflammatory   reaction  and  promoting  tissue  repair.  Macrophages  can  achieve  this  by  developing  into   functionally  distinct  populations.  The  prototypical  example  of  this  functional  plasticity  is   the  development  of  the  functional  polar  extremes:  pro-­‐inflammatory  (M1)  and  an  anti-­‐ inflammatory  (M2).     The  M1  macrophage  is  a  specialized  microbicidal  effector  cell  and  achieves  this   through  enhanced  phagocytosis,  production  of  reactive  oxygen  species,  and  release  of  pro-­‐ inflammatory  mediators  and  cytokines.  This  pro-­‐inflammatory  phenotype  is  often  referred   to  as  “classical  activation”  and  can  be  induced  in  response  to  pathogen  recognition  receptor   stimulation  e.g  in  response  to  bacterial  challenge.105  Stimulation  by  IFNγ  also  activates  an   M1  phenotype  and  in-­‐vivo,  this  may  originate  from  innate  cells  (natural  killer  cells)  or   antigen-­‐specific  Th1  cells  which  are  also  part  of  the  host  response  to  intracellular   pathogens.     In  vitro,  M1  macrophages  can  be  activated  by  simultaneous  stimulation  with  IFNγ   and  TNFα.  Stimulation  with  LPS  also  can  be  an  activator  because  LPS  induces  macrophage     20   production  of  TNFα  that  then  acts  synergistically  with  IFNγ  to  promote  an  M1   phenotype.105  Murine  M1  macrophages  are  typified  by  high  expression  of  inducible  nitric   oxide  synthase  (iNOS2)  that  converts  L-­‐arginine  into  nitric  oxide.  Nitric  oxide  can  combine   with  hydrogen  peroxide  or  superoxide  radicals  within  phagolysomes  to  produce  anti-­‐ microbial  peroxynitrite  radicals.  M1  macrophages  also  release  pro-­‐inflammatory  cytokines   (including  TNFα,  IL-­‐1β,  IL-­‐6,  IL-­‐12)  and  chemokines  (IL-­‐8)  that  function  to  stimulate  and   amplify  inflammatory  responses.  Further,  cytokines  secreted  during  M1  activation  may   influence  T-­‐lymphocyte  differentiation  (Th1  via  secretion  of  IL-­‐12,  Th17  via  secretion  of  IL-­‐ 23)106  augmenting  the  host  defense  against  pathogens.  However,  chronic  activation  of  this   pro-­‐inflammatory  phenotype  can  lead  to  tissue  damage  and  organ  dysfunction.  The  M1   phenotype  is  also  associated  with  increased  expression  of  MHC  class  II  and  of  co-­‐stimulator   molecules  (CD80  and  CD86)  resulting  in  increased  antigen-­‐presenting  activity.  M1   macrophages  are  significant  pathologic  effector  cells  in  inflammatory  bowel  disease,107   hepatotoxicity108  and  obesity  related  insulin  resistance.106     In  contrast  to  the  classical  pro-­‐inflammatory  M1  phenotype,  alternative  macrophage   phenotypes  exist  that  mediate  tissue  repair  or  regulate  inflammation.  These  phenotypes   are  collectively  called  alternatively  activated  macrophages  (M2)  and  they  evolve  in   response  to  a  range  of  stimuli  including  IL-­‐4,  IL-­‐10  and  glucocorticoids  although  it  is  now   clear  that  the  different  activating  stimuli  promote  differentiation  of  phenotypes  with   different  characteristics.109,110       Interleukin-­‐4  activated  M2  macrophages  have  an  important  role  in  tissue  repair  and   are  associated  with  host  defense  against  helminths  and  some  fungi.  The  IL-­‐4  may  be   derived  from  innate  cells  (basophils  and  mast  cells)  or  antigen  specific  Th2  cells.111  In     21   contrast  to  the  M1  phenotype,  these  M2  macrophages  have  poor  microbiocidal   mechanisms,  produce  minimal  pro-­‐inflammatory  cytokines  and  promote  tissue  repair.     M2  macrophages  are  associated  with  a  number  of  signature  proteins  (induced  by  IL-­‐4)  that   include  arginase-­‐1  (Arg-­‐1),  mannose  receptor  (CD206),  β  -­‐glucan  receptor,  YM1  and  FIZZ1   (found  in  inflammatory  zone  1).  IL-­‐4  activated  M2  macrophages  express  elevated  Arg-­‐1,   which  competes  with  iNOS  for  the  substrate  L-­‐arginine  and  converts  it  to  L-­‐ornithine.  L-­‐ ornithine  is  a  substrate  for  both  L-­‐proline,  which  is  an  essential  substrate  for  collagen,  and   polyamines,  which  regulate  cell  proliferation  and  differentiation  indicating  a  role  in  tissue   repair.   M2  macrophages  also  display  increased  expression  of  a  number  of  phagocytic   receptors  including  the  members  of  the  C-­‐type  lectin  family;  CD206112  and  β-­‐glucan   receptor  (Dectin  -­‐1).113  The  CD206  receptor  is  the  canonical  M2  associated  receptor.112  It   has  three  different  binding  domains  (cysteine  rich  domain  (CR),  fibronectin  domain  (FN)   and  C-­‐type  lectin-­‐like  carbohydrate  recognition  domain  (CTLD))  that  bestow  a  capacity  to   bind  a  variety  of  endogenous  and  exogenous  material.  It  is  predominately  expressed  intra-­‐ cellularly  within  endosomes,  and  samples  material  within  the  phagosome.114  The  CTLD   recognizes  mannose  on  many  bacteria  and  fungi  and  can  induce  production  of  pro-­‐  or  anti-­‐ inflammatory  cytokines  depending  on  the  stimulant115,116  or  modulate  the  inflammatory   cascade  induced  by  some  TLR.117  The  receptor  also  plays  an  important  role  in  tissue   homeostasis  and  the  CR  domain  recognizes  a  variety  of  endogenous  materials  such  as   thyrotropin  and  chondroitin  sulphate.115  Further,  CD206  contributes  to  the  resolution  of   inflammation  and  repair  through  its  capacity  to  endocytose  myeloperoxidase118  and   degrade  collagen.119       22   The  β-­‐glucan  receptor  (Dectin-­‐1)  recognizes  β-­‐1,3  and  β-­‐1,6  linked  glucans   (carbohydrate  polymers),  which  are  a  major  structural  component  of  fungal  cells.   Recognition  of  fungi  by  the  β-­‐glucan  receptor  induces  a  pro-­‐inflammatory  response  and  is   important  for  protection  against  fungal  infection.120–122  Stimulation  with  zymosan  (a  β-­‐ glucan  rich  yeast  particle)  can  stimulate  a  proinflammatory  response  via  a  number  of   intracellular  pathways.  Binding  of  β-­‐glucan  induces  phosphorylation  of  the  tyrosine  kinase   Syk  which  leads  to  activation  of  transcription  factors  nuclear  factor-­‐kappa-­‐B  (NFκB)  and   nuclear  factor  of  activated  T  cells  (NFAT)123  and  the  inflammasone  NLPR3.124  M2   macrophages  are  typically  considered  an  anti-­‐inflammatory  phenotype  (because  they   produce  low  quantities  of  pro-­‐inflammatory  cytokines  in  response  to  subsequent   stimulation).  However  IL-­‐4  induced  M2  macrophages  exhibit  an  enhanced  pro-­‐ inflammatory  response  to  zymosan  stimulation  which  is  mediated  by  the  β-­‐glucan   receptor,113  suggesting  that  (IL-­‐4  induced)  M2  macrophages  can  promote  a  pro-­‐ inflammatory  response  to  fungi.       IL-­‐4  induced  M2  macrophages  also  express  members  of  the  chitinase  family.  Chitin   is  a  polysaccharide  present  in  cell  walls  of  fungi,  plants  and  parasites.    YM1/YM2   (CHI3I3/CHI3I4)  are  expressed  in  murine  M2  and  can  bind  chitin  but  lack  enzymatic   function  (chitinase-­‐like  proteins)125  and  their  exact  function  is  unclear.126  The  chitinases   acidic  mammalian  chitinase  (AMCase)  and  chitotriosidase  possess  chitinolytic  activity  and   are  important  for  host  defense  against  fungi  and  parasites.  Resistin-­‐like-­‐α  (RELMα)  also   known  as  FIZZ1  contributes  to  extracellular  matrix  dynamics  by  inducing  myofibroblast   differentiation  and  survival.127     23   Alternatively  activated  macrophages  also  express  increased  anti-­‐inflammatory   cytokines  TGFβ  and  IL-­‐10.  Interleukin-­‐10  is  classically  considered  immunosuppressive  and   anti-­‐inflammatory  and  can  exert  effects  on  a  range  of  cells  including  granulocytes,   monocytes,  T-­‐cells  and  B-­‐cells.  It  can  function  to  decrease  oxidative  burst,  the  expression  of   pro-­‐inflammatory  cytokines  and  co-­‐stimulatory  molecules  (CD80,  CD86)  in   monocytes/macrophages  and  dendritic  cells.  Reduction  of  co-­‐stimulatory  molecules   impairs  T-­‐cell  differentiation,  effector  cytokine  production  and  induces  anergic  T   cells.128,129  The  anti-­‐inflammatory  response  is  mediated  by  signaling  through  the  IL-­‐10   receptor  (IL-­‐10R)  which  activates  Janus  kinase  (JAK)  1-­‐STAT3  pathway.130  Induction  of  the   transcription  factor  STAT3  selectively  reduces  the  transcriptional  rate  of  a  subset  of  LPS   induced  pro-­‐inflammatory  genes.131       TGFβ  plays  an  important  role  in  wound  healing  where  it  induces  a  broad  spectrum   of  effects  including  leukocyte  recruitment,  angiogenesis  and  collagen  synthesis.132  Further,   it  contributes  to  resolution  of  inflammation  and  regulates  many  immune  cells  including   inhibition  of  T-­‐cell  differentiation  and  proliferation,  B-­‐cell  proliferation  and  down-­‐ regulation  of  the  macrophage  inflammatory  response  to  IFNγ  or  LPS.  TGFβ  also  promotes   differentiation  of  T-­‐regulatory  cells.133  The  TGFβ  protein  is  synthesized  and  secreted  as  an   inactive  (latent)  precursor  and  activation  can  be  performed  by  a  variety  of  factors  including   reactive  oxygen  species,  matrix  metalloproteinases,  and  epithelial  integrin  αVβ6.  Binding  to   the  TGFβ  receptor  activates  the  family  of  Smad  proteins  that  regulate  gene  transcription.133   The  term  alternatively  activated  macrophage  refers  to  a  spectrum  of  macrophage   phenotypes  that  are  broadly  associated  with  immune  regulation  and  repair.106  In  addition   to  the  IL-­‐4  stimulated  M2  phenotype,  another  sub-­‐phenotype  called  a  regulatory     24   macrophage  (M-­‐reg)  has  been  described.  The  M-­‐reg  is  phenotypically  and  functionally   distinct  from  either  the  classically  activated  M1  or  IL-­‐4  induced  M2  as  described   above.110,134  Regulatory  macrophages  can  develop  subsequent  to  stimulation  with  a  variety   of  factors  including,  stimulation  with  macrophage-­‐colony  stimulating  factor  and  IFNγ,134   immune  complexes106  efferocytosis,135  glucocorticoids136  and  adenosine.137  Further,  TLR   stimulated  (pro-­‐inflammatory)  macrophages  initiate  an  intrinsic  mechanism  that  induces  a   transition  into  a  M-­‐reg  phenotype.138  Functionally,  the  M-­‐reg  modulates  the  inflammatory   response.139  They  can  inhibit  mitogen  stimulated  T-­‐cell  proliferation  and  are  less   responsive  to  LPS  stimulation.134  Furthermore,  M-­‐regs  can  preferentially  remove  allogenic   T-­‐cells  by  phagocytosis  and  delivery  of  M-­‐regs  can  prolong  organ  transplant  survival.134   Indeed,  delivery  of  regulatory  macrophages  as  an  adjunct  immunosuppressive  therapy  in   two  people  receiving  kidney  transplants  has  shown  some  promise.140       Likely,  the  repertoire  of  cytokines  and  receptors  expressed  by  M-­‐regs  may  vary   depending  on  the  exact  stimulating  conditions,  however,  the  hallmark  of  regulatory   macrophages  are  elevated  expression  of  IL-­‐10  which  usually  coincides  with  reduced   expression  of  IL-­‐12.  Like  M1,  murine  M-­‐reg  express  iNOS  but  have  low  expression  of  pro-­‐ inflammatory  cytokines  such  as  TNFα,  IL-­‐6  and  IL-­‐12  or  co-­‐stimulatory  molecule  CD80   when  compared  to  M1  phenotypes.  Furthermore,  M-­‐regs  express  low  levels  of  CD206  (in   comparison  to  IL-­‐4-­‐M2  phenotype)  but  display  elevated  levels  of  scavenger  receptors  β-­‐ glucan  receptor  and  macrophage  galactose-­‐type  c-­‐type  lectin  1  receptor.134           25   As  a  group,  the  alternatively  activated  macrophages  have  been  associated  with  host   defense  against  parasites,  down-­‐regulation  of  inflammation  and  wound  healing.  However,   the  exact  contribution  of  alternatively  activated  macrophages  is  context  specific.  In  some   instances  alternatively  activated  macrophages  are  beneficial  e.g  defense  against  certain   parasites  while  in  others  they  may  promote  pathology  e.g  M2  presence  in  tumors.141   Importantly,  macrophage  polarization  states  are  dynamic  and  individual  cells  can  shift   between  phenotypes  depending  on  external  environment,142–146    once  more  highlighting   the  dynamic  flexibility  of  these  cells.       Utilizing  Gene  Expression  Signatures  to  Identify  Macrophage  Phenotypes   Distinct  gene  signatures  that  reflect  the  opposing  functions  of  macrophages  are  used   to  identify  M1  and  M2  phenotypes.  The  genetic  signatures  of  murine  macrophage   phenotypes  are  well  established,  however,  data  from  human  macrophage  phenotypes   indicates  substantial  variation  in  the  “genetic  signature”  despite  preservation  of  general   phenotypic  function  (see  Figure  1).    Elevated  expression  of  inducible  nitric  oxide  synthase  (iNOS)  or  arginase  1  provides   well-­‐established  markers  for  murine  M1  and  M2  phenotypes,  respectively.  However,   neither  iNOS  nor  arginase  expression  is  significantly  induced  in  polarized  human   macrophage  phenotypes.147  Stimulation  with  IL-­‐4  fails  to  induce  expression  of  arginase  in   equine  neutrophils  suggesting  that  gene  expression  signatures  in  equine  M2  macrophages   may  similarly  differ  from  mice.     Further,  YM1  (chitinase-­‐like  protein)  and  RELMα  are  murine  restricted  as  humans   (and  horses)  lack  these  genetic  orthologs.126  Thus,  this  has  necessitated  investigation  of     26   different  gene  family  members  in  such  as  chitotriosidase  (CHIT  1)148  and  AMCase   (chitinase),  or  YKL-­‐40  (human  chitinase-­‐like  protein)149  or  RELMβ  in  human  macrophages.     Definitive  M2  markers  have  been  difficult  to  establish  due  to  inconsistent  findings.  The   receptor  CD163  is  commonly  used  as  a  human  M2  marker  however,  Th2  cytokines  reduced   expression  of  CD163  in  human  monocyte  derived  macrophages.150  The  lack  of  specific  gene   markers  for  M1  or  M2  phenotypes  in  humans  creates  a  challenge  in  accurately  categorizing   macrophage  phenotypes  as  M1  or  M2  and  limits  the  ability  to  infer  how  these  macrophage   phenotypes  are  induced.     Table  1.  Species  Specific  Differences  in  M1/M2  Polarized  Phenotypes     Illustrates  expression  of  select  genes  in  murine,  bovine  and  human  M1  and  M2  macrophage   phenotypes.  Arrows  indicate  increased  or  decreased  expression.  Grey  boxes  indicate  no   genetic  ortholog.  ND-­‐not  detected,  NC-­‐no  change,  question  mark  indicates  not  yet   determined.     There  are  no  data  comparing  equine  macrophage  M1  and  M2  phenotypes  and  cross  species   variations  limit  direct  translation  of  genetic  markers.  Thus,  there  is  a  fundamental  need  for  in   vitro  characterization  to  establish  reliable  genetic  markers  in  equine  alveolar  macrophage     27   phenotypes.  Establishing  these  gene  expression  templates  will  enhance  our  ability  to   understand  alveolar  macrophage  biology  in-­‐vivo.       Activation  Pathways  of  M1  and  M2  Macrophages   M1  and  M2  activating  stimuli  achieve  polarized  alveolar  macrophage  phenotypes  by   initiating  distinct  signaling  pathways.  The  latter  alter  transcriptional  responses   consequently  controlling  the  expression  pattern  of  cytokines  and  surface  receptors   resulting  in  the  functional  phenotype.   Activation  of  M1  macrophages  occurs  through  stimulation  with  IFNγ  and  LPS.   Activation  of  the  IFNγ  receptor  recruits  Janus  kinase  adaptors  that  then  activate   transcription  factors  STAT1  (signal  transducers  and  activators  of  transcription)  and   interferon  regulatory  factors  (IRF-­‐5).151  STAT1  binds  to  interferon-­‐sequence  response   elements  (ISRE)  in  the  promoter  region  of  M1  signature  genes  such  as  iNOS,  CD80  and  IL-­‐ 12.  Stimulation  of  TLR  by  pathogens  also  stimulates  a  pro-­‐inflammatory  M1  phenotype.   LPS  stimulates  TLR4  and  through  induction  of  MyD88  and  NFκβ  induces  pro-­‐inflammatory   cytokines.  Many  genes  contain  promoter  regions  that  contain  sequences  that  can  be   regulated  by  both  IFNγ  and  LPS  and  consequently  similar  gene  expression  patterns  are   seen  and  LPS  stimulated  macrophages  are  considered  M1-­‐like  phenotypes.152     IL-­‐4  and  IL-­‐13  can  induce  induction  of  M2  phenotypes  by  binding  to  IL-­‐4Rα.  Induction  of   M2  associated  genes  in  alternative  macrophages  involves  a  number  of  transcription  factors,   including  STAT6,  PPRγ  and  C/EBP  but  stimulation  with  IL-­‐4  specifically  induces  the  JAK-­‐ STAT6  pathway  and  STAT6-­‐/-­‐  mice  have  impaired  M2  polarization.  Interleukin-­‐4  also     28   activates  PI3K  (phosphoinositide  3-­‐kinase)  which  augments  STAT6  transcription  and   selectively  activates  certain  M2  associated  markers  (e.g  Arginase  1,  YM1,  FIZZ1).153     Maintenance  of  polarization  states  is  complex  and  in  addition  to  receptor-­‐mediated   activation  a  number  of  other  regulatory  proteins  play  critical  roles.  Maintenance  of  M1/M2   phenotypes  are  also  influenced  by  of  the  balance  of  SOCS1  (suppressor  of  cytokine   signaling-­‐)  and  SOCS3  proteins.  IL-­‐4  induces  SOCS1  (with  concomitant  suppression  of   SOCS3),  which  inhibits  M1  activating  pathways  (e.g  JAK/STAT1,  and  NFκβ).  In  contrast,   IFNγ/LPS  stimulation  induces  SOCS3  which  inhibits  pathways  leading  to  M2  polarization   pathways  (e.g  IL-­‐4  and  TGFβ  receptor  activated  pathways).154,155  However,  SOCS3  is  also   induced  by  IL-­‐10  stimulated  macrophages  (induced  by  STAT3)130  and  can  mediate  early   inhibition  of  LPS  induced  inflammation.156  Furthermore,  the  phosphatase  SHIP  (src   homology  2-­‐domain  containing  inositol  -­‐5’-­‐phosphatase)  is  elevated  in  M1  activation  states   and  represses  the  M2  signaling  proteins  PI3K.153,157     Polarized  lymphocyte  phenotypes  (e.g  Th1,  Th2)  are  fixed  by  chromatin   modifications,158  however,  dynamic  chromatin  remodeling  (epigenetic  regulation)  occurs   in  alternatively  activated  macrophages.  Induction  of  STAT6  (by  IL-­‐4)  results  in  lysine   demethylation  (Histone-­‐3,  Lysine-­‐27)  at  the  promoter  sites  of  M2  marker  genes,  leading  to   a  M2  gene  expression  pattern.  However,  the  demethylated  state  is  dynamic,  and  removal  of   IL-­‐4  results  in  a  return  to  repression  of  M2  marker  genes.159     Thus  there  are  many  factors  involved  in  the  control  of  macrophage  activation  states   that  allow  an  ability  to  fine  tune  function.  While  there  is  cross  regulation  between   heterologous  phenotypes  (e.g  M2  associated  proteins  suppressing  M1  signaling  pathways)   complex  environments  that  contain  both  M1  and  M2  activating  stimuli  can  result  in     29   concurrent  expression  of  both  M1  and  M2  markers.144  This  suggests  that  an  infinite   spectrum  of  macrophage  phenotypes  can  be  induced  in  complex  physiologic  systems.106,145     The  Inflammatory  Response  of  M1/  M2  to  Subsequent  PAMP  Stimulation   M1  and  M2  polarization  states  should  be  considered  “primed  states”  that  can   influence  their  response  to  subsequent  stimulation.106  The  M1  phenotype  exhibits  a   potentiated  inflammatory  response  when  subsequently  stimulated  with  LPS.  IL-­‐4  priming   may  suppress  respiratory  burst  and  production  of  pro-­‐inflammatory  cytokines  when   subsequently  stimulated  with  microbial  agonists  such  as  LPS  and  zymosan.110,160–162   However,  other  data  also  suggests  that  IL-­‐4  priming  is  not  anti-­‐inflammatory  and  actually   enhances  the  pro-­‐inflammatory  response  to  microbial  stimulation.152,163  Critically,  it  is   unknown  how  equine  alveolar  macrophage  M1  and  M2  phenotypes  will  respond  to  stimuli   that  are  relevant  to  RAO.         Macrophage  Phenotypes  and  Airway  Disease   M1  and  M2  phenotypes  prototypically  play  critical  roles  in  host  defense  against   intracellular  bacteria  and  helminths  respectively.  However,  M1  and  M2  phenotypes  are   also  associated  with  chronic  airway  disease  such  as  asthma.  In  murine  models  of  allergic   asthma  Th2  cytokines  are  crutial164  and  alveolar  macrophages  derived  from  this  IL-­‐4/IL-­‐13   enriched  condition  exhibit  prototypical  M2  (IL-­‐4  primed)  polarization  characteristics  (e.g.   YM1,  FIZZ,  ARG-­‐1,  CD206).165–168  Studies  of  murine  allergic  airway  disease  have  provided   evidence  that  the  M2  phenotype  promotes  the  inflammatory  response  and  airway   hyperresponsiveness.167,169,170  The  M2  phenotype  can  increase  Th2  cytokine  production  via     30   stimulation  of  CD4+  lymphocytes  or  can  directly  contribute  to  Th2  effector  cytokines   (increased  production  of  IL-­‐13).171,172  Further,  M2  polarization  is  enhanced  by  IL-­‐33   (overexpressed  in  asthmatics173)  and  the  depletion  of  alveolar  macrophages  reduces  IL-­‐33   driven  airway  inflammation.174  Allergen-­‐induced  acute  exacerbations  are  also  associated   with  elevated  production  TNF,  IL-­‐6,  and  IL-­‐1β  by  murine  alveolar  macrophages.171   Additionally,  the  M2  associated  production  of  pro-­‐fibrotic  mediators  (TGFβ,  L-­‐proline  and   YM1)  contribute  to  the  airway  remodeling  that  characterizes  persistent  allergic  airway   disease.170,175,176  Thus,  M2  macrophages  can  promote  the  allergic  inflammatory  response   through  a  variety  of  mechanisms.   Allergic  airway  disease  is  heterogenous  however,  and  is  not  always  characterized  by   Th2  polarization.49,177,178  Consequently,  other  macrophage  phenotypes  can  be  associated   with  allergic  airway  disease.  Indeed,  murine  neutrophilic-­‐allergic-­‐asthma  is  mediated  by   Th1  and  Th17  lymphocytes  and  isolated  alveolar  macrophages  displayed  an  M1-­‐like   phenotype  (elevated  expression  of  TNFα,  IL-­‐12p40  and  IFNγ)  when  compared  to  murine   eosinophilic-­‐allergic-­‐asthma.179  Thus,  alveolar  macrophages  may  enhance  the  allergic   response  however,  in  contrast,  some  murine  studies  suggest  that  alveolar  macrophages   may  attenuate  airway  inflammation  and  airway  hyperresponsiveness  by  antagonizing   production  of  Th2  cytokines  during  asthma.180  Further,  adoptive  transfer  of  alveolar   macrophages  from  allergy  resistant  rats  modulates  the  airway  hyperresponsiveness  that   develops  in  allergy  sensitive  rats  suggesting  that  a  protective  role  for  alveolar  macrophages   may  be  influenced  by  genetic  background.181,182  Furthermore,  the  alternative  alveolar   macrophage  phenotype  plays  an  important  role  in  healing  the  damaged  lung  and  resolving   inflammation.175       31   Thus  a  number  of  alveolar  macrophage  phenotypes  have  been  detected  in  rodent   models  of  allergic  airway  disease  and  similarly,  a  number  of  different  alveolar  macrophage   phenotypes  have  been  detected  in  human  asthma.  Alternatively  activated  macrophages  can   be  detected  in  bronchoalveolar  lavage172  and  lung  tissue  of  asthmatics183  and  analogous  to   murine  studies,  can  promote  allergic  inflammation  by  stimulating  production  of  Th2   cytokines,184  direct  production  of  Th2  cytokines185  or  promoting  Th2  cell  recruitment.150   However,  the  M2  signature  can  differ  among  studies  and  alveolar  macrophages  may   express  partial  M2  expression  profiles  (expression  of  select  IL-­‐4-­‐assocaited  genes).150         Other  studies  have  detected  M1  macrophages  in  asthmatics,186  and  allergen  challenge   induced  an  M1  phenotype  in  patients  with  eosinophilic-­‐asthma.187  Further,  patients  with   corticosteroid  resistant  asthma  exhibit  classically  activated  alveolar  macrophages  which   may  be  a  consequence  of  increased  environmental  endotoxin  exposure.188  However,   opinion  is  still  widely  divided  as  a  number  of  studies  have  found  no  evidence  for  different   alveolar  macrophage  phenotypes  between  normal  or  asthmatic  patients.47,189     Overall,  many  different  alveolar  macrophage  phenotypes  have  been  associated  with   chronic  inflammatory  airway  disease  and  these  phenotypes  may  perform  a  variety  of   functions  including  potentiation,  suppression  or  resolution  of  disease.169  This  heterogeneity  of   phenotype  may  be  a  consequence  of  a  number  of  factors  including  the  disease  (or  disease   model),  the  stage  of  disease  (chronic  versus  exacerbation)  or  the  genetic  background  and   reflects  the  integration  of  the  complexity  of  disease  and  the  dynamic  plasticity  of   macrophages.           32   Section  3.  Conclusion  and  Hypotheses   RAO  is  a  complicated  hypersensitivity  disease  that  differs  significantly  from  many   animal  models  of  allergic  airway  disease  and  atopic  human  asthma.  Principally,  RAO  does   not  appear  to  be  mediated  by  allergen-­‐specific  IgE,  infiltrating  cells  are  neutrophils  (rather   than  eosinophils)  and  it  is  not  clearly  associated  with  Th2  or  Th1  lymphocytes.  Thus  the   relative  importance  of  an  allergen-­‐specific  response  is  unclear,  and  the  immune   mechanisms  that  result  in  disease  pathogenesis  are  unknown.  The  stimulus  for  RAO-­‐ exacerbation,  HD,  contains  a  multitude  of  PAMPs.  Exposure  to  HD  induces  pulmonary   neutrophilia,  and  elevations  in  systemic  inflammation  in  both  control  horses  and  RAO-­‐ susceptible  horses  suggesting  that  HD  incites  an  inflammatory  response  via  stimulation  of   the  innate  immune  system.  However,  as  the  inflammatory  response  observed  in  RAO-­‐ susceptible  horses  is  greatly  enhanced  compared  to  control  horses,  it  is  possible  that  the   increased  responsiveness  to  HD  could  be  mediated  by  alveolar  macrophage  phenotypes   that  possess  differential  sensitivity.     Macrophage  M1  and  M2  phenotypes  represent  extremes  of  macrophage  function   with  M1  cells  performing  a  pro-­‐inflammatory  function  and  M2  cells  performing  anti-­‐ inflammatory  or  regulatory  functions.  Macrophage  M1  and  M2  phenotypes  are   characterized  by  distinct  transcriptional  expression  patterns,  however,  the  transcriptional   expression  patterns  are  species  specific  and  there  are  no  data  evaluating  transcriptional   signatures  in  equine  alveolar  macrophage  M1/M2  phenotypes.  Further,  murine  studies   have  produced  conflicting  evidence  relating  to  the  suppressive  effects  of  IL-­‐4  primed  M2   macrophages  upon  subsequent  microbial  stimulation.  To  address  this  gap  in  knowledge  the   first  hypothesis  was  generated:     33     Hypothesis  1:  The  gene  expression  markers  upregulated  in  equine  M1  and  M2  alveolar   macrophages  will  differ  from  those  established  in  other  species  but  equine  M1  and  M2  will   maintain  pro-­‐inflammatory  and  anti-­‐inflammatory  function  respectively.   Specific  Aim  1:  To  investigate  this  hypothesis,  commonly  used  markers  will  be  evaluated  in   equine  M1  and  M2  alveolar  macrophages.  The  function  of  the  M1  and  M2  phenotypes  will  be   determined  based  on  the  pro-­‐  and  anti-­‐inflammatory  gene  expression  response  following   stimulation  with  HD  and  its  individual  components.    The  results  of  this  investigation  are  presented  in  chapter  2.     It  is  possible  that  control  horses  regulate  the  inflammatory  response  to  HD  via   development  of  an  immune-­‐regulatory  M2  phenotype,  while  RAO-­‐exacerbation  is   associated  with  development  of  an  M1  phenotype  or  an  absence  of  an  immune  regulatory   M2  phenotype.  However  a  comprehensive  evaluation  of  alveolar  macrophage  M1  and  M2   phenotypes  in  response  to  HD  has  not  been  performed  in  horses.  To  address  this  gap  in   knowledge  two  hypotheses  were  generated:     Hypothesis  2:  After  inhalation  of  HD,  alveolar  macrophages  from  RAO-­‐susceptible  horses   will  exhibit  an  M1  and  control  horses  will  exhibit  an  M2  phenotype.   Specific  Aim  2:  To  investigate  this  hypothesis,  the  phenotype  of  alveolar  macrophages   isolated  from  RAO-­‐susceptible  and  control  horses  before  and  after  inhalation  of  HD  will  be   characterized  based  on  the  expression  of  M1  and  M2  associated  genes.       34   Hypothesis  3:  After  inhalation  of  HD,  alveolar  macrophages  from  RAO-­‐susceptible  horses   will  exhibit  a  pro-­‐inflammatory  function  and  control  horses  will  exhibit  an  anti-­‐inflammatory   function  when  stimulated  with  individual  components  of  HD  in  vitro.   Specific  Aim  3:  To  investigate  this  hypothesis,  alveolar  macrophages  isolated  from  RAO-­‐ susceptible  and  control  horses  before  and  after  inhalation  of  HD  will  be  stimulated  with   individual  components  of  HD  in  vitro  and  the  gene  expression  of  pro-­‐inflammatory  and  anti-­‐ inflammatory  cytokines  will  be  evaluated.      The  results  of  these  investigations  are  presented  in  chapter  3.                                 35   Chapter  2.     Polarized  Equine  Alveolar  Macrophages  Have  a  Species  Specific  Gene  Expression   Profile   Abstract   Background:  Polarized  murine  macrophage  phenotypes  M1  and  M2  express  distinct   transcriptional  signatures.  M1  (classically-­‐activated)  is  anti-­‐microbial  and  expresses  nitric   oxide  synthase,  pro-­‐inflammatory  cytokines  and  co-­‐stimulatory  molecules  (TNFα,  IL-­‐12,   CD80)  while  M2  (alternatively-­‐activated)  is  immune-­‐regulatory  and  expresses  the   regulatory  cytokines  (IL-­‐10,  TGFβ),  arginase,  and  scavenger  receptors.  It  is  unclear  if   equine  alveolar  macrophages  (AM)  develop  similar  phenotype-­‐specific  transcriptional   signatures.     Hypothesis:  Gene  expression  markers  upregulated  in  equine  M1  and  M2  alveolar   macrophages  will  differ  from  those  established  in  other  species  but  equine  M1  and  M2  will   maintain  pro-­‐inflammatory  and  anti-­‐inflammatory  function  respectively.   Methods:  Equine  AMs  were  cultured  in  IFNγ+LPS  or  IL-­‐4  generating  M1  and  M2   phenotypes  respectively.  The  gene  expression  of  predicted  surrogate  M1/M2  genes  and  the   effect  of  polarization  on  the  response  to  LPS,  peptidoglycan,  zymosan  and  hay  dust   suspension  was  evaluated.     Results:  Equine  M1s  were  characterized  by  increased  expression  of  pro-­‐inflammatory   cytokines  (TNFα,  IL-­‐1β,  IL-­‐12p40),  the  chemokine  IL-­‐8,  co-­‐stimulatory  molecule  CD80  and   elevated  regulatory  cytokine  IL-­‐10.  However,  unlike  murine  phenotypes,  equine  M1s  did   not  express  iNOS.  Equine  M2s  were  characterized  by  elevated  scavenger  receptor  CD206   and  low  expression  of  M1  associated  genes.  However,  dissimilar  to  murine  or  human  M2s     36   neither  arginase  nor  β-­‐glucan  receptors  were  M2  associated.  Further,  compared  to  the   equine  M1  that  potentiated  pro-­‐inflammatory  gene  expression  when  stimulated,  the  M2   potently  suppressed  pro-­‐inflammatory  cytokines  and  IL-­‐10.   Conclusions/clinical  importance:  The  transcriptional  profile  of  equine  M1/M2  AMs  is   species-­‐specific.  This  first  systematic  comparison  of  the  transcriptional  signature  of  equine   AM  M1  and  M2  phenotypes  will  assist  in  studying  their  role  in  equine  pulmonary  disease.     Introduction   Macrophages  can  respond  to  changes  in  their  microenvironment  (e.g.  the  presence   of  host-­‐derived  factors  and  pathogen  associated  molecular  patterns  (PAMP))  by  adapting   into  functionally  distinct  phenotypes.106,142  Though  a  spectrum  of  phenotypes  exists,  two   major  phenotypes  have  been  well  categorized  in  the  mouse.106  The  classically-­‐activated   phenotype  (M1)  develops  under  pro-­‐inflammatory  stimuli  such  as  IFNγ  or  bacterial   pathogens.  In  contrast,  IL-­‐4  (derived  from  Th2  polarized  inflammation)  fosters   development  of  so-­‐called  alternatively-­‐activated  macrophages  (M2)  with  anthelmintic  and   immune-­‐regulatory  activities.190  Accordingly,  these  primed  phenotypes  have  divergent   contributions  to  the  course  and  resolution  of  inflammation.     Distinct  gene  expression  signatures  that  reflect  their  opposing  functions  are  used  as   surrogates  to  identify  the  presence  M1  and  M2  phenotypes.  Murine  M1  macrophages   exhibit  high  expression  of  antimicrobial  nitric  oxide  synthase  (iNOS),  pro-­‐inflammatory   cytokines  and  co-­‐stimulatory  molecules  (TNFαhi,  IL-­‐6hi,  IL-­‐12hi  CD80hi).  In  contrast,  murine   M2  macrophages  exhibit  elevated  regulatory  cytokines  (IL-­‐10hi,  TGFβhi),  low  expression  of   pro-­‐inflammatory  cytokines,  simultaneously  elevated  arginase  1  (Arg1)  expression,  and  a     37   range  of  scavenger  receptors  including  mannose  receptor  1  (CD206)  and  β-­‐glucan   receptor.106126     The  M1  and  M2  phenotypes  should  be  considered  “primed”  states  that  can  influence  the   response  to  subsequent  stimulation  with  pathogen  associated  molecular  patterns190  and   while  M1  macrophages  typically  potentiate  the  pro-­‐inflammatory  response,  M2   macrophages  typically  suppress  TLR  induced  inflammation.  However,  when  stimulated   with  very  potent  microbial  ligands  (e.g  LPS),  initial  M2-­‐priming  may  have  little  influence   and  or  may  potentiate  the  release  of  certain  pro-­‐inflammatory/M1  cytokines  (such  as  IL-­‐6,   IL-­‐12a),152  underscoring  the  complexity  of  macrophage  responsiveness  to  a  broad  range  of   stimuli.     The  alveolar  macrophage  plays  an  essential  role  in  pulmonary  homeostasis,  defense   against  inhaled  substances  and  resolution  of  inflammation.  Indeed,  changes  in  M1/M2   alveolar  macrophage  phenotype  status  are  thought  to  play  an  important  role  in   pathogenesis  of  many  pulmonary  diseases  including  infectious,191  parasitic,192  allergic,167   and  occupational  disease.193  These  pulmonary  ailments  also  occur  in  horses,  however,  little   is  known  about  equine  M1/M2  transcriptional  signatures  or  if  analogous  phenotypes  occur   in  equine  lungs  during  disease.  Thus,  the  aim  of  this  study  was  to  characterize  the   transcriptional  signature  of  equine  alveolar  macrophage  M1  and  M2  phenotypes.     Conventional  equine  husbandry  (stabled  and  fed  hay)  makes  exposure  to  hay  dust   ubiquitous  even  though  inhalation  has  been  associated  with  both  development  of  lower   airway  inflammation194  and  recurrent  airway  obstruction  (RAO).1  As  M1  and  M2   phenotypes  from  other  species  typically  respond  to  inflammatory  stimuli  with  divergent     38   inflammatory  responses,  we  further  sought  to  characterize  the  effect  of  M1  and  M2   phenotypes  on  elements  present  in  hay  dust  and  select  constituents.     Using  primary  equine  alveolar  macrophages  cultured  in  M1  (IFNγ+LPS  activation)  or   alternatively  activated  M2  (IL-­‐4  activation)  polarizing  conditions,  we  evaluated  the   expression  of  surrogate  M1  and  M2  genes  and  analyzed  the  response  of  polarized  equine   macrophages  to  subsequent  stimulation  with  HD  and  individual  constituents.  This  is  the   first  comprehensive  analysis  of  equine  alveolar  macrophage  polarization  responses  to  a   group  of  relevant  inflammatory  stimuli.     Materials  and  Methods   Animals     Alveolar  macrophages  were  isolated  from  the  bronchoalveolar  lavage  fluid  (BALF)  of  eight   clinically  healthy  horses  (mean  age  13.4  years,  range  4-­‐22,  mixed  light  breeds:  TB,  STB,  QH,   Arabian,  Grade).  Horses  were  selected  based  on  clinical  history  and  physical  exam,  and  had   remained  free  from  obstructive  airway  disease  when  challenged  with  hay  straw.  All  horses   were  maintained  at  pasture  for  at  least  1  month  before  cell  collection  and  were  fed  a   supplemental  complete  pelleted  feed.  The  Michigan  State  University  Institutional  Animal   Care  and  Use  Committee  approved  all  procedures.     Collection  of  Hay  Dust   The  hay  dust  (HD)  was  collected  as  previously  described62  from  hay  with  proven    ability  to   induce  pulmonary  inflammation  in  RAO-­‐susceptible  horses.  The  same  batch  of  HD  was   used  for  all  parts  of  the  investigation.  Dry  dust  was  further  size  fractionated  using  a  series     39   of  stacked  metal  sieves  (USA  Standard  Testing  Metal  Sieves)  and  the  smallest  fraction  of  HD   (<  43  micron  diameter)  was  collected.  Analysis  of  aerodynamic  properties  (Aerodynamic   Particle  Size  Spectrometer,  TSI  3322)  indicated  a  median  aerodynamic  diameter  of  1.5um   indicating  that  the  particles  could  be  deposited  into  the  alveoli  when  inhaled.  The  HD  was   suspended  in  PBS  at  a  concentration  of  1mg/ml,  vortexed  for  5  minutes,  placed  in  aliquots   and  stored  at  -­‐80F.       Isolation  of  Alveolar  Macrophages   Horses  were  sedated  with  detomidine  hydrochloride  (10ug/kg  IV)  (Zoetis)  and   butorphanol  tartrate  (0.02mg/kg,  IV)  (Zoetis)  and  bronchoalveolar  lavage  (BAL)  was   performed  using  a  sterilized  video  endoscope  passed  intra-­‐nasally  and  wedged  in  a   peripheral  bronchus.  A  total  of  500ml  of  sterile  saline  was  infused  in  200ml  aliquots  and   lavage  fluid  was  retrieved  using  gentle  manual  suction  using  a  60ml  syringe.  The  BAL  fluid   (BALF)  was  immediately  placed  on  ice  and  processed  within  30  minutes.     The  BALF  was  passed  through  an  80μm  sterile  filter  and  centrifuged  (250xg,  4°C,  10  min).   Cell  pellets  were  re-­‐suspended  in  sterile  medium  (RPMI+  L-­‐glutamine  supplemented  with   antibiotic/antimycotic,  5%  heat  inactivated  equine  serum,  2mM  EDTA)  and  washed  twice.   Cell  number  and  viability  were  assessed  using  a  hemocytometer  and  trypan  blue  exclusion   respectively.   Alveolar  macrophages  were  isolated  from  the  mixed  cell  population  by  magnetically-­‐ activated  cell  sorting  (MACS)  using  negative  selection.  Briefly,  cells  were  resuspended  in   medium  (2x10^7  cells/ml)  and  incubated  with  antibody  against  lymphocytes  (mouse   monoclonal  IgG,  HB88)  at  4°C  for  30  minutes.  After  washing,  cells  were  incubated  with     40   secondary  antibodies  conjugated  to  metal  beads  (anti-­‐mouse  IgG  polyclonal)  at  4°C  for  20   minutes  and  washed  before  passing  through  magnetic  columns  to  remove  labeled   lymphocytes.  The  eluted  cells  (enriched  macrophage  population,  87%±8:  mean±sd)  were   collected  and  re-­‐suspended  (6.25x10^5  cells/ml)  in  sterile  medium  (minus  EDTA).  A  small   aliquot  was  collected  for  cell  cytology  (cytospin  preparation)  and  to  assess  viability  by  use   of  light  microscopy  (400  cell  count  of  Diff  Quik  stained  slides)  and  trypan  blue  exclusion,   respectively.     Cell  Culture   The  alveolar  macrophages  were  plated  into  sterile  cell  culture  dishes  (12-­‐well  at  5.x10^5   cells/well)  and  incubated  (5%  CO2,  37°C)  for  2  hours  to  allow  adherence.  After  a  medium   change,  adherent  cells  were  incubated  for  20  hours  with  either  medium  (non-­‐polarized   control),  recombinant  equine  IFNγ  (20ng/ml)  +LPS  (1ng/ml)  or  recombinant  equine  IL-­‐4   (20ng/ml).  Cells  were  then  harvested  and  RNA  was  extracted  and  stored  (-­‐80°C)  until  gene   expression  was  evaluated.     In  a  separate  experiment,  alveolar  macrophages  were  stimulated  either  with  recombinant   equine  IFNγ    (20ng/ml)  +LPS  (1ng/ml)  or  IL-­‐4  (20ng/ml)  for  24  hours  and  then  incubated   for  16  hours  with  medium  alone  or,  medium  plus  LPS  (100ng/ml),  HD  (0.02ug/ul),   peptidoglycan  (1ug/ml)  or  zymosan  (10ug/ml).         RNA  Extraction,  Reverse  Transcription  and  Quantitative  Real  Time  PCR   Cells  were  harvested  by  adding  RLT-­‐lysis  buffer  plus  (Quiagen)  and  homogenized  using   Qiagen  QIAshredderTM  spin  columns.  Total  RNA  was  extracted  and  purified  using  Qiagen     41   RNeasy®  Plus  Micro  Kit  which  included  a  step  for  genomic  DNA  removal.  RNA   concentration  was  measured  using  Qubit®  2.0  fluorometer  and  integrity  of  RNA  (RIN)  was   assessed  using  the  Agilent  2100  Bioanalyzer  (Agilent  Technologies)  and  a  RIN  score  of  >6.5   was  considered  acceptable  for  qPCR.195  Equal  concentrations  of  RNA  were  reverse   transcribed  (High  Capacity  cDNA  Reverse  Transcription  kit,  Applied  biosciences)  to  create   cDNA.    Six  µl  of  cDNA  was  then  pre-­‐amplified  using  TaqMan®  PreAmp  Master  Mix  kit   (according  to  manufacturer's  instructions)  and  amplified  cDNA  was  stored  at  -­‐80°C  until   further  analysis.  Amplification  uniformity  was  assured  for  all  gene  assays  tested.   Quantitative  PCR  was  performed  using  predesigned  TaqMan®  gene  expression  assays  (Life   Technologies)  and  TaqMan®  Gene  Expression  Master  Mix.  When  predesigned  assays  were   unavailable,  primers  and  probes  were  designed  using  the  Custom  TaqMan®  Assay  Design   Tool  or  Custom  TaqMan®  (Plus)  Assay  Design  Tool  (Table  2).    All  samples  were  run  at  once,   in  triplicate  on  384  well  optical  plates  ABI  7900HT  real  time  PCR  machine  (Applied   Biosciences)  using  standard  conditions  (50°C  (2min),  95°C  (10  min),  40  cycles  (95°C/15   seconds,  60°C/1min).  The  average  of  two  stable  endogenous  genes,  hypoxanthine   ribosyltransferase  (HPRT)  and  elongation  factor  1α  (ELF1)  were  used  to  normalize  each   sample  (See  Appendix  1).         42   Table  2.  Life  Technologies  Assay  ID  for  Proprietary  (A)  and  Custom  Designed  (B)  Taqman   Gene  Assays     A.   Gene   Life   Gene   Life   Technologies™   Technologies™   Assay  ID   Assay  ID   HPRT   Ec03470222_m1   TLR2   Ec03818334_s1   TNFα   Ec03467871_m1   TLR4   Ec03468994_m1   IL1β   Ec04260296_g1   Arg-­‐2   Ec03470258_m1   IL-­‐12p40   Ec03468777_m1   TGFβ   Ec03468030_m1   iNOS   Ec03467519_m1   IL-­‐10   Ec03468647_m1   IL-­‐8   Ec03468860_m1   Chit   Ec03818149_m1   IL-­‐6   Ec03468678_m1   B.   Gene   Life   NCBI  Gene   Technologies™   Reference   Forward  (5’-­‐3’)   Reverse  (5’-­‐3’)   Probe  (5’-­‐3’)   Assay  ID   EF1   AIXOZ4N   AY237113.1   CCACCAACTCGTC CAACTGATAAG   GACAGTACCGAT ACCACCAATTTTG   CCCTTGCGTCTGC CCC   CD206   AIGJQ7N   XM_005606899.1   TGTGCCCAATCAA ATAGCAGTAGAA   CCAGCCCTTCCGG CAGC   CD80   AII1NJ3   XM_005601958.1   CGCCAGGAATAG TGGAAGTAGAC    ACCTGACTTCCGT ATGGATTTCCAAC TTCAGCTATGGT   TCAGAGCCAACTT TCC   Relmβ*   AJAAZEG   XM_001503230.1   B-­‐glucan*   AJ89J89   XM_001499567.3   Arg  -­‐1*   AJCSVPA   XM_001503285.2   GATGTTATTGG   Assay  ID  and  NCBI  gene  reference  provided  for  all  genes.  Primer  and  probe  sequences  are   provided  when  available.  *Primer  and  probe  sequences  designed  using  Custom  TaqMan®   (Plus)  Assay  Design  Tool  are  proprietary  and  withheld  by  Life  TechnologiesTM    (Table  2B).   43   Data  Analysis   Normality  of  errors  of  each  variable  was  assessed  using  visual  inspection  of  error   histogram,  probability  plots,  and  normality  testing  using  Shapiro-­‐Wilk.  Normally   distributed  data  were  analyzed  using  an  ANOVA  with  the  fixed  effect  of  treatment  and  the   random  effect  of  horse  (SAS  Proc  Mixed).  Errors  that  were  not  normally  distributed  were   log  transformed  (IL-­‐10,  β-­‐glucan  receptor-­‐inflammatory  response  of  polarized  cells)  and   normality  of  transformed  data  was  assessed  as  described.  Non-­‐parametric  variables  (TLR2,   TLR4,  β-­‐glucan  receptor  -­‐gene  expression  of  polarized  cells)  were  analyzed  using  the   paired  Wilcoxon  signed-­‐rank  test.  Significance  was  set  at  (P<0.05).  Results  are  expressed  as   fold  change  (using  2-­‐  ΔΔCT)  compared  to  untreated,  non-­‐polarized  cells.  In  the  case  of  IL-­‐ 12p40,  expression  was  undetectable  in  control  and  IL-­‐4  treated  cells  and  an  empirically  CT   value  of  35  was  used  to  allow  approximation  of  fold  change  in  IFNγ  /LPS  treated  or  agonist   stimulated  cells.    Results   Gene  Expression  of  Polarized  Phenotypes     Our  first  goal  was  to  determine  the  effect  of  M1  polarizing  conditions  on  the  gene   expression  profile  in  equine  alveolar  macrophages.    Incubation  of  alveolar  macrophages   with  IFNγ  +LPS  induced  significant  up-­‐regulation  of  the  pro-­‐inflammatory  cytokines  (TNFα,   IL-­‐8,  IL-­‐12p40,  IL-­‐6)  that  are  typically  associated  with  the  M1  phenotype  in  other  animal   species158,109  (Table  3A).  Furthermore,  we  observed  robust  up-­‐regulation  of  CD80  co-­‐ stimulatory  molecule,  a  functional  hallmark  of  M1  macrophage  polarization.  Expression  of   the  major  M1  marker  for  rodent  species,  the  inducible  nitric  oxide  synthase  (iNOS),  was  not   44   detected  in  any  equine  alveolar  macrophage  samples.  However,  iNOS  expression  was   detected  in  separate  samples  of  equine  lung  tissue  (data  not  shown),  indicating  that  our   methodology  was  appropriate  for  the  detection  of  equine  iNOS  mRNA.     Similarly  to  non-­‐equine  species  putative  M2  marker  genes  arginases,  mannose  receptor-­‐ CD206,  and  chitotriosidase  were  not  significantly  induced  by  M1  polarizing  conditions,   however  the  latter  resulted  in  a  highly  variable  up  and  down  regulation  of  β-­‐glucan   receptor  (Table  3B).  Finally,  stimulation  with  IFNγ  +LPS  had  also  no  significant  effect  on   other  pathogen  recognition  receptors  (TLR2  and  TLR4)  mRNA  levels  but  did  induce   significant  expression  of  the  regulatory  cytokine,  IL-­‐10.     Our  second  goal  was  to  determine  the  effect  of  M2  polarizing  conditions  on  gene   expression  profile  in  equine  alveolar  macrophages.  In  contrast,  with  IFNγ  +LPS,  the  IL-­‐4   had  no  significant  effect  on  the  expression  of  M1/pro-­‐inflammatory  cytokines  TNFα  and  IL-­‐ 8  while  IL-­‐12p40  was  undetectable,  similar  to  non-­‐polarized  control  cells  (Table  3A).  Of   note,  although  not  significant,  IL-­‐6  expression  was  elevated  in  3  of  4  samples  and  when  AM   were  incubated  for  a  longer  duration  (40  hours)  this  elevation  reached  significance  (data   not  shown).  IL-­‐4  treatment  also  significantly  suppressed  expression  of  IL-­‐10.  Finally,  IL-­‐4   treatment  had  no  affected  on  the  expression  of  CD80  and  the  iNOS  remained  undetectable   in  IL-­‐4  stimulated  cells.  The  major  M2  macrophage  marker  of  rodent  species,  arginase-­‐I   was  not  detected  in  any  alveolar  macrophage  samples  but  was  detected  in  equine  liver   (data  not  shown).  Rather,  the  alternative  isoform,  arginase-­‐II,  was  detectable  but   interestingly,  significantly  suppressed  by  IL-­‐4  (compared  to  control  alveolar  macrophages).   Furthermore,  arginase  II  expression  level  post-­‐IL4  stimulation  was  not  significantly   different  from  IFNγ  +LPS  stimulated  cells.    IL-­‐4  treatment  significantly  up-­‐regulated  the   45   mannose  receptor  CD206  but  had  no  effect  on  the  β-­‐glucan  receptor,  which  showed  a  large   degree  of  variability  between  horses.  Further,  there  was  no  significant  effect  on  gene   expression  of  TGFβ  or  chitotriosidase  expression.  Relmβ  could  not  be  detected  in  any   alveolar  macrophage  samples.     Table  3.  Candidate  Genes  for  M1  (A)  and  M2  (B)  Macrophage  Phenotypes   A.   Gene   TNF   IL-­‐8   IL-­‐12p40   IL-­‐6   CD80   iNOS   TLR2   TLR4   IFNγ/LPS   IL-­‐4   48±23    *‡   17.19±14.29  *‡   305±55.82  #   38.68±28.47  *   50±65  *‡   ND   1±0.67   1.0±3.36   1.42±1.9   -­‐2.7±0.92   ND   12.8±12   -­‐6±6.37   ND   -­‐1.05±1.52   2.13±0.73   IFNγ/LPS   IL-­‐4   -­‐1.04±3.04   1.6±1.07   25.07±45.91   46.16±45.22  *   5.77±7.2   -­‐1.3±1.94   ND   -­‐11.46±7.29  *   13.72±2.9  *‡   -­‐28±48   -­‐9.4±7.2    *‡   -­‐1.36±2.7   -­‐1.29±0.168   ND   B.   Gene   Arg  2   CD206   B-­‐glucan  R   IL-­‐10   TGFβ   Chitotriosidase   Relmβ   46   Table  3  (cont'd)    The  effect  of  IFNγ  +LPS  and  IL-­‐4  stimulation  on  the  gene  expression  of  M1  and  M2   anticipated  genes  in  equine  alveolar  macrophages  (Tables  A  and  B  respectively).  Gene   expression  is  presented  as  fold  change  (mean±sd)  compared  to  untreated  control  cells.   *p≤0.05,  compared  to  non-­‐polarized  (control)  alveolar  macrophages,  ‡  p≤0.05  comparing M1  and  M2  alveolar  macrophages.  ND,  not  detectable.  #  note  fold  change  calculated  using   empirical  CT  value  in  non-­‐polarized  control  cells  (see  methods)  (n=4)   PAMP  Stimulation  of  Polarized  Alveolar  Macrophage  Phenotypes   Inflammatory  Cytokines   We  next  determined  how  AM  polarization  state  would  affect  gene  expression  of   inflammatory  cytokines  when  stimulated  with  peptidoglycan,  zymosan,  LPS  and  HD.   Compared  to  non-­‐polarized  cells,  IFNγ+LPS  polarization  tended  to  enhance  the  expression   of  pro-­‐inflammatory  cytokines  IL-­‐1β,  IL-­‐8,  IL-­‐6  and  IL-­‐12p40  (Figure  1A-­‐D).  This   potentiation  was  significant  with  LPS  and  HD  stimulation  but  not  peptidoglycan  and   zymosan  where  a  large  degree  in  variability  was  observed.     In  contrast,  the  M2  polarization  state  resulted  in  significant  down-­‐regulation  of  the   expression  of  pro-­‐inflammatory  IL-­‐1β  and  IL-­‐8  and  IL-­‐12p40  (compared  to  non-­‐polarized   control  cells)  although  the  difference  was  not  significant  in  IL-­‐8  expression  in  LPS   stimulated  cells.  However,  this  suppressive  effect  did  not  extend  to  expression  of  IL-­‐6,   which  was  expressed  similarly  in  M2  and  non-­‐polarized  cells  (Figure  1C).     47   Figure  1.  The  Effect  of  Polarization  on  PAMP  Stimulation  of  Inflammatory  Cytokines   IL-1 B. 100,000 * Fold Change 10,000 150 * 1,000 100 10 *† *† IL-8 200 non-polarized IFNγ & LPS IL-4 Fold Change A. 100 * 50 *† *† 1 Zym Pep Pep IL-6 D. 10000 15000 8000 10000 6000 5000 4000 † * 75 * Zym LPS *† LPS HD 2000 1000 800 400 † Pep Zym 200 * * 600 25 † IL-12p40 20000 125 *† 0 HD Fold Change Fold Change C. LPS *† * *† *† *† *† 0 HD Pep Zym LPS HD Gene  expression  of  inflammatory  cytokines  in  non-­‐polarized  controls  (white)  and  IFNγ/LPS   (black)  or  IL-­‐4  (grey)  stimulated  macrophages  following  stimulation  with  peptidoglycan   (Pep),  Zymosan  (Zym)  lipopolysaccharide  (LPS)  or  hay  dust  (HD,  n=3)  for  16  hours.  Data   are  expressed  as  fold  change  compared  to  non-­‐polarized-­‐non-­‐stimulated  control  (mean   ±sem).    *p≤0.05,  compared  to  non-­‐polarized  control,  †p≤0.05  compared  to  IFNγ/LPS   treatment.  Note  differing  y-­‐axis  scale  among  graphs.       48   Regulatory  Cytokines   Expression  of  IL-­‐10  was  similarly  elevated  in  both  M1  and  non-­‐polarized  cells  (Figure  2A).   However,  the  M2  phenotype  significantly  suppressed  IL-­‐10  expression  compared  to  non-­‐ polarized  controls.  Furthermore,  a  large  degree  of  variability  was  observed  in  TGFβ   expression  and  there  was  no  significant  difference  effect  of  either  polarization  state  (Figure   2B).     Figure  2.  The  Effect  of  Polarization  on  PAMP  Stimulation  of  Regulatory  Cytokines   A. IL-10 B. * 50 * Pep * * 1 Zym 125 Fold Change Fold Change 100 -20 150 non-polarized IFNγ & LPS IL-4 150 LPS TGFb 100 75 50 25 HD Pep Zym LPS HD Gene  expression  of  regulatory  cytokines  in  non-­‐polarized  controls  (white)  and  IFNγ/LPS   (black)  or  IL-­‐4  (grey)  stimulated  macrophages  following  stimulation  with  peptidoglycan   (Pep),  Zymosan  (Zym)  lipopolysaccharide  (LPS)  or  hay  dust  (HD,  n=3)  for  16  hours.  Data   are  expressed  as  fold  change  compared  to  non-­‐polarized-­‐non-­‐stimulated  control  (mean   ±sem).    *p≤0.05,  compared  to  non-­‐polarized  control.   49   Surface  Receptors   As  demonstrated  above  (Table  3B)  the  M1  phenotype  had  no  effect  on  expression  of  CD206   (Figure  3A)  except  for  zymosan  stimulated  cells  in  which  CD206  was  slightly  but   significantly  suppressed.  In  contrast,  M2  polarization  induced  significantly  greater   expression  of  CD206  across  all  stimulants  (Figure  3A).  Polarization  states  had  opposite   effects  on  CD80  expression,  which  was  significantly  enhanced  by  M1  but  significantly   suppressed  by  M2  polarized  macrophages  in  the  presence  of  all  stimuli  (Figure  3B).     Figure  3.  The  Effect  Of  Polarization  on  PAMP  Stimulation  of  Surface  Receptors   CD206 non-polarized IFNγ & LPS IL-4 25 Fold Change 20 B. *† *† 15 * 20 *† 10 5 CD80 25 *† Fold Change A. 15 0 * * 10 5 * * *† *† *† *† 0 Pep Zym LPS HD Pep Zym LPS HD Gene  expression  of  surface  receptors  in  non-­‐polarized  controls  (white)  and  IFNγ/LPS   (black)  or  IL-­‐4  (grey)  stimulated  macrophages  following  stimulation  with  peptidoglycan   (Pep),  Zymosan  (Zym)  lipopolysaccharide  (LPS)  or  hay  dust  (HD,  n=3)  for  16  hours.  Data   are  expressed  as  fold  change  compared  to  non-­‐polarized-­‐non-­‐stimulated     50   Figure  3  (cont'd).   control  (mean  ±sem).    *p≤0.05,  compared  to  non-­‐polarized  control.  †p≤0.05  compared  to   IFNγ/LPS  treatment.   Discussion     M1  and  M2  macrophage  phenotypes  have  been  associated  with  human  chronic   inflammatory  pulmonary  disease  and  animal  models  of  pulmonary  disease.168  However,   there  is  little  information  on  transcriptional  signatures  of  equine  M1  and  M2  phenotypes.   This  limits  the  ability  to  detect  or  accurately  categorize  macrophage  phenotypes  isolated   from  diseased  equine  lung.  Our  goal  was  to  characterize  equine  M1  and  M2  phenotypes  in   the  context  of  respiratory  stimuli  relevant  to  equine  respiratory  diseases,  such  as  RAO,   which  in  future  could  be  applicable  to  study  alveolar  macrophages  from  lungs  of  horses   affected  with  pulmonary  disorders.  With  this  in  mind,  we  specifically  studied  polarization   of  the  alveolar  macrophage  (rather  than  blood  derived  macrophages)  and  used  pro-­‐ inflammatory  agonists  that  represented  stimuli  to  which  horses  are  routinely  exposed   through  inhalation  of  stable  dust.  Phenotypic  characterization  was  comprised  of  1)  defining   the  transcriptional  signature  under  polarizing  stimuli  and  2)  determining  how  these   polarization  states  might  influence  the  response  to  the  pro-­‐inflammatory  stimuli.     Our  data  demonstrate  that,  similar  to  other  species,  equine  alveolar  macrophages  can  also   develop  into  distinct  M1  and  M2  polar  phenotypes  and  present  the  distinct  transcriptional   signatures.  However,  we  identify  some  key  species-­‐specific  differences  in  the   transcriptional  profile  between  equine  and  murine  polarized  macrophages.  Further,  we   demonstrate  that  M1  and  M2  states  significantly  impact  the  magnitude  of  gene  expression   51   of  pro-­‐inflammatory/regulatory  cytokines  and  surface  receptors  when  stimulated  with   stimuli  that  are  abundant  in  the  stable  environment.             The  transcriptional  signature  of  the  equine  alveolar  M1  (IFNγ+LPS  activated)  was   characterized  by  an  overall  enhanced  pro-­‐inflammatory  profile  (increased  expression  of   TNFα,  IL-­‐8,  IL-­‐12p40,  CD80),  which  is  similar  to  that  of  other  species.  However,   unexpectedly,  this  phenotype  also  demonstrated  elevated  expression  of  regulatory   cytokine  IL-­‐10,  indicating  that,  in  addition  to  their  dominant  pro-­‐inflammatory  profile   equine  M1s  are  capable  of  inducing  immune-­‐regulatory  signals.     In  contrast,  the  equine  M2  (IL-­‐4  activated)  could  be  differentiated  from  the  equine  M1   based  on  the  combination  of  no/low  induction  of  pro-­‐inflammatory  genes  and   simultaneous  high  expression  of  CD206.  This  is  consistent  with  M2  characteristics  found  in   other  species.  However,  we  also  found  important  differences.  In  mice  M2  prototypical   markers  resistin-­‐like  molecule  (Relm)α/FIZZ1  and  Ym1  are  specifically  induced  by  IL-­‐4   exposure  (and  STAT6  activation)  and  their  expression  marks  a  predominant  Th2/M2   response.  However,  expression  of  YM1  (a  member  of  the  chitinase  family)  and  FIZZ1   (member  of  resistin  family)196  are  restricted  to  mice.126,125  Thus,  in  this  study  we  explored   expression  of  alternative  equine  homologues  within  the  same  gene  families:  chitotriosidase   and  Relm  β,  however,  neither  was  upregulated  with  the  equine  M2  polarization.     A  prototypical  feature  of  murine  M1  and  M2  macrophages  phenotypes  is  the   induction  of  iNOS  or  arginase-­‐I  (respectively)  reflecting  opposing  pathways  of  L-­‐arginine   metabolism,  however,  induction  of  these  genes  was  not  associated  with  equine  M1  or  M2   phenotypes.  Interestingly,  neither  iNOS  nor  arginase  expression  are  significantly  induced  in   polarized  human  macrophage  phenotypes.147  and  our  data  show  that  in  this  respect  equine   52   and  human  macrophages  are  alike.  The  absence  of  iNOS  expression  in  stimulated  equine   macrophages  is  in  keeping  with  recent  findings.197  In  contrast  to  rodents,  it  is  arginase-­‐II   isoform  that  is  predominantly  expressed  in  human  alveolar  macrophages,198   demonstrating  another  parallel  with  human  macrophage  phenotypes.  Unlike  arginase-­‐I,   arginase-­‐II  appears  to  be  associated  with  the  pro-­‐inflammatory  M1  phenotype.199  In  this   study,  arginase-­‐II  expression  was  not  induced  by  the  equine  M1,  but  rather  polarization   with  IL-­‐4  induced  a  slight  but  significant  suppression  of  arginase-­‐II,  consistent  with   reinforcement  of  general  anti-­‐inflammatory  profile  of  M2  macrophage  in  equine  lungs.     Another  important  difference  between  equine  M2  and  those  of  human  and  mouse  is  the   differential  induction  of  the  β-­‐glucan  receptor.  Induction  of  this  receptor  was  not  a  feature   of  equine  M2,  indicating  that  while  the  transcriptional  signature  of  polarized  equine   macrophages  appears  to  share  key  similarities  with  polarized  human  macrophages  some   differences  remain.     M2  macrophages  are  generally  considered  immune-­‐regulatory  by  virtue  of  low  production   of  pro-­‐inflammatory  cytokines  and  increased  expression  of  regulatory  cytokines  (such  as   TGFβ  and  IL-­‐10).  In  this  study  the  equine  M2  produced  low  levels  of  pro-­‐inflammatory   cytokines  but  also  suppressed  expression  of  IL-­‐10.  Despite  this,  overall  the  M2  macrophage   possessed  an  immune-­‐regulatory  function  as  pro-­‐inflammatory  cytokines  IL-­‐1β,  IL-­‐8,  IL-­‐ 12p40  and  the  co-­‐stimulatory  receptor  CD80  were  significantly  suppressed  when   stimulated  with  HD  or  constituent  elements.  This  suppressive  effect  is  in  agreement  with   the  findings  of  Jackson  et  al161  who  previously  demonstrated  immunosuppressive  effects  of   IL-­‐4  on  LPS  induced  IL-­‐8  and  IL-­‐1  β  in  equine  AM.  In  the  present  investigation  however,  the   suppressive  effect  was  not  universal,  as  IL-­‐6  expression  remained  unaffected.  IL-­‐6  has  both   53   pro  and  anti-­‐inflammatory  properties200  but  is  considered  an  M1  associated  gene.   However,  IL-­‐4  has  been  shown  to  induce  and  potentiate  IL-­‐6  expression.152  Thus  in  the   horse,  expression  of  IL-­‐6  alone  cannot  differentiate  between  M1  and  M2  polarized  states.   The  agonists  selected  for  this  study  were  based  on  their  presence  in  HD,  which  has   been  associated  with  inciting  pulmonary  inflammation:  namely  induction  of  the   hypersensitivity  disease  RAO  but  also  low-­‐grade  pulmonary  inflammation  in  otherwise   healthy  horses.64,194,201  Hay  dust  is  a  complex  mixture  containing  among  other  things,  fungi   and  gram  negative  and  positive  bacteria58  thus  we  investigated  zymosan,  LPS  and   peptidoglycan  to  respectively  represent  these  components.  Our  data  indicate  that  both   non-­‐polarized  cells  (controls)  and  M1  polarized  cells  respond  to  agonist  stimulation  with   similar  gene  expression  patterns  (including  a  modest  increase  in  expression  of  CD206)  that   differed  only  in  magnitude.  This  is  not  unexpected  as  stimulation  with  either  IFNγ  or   microbial  products  can  induce  M1  phenotypes  though  different  signaling  pathways   mediate  their  development.105     In  contrast,  regardless  of  the  stimuli,  the  M2  phenotype  has  a  potent  immune-­‐ regulatory  effect.  While  the  M2  phenotype  was  still  capable  of  responding  to  the   inflammatory  stimuli,  the  suppression  of  gene  expression  of  pro-­‐inflammatory  IL-­‐1β,  IL-­‐8,   IL-­‐12p40  and  CD80  was  considerable.  Further,  the  surface  receptor  CD206  remained   significantly  elevated  under  these  conditions.  The  marked  difference  in  the  response  of  M1   and  M2  macrophages  to  these  pathogen  associated  ligands  could  certainly  contribute  to   differential  outcomes  of  inflammatory  response  that  follow  natural  hay-­‐dust  challenges  in   normal  horses  and  those  susceptible  to  recurrent  airway  diseases.       54   In  the  context  of  identifying  M1  and  M2  phenotypes  as  potential  factors  in   respiratory  disease,  the  data  in  the  present  paper  provide  an  equine  specific,   transcriptional  expression  profile  that  can  guide  interpretation  of  alveolar  macrophages   harvested  from  pulmonary  diseases.  Expression  of  certain  murine  genes  (e.g  YM1  or  FIZZ)   are  considered  pathognomonic  for  IL-­‐4  activated  M2  phenotype  but  a  unique  equine  M2   marker  was  not  identified  in  this  study.  Thus,  M1/M2  categorization  is  dependent  on   assessing  the  pattern  of  the  gene  expression  profile  in  relation  to  the  relative  magnitude   and  the  relative  response  to  stimuli.  Our  data  indicate  that  evaluating  the  pro-­‐inflammatory   cytokines  TNFα,  IL-­‐1β,  IL-­‐12p40,  IL-­‐8,  regulatory  IL-­‐10  and  surface  receptors  CD80  and   CD206  can  be  used  to  differentiate  between  equine  M1  and  M2  macrophages.  These  data   further  illustrate  that  in  the  absence  of  M2  polarizing  conditions,  stimulation  with  HD  or  its   components  has  the  capacity  to  induce  an  M1  transcriptional  profile.  This  should  be   considered  in  clinical  studies  of  equine  pulmonary  disease  and  prior  exposure  to  HD  should   be  tightly  controlled  when  assessing  macrophage  phenotype.         In  conclusion,  equine  M1  phenotypes  are  characterized  by  a  general  expression  of   pro-­‐inflammatory  cytokines  profile  with  the  exception  of  induction  of  regulatory  IL-­‐10.   When  stimulated  M1  macrophages  predominately  enhanced  the  pro-­‐inflammatory   response.    In  contrast  the  equine  M2  is  characterized  by  increased  expression  of  CD206,  in   combination  with  low  expression  of  pro-­‐inflammatory  cytokines  (e.g  TNFα,  IL-­‐1β,  IL-­‐8,  IL-­‐ 12p40,  CD80)  and  low  expression  of  regulatory  IL-­‐10,  a  profile  that  is  maintained  following   subsequent  stimulation  with  HD  and  its  components.    The  data  presented  here  expands   current  knowledge  of  the  equine  alveolar  macrophage,  and  paves  the  road  for  future   studies  regarding  the  role  of  macrophage  polarization  in  equine  pulmonary  diseases.   55   Chapter  3.   RAO-­‐Susceptible  and  Control  Horses  Possess  Different  Alveolar  Macrophage   Phenotypes   Abstract   Background:  A  central  feature  of  recurrent  airway  obstruction  (RAO)  is  a  greatly  enhanced   sensitivity  to  hay  dust  compared  to  non-­‐RAO-­‐susceptible  control  horses.  The  cellular  and   molecular  mechanisms  accounting  for  this  differential  airway  sensitivity  are  unknown.  The   presence  of  divergent  alveolar  macrophage  (AM)  phenotypes  with  differential  sensitivity  to   microbial  hay  dust  components  could  play  a  contributing  role.  To  investigate  this  two   hypothesis  were  generated:   Hypothesis  2:  After  inhalation  of  HD,  alveolar  macrophages  from  RAO-­‐susceptible  horses   will  exhibit  an  M1  and  control  horses  will  exhibit  an  M2  phenotype.   Hypothesis  3:  After  inhalation  of  HD,  alveolar  macrophages  from  RAO-­‐susceptible  horses   will  exhibit  a  pro-­‐inflammatory  function  and  control  horses  will  exhibit  an  anti-­‐ inflammatory  function  when  stimulated  with  individual  components  of  HD  in  vitro.   Methods:  To  test  these  hypotheses  RT-­‐qPCR  was  used  to  evaluate  the  expression  of  equine   M1  and  M2  transcriptional  markers  (hypothesis  2)  and  the  AM  cytokine  response  to  HD   components  (lipopolysaccharide  (LPS),  peptidoglycan  (Pep)  and  zymosan  (Zym))   (hypothesis  3)  in  RAO-­‐susceptible  (n=6)  and  control  horses  (n=5)  at  baseline  and  following   exposure  to  hay  (natural  challenge).     Results:  At  baseline,  compared  to  control  horses,  AM  of  RAO-­‐susceptible  horses  had  greater   gene  expression  of  IL-­‐10  and  enhanced  expression  of  IL-­‐10  and  IL-­‐8  when  stimulated  with   LPS.  Further,  in  RAO-­‐susceptible  horses,  natural  challenge  increased  the  expression  of   56   CD206,  but  had  no  significant  effect  on  the  cytokine  response  to  hay  dust  components.  In   contrast,  natural  challenge  induced  increased  expression  of  IL-­‐1β,  IL-­‐8,  IL-­‐10  and  CD206  in   control  horses  and  altered  the  AM  cytokine  response  to  LPS,  enhancing  the  expression  of   IL-­‐1β,  IL-­‐10  and  CD206  when  compared  to  baseline.       Conclusions/clinical  importance:  RAO-­‐susceptible  and  control  horses  possess  divergent   AM  phenotypes.  Natural  challenge  induces  an  M1-­‐like  phenotype  in  control  horses  while   RAO-­‐susceptible  horses  maintain  an  IL-­‐10  producing  macrophage  phenotype.  These  results   suggest  that  unknown  host  factors  present  in  RAO-­‐susceptible  horse  lungs  promote  an  IL-­‐ 10-­‐producing  macrophage  phenotype.      Introduction   Organic  dust  contains  a  medley  of  microbial  components  that  can  stimulate  innate  immune   cells202    and  inhalation  of  organic  dust  (e.g  hay  dust)  induces  a  pulmonary  inflammatory   response  in  horses.  A  central  feature  of  recurrent  airway  obstruction  (RAO)  is  a  greatly   enhanced  sensitivity  to  hay  dust  compared  to  non-­‐RAO-­‐susceptible,  control  horses.62  RAO-­‐ susceptible  horses  develop  prominent  pulmonary  neutrophilia,  excess  mucus  cell   production,  bronchiolitis  and  bronchoconstriction,  and  display  pronounced  clinical  signs.   Control  horses  develop  low-­‐grade  pulmonary  neutrophilia  but  show  no  clinical  signs.  The   exact  mechanisms  through  which  hay  dust  elicits  this  unregulated  inflammatory  response   in  RAO  and  what  cell  types  contribute  to  this  differential  sensitivity  remain  uncertain.203     Alveolar  macrophages  can  be  activated  to  exhibit  distinct  phenotypes  that  have   polarized  functions  and  distinct  transcriptional  expression  profiles.  Classical  (M1)   activated  macrophages  (stimulated  by  IFNγ  and  LPS)  are  characterized  by  high  levels  of   57   pro-­‐inflammatory  cytokines  and  exhibit  strong  microbicidal  activity.  In  contrast,   alternative  (M2)  activated  macrophages  (stimulated  by  IL-­‐4)  are  considered  immuno-­‐ regulatory  as  a  result  of  their  reduced  production  of  pro-­‐inflammatory  cytokines  and   increased  production  of  regulatory  cytokines.106  Importantly,  the  activation  status   influences  the  magnitude  of  response  to  subsequent  stimulation  with  pathogen-­‐derived   stimulants  and  M1  activated  macrophages  possess  greater  magnitude  of  pro-­‐inflammatory   response  compared  to  M2  activated  macrophages.  Previously  we  characterized  the   polarization  of  equine  AM,  M1  and  M2  phenotypes,  identifying  the  distinct  transcriptional   expression  profiles  and  response  to  agonist  stimulation  (Chapter  2).   Macrophage  M1  and  M2  phenotypes  have  been  associated  with  allergic  airway   disease.204,205  However,  there  are  no  data  systematically  evaluating  alveolar  macrophage   phenotype  markers  in  RAO-­‐susceptible  horses.  Thus  to  gain  a  greater  understanding  of  the   role  of  the  AM  in  disease  and  in  response  to  hay  dust  we  compared  the  gene  expression   profile  of  M1  and  M2  associated  genes  in  RAO-­‐susceptible  and  control  horses  at  pasture   and  following  chronic  exposure  to  hay/straw.  As  RAO-­‐susceptible  horses  have  a   hypersensitive  response  to  organic  dust  compared  to  control  horses,  we  hypothesized  that   after  inhalation  of  HD,  alveolar  macrophages  from  RAO-­‐susceptible  horses  will  exhibit  an   M1  and  control  horses  will  exhibit  an  M2  phenotype  (hypothesis  2)  and  further,  after   inhalation  of  HD,  alveolar  macrophages  from  RAO-­‐susceptible  horses  will  exhibit  a  pro-­‐ inflammatory  function  and  control  horses  will  exhibit  an  anti-­‐inflammatory  function  when   stimulated  with  individual  components  of  HD  in  vitro  (hypothesis  3).   58   Materials  and  Methods     Animals   The  investigation,  which  used  six  RAO-­‐susceptible  and  five  control  horses,  was  approved   by  Michigan  State  University's  Institutional  Animal  Care  and  Use  Committee.  Horses  were   classified  as  RAO-­‐susceptible  if  exposure  to  dusty  hay  and  straw  (natural  challenge)   induced  airway  obstruction  [determined  by  measurement  of  maximal  change  in  pleural   pressure  during  tidal  breathing  (ΔPplmax)]  that  was  significantly  ameliorated  by  anti-­‐ cholinergic  treatment  and  was  reversed  when  horses  were  removed  from  exposure1.   Horses  that  did  not  develop  airway  obstruction  under  natural  challenge  qualified  as   controls.  Except  during  natural  challenge,  all  horses  were  maintained  on  pasture  and   supplemented  with  complete  pelleted  feed.  RAO-­‐susceptible  horses  were  in  remission   (ΔPplmax  <10cmH20)  at  the  start  of  the  protocol.   Study  Design     Horses  were  transported  from  the  clean  air  environment  (pasture)  to  the  stable  and  were   bedded  on  straw  and  fed  hay  (natural  challenge).  In  RAO-­‐susceptible  horses,  natural   challenge  continued  until  clinical  signs  of  RAO  (labored  breathing)  were  induced  and   ΔPplmax  was  greater  than  or  equal  to  15  cmH2O.  As  the  time  to  induce  this  ΔPplmax  varied   between  RAO-­‐susceptible  horses,  the  control  horse  was  paired  with  an  RAO-­‐susceptible   horse  and  each  received  the  same  duration  of  hay/straw  exposure.  For  logistical  reasons,   the  two  horses  in  each  pair  could  not  be  sampled  on  the  same  day,  so  the  two  were   consecutively  exposed  with  overlapping  periods  in  the  same  air  space  when  possible.   Measurements  of  lung  function  were  made  and  bronchoalveolar  lavage  fluid  (BALF)  was   59   harvested  at  two  time  points;  baseline  (BL:  immediately  after  being  brought  to  the   laboratory  from  pasture)  and  at  the  end  of  natural  challenge  (Figure  4).     Figure  4.  Natural  Challenge  Study  Design   Pulmonary  Function  Tests   Pulmonary  function  was  assessed  by  measurement  of  ∆Pplmax,  pulmonary  resistance  (RL)   and  dynamic  compliance  (Cdyn)  as  previously  described206.       Isolation  of  Alveolar  Macrophages   Horses  were  sedated  with  detomidine  hydrochloride  (10ug/kg  IV)  (Zoetis)  and   butorphanol  tartrate  (0.02mg/kg,  IV)  (Zoetis)  and  bronchoalveolar  lavage  (BAL)  was   60   performed  as  previously  described  (Chapter  2).  The  BAL  fluid  (BALF)  was  immediately   placed  on  ice  and  processed  within  30  minutes.     The  BALF  was  passed  through  an  80μm  sterile  filter  and  centrifuged  (250xg,  4°C,  10  min).   Cell  pellets  were  re-­‐suspended  in  sterile  medium  (RPMI+  L-­‐glutamine  supplemented  with   antibiotic/antimycotic,  5%  heat  inactivated  equine  serum,  2mM  EDTA)  and  washed  twice.   Cell  number  and  viability  were  assessed  using  a  hemocytometer  and  trypan  blue  exclusion   respectively.   Alveolar  macrophages  were  isolated  from  the  mixed  cell  population  by  magnetically-­‐ activated  cell  sorting  (MACS)  using  negative  selection.  Briefly,  BALF  cells  were   resuspended  in  medium  (2x10^7  cells/ml)  and  incubated  with  antibodies  against   lymphocytes  and  neutrophils  (mouse  monoclonal  IgG,  HB88,  mouse  monoclonal  IgM  DH24   respectively  {Monoclonal  Antibody  Center,  Washington  State  University})  at  4°C  for  30min.   After  washing,  BALF  cells  were  incubated  with  secondary  antibodies  conjugated  to  metal   beads  (anti-­‐mouse  IgG  polyclonal  and  anti-­‐mouse  IgM)  at  4°C  for  20  min  and  washed   before  passing  through  magnetic  columns  to  remove  labeled  cells.  The  eluted  cells   (enriched  macrophage  population)  were  collected  and  re-­‐suspended  (6.25x10^5  cells/ml)   in  sterile  medium  (minus  EDTA).  An  aliquot  of  cells  was  collected  for  RNA  analysis  -­‐“freshly   harvested  alveolar  macrophages,”(Figure  4)  (n=6  RAO-­‐susceptible  horses,  n=5  control   horses).  A  small  aliquot  was  collected  for  cell  cytology  (cytospin  preparation)  and  to  assess   viability  by  use  of  light  microscopy  (400  cell  count  of  Diff  Quik  stained  slides)  and  trypan   blue  exclusion,  respectively.     61   Cell  Culture   The  remaining  cells  were  plated  into  sterile  tissue  culture  dishes  (12-­‐well  at  5.x10^5   cells/well)  and  incubated  (5%  CO2,  37°C)  for  2  hours  to  allow  adherence.  After  a  medium   change,  adherent  cells  were  incubated  for  16  hours  with  medium  alone  (control)  or,   medium  plus  lipopolysaccharide  (LPS,  10ng/ml),  peptidoglycan  (Pep,  1ug/ml)  or  zymosan   (Zym,  10ug/ml).    Cells  were  then  harvested  -­‐“plated  AM”  (Figure  4)  (n=5  RAO-­‐susceptible,   n=5  control  horses)  and  RNA  was  extracted  and  stored  (-­‐80°C)  until  gene  expression  was   evaluated.     RNA  Extraction,  Reverse  Transcription  and  Quantitative  Real  Time  PCR   Cells  were  harvested  by  adding  RLT-­‐Lysis  Buffer  (Qiagen)  and  homogenized  using  Qiagen   QIAshredderTM  spin  columns.  Total  RNA  was  extracted  and  purified  using  Qiagen  RNeasy®   Mini  Kit  (freshly  harvested  AM)  or  Qiagen  RNeasy®  Plus  Micro  Kit  (plated  AM)  which   include  steps  for  genomic  DNA  removal.  RNA  concentration  was  measured  using  Qubit®  2.0   fluorometer  and  integrity  of  RNA  (RIN)  was  assessed  using  the  Agilent  2100  Bioanalyzer®   (Agilent  Technologies)  and  a  RIN  score  of  >6.5  was  considered  acceptable  for  qPCR.195   Equal  concentrations  of  RNA  were  reverse  transcribed  (High  Capacity  cDNA  Reverse   Transcription  Kit®,  Applied  Biosystems)  to  create  cDNA.    Six  µl  of  cDNA  was  then  pre-­‐ amplified  (TaqMan®  PreAmp  Master  Mix  Kit,  Life  Technologies)  according  to   manufacturer's  instructions)  and  amplified  cDNA  was  stored  at  -­‐80°C  until  further  analysis.   Amplification  uniformity  was  assured  for  all  gene  assays  tested.   Quantitative  real  time  PCR  was  performed  using  predesigned  TaqMan®  Gene  Expression   Assays  (Life  Technologies)  and  TaqMan®  Gene  Expression  Master  Mix  (Life  Technologies).   62   When  predesigned  assays  were  unavailable,  primers  and  probes  were  designed  using  the   Custom  TaqMan®  Assay  Design  Tool  or  Custom  TaqMan®  (Plus)  Assay  Design  Tool  (Life   Technologies).    For  each  experiment,  all  samples  were  run  at  once,  in  triplicate  on  384  well   optical  plates  in  the  ABI  7900HT  Real-­‐Time  PCR  System®  (Applied  Biosystems).  The   average  of  two  stable  endogenous  genes,  hypoxanthine  ribosyltransferase  (HPRT)  and   elongation  factor  1α  (ELF1)  were  used  to  calculate  the  deltaCT  of  each  sample.     Data  Analysis   Normality  of  errors  of  each  variable  was  assessed  using  visual  inspection  of  error   histogram,  probability  plots,  and  normality  testing  using  SAS-­‐Proc  Univariate  procedure.   Normally  distributed  data  were  analyzed  using  an  ANOVA  with  the  fixed  effect  of  group  and   time  and  the  random  effect  of  horse  (cytology,  pulmonary  function,  freshly  harvested  AM   gene  expression).  Variables  from  “stimulated  alveolar  macrophages”  were  analyzed  with   the  fixed  effect  of  group,  time  and  treatment  and  the  random  effect  of  horse  (SAS  Proc   Mixed).  Errors  that  were  not  normally  distributed  were  log  transformed  (IL-­‐1β-­‐plated   alveolar  macrophages)  and  normality  of  transformed  data  was  assessed  as  described.  Data   from  “plated  alveolar  macrophages”  was  corrected  for  multiple  treatment  comparisons   using  Bonnferroni  correction.  Results  were  considered  significant  if  ΔCT  between   comparisons  was  ≥1  (2  fold  difference)  and  p≤0.05.  All  statistical  analyses  were  run  on  SAS   9.4®    (SAS  Institute  Inc.,  SAS  Campus  Drive,  Cary,  NC).     63   Results   BALF  Cytology  and  Evaluation  of  Pulmonary  Function     The  median  duration  of  natural  challenge  was  9  days  (range  7-­‐21).  At  baseline,  there  were   no  significant  differences  in  macrophage  or  neutrophil  percentage  between  the  RAO-­‐ susceptible  and  control  horses  (Table  4)  but  lymphocyte  percentage  was  significantly   greater  in  RAO-­‐susceptible  horses  compared  to  control  horses.  Natural  challenge  induced  a   significant  increase  in  airway  neutrophil  percentage  in  both  control  and  RAO-­‐susceptible   horses  although  the  magnitude  of  airway  neutrophils  was  significantly  greater  in  RAO-­‐ susceptible.  The  increase  in  airway  neutrophils  was  accompanied  by  a  significant  decrease   in  lymphocyte  percentage  in  RAO-­‐susceptible  population  only.  In  keeping  with  RAO   phenotype,  RAO-­‐susceptible  horses  but  not  controls  develop  impairments  in  pulmonary   function  following  natural  challenge  (significant  elevations  in  ∆Ppl,  and  RL  and  significantly   reduced  Cdyn).     64   Table  4.  Effect  of  Natural  Challenge  on  BALF  Cytology  and  Pulmonary  Function   ! Baseline! Natural!Challenge! Control! RAO4susceptible! Control! RAO4susceptible! Macrophage!(%)! 69.5±9.2! 54.6±12.2! 54.3±18.94! 42.5±13.13! Lymphocyte!(%)! 29.6±9! 41.3±11.84†! 28.9±8.8! 24.67±4.3*! Neutrophil!(%)! 0.7±0.5! 3.23±2.6! 16.1±11.9*! 31.83±12.59*†! Eosinophils!(%)! 0.18±0.36! 0! 0.17±0.3! 0.067±0.1! DpPl!max!(mmHg)! 3.9±0.67! 5.19±0.85! 4.32±0.46! 29.42±13.5*†! Resistance! (cmH2O/l/s)! 0.48±0.14! 0.58±0.24! 0.78±0.34! 2.5±0.89*†! 1.27±0.46! 1.41±0.36! 0.9±0.29! 0.27±0.19*†! Dynamic!Compliance! (cmH2O/l)! ! Data  presented  as  (mean±sd),  *p≤0.05  compared  to  pasture,  †  p≤0.05  compared  to  control   horses.     65    Freshly  Harvested  Alveolar  Macrophages     Macrophage  purity  was  not  significantly  different  between  groups  or  time  points   (85%±8.9,  mean±sd)  and  contaminating  cells  were  lymphocytes.  We  first  compared  gene   expression  between  RAO-­‐susceptible  and  control  horses  at  baseline  and  following  natural   challenge  (Table  5).  There  were  no  significant  differences  in  gene  expression  of  M1  or  M2   regulated  genes  or  pathogen  receptors  at  either  time  point.  However,  at  baseline,  RAO-­‐ susceptible  horses  expressed  greater  quantities  of  the  immune-­‐regulatory  IL-­‐10  mRNA   transcript  compared  to  control  horses  but  this  difference  was  not  present  following  natural   challenge.     We  next  evaluated  the  effect  of  natural  challenge  within  each  group  (Table  6).  Within  the   control  group,  natural  challenge  had  no  significant  effect  on  M1  or  M2  regulated  genes  or   pathogen  receptors  but  IL-­‐10  expression  was  significantly  increased.  Within  the  RAO-­‐   susceptible  horses  there  was  a  modest  but  significant  reduction  in  IL-­‐12p40  expression.   There  was  no  effect  on  any  other  genes  tested.   66   Table  5.  Gene  Expression  in  Freshly  Harvested  Alveolar  Macrophages  in  RAO-­‐Susceptible   Horses  Relative  to  Control  Horses  at  Pasture  and  Following  Natural  Challenge     Gene$Name$$ Baseline$$$$$$$$$$P$ value$ Genes!regulated!in!Equine!M1!! ! Natural$ Challenge$ P$value! ! ! ! TNFα! 1.26!!±!0.94! NS! 1.06!!±!0.37! NS! ILB1β! 1.334%%±%0.40! NS! 1.79%±%2.16! NS! ILB12!p40! 1.38%±%0.83! NS! 0.35%±%0.18! NS(p<0.08)! CD80! 1.03%%±%0.33! NS! 0.98%%±%0.37! NS! ILB8! 1.07%±%0.83! NS! 0.84%%±%0.47! NS! ILB6! 1.70%±%1.75! NS! 1.40%±%1.69! NS! % % ! Genes!regulated!in!Equine!M2!% % ARG!2! 1.24%%±%0.61! NS! 0.82%%±%0.14! NS! CD206! 1%%±%0.34! NS! 1.1%%±%0.51! NS! ! ! ! ! 0.76%%±%0.17! NS! 0.89%%±%0.16! NS! 0.87%±%0.35! NS! 0.91%±%0.11! NS! NS! 0.86%0.45! NS! ! ! ! 0.92%%±%0.22! NS! Pathogen!Recognition! Receptors! ! βBGlucan!Receptor! TLR4! TLR2! Regulatory!Cytokines! 1.27%%±%0.35! ! TGFβ! 1.13%%±%0.38! NS! IL610! 4.35%±%2.32! 0.006! 0.8%±%0.7%! NS! ! Data  are  presented  as  relative  fold  difference  (mean  ±sd)  with  1  being  equivalent   expression  >1  greater  expression  and  <1  lower  expression.  NS  not  significant.   67   Table  6.  The  Effect  of  Natural  Challenge  on  Gene  Expression  of  Freshly  Harvested  Alveolar   Macrophages  in  RAO-­‐Susceptible  and  Control  Horses     Gene$Name$$ Control$$$$$$$$$$$ Genes!regulated!in!Equine!M1!! ! P$ RAO3susceptible$ P$ value$ value! ! ! ! TNFα! 0.98!±0.25! NS! 1.27!±0.74! NS! ILB1β! 2.89!±3.04! NS! 2.21!±1.2! NS! ILB12!p40! 1.79!±1.24! NS! 0.59!±0.47! 0.027! CD80! 0.70!±0.16! NS! 0.70!±0.65! NS! ILB8! 1!±0.2! NS! 1!±0.47! NS! ILB6! 1.71!±0.9! NS! 1.47!±1.1! NS! ! ! ! Genes!regulated!in!Equine!M2!! ! ARG!2! 1.42!±0.39! NS! 0.97!±0.28! NS! CD206! 1.02!!±0.16! NS! 1.15!±0.34! NS! ! ! ! ! 0.81!±0.14! NS! 0.99!±0.07! NS! TLR4! 0.74!±0.24! NS! 0.99!±0.39! NS! TLR2! 1!±0.09! NS! 0.98!±0.35! NS! ! ! ! NS! 0.89!±0.4! NS! Pathogen!Recognition! Receptors! ! βBGlucan!Receptor! Regulatory!Cytokines! ! TGFβ! 1.04!±0.29! IL$10! 15.59!±13.24! 0.006! 1.76!±0.77!! NS! ! Data  presented  as  fold  change  (mean  ±sd)  at  post  natural  challenge  relative  to  baseline   with  1  being  equivalent  expression  >1  greater  expression  and  <1  lower  expression.  NS  not   significant.   68   Phenotype  of  Plated  Alveolar  Macrophages   We  next  evaluated  the  response  of  the  alveolar  macrophages  to  incubation  with  pathogen-­‐ derived  stimuli  (LPS,  Pep  and  Zym)  for  16  hours.  At  baseline,  in  keeping  with  freshly   harvested  cells,  the  unstimulated  (media  control)  alveolar  macrophages  of  RAO-­‐susceptible   horses  had  significantly  greater  expression  of  IL-­‐10  than  those  from  control  horses  (Figure   5A).  Further,  natural  challenge  was  associated  with  significantly  increased  expression  of  IL-­‐ 1β,  IL-­‐8,  IL-­‐10  in  the  control  horses  only.  Both  RAO-­‐susceptible  and  control  horses   developed  increased  gene  expression  of  CD206  following  natural  challenge  (Figure  5B).       In  comparing  the  response  of  stimulated  alveolar  macrophages  from  RAO-­‐susceptible  and   control  horses  at  baseline  (Figure  6),  there  were  no  significant  differences  in  expression   TNFα,  IL-­‐1β,  CD80  or  CD206  following  stimulation  with  LPS,  Pep  or  Zym  (Fig.  6A-­‐C).   However,  compared  to  control  horses,  RAO  horses  demonstrated  significantly  greater   expression  of  both  IL-­‐10  and  IL-­‐8  when  stimulated  with  LPS  (Fig  6A).  Following  natural   challenge,  there  were  no  significant  differences  between  RAO-­‐susceptible  and  control   horses  in  gene  expression  of  TNF,  IL-­‐1,  IL-­‐8,  IL-­‐10,  CD80  and  CD206  by  LPS,  Pep  and  Zym   stimulated  alveolar  macrophages.     Within  the  control  group  there  was  a  significant  effect  of  natural  challenge  on  the  response   to  LPS  stimulation:  following  natural  challenge,  LPS  stimulated  alveolar  macrophages   expressed  significantly  more  IL-­‐1β  and  IL-­‐10  than  at  baseline  (Fig.  6A).  However,  gene   expression  of  IL-­‐1β  and  IL-­‐10  was  not  significantly  different  from  the  unstimulated  cells  at   the  natural  challenge  time  point,  although  IL-­‐1β  did  approach  significance  (Fig.7).  By   contrast,  there  was  no  effect  of  natural  challenge  on  TNFα,  IL-­‐1β,  or  IL-­‐10  in  RAO-­‐ susceptible  horses  and  gene  expression  was  equivalent  for  all  stimulants  at  baseline  and   69   after  natural  challenge.    However,  natural  challenge  induced  elevations  in  CD206  of  LPS   stimulated  alveolar  macrophages  in  both  control  horses  and  RAO-­‐susceptible  horses  and  in   zymosan  stimulated  alveolar  macrophages  in  RAO-­‐susceptible  horses.  Further,  there  were   also  no  changes  (between  groups  or  over  natural  challenge)  in  pro-­‐inflammatory  IL-­‐6,   regulatory  TGFβ  (data  not  shown).         70   Figure  5.  Alveolar  Macrophages  From  RAO-­‐Susceptible  and  Control  Horses  Differ  in  Gene   Expression  of  Pro-­‐Inflammatory,  Regulatory  and  Surface  Receptors  at  Baseline  and   Following  Natural  Challenge     A.  Pro-­‐inflammatory  and  regulatory  cytokines   12 TNFα# IL+1β# IL+8# IL+10# 10 †## *# Delta CT 8 6 *# 4 2 *# 0 -2 -4 [ [ [ [ [ [ [ [ BL# NC# BL# NC# BL# NC# BL# NC# B.  Surface  Receptors   CD206" CD80" 5 Delta CT 4 3 2 *" 1 [ [ [ [ 0 *" BL" BL" NC" NC" Gene  expression  (deltaCT,  mean  ±  sem)  of  unstimulated  (media  control)  alveolar   macrophages  from  control  (dark  bars,  n=5)  and  RAO-­‐susceptible  (light  bars,  n=5)  horses  at   baseline  (BL)  and  following  natural  challenge  (NC).  (A)  Pro-­‐inflammatory  and  regulatory   cytokines  (B)  Surface  receptors.  Note,  a  lower  deltaCT  indicates  greater  mRNA  expression.    *   p≤0.05  compared  to  baseline  †  p≤0.05  compared  to  control  group.   71   Figure  6.  Alveolar  Macrophages  From  RAO-­‐Susceptible  Horses  Exhibit  Enhanced  LPS   Responsiveness  at  Baseline  but  Not  After  Natural  Challenge     * * 2 1 _________ Baseline 0 8 IL-8 expression IL-1 expression 3 50 * † 6 _________ Nat. Challenge * 2 _________ Baseline _________ Nat. Challenge CD80 expression 4 3 * * 2 1 0 _________ Baseline 30 * * _________ Nat. Challenge 72   ‡ * 20 10 _________ Baseline _________ Nat. Challenge ‡ 50 4 0 40 0 IL-10 expression TNF expression 4 40 30 20 10 0 CD206 expression A. $ * *† _________ Baseline * _________ Nat. Challenge 10 ‡ 8 ‡ 6 4 2 0 _________ Baseline _________ Nat. Challenge Figure  6  (cont'd).   * 5 _________ Baseline IL-8 expression 15 10 5 _________ Baseline 6 4 2 0 _________ Baseline _________ Baseline _________ Nat. Challenge 73   _________ Nat. Challenge * 150 100 50 0 _________ Nat. Challenge * 8 100 200 20 0 200 0 _________ Nat. Challenge * 25 CD80 expression IL-1 expression 10 0 * 300 IL-10 expression TNF expression 15 CD206 expression B.$ _________ Baseline _________ Nat. Challenge * 20 15 ‡ 10 5 0 _________ Baseline _________ Nat. Challenge Figure  6  (cont'd).   * 10 _________ Baseline IL-8 expression 20 10 _________ Baseline 20 _________ Nat. Challenge * 15 10 5 _________ Baseline 400 200 _________ Baseline _________ Nat. Challenge _________ Nat. Challenge * 2000 30 0 600 0 _________ Nat. Challenge * 40 CD80 expression IL-1 expression 20 0 * 800 IL-10 expression TNF expression 30 1500 1000 500 0 CD206 expression C.$ _________ Baseline _________ Nat. Challenge * 15 ‡ 10 5 0 _________ Baseline _________ Nat. Challenge Graphs  represent  the  gene  expression  response  of  AM  from  control  horses  (dark)  and  RAO-­‐ susceptible  (light)  horses  when  stimulated  with  LPS  (A),  peptidoglycan  (B),  or  zymosan  (C)   at  baseline  and  following  natural  challenge  (Nat.  Challenge).  Data  represents  gene   expression  relative  to  media  treated  cells  at  baseline  (fold  change,  mean  ±  sem).  *  p≤0.05   compared  to  media  treated  cells  within  same  time  point,  ‡  p≤0.05  compared  to  same   stimulation  conditions  at  baseline,  †  p≤0.05  compared  to  non-­‐RAO  susceptible  group.     74   Figure  7.  Natural  Challenge  Differentially  Alters  the  Phenotype  of  Alveolar  Macrophages   From  RAO-­‐Susceptible  and  Control  Horses     IL#1β% TNFα% 25 Relative Fold Change Relative Fold Change 2.5 2.0 1.5 1.0 0.5 *% 2 1 0 5 *% 40 30 20 *% 10 0 CD80% CD206% 6 Relative Fold Change 2.0 Relative Fold Change *% 10 50 3 1.5 1.0 0.5 0.0 15 IL#10% Relative Fold Change Relative Fold Change 4 *% 0 0.0 IL#8% 20 Con% LPS% 5 4 3 2 *% 1 0 Pep% Zym% *%*% *%*% Con% LPS% Pep% Zym% CD80 DCT(3.4.) 150 100 Graphs  represent  AM  gene  expression  in  control  horses  (black  bars)  and  RAO-­‐susceptible   50 (light  bars)  following  incubation  with  media  control  (Con),  lipopolysaccharide  (LPS),   2 1 0 peptidoglycan  (Pep),  or  zymosan  (Zym).  Data  from  figure  2  are  presented  as  the  fold   change  (mean  ±  sem)  following  natural  challenge  relative  to  its  expression  at  baseline.   *p<0.05  significant  effect  of  natural  challenge. 75   Discussion   Our  findings  indicate  that  the  phenotype  of  alveolar  macrophages  from  RAO-­‐ susceptible  horses  at  baseline  (in  remission)  differs  from  that  of  control  horses  in  that  the   former  have  greater  gene  expression  of  IL-­‐10,  and  display  an  enhanced  responsiveness  to   LPS  stimulation  (increased  expression  of  IL-­‐8  and  IL-­‐10).   The  term  "alternatively  activated  macrophages"  (M2  phenotype)  refers  to  a   spectrum  of  phenotypes  that  are  broadly  associated  with  immune  regulation,  resolution  of   inflammation  and  wound  repair.  The  term  “alternatively  activated”  refers  to  the  fact  that  it   differs  from  the  classical,  microbial/IFNγ-­‐activated  cell.207  In  addition  to  IL-­‐4,  other   mediators  can  induce  different  M2  sub-­‐phenotypes,  including  regulatory  macrophages  (M-­‐ reg).  The  M-­‐reg  phenotype  predominately  expresses  IL-­‐10  and  can  be  induced  in  response   to  stimulation  with  immune  complexes208,  certain  TLR  ligands209  corticosteroids,136   adenosine137,138  and  phagocytosis  of  apoptotic  cells.135  Additionally,  activated  M1   macrophages  can  transition  to  become  M-­‐regs  as  a  means  of  controlling  and  resolving   inflammation.138  The  IL-­‐10  producing  alveolar  macrophage  phenotype  detected  in  RAO-­‐ susceptible  horses  is  suggestive  of  an  M-­‐reg  phenotype.  Typically,  M-­‐reg  also  display   significantly  suppressed  IL-­‐12  expression.  In  the  present  study,  baseline  expression  of  IL-­‐ 12p40  was  similar  in  RAO-­‐susceptible  horses  and  control  horses.  However,  in  support  of   the  M-­‐reg  phenotype,  natural  challenge  induced  a  modest  but  significant  decrease  (mean  -­‐ 1.7  fold  difference)  in  RAO-­‐susceptible  horses  (Table  6)  and  there  was  a  trend  for  reduced   IL-­‐12p40  expression  in  RAO-­‐susceptible  horses  relative  to  controls  (mean  -­‐2.8  fold   difference  (p=0.08)(Table  5).     76   Interleukin-­‐10  is  a  pleotropic  cytokine  that  can  induce  tolerance  in  T-­‐cells,  inhibit  T-­‐ cell  proliferation  and  Th1/Th2  cytokines  and  decrease  the  pro-­‐inflammatory  response  of   monocytes/macrophages,  thus  it  is  classically  considered  immunosuppressive  and  anti-­‐ inflammatory.128,129  In  RAO,  it  is  possible  that  alveolar  macrophage  derived  IL-­‐10   contributes  to  the  regulation  of  the  subclinical  pulmonary  inflammation  that  remains   during  remission.10,11,67,210–212  Furthermore,  it  may  contribute  to  the  down-­‐regulation  of   Th1/Th2  cytokines  in  BAL  lymphocytes  that  has  been  documented  during  RAO-­‐ remission.43      Increased  numbers  of  IL-­‐10  producing  macrophages  have  been  detected  in  atopic   asthmatics  213–216  however,  there  are  opposing  accounts  on  the  exact  role  that  IL-­‐10  plays   in  the  context  of  chronic  pulmonary  inflammatory  disease.  Interleukin-­‐10  has  been   associated  with  reduced  allergic  airway  inflammation  and  airway   hyperresponsiveness.217,218  Yet,  IL-­‐10  can  also  potentiate  pulmonary  pathology  by   promoting  Th2  effector  cytokines,213,219  airway  hypersensitivity,220  mucus  metaplasia,  and   airway  remodeling.221  There  are  opposing  data  on  the  contribution  of  Th2  cytokines  to  RAO   pathogenesis38,39,42,43,46  thus  it  is  unclear  if  alveolar  macrophage  derived  IL-­‐10  could   promote  disease  pathogenesis  by  augmenting  the  Th2  pathway.     Generally,  exogenous  IL-­‐10  significantly  reduces  the  pro-­‐inflammatory  response  of   LPS  stimulated  macrophages.131,222  Thus  it  was  somewhat  surprising  that  the  alveolar   macrophage  from  RAO-­‐susceptible  horses  maintained  responsiveness  to  agonist   stimulation  and  in  fact,  expressed  elevated  neutrophilic  chemokine  IL-­‐8  when  stimulated   with  LPS.  However,  stimulation  of  TLR-­‐4  or  TLR-­‐2  can  modify  IL-­‐10-­‐receptor  function  in   alveolar  macrophages,  which  could  contribute  to  the  lack  of  autocrine  immunosuppression   77   observed.101  Further,  while  M2  macrophages  are  typically  considered  to  exhibit  a   diminished  pro-­‐inflammatory  response  to  microbial  stimuli,  M2  macrophages  induced  in   murine  allergic  asthma223  or  in  presence  of  in-­‐vitro  IL-­‐33174,224  (associated  with  Th2   polarized  pathology225)  possess  enhanced  expression  of  pro-­‐inflammatory  genes  when   stimulated  with  LPS.  In  the  present  study,  enhanced  response  to  LPS  was  not  mediated  by   differences  in  expression  of  TLR-­‐4  (Table  5)  and  the  molecular  mechanisms  that  mediate   this  sensitivity  requires  further  investigation.     Furthermore,  stimulation  of  isolated  alveolar  macrophages  with  LPS  generally   induces  expression  of  pro-­‐inflammatory  TNFα,  IL-­‐1β,  IL-­‐8,  and  anti-­‐inflammatory  IL-­‐10,   thus  it  is  interesting  that  only  IL-­‐8  and  IL-­‐10  were  enhanced  by  LPS  stimulation  at  baseline   (Figure  6).  This  could  suggest  that  there  is  differential  regulation  of  IL-­‐8  and  IL-­‐10  within   the  RAO-­‐susceptible  alveolar  macrophage.  One  possible  explanation  could  be  exposure  to   adenosine  which  is  elevated  in  horses  with  lower  airway  inflammation.226  Adenosine  can   induce  regulatory  macrophages  and  activation  of  adenosine  receptors  can  modulate  the   inflammatory  response  of  equine  monocytes  to  LPS,  differentially  enhancing  production  of   IL-­‐10  and  IL-­‐8.227     In  vivo,  RAO-­‐susceptible  horses  are  more  sensitive  to  LPS  than  control  horses,   developing  enhanced  neutrophilic  inflammation  at  lower  LPS  doses.61  It  is  possible  that   enhanced  IL-­‐8  production  by  alveolar  macrophage  could  contribute  to  the  early  neutrophil   recruitment.  In  this  respect,  our  data  are  in  keeping  with  Laan  et  al87  who  also  reported   that  AMs  from  RAO-­‐susceptible  horses  display  increased  sensitivity  to  nebulized  LPS  and   hay  dust  suspension.  On  the  other  hand,  in  the  present  study,  IL-­‐10  is  similarly  enhanced   78   by  LPS  stimulation,  (Figure  6A)  and  as  IL-­‐10  can  reduce  neutrophil  recruitment  to  the   lung228,229  it  is  possible  that  elevated  IL-­‐10  could  counteract  the  effects  of  IL-­‐8.   In  contrast  to  our  study,  Laan  et  al87  failed  to  detect  increased  expression  of  IL-­‐10   from  alveolar  macrophages  from  RAO-­‐susceptible  horses.  It  is  possible  that  differences  in   macrophage  isolation  technique  and  culture  conditions  could  account  for  disparity   amongst  investigators.  Polarized  M2  phenotypes  may  be  less  adherent,161,222,230,231  and   isolation  of  alveolar  macrophages  by  adhesion  (as  used  by  Laan  et  al.)  could  select  against   this  phenotype.  Interestingly,  in  the  present  study  we  did  not  detect  an  enhanced  response   to  agonists  peptidoglycan  and  zymosan.  It  is  possible  that  alveolar  macrophages   specifically  develop  sensitivity  to  LPS  but  not  to  other  microbial  stimuli,  but  as  the   concentrations  of  peptidoglycan  and  zymosan  induced  a  much  greater  inflammatory   response  compared  to  LPS  stimulation,  it  is  certainly  possible  that  this  masked  any  subtle   differences  in  macrophage  sensitivity.     We  also  compared  the  phenotypic  response  of  both  groups  to  hay  and  interestingly,   alveolar  macrophages  from  each  group  responded  differently  to  natural  challenge.  Control   horses  developed  an  M1-­‐like  phenotype  (i.e.  mixed  expression  of  pro-­‐inflammatory  IL-­‐1β,   IL-­‐8  and  immune-­‐regulatory  IL-­‐10,  and  enhanced  response  to  LPS)  in  conjunction  with   increased  CD206  expression  (M2  marker).  In  contrast,  the  phenotype  in  RAO-­‐susceptible   horses  remained  relatively  stable  only  showing  increased  CD206  expression.  CD206  is  the   canonical  M2  associated  receptor  induced  by  IL-­‐4.112  As  other  characteristics  of  IL-­‐4   activated  equine  M2  macrophages  were  not  observed  in  either  group  (Chapter  2)  it  is   unlikely  that  a  pulmonary  environment  dominated  by  IL-­‐4  induced  the  CD206  expression   observed.  CD206  is  a  multifunctional  phagocytic  receptor  and  may  have  a  number  of   79   functions  in  the  context  of  chronic  pulmonary  inflammatory  disease.  Recognition  of   bacteria  and  fungi  can  induce  production  of  pro-­‐  or  anti-­‐inflammatory  cytokines  depending   on  the  stimulant.115  The  CD206  receptor  can  also  phagocytose  a  variety  of  allergens  (e.g   hay  dust  mite,  dog  and  cat  allergens)  and  promote  Th2  polarization.232  Furthermore,   CD206  may  contribute  to  the  resolution  of  inflammation  through  its  capacity  to   phagocytose  myeloperoxidase.118         The  present  data  demonstrates  that  organic  dust  is  a  noxious  stimulant  that  elicits  an  M1-­‐ like  phenotype  in  control  horses  and  suggests  that  the  alveolar  macrophage  participates  in   the  normal  inflammatory  response  to  hay  dust.64,194,201,233  In  partial  agreement,  Joubert  et   al88  reported  that  24  hours  of  natural  challenge  induced  increased  expression  of  neutrophil   chemokines  (IL-­‐8  and  MIP2)  in  equine  alveolar  macrophage  from  control  horses,  but  in   contrast  to  our  findings  RAO-­‐susceptible  horses  had  an  equivalent  response.  However,   Joubert  et  al  also  suggested  that  variations  in  natural  challenge  had  a  greater  influence  on   the  inflammatory  response  as  opposed  to  disease  state.  Thus  presumably,  variations  in   composition  or  concentrations  of  organic  dust  could  influence  the  response  on  the  alveolar   macrophage,  and  account  for  differences  between  studies.   In  vitro  activated  M1  and  M2  phenotypes  typically  have  distinct  gene  expression   patterns  and  functions.  It  is  interesting  that  complex  macrophage  phenotypes  are  present   in  both  populations.  At  baseline  the  RAO-­‐susceptible  expresses  an  immune-­‐regulatory  M2-­‐ like  phenotype  (producing  IL-­‐10)  that  surprisingly  displays  enhanced  LPS  responsiveness.   Further,  following  natural  challenge  control  horses  display  mixed  pro-­‐inflammatory  M1   markers,  the  M2  marker  CD206  and  enhanced  LPS  responsiveness.  These  phenotypes  are   likely  a  consequence  of  the  complex  physiological  environment  of  the  inflamed  lung.   80   Macrophage  activation  states  are  dynamic  and  can  change  with  the  course  of   inflammation138,143  and  dual  up-­‐regulation  of  M1  and  M2  phenotypic  markers  can  result   from  represent  exposure  to  a  mixture  of  M1  and  M2  activating  stimuli144  or  could  be   indicative  of  dynamic  transition  between  activation  states.138     In  summary,  RAO-­‐susceptible  horses  in  remission  possess  a  divergent  alveolar   macrophage  phenotype  that  responds  differently  to  natural  challenge.  Overall,  this   suggests  that  the  alveolar  macrophage  plays  a  prominent  role  in  RAO  immunopathology.   However,  exactly  how  this  phenotype  affects  other  cells  of  the  innate  and  adaptive  immune   system  in  this  complex  inflammatory  environment  remains  to  be  determined.     81   Chapter  4.   Concluding  Discussion   A  central  feature  of  RAO  is  a  greatly  enhanced  inflammatory  response  to  HD   compared  to  control  horses.62  However,  the  underlying  immune  mechanisms  that  result  in   this  differential  inflammatory  response  are  unknown.  It  is  clear,  however,  that  HD  induces   an  inflammatory  response  from  the  innate  immune  system  and  that  RAO-­‐susceptible   horses  have  a  hyperresponsive  innate  immune  system.61  However,  which  immune  cells  or   exact  mechanisms  that  underlie  this  are  unknown.  The  work  presented  here  addressed  the   overarching  hypothesis  that  the  differential  response  to  HD  would  be  associated  with   differential  alveolar  macrophage  phenotypes.     The  equine  specific,  transcriptional,  gene  expression  markers  of  alveolar  M1  and  M2   phenotypes  have  been  presented.  Data  from  other  species  indicate  that  gene  expression   profiles  can  differ  between  monocyte/macrophage  subpopulations234  and  macrophages   from  different  sources  exhibit  differences  in  M2  gene  expression  profiles  and  function.235   The  data  presented  in  Chapter  2  is  thus  particularly  relevant  for  the  study  of  equine   pulmonary  disease  as  alveolar  macrophages  (and  not  monocyte-­‐derived-­‐macrophages)   were  used  to  investigate  M1/M2  phenotypes  and  the  agonists  used  to  study  the   inflammatory  response  were  selected  due  to  their  presence  in  hay  dust.  These   investigations  could  be  extended  in  the  future  to  include  gene  expression  characterization   of  additional  in-­‐vitro  derived  alternative  macrophages.110  The  characterization  of  equine   macrophage  phenotypes  provides  novel  and  important  information  about  equine  alveolar   82   macrophage  biology  and  adds  to  a  growing  body  of  evidence  of  unique  equine  immune   mechanisms.     The  data  presented  in  Chapter  3  demonstrates  that  RAO-­‐susceptible  horses  in   remission  possess  a  divergent  alveolar  macrophage  phenotype  compared  to  control  horses.   Based  on  the  increased  expression  of  IL-­‐10,  this  could  be  considered  an  M2-­‐immune   regulatory  phenotype  that  retains  LPS  responsiveness.  The  presence  of  different   macrophage  phenotypes  suggests  that  an  M2-­‐immune  regulatory  macrophage  plays  a  role   in  RAO  immunopathology.  However,  further  studies  are  required  to  determine  what  factors   induce  this  phenotype  and  how  this  phenotype  contributes  to  individual  pulmonary   responses.  It  will  be  important  to  next  investigate  interactions  with  adjacent  cells  in  the   lung  environment  to  explore  if  this  phenotype  promotes  or  attenuates  the  inflammatory   response.  Although  IL-­‐10  is  an  immune  regulatory  cytokine  it  is  possible  that  this   phenotype  promotes  neutrophilic  inflammation.  Macrophage  phagocytosis  of  apoptotic   neutrophils  is  an  important  mechanism  for  resolution  of  airway  neutrophilia,9  however,   persistent  neutrophilia  is  common  weeks  after  antigen  avoidance.20  Alternatively-­‐activated   macrophages  show  impaired  phagocytosis  of  bacteria  and  zymosan,163  thus  it  is  possible   that  phagocytosis  of  apoptotic  neutrophils  is  similarly  impaired.  Further,  impaired   phagocytosis  of  hay  dust  components  could  promote  persistence  of  inflammation  as  it   could  prolong  the  presence  of  these  stimulatory  molecules  within  the  airways.  As  a   counterbalance,  IL-­‐10  can  reduce  neutrophil  pro-­‐inflammatory  cytokine  production236  thus   it  would  also  be  interesting  to  investigate  the  effects  of  alveolar  macrophage  products  on   the  neutrophil  inflammatory  response.     83   In  summary,  these  data  provide  novel  information  that  contributes  to  our   knowledge  of  equine  alveolar  macrophage  biology  and  RAO  immunopathology  and   supports  that  continued  investigations  of  the  alveolar  macrophage  in  RAO  disease   pathogenesis  are  warranted.       84   APPENDICES   85   Appendix  1.   Selection  of  Reference  Genes  for  Quantitative  Real  Time  PCR   Introduction     In  the  reported  studies,  the  alterations  of  alveolar  macrophage  phenotype  are   evaluated  by  measuring  gene  expression.  These  changes  were  measured  using  relative   quantification;  which  standardizes  the  target  gene  expression  against  that  of  a  stable   reference  gene  (ΔCT=CT  target  gene  -­‐  CT  reference  gene).  This  allows  correction  for  differences  in   sample  quantity  and  quality  so  that  changes  in  target  gene  expression  reflect  gene-­‐specific   variation.  A  variety  of  reference  genes  are  commonly  employed  to  perform  this  function,   however,  the  stability  of  reference  genes  can  vary  between  biological  sample  types  (e.g   macrophage  versus  neutrophil)  and  experimental  methodology.  It  is  now  recommended   that  the  reference  gene  stability  be  validated  for  each  tissue  type  or  experimental   methodology,  and  it  is  also  suggested  that  an  average  of  3  or  more  endogenous  genes  be   used.237  Thus  to  study  equine  alveolar  macrophage  phenotype  characterization  using  qRT-­‐ PCR,  it  was  first  important  to  establish  the  most  appropriate  reference  genes.  The  stability   of  9  candidate  reference  genes  was  evaluated  (Table  7).  These  were  tested  in  alveolar   macrophages  enriched  from  6  control  horses  (using  adhesion  technique)  that  were   untreated  or  stimulated  with  1ug/ml  LPS  (n=15  samples  total).  The  stability  of  the   endogenous  genes  was  then  evaluated  using  Best  Keeper  analysis.238     86   Table  7.  Panel  of  Candidate  Reference  Genes   Reference  Gene   β-­‐actin   Ubiquitin   β-­‐2  microglobulin   Elongation  factor  1α   18s  ribosomal  RNA   Succinate  dehydrogenase   complex   Abbreviation   ACTIN   UBIQ   B2M   ELF1   18S   SDHA   Glyceraldehyde-­‐3-­‐phosphate   GAPDH   dehydrogenase   Hypoxanthine   HPRT   phosphoribosyltransferase   BGUS   β  glucuronidase   Results  and  Discussion   Alveolar  macrophages  were  isolated  from  six  control  horses.  Bronchoalveolar   lavage  was  performed  as  standard  and  BAL  fluid  (BALF)  was  immediately  placed  on  ice.   BALF  was  then  transferred  to  sterile  50ml  tubes  and  centrifuged  at  400xg  for  8  min.   Supernatant  was  discarded  and  cells  were  washed  twice  more  then  counted  and   resuspended  at  a  concentration  of  1x10^6  cells/ml  in  RPMI  containing  10%  heat   inactivated  equine  serum,  antimycotic  and  antibiotic.  Cells  were  then  transferred  onto  6   well  culture  dishes  (3mls),  allowed  to  adhere  for  2  hours  at  37°C,  5%  CO2,  then  non   adherent  cells  were  washed  off.  Cells  were  then  bathed  in  equine  media  alone  (control)  or   equine  media  containing  LPS  (1µg/ml  or  ml)  and  incubated  for  1hr  (n=8),  or  9  hours  (n=7)   after  which  cells  were  rinsed  once  with  HBSS,  harvested  by  adding  RLT-­‐Lysis  Buffer   (Qiagen)  and  homogenized  using  Qiagen  QIAshredderTM  spin  columns.  Total  RNA  was   extracted  and  purified  using  Qiagen  RNeasy®  Mini  Kit  and  then  frozen  at  -­‐80°C  until  RNA   extraction.  RNA  concentration  was  measured  using  a  spectrophotometer  and  integrity  of   87   RNA  (RIN)  was  assessed  using  the  Agilent  2100  Bioanalyzer®  (Agilent  Technologies)  and  a   RIN  score  of  >6.5  was  considered  acceptable.     Equal  concentrations  of  RNA  were  reverse  transcribed  (High  Capacity  cDNA   Reverse  Transcription  Kit®,  Applied  Biosystems)  to  create  cDNA.    The  cDNA  was  then  pre-­‐ amplified  (TaqMan®  PreAmp  Master  Mix  Kit,  Life  Technologies)  according  to   manufacturer's  instructions).  Quantitative  PCR  was  performed  using  TaqMan®  Gene   Expression  Master  Mix  (Life  Technologies)  and  TaqMan®  Expression  Assays.  PCR  was   performed  in  triplicate  on  ABI  7900HT  Real-­‐Time  PCR  System®  (Applied  Biosystems).   The  descriptive  statistics  for  each  candidate  reference  gene  are  shown  (Table  8).   Based  on  the  standard  deviation  (SD),  the  genes  were  ranked  1-­‐9  with  1  being  the  lowest   SD.  Genes  with  SD>1  was  considered  too  variable  and  excluded  from  further  analysis  in   BestKeeper.   Using  the  CT  values,  BestKeeper  analysis  creates  a  “BestKeeper  Index”  for  each  sample  and   then  performs  repeated  pair-­‐wise  correlation  analysis  creating  correlation  coefficient  (r)   and  p-­‐value  (p)  for  each  gene  (Table  9).  The  genes  with  the  highest  correlation  coefficient   and  lowest  p-­‐values  were  β-­‐actin,  ELF1  and  HPRT.  Analysis  of  the  average  CT  of  ELF1  and   HPRT  together  slightly  improved  the  correlation  coefficient  and  p-­‐value.  The  average  of  β-­‐ actin,  ELF1  and  HPRT  was  almost  identical  to  the  combination  ELF1  and  HPRT.  Thus  based   on  this  data  ELF1  and  HPRT  were  selected  for  as  the  most  appropriate  reference  genes.  The   alveolar  macrophages  used  for  this  validation  experiment  were  isolated  from  healthy   control  horses  and  did  not  include  RAO-­‐susceptible  horses.  However,  there  was  no   difference  in  reference  gene  values  (average  of  ELF/HPRT)  between  groups  (RAO-­‐ 88   susceptible,  15.844±0.40,  control  horses,  15.873±0.419  (mean  CT  ±sd,  chapter  1)   indicating  that  these  genes  were  appropriate  for  use.     Table  8.  Descriptive  Statistics  of  Candidate  Reference  Genes  Based  on  the  Cross  Threshold   Point  (CT)     geo Mean [CT] ar Mean [CT] min [CT] max [CT] std dev [± CT] CV [% CT] ranking ACTIN 8.84 8.86 8.28 10.04 0.48 5.40 4 UBIQ 23.46 23.50 20.38 26.50 *1.06 4.51 7 B2M 6.61 6.62 6.24 7.20 0.20 3.09 1 ELF1 7.36 7.37 6.73 8.24 0.28 3.83 2 18S 12.34 12.35 11.62 13.99 0.43 3.44 3 SDHA 15.47 15.49 14.38 17.20 0.58 3.72 6 GAPDH 25.81 25.99 19.48 30.75 *2.23 8.58 9 HPRT BGUS 15.92 16.33 15.94 16.38 14.74 14.70 17.47 18.61 0.55 *1.06 3.46 6.52 5 8 Data  represents  descriptive  statistics  (CT)  for  each  candidate  reference  gene  from  15   samples.  *Indicates  genes  excluded  from  analysis  due  to  standard  deviation  >1     Table  9.  BestKeeper  Analysis  Showing  Correlation  Coefficient  (r)  and  P-­‐Value  (p)  From   Candidate  Reference  Genes   !! Correla:on!coefficient! (r)! pJvalue!(p)! ACTIN! B2M! ELF! 18S! SDHA! HPRT! HPRT/ELF! ACTIN/ELF/! HPRT! 0.866! 0.001! 0.301! 0.275! 0.93! 0.001! 0.68! 0.005! 0.36! 0.1! 0.85! 0.001! 0.94! 0.001! 0.958! 0.001! 89   Appendix  2.   Characterization  of  β-­‐Glucan  Receptor  Isoforms  in  the  Horse   Introduction   Equine  recurrent  airway  obstruction  (RAO)  is  a  chronic  inflammatory  disease  that  is   triggered  with  the  inhalation  of  hay  dust.  The  exact  immunopathology  of  RAO  is  unclear,   and  while  an  allergic  etiology  has  been  proposed  the  data  supporting  this  is  conflicting.   However,  in  addition  to  containing  potential  allergens,  hay  dust  contains  a  complex   mixture  of  microbial  cell  wall  components202  which  can  stimulate  the  innate  immune   system.  The  presence  of  fungal  elements  in  hay  dust  is  particularly  significant  as  both   moldy  hay27  (overgrowth  of  fungal  species)  and  the  β-­‐glucan  content  of  hay  dust62  are   potent  triggers  for  inducing  exacerbation  of  RAO.     Fungal  cell  walls  contain  the  complex  carbohydrate  β-­‐glucan,239  which  binds  to  the   innate  pathogen  recognition  receptor,  the  β-­‐glucan  receptor  ((βGR),  Dectin  1  Dendritic  cell   associated  C-­‐type  lectin)).  This  receptor  is  expressed  by  many  innate  immune  cells   including  neutrophils  and  macrophages,  and  stimulation  of  this  receptor  is  important  in   fungal  immunity.116  The  βGR  is  a  C-­‐type  lectin  transmembrane  receptor  and  is  composed  of   a  glycosylated  extracellular  C-­‐type  lectin  binding  domain,  a  short  extra-­‐cellular  stalk,  a   transmembrane  domain  and  a  cytoplasmic  tail  that  contains  a  signaling  domain.  In  other   species a  number  of  splice  variants  of  the  βGR  gene  (CLEC7A)  are  expressed.240–244 Principally,  2  major  splice  isoforms  are  expressed  that  differ  in  their  presence  of  the   extracellular  stalk.  The  full  length  (βGRA)  or  stalkless  isoforms  (βGRB)  are  both  functional   in  recognizing  β-­‐glucan.240,241  Additionally,  6  less  abundant  splice  variants  (minor  variants)   90   have  been  described  in  humans  and  1  additional  minor  variant  has  been  described  in   pigs.242  In  humans,  these  minor  variants  have  no  β-­‐glucan  binding  capacity  but  it  is   proposed  that  their  expression  could  serve  a  regulatory  role.241   Thus,  despite  β-­‐glucan  being  a  major  component  of  hay  dust  there  are  no   investigations  characterizing  the  equine  βGR.  Thus  our  objective  was  to  determine  if   equine  splice  isoforms  of  the  βGR  are  expressed  in  pulmonary  cells.      Materials  and  Methods   To  clone  the  β-­‐glucan  receptor  the  human  and  equine  CLEC7A  gene  sequences  were   aligned  (UCSC  Genome  Browser/EquCab2.0).  Primers  were  designed  using  Primer3   software  to  amplify  the  full-­‐length  coding  sequence  (forward  primer  5’-­‐3’  TCA  AAC  GCT   ATG  TCA  ATT  CAG  G,  reverse  primer  5’-­‐3’  TGG  TCG  TAA  ATG  ATT  GAT  AGG  TG).  Equine   pulmonary  cells  were  obtained  from  bronchoalveolar  lavage  (as  previously  described245)   from  RAO-­‐susceptible  and  control  horses.  Some  samples  were  collected  from  RAO-­‐ susceptible  and  control  horses  prior  to  and  following  >1week  exposure  to  hay  and  straw.   Tissues  (spleen,  respiratory  mucosa)  were  collected  post  euthanasia  from  healthy  horses   and  immediately  snap  frozen  using  liquid  nitrogen  and  the  stored  at  -­‐80°C  until  RNA   isolation.  BAL  cells  were  lysed  using  RLT-­‐Lysis  Buffer  (Qiagen)  and  homogenized  using  a   Qiagen  QIAshredder™  spin  columns.  Tissue  samples  were  disrupted  using  a  mortar  and   pestle  and  RLT-­‐Lysis  Buffer  and  homogenized  using  Qiagen  QIAshredder™.  RNA  was   isolated  and  purified  from  all  samples  using  Qiagen  RNeasy  Mini®Kit  which  involves  a   DNAase  digest  step.  RNA  concentration  was  measured  using  Qubit®  2.0  fluorometer  and   RNA  and  was  reverse  transcribed  (High  Capacity  cDNA  Reverse  Transcription  Kit,  Applied   91   Biosystems)  to  create  cDNA.  PCR  was  performed  (4  minutes  at  94°C,  40  cycles  of  1  minute   94°C,  1  minute  60°C,  1  minute  72°C)  using  Taq  DNA  Polymerase  (Life  Technologies™)  and   PCR  products  were  visualized  using  gel  electrophoresis  using  ethidium  bromide  stained   2%  agarose  gel.  Amplified  PCR  products  from  1  RAO-­‐susceptible  and  1  control  horse  were   mixed  together  and  cloned  using  Zero  Blunt®  TOPO®  PCR  Cloning  Kit  (Life  Technologies™).   Transformed  bacteria  were  plated  overnight  on  selective  media  (LB  with  50ug/ml   kanamycin)  after  which  positive  transformants  were  transferred  onto  a  reference  agar   plate  (plate  with  labeled  grid  lines).  For  each  positive  transformant,  PCR  and  2%  agarose   gel  electrophoresis  was  used  to  assess  the  cloned  product  size.  A  selection  of  clones  with   variable  product  sizes  were  sequenced  using  ABI-­‐3730  genetic  analyzer.  The  resulting   sequences  from  each  sample  were  aligned  next  to  the  human  and  equine  CLEC7A  gene   sequences  using  the  UCSC  Genome  Browser.  Using  the  sequences  derived  from  cloning,   primers  were  designed  to  specifically  amplify  the  splice  variants  and  the  expression  of  the   splice  variants  was  investigated  in  a  variety  of  tissues  using  RT-­‐PCR  and  gel  electrophoresis   as  above.       Results  and  Discussion   RT-­‐PCR  and  gel  electrophoresis  indicated  two  main  variants  with  a  large  isoform   being  approximately  800  base  pair  (bp)  in  size  and  a  smaller  isoform  being  approximately   600  bp  (Figure  8).  These  bands  approximately  corresponded  to  the  full-­‐length  isoform   (βGRA)  (comprised  of  exons  1-­‐6,  744  base  pairs)  and  the  stalkless  isoform  (βGRB,  spliced   exon  3,  606  base  pairs).  Sequencing  of  the  cloned  PCR  products  confirmed  that  these   isoform  variants  corresponded  to  human  βGRA  and  βGRB  respectively  (Figure  9).  Thus,   92   similar  to  other  species,240–244  the  two  major  isoforms  of  the  βGR  (βGRA  and  βGRB)  are   expressed  in  the  horse.  The  intensity  of  the  lower  band  was  greater  suggesting  that  in  BAL   cells  and  equine  spleen  the  βGRB  isoform  is  the  predominantly  expressed.  This  is  similar  to   the  expression  pattern  of  human  monocytes,  however,  expression  patterns  of  βGRA  and   βGRB  can  differ  amongst  cell  types  suggesting  cell  specific  regulation.241  Regulation  of   isoform  expression  could  impact  the  inflammatory  response,  as  zymosan  stimulated  βGRA   and  βGRB  produce  significantly  different  quantities  of  TNFα.246   93   Control!horse RAO$horse Spleen Figure  8.  Expression  of  β-­‐Glucan  Receptor  Splice  Isoforms  Using  RT-­‐PCR   800 600 ! PCR  amplified  cDNA  from  equine  bronchoalveolar  lavage  cells  (taken  from  a  normal   healthy  horse  and  a  RAO-­‐susceptible  horse)  and  spleen  was  analyzed  for  the     presence  of  the  β-­‐glucan  receptor  splice  variants.  The  two  major  bands  at  800  and  600bp   corresponding  to  βGRA  and  βGRB  respectively  are  indicated  by  arrows.   94   In  addition,  sequencing  also  identified  3  further  isoforms,  which  were  variations  of  βGRA   (Figure  9).  Two  of  these  isoforms  (minor  isoform  1  and  2)  contained  small  insertions   (between  exons  5  and  6)  and  do  not  correspond  to  any  human  isoform  variants.  Exons  3   and  5  were  deleted  from  minor  isoform  3  and  this  configuration  corresponds  with  human   βGRD.241  Minor  isoforms  1,2,  and  3  were  720,  663  and  471  nucleotides  long  respectively.   Exons  1  and  2  code  for  the  cytoplasmic  tail  and  transmembrane  domain,  exon  3  codes  for   the  extracellular  stalk  and  exons  4-­‐6  code  for  the  carbohydrate  recognition  domain  (CRD).   As  each  of  the  3  minor  variants  detected  had  sequence  alterations  that  affected  the  CRD     these  variants  would  be  predicted  to  have  impaired  βGR  function  and  indeed,  human  minor   variants  βGRC-­‐H  do  not  have  zymosan  binding  capacity.241  These  minor  isoforms  were  not   observed  as  distinct  bands  when  amplified  PCR  products  (primers  amplified  full  length   coding  sequence)  were  analyzed  using  ethidium  bromide  agarose  gel  electrophoresis   (Figure  8),  thus  it  is  likely  that  the  level  of  expression  of  these  variants  is  minor  relative  to   βGRA  and  βGRB  isoforms  and  below  the  level  of  detection  using  this  method.   95   Figure  9.  Equine  β-­‐Glucan  Receptor  Splice  Isoforms  Determined  by  Cloning  and  Sequencing     The  cDNA  from  1  RAO-­‐susceptible  and  1  control  horse  were  amplified  with  primers   designed  to  include  the  entire  coding  region  of  the  CLEC7A  gene.  PCR  product  was  then   cloned  in  E.coli  and  positive  transformants  were  selected  for  sequencing  and  Figure  9     illustrates  the  splice  variants  that  were  detected.  The  colored  bar  illustrates  the  regions  of   the  receptor  that  each  exon  encodes  (based  on  human  β-­‐glucan  receptor).  The  red  (5a)  and   blue  (5b)  boxes  represent  exon  insertions.  All  minor  isoform  variants  possess  alterations  in   the  carbohydrate  recognition  domain.   96   β-­‐glucan  expression  on  myeloid  cells  is  well  described.  However,  βGR  is  also  expressed  on   airway  epithelium.247  Using  primers  designed  to  specifically  amplify  individual  isoforms   (Table  10),  βGRA  and  βGRB  were  found  to  be  expressed  in  respiratory  mucosa  and  lung  in   addition  to  non-­‐respiratory  tissues  liver  and  spleen  (Table  11).  However,  the  relative   quantities  the  isoforms  were  not  compared.  Further,  minor  isoform  3  was  expressed  in  all   tissues  tested  but  minor  isoforms  1  and  2  were  not  consistently  expressed,  suggesting  a   degree  of  tissue  specificity  for  isoform  expression.     97   ! Table  10.  Location  of  Primers  Used  to  Identify  β-­‐Glucan  Splice  Isoforms   Splice'Variant' βGRA! βGRB! Minor!isoform!1! Minor!isoform!2! Minor!isoform!3! ! Forward'Primer'' ' Exon!4*3!junction! Exon!4! Exon!6*5a!junction! Exon!5b*5!junction! Exon!6*4!junction! Reverse'Primer' Exon!3! Exon!4*2!junction! Exon!5a! Exon!5! Exon!4! Table  11.  Tissue  Specific  Expression  of  β-­‐Glucan  Receptor  Splice  Isoforms   Equine'Splice' Nasal'Mucosa' Tracheal' Lung' Liver' isoform' mucosa' βGRA' 2/2! 2/3! 2/2! 1/2! βGRB' 2/2! 2/3! 2/2! 2/2! Minor'isoform'1' 0/1! 1/2! 1/2! 0/1! Minor'isoform'2' 2/2!(weak)! 0/3! 1/3!(weak)! 1/2!(weak)! Minor'isoform'3' 2/2! 3/3! 2/2! 2/2! Spleen' 2/2! 2/2! 1/1! 0/2! 2/2! ! For  each  isoform  the  numbers  indicate  the  frequency  positive  samples  /  the  total  number   of  samples  tested  in  each  tissue  e.g  2  of  3  (2/3)  tracheal  mucosal  samples  expressed  βGRA.   Minor  isoform  2  was  only  weakly  expresses  in  certain  tissues  as  indicated.     98   Similarly,  expression  of  isoform  variants  was  evaluated  in  alveolar  macrophages   isolated  from  normal  and  RAO-­‐susceptible  horses  preceding  and  following  exposure  to  hay.   Both  groups  of  horses  expressed  each  of  the  5  isoform  variants  both  preceding  and   following  exposure  to  hay/straw  (Figure  10),  suggesting  that  major  differences  in  alveolar   macrophage  regulation  of  βGR  isoform  variants  was  not  associated  with  exposure  to   hay/straw  in  either  group.     Figure  10.  Expression  Profile  of  β-­‐Glucan  Receptor  Splice  Isoforms  in  Alveolar   Macrophages  From  RAO-­‐Susceptible  and  Control  Horses     RT-­‐PCR  showing  the  β-­‐glucan  receptor  splice  variants  detected  in  alveolar  macrophages.   Lanes  1&  2  are  representative  of  alveolar  macrophages  from  RAO-­‐susceptible  horses  (n=3)   before  and  after  hay  exposure  respectively.  Lanes  3  &  4  are     99   Figure  10  (cont'd)    representative  of  alveolar  macrophages  from  normal  horses  (n=3)  before  and  after  hay   exposure  respectively.  All  isoforms  were  detected  in  all  samples   The  functional  consequence  of  alternative  splicing  is  not  always  apparent.  It  is   suggested  that  alternative  splicing  is  often  a  non-­‐functional  stochastic  event,248  however,   alternative  splicing  may  also  serve  as  a  means  of  gene  regulation  by  generating  splice   isoforms  that  are  subjected  to  nonsense  mediated  decay  or  generating  proteins  with   different  functions.249,250  It  is  not  known  if  changes  in  the  expression  of  βGR  isoforms  could   contribute  to  regulation  the  inflammatory  response  to  fungi.     In  conclusion,  βGRA  and  βGRB  are  expressed  in  horses  in  addition  to  3  minor   isoforms.  Expression  is  present  in  white  blood  cells  but  also  respiratory  epithelium.  This  is   the  first  study  to  characterize  the  β-­‐glucan  receptor  in  horses  and  will  provide  a   fundamental  platform  for  future  studies  investigation  interaction  of  β-­‐glucan  and  the   equine  innate  immune  system.     100   BIBLIOGRAPHY   101   BIBLIOGRAPHY   1. Robinson  NE.  International  Workshop  on  Equine  Chronic  Airway  Disease.  Michigan  State University  16-­‐18  June  2000.  Equine  Vet  J  2001;33:5–19.   2. Couëtil  LL,  Ward  MP.  Analysis  of  risk  factors  for  recurrent  airway  obstruction  in  North American  horses:  1,444  cases  (1990-­‐1999).  J  Am  Vet  Med  Assoc  2003;223:1645–50.   3. Hotchkiss  JW,  Reid  SWJ,  Christley  RM.  A  survey  of  horse  owners  in  Great  Britain regarding  horses  in  their  care.  Part  2:  Risk  factors  for  recurrent  airway  obstruction.  Equine   Vet  J  2007;39:301–308.   4. Thomson  JR,  Mcpherson  EA.  Effects  of  environmental  control  on  pulmonary  function  of horses  affected  with  chronic  obstructive  pu  I  mona  ry  disease.  1984;16:35–38.   5. Vandenput  S,  Votion  D,  Duvivier  DH,  et  al.  Effect  of  a  set  stabled  environmental  control on  pulmonary  function  and  airway  reactivity  of  COPD  affected  horses.  Vet  J  1998;155:189– 95.   6. Dixon  PM,  Railton  DI,  McGorum  BC.  Equine  pulmonary  disease:  a  case  control  study  of 300  referred  cases.  Part  2:  Details  of  animals  and  of  historical  and  clinical  findings.  Equine   Vet  J  1995;27:422–7.   7. Aviza  GA,  Ainsworth  DM,  Eicker  SW,  et  al.  Outcome  of  horses  diagnosed  with  and  treated for  RAO_  Case  Report.  2001;13:243–246.   8. Fairbairn  SM,  Page  CP,  Lees  P,  et  al.  Early  neutrophil  but  not  eosinophil  or  platelet recruitment  to  the  lungs  of  allergic  horses  following  antigen  exposure.  Clin  Exp  Allergy   1993;23:821–8.   9. Brazil  TJ,  Dagleish  MP,  McGorum  BC,  et  al.  Kinetics  of  pulmonary  neutrophil  recruitment and  clearance  in  a  natural  and  spontaneously  resolving  model  of  airway  inflammation.  Clin   Exp  Allergy  2005;35:854–65.   10. Art  T,  Franck  T,  Lekeux  P,  et  al.  Myeloperoxidase  concentration  in  bronchoalveolar lavage  fluid  from  healthy  horses  and  those  with  recurrent  airway  obstruction.  Can  J  Vet  Res   2006;70:291–6.   11. Art  T,  Kirschvink  N,  Smith  N,  et  al.  Indices  of  oxidative  stress  in  blood  and  pulmonary epithelium  lining  fluid  in  horses  suffering  from  recurrent  airway  obstruction.  Equine  Vet  J   1999;31:397–401.   102   12. Joubert  P,  Silversides  DW,  Lavoie  JP.  Equine  neutrophils  express  mRNA  for  tumour necrosis  factor-­‐alpha,  interleukin  (IL)-­‐1beta,  IL-­‐6,  IL-­‐8,  macrophage-­‐inflammatory-­‐protein-­‐ 2  but  not  for  IL-­‐4,  IL-­‐5  and  interferon-­‐gamma.  Equine  Vet  J  2001;33:730–3.   13. Ainsworth  DM,  Wagner  B,  Franchini  M,  et  al.  Time-­‐dependent  alterations  in  gene expression  of  interleukin-­‐8  in  the  bronchial  epithelium  of  horses  with  recurrent  airway   obstruction.  Am  J  Vet  Res  2006;67:669–77.   14. Debrue  M,  Hamilton  E,  Joubert  P,  et  al.  Chronic  exacerbation  of  equine  heaves  is associated  with  an  increased  expression  of  interleukin-­‐17  mRNA  in  bronchoalveolar  lavage   cells.  Vet  Immunol  Immunopathol  2005;105:25–31.   15. Deluca  L,  Erb  HN,  Young  JC,  et  al.  The  Effect  of  Adding  Oral  Dexamethasone  to  Feed Alterations  on  the  Airway  Cell  Inflammatory  Gene  Expression  in  Stabled  Horses  Affected   with  Recurrent  Airway  Obstruction.  J  Vet  Intern  Med  2008:427–435.   16. Padoan  E,  Ferraresso  S,  Pegolo  S,  et  al.  Real  Time  Rt-­‐Pcr  Analysis  Of  Inflammatory Mediator  Expression  In  Recurrent  Airway  Obstruction-­‐Affected  Horses.  Vet  Immunol   Immunopathol  2013;156:190–199.   17. Franchini  M,  Gill  U,  von  Fellenberg  R,  et  al.  Interleukin-­‐8  concentration  and  neutrophil chemotactic  activity  in  bronchoalveolar  lavage  fluid  of  horses  with  chronic  obstructive   pulmonary  disease  following  exposure  to  hay.  Am  J  Vet  Res  2000;61:1369–74.   18. Riihimäki  M,  Raine  A,  Art  T,  et  al.  Partial  divergence  of  cytokine  mRNA  expression  in bronchial  tissues  compared  to  bronchoalveolar  lavage  cells  in  horses  with  recurrent  airway   obstruction.  Vet  Immunol  Immunopathol  2008;122:256–64.   19. Ainsworth  DM,  Wagner  B,  Erb  HN,  et  al.  The  effects  of  in  vitro  exposure  to  hay  dust  on expression  of  interleukin-­‐17,  23,  8,  1B  and  chemokine  (C-­‐X-­‐C  motif)  ligand  2  by  pulmonary   mononuclear  cells  isolated  from  hroses  chronically  affected  with  recurrent  airway  disease.   AJVR  2007;12:1361–1369.   20. Giguère  S,  Viel  L,  Lee  E,  et  al.  Cytokine  induction  in  pulmonary  airways  of  horses  with heaves  and  effect  of  therapy  with  inhaled  fluticasone  propionate.  Vet  Immunol   Immunopathol  2002;85:147–58.   21. Simonen-­‐Jokinen  T,  Pirie  RS,  McGorum  B,  et  al.  Dose  responses  to  inhalation  of endotoxin,  hay  dust  suspension  and  Aspergillus  fumigatus  extract  in  horses  as  measured  by   levels  and  activation  of  matrix  metalloproteinase-­‐9.  Equine  Vet  J  2005;37:155–60.   22. Lindberg  Å,  Robinson  NE,  Näsman-­‐glaser  B,  et  al.  Assessment  of  leukotriene  B4 production  in  leukocytes  from  horses  with  recurrent  airway  obstruction.  Am  J  Vet  Res   2004;65:289–295.   103     23.  Gray  PR,  Derksen  FJ,  Broadstone  R  V,  et  al.  Increased  pulmonary  production  of   immunoreactive  15-­‐hydroxyeicosatetraenoic  acid  in  an  animal  model  of  asthma.  Am  Rev   Respir  Dis  1992;145:1092–7.   24.  Gray  PR,  Derksen  FJ,  Robinson  NE,  et  al.  The  role  of  cyclooxygenase  products  in  the   acute  airway  obstruction  and  airway  hyperreactivity  of  ponies  with  heaves.  Am  Rev  Respir   Dis  1989;140:154–60.   25.  Pirie  RS,  Dixon  PM,  McGorum  BC.  Evaluation  of  nebulised  hay  dust  suspensions  (HDS)   for  the  diagnosis  and  investigation  of  heaves.  3:  Effect  of  fractionation  of  HDS.  Equine  Vet  J   2002;34:343–7.   26.  McPherson  EA,  Lawson  GH,  Murphy  JR,  et  al.  Chronic  obstructive  pulmonary  disease   (COPD)  in  horses:  aetiological  studies:  responses  to  intradermal  and  inhalation  antigenic   challenge.  Equine  Vet  J  1979;11:159–66.   27.  Mcgorum  C,  Dixon  PM,  Halliwell  REW.  Responses  of  horses  affected  with  chronic   obstructive  pulmonary  disease  to  inhalation  challenges  with  mould  antigens.  Equine  Vet  J   1993;25:261–267.   28.  Halliwell  RE,  McGorum  BC,  Irving  P,  et  al.  Local  and  systemic  antibody  production  in   horses  affected  with  chronic  obstructive  pulmonary  disease.  Vet  Immunol  Immunopathol   1993;38:201–15.   29.  Schmallenbach  KH,  Rahman  I,  Sasse  HH,  et  al.  Studies  on  pulmonary  and  systemic   Aspergillus  fumigatus-­‐specific  IgE  and  IgG  antibodies  in  horses  affected  with  chronic   obstructive  pulmonary  disease  (COPD).  Vet  Immunol  Immunopathol  1998;66:245–56.   30.  Tahon  L,  Baselgia  S,  Gerber  V,  et  al.  In  vitro  allergy  tests  compared  to  intradermal   testing  in  horses  with  recurrent  airway  obstruction.  Vet  Immunol  Immunopathol   2009;127:85–93.   31.  Eder  C,  Crameri  R,  Mayer  C,  et  al.  Allergen-­‐specific  IgE  levels  against  crude  mould  and   storage  mite  extracts  and  recombinant  mould  allergens  in  sera  from  horses  affected  with   chronic  bronchitis.  Vet  Immunol  Immunopathol  2000;73:241–53.   32.  Wong  DM,  Buechner-­‐maxwell  VA,  Manning  TO,  et  al.  Comparison  of  results  for   intradermal  testing  between  clinically  normal  horses  and  horses  affected  with  recurrent   airway  obstruction.  Am  J  Vet  Res  2004;8:1348–1355.   33.  Wagner  B.  IgE  in  horses:  occurrence  in  health  and  disease.  Vet  Immunol  Immunopathol   2009;132:21–30.   34.  Marti  E,  Gerber  V,  Wilson  AD,  et  al.  Report  of  the  3rd  Havemeyer  workshop  on  allergic   diseases  of  the  Horse,  Hólar,  Iceland,  June  2007.  Vet  Immunol  Immunopathol   2008;126:351–61.     104   35. Künzle  F,  Gerber  V,  Van  Der  Haegen  A,  et  al.  IgE-­‐bearing  cells  in  bronchoalveolar  lavage fluid  and  allergen-­‐specific  IgE  levels  in  sera  from  RAO-­‐affected  horses.  J  Vet  Med  A  Physiol   Pathol  Clin  Med  2007;54:40–7.   36. Van  der  Haegen  a,  Künzle  F,  Gerber  V,  et  al.  Mast  cells  and  IgE-­‐bearing  cells  in  lungs  of RAO-­‐affected  horses.  Vet  Immunol  Immunopathol  2005;108:325–34.   37. Deaton  CM,  Deaton  L,  Jose-­‐Cunilleras  E,  et  al.  Early  onset  airway  obstruction  in response  to  organic  dust  in  the  horse.  J  Appl  Physiol  2007;102:1071–7.   38. Cordeau  M,  Joubert  P,  Dewachi  O,  et  al.  IL-­‐4,  IL-­‐5  and  IFN-­‐gamma  mRNA  expression  in pulmonary  lymphocytes  in  equine  heaves.  Vet  Immunol  Immunopathol  2004;97:87–96.   39. Lavoie  J,  Maghni  K,  Desnoyers  M,  et  al.  Neutrophilic  Airway  Inflammation  in  Horses with  Heaves  Is  Characterized  by  a  Th2-­‐type  Cytokine  Profile.  Crit  Care  Med:2–5.   40. Bodey  KJ,  Semper  AE,  Redington  AE,  et  al.  Cytokine  profiles  of  BAL  T  cells  and  T-­‐cell clones  obtained  from  human  asthmatic  airways  after  local  allergen  challenge.  Allergy   1999;54:1083–93.   41. Robinson  DS,  Hamid  Q,  Ying  S,  et  al.  Predominant  Th2-­‐Like  Bronchoalveolar  T-­‐ Lymphocyte  Poluation  in  Atopic  Asthma.  N  Engl  J  Med  1992;30:298–304.   42. Ainsworth  D,  Grunig  G,  Matychak  MB,  et  al.  Recurrent  airway  obstruction  (RAO)  in horses  is  characterized  by  IFN-­‐gamma  and  IL-­‐8  production  in  bronchoalveolar  lavage  cells.   Vet  Immunol  Immunopathol  2003;96:83–91.   43. Kleiber  C,  McGorum  BC,  Horohov  DW,  et  al.  Cytokine  profiles  of  peripheral  blood  and airway  CD4  and  CD8  T  lymphocytes  in  horses  with  recurrent  airway  obstruction.  Vet   Immunol  Immunopathol  2005;104:91–7.   44. Pietra  M,  Cinotti  S,  Ducci  A,  et  al.  Time-­‐dependent  changes  of  cytokines  mRNA  in bronchoalveolar  lavage  fluid  from  symptomatic  Recurrent  Airway  Obstruction-­‐affected   horses.  Pol  J  Vet  Sci  2011;14:343–351.   45. Horohov  DW,  Beadle  RE,  Mouch  S,  et  al.  Temporal  regulation  of  cytokine  mRNA expression  in  equine  recurrent  airway  obstruction.  Vet  Immunol  Immunopathol   2005;108:237–45.   46. Ryhner  T,  Müller  N,  Balmer  V,  et  al.  Increased  mucus  accumulation  in  horses  chronically affected  with  recurrent  airway  obstruction  is  not  associated  with  up-­‐regulation  of  CLCA1,   EGFR,  MUC5AC,  Bcl-­‐2,  IL-­‐13  and  INF-­‐gamma  expression.  Vet  Immunol  Immunopathol   2008;125:8–17.   47. Woodruff  PG,  Modrek  B,  Choy  DF,  et  al.  T-­‐helper  type  2-­‐driven  inflammation  defines major  subphenotypes  of  asthma.  Am  J  Respir  Crit  Care  Med  2009;180:388–95.   105   48. Swinburne  JE,  Bogle  H,  Klukowska-­‐Rötzler  J,  et  al.  A  whole-­‐genome  scan  for  recurrent airway  obstruction  in  Warmblood  sport  horses  indicates  two  positional  candidate  regions.   Mamm  Genome  2009;20:504–15.   49. Ruth  H,  Green  RH,  Brightling  CE,  et  al.  The  reclassification  of  asthma  based  on subphenotypes.  Curr  Opin  Allergy  Clin  Immunol  2014;7:43–50.   50. Jost  U,  Klukowska-­‐Rötzler  J,  Dolf  G,  et  al.  A  region  on  equine  chromosome  13  is  linked  to recurrent  airway  obstruction  in  horses.  Equine  Vet  J  2007;39:236–241.   51. Gerber  V,  Baleri  D,  Klukowska-­‐Rötzler  J,  et  al.  Mixed  inheritance  of  equine  recurrent airway  obstruction.  J  Vet  Intern  Med  2009;23:626–30.   52. Lanz  S,  Gerber  V,  Marti  E,  et  al.  Effect  of  hay  dust  extract  and  cyathostomin  antigen stimulation  on  cytokine  expression  by  PBMC  in  horses  with  recurrent  airway  obstruction.   Vet  Immunol  Immunopathol  2013;155:229–37.   53. Leclere  M,  Lavoie-­‐Lamoureux  A,  Lavoie  J-­‐P.  Heaves,  an  asthma-­‐like  disease  of  horses. Respirology  2011;16:1027–46.   54. Corrigan  C.  Mechanisms  of  intrinsic  asthma.  Curr  Opin  Allergy  Clin  Immunol 2004;4:53–56.   55. Derksen  F,  Robinson  N,  Scott  J,  et  al.  Aerosolized  Micropolyspora  faeni  antigen  as  a cause  of  pulmonary  dysfunction  in  ponies  with  recurrent  airway  obstruction  (heaves).  Am  J   Vet  Res  1988;49:933–938.   56. Gregory  LG,  Lloyd  CM.  Orchestrating  house  dust  mite-­‐associated  allergy  in  the  lung. Trends  Immunol  2011;32:402–411.   57. Clarke  A,  Madelin  T.  Technique  for  assessing  respiratory  health  hazards  from  hay  and other  source  materials.  Equine  Vet  J  1987;19:442–7.   58. Pirie  RS,  McLachlan  G,  McGorum  BC.  Evaluation  of  nebulised  hay  dust  suspensions (HDS)  for  the  diagnosis  and  investigation  of  heaves.  1:  Preparation  and  composition  of   HDS.  Equine  Vet  J  2010;34:332–336.   59. Pirie  RS,  Dixon  PM,  McGorum  BC.  Endotoxin  contamination  contributes  to  the pulmonary  inflammatory  and  functional  response  to  Aspergillus  fumigatus  extract   inhalation  in  heaves  horses.  Clin  Exp  Allergy  2003;33:1289–96.   60. Pirie  RS,  Collie  DDS,  Dixon  PM,  et  al.  Inhaled  endotoxin  and  organic  dust  particulates have  synergistic  proinflammatory  effects  in  equine  heaves  (organic  dust-­‐induced  asthma).   Clin  Exp  Allergy  2003;33:676–83.   106   61. Pirie  RS,  Dixon  PM,  Collie  DD,  et  al.  Pulmonary  and  systemic  effects  of  inhaled  endotoxin in  control  and  heaves  horses.  Equine  Vet  J  2001;33:311–8.   62. Pirie  RS,  Collie  DDS,  Dixon  PM,  et  al.  Evaluation  of  nebulised  hay  dust  suspensions (HDS)  for  the  diagnosis  and  investigation  of  heaves.  2:  Effects  of  inhaled  HDS  on  control   and  heaves  horses.  Equine  Vet  J  2002;34:337–42.   63. Robinson  NE,  Berney  C,  Eberhart  S,  et  al.  Coughing,  mucus  accumulation,  airway obstruction,  and  airway  inflammation  in  control  horses  and  horses  with  recurrent  airway   obstruction.  Am  J  Vet  Res  2003;64:550–557.   64. Gerber  V,  Lindberg  A,  Berney  C,  et  al.  Airway  mucus  in  recurrent  airway  obstruction-­‐-­‐ short-­‐term  response  to  environmental  challenge.  J  Vet  Intern  Med  2004;18:92–7.   65. Berndt  A,  Derksen  FJ,  Venta  PJ,  et  al.  Expression  of  toll-­‐like  receptor  2  mRNA  in bronchial  epithelial  cells  is  not  induced  in  RAO-­‐affected  horses.  Equine  Vet  J  2009;41:76– 81.   66. Lugo  J,  Harkema  JR,  deFeijter-­‐Rupp  H,  et  al.  Airway  inflammation  is  associated  with mucous  cell  metaplasia  and  increased  intraepithelial  stored  mucosubstances  in  horses.  Vet   J  2006;172:293–301.   67. Lavoie-­‐Lamoureux  A,  Leclere  M,  Lemos  K,  et  al.  Markers  of  systemic  inflammation  in horses  with  heaves.  J  Vet  Intern  Med  2012;26:1419–26.   68. Lavoie-­‐Lamoureux  A,  Moran  K,  Beauchamp  G,  et  al.  IL-­‐4  activates  equine  neutrophils and  induces  a  mixed  inflammatory  cytokine  expression  profile  with  enhanced  neutrophil   chemotactic  mediator  release  ex  vivo.  Am  J  Physiol  Lung  Cell  Mol  Physiol  2010;299:L472– 82.   69. Berndt  A,  Derksen  FJ,  Venta  PJ,  et  al.  Elevated  amount  of  Toll-­‐like  receptor  4  mRNA  in bronchial  epithelial  cells  is  associated  with  airway  inflammation  in  horses  with  recurrent   airway  obstruction.  Am  J  Physiol  Lung  Cell  Mol  Physiol  2007;292:L936–43.   70. Alexis  N,  Eldridge  M,  Reed  W,  et  al.  CD14-­‐dependent  airway  neutrophil  response  to inhaled  LPS:  role  of  atopy.  J  Allergy  Clin  Immunol  2001;107:31–5.   71. Schwartz  DA,  Cook  DN.  Polymorphisms  of  the  Toll-­‐like  receptors  and  human  disease. Clin  Infect  Dis  2005;41  Suppl  7:S403–7.   72. Werners  AH,  Bull  S,  Vendrig  JC,  et  al.  Genotyping  of  Toll-­‐like  receptor  4,  myeloid differentiation  factor  2  and  CD-­‐14  in  the  horse:  an  investigation  into  the  influence  of   genetic  polymorphisms  on  the  LPS  induced  TNF-­‐alpha  response  in  equine  whole  blood.  Vet   Immunol  Immunopathol  2006;111:165–73.   107   73. Simonen-­‐Jokinen  T,  Pirie  RS,  McGorum  BC,  et  al.  Effect  of  composition  and  different fractions  of  hay  dust  suspension  on  inflammation  in  lungs  of  heaves-­‐affected  horses:  MMP-­‐ 9  and  MMP-­‐2  as  indicators  of  tissue  destruction.  Equine  Vet  J  2005;37:412–7.   74. Lilly  LM,  Gessner  MA,  Dunaway  CW,  et  al.  The  β-­‐glucan  receptor  dectin-­‐1  promotes  lung immunopathology  during  fungal  allergy  via  IL-­‐22.  J  Immunol  2012;189:3653–60.   75. Poole  JA,  Dooley  GP,  Saito  R,  et  al.  Muramic  acid,  endotoxin,  3-­‐hydroxy  fatty  acids,  and ergosterol  content  explain  monocyte  and  epithelial  cell  inflammatory  responses  to   agricultural  dusts.  J  Toxicol  Environ  Health  A  2010;73:684–700.   76. Elfman  L,  Riihimäki  M,  Pringle  J,  et  al.  Influence  of  horse  stable  environment  on  human airways.  J  Occup  Med  Toxicol  2009;4:10.   77. Poole  JA,  Wyatt  TA,  Kielian  T,  et  al.  Toll-­‐like  receptor  2  regulates  organic  dust-­‐induced airway  inflammation.  Am  J  Respir  Cell  Mol  Biol  2011;45:711–9.   78. Derksen  FJ,  Scott  JS,  Slocombe  RF,  et  al.  Micropolyspora  faeni  causes  airway inflammation  but  not  hyperresponsiveness  in  sensitized  ponies.  J  Appl  Physiol   1987;62:1398–404.   79. Bogaert  P,  Tournoy  KG,  Naessens  T,  et  al.  Where  asthma  and  hypersensitivity pneumonitis  meet  and  differ:  noneosinophilic  severe  asthma.  Am  J  Pathol  2009;174:3–13.   80. Rose  CS,  Lara  AR.  Murray  and  Nadel’s  Textbook  of  Respiratory  Medicine, Hypersensitivity  Pneumonitis.  Fifth  Edit.  (Mason  RJ,  Broaddus  VC,  Martin  TR,  eds.).  Elsevier   Inc.;  2010:1587–1600.   81. Ainsworth  DM,  Matychak  M,  Reyner  CL,  et  al.  Effects  of  in  vitro  exposure  to  hay  dust  on the  gene  expression  of  chemokines  and  cell  surface  receptors  in  primary  bronchial   epithelial  cell  cultures  established  from  horses  with  chronic  recurrent  airway  obstruction.   AJVR  2009;70:365–372.   82. Wilson  ME,  Berney  C,  Behan  AL,  et  al.  The  Effect  of  Intravenous  Lidocaine  Infusion  on Bronchoalveolar  Lavage  Cytology  in  Equine  Recurrent  Airway  Obstruction.  J  Vet  Intern   Med  2012;26:1427–32.   83. Jackson  CA,  Berney  C,  Jefcoat  AM,  et  al.  Environment  and  prednisone  interactions  in  the treatment  of  recurrent  airway  obstruction  (heaves).  Equine  Vet  J  2000;32:432–8.   84. Kirschvink  N,  Smith  N,  Fiévez  L,  et  al.  Effect  of  chronic  airway  inflammation  and exercise  on  pulmonary  and  systemic  antioxidant  status  of  healthy  and  heaves-­‐affected   horses.  Equine  Vet  J  2002;34:563–71.   108   85. Niedzwiedz  A,  Jaworski  Z,  Tykalowski  B,  et  al.  Neutrophil  and  macrophage  apoptosis  in bronchoalveolar  lavage  fluid  from  healthy  horses  and  horses  with  recurrent  airway   obstruction  (RAO).  BMC  Vet  Res  2014;10:29.   86. Jean  D,  Beauchamp  G,  Lavoie  J.  Evaluation  of  variations  in  brochoalveolar  lavage  fluid  in horses  with  recurrent  airway  obstruction.  Am  J  Vet  Res  2011;72:838–842.   87. Laan  TTJM,  Bull  S,  Pirie  R,  et  al.  The  role  of  alveolar  macrophages  in  the  pathogenesis  of recurrent  airway  obstruction  in  horses.  J  Vet  Intern  Med  2006;20:167–74.   88. Joubert  P,  Cordeau  M-­‐E,  Lavoie  J-­‐P.  Cytokine  mRNA  expression  of  pulmonary macrophages  varies  with  challenge  but  not  with  disease  state  in  horses  with  heaves  or  in   controls.  Vet  Immunol  Immunopathol  2011;142:236–242.   89. Landsman  L,  Jung  S.  Lung  Macrophages  Serve  as  Obligatory  Intermediate  between Blood  Monocytes  and  Alveolar  Macrophages.  J  Immunol  2007;179:3488–3494.   90. Sieweke  MH,  Allen  JE.  Beyond  stem  cells:  self-­‐renewal  of  differentiated  macrophages. Science  2013;342:1242974.   91. Murphy  J,  Summer  R,  Wilson  AA,  et  al.  The  prolonged  life-­‐span  of  alveolar  macrophages. Am  J  Respir  Cell  Mol  Biol  2008;38:380–5.   92. Hashimoto  D,  Chow  A,  Noizat  C,  et  al.  Tissue-­‐resident  macrophages  self-­‐maintain  locally throughout  adult  life  with  minimal  contribution  from  circulating  monocytes.  Immunity   2013;38:792–804.   93. Shi  C,  Pamer  EG.  Monocyte  recruitment  during  infection  and  inflammation.  Nat  Rev Immunol  2011;11:762–774.   94. Gautier  EL,  Shay  T,  Miller  J,  et  al.  Gene-­‐expression  profiles  and  transcriptional regulatory  pathways  that  underlie  the  identity  and  diversity  of  mouse  tissue  macrophages.   Nat  Immunol  2012;13:1118–28.   95. Lohmann-­‐Matthes  M-­‐L,  Steinmüller  C,  Franke-­‐Ullmann  G.  Pulmonary  macrophages.  Eur Respir  J  1994;7:1678–1689.   96. Hodgson  JL,  Hodgson  DR.  Collection  and  Analysis  of  Respiratory  Tract  Samples.  Equine Respiratory  Medicine  and  Surgery.  (McGorum  BC,  Dixon  PM,  Robinson  NE,  et  al.,  eds.).   Elsevier  Ltd;  2007:119–150.   97. Lambrecht  BN.  Alveolar  macrophage  in  the  driver’s  seat.  Immunity  2006;24:366–8. 98. Hussell  T,  Bell  TJ.  Alveolar  macrophages:  plasticity  in  a  tissue-­‐specific  context.  Nat  Rev Immunol  2014;14:81–93.   109   99. Munger  JS,  Huang  X,  Kawakatsu  H,  et  al.  The  integrin  alpha  v  beta  6  binds  and  activates latent  TGF  beta  1:  a  mechanism  for  regulating  pulmonary  inflammation  and  fibrosis.  Cell   1999;96:319–28.   100.  Gardai  SJ,  Xiao  Y-­‐Q,  Dickinson  M,  et  al.  By  binding  SIRPalpha  or  calreticulin/CD91,  lung   collectins  act  as  dual  function  surveillance  molecules  to  suppress  or  enhance  inflammation.   Cell  2003;115:13–23.   101.  Fernandez  S,  Jose  P,  Avdiushko  MG,  et  al.  Inhibition  of  IL-­‐10  receptor  function  in   alveolar  macrophages  by  Toll-­‐like  receptor  agonists.  J  Immunol  2004;172:2613–20.   102.  Serezani  CH,  Chung  J,  Ballinger  MN,  et  al.  Prostaglandin  E2  suppresses  bacterial  killing   in  alveolar  macrophages  by  inhibiting  NADPH  oxidase.  Am  J  Respir  Cell  Mol  Biol   2007;37:562–70.   103.  Westphalen  K,  Gusarova  GA,  Islam  MN,  et  al.  Sessile  alveolar  macrophages   communicate  with  alveolar  epithelium  to  modulate  immunity.  Nature  2014;506:503–6.   104.  Guth  AM,  Janssen  WJ,  Bosio  CM,  et  al.  Lung  environment  determines  unique  phenotype   of  alveolar  macrophages.  Am  J  Physiol  Lung  Cell  Mol  Physiol  2009;296:936–946.   105.  Benoit  M,  Desnues  B,  Mege  J-­‐L.  Macrophage  polarization  in  bacterial  infections.  J   Immunol  2008;181:3733–9.   106.  Mosser  DM,  Edwards  JP.  Exploring  the  full  spectrum  of  macrophage  activation.  Nat   Rev  Immunol  2008;8:958–69.   107.  Heinsbroek  SEM,  Gordon  S.  The  role  of  macrophages  in  inflammatory  bowel  diseases.   Expert  Rev  Mol  Med  2009;11:e14.   108.  Laskin  DL,  Sunil  VR,  Gardner  CR,  et  al.  Macrophages  and  tissue  injury:  agents  of   defense  or  destruction?  Annu  Rev  Pharmacol  Toxicol  2011;51:267–88.   109.  Vega  MA,  Corbí  AL.  Human  macrophage  activation :  Too  many  functions  and   phenotypes  for  a  single  cell  type.  Immunologia  2006;25:248–272.   110.  Edwards  JP,  Zhang  X,  Frauwirth  KA,  et  al.  Biochemical  and  functional  characterization   of  three  activated  macrophage  populations.  J  Leukoc  Biol  2006;80:1298–1307.   111.  Loke  P,  Gallagher  I,  Nair  MG,  et  al.  Alternative  Activation  Is  an  Innate  Response  to   Injury  That  Requires  CD4+  T  Cells  to  be  Sustained  during  Chronic  Infection.  J  Immunol   2007;179:3926–3936.   112.  Stein  BM,  Keshav  S,  Harris  N,  et  al.  Interleukin  4  Potently  Enhances  Murine   Macrophage  Mannose  Receptor  Activity:  A  Marker  of  Alternative  Immunologic  Macrophage   Activation.  J  Exp  Med  1992;176:287–292.   110   113.  Willment  JA,  Lin  H-­‐H,  Reid  DM,  et  al.  Dectin-­‐1  Expression  and  Function  Are  Enhanced   on  Alternatively  Activated  and  GM-­‐CSF-­‐Treated  Macrophages  and  Are  Negatively  Regulated   by  IL-­‐10,  Dexamethasone,  and  Lipopolysaccharide.  J  Immunol  2003;171:6297–6297.   114.  Heinsbroek  SEM,  Taylor  PR,  Martinez  FO,  et  al.  Stage-­‐specific  sampling  by  pattern   recognition  receptors  during  Candida  albicans  phagocytosis.  PLoS  Pathog   2008;4:e1000218.   115.  Gazi  U,  Martinez-­‐Pomares  L.  Influence  of  the  mannose  receptor  in  host  immune   responses.  Immunobiology  2009;214:554–61.   116.  Vautier  S,  MacCallum  DM,  Brown  GD.  C-­‐type  lectin  receptors  and  cytokines  in  fungal   immunity.  Cytokine  2012;58:89–99.   117.  Taylor  PR,  Gordon  S,  Martinez-­‐Pomares  L.  The  mannose  receptor:  linking  homeostasis   and  immunity  through  sugar  recognition.  Trends  Immunol  2005;26:104–10.   118.  Shepherd  VL,  Hoidal  JR.  Clearance  of  neutrophil-­‐derived  myeloperoxidase  by  the   macrophage  mannose  receptor.  Am  J  Respir  Cell  Mol  Biol  1990;2:335–40.   119.  Madsen  DH,  Leonard  D,  Masedunskas  A,  et  al.  M2-­‐like  macrophages  are  responsible   for  collagen  degradation  through  a  mannose  receptor-­‐mediated  pathway.  J  Cell  Biol   2013;202:951–66.   120.  Marakalala  MJ,  Kerrigan  AM,  Brown  GD.  Dectin-­‐1:  a  role  in  antifungal  defense  and   consequences  of  genetic  polymorphisms  in  humans.  Mamm  Genome  2011;22:55–65.   121.  Gessner  MA,  Werner  JL,  Lilly  LM,  et  al.  Dectin-­‐1-­‐dependent  interleukin-­‐22  contributes   to  early  innate  lung  defense  against  Aspergillus  fumigatus.  Infect  Immun  2012;80:410–7.   122.  Hohl  TM,  Van  Epps  HL,  Rivera  A,  et  al.  Aspergillus  fumigatus  triggers  inflammatory   responses  by  stage-­‐specific  beta-­‐glucan  display.  PLoS  Pathog  2005;1:e30.   123.  Plato  A,  Willment  JA,  Brown  GD.  C-­‐type  lectin-­‐like  receptors  of  the  dectin-­‐1  cluster:   ligands  and  signaling  pathways.  Int  Rev  Immunol  2013;32:134–56.   124.  Kankkunen  P,  Teirilä  L,  Rintahaka  J,  et  al.  (1,3)-­‐beta-­‐glucans  activate  both  dectin-­‐1  and   NLRP3  inflammasome  in  human  macrophages.  J  Immunol  2010;184:6335–42.   125.  Shuhui  L,  Mok  Y-­‐K,  Wong  WSF.  Role  of  mammalian  chitinases  in  asthma.  Int  Arch   Allergy  Immunol  2009;149:369–77.   126.  Martinez  FO,  Helming  L,  Gordon  S.  Alternative  activation  of  macrophages:  an   immunologic  functional  perspective.  Annu  Rev  Immunol  2009;27:451–83.   111   127.  Chung  MJ,  Liu  T,  Ullenbruch  M,  et  al.  Antiapoptotic  effect  of  found  in  inflammatory   zone  (FIZZ)1  on  mouse  lung  fibroblasts.  J  Pathol  2007;212:180–187.   128.  Akdis  CA,  Blaser  K.  Mechanisms  of  interleukin-­‐10-­‐mediated  immune  suppression.   Immunology  2001;103:131–6.   129.  Chung  F.  Anti-­‐inflammatory  cytokines  in  asthma  and  allergy:  interleukin-­‐10,   interleukin-­‐12,  interferon-­‐gamma.  Mediators  Inflamm  2001;10:51–9.   130.  Lang  R,  Patel  D,  Morris  JJ,  et  al.  Shaping  Gene  Expression  in  Activated  and  Resting   Primary  Macrophages  by  IL-­‐10.  J  Immunol  2002;169:2253–2263.   131.  Murray  PJ.  The  primary  mechanism  of  the  IL-­‐10-­‐regulated  antiinflammatory  response   is  to  selectively  inhibit  transcription.  Proc  Natl  Acad  Sci  U  S  A  2005;102:8686–91.   132.  Finnson  KW,  Arany  PR,  Philip  A.  Transforming  Growth  Factor  Beta  Signaling  in   Cutaneous  Wound  Healing:  Lessons  Learned  from  Animal  Studies.  Adv  wound  care   2013;2:225–237.   133.  Li  MO,  Wan  YY,  Sanjabi  S,  et  al.  Transforming  growth  factor-­‐beta  regulation  of  immune   responses.  Annu  Rev  Immunol  2006;24:99–146.   134.  Riquelme  P,  Tomiuk  S,  Kammler  A,  et  al.  IFN-­‐γ-­‐induced  iNOS  expression  in  mouse   regulatory  macrophages  prolongs  allograft  survival  in  fully  immunocompetent  recipients.   Mol  Ther  2013;21:409–22.   135.  Ferracini  M,  Rios  FJO,  Pecenin  M,  et  al.  Clearance  of  apoptotic  cells  by  macrophages   induces  regulatory  phenotype  and  involves  stimulation  of  CD36  and  platelet-­‐activating   factor  receptor.  Mediators  Inflamm  2013;2013:950273.   136.  Frankenberger  M,  Häussinger  K,  Ziegler-­‐Heitbrock  L.  Liposomal  methylprednisolone   differentially  regulates  the  expression  of  TNF  and  IL-­‐10  in  human  alveolar  macrophages.   Int  Immunopharmacol  2005;5:289–99.   137.  Ferrante  CJ,  Pinhal-­‐Enfield  G,  Elson  G,  et  al.  The  adenosine-­‐dependent  angiogenic   switch  of  macrophages  to  an  M2-­‐like  phenotype  is  independent  of  interleukin-­‐4  receptor   alpha  (IL-­‐4Rα)  signaling.  Inflammation  2013;36:921–31.   138.  Cohen  HB,  Mosser  DM.  Extrinsic  and  intrinsic  control  of  macrophage  inflammatory   responses.  J  Leukoc  Biol  2013;94:913–9.   139.  Fleming  BD,  Mosser  DM.  Regulatory  macrophages:  setting  the  threshold  for  therapy.   Eur  J  Immunol  2011;41:2498–502.   112     140.  Hutchinson  J  a,  Riquelme  P,  Sawitzki  B,  et  al.  Cutting  Edge:  Immunological   consequences  and  trafficking  of  human  regulatory  macrophages  administered  to  renal   transplant  recipients.  J  Immunol  2011;187:2072–8.   141.  Jang  JC,  Nair  MG.  Alternatively  Activated  Macrophages  Revisited:  New  Insignts  into  the   Regulation  of  Immunity,  Inflammation  and  Metabolic  Function  following  Parasite  Infection.   Curr  Immunol  Rev  2013;9:147–156.   142.  Stout  RD,  Jiang  C,  Matta  B,  et  al.  Macrophages  Sequentially  Change  Their  Functional   Phenotype  in  Response  to  Changes  in  Microenvironmental  Influences.  J  Immunol   2005;175:342–349.   143.  Davis  MJ,  Tsang  TM,  Qiu  Y,  et  al.  Macrophage  M1  /  M2  Polarization  Dynamically   Adapts  to  Changes  in  Cytokine  Microenvironments  in  Cryptococcus  neoformans  Infection.   2013;4:1–10.   144.  Arora  S,  Olszewski  MA,  Tsang  TM,  et  al.  Effect  of  cytokine  interplay  on  macrophage   polarization  during  chronic  pulmonary  infection  with  Cryptococcus  neoformans.  Infect   Immun  2011;79:1915–26.   145.  Stout  RD.  Editorial:  macrophage  functional  phenotypes:  no  alternatives  in  dermal   wound  healing?  J  Leukoc  Biol  2010;87:19–21.   146.  Daley  JM,  Brancato  SK,  Thomay  AA,  et  al.  The  phenotype  of  murine  wound   macrophages.  J  Leukoc  Biol  2009;87:59–67.   147.  Raes  G,  Baetselier  P  De,  Ghassabeh  GH,  et  al.  Arginase-­‐1  and  Ym1  Are  Markers  for   Murine,  but  Not  Human,  Alternatively  Activated  Myeloid  Cells.  J  Immunol  2012;174:6–9.   148.  Létuvé  S,  Kozhich  A,  Humbles  A,  et  al.  Lung  chitinolytic  activity  and  chitotriosidase  are   elevated  in  chronic  obstructive  pulmonary  disease  and  contribute  to  lung  inflammation.   Am  J  Pathol  2010;176:638–49.   149.  Letuve  S,  Kozhich  A,  Arouche  N,  et  al.  YKL-­‐40  Is  Elevated  in  Patients  with  Chronic   Obstructive  Pulmonary  Disease  and  Activates  Alveolar  Macrophages.  J  Immunol   2008;181:5167–5173.   150.  Staples  KJ,  Hinks  TSC,  Ward  JA,  et  al.  Phenotypic  characterization  of  lung  macrophages   in  asthmatic  patients:  overexpression  of  CCL17.  J  Allergy  Clin  Immunol  2012;130:1404– 1412.   151.  Krausgruber  T,  Blazek  K,  Smallie  T,  et  al.  IRF5  promotes  inflammatory  macrophage   polarization  and  TH1-­‐TH17  responses.  Nat  Immunol  2011;12:231–8.   152.  El  Chartouni  C,  Rehli  M.  Comprehensive  analysis  of  TLR4-­‐induced  transcriptional   responses  in  interleukin  4-­‐primed  mouse  macrophages.  Immunobiology  2010;215:780–7.     113   153.  Weisser  SB,  McLarren  KW,  Voglmaier  N,  et  al.  Alternative  activation  of  macrophages   by  IL-­‐4  requires  SHIP  degradation.  Eur  J  Immunol  2011;41:1742–53.   154.  Wilson  HM.  SOCS  Proteins  in  Macrophage  Polarization  and  Function.  Front  Immunol   2014;5:1–5.   155.  Whyte  CS,  Bishop  ET,  Rückerl  D,  et  al.  Suppressor  of  cytokine  signaling  (SOCS)1  is  a   key  determinant  of  differential  macrophage  activation  and  function.  J  Leukoc  Biol   2011;90:845–54.   156.  Qasimi  P,  Ming-­‐Lum  A,  Ghanipour  A,  et  al.  Divergent  mechanisms  utilized  by  SOCS3  to   mediate  interleukin-­‐10  inhibition  of  tumor  necrosis  factor  alpha  and  nitric  oxide   production  by  macrophages.  J  Biol  Chem  2006;281:6316–24.   157.  Rauh  MJ,  Ho  V,  Pereira  C,  et  al.  SHIP  represses  the  generation  of  alternatively  activated   macrophages.  Immunity  2005;23:361–74.   158.  Murray  PJ,  Wynn  TA.  Obstacles  and  opportunities  for  understanding  macrophage   polarization.  J  Leukoc  Biol  2011;89:557–63.   159.  Ishii  M,  Wen  H,  Corsa  C  a  S,  et  al.  Epigenetic  regulation  of  the  alternatively  activated   macrophage  phenotype.  Blood  2009;114:3244–54.   160.  Abramson  SL,  Gallin  JI.  IL-­‐4  inhibits  superoxide  procution  by  human  mononuclear   phagocytes.  J  Immunol  1990;144:625–630.   161.  Jackson  KA,  Stott  JL,  Horohov  DW,  et  al.  IL-­‐4  induced  CD23  (FcepsilonRII)  up-­‐ regulation  in  equine  peripheral  blood  mononuclear  cells  and  pulmonary  alveolar   macrophages.  Vet  Immunol  Immunopathol  2004;101:243–50.   162.  Bonder  CS,  Davies  KVL,  Hosszu  EK,  et  al.  IFN-­‐gamma  downregulates  interleukin-­‐4   functional  activity  on  monocytes  by  multiple  mechanisms.  J  Interferon  Cytokine  Res   2002;22:287–93.   163.  Varin  A,  Mukhopadhyay  S,  Herbein  G,  et  al.  Alternative  activation  of  macrophages  by   IL-­‐4  impairs  phagocytosis  of  pathogens  but  potentiates  microbial-­‐induced  signalling  and   cytokine  secretion.  2010;115:353–362.   164.  Finkelman  FD,  Hogan  SP,  Hershey  GKK,  et  al.  Importance  of  cytokines  in  murine   allergic  airway  disease  and  human  asthma.  J  Immunol  2010;184:1663–74.   165.  Moreira  AP,  Hogaboam  CM.  Macrophages  in  allergic  asthma:  fine-­‐tuning  their  pro-­‐  and   anti-­‐inflammatory  actions  for  disease  resolution.  J  Interferon  Cytokine  Res  2011;31:485– 91. 114   166.  Welch  JS,  Escoubet-­‐Lozach  L,  Sykes  DB,  et  al.  TH2  cytokines  and  allergic  challenge   induce  Ym1  expression  in  macrophages  by  a  STAT6-­‐dependent  mechanism.  J  Biol  Chem   2002;277:42821–9.   167.  Ford  AQ,  Dasgupta  P,  Mikhailenko  I,  et  al.  Adoptive  transfer  of  IL-­‐4Rα+  macrophages  is   sufficient  to  enhance  eosinophilic  inflammation  in  a  mouse  model  of  allergic  lung   inflammation.  BMC  Immunol  2012;13:6.   168.  Byers  DE,  Holtzman  MJ.  Alternatively  activated  macrophages  and  airway  disease.   Chest  2011;140:768–74.   169.  Balhara  J,  Gounni  AS.  The  alveolar  macrophages  in  asthma:  a  double-­‐edged  sword.   Mucosal  Immunol  2012;5:605–9.   170.  Moreira  AP,  Cavassani  KA,  Hullinger  R,  et  al.  Serum  amyloid  P  attenuates  M2   macrophage  activation  and  protects  against  fungal  spore-­‐induced  allergic  airway  disease.  J   Allergy  Clin  Immunol  2010;126:712–721.e7.   171.  Herbert  C,  Scott  MM,  Scruton  KH,  et  al.  Alveolar  macrophages  stimulate  enhanced   cytokine  production  by  pulmonary  CD4+  T-­‐lymphocytes  in  an  exacerbation  of  murine   chronic  asthma.  Am  J  Pathol  2010;177:1657–64.   172.  Kim  EY,  Battaile  JT,  Patel  AC,  et  al.  Persistent  activation  of  an  innate  immune  response   translates  respiratory  viral  infection  into  chronic  lung  disease.  Nat  Med  2008;14:633–40.   173.  Préfontaine  D,  Lajoie-­‐Kadoch  S,  Foley  S,  et  al.  Increased  expression  of  IL-­‐33  in  severe   asthma:  evidence  of  expression  by  airway  smooth  muscle  cells.  J  Immunol  2009;183:5094– 103.   174.  Kurowska-­‐Stolarska  M,  Stolarski  B,  Kewin  P,  et  al.  IL-­‐33  amplifies  the  polarization  of   alternatively  activated  macrophages  that  contribute  to  airway  inflammation.  J  Immunol   2009;183:6469–77.   175.  Alber  A,  Howie  SEM,  Wallace  WAH,  et  al.  The  role  of  macrophages  in  healing  the   wounded  lung.  Int  J  Exp  Pathol  2012;93:243–51.   176.  Zhong  B,  Yang  X,  Sun  Q,  et  al.  Pdcd4  modulates  markers  of  macrophage  alternative   activation  and  airway  remodeling  in  antigen-­‐induced  pulmonary  inflammation.  J  Leukoc   Biol  2014;96:1–11.   177.  Woodruff  PG,  Modrek  B,  Choy  DF,  et  al.  T-­‐helper  type  2-­‐driven  inflammation  defines   major  subphenotypes  of  asthma.  Am  J  Respir  Crit  Care  Med  2009;180:388–95.   178.  Pappas  K,  Papaioannou  AI,  Kostikas  K,  et  al.  The  role  of  macrophages  in  obstructive   airways  disease:  chronic  obstructive  pulmonary  disease  and  asthma.  Cytokine   2013;64:613–25.   115   179.  Bogaert  P,  Naessens  T,  Koker  S  De,  et  al.  Inflammatory  signatures  for  eosinophilic  vs  .   neutrophilic  allergic  pulmonary  inflammation  reveal  critical  regulatory  checkpoints.  Am  J   Physiol  Lung  Cell  Mol  Physiol  2011;300:679–690.   180.  Tang  C,  Inman  MD,  van  Rooijen  N,  et  al.  Th  Type  1-­‐Stimulating  Activity  of  Lung   Macrophages  Inhibits  Th2-­‐Mediated  Allergic  Airway  Inflammation  by  an  IFN-­‐γ-­‐Dependent   Mechanism.  J  Immunol  2001;166:1471–1481.   181.  Careau  E,  Turmel  V,  Lauzon-­‐Joset  J-­‐F,  et  al.  Alveolar  macrophages  reduce  airway   hyperresponsiveness  and  modulate  cytokine  levels.  Exp  Lung  Res  2010;36:255–61.   182.  Sirois  J,  Bissonnette  EY.  Alveolar  macrophages  of  allergic  resistant  and  susceptible   strains  of  rats  show  distinct  cytokine  profiles.  Clin  Exp  Immunol  2001;126:9–15.   183.  Melgert  BN,  ten  Hacken  NH,  Rutgers  B,  et  al.  More  alternative  activation  of   macrophages  in  lungs  of  asthmatic  patients.  J  Allergy  Clin  Immunol  2011;127:831–3.   184.  Tang  C,  Rolland  JM,  Ward  C,  et  al.  Differential  regulation  of  allergen-­‐specific  T(H2)-­‐  but   not  T(H1)-­‐type  responses  by  alveolar  macrophages  in  atopic  asthma.  J  Allergy  Clin   Immunol  1998;102:368–75.   185.  Prieto  J,  Lensmar  C,  Roquet  A,  et  al.  Increased  interleukin-­‐13  mRNA  expression  in   bronchoalveolar  lavage  cells  of  atopic  patients  with  mild  asthma  after  repeated  low-­‐dose   allergen  provocations.  Respir  Med  2000;94:806–14.   186.  Lay  JC,  Alexis  NE,  Zeman  KL,  et  al.  In  vivo  uptake  of  inhaled  particles  by  airway   phagocytes  is  enhanced  in  patients  with  mild  asthma  compared  with  normal  volunteers.   Thorax  2009;64:313–20.   187.  Lensmar  C,  Katchar  K,  Eklund  A,  et  al.  Phenotypic  analysis  of  alveolar  macrophages   and  lymphocytes  following  allergen  inhalation  by  atopic  subjects  with  mild  asthma.  Respir   Med  2006;100:918–25.   188.  Goleva  E,  Hauk  PJ,  Hall  CF,  et  al.  Corticosteroid-­‐resistant  asthma  is  associated  with   classical  antimicrobial  activation  of  airway  macrophages.  J  Allergy  Clin  Immunol   2008;122:550–559.   189.  Woodruff  PG,  Koth  LL,  Yang  YH,  et  al.  A  distinctive  alveolar  macrophage  activation   state  induced  by  cigarette  smoking.  Am  J  Respir  Crit  Care  Med  2005;172:1383–92.   190.  Gordon  S,  Martinez  FO.  Alternative  activation  of  macrophages:  mechanism  and   functions.  Immunity  2010;32:593–604.   191.  Hardison  SE,  Ravi  S,  Wozniak  KL,  et  al.  Pulmonary  infection  with  an  interferon-­‐ gamma-­‐producing  Cryptococcus  neoformans  strain  results  in  classical  macrophage   activation  and  protection.  Am  J  Pathol  2010;176:774–85.   116   192.  Flynn  RJ,  Mulcahy  G.  Possible  role  for  Toll-­‐like  receptors  in  interaction  of  Fasciola   hepatica  excretory/secretory  products  with  bovine  macrophages.  Infect  Immun   2008;76:678–84.   193.  Poole  JA,  Alexis  NE,  Parks  C,  et  al.  Repetitive  organic  dust  exposure  in  vitro  impairs   macrophage  differentiation  and  function.  J  Allergy  Clin  Immunol  2008;122:375–82,.   194.  Holcombe  SJ,  Jackson  C,  Gerber  V,  et  al.  Stabling  is  associated  with  airway   inflammation  in  young  Arabian  horses.  Equine  Vet  J  2001;33:244–9.   195.  Fleige  S,  Pfaffl  MW.  RNA  integrity  and  the  effect  on  the  real-­‐time  qRT-­‐PCR   performance.  Mol  Aspects  Med  2006;27:126–39.   196.  Raes  G,  Noël  W,  Beschin  A,  et  al.  FIZZ1  and  Ym  as  Tools  to  Discriminate  between   Differentially  Activated  Macrophages.  Dev  Immunol  2002;9:151–159.   197.  Karagianni  AE,  Kapetanovic  R,  McGorum  BC,  et  al.  The  equine  alveolar  macrophage:   Functional  and  phenotypic  comparisons  with  peritoneal  macrophages.  Vet  Immunol   Immunopathol  2013;155:219–28.   198.  Warnken  M,  Haag  S,  Matthiesen  S,  et  al.  Species  differences  in  expression  pattern  of   arginase  isoenzymes  and  differential  effects  of  arginase  inhibition  on  collagen  synthesis  in   human  and  rat  pulmonary  fibroblasts.  Naunyn  Schmiedebergs  Arch  Pharmacol   2010;381:297–304.   199.  Ming  X-­‐F,  Rajapakse  AG,  Yepuri  G,  et  al.  Arginase  II  Promotes  Macrophage   Inflammatory  Responses  Through  Mitochondrial  Reactive  Oxygen  Species,  Contributing  to   Insulin  Resistance  and  Atherogenesis.  J  Am  Heart  Assoc  2012;1:e000992.   200.  Scheller  J,  Chalaris  A,  Schmidt-­‐Arras  D,  et  al.  The  pro-­‐  and  anti-­‐inflammatory   properties  of  the  cytokine  interleukin-­‐6.  Biochim  Biophys  Acta  2011;1813:878–88.   201.  Tremblay  GM,  Ferland  C,  Lapointet  J,  et  al.  Effect  of  stabling  on  bronchoalveolar  cells   obtained  from  normal  and  COPD  horses.  1993;25:194–197.   202.  Pirie  RS,  McLachlan  G,  McGorum  BC.  Evaluation  of  nebulised  hay  dust  suspensions   (HDS)  for  the  diagnosis  and  investigation  of  heaves.  1:  Preparation  and  composition  of   HDS.  Equine  Vet  J  2010;34:332–336.   203.  Pirie  RS.  Recurrent  airway  obstruction:  a  review.  Equine  Vet  J  2014;46:276–88.   204.  Yang  M,  Kumar  RK,  Hansbro  PM,  et  al.  Emerging  roles  of  pulmonary  macrophages  in   driving  the  development  of  severe  asthma.  J  Leukoc  Biol  2012;91:557–69.   205.  Boorsma  CE,  Draijer  C,  Melgert  BN.  Macrophage  heterogeneity  in  respiratory  diseases.   Mediators  Inflamm  2013;2013:769214.   117     206.  Behan  AL,  Hauptman  JG,  Robinson  NE.  Telemetric  analysis  of  breathing  pattern   variability  in  reccurent  airway  obstruction  (heaves  )–  affected  horses.  Am  J  Vet  Res   2013;74:925–933.   207.  Mantovani  A,  Sica  A,  Locati  M.  Macrophage  polarization  comes  of  age.  Immunity   2005;23:344–6.   208.  Gerber  JS,  Mosser  DM.  Reversing  Lipopolysaccharide  Toxicity  by  Ligating  the   Macrophage  Fc    Receptors.  J  Immunol  2001;166:6861–6868.   209.  Porta  C,  Rimoldi  M,  Raes  G,  et  al.  Tolerance  and  M2  (alternative)  macrophage   polarization  are  related  processes  orchestrated  by  p50  nuclear  factor  kappaB.  Proc  Natl   Acad  Sci  U  S  A  2009;106:14978–83.   210.  Deaton  CM,  Marlin  DJ,  Smith  NC,  et  al.  Pulmonary  epithelial  lining  fluid  and  plasma   ascorbic  acid  concentrations  in  horses  affected  by  recurrent  airway  obstruction.  Am  J  Vet   Res  2004;65:80–7.   211.  Leclere  M,  Lavoie-­‐Lamoureux  A,  Joubert  P,  et  al.  Corticosteroids  and  antigen  avoidance   decrease  airway  smooth  muscle  mass  in  an  equine  asthma  model.  Am  J  Respir  Cell  Mol  Biol   2012;47:589–96.   212.  Wagner  B,  Ainsworth  DM,  Freer  H.  Analysis  of  soluble  CD14  and  its  use  as  a  biomarker   in  neonatal  foals  with  septicemia  and  horses  with  recurrent  airway  obstruction.  Vet   Immunol  Immunopathol  2013;155:124–8.   213.  Prasse  A,  Germann  M,  Pechkovsky  D  V,  et  al.  IL-­‐10-­‐producing  monocytes  differentiate   to  alternatively  activated  macrophages  and  are  increased  in  atopic  patients.  J  Allergy  Clin   Immunol  2007;119:464–71.   214.  Messenger  II-­‐,  Expression  RNA,  Robinson  DS,  et  al.  Increased  Interleukin-­‐10   Messenger  RNA  Expression  in  Atopic  Allergy  and  Asthma.  Am  J  Respir  Cell  Mol  Biol   1996;14:113–117.   215.  Zeibecoglou  K,  Ying  S,  Meng  Q,  et  al.  Macrophage  subpopulations  and  macrophage-­‐ derived  cytokines  in  sputum  of  atopic  and  nonatopic  asthmatic  subjects  and  atopic  and   normal  control  subjects.  J  Allergy  Clin  Immunol  2000;106:697–704.   216.  Colavita  AM,  Hastie  AT,  Musani  AI,  et  al.  Kinetics  of  IL-­‐10  production  after  segmental   antigen  challenge  of  atopic  asthmatic  subjects.  J  Allergy  Clin  Immunol  2000;106:880–6.   217.  Tournoy  KG,  Kips  JC,  Pauwels  RA.  Endogenous  interleukin-­‐10  suppresses  allergen-­‐ induced  airway  inflammation  and  nonspecific  airway  responsiveness.  Clin  Exp  Allergy   2000;30:775–83.     118     218.  Zuany-­‐Amorim  C,  Hailé  S,  Leduc  D,  et  al.  Interleukin-­‐10  inhibits  antigen-­‐induced   cellular  recruitment  into  the  airways  of  sensitized  mice.  J  Clin  Invest  1995;95:2644–51.   219.  Barbarin  V,  Xing  Z,  Delos  M,  et  al.  Pulmonary  overexpression  of  IL-­‐10  augments  lung   fibrosis  and  Th2  responses  induced  by  silica  particles.  Am  J  Physiol  Lung  Cell  Mol  Physiol   2005;288:841–848.   220.  Mäkelä  MJ,  Kanehiro  a,  Borish  L,  et  al.  IL-­‐10  is  necessary  for  the  expression  of  airway   hyperresponsiveness  but  not  pulmonary  inflammation  after  allergic  sensitization.  Proc  Natl   Acad  Sci  U  S  A  2000;97:6007–12.   221.  Lee  CG,  Homer  RJ,  Cohn  L,  et  al.  Transgenic  overexpression  of  interleukin  (IL)-­‐10  in   the  lung  causes  mucus  metaplasia,  tissue  inflammation,  and  airway  remodeling  via  IL-­‐13-­‐ dependent  and  -­‐independent  pathways.  J  Biol  Chem  2002;277:35466–74.   222.  Florentino  DF,  Zlotnik  A,  Mosmann  TR,  et  al.  IL-­‐10  Inhibits  Cytokine  Production  by   Activated  Macrophages.  J  Immunol  1991;147:815–3822.   223.  Naessens  T,  Vander  Beken  S,  Bogaert  P,  et  al.  Innate  imprinting  of  murine  resident   alveolar  macrophages  by  allergic  bronchial  inflammation  causes  a  switch  from   hypoinflammatory  to  hyperinflammatory  reactivity.  Am  J  Pathol  2012;181:174–84.   224.  Espinassous  Q,  Garcia-­‐de-­‐Paco  E,  Garcia-­‐Verdugo  I,  et  al.  IL-­‐33  enhances   lipopolysaccharide-­‐induced  inflammatory  cytokine  production  from  mouse  macrophages   by  regulating  lipopolysaccharide  receptor  complex.  J  Immunol  2009;183:1446–55.   225.  Lloyd  CM.  IL-­‐33  family  members  and  asthma  -­‐  bridging  innate  and  adaptive  immune   responses.  Curr  Opin  Immunol  2010;22:800–6.   226.  Zhang  L,  Franchini  M,  Wehrli  Eser  M,  et  al.  Increased  adenosine  concentration  in   bronchoalveolar  lavage  fluid  of  horses  with  lower  airway  inflammation.  Vet  J   2012;193:268–70.   227.  Sun  W,  Moore  JN,  Hurley  DJ,  et  al.  Differential  modulation  of  lipopolysaccharide-­‐ induced  expression  of  inflammatory  genes  in  equine  monocytes  through  activation  of   adenosine  A2A  receptors.  Vet  Immunol  Immunopathol  2010;134:169–77.   228.  Sun  L,  Guo  R-­‐F,  Newstead  MW,  et  al.  Effect  of  IL-­‐10  on  neutrophil  recruitment  and   survival  after  Pseudomonas  aeruginosa  challenge.  Am  J  Respir  Cell  Mol  Biol  2009;41:76– 84.   229.  Andrade  EB,  Alves  J,  Madureira  P,  et  al.  TLR2-­‐induced  IL-­‐10  production  impairs   neutrophil  recruitment  to  infected  tissues  during  neonatal  bacterial  sepsis.  J  Immunol   2013;191:4759–68.     119     230.  Hammond  SA,  Horohov  D,  Montelaro  RC.  Functional  characterization  of  equine   dendritic  cells  propagated  ex  vivo  using  recombinant  human  GM-­‐CSF  and  recombinant   equine  IL-­‐4.  Vet  Immunol  Immunopathol  1999;71:197–214.   231.  Nair  MG,  Cochrane  DW,  Allen  JE.  Macrophages  in  chronic  type  2  inflammation  have  a   novel  phenotype  characterized  by  the  abundant  expression  of  Ym1  and  Fizz1  that  can  be   partly  replicated  in  vitro.  Immunol  Lett  2003;85:173–80.   232.  Royer  P-­‐J,  Emara  M,  Yang  C,  et  al.  The  mannose  receptor  mediates  the  uptake  of   diverse  native  allergens  by  dendritic  cells  and  determines  allergen-­‐induced  T  cell   polarization  through  modulation  of  IDO  activity.  J  Immunol  2010;185:1522–31.   233.  Gerber  V,  Robinson  NE,  Luethi  S,  et  al.  Airway  inflammation  and  mucus  in  two  age   groups  of  asymptomatic  well-­‐performing  sport  horses.  Equine  Vet  J  2003;35:491–5.   234.  Li  J,  Pritchard  DK,  Wang  X,  et  al.  cDNA  microarray  analysis  reveals  fundamental   differences  in  the  expression  profiles  of  primary  human  monocytes,  monocyte-­‐derived   macrophages,  and  alveolar  macrophages.  J  Leukoc  Biol  2007;81:328–35.   235.  Gundra  UM,  Girgis  NM,  Ruckerl  D,  et  al.  Alternatively  activated  macrophages  derived   from  monocytes  and  tissue  macrophages  are  phenotypically  and  functionally  distinct.   2014;123:110–122.   236.  Keel  M,  Ungethüm  U,  Steckholzer  U,  et  al.  Interleukin-­‐10  counterregulates   proinflammatory  cytokine-­‐induced  inhibition  of  neutrophil  apoptosis  during  severe  sepsis.   Blood  1997;90:3356–63.   237.  Vandesompele  J,  De  Preter  K,  Pattyn  F,  et  al.  Accurate  normalization  of  real-­‐time   quantitative  RT-­‐PCR  data  by  geometric  averaging  of  multiple  internal  control  genes.   Genome  Biol  2002;3:RESEARCH0034.   238.  Pfaffl  MW,  Tichopad  A,  Prgomet  C,  et  al.  Determination  of  stable  housekeeping  genes,   differentially  regulated  target  genes  and  sample  integrity:  BestKeeper-­‐-­‐Excel-­‐based  tool   using  pair-­‐wise  correlations.  Biotechnol  Lett  2004;26:509–15.   239.  Herre  J.  Dectin-­‐1  and  its  role  in  the  recognition  of  β-­‐glucans  by  macrophages.  Mol   Immunol  2004;40:869–876.   240.  Heinsbroek  SEM,  Taylor  PR,  Rosas  M,  et  al.  Expression  of  functionally  different  dectin-­‐ 1  isoforms  by  murine  macrophages.  J  Immunol  2006;176:5513–8.   241.  Willment  JA,  Gordon  S,  Brown  GD.  Characterization  of  the  human  beta  -­‐glucan   receptor  and  its  alternatively  spliced  isoforms.  J  Biol  Chem  2001;276:43818–23.   242.  Sonck  E,  Stuyven  E,  Goddeeris  B,  et  al.  Identification  of  the  porcine  C-­‐type  lectin   dectin-­‐1.  Vet  Immunol  Immunopathol  2009;130:131–4.     120     243.  Willcocks  S,  Yamakawa  Y,  Stalker  A,  et  al.  Identification  and  gene  expression  of  the   bovine  C-­‐type  lectin  Dectin-­‐1.  Vet  Immunol  Immunopathol  2006;113:234–42.   244.  Zhou  H,  Hu  J,  Luo  Y,  et  al.  Variation  in  the  ovine  C-­‐type  lectin  dectin-­‐1  gene  (CLEC7A).   Dev  Comp  Immunol  2010;34:246–9.   245.  Robinson  NE,  Berney  C,  Behan  A,  et  al.  Fluticasone  propionate  aerosol  is  more   effective  for  prevention  than  treatment  of  recurrent  airway  obstruction.  J  Vet  Intern  Med   2009;23:1247–53.   246.  Heinsbroek  SEM,  Taylor  PR,  Rosas  M,  et  al.  Expression  of  functionally  different  dectin-­‐ 1  isoforms  by  murine  macrophages.  J  Immunol  2006;176:5513–8.   247.  Sun  W-­‐K,  Lu  X,  Li  X,  et  al.  Dectin-­‐1  is  inducible  and  plays  a  crucial  role  in  Aspergillus-­‐ induced  innate  immune  responses  in  human  bronchial  epithelial  cells.  Eur  J  Clin  Microbiol   Infect  Dis  2012;31:2755–64.   248.  Editor  KJH.  Spliceosomal  Pre-­‐mRNA  Splicing,  Methods  and  Protocols,  Methods  in   Molcular  Biology  1126.;  2014:45–54.   249.  Lewis  BP,  Green  RE,  Brenner  SE.  Evidence  for  the  widespread  coupling  of  alternative   splicing  and  nonsense-­‐mediated  mRNA  decay  in  humans.  Proc  Natl  Acad  Sci  U  S  A   2003;100:189–92.   250.  Smith  CW,  Valcárcel  J.  Alternative  pre-­‐mRNA  splicing:  the  logic  of  combinatorial   control.  Trends  Biochem  Sci  2000;25:381–8.       121