Double haploid in maize

     Doubled Haploid Technology in Maize Breeding: Theory and Practice               Editors   BM  Prasanna,  Vijay  Chaikam,  and  George  Mahuku                                       Headquartered   in   Mexico,   the   International   Maize   and   Wheat   Improvement   Center   (known   by   its   Spanish   acronym,   CIMMYT)   is   a   not-­‐for-­‐profit   agriculture   research   and   training   organization.   The   center   works   to   reduce  poverty  and  hunger  by  sustainably  increasing  the  productivity  of  maize  and  wheat  in  the  developing   world.  CIMMYT  maintains  the  world’s  largest  maize  and  wheat  seed  bank  and  is  best  known  for  initiating  the   Green   Revolution,   which   saved   millions   of   lives   across   Asia   and   for   which   CIMMYT’s   Dr.   Norman   Borlaug   was   awarded  the  Nobel  Peace  Prize.  CIMMYT  is  a  member  of  the  CGIAR  Consortium  and  receives  support  from   national  governments,  foundations,  development  banks,  and  other  public  and  private  agencies.   ©International   Maize   and   Wheat   Improvement   Center   (CIMMYT)   2012.   All   rights   reserved.   The   designations  employed  in  the  presentation  of  materials  in  this  publication  do  not  imply  the  expression  of  any   opinion   whatsoever   on   the   part   of   CIMMYT   or   its   contributory   organizations   concerning   the   legal   status   of   any   country,   territory,   city,   or   area,   or   of   its   authorities,   or   concerning   the   delimitation   of   its   frontiers   or   boundaries.  The  opinions  expressed  are  those  of  the  author(s),  and  are  not  necessarily  those  of  CIMMYT  or   our  partners.  CIMMYT  encourages  fair  use  of  this  material.  Proper  citation  is  requested.   Correct   citation:   B.M.   Prasanna,   Vijay   Chaikam   and   George   Mahuku   (eds).   2012.   Doubled   Haploid     Technology  in  Maize  Breeding:  Theory  and  Practice.    Mexico,  D.F.:  CIMMYT.     Abstract:  This  manual  is  primarily  intended  for  maize  breeders  in  national  agricultural  research  systems  and   small  and  medium  enterprise  seed  companies  in  developing  countries  who  would  like  to  better  understand   and   utilizes   the   doubled   haploid   (DH)   technology   in   breeding   programs.     It   is   a   compilation   and   consolidation   of   knowledge   accumulated   through   scientific   contributions   of   several   maize   geneticists   and   breeders   worldwide   as   well   as   protocols   successfully   developed   (in   collaboration   with   the   University   of   Hohenheim,   Germany)   and   being   used   by   the   CIMMYT   Global   Maize   Program   in   DH   line   development,   especially   in   Mexico.  An  overview  of  the  utility  and  applications  of  DH  technology  in  maize  breeding  is  presented  first  in   the   manual,   followed   by   chapters   on   in   vivo   maternal   haploid   induction   using   haploid   inducers,   haploid   kernel   detection   using   anthocyanin   markers,   chromosome   doubling   of   haploids,   deriving   DH   seed   from   colchicine-­‐treated   plants,   integrating   molecular   markers   in   DH-­‐based   breeding   pipeline,   DH   in   commercial   maize  breeding,  and  finally,  access  to  tropicalized  haploid  inducers  and  DH  service  on  cost-­‐recovery  basis  to   CIMMYT  partners.     ISBN:  978-­‐607-­‐95844-­‐9-­‐8     AGROVOC   descriptors:   Zea   mays;   plant   breeding;   doubled   haploids;   haploid   induction;   tropicalized   inducers;   agronomic  management;  molecular  markers;  phenotypic  selection;  intellectual  property.     AGRIS  category  codes:  F30  Plant  Genetics  and  Breeding     Dewey  decimal  classification:  633.150575     Printed  in  Mexico.       ii   Acknowledgements     The  authors  of  this  manual  would  like  to  express  their  sincere  thanks  to  a  number  of  people,   institutions,   and   funding   agencies   for   their   great   support   and   inputs   to   the   doubled   haploid   (DH)  research  work  undertaken  at  CIMMYT  over  the  last  five  years.  The  present  manual  would   not  have  been  possible  without  them.  Special  thanks  go  to:     Prof.   Dr.   Albrecht   E.   Melchinger,   Dr   Wolfgang   Schipprack,   and   the   University   of   Hohenheim   (Germany)   DH   team,   including   the   PhD   students   (Dr.   Vanessa   Prigge   and   Aida   Kebede),   for   their   significant   support   in   transferring   the   DH   technology   to   CIMMYT   and  for  optimizing  various  steps  in  the  DH  production  pipeline;     Dr    Marianne  Banziger  for  her  insight,  vision,  and  support  in  forging  the  collaboration  of   CIMMYT  with  University  of  Hohenheim;   Dr   José   Luis   Araus,   who   led   the   initial   collaboration   of   CIMMYT   with   the   University   of   Hohenheim;   Dr   Natalia   Palacios,   Luis   Galicia   and   Miguel   Bojorges   Cortés     for   critical   support   with   the   colchicine   doubling   process,   and   for   providing   training   on   the   safe   handling   of   colchicine;     CIMMYT   colleagues   (especially   Drs   Gary   Atlin,   Daniel   Jeffers,   and   Kevin   Pixley)   for   providing  important  inputs  for  strengthening  the  DH  program;   The  committed  DH  team  in  CIMMYT-­‐Mexico,  namely  Leocadio  Martinez  Hernandez,  Luis   Antonio   López   Rodrígues,   Juana   Roldán   Valencia,   Ana   Mely   Islas   Montes,   Gustavo   Alberto   Martínez   Rodríguez,   Zaira   Ivette   Mata   Carrillo,   Belem   Adriana   Cervantes   Hernández,   Juan   Antonio   Díaz   Ríos,   César   Muñoz   Galindo,   Reymundo   Blancas,   Pablo   Ostria,  Jabed  Bahena  Torrez,  Miguel  Ángel  Díaz,  Marta  Cano,  Germán  Bastián  De  León,   Blanca   Flor   Pineda   González,   Dany   Fajardo   Romero,   Edith   Hernández   Márquez,   Francisco  Hernández  Fajardo,  Emiliano  Reyes  Espinoza,  Leonardo  Juárez  Agustín,  Ignacio   Morales   Guzmán,   Faustino   López   Hernández,   Eduardo   Fernández   Cruz,   Yesenia   López   Vázquez,   Marcela   Santos   Hernández,   Verónica   Román   Álvarez,   and   Belén   Paredes   Hernández;     The   station   managers   of   CIMMYT’s   Agua   Fría   station,   Raymundo   López   and   Ciro   Sánchez,   for   efficiently   handling   the   field   logistics   required   to   manage   the   DH   line   production  pipeline;     Visiting   scientists   –   Alba   Lucía  Arcos   (Colombia),     Thanh   Duc   Nguyen,   Hung   Huu  Nguyen   and  Ha  Thai  (Vietnam)  –  for  providing  useful  feedback  for  improving  some  of  the  aspects   of  DH  line  development;     The   CIMMYT   Intellectual   Property   &   legal   teams   (especially   Rodrigo   Sara   and   Carolina   Roa)  for  valuable  support  in  steering  through  the  IP  process;     All   the   funding   agencies   that   have   generously   supported   DH   research   at   in   CIMMYT,   especially   the   Bill   and   Melinda   Gates   Foundation,   CGIAR   (through   MAIZE   CRP),   SAGARPA,   Howard   G   Buffet   Foundation,   USAID,   and   Vilmorin,   under   various   projects,   including  DTMA,  and  MasAgro-­‐IMIC;  and     The   CIMMYT   Corporate   Communications   Team,   especially   Mike   Listman   and   Miguel   Mellado,  for  the  design  and  layout  of  the  manual.           iii   Contents     Chapter     Author(s)   1   Doubled  haploid  (DH)  technology  in  maize  breeding:   BM  Prasanna   an  overview   1   2   In  vivo  maternal  haploid  induction  in  maize   Vijay  Chaikam   9   3   Design  and  implementation  of  maternal  haploid   induction   Vijay  Chaikam,   George  Mahuku,  &   BM  Prasanna   14   4   Maternal  haploid  detection  using  anthocyanin   markers   Vijay  Chaikam  &   BM  Prasanna   20   5   Chromosome  doubling  of  maternal  haploids   24   6   Putative  DH  seedlings:  from  the  lab  to  the  field   Vijay  Chaikam  &   George  Mahuku     George  Mahuku   7   Integrating  marker-­‐assisted  selection  in  the  DH-­‐ based  breeding  pipeline  for  rapid  development  and   delivery  of  superior  parental  lines  and  cultivars     39   8   DH  in  commercial  maize  breeding:  phenotypic   selections       R  Babu,  Sudha  K   Nair,  BS  Vivek,   Felix  San  Vicente,  &   BM  Prasanna     Daniel  Jeffers  &   George  Mahuku   9   Access  to  tropicalized  haploid  inducers  and  DH   service  to  CIMMYT  partners       iv   BM  Prasanna,   Vijay  Chaikam,   George  Mahuku,  &   Rodrigo  Sara   Page   30   45   48   1. Doubled  Haploid  (DH)  Technology  in  Maize  Breeding:  An  Overview   BM  Prasanna     Introduction   A   “doubled   haploid”   (DH)   is   a   genotype   formed   when   haploid   (n)   cells   successfully   undergo   either   spontaneous   or   artificially   induced   chromosome   doubling.   Chase   (1947,   1951,   1952,   1969)   pioneered   the  studies  on  maize  monoploids  (synonymous  to  haploids,  in  the  case  of  maize)  and  the  use  of  DH  lines   in   breeding.   The   DH   technology   shortens   the   breeding   cycle   significantly   by   rapid   development   of   completely   homozygous   lines   (in   2–3   generations),   instead   of   the   conventional   inbred   line   development   process,   which   takes   at   least   6–8   generations   to   derive   lines   with   ~99%   homozygosity   (Forster   and   Thomas,   2005;   Geiger   and   Gordillo,   2009;  Chang  and  Coe,  2009).     Chase  initially  relied  on  spontaneous   haploid   induction   and   doubling,   which   was   not   quite   conducive   (due   to  very  low  frequency)  to  commercial   application.   The   foundation   for   in   vivo   haploid   induction   using   haploid   inducers   was   laid   when   Coe   (1959)   described   “a   line   of   maize   with   high   haploid   frequency”   of   2.3%,   designated  as  “Stock  6.”  This  genetic   stock   served   as   a   founder   for   an   array  of  inducers  with  higher  haploid   induction   rates   (HIR   =   number   of   kernels   with   haploid   embryo   divided   Figure   1.   Number   of   generations   to   reach   genetic   purity   by   all   kernels   investigated)   through   (homozygosity)   through:   (A)   conventional   inbreeding;   (B)   the   subsequent   efforts   of   maize   doubled  haploid  technology.   geneticists  worldwide.         Although   DH   lines   in   maize   have   been   produced   by   several   institutions   using   either   in   vitro   or   in   vivo   methods,   the   in   vitro   methods   had   very   limited   success   due   to   non-­‐responsiveness   of   many   maize   genotypes,   besides   the   need   to   have   a   good   laboratory   and   skilled   staff.   In   contrast,   in   vivo   haploid   induction-­‐based   DH   line   development   in   maize   is   relatively   easier,   thanks   to   the   efforts   made   by   the   maize   geneticists   in   identifying   “haploid   inducer   genetic   stocks”   (Coe,   1959;   Coe   and   Sarkar,   1964),   further  incorporating  an  anthocyanin  color  marker  in  the  inducer  genetic  backgrounds  to  facilitate  easy   identification  of  haploids  at  both  the  seed  and  seedling  stages  (Nanda  and  Chase,  1966;  Greenblatt  and   Bock,  1967;  Chase,  1969),  and  deriving  new  haploid  inducers  with  higher  HIR.       The   DH   technology   in   maize   breeding,  based   on   in   vivo   haploid   induction,   is   recognized   worldwide   as   an   important   means   for   enhancing   breeding   efficiency.   In   the   last   10-­‐15   years,   the   technology   has   been   well  adapted  by  several  commercial  maize  breeding  programs  in  Europe  (Schmidt,  2003),  North  America   (Seitz,   2005),   and   more   recently   in   China   (Chen   et   al.,   2009),   almost   as   soon   as   haploid   inducer   lines   became  available  for  temperate  environments  (Prigge  and  Melchinger,  2011).  However,  several  of  the   maize   breeding   institutions   in   the   public   sector,   as   well   as   small   and   medium   enterprise   (SME)   seed   1       companies   in   tropical   maize   growing   countries   in   Latin   America,   sub-­‐Saharan   Africa   and   Asia,   have   lagged  behind  (Prasanna  et  al.,  2010;  Kebede  et  al.,  2011).  This  may  be  due  to  several  factors,  including   inadequate  awareness  about  the  DH  technology,  lack  of  access  to  the  tropicalized  haploid  inducers,  or   lack   of   relevant   “know-­‐how”   for   effectively   integrating   DH   in   breeding   programs.   The   purpose   of   this   manual  is,  therefore,  to  introduce  the  theory  and  practice  of  DH  technology  in  maize  breeding.     Why  DH  in  maize  breeding?   The   DH   technology   offers   an   array   of   advantages   in   maize   genetics   and   breeding   (Röber   et   al.,   2005;   Geiger,  2009;  Geiger  and  Gordillo,  2009);  salient  among  these  are  that  it:   (1)     Significantly   shortens   the   breeding   cycle   by   development     of   completely   homozygous   lines   in   two  generations;     (2)     Simplifies  logistics  (Geiger  and  Gordillo,  2009),  including  requiring  less  time,  labor,  and  financial   resources   for   developing   new   breeding   lines;   the   time   and   resources   thus   saved   could   be   potentially  channelized  for  implementing  more  effective  selections  and  for  accelerated  release   of  elite  cultivars;   (3)   Enables   greater   efficiency   and   precision   of   selection   (Röber   et   al.,   2005;   Geiger   and   Gordillo,   2009),  especially  when  used  in  combination  with  molecular  markers  and  year-­‐round  nurseries;     (4)     Accelerates  product  development  by  allowing  rapid  pyramiding  of  favorable  alleles  for  polygenic   traits  influencing  maize  productivity  and  stress  resilience,  which  are  otherwise  difficult  and  time-­‐ consuming  to  combine  in  adapted  germplasm  using  conventional  breeding  practices;     (5)   Perfectly   fulfills   the   requirements   of   DUS   (distinctness,   uniformity,   and   stability)   for   plant   variety   protection   due   to   the   complete   homozygosity   and   homogeneity   of   DH-­‐based   parental   lines   (Geiger  and  Gordillo,  2009);     (6)   Reduces  the  effort  for  line  maintenance  (Röber  et  al.,  2005);   (7)   Can,  in  combination  with  molecular  markers,  facilitate  access  to  the  germplasm  present  within   either  the  female  or  the  male  parental  lines  of  hybrid  cultivars  (Heckenberger  et  al.,  2005);  and   (8)   Provides   opportunities   for   undertaking   marker-­‐trait   association   studies,   marker-­‐based   gene   introgression   (Forster   and   Thomas,   2005),   functional   genomics,   molecular   cytogenetics,   and   genetic  engineering  (Forster  et  al.,  2007;    Wijnker  et  al.,  2007).         In  vivo  maternal  haploid  induction-­‐based  DH  development     Haploid  induction   The   haploid   inducers   are   specialized   genetic   stocks   which,   when   crossed   to   a   diploid   (normal)   maize   plant,  result  in  progeny   kernels  in  an  ear  with  segregation  for  diploid  (2n)  kernels  and  certain  fraction  of   haploid   (n)   kernels   due   to   anomalous   fertilization.   Kernels   with   a   haploid   embryo   have   a   regular   triploid   (3n)   endosperm,   and   therefore,   these   kernels   are   capable   of   displaying   germination   similar   to   those   kernels  with  a  diploid  embryo  (Coe  and  Sarkar,  1964).       The   in   vivo   maternal   haploid   induction   scheme   at   present   relies   on   the   presence   of   a   dominant   anthocyanin  color  marker,  referred  as  R1-­‐Navajo  (R1-­‐nj),  that  expresses  in  the  aleurone  (the  outermost   layer   of   the   maize   endosperm)   as   well   as   in   the   embryo   (scutellum)   in   the   haploid   inducer,   unlike   the   source   populations,   which   do   not   usually   have   any   anothocyanin   coloration   in   the   embryo   or   the   endosperm.   Thus,   R1-­‐nj   as   a   dominant   color   marker   helps   in   differentiation   of   monoploid/haploid   (n)   kernels   (with   no   expression   of   purple/red   colored   anthocynanin   in   the   scutellum,   but   with   the   typical   crown-­‐coloration  on  the  endosperm),   from  the  diploid  (2n)  kernels  (with  expression  of  anthocyanin  in   both  the  endosperm  and   scutellum)   (Nanda  and  Chase,  1966;  Greenblatt   and   Bock,   1967;   Chase,  1969).   Normal  colorless  kernels  are  the  result  of  either  selfing  or  contamination  due  to  outcrossing.  However,  it   must   be   noted   that   the   expression   of   the   R1-­‐nj   color   marker   can   vary   significantly   depending   on   the   2       genetic  background  of  the  source  genotype  (in  which  maternal  haploids  have  to  be  induced),  the  genetic   background   of   the   haploid   inducer,   as   well   as   environmental   factors   (Chase,   1952;   Röber   et   al.,   2005;   Kebede  et  al.,  2011;  Prigge  et  al.,  2011).       Temperate   haploid   inducers:   A  number  of  haploid  inducer  lines  with  high  HIR  and  for  commercial  use   have   been   derived   over   the   years,   with   Stock   6   as   the   founder;   these   include:   (1)   KMS   (Korichnevy   Marker  Saratovsky)  and  ZMS,  both  derived  from  Stock  6  (Tyrnov  and  Zavalishina  1984,  cited  in  Chebotar   and  Chalyk,  1996);  (2)  WS14,  developed  from  a  cross  between  lines  W23ig  and  Stock  6  (Lashermes  and   Beckert,  1988);  (3)  KEMS   (Krasnador  Embryo  Marker  Synthetic),  derived  from  a  cross  (Shatskaya  et  al.,   1994);  (4);  MHI  (Moldovian  Haploid  Inducer),  derived  from  a  cross  KMS  ×  ZMS  (Eder  and  Chalyk,  2002);   (5)   RWS   (Russian   inducer   KEMS   +   WS14),   descendant   of   the   cross   KEMS   ×   WS14   (Röber   et   al.,   2005);   (6)   UH400,   developed   at   University   of   Hohenheim   from   KEMS   (cited   in   Chang   and   Coe,   2009);   (7)   PK6   (Barret  et  al.,  2008);  (8)  HZI1,  derived  from  Stock  6  (Zhang  et  al.,  2008);  (9)  CAUHOI,  derived  at  China   Agricultural  University  from  a  cross  between  Stock  6  and  Beijing  High  Oil  Population  (Li  et  al.,  2009),  and   (10)   PHI   (Procera   Haploid   Inducer),   derived   from   a   cross   between   MHI   and   Stock   6   (Rotarenco   et   al.,   2010).     The   temperate   inducers   UH400,   RWS,   and   RWS   ×   UH400   were   successfully   employed   for   haploid   induction  and  DH  line  development  in  CIMMYT’s  tropical  and  subtropical  source  germplasm  from  2007   to  2011,  although  these  temperate  inducers  are  poorly  adapted  to  tropical  lowland  conditions  (Prigge  et   al.,   2011).   However,   efficient   and   large-­‐scale   production   of   DH   lines   in   tropical   maize-­‐growing   environments   using   temperate   haploid   inducers   could   be   severely   constrained   as   these   inducers   display   poor  vigor,  poor  pollen  production,  poor  seed  set,  and  high  susceptibility  to  tropical  maize  diseases.     Tropicalized  haploid  inducers:  Since  2007,  CIMMYT  Global  Maize  Program  has  been  intensively  engaged   in   optimization   of   the   DH   technology   especially   for   the   tropical/subtropical   maize   growing   environments,   in   partnership   with   the   University   of   Hohenheim,   Germany.   Tropically   adapted   inducer   lines   (TAILs;   with   8–10%   HIR)   have   been   developed   through   this   collaboration   (Prigge   et   al.,   2011).   Experimental  evaluation  of  the  first-­‐generation  TAILs  in  two  environments  (Agua  Fría  and  Tlatizapan  in   Mexico)  over  two  seasons  consistently  resulted  in  average  HIR  ranging  from  9%  to  14%.  A  single-­‐cross   hybrid   haploid   inducer   (with   high   HIR)   has   been   developed   using   a   sub-­‐set   of   TAILs.   The   tropicalized   haploid   inducers   are   now   available   for   sharing   with   interested   institutions   for   research   or   commercial   use   under   specific   terms   and   conditions   (http://www.cimmyt.org/en/about-­‐us/media-­‐resources/recent-­‐ news/1399-­‐now-­‐available-­‐tropicalized-­‐maize-­‐haploid-­‐inducer-­‐lines).  The  availability  of  TAILs  is  expected   to  significantly  enhance  the  efficiency  of  DH  line  production,  increasing  seed  set  and  rates  of  induction,   and  reducing  the  costs  of  inducer  line  maintenance  and  seed  production.     Pathway  for  DH  development  and  scope  for  further  refinement     It  must  be  noted  that  efficient  DH  development  is  dependent  not  only  on  access  to  tropicalized  haploid   inducers  with  high   HIR,  but  also  on  a  number  of  other  important  steps  in  the  DH  production  pipeline.   The   salient   steps   in   DH   development   are:   (1)   crossing   the   source   population   (usually   a   hybrid   generated   using   desired   lines   or   F2   derived   by   selfing   of   the   hybrid)   as   female   parent   with   pollen   of   the   haploid   inducer;   (2)   identification   of   haploid   kernels   (at   the   dry   seed   stage)   using   the   anthocyanin   color   marker;   (3)   germination   of   the   haploid   seeds;   (4)   safe   application   of   colchicine   or   any   other   effective   chromosome   doubling   agent   to   the   haploid   seedlings;   (5)   proper   agronomic   management   of   D0   seedlings   and   derivation   of   D1   (DH)   seed   by   self-­‐pollinating   D0   plants;   and   (6)   further   selection   and   utilization   of   DH   lines   in   breeding   programs.   The   manual,   in   the   subsequent   chapters,   provides   both   theoretical   and   practical   details   for   each   of   the   above   steps.   Some   important   steps   that   are   further   being  refined  through  ongoing  research  in  different  institutions  worldwide  are  highlighted  below.     3       Haploid   identification:   Although   the   R1-­‐nj-­‐based   haploid   identification   scheme   is,   in   general,   quite   effective,  it  is  not  without  a  pitfall.  Presence  of  dominant  anthocyanin  inhibitor  genes  (such  as  C1-­‐I,  C2-­‐ Idf,  and  In1-­‐D)  in  the  source  population  or  donor  genome  (Coe,  1994)  or  dosage  effects  can  sometimes   make   this   marker   scheme   ineffective.   CIMMYT’s   elite   germplasm   is   currently   being   surveyed   to   determine   in   what   proportion   the   seed   color   marker   will   function,   permitting   efficient   haploid   seed   detection.   Currently,   it   appears   that   R1-­‐nj   color   expression   is   inhibited   in   only   about   8%   of   crosses   of   haploid  inducers  with  diverse  source  populations.       The   use   of   haploid   inducers   with   anthocyanin   genes   B1   (Booster1)   and   Pl1   (Purple1)   that   result   in   sunlight-­‐independent   purple   pigmentation   in   the   plant   tissue   (coleoptile   and   root)   was   found   suitable   for  cases  where  haploid  sorting  is  not  possible  at  dry  seed  stage  (Rotarenco  et  al.,  2010).  In  this  case,  a   pigmented   coleoptile   or   root   in   the   early   developmental   stage   indicates   diploid   state,   while   the   non-­‐ pigmented  seedlings  could  be  designated  as  haploids  (Geiger  and  Gordillo,  2009;  Rotarenco  et  al.,  2010).   Although   CIMMYT   has   a   few   backcross   populations   that   combine   the   root   coloration   marker   with   the   R1-­‐nj  gene,  the  HIR,  agronomic  stability,  and  utility  of  this  alternative  marker  scheme  in  DH  production   need  to  be  established.     To  avoid  possible  misclassification  of  haploids  due  to  poor  expression  of  anthocyanin  color  marker  in  the   dry   seed,   Rotarenco   et   al.   (2007)   proposed   haploid   identification   based   on   kernel   oil   content,   determination   of   which   can   be   potentially   automated   using   nuclear   magnetic   resonance   (NMR)-­‐based   techniques.    Li  et  al.  (2009)  developed  CAUHOI,  a  Stock  6-­‐derived  inducer  with   ~2%  HIR  and  high  kernel   oil   content   (78   g   kg−1),   that   allows   identification   of   haploids   based   on   both   lack   of   R1-­‐nj   conferred   scutellum   coloration   and   low   embryo   oil   content.   This   novel   approach   looks   promising,   but   its   reliability   and   applicability   for   high-­‐throughput   DH   production   in   tropical   genetic   backgrounds   remains   to   be   investigated.   Jones   et   al.   (2012)   examined   the   utility   of   Near-­‐infrared   spectroscopy   to   differentiate   haploids  from  hybrid  maize  kernels  after  maternal  haploid  induction.     Chromosome   doubling:   Several   institutions,   including   CIMMYT,   currently   use   colchicine   as   a   chromosome   doubling   agent   (or   mitotic   inhibitor)   in   DH   production,   as   spontaneous   duplication   of   chromosomes  occurs  at  a  very  low  rate  (Chase,  1969;  Deimling  et  al.,  1997).  However,  treatment  with   colchicine   is   not   always   completely   effective,   and   sectoral   diploidization   of   male   and/or   female   inflorescences   can   occur.   More   importantly,   colchicine   is   highly   carcinogenic,   requiring   very   careful   handling   and   safe   disposal   after   use.   Herbicides   such   as   pronamid,   APM,   trifluralin,   and   oryzalin   have   been   reported   to   be   efficient   as   mitotic   inhibitors   (Häntzschel   and   Weber,   2010).   These   are   less   expensive   and   less   toxic   than   colchicine   and   are   easier   to   handle   and   dispose   of   safely.   Several   commercial  breeding  companies  apply  proprietary  artificial  chromosome  doubling  treatments  that  are   less  toxic  and  safer  than  colchicine  (Geiger  and  Gordillo,  2009).     Agronomic  management:  Optimal  agronomic  management  of  the  colchicine-­‐treated  D0  seedlings,  first   in   the   greenhouse   and   later   in   the   field,   is   highly   crucial   for   the   success   of   DH   line   development,   as   discussed   in   detail   in   chapter   5   of   the   manual.   Optimization   of   irrigation   regime,   fertilizer   application,   possible   mechanization   of   operations,   and   effective   management   of   weeds,   diseases,   and   insects   are   crucial   for   minimizing   stress   on   the   D0   plants   and   improving   the   success   rates   of   DH   line   production.     In   addition   to   proper   agronomic   management,   the   soil   and   climatic   conditions   at   the   DH   operations   site   should  be  optimal.               4       Mechanism(s)  underlying  maternal  haploid  induction   As  explained  in  detail  in  chapter  2  of  this  manual,  several  studies  have  been  undertaken  since  the  1960s   (reviewed   by   Eder   and   Chalyk,   2002;   Geiger   and   Gordillo,   2009)   to   understand   the   biological   mechanism(s)   underlying   in   vivo   maternal   haploid   induction.   Although   some   important   leads   are   available,  the  exact  mechanism(s)  behind  maternal  haploid  induction  are  yet  to  be  fully  understood.  This   has   not,   however,   limited   large-­‐scale   derivation   of   DH   lines   and   utilization   of   DH   parental   lines   in   developing   and   deploying   commercial   maize   cultivars,   especially   by   the   major   commercial   maize   breeding  programs.       Genetic  analyses  of  maternal  haploid  induction  revealed  polygenic  control  of  the  trait  (Lashermes  and   Beckert,  1988;  Deimling  et  al.,  1997;  Röber  et  al.,  2005).  Quantitative  Trait  Loci  (QTL)  mapping  for  in  vivo   haploid   induction   ability   suggested   that   the   trait   is   controlled   by   one   or   a   few   major   QTL   and   several   small-­‐effect   and/or   modifier   QTL.   A   major   QTL   on   chromosome   1   (qhir1,   in   bin   1.04)   explained   up   to   66%   of   the   genetic   variance   for   haploid   induction   ability   in   three   populations   involving   a   non-­‐inducer   parent   and   the   HIR-­‐enhancing   QTL   (Prigge   et   al.   2012).   Identification   and   validation   of   breeder-­‐ready   markers   for   this   major   QTL   and   marker-­‐assisted   introgression   of   the   favorable   allele   could   potentially   speed  up  the  development  of  improved  tropicalized  haploid  inducers  with  high  HIR  and  local  adaptation.     DH  technology  and  molecular  markers,  makes  a  very  powerful  combination   Because   DH   technology   offers   a   faster   way   to   obtain   completely   homozygous   lines,   it   can   save   significant   time   and   resources   for   implementing   genetic   studies   and/or   molecular   breeding   projects,   including:   1. Developing   genetic   maps   (Chang   and   Coe,   2009;   Forster   et   al.,   2007),   which   is   one   of   the   widespread   applications   of   DH   populations   in   many  crop  plants;   2. Identification   of   marker-­‐trait   associations   using   relevant   DH   populations   (with   parents   of   source   populations   showing   significant   phenotypic   contrast),   further   leading   to   potential   use   of   markers   in   marker-­‐assisted   selection  (MAS);   3. High-­‐density   genotyping   of   the   DH   lines   for   selection   of   parental   lines   with   complementary   genotypes   (or   haplotypes)   in   generating  hybrids  for  further  testing;   4. Combining   seed-­‐chipping   technology   in   MAS   of  DH  lines  for  relatively  simply  inherited  traits   (e.g.,   provitamin-­‐A   enrichment)   using   reliable   markers   for   favorable   genes/alleles   with   high   contribution   to   phenotypic   variation,   which   could   be   cheaper,   faster,   and   more   effective   than  phenotyping  the  DH  lines;   5. Potential   usefulness   of   DH   lines   in Figure   2.   An   illustrative   scheme   for   enhancing   implementing   genome-­‐wide   selection   (or   breeding   efficiency   and   genetic   gains   through   a   genomic   selection   or   GS;   Meuwissen   et   al.,   combination  of  modern   technologies/strategies  in   2001;   Jannink   et   al.,   2010)   for   improving   maize  breeding.   complex  polygenic  traits  with  low  heritability   (e.g.,   grain   yield   (GY),   abiotic   stress   tolerance),   and   when   N   (population   size)   is   small   (Bernardo   and  Yu,  2007;  Lorenzana  and  Bernardo,  2009;  Mayor  and  Bernardo,  2009);  and   5       6. Potential  complementary  of  DH  and  MAS  for  deriving  DH  lines  from  bi-­‐parental  crosses  when  the   objective  is  to  obtain  lines  genetically  similar  to  either  parent  of  the  cross  (Smith  et  al.,  2008)  or  to   identify  recombinants  at  or  flanking  specific  loci.  The  most  frequent  application  of  this  approach   would   likely   be   the   use   of   DH   line   conversion   protocols   instead   of   slower   conventional   backcrosses  (Forster  and  Thomas,  2005).       Future  Perspective   The  DH  technology,  undoubtedly,  provides  powerful  means  to  modernize  the  maize  breeding  operations   through   simplified   logistics   and   significantly   lesser   investment   of   resources   for   deriving   completely   homozygous  lines  for  hybrid  development  and  deployment.  Implementation  of  DH  technology  requires   new   skills   on   the   part   of   breeding   programs,   for   both   DH   line   production   and   integrating   DH   lines   efficiently   in   the   breeding   pipeline.   Firstly,   the   major   steps   in   DH   line   production   (haploid   induction,   haploid  identification,  chromosome  doubling,  and  DH  line  recovery)  require  implementation  of  effective   (and   safe)   operational   practices,   and   proper   training   of   the   concerned   scientific/technical   personnel.   Secondly,   the   haploid   maize   plants   derived   through   in   vivo   induction   and   chromosome   doubling   are   often  weak  and  vulnerable  to  various  environmental  stresses,  including  excessive  heat,  insect  pests,  and   diseases.  Thirdly,  the  power  of  DH  technology  in  enhancing  genetic  gains  and  breeding  efficiency,  and   ultimately   for   fast-­‐track   development   of   elite   hybrids,   can   be   realized   when   it   is   effectively   combined   with   MAS   and   year-­‐round   nurseries.   Therefore,   to   be   able   to   effectively   scale-­‐up   DH   development   by   institutions  based  in  the  tropical/subtropical  maize-­‐growing  countries,  these  factors  need  to  be  carefully   considered.       With   financial   support   from   the   Bill   &   Melinda   Gates   Foundation,   CIMMYT   will   soon   be   establishing   a   centralized   maize   DH   facility   for   sub-­‐Saharan   Africa.   The   facility   is   expected   to   serve   primarily   the   DH   requirements  of  public  (not-­‐for-­‐profit)  research  institutions  in  CIMMYT-­‐  and  IITA  (International  Institute   of   Tropical   Agriculture)-­‐led   breeding   networks,   and   to   provide   (over   a   period   of   time)   low-­‐cost   DH   service  to  SME  seed  companies  in  the  region.  CIMMYT  also  plans  to  operationalize  a  DH  service  facility   in   Latin   America,   followed   by   a   similar   facility   in   Asia,   through   the   International   Maize   Improvement   Consortium.     References   Barret   P,   Brinkmann   M,   Beckert   M   (2008)   A   major   locus   expressed   in   the   male   gametophyte   with   incomplete   penetrance  is  responsible  for  in  situ  gynogenesis  in  maize.  Theor.  Appl.  Genet.  117:  581–594.   Bernardo  R,  Yu  J  (2007)  Prospects  for  genomewide  selection  for  quantitative  traits  in  maize.  Crop  Sci.  47:  1082– 1090.     Chang  MT,  Coe  EH  (2009)  Doubled  haploids.  In:  AL  Kriz,  BA  Larkins  (eds)  Biotechnology  in  Agriculture  and  Forestry.   Vol.  63.  Molecular  Genetic  Approaches  to  Maize  Improvement.  Springer  Verlag,  Berlin,  Heidelberg,  pp.  127– 142.   Chase  SS  (1947)  Techniques  for  isolating  monoploid  maize  plants.  J.  Bot.  34:  582.   Chase  SS  (1951)  Production  of  homozygous  diploids  of  maize  from  monoploids.  Agron.  J.  44:  263–267.     Chase  SS  (1952)  Monoploids  in  maize.  Iowa  State  College  Press,  Ames,  Iowa,  pp.  389–399.   Chase   SS   (1969)   Monoploids   and   monoploid-­‐derivatives   in   maize   (Zea   mays   L.).   The   Botanical   Reviews   35:   117– 167.   Chebotar   OD,   Chalyk   ST   (1996)   The   use   of   maternal   haploids   for   genetic   analysis   of   the   number   of   kernel   rows   per   ear  in  maize.  Hereditas  124:  173–178.   Chen  S,  Li  L,  Li  H  (2009)  Maize  doubled  haploid  breeding  [in  Chinese].  China  Agricultural  University  Press,  Beijing.   Coe  EH  (1959)  A  line  of  maize  with  high  haploid  frequency.  Am.  Naturalist  93:  381–382.   Coe   EH   (1994)   Anthocyanin   genetics.   In:   M   Freeling,   V   Walbot   (eds)   The   maize   handbook.   Springer-­‐Verlag,   New   York,  pp.  279–281.   Coe  EH,  Sarkar  KR  (1964)  The  detection  of  haploids  in  maize.  J.  Heredity  55:  231–233.   Deimling   S,   Röber   FK,   Geiger   HH   (1997)   Methodology   and   genetics   of   in   vivo   haploid   induction   in   maize   [in   German].  Vortr  Pflanzenz üchtg  38:  203–224.   6       Eder  J,  Chalyk  ST  (2002)  In  vivo  haploid  induction  in  maize.  Theor.  Appl.  Genet.  104:  703–708.   Forster  BP,  Heberle-­‐Bors  E,  Kasha  KJ,  Touraev  A  (2007)  The  resurgence  of  haploids  in  higher  plants.  Trends  in  Plant   Sci.  12:  368–375.   Forster  BP,  Thomas  WTB  (2005)  Doubled  haploids  in  genetics  and  plant  breeding.  Plant  Breed  Rev.  25:  57–88.   Geiger   HH   (2009)   Doubled   haploids.   In:   JL   Bennetzen,   S   Hake   (eds.)   Maize   handbook   –   volume   II:   genetics   and   genomics.  Springer  Science  and  Business  Media,  New  York,  pp.  641–657.   Geiger  HH,  Gordillo  GA  (2009)  Doubled  haploids  in  hybrid  maize  breeding.  Maydica  54:  485–499.   Greenblatt   IM,   Bock   M   (1967)   A   commercially   desirable   procedure   for   detection   of   monoploids   in   maize.   J.   Hered.   58:  9–13.   Häntzschel   KR,   Weber   G   (2010)   Blockage   of   mitosis   in   maize   root   tips   using   colchicine-­‐alternatives.   Protoplasma   241:  99–104.   Heckenberger   M,   Bohn   M,   Melchinger   AE   (2005)   Identification   of   essentially   derived   varieties   obtained   from   biparental   crosses   of   homozygous   lines:   I.   Simple   sequence   repeat   data   from   maize   inbreds.   Crop   Sci.   45:   1120–1131.   Jannink   J-­‐L,   Lorenz   AJ,   Iwata   H   (2010)   Genomic   selection   in   plant   breeding:   from   theory   to   practice.   Briefings   in   Functional  Genomics  9:  166–177.     Jones  RW,  Reinot  T,  Frei  UK,  Tseng  Y,  Lubberstedt  T,  McClelland  JF  (2012)  Selection  of  haploid  maize  kernels  from   hybrid   kernels   for   plant   breeding   using     near-­‐infrared   spectroscopy   and   SIMCA   analysis.   Applied   Spectroscopy   66:447–450.   Kebede   AZ,   Dhillon,   B.S.,   Schipprack,   W.,   Araus,   J.L.,   Banziger,   M.,   Semagan,   K.,   Alvarado,   G.,   and   Melchinger   AE   (2011)   Effect   of   source   germplasm   and   season   on   the   in   vivo   haploid   induction   rate   in   tropical   maize.   Euphytica  180:  219–226.   Lashermes   P,   Beckert   M   (1988)   Genetic   control   of   maternal   haploidy   in   maize   (Zea   mays   L.)   and   selection   of   haploid  inducing  lines.  Theor.  Appl.  Genet.  76:  404–410.   Li   L,   Xu   X,   Jin   W,   Chen   S   (2009)   Morphological   and   molecular   evidences   for   DNA   introgression   in   haploid   induction   via  a  high  oil  inducer  CAUHOI  in  maize.  Planta  230:  367–376.   Lorenzana   RE,   Bernardo   R   (2009)   Accuracy   of   genetic   value   predictions   for   marker-­‐based   selection   in   biparental   plant  populations.  Theor.  Appl.  Genet.  120:  151–161.     Mayor   PJ,   Bernardo   R   (2009)   Genomewide   selection   and   marker-­‐assisted   recurrent   selection   in   doubled   haploid   versus  F2  populations.  Crop  Sci.  49:  1719–1725.     Meuwissen  THE,  Hayes  BJ,  Goddard  ME  (2001)  Prediction  of  total  genetic  value  using  genome-­‐wide  dense  marker   maps.  Genetics  157:  1819–1829.   Nanda  DK,  Chase  SS  (1966)  An  embryo  marker  for  detecting  monoploids  of  maize  (Zea  mays  L.).  Crop  Sci.  6:  213– 215.   Prasanna   BM,   Pixley   K,   Warburton   ML,   Xie   CX   (2010)   Molecular   marker-­‐assisted   breeding   options   for   maize   improvement  in  Asia.  Mol.  Breed.  26:  339–356.   Prigge   V,   Melchinger   AE   (2011)   Production   of   haploids   and   doubled   haploids   in   maize.   In:   VM   Loyola-­‐Vargas,   Ochoa-­‐Alejo   N   (eds)   Plant   cell   culture   protocols,   3rd   edition.   Humana   Press   -­‐   Springer   Verlag,   Totowa,   New   Jersey.   Prigge   V,   C   Sanchez,   BS   Dhillon,   W   Schipprack,   JL   Araus,   M   Banziger,   AE   Melchinger   (2011)   Doubled   haploids   in   tropical   maize:   I.   Effects   of   inducers   and   source   germplasm   on   in   vivo   haploid   induction   rates.   Crop   Sci.   51:   1498–1506.   Prigge   V,   Xu   XW,   Li   L,   Babu   R,   Chen   SJ,   Atlin   GN,   Melchinger   AE   (2012)   New   insights   into   the   genetics   of   in   vivo   induction  of  maternal  haploids,  the  backbone  of  doubled  haploid  technology  in  maize.  Genetics  111:  781–793.   Röber   FK,   Gordillo   GA,   Geiger   HH   (2005)   In   vivo   haploid   induction   in   maize   –   performance   of   new   inducers   and   significance  of  doubled  haploid  lines  in  hybrid  breeding.  Maydica  50:  275–283.   Rotarenco  VA,  Kirtoca  IH,  Jacota  AG  (2007)  Possibility  to  identify  kernels  with  haploid  embryo  by  oil  content.  Maize   Genet.  Coop.  Newslett.  81:  11.   Rotarenco   VA,   Dicu   G,   State   D,   Fuia   S   (2010)   New   inducers   of   maternal   haploids   in   maize.   Maize   Genet.   Coop.   Newslett.  84:  15.   Schmidt  W  (2003)  Hybrid  maize  breeding  at  KWS  SAAT  AG.  In:  Bericht  über  die  Arbeitstagung  der  Vereinigung  der   Pflanzenz  üchter  und  Saatgutkaufleute  Österreichs,  Gumpenstein,  Österreich,  25–27  November,  pp.  1–6.     Seitz  G  (2005)  The  use  of  doubled  haploids  in  corn  breeding.  In:  Proc.  41st  Annual  Illinois  Corn  Breeders’  School   2005.  Urbana-­‐Champaign,  Illinois,  pp.  1–7.     7       Shatskaya  OA,  Zabirova  ER,  Shcherbak  VS,  Chumak  MV  (1994)  Mass  induction  of  maternal  haploids.  Maize  Genetics   Coop.  Newslett.  68:  51.   Smith  JSC,  Hussain  T,  Jones  ES,  Graham  G,  Podlich  D,  Wall  S,  Williams  M  (2008)  Use  of  doubled  haploids  in  maize   breeding:   implications   for   intellectual   property   protection   and   genetic   diversity   in   hybrid   crops.   Mol.   Breed.   22:  51–59.   Tyrnov  VS,  Zavalishina  AN  (1984)  Inducing  high  frequency  of  matroclinal  haploids  in  maize  [in  Russian].  Dokl  Akad   Nauk  SSSR  276:  735–738.   Wijnker   E,   Vogelaar   A,   Dirks   R,   van   Dun   K,   de   Snoo   B,   van   den   Berg   M,   Lelivelt   C,   de   Jong   H,   Chunting   L   (2007)   Reverse  breeding:  reproduction  of  F1  hybrids  by  RNAi-­‐induced  asynaptic  meiosis.  Chromosome  Research  15:   87–88.     8       2.  In  vivo  Maternal  Haploid  Induction  in  Maize     Vijay  Chaikam       In  vivo  versus  in  vitro  haploid  induction   Haploids   in   maize   can   be   obtained   either   through   in   vitro   (androgenesis)   or   in   vivo   methods.   Androgenesis   refers   to   the   development   of   haploid   plants   from   immature   pollen   either   by   anther   culture  or  microspore  culture.  In  anther  culture  systems,  microspores  within  the  anther  are  induced  to   undergo   androgenesis   to   form   microspore-­‐derived   embryo-­‐like   structures.   In   pollen   culture,   microspores   are   isolated   from   anthers   and   cultured   on   a   medium   to   produce   embryo-­‐like   structures.   Embryo-­‐like   structures   can   either   directly   regenerate   into   haploid   plants   or   indirectly   regenerate   via   the   formation   of   regenerable   calli.   As   microspores   are   produced   in   abundance   in   plant   anthers,   they   are   relatively   easy   to   access   and   manipulate   in   cultures.   Although   androgenesis   protocols   are   well   established   and   routinely   used   in   some   crop   species,   obtaining   haploids   and   doubled   haploids   (DH)   through  androgenesis  has  not  proved  to  be  efficient  in  maize.  Androgenesis  in  maize  was  found  to  be   highly   genotype-­‐dependent;   most   maize   genotypes   are   recalcitrant   and   do   not   show   any   response   in   culture  (Brettell  et  al.,  1981;  Genovesi  and  Collins,  1982;  Miao  et  al.,  1981;  Spitkó  et  al.,  2006).  Even  in   genotypes  that  respond  to  androgenesis,  this  process  is  highly  influenced  by  many  conditions,  including   anther   stage,   donor   plant,   and   anther   pretreatment.   (Wan   et   al.,   1991;   Chu   et   al.,   1975;   Ku   et   al.,   1978;   Genovesi   and   Collins,   1982;   Miao   et   al.,   1978;   Spitkó   et   al.,   2006).   Therefore,   in   vitro   approaches   for   DH   development  are  not  very  commonly  used  in  maize.       In  contrast,  in  vivo  haploid  induction  has  been  highly  successful  in  maize  and  is  now  extensively  followed   by   several   commercial   breeding   programs   (as   discussed   in   chapter   1   in   this   manual).   Haploids   were   reported   to   occur   naturally   in   maize   plantings   at   a   frequency   of   about   0.1%   (Chase,   1951).   Such   a   frequency   of   induction   cannot   be   exploited   efficiently   for   large-­‐scale   DH   operations.   The   discovery   of   Stock   6   (Coe,   1959)   and   further   derivation   of   an   array   of   maternal   haploid   inducers   in   maize,   as   described  earlier  in  this  manual,  revolutionized  the  application  of  DH  technology  in  maize  breeding,  as   this  method  is  much  less  dependent  on  the  donor  genotypes  (source  germplasm)  from  which  DH  lines   are  derived.       Maternal  versus  paternal  haploids   The   induction   of   paternal   (androgenetic)   haploids   is   based   on   a   mutant   gene,   ig1   (indeterminate   gametophyte),  which  can  increase  the  frequency  of  haploids  in  its  progeny  (Kermicle,  1969,  1971;  Lin,   1981).   Homozygous   ig1   mutants   show   several   embryological   abnormalities   including   egg   cells   without   a   nucleus.   After   fusion   with   one   of   the   two   paternal   sperm   cells,   such   an   egg   cell   may   develop   into   a   haploid  embryo  possessing  the  maternal  cytoplasm  and  only  paternal  chromosomes.  In  selected  genetic   backgrounds,  the  haploid  induction  rate  ranges  from  1%  to  2%  (Kermicle,  1994).       To   produce   paternal   haploids,   the   inducer   (with   ig1)   is   used   as   the   female   parent   and   the   donor   (source   germplasm)   as   the   male   parent.   Hence,   paternal   haploids   contain   the   cytoplasm   of   the   inducer   and   chromosomes   from   the   donor   plant.   Low   frequency   of   haploids   and   changes   in   the   constitution   of   cytoplasm   from   the   donor   genotype   make   this   system   not   very   attractive   to   derive   inbred   lines   for   breeding.   However,   the   ig1/ig1   genetic   stock   can   be   useful   for   the   conversion   of   an   inbred   line   to   its   cytoplasmic   male   sterile   form.   The   DH   plants   obtained   in   this   method   are   isogenic   with   the   male   parent   except  that  they  carry  male-­‐sterile  cytoplasm.  Inducer  lines  with  various  Cytoplasmic  Male  Sterile  (CMS)-­‐ inducing   cytoplasms  have  been  created,   which   can   be   used   to   transfer   new   breeding   lines   into   the   CMS   cytoplasm  (Pollacsek,  1992;  Schneerman  et  al.,  2000).       9   For   producing   maternal   haploids,   the   haploid   inducer   is   used   as   the   male   parent   in   induction   crosses,   with   the   source   germplasm   or   donor   as   the   female   parent.   Maternal  haploids   carry   both   cytoplasm   and   chromosomes   from   the   donor.   Many  haploid   inducer   lines   with   commercially   usable   and   higher   haploid   induction  rates  (HIR)  are  now  available,  the  details  of  which  were  provided  in  the  introductory  chapter   of  this  manual.     To   develop   improved   haploid   inducers   adapted   to   tropical   conditions,   segregating   populations   were   developed  at  CIMMYT  from  crosses  between  temperate  inducers  (RWS,  UH400,  and  RWS  x  RWK-­‐  with   HIR   of   8–10%)   and   three   tropical   maize   lines   developed   by   CIMMYT   (CML494,   CML451,   and   CL02450).   A   pedigree   breeding   scheme   was   followed   with   mass   selection   for   highly   heritable   and   visually   scorable   traits  on  individual  F2  plants  and  family-­‐based  selection  for  HIR  and  other  agronomic  characteristics  in   advanced   selfing   and   backcross   generations   (Prigge   et   al.,   2011).   Tropically   adapted   inducer   lines   so   developed   combined   high   HIR   (ranging   from   6%   to   13%)   with   improved   pollen   production,   disease   resistance,  and  plant  vigor  compared  to  the  temperate  inducers  under  tropical  conditions.       Mechanism  of  in  vivo  maternal  haploid  induction     The  exact  sequence  of  events  underlying  maternal  haploid  induction  has  not  been  clearly  understood.   Several  hypotheses  were  proposed   to   explain  in  vivo   maternal   haploid   induction.  As   haploid   induction   is   achieved   when   an   inducer   line   is   used   as   a   pollen   parent,   hypotheses   were   proposed   that   the   regular   double   fertilization   is   distorted   after   pollination   with   the   pollen   of   a   haploid   inducer   line.   In   normal   double   fertilization,   one   of   the   two   sperm   cells   from   the   pollen   grain   fertilizes   the   egg   cell   to   form   a   diploid  zygote  and  the  other  sperm  cell  fertilizes  the  two  polar  nuclei  of  the  central  cell  in  the  female   gametophyte,   which   ultimately   develops   into   triploid   endosperm.   Pollen   from   haploid   inducers   was   proposed  to  cause  a  distortion  in  double  fertilization  in  such  a  way  that  one  sperm  cell  fuses  with  the   central  cell  but  the  other  sperm  cell  does  not  fuse  with  the  egg  cell.  But  a  fertilized  and  dividing  central   cell   stimulates   the   unfertilized   haploid   egg   cell   to   develop   into   a   haploid   embryo   (Chase,   1969).   Such   single  fertilization  could  be  a  result  of  morphological  defects  in  pollen  grains  or  existence  of  only  a  single   normal  sperm  in  a  pollen  grain.  Pogna  and  Marzetti  (1977)  germinated  pollen  grains  from  inducers  and   non-­‐inducers  in  vitro  and  observed  that  pollen  grains  from  inducers  exhibited  two  pollen  tubes  at  high   frequency.   They   proposed   that   such   an   abnormality   in   pollen   tube   growth   may   be   related   to   haploid   induction  capability.       Bylich   and   Chalyk   (1996)   noticed   about   6.3%   of   pollen   grains   with   a   pair   of   morphologically   different   sperm   nuclei   in   haploid   inducer   line   ZMS.   They   proposed   that   the   morphological   differences   could   possibly  arise  as  two  sperms  cells  develop  at  different  speeds,  which  could  lead  to  development  of  one   sperm   that   is   in   a   state   ready   for   fertilization   and   another   that   is not.   High   heterofertilization   frequency   was  noticed  with  Stock  6  (Sarkar  and  Coe  1966,  1971).  Similar  observations  were  made  with  inducer  line   MHI   by   Rotarenko   and   Eder   (2003).   Heterofertilization,   usually   caused   by   delayed   fertilization,   is   proposed   to   be   related   to   the   mechanism   of   haploid   induction   as   well   as   the   HIR.   Mahendru   and   Sarkar   (2000),  however,  could  not  find  any  difference  between  the  two  sperms  in  pollen  of  a  haploid  inducing   line.  Swapna  and  Sarkar  (2011)  also  could  not  find  any  defects  in  pollen  tube  growth  and  did  not  observe   delayed  fertilization.  They  proposed  attenuation  of  sperm  nuclei  after  the  release  from  the  synergid  into   the  embryo  sac  as  a  possible  cause  of  haploid  induction.       Chalyk  et  al.  (2003)  found  10%  to  15%  aneuploid  microsporocytes  in  the  haploid  induction  lines  MHI  and   M471H.  They  proposed  that  in  the  haploid  inducers,  abnormal  division  of  chromosomes  occurs  during   microsporocyte  formation,  which  may  lead  to  development  of  aneuploid  sperm.  Aneuploid  gametes  can   break  doubled  fertilization  and  stimulate  egg  cell  development  into  embryo  without  fertilization.       10   In   contrast   to   the   above,   some   researchers   (Wedzony   et   al.,   2002)   indicated   that   during   maternal   haploid   induction,   normal   fertilization   might   still   occur,   but   during   the   subsequent   cell   divisions,   the   inducer   chromosomes   degenerate   and   are   then   eliminated   from   the   primordial   cells.   Fischer   (2004)   used  microsatellite  markers  to  check  for  strictly  maternal  origin  of  haploids  induced  by  RWS.  About  1.4%   of   the   genotypes   possessed   one   or,   rarely,   several   inducer   chromosome   segments.   Generally,   these   segments   had   replaced   the   homologous   maternal   segments.   Li   et   al.   (2009)   and   Zhang   et   al.   (2008)   demonstrated  that  chromosomal  segments  from  inducer  parent  are  integrated  into  the  genome  of  the   haploids   and   doubled   haploids,   suggesting   elimination   of   chromosomes   from   the   inducer   parent   after   fertilization.     Taking   all   of   this   information   together,   the   mechanism   of   haploid   induction   is   yet   to   be   conclusively   elucidated.   However,   it   is   certain   that   some   reproductive   abnormalities   are   involved,   and   it   is   also   possible   that   different   inducers   may   cause   different   reproductive   abnormalities   leading   to   maternal   haploid  formation.       Genetics  and  molecular  marker  analysis  of  maternal  haploid  induction     Studies   on   segregating   generations   derived   from   crosses   between   inducer   and   non-­‐inducer   parents   revealed   continuous   variation   for   haploid   induction   associated   traits   and   indicated   that   the   in   vivo   haploid  induction  trait  is  under  polygenic  control  (Lashermes  and  Beckert,  1988;  Deimling  et  al.,  1997;   Röber   et   al.,   2005,   Vanessa   et   al.,   2011).   Lashermes   and   Beckert   (1988)   inferred   that   the   haploid   induction   trait   of   the   Stock   6   inducer   line   is   a   dominant   character   with   nuclear   determination   and   is   controlled  by  a  few  major  genes.     Deimling  et  al.  (1997)  and  Röber  (1999)  used  Restriction  Fragment  Length  Polymorphism  (RFLP)  markers   and   identified   two   QTL   (on   chromosomes   1   and   2)   responsible   for   haploid   induction   in  an   F3   population   involving  Stock  6  and  W23ig  as  parents.  These  QTL  together  explained  17.9%  of  the  phenotypic  variance   and  40.7%  of  the  genotypic  variance  in  haploid  induction  rates.  The  positive  QTL  allele  on  chromosome   1   was   dominant   and   originated   from   Stock   6   whereas   the   one   on   chromosome   2   was   additive   and   originated  from  W23ig.     In   another   study,   Barret   et   al.   (2008)   found   segregation   distortion   in   a   population   developed   from   a   cross   between   a   non-­‐inducer   and   an   inducer   line   (PK6).   This   analysis   revealed   a   major   locus   on   chromosome  1  covering  11.6  cM  in  bin  1.04  for  haploid  induction.  Fine  mapping  based  on  synteny  with   rice  chromosomes  led  to  identification  of  two  Sequence-­‐Tagged  Site  (STS)  markers  closely  linked  to  the   induction  locus  (4.5  and  4.9   cM,   respectively).   This   fine-­‐mapped  region  contained  28  putative  expressed   genes.     Prigge   et   al.   (2012)   conducted   comparative   QTL   mapping   involving   four   segregating   mapping   populations,   which   were   developed   by   crossing   haploid   inducer   line   UH400   with   two   temperate   (CAUHOI,  1680)  and  two  tropical  (CML395,  CML495)  inbreds.  In  three  of  these  populations  a  major  QTL   was   identified   for   haploid   induction   on   chromosome   1   (bin   1.04)   explaining   up   to   66%   of   the   genetic   variance.   The   loci   in   bin   1.04   exhibited   segregation   distortion   against   the   UH400   allele   in   these   three   populations.   In   another   segregating   population   involving   two   inducer   lines   as   parents   (CAUHOI   ×   UH400),  seven  QTL  were  identified  on  five  chromosomes,  with  one  QTL  on  chromosome  9  contributing   20%  in  three  generations  of  this  cross.  The  results  led  to  the  suggestion  of  pyramiding  of  major  QTL  on   chromosome  1  and  minor  QTLs  could  lead  to  further  improvement  in  induction  capabilities.             11   Source  germplasm  for  haploid  induction   The   choice   of   source   germplasm   or   donor   for   haploid   induction   depends   on   the   objectives   of   the   breeding   programs.   Usually   breeders   induce   haploids   on   the   F1   or   F2   populations.   It   was   estimated   that   an   F2-­‐derived   DH   may   contain   almost   50%   more   of   the   best   recombinants   than   an   F1-­‐derived   population  (Gallais,  1990).  However,  the  difference  in  the  frequency  of  the  best  recombinants  between   F2-­‐   and   F3-­‐derived   populations   is   small.   This   implies   that   the   DH   approach   is   better   followed   on   F2   populations  when  linkage  is  observed  between  genes  (Gallais,  1990;  Bernardo,  2009).     In  maize,  a  high  mutational  load  of  deleterious  recessive  alleles  hampers  exploiting  the  genetic  potential   of   allogamous   landraces   in   hybrid   breeding.   It   was   proposed   that   the   DH   technology   could   be   an   effective   approach   for   eliminating   deleterious   recessives   from   a   gene   pool   (Gallais,   1990,   Wilde   et   al.,   2010).   Even   though   landrace-­‐derived   lines   may   not   be   directly   used   as   parents   in   hybrid   breeding   programs   because   of   significant   differences   in   performance   for   agronomically   important   traits   as   compared  to  elite   inbred  lines,  they  may  be  valuable  genetic  resources  for  marker-­‐assisted  backcrossing   or  pre-­‐breeding  activities  (Wilde  et  al.,  2010).  Compared  to  elite  inbred  lines,  landrace-­‐derived  DH  lines   are  much  closer  to  Hardy-­‐Weinberg  equilibrium,  which  allow  detection  and  mapping  of QTL  with  high   accuracy   and   resolution.   So   land   race   derived   DH   lines   are   ideally   suited   for   marker-­‐trait   association   studies    (Wilde  et  al.,  2010).   References   Barret   P,   Brinkmann   M,   Beckert   M   (2008)   A   major   locus   expressed   in   the   male   gametophyte   with   incomplete   penetrance  is  responsible  for  in  situ  gynogenesis  in  maize.  Theor.  Appl.  Genet.  117:  581–594.     Bernardo   R   (2009)   Should   maize   doubled   haploids   be   induced   among   F1   or   F2   plants?   Theor.   Appl.   Genet.   119:   255–262.   Brettell  RIS,  Thomas  E,  Wernicke  W  (1981)  Production  of  haploid  maize  plants  by  anther  culture.   Maydica  26:  101– 111.   Bylich   VG,   Chalyk   ST   (1996)   Existence   of   pollen   grains   with   a   pair   of   morphologically   different   sperm   nuclei   as   a   possible  cause  of  the  haploid-­‐inducing  capacity  in  ZMS  line.  Maize.  Genet.  Coop.  Newslett.  70:  33.     Chalyk  S,  Baumann  A,  Daniel  G,  Eder  J  (2003)  Aneuploidy  as  a  possible  cause  of  haploid-­‐induction  in  maize.  Maize   Genet.  Coop.  Newslett.  77:  29.     Chase  SS  (1951)  Production  of  homozygous  diploids  of  maize  from  monoploids.  Agron.  J.  44:  263–267.   Chase  SS  (1969)  Monoploids  and  monoploid-­‐derivatives  of  maize  (Zea  mays  L.).  Bot.  Rev.  35:  117–167.   Chu   CC,   Wang   CC,   Sun   C,   Shu   KC,   Yin   CY,   Chu   FY   (1975)   Establishment   of   an   efficient   medium   for   anther   culture   of   rice  through  comparative  experiments  on  the  nitrogen  sources.  Sci.  Sinica  18:  659–668.   Coe  EH  (1959)  A  line  of  maize  with  high  haploid  frequency.  Am.  Naturalist  93:  381–382.   Deimling   S,   Röber   FK,   Geiger   HH   (1997)   Methodik   und   Genetik   der   in-­‐vivo-­‐Haploideninduktion   bei   Mais.   Vortr.   Pflanzenzüchtung  38:  203–224.   Fischer  E  (2004)  Molekulargenetische  Untersuchungen  zum  Vorkommen  paternaler  DNA-­‐Übertragung  bei  der  in-­‐ vivo-­‐Haploiderinduktion   bei   Mais   (Zea   mays   L.).   PhD   dissertation,   University   of   Hohenheim.   Grauer   Verlag,   Stuttgart,  Germany.   Gallais   A   (1990)   Quantitative   genetics   of   doubled   haploid   populations   and   application   to   the   theory   of   line   development.  Genetics  124:  199–206.   Genovesi   AD,   Collins   GB   (1982)   In   vitro   production   of   haploid   plants   of   corn   via   anther   culture.   Crop   Sci.   22:   1137– 1144.   Kermicle  JL  (1969)  Androgenesis  conditioned  by  a  mutation  in  maize.  Science  166:  1422–1424.   Kermicle   JL   (1971)   Pleiotropic   effects   on   seed   development   of   the  indeterminate   gametophyte   gene   in   maize.   Am.   J.  Bot.  58:  1–7.   Kermicle   JL   (1994)   Indeterminate   gametophyte   (ig):   biology   and   use.   In:   M   Freeling,   V   Walbot   (eds)   The   maize   handbook,  New  York.  Springer-­‐Verlag,  pp.  388–393.   Ku   MG,   Cheng   WC,   Kuo   LC,   Kuan,   YL,   An   HP,   Huang   CH   (1978)   Induction   factors   and   morpho-­‐cytological   characteristics  of  pollen  derived  plants  in  maize  (Zea  mays).  In:  Proc.  Symp.  Plant  Tissue  Culture  May  25-­‐30,   1978,  Beijing,  China,  pp.  35–42.  Science  Press,  Beijing,  China.     12   Lashermes   P,   Beckert   M   (1988)   Genetic   control   of   maternal   haploidy   in   maize   (Zea   mays   L.)   and   selection   of   haploid  inducing  lines.  Theor.  Appl.  Genet.  76:  404–410.   Li   L,   Xu   X,   Jin   W,   Chen   S   (2009)   Morphological   and   molecular   evidences   for   DNA   introgression   in   haploid   induction   via  a  high  oil  inducer  CAUHOI  in  maize.  Planta  230:  367–376.   Lin   BY   (1981)   Megagametogenetic   alterations   associated   with   the   indeterminate   gametophyte   (ig)   mutation   in   maize.  Rev.  Bras.  Biol.  41:  557–563.   Mahendru  A,  Sarkar  KR  (2000)  Cytological  analysis  of  the  pollen  of  haploidy  inducer   lines   in   maize   (Zea  mays  L.)   Indian  J.  Genet.  Plant  Breed.  60:  37–43.   Miao   SH,   Kuo   CS,   Kwei   YL,   Sun   AT,   Lu   WL,   Wang   YY   (1981)   Induction   of   pollen   plants   of   maize   and   observations   on   their   progeny.   In:   Proc.   Symp.   Plant   Tissue   Culture,   date,   Beijing,   China,   pp.   23–24.   Science   Press,   Beijing,   China.   Pogna   NE,   Marzetti   A   (1977)   Frequency   of   two   tubes   in   in   vitro   germinated   pollen   grains.   Maize   Genet.   Coop.   Newslett.  51:  44.   Pollacsek  M  (1992)  Management  of  the  ig  gene  for  haploid  induction  in  maize.  Agronomie  12:  247–251.   Prigge   VC,   Sanchez     BS,   Dhillon   W,   Schipprack,   Araus   JL,   Banziger   M,   Melchinger   AE   (2011)   Doubled   haploids   in   tropical   maize:   I.   Effects   of   inducers   and   source   germplasm   on   in   vivo   haploid   induction   rates.   Crop   Sci.   51:   1498–1506.   Prigge   V,   Xu   X,   Li   L,   Babu   R,   Chen   S,   Atlin   GN,   Melchinger   AE   (2012)   New   insights   into   the   genetics   of   in   vivo   induction  of  maternal  haploids,  the  backbone  of  doubled  haploid  technology  in  maize.  Genetics  190:  781–793.   Röber   FK   (1999)   Fortpflanzungsbiologische   und   genetische   Untersuchungen   mit   RFLP-­‐Markern   zur   in-­‐vivo   Haploideninduktion  bei  Mais.  Ph.D.  dissertation,  University  of  Hohenheim,  Stuttgart,  Germany.   Röber   FK,   Gordillo   GA,   Geiger   HH   (2005)   In   vivo   haploid   induction   in   maize   –   performance   of   new   inducers   and   significance  of  doubled  haploid  lines  in  hybrid  breeding.  Maydica  50:  275–283.   Rotarenco  VA,  Eder  J  (2003)  Possible  effect  of  heterofertilization  on  the  induction  of  maternal  haploids  in  maize.   Maize  Genet.  Coop.  Newslett.  77:  30.   Sarkar  KR,  Coe  EH  Jr  (1966)  A  genetic  analysis  of  the  origin  of  maternal  haploids  in  maize.  Genetics  54:  453–464.   Sarkar  KR,  Coe  EH  Jr  (1971)  Analysis  of  events  leading  to  heterofertilization  in  maize.  J.  Hered.  62:  118–120.   Schneerman   MC,   Charbonneau   M,   Weber   DF   (2000)   A   survey   of   ig   containing   materials.   Maize   Genet.   Coop.   Newslett.  74:92–93.   Spitkó  T,  Sági  L,    Pintér  J,    Marton  LC,  Barnabás  B  (2006)  Haploid  regeneration  aptitude  maize  (Zea  mays  L.)  lines  of   various  origin  and  of  their  hybrids.  Maydica  51:  537–542   Swapna  M,  Sarkar  KR  (2011)  Anomalous  fertilization  in  haploidy  inducer  lines  in  maize  (Zea  mays  L).    Maydica  56:   221–225   Wan   Y,   Duncan   DR,   Rayburn   AL,   Petolino   JF,   Widholm   JM   (1991)   The   use   of   antimicrotubule   herbicides   for   the   production  of  doubled  haploid  plants  from  anther-­‐derived  maize  callus.  Theor.  Appl.  Genet.  81:  205–211.   Wedzony  M,  Röber  FK,  Geiger  HH  (2002)  Chromosome  elimination  observed  in  selfed  progenies  of  maize  inducer   line   RWS.   In:   XVIIth   International   Congress   on   Sex   Plant   Reports   Maria   Curie-­‐Sklodowska   University   Press,   Lublin,  p.  173.   Wilde   K,   Burger   H,   Prigge   V,   Presterl   T,   Schmidt   W,   Ouzunova   M,   Geiger   HH   (2010)   Testcross   performance   of   doubled-­‐haploid  lines  developed  from  European  flint  maize  landraces.  Plant  Breeding  129:  181–185.     Zhang  Z,  Qiu  F,  Liu  Y,  Ma  K,  Li  Z,  Xu  S  (2008)  Chromosome  elimination  and  in  vivo  haploid  production  induced  by   Stock  6-­‐derived  inducer  line  in  maize  (Zea  mays  L.).  Plant  Cell  Rep.  27:  1851–1860.       13   3.  Design  and  Implementation  of  Maternal  Haploid  Induction     Vijay  Chaikam,  George  Mahuku,  and  BM  Prasanna     For  successfully  producing  an  optimal  number  of  doubled  haploid  (DH)  lines  from  a  source  population,   the   first   critical   step   is   to   produce  enough   haploid   seeds   from  the  induction  crosses.  This   will   depend   on   three   important   factors:   (1)   haploid   induction   rate   (HIR)   and   pollen   production   capabilities   of   inducer   used,  (2)  total  number  of  successful  induction  crosses,  and  (3)  lack  of  anthocyanin  color  inhibitors  in  the   source   population.   The   design   of   the   induction   nursery   also   affects   the   efficiency   in   handling   the   pollinations   and   the   number   of   successful   induction   crosses.   These   factors   need   to   be   thoroughly   considered  before  planting  the  induction  nursery.     Selection  of  inducer  lines  for  haploid  induction   Inducer  lines  for  the  haploid  induction  nursery  should  be  selected  based  on  HIR,  pollen  production,  plant   height,   vigor   and   per   se   performance   of   the   inducer,   flowering   behavior,   resistance   to   diseases   and   insects,  and  ease  of  maintenance  of  the  inducer  in  the  target  environment.  For  large-­‐scale  commercial   application   of   DH   technology,   haploid   inducers   with   high   HIR   should   be   chosen   for   the   induction   nursery.   As   mentioned   in   chapter   1,   several   inducer   lines   have   been   developed   with   an   average   HIR   above   6%.   However,   most   of   these   inducer   lines   are   better   adapted   to   temperate   environments.   CIMMYT  has  been  using  temperate  inducer  lines  UH400,  RWS,  and  their  hybrid  for  haploid  induction  in   tropical  and  subtropical  environments  in  Mexico.  The  HIR  of  these  temperate  inducers  is  maintained  at   similar   levels   (~8–10%)   in   tropical   and   subtropical   environments.   However,   these   temperate   inducer   lines   and   their   hybrids   exhibit   poor   agronomic   characteristics   and   disease   vulnerability   in   tropical   environments.  Nevertheless,  it  is  possible  to  obtain  pollen  for  haploid  induction  with  multiple  sprays  of   fungicides,   insecticides,   foliar   nutrition,   and   other   best   agronomic   management   practices.   Temperate   inducer  lines  and  their  hybrids  are  very  short  in  height,  which  makes  them  almost  impossible  to  use  in   isolation  blocks  with  open  pollinations,  thereby  necessitating  expensive  manual  pollinations.  Extremely   early  flowering  and  a  very  brief  period  of  pollen  shedding  also  make  it  necessary  to  stagger  inducer  lines   multiple   times   to   coincide   flowering   with   tropical   source   germplasm.   Seed   production   and   maintenance   of  temperate  inducers  are  also  problems  in  tropical  conditions  as  they  produce  very  small  ears  which  are   susceptible  to  ear  rots.  Temperate  inducers  show  comparatively  better  performance  in  winter  induction   nurseries  than  summer  nurseries  in  tropical  environments  in  Mexico.     In   contrast,   the   tropicalized   haploid   inducer   lines   (TAILs)   developed   at   CIMMYT-­‐Mexico,   in   collaboration   with   the   University   of   Hohenheim,   exhibit   better   agronomic   characteristics   in   terms   of   flowering   and   pollen  production  in  tropical  environments,  while  maintaining  high  HIR  of  8–12%.The  TAILs  also  exhibit   better   resistance   to   tropical   diseases   and   insects,   making   agronomic   management   less   expensive   in   tropical  environments.  Hybrids  of  tropical  inducers  are  taller  compared  to  temperate  inducer  hybrids,  so   they  can  be  used  in  an  isolation  nursery  with  open  pollination.  Seed  production  and  line  management   are  also  comparatively  easy  for  TAILs  in  tropical  environments.                     14                                  Figure   1.   Temperate   and   tropical   inducer   lines   at   CIMMYT   El   Batán   experimental   station,  Mexico.         Number  of  induction  crosses   The  number  of  induction  crosses  per  source  population  depends  on  the  number  of  haploid  seeds  to  be   produced   per   source   population,   which   in   turn   depends   on   the   target   number   of   DH   lines   to   be   produced.  At  CIMMYT,  we  aim  to  produce  200  DH  lines  from  each  source  population.  At  a  10%  success   rate   in   chromosomal   doubling   we   need   at   least   2,000   haploid   seeds   to   obtain   200   DH   lines.   At   an   8%   induction  rate  and  200  kernels  per  ear  we  need  to  have  125  successful   crosses  to  obtain  2,000  haploids.   We   plant   at   least   150   plants   to   allow   for   plant   losses   due   to   non-­‐germination   and   post-­‐germination   death  of  the  plants.   Manual  vs.  open  pollination  in  the  haploid  induction  nursery   Induction  crosses  can  be  conducted  in  an  isolation  nursery  using  open  pollination  or  could  be  conducted   in   a   nursery   using   manual   pollinations.   The   decision   to   use   open   pollination   or   manual   pollination   depends  on  several  factors.     An  isolation  nursery  with  open  pollinations  can  be  a  chosen  when:   It  is  possible  to  plant  an  induction  nursery  at  least  one  month  earlier  than  the  rest  of  the  maize   plantings  in  the  surrounding  area;   The  source  populations  do  not  differ  in  their  silk  emergence  date  by  more  than  15-­‐20  days;     There  are  more  than  50  source  populations;  and     A  taller  inducer  or  inducer  hybrid  is  available  that  reaches  at  least  the  height  of  the  ears  of  the   source  population  plants.      Manual  pollinations  may  be  preferred  in  an  induction  nursery  when:   Few  populations  to  be  induced;   When  flowering  time  information  is  not  available  for  source  populations   Source  populations  have  a  wide  range  of  maturity;   Isolation  by  early  planting  would  not  be  possible;  and     The  inducers  used  are  very  short  in  height  relative  to  the  source  populations.       15   Design  of  the  induction  nursery   A  good  design  of  the  induction  nursery  is  important  for  efficient  handling  of  pollinations  and  to  achieve   success  in  haploid  induction.  The  same  field  design  can  be  used  for  the  induction  nursery  with  isolation   using  open  pollinations  and  an  induction  nursery  using  manual  pollinations.  Flowering  time  information   for  the  source  population  (days  to  silking)  and  inducers  (days  to  anthesis)  is  necessary  for  designing  the   induction   nursery.   All   source   populations   with   similar   silking   time   can   be   grouped   and   planted   in   the   same  area  of  the  nursery  so  that  pollinations  can  be  handled  easily.     At  CIMMYT’s  DH  nursery  in  the  Agua  Fría  experimental  station,  seeds  from  the  source  populations  are   sown  in  4.5  m  long  rows  at  a  spacing  of  25  cm.  Each  row  accommodates  19  plants.  The  spacing  between   rows   is   maintained   at   75   cm.   Haploid   inducer   lines   are   planted   in   long   ranges.   A   typical   design   for   an   induction   nursery   is   represented   in   Figures   2   and   3.   Since   the   tropical   and   temperate   inducer   lines   flower   much   earlier   than   the   tropical   source   germplasm,   planting   of   source   populations   can   be   done   earlier.  All  the  source  populations  with  early  to  intermediate  silking  dates  can  be  accommodated  in  the   front   to   the   middle   of   the   nursery.   Other   populations   with   intermediate   to   late   silking   dates   can   be   accommodated  from  the  middle  to  the  end  of  the  nursery.  For  each  source  population,  eight  rows  are   planted.   After   every   four   rows   of   source   populations,   two   long   ranges   (highlighted   in   yellow   in   Figure   3)   are  left  for  inducer  planting.  Also,  two  horizontal  ranges  in  the  front  and  two  horizontal  ranges  in  the   back   (highlighted   in   yellow   in   Figure   3)   of   the   induction   nursery   are   left   for   inducer   planting.   Inducer   plantings  need  to  be  staggered  at  weekly  intervals  depending  on  the  variability  in  silking  dates  of  source   populations.   The   first   inducer   planting   is   done   one   week   after   planting   the   source   population   in   the   first   vertical  long  ranges.  The  second  inducer  planting  is  done  14  days  after  planting  the  source  population  in   the  second  vertical  long  ranges.  The  third  planting  of  inducers  is  done  in  the  front  two  ranges  21  days   after   planting   the   source   population.  The   fourth   inducer   planting  is  done   in   one   or   two  of   the  horizontal   ranges  at  the  end  of  the  induction  nursery  28  days  after  planting  the  source  populations.  If  needed,  a   fifth  inducer  planting  can  be  done  in  the  second  horizontal  range  after  another  week.  In  this  design,   150   source  populations  can  be  accommodated  per  hectare  along  with  necessary  numbers  of  inducer  plants.               Figure  2.  Haploid  induction  nursery  in  CIMMYT  Agua  Fría  experimental  station,  Mexico.                   16                                                                            H1,   H2,   H3,   H4,   and   H5:   First,   second,   third,   fourth,   and   fifth   plantings   of   the   haploid  inducer,  respectively   P1,  P6:  Early  maturing  source  populations   P2,  P5:  Medium  maturity  source   populations   P3,  P4:  Late  maturing  source  populations     Figure   3.   Typical   design   of   a   haploid   induction  nursery  at  CIMMYT.         Management  of  induction  nursery   Since  the  isolation  block  is  planted  very  early  compared  to  other  maize  plantings,   seeds  and  seedlings   may  be  prone  to  fungal  or  insect  attacks  depending  on  the  environment  and  local  biotic  stress  pressure.   In  such  cases,  seed  treatment  will  aid  in  combating  fungal  pathogens  and  insects  during  germination  and   early   seedling   stages.   Seeds   may   be   treated   with   a   mixture   of   fungicides   and   insecticides.   Gaucho,   a   systemic   insecticide   used   for   seed   treatment,   is   effective   for   combating   insect   attack   during   seedling   stages.   During   soil   preparation,   fertilizer   (75-­‐80-­‐60   NPK/ha),   pre-­‐emergent   herbicide   (Atrazine),   and   insecticide   (Lorsban   5G)   are   incorporated   into   the   soil.   Before   planting,   plots   are   irrigated.   After   germination,  plots  are  irrigated  based  on  the  soil  conditions.  A  second  application  of  fertilizer  (150-­‐80-­‐   17   60  NPK/ha)  may  be  done  after  40  days.  Paraquat  may  be  applied  when  plants  are  40  to  50  days  old  for   managing  the  weeds.   In  the  induction  nursery,  inducers  need  to  be  given  special  care,  as  they  could  be  weak  and  vulnerable  to   diseases  and  insects.  In  such  cases,  the  inducer  plants  may  be  sprayed  with  fungicides  and  insecticides.   Turcicum   leaf   blight,   rust,   tar   spot   complex,   Bipolaris   maydis,   and   ear   rots   are   common   diseases   affecting  maize  in  the  tropical  environments  of  Mexico.  These  diseases  can  be  effectively  controlled  by   application  of  Tilt  (Propicanazole-­‐0.5L/ha)  at  15-­‐day  intervals.     Armyworm  (Spodoptera  frugiperda)  larvae  may  also  cause  damage  in  the  induction  nursery,  depending   on  the  local  conditions.  This  can  be  controlled  by  the  insecticide  Palgus  (Spinetoram)  during  the  seedling   stage   and   Larsbon   3G   (Chlorpyrifos   ethyl)   during   later   stages   of   plant   growth.   Karate   Zeon   (Lambda   Cyalotrina)   can   also   be   applied   to   control   armyworm.   When   severe   infection   of   armyworm   occurs,   a   mixture  of  Larsbon,  Karate  Zeon,  and  Palgus  is  applied.  In  some  environments/locations,  greenhoppers,   which  are  carriers  for  corn  stunt  complex  (Spiroplasma,  Phytoplsama,  Raillophena),  need  to  be  managed   in   early   plantings.   During   ear   development,   Spodoptera   litura   may   cause   considerable   damage,   which   needs  to  be  effectively  managed.     Pollinations  in  the  induction  nursery   For   both   manual   and   open   pollinations,  detasseling,   i.e.,   removal   of   the   tassels   from   source   populations   immediately  after  they  appear  (to  minimize  pollen  contamination),  is  done.  In  an  isolation  nursery  with   open  pollinations,  the  ears  should  not  be  covered  with  shoot  bags.  Open  pollinations  can  be  aided  by   dispersing  the  pollen  using  a  hand  blower.   In   an   induction   nursery   with   manual   pollinations,   ears   should   be   covered   by   shoot   bags   before   silking   occurs.   Ear   shoot   may   be   neatly   cut   at   the   tip   one   day   before   conducting   manual   pollinations   to   aid   uniform  growth  of  silks.  Pollen  from  ~10  inducer  parents  is  bulked  and  adequately  applied  on  the  silks.   Each   ear   may   be   pollinated   twice,   if   needed,   on   two   consecutive   days   to   obtain   ears   with   good/complete   seed   sets.   During   development   and   drying,   the   ears   need   to   be   properly   protected   from   birds.     Handling  of  the  harvested  ears     Ears  from  each  of  the  source  populations  in  an  induction  nursery  can  be  harvested  independent  of  other   source   populations   when   all   the   ears   in   that   population   reach   physiological   maturity.   This   prevents   losses  due  to  ear  damage  by  pathogens  and  insect  pests.  All  the  ears  from  same  population  should  be   harvested  in  one  or  two  bigger  mesh  bags  that  are  clearly  labeled.  Harvested  ears  are  dipped  briefly  in   the  insecticide  deltametrina  (125  ml/200  lts  water)  to  control  insect  pests  and  are  then  dried  completely   in   sunlight   for   two   to   three   days.   Once   the   ears   are   dried,   they   are   shelled   and   seed   is   collected   in   a   labeled  mesh  bag  and  kept  in  a  cold  storage  room  until  ready  for  sorting.   Note:   Mention   of   specific   brand   names   of   commercial   chemicals   (including   fertilizers,   fungicides,   and  pesticides)  is  not  intended  as  an  official  endorsement  of  the  product  by  CIMMYT.  There  may  be   other  equal  or  better  products  available  in  the  market  for  achieving  the  same  task.         HIR  assessment  of  the  haploid  inducers   Assessment   of   the   haploid   inducer   lines   for   HIR   requires   suitable   testers   that   allow   unambiguous   identification   of   haploids   at   the   early   seedling   or   kernel   stage.   Most   commonly   used   testers   possess   recessive  genes  like  liguleless  and  glossy.  When  the  testcross  kernels  from  the  cross  liguleless  ×  inducer   or   glossy   ×   inducer   are   germinated,   only   haploid   seedlings   exhibit   the   glossy   or   liguleless   phenotypes.   These  assays  can  be  conducted  on  seedlings  (at  the  three-­‐to  four-­‐leaf  stage)  in  a  greenhouse.     18   It   is   also   possible   to   use   an   R1-­‐nj   anthocyanin   marker   system   for   assessment   of   HIR   in   inducer   lines   that   incorporate   R1-­‐nj.   For   this   system   to   be   effective,   testers   should   be   identified   that   do   not   possess   inhibitor  genes  and  that  express  R1-­‐nj  very  well  under  different  environments.  Inducers  can  be  crossed   to  such  testers  and  HIR  determined  based  on  R1-­‐nj  expression.         A       B     Figure   4.   Liguleless   phenotype   in   the   (A)   seedling   and   (B)   adult   plant   stages.   The   liguleless   phenotype   is   characterized   by  lack  of  the  ligule  and  the  auricle,  and  by  erect  leaves.     Maintenance  breeding  of  the  haploid  inducers   A   continuous   selection   and   testing   system   should   be   established   for   maintaining   the   high   haploid   induction   rate   as   the   genetic   factors   controlling   haploid   induction   undergo   segregation   distortion   and   are   selected   against   by   the   nature.   In   maintenance   breeding   of   the   inducer   lines,   the   greatest   emphasis   should   be   given   to   retaining   the   high   haploid   induction   rate   and   maintaining   the   anthocyanin   marker   system.  Importance  should  also  be  given  to  pollen  production  characteristics  and  vigor  of  the  plants.  Sib-­‐ mating  is  recommended  rather  than  selfing  to  maintain  the  vigor  of  the  inducer.       Inducer   plants   in   the   seed   multiplication   plot   are   scored   for   various   agronomic   and   phenotypic   traits   (e.g.,   pollen   production   ability,   plant   height,   vigor,   expression   of   purple   color   on   the   stem,   disease   resistance,   and   ear   traits).   Plants   with   the   best   per   se   performance   are   selected   and   tagged.   Pollen   is   collected  from  selected  plants  in  a  row  and  bulked,  and  the  bulked  pollen  is  used  for  sib-­‐mating  (within   the   same   row)   and   for   testcross   to   a   recessive   tester   suitable   for   HIR   assessment.   All   the   ears   harvested   from  a  row  are  shelled  and  kept  separately.     Testcross  seeds  from  each  row  of  the  inducer  multiplication  plot  are  planted  separately  in  a  greenhouse.   Seedlings   are   evaluated   for   the   recessive   trait   at   the   three-­‐   to   four-­‐leaf   stage   and   HIR   is   assessed.   Inducer  rows  with  high  HIR  are  identified,  and  seeds  from  each  ear  of  that  row  are  scored  for  intensity   and   proper   expression   of   purple   coloration   on   the   embryo   and   endosperm.   Ears   from   plants   with   the   best   agronomic   scores   and   with   good   expression   of   color   marker   on   the   embryo   and   endosperm   are   selected.  These  ears  are  used  for  maintaining  the  inducer  stocks.       19   4.  Maternal  Haploid  Detection  using  Anthocyanin  Markers   Vijay  Chaikam  and  BM  Prasanna   Introduction   Haploid  plants   can  be  distinguished  from  diploid  plants  by  characteristics  like   erect  leaves,  poor  vigor,   and   sterility.   These   characteristics   can   only   be   observed   after   sufficient   growth   of   haploid   plants.   Distinguishing  haploids  from  diploids  at  seed  level  offers  many  advantages  like  saving  costs  involved  in   artificial   chromosomal   doubling   and   saving   greenhouse   and   field   space   and   labor.   So   identification   of   haploids   at   seed   level   is   critical   for   adapting   DH   technology   on   a   commercial   scale.   A   commercially   usable,   ingenious   phenotypic   marker   system   based   on   anthocyanin   coloration   was   identified   in   the   1960s  (Nanda  and  Chase,  1966;  Greenblatt  and  Bock,  1967)  to  distinguish  haploids  from  diploid  at  the   seed  stage.  Integration  of  anthocyanin  markers  in  haploid  inducer  lines  facilitated  haploid  identification   not  only  at  the  seed  level  but  also  at  different  stages  of  plant  growth.     Identification  of  haploid  kernels  using  an  R1-­‐nj  marker  system   R1-­‐nj   (R1-­‐Navajo),   a   dominant   variant   allele   of   the   R1   locus,   is   now   widely   used   for   the   screening   of   haploids   in   kernels.   R1-­‐nj   in   combination   with   other   dominant   genes   in   the   anthocyanin   synthesis   pathway   (A1,   A2,   Bz1,   Bz2,   C1,   and   C2)   causes   deep   pigmentation   of   the   aleurone   (endosperm   tissue)   in   the   crown   (top)   region   of   the   kernel   (Coe,   1994).   In   addition,   it   conditions   purple   pigmentation   in   the   scutellum  (embryo  tissue).  This  phenotype  is  called  the  Navajo  kernel  phenotype.     In   haploid   inducer   lines   that   are   commonly   used   now,   the   R1-­‐nj   allele   is   integrated   along   with   other   genes  necessary  for  anthocyanin  biosynthesis.  Most  of  the  maize  germplasm  used  in  breeding  programs   does  not  have  R1-­‐nj  allele  or  anthocyanin  biosynthetic  genes  that  confer  purple/red  pigmentation  in  the   kernel/plant  tissues.  When  the  inducer  lines  are  crossed  (as  male  parent)  to  the  source  germplasm  (as   female  parent)  not  having  the  anthocyanin  color  markers,  all  the  resulting  hybrid  kernels  are  expected   to  express  the  Navajo  phenotype  in  the  endosperm  and  in  the  scutellum  (embryo)  as  R1-­‐nj  is  dominant   over  the  colorless  r1  allele.  Thus,  the  differential  expression  of  R1-­‐nj  facilitates  identification  of  maternal   haploids   from   the   diploid   kernels.   When   haploid   inducers   with   a   high   haploid   induction   rate   (HIR)   are   used  in  the  induction  cross,  maternal  haploids  usually  occur  at  a  frequency  of  6–10%.       In  practice,  three  types  of  kernels  may  be  obtained  from  the  induction  cross:   (1) Normal   diploid   or   hybrid   kernels   with   purple   coloration   on   the   endosperm   (aleurone)   and   the   embryo  (scutellum);     (2) Haploid  kernels  with  purple  endosperm  but  no  coloration  on  the  embryo;  and     (3) Kernels   without   purple   coloration   on  the   embryo   and   endosperm,   which   could   be   due   to  pollen   contamination.     20        Figure   1.   Illustration   of   maternal   haploid   induction   and   kernel   types   obtained   through   a   typical  induction  cross.     Even  though  the  R1-­‐nj  marker  system  offers  an  efficient  way  to  identify  haploids,  R1-­‐nj  expression  could   be  highly  influenced  by  the  genetic  background  of  the  female  parent.  The  Navajo  crown  pigmentation   might  vary  from  a  small  spot  (at  the  silk  attachment  region  of  the  kernel)  to  covering  the  entire  aleurone   (except   the   base).   Also,   the   intensity   of   color   on   the   aleurone   may   vary   from   very   pale   to   deep.   Expression  of  color  on  the  scutellum  may  also  vary  from  pale  to  deep  (Figure  2).             On  the  scutellum  (embryo)  expression         On  the  endosperm  expression     Figure  2.  Variation  in  R1-­‐nj  expression  on  the  embryo  and  endosperm.       The  variation  in  R1-­‐nj  expression  can  lead  to  different  outcomes  as  follows:       (1) Whole  endosperm  and  all  the  embryo  tissues  become  colored:  haploid  identification  is  easy.   (2) Good  coloration  on  the  crown  of  the  endosperm  and  scutellum:  haploid  identification  is  easy.       21   (3) Only  a  purple  spot  on  the  crown  of  the  endosperm  and  slight  expression  on  the  embryo:  haploid   identification   is   possible,   but   high   false   positives   could   happen   due   to   difficulties   in   haploid   identification.   (4) Completely  inhibited  on  both  endosperm  and  embryo:  impossible  to  identify  haploids.   (5) Completely   inhibited   on   the   endosperm   but   embryo   tissue   is   marked   to   some   extent:   in   such   cases,  all  the  kernels  with  colored  embryos  can  be  considered  diploid.  But  it  is  not  possible  to   distinguish  haploids  and  pollen  contaminants  from  this  category.       Limitations  of  R1-­‐nj  system   (1) When   the   source   populations   contain   dominant   anthocyanin   inhibitor   genes   such   as  C1-­‐I,   which   are   common   in   flint   maize   (Röber   et   al.,   2005),   R1-­‐nj   color   marker   expression   is   completely   suppressed  and  haploid  kernel  identification  is  almost  impossible.  CIMMYT’s  elite  germplasm  is   currently  being  surveyed  to  determine  in  what  proportion  the  seed  color  marker  will  function,   permitting   efficient   haploid   seed   detection.   Currently,   it   appears   that   R1-­‐nj  color  expression  is   inhibited  in  only  about  8%  of  crosses  of  haploid  inducers  with  diverse  source  populations.   (2) When  F1  or  F2  populations  are  used  as  source  materials  and  if  only  one  parent  has  an  inhibitor   gene,  kernels  will  be  segregating  for  Navajo  phenotype.  In  such  cases,  one  may  not  be  able  to   identify  all  the  haploid  kernels  efficiently  and  could  potentially  lose  half  to  three-­‐fourths  of  the   haploids.     (3) The   accuracy   and   speed   of   haploid   identification   depends   on   trained   staff   with   good   understanding  of  haploid  detection  through  the  color  expression  on  endosperm  and  embryo.     (4) Automation  of  haploid  identification  is  difficult,  if  not  impossible,  using  this  system.   (5) Moisture   of   kernels   at   the   time   of   harvest   could   potentially   affect   the   intensity   of   color   expression  (Rotarenco  et  al.,  2010).     Purple  root  and  purple  stem  markers  for  haploid  identification   In   view   of   the   above-­‐mentioned   limitations   of   the   R1-­‐nj   color   marker   system   in   haploid   detection,   some   researchers  have  explored  the  possibility  of  additional  color  markers,  especially  those  expressed  in  root   and  stem,  for  reliable  identification  of  maternal  haploids  (Rotarenco  et  al.,  2010).  Two  such  genes  that   can   impart   purple   or   red   color   to   the   plant   tissues   are   Pl1   (Purple1),   which   conditions   sunlight-­‐ independent   purple   pigmentation   in   plant   tissues,   and   B1   (Booster1),   which   conditions   sunlight-­‐ dependent  purple  pigmentation  in  most  of  the  above-­‐ground  plant  tissues  (Coe,  1994).     The  B1  and  Pl1  genes  can  be  integrated  into  the  inducer  lines  along  with  the  R1-­‐nj  marker  system.  When   Navajo   coloration   is   not   expressed   on   the   kernels,   haploids   can   be   identified   based   on   the   seedling   root   color  or  stem  color  in  the  field.  When  such  an  inducer  is  crossed  with  source  material,  diploid  plants  will   have  purple  roots  and  stems,  while  the  putative  doubled  haploid  plants  will  not  express  such  coloration.     Some  temperate  inducer  lines  like  MHI  (Eder  and  Chalyk,  2002)  and  Procera  Haploid  Inducer  (Rotarenco   et  al.,  2010)  combine  R1-­‐nj  with  B1  and  Pl1  genes  for  more  effective  identification  of  haploids.                         22                        Figure  3.  (A)  Purple  color  expression  in  the  roots;     (B)  purple  stem  color  in  diploid  plants  and  normal     green  stem  in  putative  doubled  haploid  plants.   B   A       Limitations  of  B1  and  Pl1  system:     (1) Many  source  materials  contain  B1  and  Pl1  genes.  In  such  source  populations,  haploid  plants  also   express  coloration  in  the  root  and  stem,  making  it  almost  impossible  to  reliably  identify  haploid   plants.   (2) Expression  of  the  B1  and  Pl1  genes  are  affected  by  plant  growth  conditions,  especially  sunlight   and   temperature.   It   was   observed   that   purple   pigments   accumulate   best   under   low   temperatures.     Further  possibilities   Some   research   teams   are   exploring   novel   marker   systems   that   can   potentially   facilitate   automated   haploid   detection   with   minimal   false   positives.  Rotarenco  et  al.  (2007)  proposed   haploid   identification   based   on   kernel   oil   content,   determination   of   which   can   be   potentially   automated   using   nuclear   magnetic   resonance–based   techniques.   Li   et   al.   (2009)   recently   developed   CAUHOI,   a   Stock6-­‐derived   inducer  (with  ~2%  HIR  and  high  kernel  oil  content  (78  g  kg−1))  that  allows  identification  of  haploids  based   on  both  lack  of  R1-­‐nj  conferred  scutellum  coloration  and  low  embryo  oil  content.  This  novel  approach   looks  promising,  but  its  reliability  and  applicability  for  high-­‐throughput  DH  production  in  tropical  genetic   backgrounds  remains  to  be  investigated.     References   Coe   EH   (1994)   Anthocyanin   genetics.   In:   M   Freeling,   V   Walbot   (eds)   The   maize   handbook.   Springer-­‐Verlag,   New   York,  pp.  279–281.   Eder  J,  Chalyk  ST  (2002)  In  vivo  haploid  induction  in  maize.  Theor.  Appl.  Genet.  104:  703–708.   Greenblatt   IM,   Bock   M   (1967)   A   commercially   desirable   procedure   for   detection   of   monoploids   in   maize.   J.   Hered.   58:  9–13.   Li   L,   Xu   X,   Jin   W,   Chen   S   (2009)   Morphological   and   molecular   evidences   for   DNA   introgression   in   haploid   induction   via  a  high  oil  inducer  CAUHOI  in  maize.  Planta  230:  367–376.   Nanda  DK,  Chase  SS  (1966)  An  embryo  marker  for  detecting  monoploids  of  maize  (Zea  mays  L.).  Crop  Sci.  6:  213– 215.   Rotarenco  VA,  Kirtoca  IH,  Jacota  AG  (2007)  Possibility  to  identify  kernels  with  haploid  embryo  by  oil  content.  Maize   Genet.  Coop.  Newslett.  81:  11.   Rotarenco  V,  DG,  State  D,  Fuia  S.  (2010)  New  inducers  of  maternal  haploids  in  maize.  Maize  Genet.  Coop.  Newslett.   84:    1–7.   Röber   FK,   Gordillo   GA,   Geiger,   HH   (2005)   In   vivo   haploid   induction   in   maize   –   performance   of   new   inducers   and   significance  of  doubled  haploid  lines  in  hybrid  breeding.  Maydica  50:  275–283.       23   5.  Chromosome  Doubling  of  Maternal  Haploids     Vijay  Chaikam  and  George  Mahuku     Introduction   Diploid   plants   contain   two   copies   of   each   chromosome   in   their   cells,   of   which   one   copy   is   received   from   the  male  parent  and  the  other  from  the  female  parent.  In  the  reproductive  structures  (tassel  and  ear  in   maize),   haploid   male   (pollen   grain)   and   female   (embryo   sac)   gametophytes   are   results   of   meiotic   cell   divisions  which  involve  pairing  of  homologous  chromosomes  and  recombination.     Haploid  plants  contain  only  one  copy  of  each  chromosome  in  their  cells.  Haploids  derived  by  maternal  in   vivo   induction   contain   chromosomes   only   from   the   female   parent.   In   the   reproductive   structures   of   haploid   plants,   meiotic   cell   divisions   cannot   proceed   as   homologous   chromosomal   pairs   cannot   form,   resulting  in  non-­‐production  of  male  and  female  gametophytes  and  gametic  cells.  So  haploid  plants  are   usually  sterile.  The  purpose  of  chromosomal  doubling  is,  therefore,  to  achieve  fertility  in  haploid  plants   by   generating   a   doubled   haploid   (2n)   plant   out   of   a   haploid   (n),   so   that   these   plants   can   be   selfed   to   derive  doubled  haploid  (DH)  lines.     Mechanism  of  chromosomal  doubling   Spontaneous   chromosomal   doubling   occurs   at   low   frequency,   resulting   in   fertility   of   some   haploid   plants.   The   frequency   of   spontaneous   doubling   is   dependent   on   the   genotype   of   the   source   population.   To   achieve   consistent   and   high   frequency   of   chromosomal   doubling,   haploid   plants   are   treated   with   chemicals  called  mitotic  inhibitors.  These  chemicals  alter  the  regular  mitosis  in  such  a  way  that  only  a   single  cell  with  double  the  number  of  chromosomes  results  after  mitosis.  A  commonly  used  chemical  is   colchicine,   which   is   a   water-­‐soluble   alkaloid   produced   from   the   bulbs   of   Colchicum   autumnale.   In   the   presence   of   colchicine,   replication   of   chromosomes   occurs   normally   in   interphase.   Colchicine   binds   to   tubulins   and   prevents   the   formation   of   spindle   microtubules   during   the   metaphase   stage   of   mitosis.   During  anaphase,  two  sister  chromatids  in  a  replicated  chromosomes  are  separate  but  cannot  move  to   opposite   poles   of   the   cell   and   instead   stay   at   the   center   of   the   cell.   In   telophase,   a   nuclear   membrane   is   formed   around   the   unseparated   chromosomes.   So   after   mitosis   a   cell   with   double   the   number   of   chromosomes  results.       In   plants,   all   the   above-­‐ground   organs   including   reproductive   structures   arise   from   the   shoot   apical   meristem   (SAM).   The   SAM   contains   meristematic   cells,   which   divide   and   differentiate   into   organ   primordia.  To  achieve  complete  fertility  in  reproductive  tissues  of  haploid  plants,  chromosomal  doubling   of   meristematic   cells   should   occur   before   they   differentiate   into   reproductive   organs.   Therefore,   exposing  very  young  seedlings  (three  to  five  days  after  sowing)  to  mitotic  inhibitors  is  recommended.       Facilities  required  for  chromosomal  doubling   For  operational  convenience  and  safety  of  the  workers,  chromosomal  doubling  work  can  be  segregated   into  three  work  areas:   (1) A  germination  room  where  the  seeds  are  processed  and  germinated  and  seedlings  are  prepared   for  colchicine  treatment.  This  germination  room  is  equipped  with  work  benches  for  workers  to   process  the  seed  and  to  prepare  the  seedlings  for  treatment.  This  room  should  also  be  equipped   with  an  incubator  for  seed  germination.   (2) A   colchicine   treatment   lab   where   chemicals   are   stored,   colchicine   solution   is   prepared,   and   seedlings   are   treated.   This   room   should   be   equipped   with   a   refrigerator   to   store   chemicals,   a   fume  hood  to  prepare  solutions,  colchicine  treatment  tanks,  and  an  exhaust.     (3) A  room  where  colchicine  waste  is  stored  until  processed  through  chemical  waste  management.   This  room  should  be  equipped  with  an  exhaust  fan.       24   Supplies  needed  for  seed  germination  and  seedling  processing:   Germination  paper   Plastic  tubs   Seed  spreaders     Scalpels  and  blades         Supplies  needed  for  colchicine  treatment  and  waste  management:   Refrigerator     Weighing  balance   Measuring  cylinders:  5,000ml;  1,000ml;  500  ml;  100ml   Pipettes:  1,000  microliters;  200  microliters;  100  microliters   Magnetic  stirrer  with  magnets   Containers  to  prepare  solutions   Metallic  tanks  for  treatment   Metallic  tanks  or  polypropylene  containers  to  collect  waste   Protective  clothes,  gloves,  and  masks     Chemical  supplies  needed:   Colchicine     DMSO   Bleach       Steps  in  chromosome  doubling     Seed  germination:       § Germination  paper  is  marked  by  a  cut  at  one  of  the  corners  and  moistened  with  0.05%  bleach   solution  to  prevent  fungal  growth.     § Two  germination  papers  are  spread  on   top  of  one  another,  aligning  the  cut  ends,  and  seeds  are   spread  evenly  using  a  spreader  (Figure  1).     § Seeds   are   placed   with   the   embryo   side   facing   down   and   the   radicle   emergence   side   placed   towards  the  cut  end  of  the  paper  (Figure  2).  Then  seeds  are  covered  with  one  more  paper  on   the  top  aligning  the  cut  ends  (Figure  3).     § These  three  layers  of  germination  paper  with  seeds  are  folded  tightly  into  a  bundle  and  tied  with   rubber  bands  at  both  ends  (Figures  3,  4,  and  5).     § Then  bundles  of  seed  from  the  same  population  are  kept  in  a  mesh  bag  vertically  with  cut  ends   facing   down   and   placed   in   plastic   containers   with   bleach   solution   (Figure   6   and   7).   Bleach   solution  in  the  tub  helps  to  prevent  fungal  growth  and  maintains  humidity.     § Plastic  tubs  are  placed  in  the  incubation  chamber  (Figure  8),  where  temperature  is  maintained   around  25  to  28 oC.  Seeds  are  allowed  to  germinate  for  72  hours.       Preparation  of  seedlings:   § Three   days   after   incubation,   the   plastic   containers   with   seed   bundles   are   removed   from   the   incubation  chamber.  The  bundles  are  spread  out  on  a  work  table  (Figures  9  and  10).     § Seedlings   with   a   root   length   of   3–5   cm   and   coleoptile   length   of   about   2   cm   are   ideal   for   colchicine  treatment.  Before  colchicine  treatment,  root  and  shoot  tissues  are  cut  at  about  2  cm   and  1  cm  from  the  tip,  respectively,  using  a  sterile  blade  fixed  to  a  scalpel  blade  holder  (Figure   11).  Blades  are  sterilized  by  heating  them  over  an  alcohol  lamp.  Cutting  the  root  tips  aids  in  easy   handling  of  seedlings  during  transplanting,  and  cutting  shoot  tips  enhances  the  exposure  of  the   SAM  to  colchicine  treatment.       25   § § § Cut  seedlings  that  belong  to  the  same  population  are  kept  in  a  mesh  bag  (Figure  12).     Mesh   bags   with   seedlings   are   kept   in   water   for   a   few   hours   before   transferring   to   the   colchicine   tank  (Figure  13).     The   germination   paper   with   very   small   seedlings   and   non-­‐germinated   seeds   can   be   bundled   again  and  kept  in  a  growth  chamber  for  one  more  day.  The  same  procedure  of  seedling  cutting   can  be  followed  for  the  next  two  days.                       Fig.%6.%%Bundles*in*a*mesh*   Fig.%1.%%Spreading*of*seeds*on*germina/on*paper*                 Fig.%2.%%Aligning*the*seed*with*the*radicle*side*facing* the*cut*end*of*the*paper,*and*embryo*side*facing*down* Fig.%7.%Bundles*kept*in*a*plas/c*tub*with*bleach*   solu/on*                 Fig.%3.%Covering*the*seeds*with*a*germina/on*paper* Fig.%8.%%Plas/c*tubs*(with*seeds)*in*an*incubator* Fig.%4.%Bundling*the*germina/on*papers*with*seeds* Fig.%9.%%Opening*the*bundles* Fig.%5.%%Bundle*formed*a=er*rolling*   Fig.%10.%Germinated*haploid*seedlings* 26   Fig.%11.%%CuDng*the*root*and*shoot*of*germinated* seedlings* Fig.%12.%Cut*seedlings*in*a*mesh*bag* Fig.%13.%Mesh*bags*with* seedlings*kept*in*water*un/l* treated*with*colchicine*       Colchicine  treatment:   Since  colchicine  is  very  toxic  and  carcinogenic  to  human  beings,  care  must  be  taken  to  avoid  exposure  to   it   by   taking   necessary   precautions.   Seedlings   can   be   treated   with   colchicine   in   the   dark   in   specialized   tanks  that  allow  workers  to  avoid  direct  contact  with  colchicine.  These  tanks  are  made  up  of  stainless   steel   to   avoid   corrosion   (Figure   14).   The   lid   of   the   tank   has   an   opening,   which   can   be   connected   to   a   running  water  supply  or  to  a  container  with  colchicine  solution  (Figure  15A).  The  tank  is  equipped  with   an   outlet   at   the   bottom   center   to   allow   drainage   of   the   spent   liquid   (Figure   15B).   The   tank   is   placed   at   a   height   on   a   stand   with   iron   legs.   This   permits   placement   of   containers   under   the   tank   to   collect   the   waste.  The  base  has  wheels  for  easy  movability  (Figure  14).     The  volume  of  the  colchicine  solution  required  is  estimated  by  placing  the  cut  seedlings  in  the  treatment   tank   (Figure   16)   and   pumping   water   gently   until   all   the   seedlings   are   immersed.   Water   is   emptied   into   a   container   from   the   bottom   opening   (Figure   15B),   and   the   collected   water   is   measured.   This   volume   represents  the  amount  of  colchicine  solution  that  needs  to  be  prepared.       A  solution  with  0.04%  colchicine  and  0.5%  DMSO  is  used  for  chromosomal  doubling.  Colchicine  powder   is  weighed  in  a  fume  hood  and  dissolved  in  water  in  a  plastic  tank  wrapped  with  aluminum  foil.  With  the   aid   of   a   magnetic   stirrer,   colchicine   powder   is   dissolved   in   water   along   with   DMSO   for   two   to   three   hours.  The  person  preparing  this  solution  should  wear  overalls,  gloves,  and  protective  facial  cover.  The   container   in   which   the   colchicine   solution   is   prepared   has   an   outlet   at   the   bottom   which   can   be   connected   to   a   pipe   and   an   automatic   dispenser   pump   to   dispense   colchicine   solution   into   the   treatment  tank  automatically.  Seedlings  are  kept  in  the  colchicine  tank  for  12  hours.  For  convenience,   the  treatment  can  be  started  at  8   P.M.  and  stopped  at  8   A.M.  The  spent  colchicine  solution  is  collected   into   plastic   containers   by   opening   the   outlet   at   the   bottom   of   the   treatment   tank.   Seedlings   are   washed   at   least   three   times   by   pumping   distilled   water   into   the   tank.   Waste   is   collected   in   big   plastic   containers   and  stored  in  a  separate  room  along  with  spent  colchicine  solution  until  processed  by  chemical  waste   management.       27   Fig.%14.%Colchicine*tank*and*pumping*of* colchicine*solu/on*from*container*to*the*tank* A* B* Fig.%15.%(A)*Lid*of*the*colchicine*tank*with*connec/on* to*water*supply*or*colchicine*container;*(B)*Collec/on* of*colchicine*waste*from*the*boHom*outlet.* Fig.%16.%Cut*seedlings*in* mesh*bags*in*colchicine*tank*     Seedling  transplanting  and  greenhouse  care:       § Seedlings   taken   out   of   the   treatment   tank   are   immediately   transplanted   into   Styrofoam   trays   containing  promix  (peat  moss)(Figure  17,18  and  19).  Seedlings  should  be  handled  very  carefully   as   they   become   brittle   after   colchicine   treatments.   Seedlings   with   long   hypocotyls   are   more   susceptible  to  damage.     § Seedlings   are   maintained   in   the   Styrofoam   trays   for   three   weeks   in   a   greenhouse   where   temperature   is   maintained   at   28–30°C.   Seedlings   are   irrigated   gently   from   the   top   every   evening.  For  the  first  irrigation,  water  is  used.  From  the  second  irrigation,  Hakaphos  (13-­‐40-­‐13   NPK  and  micronutrients)  is  applied,  which  helps  in  root  growth  and  seedling  establishment.     § To   prevent   fungal   attacks,   the   fungicide   Tecto   (Thiabendazol)   is   applied   every   third   day.   Gaucho   (Imidacloprid),   which   is   a   systemic   insecticide,   is   applied   once   a   week   before   transplanting   to   prevent  insect  damage.  Hakaphos  and  Gaucho  can  be  combined  for  application.     Success  rates  in  different  steps  of  chromosomal  doubling   At   CIMMYT   a   germination   percentage   of   85–90%   is   commonly   achieved   among   the   putative   haploid   kernels.   During   chromosomal   doubling,   a   considerable   number   of   seedlings   could   be   lost   due   to   the   toxic  effects  of  colchicine,  which  is  dependent  on  the  genetic  background  of  the  source  material  and  the   procedure   followed   for   application.   Only   40–80%   of   treated   putative   haploid   seedlings   will   be   established   in   the   field.   Among   the   established   plants,   10–30%   false   positives   (diploids)   may   be   noticed,   and  these  should  be  spotted  and  removed  before  flowering.  Among  the  remaining  true  haploids,  0–40%   of  plants  produce  both  pollen  and  silks  so  that  successful  pollinations  can  be  conducted.  Pollen  fertility   is   again   dependent   on   the   genotype   of   the   source   germplasm.   Only   30–50%   of   the   pollinated   plants   usually  produce  seed.         28                      Fig.%17.%Transplan/ng*of*treated*seedlings* Fig.%18.%Transplanted*seedlings*a=er*one*week**                     Fig.%19.%Transplanted*   seedlings*a=er*3*weeks*     Colchicine  toxicity  and  precautions  to  be  taken     Colchicine  toxicity:       At  concentrations  of  0.1–1  g/ml,  colchicine  can  cause  the  mitotic  arrest  of  dividing  cells  (both  plant  and   animal   cells)   at   metaphase   by   interfering   with   microtubule   organization,   in   particular   those   of   the   mitotic   spindle.   Colchicine   is   fatal   if   swallowed,   inhaled,   or   absorbed   through   skin.   Exposure   to   colchicine  causes  respiratory  tract  irritation,  skin  irritation,  eye  irritation,  and  serious  eye  damage,  and   can  be  carcinogenic.       Precautions  to  be  taken:   To   avoid   exposure   of   workers   to   colchicine,   a   separate   room   should   be   assigned   for   storing   colchicine  powder  and  for  colchicine  solution  preparation  and  treatment.     The  laboratory  should  be  equipped  with  a  fume  hood  to  handle  colchicine  and  an  exhaust  fan  to   remove  chemical  odors  and  vapors.     A  cart  should  be  specifically  assigned  to  the  lab  to  move  solutions.   Colchicine  should  be  stored  in  a  refrigerator,  which  should  be  securely  locked.     The   process   of   colchicine   treatment   should   be   automated   as   much   as   possible   to   reduce   exposure.     Material  safety  data  sheets  should  be  easily  accessible  in  the  lab  for  all  the  chemicals  used.     Persons   working   with   colchicine   should   wear   protective   gloves,   respiratory   protection,   eye   protection,  and  whole  body  cover.     Workers  should  wash  hands  thoroughly  every  time  after  handling  colchicine.     Colchicine  waste  should  be  stored  in  a  secluded  room,  which  should  be  locked.     Waste  should  be  disposed  of  by  a  well-­‐trained  hazardous  waste  disposal  team.     In   case   of   exposure,   the   exposed   part   should   be   rinsed   cautiously   with   water   for   several   minutes.   Immediately   call   a   poison   center   or   a   doctor   with   experience   in   occupational   safety.   Material  safety  documents  should  be  presented  to  the  doctor.   29   6.    Putative  DH  Seedlings:  From  the  Lab  to  the  Field   George  Mahuku   Management  of  haploid  seedlings  is  crucial  for  the  success  of  doubled  haploid  (DH)  line  development.   There   are   two   critical   steps:   (1)   managing   colchicine-­‐treated   D0   seedlings   and   reestablishing   these   seedlings  under  greenhouse  conditions;  and  (2)  managing  putative  DH  plants  under  field  conditions.  At   each   of   these   steps,   loss   of   putative   DH   plants   can   occur,   affecting   the   success   rate   of   achieving   DH   lines.   This   chapter   addresses   some   of   the   pertinent   issues   (handling   of   DH   seedlings,   availability   of   suitable  facilities  to  raise  DH  plants  for  maintenance,  and  seed  multiplication  and  optimizing  handling  of   putative  DH  lines  under  greenhouse  and  field  conditions)  that  are  required  for  optimal  recovery  of  DH   lines.       Managing  D0  Seedlings   After  treating  the  haploid  seedlings  with  colchicine,  drain  the  solution  from  the  treatment  container  and   collect  it  in  specially  designated  residual  waste  containers.  Rinse   the  treated  seedlings  with  tap  water  at   least  thrice  to  remove  residual  colchicine.  Rinsing  is  performed  by  filling  the  treatment  container  with   tap   water   until   all   seedlings   are   fully   submerged,   followed   by   draining   and   collection   of   the   waste   solution   into   special   toxic   waste   containers   for   proper   disposal.   A   final   wash/rinse   should   be   conducted   using  100  ppm  of  chlorox,  which  acts  as  a  disinfectant  and  minimizes  contamination  of  seeds  by  bacteria   and  fungi.  After  this,  the  seedlings  are  ready  for  transplanting  in  the  greenhouse.       Note:  All  the  colchicine  waste  must  be  collected  in  specially  designated  and  clearly  labeled  container(s)   and  disposed  of  by  an  authorized  company/agency.  Please  follow  the  relevant  rules  and  regulations  for   safe   disposal   of   dangerous   chemical   wastes,   as   applicable   in   your   specific   institution   and   country.   Under   no  circumstances  should  these  toxic  residuals  be  disposed  of  through  the  common  sink!     Handling  treated  seedlings:  Take  utmost  care  while  handling  the  seedlings,  especially  after  treatment.   The   coleoptile   is   a   very   vulnerable   tissue   and   could   easily   break   if   not   handled   properly.   Therefore,   handle   the   seedlings   by   holding   the   kernel,   and   do   not   touch   the   root   or   coleoptile   to   avoid   possible   breakage.   Damage   to   the   tissue   during   preparation   of   seedlings   for   treatment   or   during   the   subsequent   handling  of  treated  seedlings  can  lead  to  necrotic  shoot  tissue  and  subsequent  seedling  death.     Transplanting  materials:   § Colchicine   treated   seedlings:   Each   population   should   be   clearly   labeled   to   avoid   misidentification  while  transplanting.   § Tray   with   sterile   distilled   water:   Seedlings   should   be   transported   to   the   greenhouse   in   a   tray   with  water  to  avoid  dehydration.         § Lab   coats,   gloves,   and   work   suits:   Remember   that   seedlings   were   treated   with   colchicine;   so   take  adequate  operational  health  and  safety  measures  while  handling  the  treated  seedlings.   § Greenhouse   or   screen   house:   This   should   have   controlled   conditions   (temperature,   light,   and   humidity).   § Labeling   stacks:   These   are   required   for   proper   identification   of   the   populations   being   transplanted.   § Masking  tape  and  permanent  markers:  For  recording  all  necessary  information.           30   § Jiffy  pots  (in  trays)  with  potting  medium:  Use  greenhouse  soil  (peat  moss)  if  possible,  but  any   soil  can  be  used  as  long  as  it  is  properly  sterilized.  Use  soil  with  high  organic  matter  content  and   avoid  soil  with  a  high  clay  content,  if  possible.  In  CIMMYT,  we  use  either  promix  or  premier  peat   moss.  The  promix  is  more  compressed  and  is  ready  to  use,  while  the  peat  moss  should  be  mixed   with  at  least  10%  perlite  (agrolita).   Greenhouse  soil:  Use  sterile  soil  with  high  organic  matter.     §   Transplanting  procedure:     Take   the   treated   and   washed   seedlings   to   the   greenhouse   for   transplanting,   making   sure   that   they  are  in  a  tray  with  water  to  avoid  dehydration.     Different   types   of   pots   can   be   used   (see   below)   for   transplanting;   the   choice   depends   on   the   budget,  availability  of  materials,  and  transplanting  methods.     First  fill  each  pot  about  halfway  with  soil  (see  below  for  the  type  of  soil).    Then  carefully  place   the  seedling  (holding  the  attached  seed  rather  than  the  shoot  or  root)  onto  the  soil  and  hold  it   while   filling   more   soil   around   it   until   the   pot   is   well   filled   and   only   the   tip   of   the   coleoptile   is   visible   (Figure   1).   Gently   push   to   make   soil   compact   and   prevent   soil   run-­‐off   during   watering.   Leaving   a   large   part   of   the   elongated   coleoptile   outside   will   increase   the   chance   of   damage   and   will  reduce  the  number  of  plants  that  can  successfully  be  transplanted  in  the  field.   Care  should  be  taken  to  avoid  or  minimize  breaking  the  coleoptile,  as  seedlings  are  very  fragile   and  break  easily  after  treatment.  While  transplanting,  make  sure  that  the  size  of  the  coleoptile   outside   the   soil   is   less   than   2   cm,   as   any   length   greater   than   this   may   increase   the   risk   of   coleoptile  breakage  and  thereby  affect  the  number  of  successfully  established  DH  plants.         Note:  Protective  gloves  should  be  worn  all  the  time  when  handling  colchicine-­‐treated  seedlings!       Figure  1.  Transplanting  treated  putative  DH  (D0)  seedlings  into  jiffy  pots  in  the  greenhouse.     Make  sure  that  only  the  tip  of  the  coleoptile  is  visible  so  as  to  avoid  damage  and  loss  of  plants.     [Photos:  G.  Mahuku]     Managing  the  D0  greenhouse     Types  of  pots  for  greenhouse  transplanting:  Various  types  of  pots  can  be  used:  (1)  typical  plastic  pots   (approximately  5×5×8  cm)  commonly  used  for  greenhouse  experiments  and  horticulture  that  are  made   of   durable   plastic   and   can   be   reused;   (2)   Styrofoam   cups,   more   commonly   used   for   coffee   or   tea,   which   are   very   cheap   but   less   durable;   or   (3)   pots   made   of   biodegradable   material   that   decays   in   the   soil   allowing  transplanting  of  seedlings  along  with  pots  and  thereby  enabling  the  use  of  a  planting  machine   (Figure  2).  All  pots  must  have  perforated  bottoms  to  allow  drainage  of  excess  water.       31      A   B   C     Figure   2.   Transplanting   treated   seedlings   into   pots:   (A)   Styrofoam   cups   normally   used   for   coffee,   (B)   jiffy   pots,   and   (C)   plastic   pots.   The   pots   are   filled   with   sterile   soil   containing   high   organic   matter   content.   Pots   are   placed   in   special   containers   for   easy   handling   and   management.  [Photos  A  &  B:  V.  Prigge,  photo  C:  G.  Mahuku]     Maintenance   of   seedlings   in   the   greenhouse:   Place   the   potted   seedlings   in   special   containers   in   the   greenhouse   (Figure   2)   for   approximately   10   days   to  two   weeks   so   that   they   recover   from   the   colchicine   treatment  and  grow  to  the   three-­‐  or  four-­‐leaf  stage.  It  is  important  that  proper  fertilization  and  control   of  insects  and  diseases  is  undertaken  so  that  the  treated  plants  recover  well  and  are  vigorous.  During   this  period,  maintain  the  following  conditions  favorable  for  seedling  growth:     Keep  the  soil  moist  but  avoid  excess  water.  Watering  can  be  done  once  per  day  or  as  needed.   Apply   the   required   dose   of   fertilizer   in   a   soluble   form   with   the   irrigation   water.   It   is   also   advisable  to  use  a  fertilizer  with  high  phosphorus  content  to  stimulate  root  growth.       Three  days  after  transplanting,  fertilize  plants  with  Triple  20  –  this  may  be  done  by  dissolving  2   tablespoons  of  fertilizer  in  20  liters  of  water  and  using  this  to  irrigate  plants  until  they  are  ready   for  field  transplanting.     A  week  after  transplanting,  do  a  foliar  application  of  Hakaphos  violet  (13-­‐40-­‐13;  NPK),  at  a  rate   of  2  grams  per  1  liter  of  water.  Hakaphos  stimulates  root  development  and  growth.   Apply  Gaucho  (a  systemic  insecticide)  10  days  after  transplanting  or  a  week  before  transplanting   in   the   field;   use   just   enough   water   to   wet   the   trays   without   having   an   overflow.   Gaucho   is   applied   at   2.4   grams   per   20   liters   of   water;   the   quantity   depends   on   the   number   of   plants   to   be   treated.           Greenhouse   conditions:   Greenhouse   conditions   should   be   optimal   for   plant   growth.   Temperature   should   be   maintained   at   less   than   30°C   and   should   not   go   below   20°C   at   night.   Too   high   or   too   low   temperatures  will  stress  the  plants  and  affect  plant  establishment  and  development.  Use  data  loggers,  if   possible,  to  monitor  temperatures  and  relative  humidity  within  the  greenhouse.       Transplanting  D0  seedlings  to  the  field     Field  conditions:  Selection  of  a  proper  site  is  crucial  to  the  success  of  DH  line  development.  If  possible,   select   a   site   that   has   no   or   minimal   disease   and   insect   pressure.   Temperatures   should   seldom   go   above   30°C   and   night   temperatures   should   not   go   below   20°C   during   the   growing   cycle   of   the   plants.   Therefore,  it  is  important  to  have  data  loggers  in  the  field,  to  monitor  the  climatic  variables  (Figure  3).    If   light  intensity  is  too  high,  use  50%  aluminet  shade  cloth  (http://www.greenhousemegastore.com/Stock-­‐ Shadecloth-­‐50-­‐Aluminet/productinfo/SC-­‐ST50A/);   this   will   both   shade   the   plants   and   reduce   the   temperature  underneath  the  shade  cloth.           32   Note:   There   is   aluminet   for   external   use   and   for   internal   use   inside   greenhouses.   Aluminum   cloth   shading  will  reduce  light  intensity  by  50%,  and  this  will  help  with  plant  establishment  and  pollen  set.  The   cloth   should   be   put   4   to   5   meters   above  the   plants   and   leave   enough   space   so   that   the   tract   operations   can   be   done   inside   the   field.   Also,   make   sure   that   there   is   very   good   air   circulation;   otherwise   temperatures  can  rise  and  this  will  affect  pollen  set  and  shedding.       B   A   C     Figure   3.   Agronomic   management   of   D0   nursery   to   minimize   stress   on   the   plants:   (A)   50%   aluminet   shade  cloth  used  to  reduce  the  light  intensity  and  temperature  so  as  to  avoid  stressing  the  plants  and   avoid  tassel  blasting.  The  shading  cloth  is  put  4  meters  high  to  allow  proper  circulation  of  air,  movement   of  tractors,  etc.  (B)  Data  logger  for  measuring  light  intensity  and  temperature.  (C)  Hobo  data  logger  for   measuring   relative   humidity   and   temperature.   The   data   loggers   are   programmed   to   record   every   30   minutes.  [Photos:  G.  Mahuku]     Materials:   Seedlings:  These  should  be  two-­‐week  seedlings  previously  established  in  the  greenhouse.     Plastic   storage   containers:   These   are   necessary   for   transporting   seedlings   to   the   field   while   minimizing  damage  to  plants.   Prepared   field   ready   for   establishing   the   D0   nursery:   The   land   should   preferably   have   drip   irrigation   and   plastic   sheeting   established.   If   necessary,   shading   cloth   should   also   be   put   in   place.     Transport   to   carry   seedlings   to   the   field:   Depending   on   the   number   of   seedlings   to   be   handled,   one  can  use  a  tractor-­‐mounted  trailer  or  a  pickup  truck.     Transplanting   to   the   field:   Well-­‐established   seedlings   should   be   transplanted   in   the   field   after   two   weeks  (maximum),  as  follows:         Take   out   the   seedlings   from   the   greenhouse   and   put   them   in   a   screen   house   close   to   the   D0   nursery;  leave  the  seedlings  for  one  to  three  days  so  that  they  acclimatize  to  field  conditions.   Transport  the  seedlings  in  plastic  trays  using  a  tractor  or  pickup  truck,  to  minimize  damage  or   stress  to  the  plants.     Organize   the   seedlings   according   to   population   and   transplant   them   together,   so   as   to   avoid   confusion  and  mixing  of  populations.   Water  the  seedlings  well,  approximately  one  hour  before  transplanting.   Transplanting   should   be   conducted   early   in   the   morning   to   avoid   mid-­‐day   high   temperatures,   and   the   transplanted   plants   should   be   irrigated   immediately   to   avoid   stressing   them.   If   soils   have   a   high   clay   content,   jiffy   pots   may   not   degrade   well   and   thus   will   create   a   stressed   environment  for  the  plants.  In  such  instances,  eliminate  the  jiffy  pots  just  before  planting.     Immediately  after  transplanting  (or  when  a  row  has  been  completed),  open  the  drip  irrigation   valve  and  start  watering.   Make  a  list  of  the  total  number  of  plants  that  have  been  transplanted.     33      Note:     Field   transplanting   can   be   done   manually   or   with   a   transplanter   (Figure   4).   This   is   convenient   especially  for  large-­‐scale  applications.       A   B     Figure   4.   Depending   on   soil   type,   availability   of   labor,   and   size   of   the   populations,   transplanting   seedlings  in  the  field  can  be  done  either  manually  (A)  or  mechanically  (B)  using  a  tractor  mounted   planter.  [Photos:  G.  Mahuku]     Agronomic  management:  This  is  the  single  most  critical  factor  for  successful  recovery  of  D0  seedlings   along  the  DH  line  development  pipeline.  If  this  process  is  not  managed  well,  success  rates  will  be  low,   even   if   the   other   steps   were   successfully   executed.   Optimization   of   irrigation   regime,   fertilizer   application,  and  effective  management  of  weeds,  diseases,  and  insects  are  crucial  for  minimizing  stress   on   the   D0   plant.   As   the   D0   plants   are   weak   from   the   start,   any   type   of   stress   will   contribute   to   reducing   the   success   and   recovery   rate   of   DH   lines.   Optimum   climatic   conditions   are   required;   where   possible,   select   a   site   that   meets   those   conditions,   soil   type,   and   fertility   regimes   with   minimal   or   no   pressure   from   insects   and   diseases.   Timely   application   of   inputs   such   as   irrigation,   fertilizer,   herbicides,   and   pesticides   is   critical   for   proper   plant   establishment.   Shading   with   nets   can   be   very   helpful   during   anthesis,   particularly   when   temperatures   are   abnormally   high.   Shading   nets   will   reduce   temperature   and  radiation  stress  to  plants  (Figure  3).       Irrigation:   D0   seedlings   have   weak   roots   and   use   much   less   water   than   a   normal   inbred   or   hybrid.   Therefore,  it  is  crucial  that  the  right  amount  of  water  is  applied  for  optimal  plant  development.  Too  little   water   will   stress   the   plants   and   affect   normal   establishment   and   development.   Too   much   water   will   result  in  chlorotic  lines  with  thin  stalks,  and  this  will  affect  subsequent  cob  size  and  pollen  production.   Depending  on  soil  type  and  water  holding  capacity,  a  proper  irrigation  schedule  should  be  worked  out   that   optimizes   water   and   nutrient   use   efficiency.   In   this   regard,   drip   irrigation   is   particularly   suitable   in   the   D0   nursery;   where   possible,   this   should   be   accompanied   by   probes   at  various  points  in  the  D0  nursery,  to  monitor   soil  moisture  and  assist  in  proper  scheduling  of   irrigation  regimes  (Figure  5).       Figure   5.   Drip   irrigation   to   manage   water   application   in   the   D0   nursery.   Fertilizers   can   also   be   effectively   applied   using   the   drip   irrigation  system.  [Photos:  G.  Mahuku]       34   Proper  fertilization:  This  is  critical  to  plant  development,  and  where  the  drip  irrigation  is  being  used,  this   should   be   applied   as   a   solution   along   with   the   irrigation   water.   The   first   irrigation   following   transplanting   should   contain   high   phosphorus   fertilizer   [e.g.,   Haifa   polyfeed   drip   13-­‐36-­‐13   or   Peter’s   Professional  9-­‐45-­‐15  (NPK)]  for  improved  root  development  and  plant  establishment.    Too  much  water   or   heavy   rains   can   affect   nutrient   availability,   as   most   will   be   leached   out.   This   can   be   problematic   during   rainy   seasons   and   if   there   are   frequent   rains,   fertilizer   should   be   banded,   and   avoid   saturating   the  soil.  Plastic  sheeting  and  growing  the  plants  on  raised  beds  will  minimize  this  problem.  Where  drip   irrigation  is  being  used,  connect  a  fertilizer  tank  to  the  main  irrigation  by  a  venturi  valve,  calculate  the   quantity  of  fertilizer  you  want  to  apply  per  hectare,  and  feed  a  concentrated  solution  in  the  irrigation   water.   Micronutrients   are   crucial   to   improved   plant   establishment   and   subsequent   flowering   promotion.   These   should   be   foliar-­‐applied   during   the   entire   plant   growth   period,   following   the   recommended  dosages  and  frequency  of  applications.  During  land  preparation,  75%  N,  100%  P,    100%  K   is   incorporated   into   soil,   and   this   is   also   applied   through   drip   irrigation   just   before   flowering.   Foliar   application   of   nutrients   is   essential   to   enhance   plant   growth   and   development.   Three   days   after   transplanting,   Hakaphos   Violet   A   13–40–13   (NPK)   is   applied   at   2.4   grams   for   20   liters   of   water   once   every  week,  and  Impulsor  at  40  ml  per  15  liters  of  water  (at  the  rate  of  0.75  liters  Impulsor  per  hectare).   It  is  important  to  consistently  monitor  the  plants  and  apply  foliar  nutrition  or  fertilizers  as  needed.     Weed   control:   Proper   weed   control   is   essential,   to   avoid   competition   and   maximize   nutrient   availability   and  use  by  D0  plants.  Hand  weeding  is  preferred  and  where  possible  minimizes  the  use  of  herbicides,  as   most   D0   plants   are   highly   sensitive   to   residual   herbicides,   and   hence   this   will   affect   proper   plant   development  and  establishment.  Plastic  sheets  (foils)  are  an  excellent,  low-­‐cost  way  to  manage  weeds,   and   these   are   routinely   used   in   horticultural   crops.   Apart   from   managing   weeds,   plastic   foil   will   help   regulate   soil   temperature   and   humidity   within   the   rooting   system,   resulting   in   uniform   DH   plant   establishment  and  growth.  Drip  irrigation  tubing  and  plastic  sheeting  can  be   placed  in  one  step,  using  a   tractor-­‐mounted  device  (Figure  6).       Figure  6.  Plastic   foil  is  used  to  better  manage  weeds  and   regulate   soil   moisture   and   humidity.   DH   plants   under   plastic   sheeting   were   found   to   perform   better   than   those   without   plastic   covers.   [Photos:   G.   Mahuku]     Disease  and  insect  control:  In  the  tropics,  disease  and  insect  pressure  is  a  major  problem  that  can  affect   the   rate   of   recovery   of   DH   lines.   A   judicious   insect   and   disease   management   regime   is   required   to   minimize  plant  damage  and  increase  the  rate  of  DH  line  recovery.  The  time  of  fungicide  and  insecticide   application   is   crucial   to   minimize   the   effect   on   flowering,   especially   pollen   shedding.   Stop   fungicide   and   insecticide  two  weeks  before  flowering  to  minimize   the  possible  effects  on  flowering.  Depending  on  the   incidence  of  diseases  and  insect  pests,  use  the  recommended  fungicides  or  pesticides.    Foliar  diseases     35   such   as   Northern   and   Southern   Corn   Leaf   Blights   are   controlled   using   the   fungicide   Tilt   (PROPICONAZOLE),   which   is   applied   one   month   after   planting   or   when   symptoms   are   noticed,   and   thereafter  once  every  two  weeks  at  0.75  liters  per  hectare.  This  fungicide  is  effective  against  most  foliar   diseases,   but   application   should   be   stopped   a   week   or   two   before   flowering.   Gaucho   is   a   systemic   insecticide   that   is   applied   during   the   seedling   stage   and   will   protect   plants   from   most   insects   (Figure   7).   Cutworms   are   controlled   using   Lorsban   Grana   lade   3%   insecticide   (40   kg/ha),   and   this   is   incorporated   into  the  rows  during  land  preparation.  Stem  borers  are  controlled  using  Lorsban  480  EM  at  a  rate  of  1   liter  per  hectare  or  using  Karate  Zeon  at  0.5  liters  per  hectare.     Note:   Mention   of   specific   brand   names   of   commercial   chemicals   (including   fertilizers,   fungicides,   and   pesticides)  is  not  intended  as  an  official  endorsement  of  the  product  by  CIMMYT.  There  may  be  other   equal  or  better  products  available  in  the  market  for  achieving  the  same  task.             A   B   Figure   7.   Ear   worms   can   be   devastating   in   the   D0   nursery,   and   a   judiciary   schedule  for  managing  such  insect  pests   is   required.   The   figure   shows   the   damage  that  can  be  done  by  (A)  corn  ear   worm   (Heliothis   sp.)   and   (B)   insecticide   application   to   manage   the   pest.   [Photos:   G.  Mahuku]     Self-­‐pollination  of  D0  plants  for  deriving  new  DH  lines  In  this  step,  the  putative  doubled  haploid  (D0)   plants  are  carefully  self-­‐pollinated  to  derive  the  D1  seed,  which  in  essence  is  the  new  DH  line  for  further   seed  multiplication  or  use  by  the  breeder.  Please  note  that  colchicine  treatment  may  or  may  not  lead  to   uniform   or   complete   doubling   of   the   chromosomes   of   all   cells   of   a   seedling;   this   is   called   “sectoral   diploidization.”   The   effect   may   be   variable,   especially   on   the   genotype   and   the   colchicine   application.   Some   plants   may   have   tassels   producing   abundant   pollen   while,   in   most   instances,   tassels   may   have   limited   pollen-­‐producing   anthers   or   none   at   all   (Figure   8).   Consequently,   self-­‐pollination   may   prove   difficult.  Therefore,  well-­‐trained  staff  are  required  to  avoid  losing  such  genotypes  due  to  unsuccessful   self-­‐pollination.       A   B   C     Figure   8.   Pollen   production   by   a   putative   DH   plant:   (A)   good   quantity   of   pollen   produced;   (B)  only   a   few   branches   are   producing   pollen   while   the   rest   are   sterile   due   to   sectoral   diploidization;   and   (C)   a   sterile   tassel,   a   common   problem   that   could   be   observed   in   the   D0   nursery.  [Photos:  G.  Mahuku]       36   Identifying  and  discarding  “false”  (diploid)  plants  in  the  D0  nursery:  Misclassification  of  haploid  kernels   can  sometimes  result  from  insufficient  expression  of  phenotypic  color  markers,  presence  of  dominant   anthocyanin  color  inhibitor  genes,  and  lack  of  trained  personnel.  However,  putative  DH  plants  can  easily   be  distinguished  from  normal  diploid  plants  under  field  conditions.  The  DH  plants  can  be  differentiated   from   the   normal   diploid   plants   on   the   basis   of   plant   vigor,   leaf   habit,   tassel   size,   and   anthocyanin   pigmentation  on  the  stalk.    Monitor  coloration  of  the  stalk  of  putative  DH  plants  before  flowering  time   and   eliminate   plants   that   show   purple   stalk   coloration.   False   plants   need   to   be   discarded   in   time   to   avoid   competition   for   light,   water,   and   nutrients,   avoiding   pollen   contamination   to   correct   DH   plants,   and  focusing  the  efforts  on  correct  (DH)  plants  for  maintenance.       Self-­‐pollination  for  maintenance  and  seed  multiplication  of  new  DH  lines       Materials:   1.   Custom-­‐made   glassine   bags   (or   “silk   bags”)   and   common   pollination   bags   (or   “tassel   bags,”   Lawson   No.  501).   2.  Trained  and  dedicated  personnel     Monitor  anther  emergence:  Well-­‐trained  field  staff  are  crucial  to  perform  this  step.  The  DH  plants  are   generally  weak,  often  have  only  a  few  pollen-­‐shedding  anthers,  and  may  only  shed  limited  pollen  for  a   few   days.   Hence,   constant   monitoring   to   spot   the   D0   plants   shedding   pollen   (among   the   many   plants   which   may   not   have   fertile   tassels),   immediately   collecting   the   pollen   and   performing   the   self-­‐ pollination  are  critical  for  recovery  of  D1  seed  and   DH  line  development.  It  should  be  noted  that  the  DH   genotype   will   be   lost   if   self-­‐pollination   is   not   properly   undertaken,   even   if   all   the   previous   steps   are   performed   perfectly.   Pollination   is   a   labor-­‐intensive   step,   and   during   this   period,   a   skilled   workforce   must  be  constantly  in  the  field  to  avoid  missing  any  plant  that  is  ready  for  pollination.       Note:   Success   of   DH   operations   depends   on   well-­‐trained   staff,   especially   the   lab   and   field   workers.   Experience  does  matter,  and  with  each  cycle  efficiency  is  increased.  Therefore,  avoid  high  staff  turnover   as  this  can  significantly  affect  the  success  rates.       Pollination:  Cover  the  ear  shoots  before  any  silk  emergence.  Non-­‐coated,  transparent  glassine  bags  or   “silk  bags”  (approximately  6×20  cm)  are  most  suitable  to  collect  pollen  from  the  putative  DH  plants  for   self-­‐pollination.  As  pollen  production  is  often  limited  in  these  plants,  the  transparent  bags  allow  visual   assessment   of   the   quantity   of   pollen   collected   for   self-­‐pollination.   If   necessary,   the   pollination   can   be   repeated   the   next   day.   After   successful   pollination,   common   pollination   bags   or   “tassel   bags”   can   be   used  to  cover  and  protect  the  pollinated  ears.  For  pollination,  cover  tassels  with  pollination  bags  in  time   before   the   intended   pollination,   and   try   to   self-­‐ pollinate   each   plant.   Cover   the   pollinated   ears   properly   with   tassel   bags   for   protection   and   A   B   fasten  them  tightly  with  stapler  pins  (Figure  9).       Figure  9.  Pollination  in  the  D0  nursery:  (A)  non-­‐ coated,   transparent   glassine   bags   or   “silk   bags”   are   used   to   collect   pollen   from   putative   DH   plants   for   self-­‐pollination;   (B)   a   successfully   pollinated  plant.  [Photos:  G.  Mahuku]       37   Harvesting   self-­‐pollinated   ears   after   physiological   maturity:   Often   only   few   seeds   are   set   on   the   ear   of   a   DH   plant   (Figure   10).   Therefore,   utmost   care   is   needed   to   avoid   seed   loss   in   the   field.   The   ears   should   be  harvested  carefully  and  kept  in  proper  cover  bags  during  transport  to  the  warehouse  for  dehusking   and  drying.  This  seed  represents  the  newly  developed,  completely  homozygous  DH  line,  which  may  be   planted   again   for   seed   multiplication   to   be   able   to   use   the   DH   line   in   further   research   and   breeding   activities.   If   there   are   some   ears   bearing   purple-­‐colored   seeds,   discard   these   because   they   are   misclassified  plants  (normal  diploids  and  not  DH)  that  could  have  been  missed  in  the  earlier  steps.    Seed   production   on   DH   plants   is   expected   to   improve   in   subsequent   cycles   of   DH   production   because   (1)  selection   occurs   in   source   germplasm   for   genes   imparting   favorable   response   to   haploid   induction   and   artificial   chromosome   doubling,   and   (2)   the   technical   and   field   personnel   involved   in   DH   operations   gain  experience.     A   B     Figure   10.   Harvested   ears   from   the   D0   nursery.   There   could   be   considerable   variation  in  the  amount  of  seed  produced  on  the  D1  (DH)  ears,  varying  from  one   or   two   seeds   to   more   than   50.   Therefore,   adequate   care   should   be   taken   to   avoid  any  loss  of  seed  while  harvesting.  [Photos:  G.  Mahuku]       38   7.    Integrating  Marker-­‐Assisted  Selection  in  the  DH-­‐Based  Breeding  Pipeline  for   Rapid  Development  and  Delivery  of  Superior  Parental  Lines  and  Cultivars   R  Babu,  Sudha  K  Nair,  BS  Vivek,  Felix  San  Vicente,  and  BM  Prasanna     Introduction   In   recent   years,   doubled   haploids   (DH)   and   molecular   markers   have   emerged   as   two   of   the   most   powerful   technologies   that   are   revolutionizing   the   way   homozygous   lines   are   developed   in   applied   maize   breeding   programs   (Mayor   and   Bernardo,   2009).   As   discussed   earlier   in   this   manual,   the   DH   technology   significantly   reduces   the   time   required   to   obtain   genetically   homozygous   and   pure   lines   compared   to   conventional   inbreeding.     Important   advantages   of   using   DH   lines   in   the   breeding   program   include   a   maximum   genetic   variance   between   selection   units   and   an   increased   precision   in   estimating   the   genotypic   value   of   DH   lines   and   their   testcrosses   (TCs)   (Gordillo   and   Geiger,   2008).   In   addition,   utilizing  DH  lines  in  the  breeding  program  permits  early  selection  of  prospective  hybrids,  simplifies  the   logistics   of   inbred   seed   increase   and   maintenance,   and   allows   quick   fixation   of   favorable   alleles   at   quantitative  trait  loci  (QTL)  (Mayor  and  Bernardo,  2009;  Lubberstedt  and  Ursula  ,  2012).  When  coupled   with   seed   DNA-­‐based   genotyping   (Gao   et   al.,   2009),   especially   for   large   effect   genomic   regions   conditioning   nutritional   quality   (e.g.,   crtRB1-­‐governed   beta   carotene   content)   or   disease   resistance   traits   (e.g.,   msv1-­‐driven   Maize   Streak   Virus   resistance),   DH-­‐based   molecular   breeding   results   in   enormous  saving  of  time,  labor,  land,  and  other  resources.       Line   development   and   recurrent   selection   (RS)-­‐based   population   improvement   are   the   two   most   routinely   applied   activities   in   maize   breeding   programs.   The   improved   source   populations   obtained   through  RS  are  used  either  as  new  source  germplasm  for  deriving  inbred  lines  or  directly  as  synthetics   that   could   be   released   for   farmer   cultivation   in   resource-­‐poor   regions.   Here,   we   discuss   two   possible   and   pragmatic   approaches   to   integrating   marker-­‐assisted   selection   (MAS)   strategies   in   DH-­‐based   breeding  programs.         Integrating  MAS  in  DH-­‐based  pedigree  breeding  pipeline   Pedigree   breeding   along   with   extensive   multi-­‐location   testing   across   a   wide   range   of   target   environments   has   been   the   mainstay   of   maize   improvement   programs   worldwide.   Pedigree   breeding   starts   with   crossing   of   two   elite   genotypes   that   have   complementary   traits   (such   as   good   agronomy,   abiotic   stress   tolerance,   disease   resistance,   and   nutritional   quality),   and   in   the   successive   generations,   superior  progenies  combining  the  different  desirable  traits  are  selected  until  homozygosity  is  achieved   in  F7  or  F8  generation.  A  selection  history  is  maintained  throughout  the  breeding  generations.    With  the   advent  of  DH  technology,  it  is  possible  to  obtain  homozygous  lines  in  only  two  generations  as  against  the   seven   to   eight   generations   that   are   mandatory   in   a   typical   conventional   pedigree   breeding.   While   the   DH  technology  makes  it  possible  to  save  time  significantly,  it  removes,  to  a  certain  extent,  the  selection   opportunities   that   a   breeder   generally   has   during   multiple   filial   generations.   In   general,   it   has   been   a   routine   practice   in   maize   breeding   to   induce   haploids   at   F1   generation   to   save   time.   However,   F1-­‐ derived   doubled   haploids   tend   to   have   reduced   recombinations   (because   there   is   only   one   round   of   meiosis)   and   have   been   found   to   decrease   the   responses   to   single   or   multiple   cycles   of   selection   pressure  (Riggs  and  Snape,  1977;  Jannink  and  Abadie,  1999).       For  a  trait  controlled  by  100  or  more  QTLs,  Bernardo  (2009)  reported,  based  on  simulation  experiments,   that   the   cumulative   responses   to   selection   were   up   to   4–6%   larger   among   F2-­‐derived   DH   lines   than   among  F1-­‐derived  DH  lines  and  hence  suggested  inducing  haploids  from  F2  rather  than  F1  for  sustaining   long-­‐term   enhanced   selection   response.   However,   deciding   between   F1   and   F2   involves   a   certain   trade-­‐ off  between  time  and  resources  for  the  breeding  program.  As  proposed  by  Bernardo  (2009),  if  initial  F1s     39   are   made   on   a   speculative   basis   in   the   breeding   program,   inducing   haploids   at   F2   may   not   consume   additional  time.       Besides  enhanced  recombination,  an  important  advantage  of  inducing  haploids  at  F2  is  the  opportunity   to  subject  F2  individuals  (both  at  the  seed  and  plant  level)  to  required  genotypic  as  well  as  phenotypic   selection   pressure.   Depending   on   the   target   traits   of   the   breeding   program   and   availability   of   molecular   marker  information  for  the  specific  locus/loci  governing  those  traits,  F2  seeds  could  be  individually  seed-­‐   genotyped   through   non-­‐destructive   sampling   (Gao   et   al.,   2009),   and   the   seeds   carrying   unfavorable   alleles   in   homozygous   condition   could   be   discarded.   This   procedure   has   been   described   as   “F2-­‐ enrichment”  (Howes  et  al.,  1998;  Bonnett  et  al.,  2005;  Wang  et  al.,  2007).  Currently  in  maize,  few  loci   have  been  identified  that  confer  large  phenotypic  effects  for  which  such  an  approach  would  be  feasible.   Some  examples  are  as  follows:     Crtrb1,   a   carotene   hydroxylase   gene   (Yan   et   al.,  2010)   has   been   demonstrated   to   have  a   2-­‐  to   10-­‐fold  effect  on  beta-­‐carotene  content  across  diverse  genetic  backgrounds  in  the  tropical  maize   germplasm  (Babu  et  al.,  forthcoming).  Seed  DNA  based  genotyping  of  crtRB1,  especially  in  the   upstream   generations   such   as   F2   or   F3,   is   routinely   employed   in   the   HarvestPlus–Maize   breeding   program   at   CIMMYT,   which   has   paid   rich   dividends   leading   to   development   of   lines   that  have  significantly  higher  levels  (15–20  ppm)  of  provitamin  A  as  compared  to  1–2  ppm  found   in  normal  yellow  maize.     Opaque2  (o2)  is  a  transcriptional  regulator  in  maize  whose  mutant  allele  confers  twice  as  much   lysine   and   tryptophan   in   the   endosperm   as   normal   maize,   which   along   with   associated   endosperm   modifiers   is   known   as   Quality   Protein   Maize   (QPM)   (Prasanna   et   al.,   2001).   Molecular  markers  located  within  o2  have  been  successfully  used  in  the  rapid  development  of   QPM  versions  of  normal  maize  lines  (Babu  et  al.,  2005).     Maize  Streak  Virus  (MSV)  is  a  major  disease  in  most  parts  of  sub-­‐Saharan  Africa;  a  large  effect   QTL   conditioning   MSV   resistance   has   been   identified   on   chr.1   (Welz   et   al.,   1998;   Lu   et   al.,   1999;   Kyetere  et  al.,  1999)  across  different  genetic  backgrounds.  The  CIMMYT  Global  Maize  Program   has  recently  identified  (and  is  presently  validating)  a  set  of  SNP  markers  in  this  genomic  region   which  could  be  potentially  utilized  for  effective  differentiation  of  MSV  resistant  and  susceptible   genotypes   without   phenotypic   selection.   Though   additional   minor   loci   influencing   MSV   resistance   may   exist   in   other   parts   of   the   maize   genome   or   in   different   genetic   backgrounds,   msv1   may   be   considered   as   an   essential   prerequisite   for   achieving   reasonable   levels   of   resistance  to  the  disease  (Sudha,  personal  communication).     With   wider   adoption   of   Genome-­‐Wide   Association   Studies,   the   maize   genetics   community   is   likely   to   unravel   and   validate   further   a   large   number   of   marker-­‐trait   associations   in   the   coming   years,   which   is   expected  to  enable   F2   enrichment  as  a  preferred   approach   in  maize,  thereby  providing  scope  for  pre-­‐ selecting   source   germplasm   before   DH   induction.   When   the   marker-­‐selected   individuals   are   grown   in   the  field,  additional  phenotypic  selection  for  general  plant  vigor,  type,  and  other  per  se  traits  could  be   exercised,  ensuring  that  only  “good”  genotypes  are  subjected  to  the  DH  induction  procedure.       An   illustrative   DH-­‐based   MAS   scheme   is   presented   in   Figure   1,   which   is   aimed   at   combining   drought   tolerance   with   one   of   the   disease   resistance   traits   during   pedigree   breeding.   As   mentioned   earlier,   a   large  effect  genomic  region  influencing  MSV  resistance  has  been  identified  and  its  phenotypic  effect  has   been  validated  in  diverse  genetic  backgrounds.  Marker-­‐based  screening  of  individual  F2  seeds  for  msv1   could   help   in   discarding   individuals   with   an   unfavorable   msv1   allele   in   homozygous   condition,   and   further   phenotypic   screening   for   per   se   and   plant   vigor   related   traits   will   ensure   elimination   of   weak   plants   from   being   subjected   to   haploid   induction.   Subsequently,   the   marker-­‐screened   and   phenotypically   selected   plants   are   crossed   to   the   tropicalized   haploid   inducer   and   putative   haploid     40   kernels   are   identified.   Upon   chromosome   doubling,   and   selfing   of   the   D0   plants   to   D1   seed   stage,   simultaneously,  pollen  from  the  D0  plants  (if  available  in  adequate  quantity)  can  be  used  for  making  TCs.   Once   sufficient   DH-­‐TCs   are   produced   (using   D0   or   D1s),   they   can   be   evaluated   for   performance   under   drought   and   optimal   conditions   at   multiple   locations,   representing   target   population   of   environments   and   best-­‐bet   drought   tolerant   hybrids   combining   reasonable   levels   of   MSV   resistance   identified   and   nominated   for   National   Performance   Trials,   and   their   corresponding   parental   lines   maintained.   Additionally,   genotyping   the   DH   lines   enables   estimation   of   marker   effects   for   drought   and   optimal   performances,   which   over   the   years   can   potentially   contribute   to   genome-­‐enabled   prediction   of   untested   DH   lines,   being   generated   year   after   year   in   the   breeding   program,   thereby   minimizing   the   managed  screening/phenotyping  requirements.       P1#(Drought#tolerant)#x#P2#(MSV#resistant)#     F1#     F2#   • Chip 500 seeds and genotype with validated SNPs flanking   major QTL for MSV resistance   • Select 100 seeds with favorable allele and grow in the field   • Select 40-50 plants based on vigor and per se performance and cross them (as female) to tropicalized haploid inducer (as male).     Haploid#Kernels#   • Chromosome doubling   • D1 seed obtained by selfing D0 plants   • Seed increase, only if needed.   Doubled#Haploid#(DH)#Lines#       DH7TC#forma<on# Mul<7loca<on#evalua<on#of# High#density#genotyping# using#DH#lines# DH7TCs#under#op<mal#and# of#successful#DH#lines#   drought#condi<ons#       Iden<fica<on#of#best7bet# Es<mate#marker#effects#for#   Maintenance#of#superior#DH# hybrids#combining#drought# lines#(parents#of#iden<fied# further#use#in#Genomic#Selec<on#   best7bet#hybrids#)# tolerance#and#MSV#resistance## (GS)#as#in#(Fig.#2)#   Figure  1.  An  illustrative  scheme  for  DH-­‐based,  marker-­‐assisted  selection  for  potentially  combining   drought   tolerance   and   disease   resistance   in   a   pedigree   breeding   program.   MSV   =   Maize   Streak   Virus;   D0   =   putative   DH   seedling   after   chromosome   doubling   treatment   of   haploids;   D1   =   seed   derived  from  D0  plants;  TC  =  testcross.     Rapid-­‐cycle,  open-­‐source  genomic  selection   Recurrent  selection  (RS)  has  been  an  important  tool  in  maize  breeding  for  developing  improved  source   populations  that  are  significantly  enhanced  in  desirable  allele  frequencies,  especially  for  highly  complex,   polygenic   target   traits   like   drought   and   heat   tolerance.   Despite   being   effective,   RS   based   only   on   phenotypes   is   time   consuming   and   resource   demanding   as   it   involves   developing   TC   progenies   and   evaluation   in   replicated   multiple   locations   before   every   selection   step.   Thus,   if   four   RS   cycles   are   intended,   it   entails   a   minimum   of   four   seasons   of   TC   generation   and   another   four   seasons   of   performance  evaluation  in  multiple  locations.  Genomic  selection  (GS)  is  a  novel  approach  that  exploits   high-­‐density   genotyping   to   predict   the   total   genetic   value   of   an   individual   based   on   a   model   set   of   training   individuals   that   are   phenotyped   at   representative   locations.   GS-­‐based   approaches   typically     41   ignore   information   on   the   number   and   location   of   QTL   and   focus   on   the   genetic   improvement   of   quantitative  traits  rather  than  on  understanding  their  genetic  basis  (Jannink  et  al.,  2010).  The  usefulness   of   high-­‐density   genotyping   based   procedures   that   focus   on   predicting   performance   indicates   that   markers  can  be  used  as  a  selectable  tool  to  improve  a  complex  trait,  even  without  a  clear  understanding   of  the  underlying  genetics  of  the  trait.  Rapid-­‐cycle  GS  (RCGS)  is  a  convenient  tool  to  augment  the  pace  of   RS  cycles  without  having  to  phenotype  each  set  of  intermated  progenies.  RCGS  also  saves  a  considerable   amount   of   material   resources   as   it   involves   only   one   season   of   phenotyping   at   representative   locations.   In   the   subsequent   generations,   intermated   individuals   are   only   genotyped   and   their   genetic   worth   is   predicted  based  on  previously  estimated  marker  effects.       An   illustrative   narration   of   RCGS   in   a   closed   multi-­‐parent   population   context   is   presented   in   Figure   2.   RS   in  a  multi-­‐parent  population  can  be  very  effective  as  compared  to  in  a  bi-­‐parental  population,  which  is   resource  demanding  and  doesn’t  permit  evaluation  of  a  large  number  of  populations.  Typically,  8  to  12   elite   maize   lines   with   trait   complementarity   such   as   drought   tolerance,   disease   resistance,   and   enhanced  nutritional  quality  are  chosen  within  each  heterotic  group  and  half-­‐diallels  are  made  so  as  to   obtain   all   possible   combinations.   The   F1s   are   intermated   either   in   isolation   or   through   controlled   pollination  to  form  a  large  F2  population.  Depending  on  the  target  traits  for  the  particular  agro-­‐ecology   and  availability  of  molecular  information  for  such  traits,  F2  enrichment  could  be  pursued  as  described   earlier   in   the   pedigree   breeding   context.   In   this   illustration,   F2   seeds   are   screened   with   four  validated   SNPs,   which   are   flanking   a   disease   resistance   QTL   and   a   major   gene,   crtRB1,   which   enhances   beta-­‐ carotene  content  in  the  maize  endosperm.       In   the   subsequent   season,   following   F2   screening,   at   least   500   S2   families   (C0)   will   be   established   for   each   multi-­‐parent   population,   which   will   be   genotyped   and   testcrossed.   In   the   following   seasons,   TCs   will   be   phenotyped   under   drought   and   optimal   conditions   in   representative   locations,   and   marker/haplotype/QTL   effects   will   be   estimated.   The   top   5–10%   of   the   C0   families   will   be   selected   based   on   the   test   cross   data   and   recombined   to   form   C1   (cycle   1).   The   individuals   of   C1   will   be   genotyped   and   based   on   the   previously   calculated   C0   marker   effects,   GEBVs   (Genomic   Estimated   Breeding   Values)   will   be   estimated   and   the   top   5–10%   of   the   GEBV-­‐selected   C1   individuals   will   be   recombined  to  form  C2,  without  phenotypic  evaluation.  This  would  be  repeated  for  one  more  cycle  until   C3,  wherein  the  GEBV-­‐selected  individuals  will  be  crossed  to  a  haploid  inducer  for  generating  DH  lines.  If   the   breeding   program   manages   to   obtain   a   large   number   of   DH   lines   from   C3,   one   way   of   selecting   a   smaller   portion   of   superior   lines   without   phenotypic   evaluation   could   be   based   on   GEBVs   (marker   effects   of   C0   +   genotypic   information   of   DH   lines).   The   GEBV-­‐selected   DH   lines   may   be   distributed   to   different   small   and   medium   enterprises   (SMEs)   and   national   agricultural   research   system   (NARS)   partners   for   respective   evaluation   (per   se   and   TCs)   in   their   target   production   environments.   The   phenotypic   information   generated   by   the   SME   seed   companies   and   NARS   partners   could   be   shared,   which  will  contribute  to  revised  or  updated  marker  effects  to  aid  in  future  predictions.  The  pre-­‐selection   of   F2   individuals   (using   specific   marker   information   for   nutritional   quality   and/or   disease   resistance   traits)  coupled  with  multiple  cycles  of  recurrent  selection  based  on  robust  marker  effect  estimates  for   drought   and   optimal   conditions   ensure   that   the   C3-­‐derived   DH   lines   are   superior   in   nutritional   quality   and   disease   resistance   as   well   as   competent   in   terms   of   performance   under   drought   without   yield   penalty  in  optimum  conditions.  The  open-­‐source  nature  of  the  proposed  scheme  also  ensures  that  the   phenotype   information   generated   by   different   partner   institutions   is   shared   while   the   institutions   maintain  proprietary  rights  over  the  material  resources.  The  improved  source  population,  C3,  can  also   be   shared   with   the   interested   NARS,   which   in   turn   may   promote   this   as   superior   synthetic/OPV   for   farmer   cultivation   or   use   it   in   their   own   breeding   program   for   deriving   superior   inbred   lines.   One   can   derive  greater  benefit  from  the  proposed  scheme  when  year-­‐round  nurseries  are  used  for  accelerating   the  breeding  cycle.       42   DT1,%DT2,%DT3,%DT4,%DR1,%DR2,%DR3,%DR4,%PA1,%PA2,%PA3,%PA4% Half-diallel of parental lines F1# Intermating of F1s to form F2 F2# • Chip 5000 seeds and genotype with 4 validated SNPs (two each flanking a major DR QTL and CrtRB1 (ProA) gene) • Discard seeds homozygous for unfavorable alleles and self the rest F2:3#families#(C0)# F2:3#family#TCs#(C04TC)# C0#observa9on# nursery# C04TC#evalua9on#under# op9mal#and#drought# Low/high#density# genotyping#of#C0#families# Select#top#5410%#of#C0s#based#on#C04TH# and#recombine#to#form#C1# Recombine#the#top#5410%#of#C1#plants#based#on# GEBVs#to#form#C2#without#phenotypic#evalua9on# Marker effects from C0 + C1 genotypes = GEBV-C1 Recombine#the#top#5410%#of#C2#plants#based#on# GEBVs#to#form#C3#without#phenotypic#evalua9on# Marker effects from C0 + C2 genotypes = GEBV-C2 Select#the#top#5410%#of#C3#plants#based#on#GEBVs#and# cross#them#(as#female)#to#haploid#inducer#(as#male)# Marker effects from C0 + C3 genotypes = GEBV-C3 Haploids# Same steps as presented in Fig. 1 DH#lines# Genotyping#of#DH#lines#and#Selec9on#of#superior#lines# based#on#GEBVs#without#phenotypic#evalua9on# Phenotyping#of## DH4TC#and#DH#lines# per$se$by#SME1# Phenotyping#of## DH4TC#and#DH#lines# per$se$by#SME2# Commercial#release#of#elite#hybrids#by# SMEs#aSer#NPTs# Phenotyping#of## DH4TC#and#DH#lines# per$se$by#NARS1# Phenotyping#of## DH4TC#and#DH#lines# per$se$by#NARS2# Release#of#elite#hybrids#by#NARS# aSer#NPTs# Phenotypes#from#SMEs#and#NARS#contribute#to# revised/updated#marker#effects#     Figure   2.   An   open-­‐source,   multi-­‐parent   RS   model   for   integrating   molecular   markers   and   DH   technology   to   rapidly  deliver  superior  lines  with  drought  tolerance,  disease  resistance,  and  nutritional  quality.  DT   =  drought   tolerance;  DR  =  disease  resistance;  PA  =  provitamin  A;  C0  =  cycle  0;    C1  =  cycle  1;  C2  =  cycle  2;  C3  =  cycle  3;  TC   =   testcross;   GEBV   =   genomic   estimated   breeding   value;   SME   =   small   and   medium   enterprises;   NARS   =   national  agricultural  research  system;  NPTs  =  national    performance  trials.     43   References   Babu  R,  Nair  SK,  Kumar  A,  Venkatesh  S,  Sekhar  JC,  Singh  NN,  Srinivasan  G,  Gupta  HS  (2005)  Two-­‐generation  marker   aided  backcrossing  for  rapid  conversion  of  normal  maize  lines  to  Quality  Protein  Maize  (QPM).  Theor.  Appl.   Genet.  111:  888–897.     Babu   R,   Palacios   N,   Gao   S,   and   Yan   J,   and   Pixley   K   (forthcoming)   Validation   of   the   effects   of   molecular   marker   polymorphisms  in  LcyE  and  CrtRB1  on  provitamin  A  concentrations  for  26  tropical  maize  populations.  Theor.   Appl.  Genet.   Bernardo   R   (2009).   Should   maize   doubled   haploids   be   induced   among   F1   or   F2   plants?   Theor.   Appl.   Genet.   119:   255–262.   Bonnett   DG,   Rebetzke   GJ,   Spielmeyer   W   (2005)   Strategies   for   efficient   implementation   of   molecular   markers   in   wheat  breeding.  Mol.  Breed.  15:  75–85.   Gao   S,   Martinez   C,   Debra   J,   Krivanek   AF,   Crouch   JH,   Xu   Y   (2009)   Development   of   a   seed   DNA-­‐based   genotyping   system  for  marker-­‐assisted  selection  in  maize.  Mol.  Breed.  22:  477–494.   Gordillo  A,  Geiger  HH  (2008)  Alternative  recurrent  selection  strategies  using  doubled  haploids  lines  in  hybrid  maize   breeding.  Crop  Sci.  48:  911–922.   Howes  NK,  Woods  SM,  Townley-­‐Smith  TF  (1998)  Simulations  and  practical  problems  of  applying  multiple  marker   assisted  selection  and  doubled  haploids  to  wheat  breeding  programs.  Euphytica  100:  225–230.   Jannink   JL,   Abadie   TE   (1999)   Inbreeding   method   effects   on   genetic   mean,   variance,   and   structure   of   recurrent   selection  populations.  Crop  Sci.  39:  988–997.   Jannink   JL,   Lorenz   AJ,   Iwata   H   (2010)   Genomic   selection   in   plant   breeding:   from   theory   to   practice.   Briefings   in   Functional  Genomics  9:  166–177.     Kyetere  DT,  Ming  R,  McMullen  MD,  Pratt  RC,  Brewbaker  J,  Musket  T  (1999)  Genetic  analysis  of  tolerance  to  maize   streak  virus  in  maize.  Genome  42:  20–26.   Lu   XW,   Brewbaker   JL,   Nourse   SM,   Moon   HG,   Kim   SK,   Khairallah   M   (1999)   Mapping   of   quantitative   trait   loci   conferring  resistance  to  maize  streak  virus.  Maydica  44:  313–318.     Lubberstedt  T,  Ursula  K  (2012)  Application  of  doubled  haploids  for  target  gene  fixation  in  backcross  programmes  of   maize.  Plant  Breed.  doi:10.1111/j.1439-­‐0523.2011.01948.x   Mayor  P,  Bernardo  R  (2009)  Doubled  haploids  in  commercial  maize  breeding:   one-­‐stage  and  two-­‐stage  selection   versus  marker-­‐assisted  recurrent  selection.  Maydica  54:  439–448.   Prasanna  BM,  Vasal  SK,  Kassahun  B,  Singh  NN  (2001)  Quality  protein  maize.  Curr.  Sci.  81:  1308–1319.   Riggs  TJ,  Snape  JW  (1977)  Effects  of  linkage  and  interaction  in  a  comparison  of  theoretical  populations  derived  by   diploidized  haploid  and  single  seed  descent  methods.  Theor.  Appl.  Genet.  49:  111–115.   Wang   J,   Chapman   SC,   Bonnett,   DG,   Rebetzke   GJ,   Crouch   J   (2007)   Application   of   population   genetic   theory   and   simulation  models  to  efficiently  pyramid  multiple  genes  via  marker-­‐assisted  selection.  Crop  Sci.  47:  582–588.   Welz   HG,   Schechert   A,   Pernet   A,   Pixley   KV,   Geiger   HH   (1998)   A   gene   for   resistance   to   maize   streak   virus   in   the   African  CIMMYT  maize  inbred  CML202.  Mol.  Breed.  4:  147–154.   Yan   J,   Kandianis   C,   Harjes   CE,   Li   B,   Kim   EH,   Yang   X,   Skinner   DJ,   Zhiyuan   F,   Mitchell   S,   Li   Q,   Salas   Fernandez   MG,   Zaharieva   M,   Babu   R,   Yang   F,   Palacios   Rojas   N,   Li   J,   Dellapenna   D,   Brutnell   T,   Buckler   ES,   Warburton   ML,   Rocheford   T   (2010)   Rare   genetic   variation   at   Zea   mays   crtRB1   increases   beta-­‐carotene   in   maize   grain.   Nat.   Genet.  42:  322–327.       44   8.    DH  in  Commercial  Maize  Breeding:  Phenotypic  Selections   Daniel  Jeffers  and  George  Mahuku     Introduction   The   use   of   doubled   haploids   (DH)   in   commercial   maize   breeding   programs   offered   several   benefits   to   the  seed  industry,  including  reduction  of  costs  related  to  running  the  breeding  operations,  accelerated   breeding   cycles   to   bring   products   to   market,   and   improved   efficiencies   in   characterization   and   exploitation   of   new   germplasm.   DH   technology   has   thus   become   a   key   component   of   the   product   development   process   of   the   large   seed   companies.   This   chapter   will   discuss   how   the   use   of   DH,   coupled   with   high-­‐throughput   and   reasonably   precise   phenotyping,   is   being   used   in   the   commercial   breeding   programs.     Doubled  haploids  in  maize  have  been  produced  for  maize  breeding  since  1940s  in  the  US  (Chase,  1947,   1949),   and   as   parental   lines   of   commercial   hybrids   since   the   early   1960s   (Troyer,   2004;   Forster   and   Thomas,   2005).   DeKalb   640   was   the   first   widely-­‐accepted,   high   density   planting   tolerant   commercial   hybrid   in   the   US,   and   contained   three   DH   lines   in   its   pedigree   (Chang   and   Coe,   2009).   Though   the   induction   rate   and   chromosome   doubling   rate   initially   occurred   at   a   low   frequency,   the   homozygous   lines   developed   from   elite   pedigree   breeding   programs   proved   very   useful   in   commercially   oriented   maize  breeding  programs.         For  commercial  breeding  activities,  speeding-­‐up  of  the  breeding  cycle  through  DH  technology  has  great   benefit   due   to   significant   reduction   in   resources   needed   for   line   development.     Commercial   breeding   companies,   through   the   use   of   DH   technology,   also   eliminate   Stage   1   testing   activities   on   early   generation  inbred  lines.  In  as  little  as  3-­‐4  seasons,  following  the  development  of  D1  lines,  the  value  of   the   new   lines   for   use   in   commercial   hybrids   can   be   evaluated   in   stress   screening   nurseries   and   multi-­‐ location   trials,   and   passed   to   the   commercial   seed   production   units   of   the   company   for   use   in   pre-­‐ commercial   hybrid   strip   plot   testing.   The   breeding   operations   of   several   large   multinational   seed   companies  are  currently  based  on  the  use  of  DH  lines  for  majority  of  breeding  activities.  During  2011,   Pioneer   has   reportedly   generated   more   DH   lines   than   the   total   number   of   inbreds   generated   in   the   first   80  years  of  their  breeding  efforts.  This  is  representative  of  the  multinational  seed  industry  as  a  whole.     The  emphasis  has  now  shifted  on  marker-­‐assisted  selection  (MAS)  and  high  throughput  phenotyping  of   the  newly  generated  DH  lines.     DH   improves   the   capacity   to   identify   breeding   materials   with   superior   performance   under   diverse   environmental  conditions   In   the   large   commercial   breeding   programs,   the   DH   lines   are   now   quickly   screened   using   molecular   markers   and   selections   done,   before   further   characterization   for   agronomic   performance   across   many   environments,   and   against   relevant   abiotic   and   biotic   stresses.   These   include   managed   stress   environments   (Campos   et   al.,   2004)   that   provide   information   on   yield   performance   under   less   than   optimum   conditions.   Evaluation   of   the   completely   homozygous   DH   lines   and   their   hybrid   products   provide   an   excellent   opportunity   to   link   phenotypic   performance   with   the   genotype.   Utilizing   a   commercial-­‐scale   DH   program   coupled   with   good   phenotypic   characterization   for   reaction   to   biotic   diseases,   Dow   AgroSciences   in   Brazil   rapidly   shifted   their   elite   inbred   disease   phenotypic   profiles   to   multiple   disease   resistance   by   rapid   recycling,   and   have   developed   a   strong   commercial   pipeline   of   multiple  disease  resistant  hybrids.  This  was  done  prior  to  the  routine  use  of  molecular  tools  to  assist  in   genotyping  the  DH  inbreds  (D.  Jeffers,  personal  information).             45   The   large   international   seed   companies   in   the   last   decade   have   made   huge   gains   in   their   genotyping   capacity,   and   realized   that   their   ability   to   phenotype   was   not   keeping   pace   (Campos   et   al.,   2004).     Therefore,   heavy   investments   were   made   on   improving   high-­‐throughput   phenotyping   capacity   to   evaluate   maize   germplasm   under   both   optimum   and   stress   conditions   using   “phenotyping   platforms”.     The   term   “phenomics”   has   been   used   for   the   whole   field   of   improved   phenotypic   characterization   through   the   use   of   modern   technology,   including   techniques   such   as   digital   imaging,   spectral   analysis,   and   canopy   temperature   measurements,   linked   with   bioinformatics   (Finkle,   2009;   González-­‐Pérez   et   al.,   2011;   Montes   et   al.,   2011;   Patil   and   Kumar,   2011).   These   techniques   have   been   used   to   examine   agronomic   performance   under   optimum   and   stressed   conditions,   both   for   abiotic   and   biotic   stresses,   and  provide  a  more  quantitative  measure  of  the  responses  of  the  germplasm.  The  improved  precision   has  also  provided  the  opportunity  to  better  understand  the  genetic  basis  of  response  to  various  stresses.     Linking  DH  with  other  technologies  to  accelerate  breeding  gains   Doubled   haploids   are   just   one   component   of   a   technological   package   that   has   allowed   commercial   breeding   programs   to   improve   their   breeding   efficiency,   and   increase   the   genetic   gains   per   breeding   cycle.    Examples  from  the  seed  industry  are  Pioneer’s  use  of  Accelerated  Yield  Technology,  AYTTM  System   which   encompasses     molecular   breeding,     bioinformatics,   doubled   haploids,   plant   genomics,   precision   phenotyping   and   decision   support   tools   to   develop   and   deliver   better   commercial   products.   Precision   phenotyping  provides  the  capacity  to  examine  the  phenotypic  response  at  the  macro  level,  but  also  at   the   molecular   level   once   an   understanding   of   the   genetic   basis   of   response   is   known.   Phenotyping   tools   such   as   “Proteomics”   (Liebler,   2002)   and   “Metalabolomics”   (Daviss,   2005)   can   then   be   used   to   better   characterize  the  germplasm.         Future  perspective   High   throughput   field-­‐based   phenotyping   with   reasonable   precision   plays   a   key   role   in   the   modern   maize  breeding  operations,  with  significant  advances  in  understanding  the  maize  plant’s  response  to  its   environment,   and   finally   agronomic   performance.     As   more   information   is   obtained   on   the   genetic   basis   of   this   response,   molecular   phenotyping   will   become   a   larger   component   of   the   phenotyping   process   that  predicts  genotypic  response  for  commercial  maize  products.  These  activities  can  be  carried  out  in   large   institutions   including   multinational   seed   companies,   since   it   requires   a   significant   investment   in   infrastructure.     Haploid   techniques   can   be   a   valuable   tool   for   the   rapid   production   of   homozygous   transgenic  plants,  thus  assisting  in  the  establishment  of  transformation  techniques.  Combining  haploidy   with  other  technologies,  such  as  MAS,  induced  mutagenesis,  and  transgenic  technology,  would  acceler-­‐ ate   crop   improvement.   Doubled   haploids   will   provide   the   products   to   facilitate   these   activities,   and   a   rapid  mechanism  to  deploy  improved  genetics  in  commercial  products.     References   Campos   H,   Cooper   M,   Habben   JE,   Edmeades   GO,   Schussler   JR   (2004)   Improving   drought   tolerance   in   maize:   a   view   from  industry.  Field  Crops  Res.  90:  19-­‐34.     Chang  MT,  Coe  EH  (2009)  Doubled  haploids.  In:  AL  Kriz,  BA  Larkins  (eds)  Biotechnology  in  Agriculture  and  Forestry.   Vol.  63.  Molecular  Genetic  Approaches  to  Maize  Improvement.  Springer  Verlag,  Berlin,  Heidelberg,  pp.  127– 142.   Chase  SS  (1947)  Techniques  for  isolating  monoploid  maize  plants.  Am.  J.  Bot.  34:  582.   Chase  SS  (1949)  The  reproductive  success  of  monoploid  maize.  Am.  J.  Bot.  36:  795-­‐796.   Daviss  B  (2005)  Growing  pains  for  metabolomics.  The  Scientist  19:  25-­‐28.   Finkle  E  (2009)  With  ‘Phenomics”  plant  scientists  hope  to  shift  breeding  into  overdrive.  Science  325:  380-­‐381.   Forster  BP,  Thomas  WTB  (2005)  Doubled  haploids  in  genetics  and  plant  breeding.  Plant  Breed  Rev.  25:  57–88.   González-­‐Pérez   JL,   Espino-­‐Gudiño   MC,   Torres-­‐Pacheco   I,   Guevara-­‐González   RG,   Herrera-­‐Ruiz   G,   Rodríguez-­‐ Hernández  V  (2011)  Quantification  of  virus  syndrome  in  chili  peppers.  African  J.  Biotech.  10:  5236-­‐5250.    Liebler  DC  (2002)  Introduction  to  Proteomics:  Tools  for  the  New  Biology.  Humana  Press,  Totowa,  NJ,  USA,  201  pp.     46   Montes   JM,   Technow   F,   Dhillon   BS,   Mauch   F,   Melchinger   AE   (2011)   High-­‐throughput   non-­‐destructive   biomass   determination   during   early   plant   development   in   maize   under   field   conditions.   Field   Crops   Research   121:   268-­‐ 273.   Patil   JK,   Kumar   R   (2011)   Advances   in   image   processing   for   detection   of   plant   diseases.   Journal   of   Advanced   Bioinformatics  Applications  and  Research  2:  135-­‐141.     Troyer  AF  (2004)  Persistent  and  popular  germplasm  in  seventy  centuries  of  corn  evolution.  In:  CW  Smith,  J  Betran,   ECA  Runge  (eds)  Corn:  Origin,  History,  Technology  and  Production.  Wiley-­‐Hoboken,  pp.133-­‐232.       47   9.    Access  to  Tropicalized  Haploid  Inducers  and  DH  Service  to  CIMMYT  partners   BM  Prasanna,  Vijay  Chaikam,  George  Mahuku,  and  Rodrigo  Sara     Access  to  tropicalized  haploid  inducers   Adoption  of  DH  technology  by  several  of  the  maize  breeding  institutions  under  the  national  agricultural   research   systems   (NARS)   as   well   as   small   and   medium   enterprise   seed   companies,   especially   in   the   developing   countries,   is   limited   by   the   lack   of   inducers   adapted   to   tropical/sub-­‐tropical   conditions.     CIMMYT's   Global   Maize   Program,   in   collaboration   with   the   Institute   of   Plant   Breeding,   Seed   Science   and   Population   Genetics   of   the   University   of   Hohenheim   (UHo),   addressed   this   limitation   and   now   has   haploid   inducers   ready   for   sharing   with   interested   institutions,   under   specific   terms   and   conditions   as   outlined  below.     The   tropically   adapted   inducer   lines   developed   by   CIMMYT   and   UHo   have   high   haploid   induction   capacity   (~8–10%)   and   were   found   to   exhibit   better   agronomic   performance   compared   to   the   temperate   inducers   in   the   CIMMYT   experimental   stations   in   Mexico   (Agua   Fría   and   Tlaltizapan).   A   haploid   inducer   hybrid   developed   using   these   TAILs   revealed   heterosis   for   plant   vigor   and   pollen   production   characteristics   under   tropical   conditions,   while   maintaining   similar   haploid   induction   rate   (~8-­‐10%).  CIMMYT  and  UHo  have  decided  to  share  the  seed  and  grant  authorization  for  use  of  one  of   the  tropicalized  haploid  inducer  lines  (one  of  the  parent  of  a  hybrid  inducer)  and  the  hybrid  inducer  to   interested  applicants  after  signing  of  the  relevant  Material  Transfer  Agreement  (MTA)  and  with  certain   restrictions  to  protect  the  intellectual  property  rights  of  both  institutions  for  the  inducer  lines.     Guidelines  to  obtain  tropical  haploid  inducers   The  general  guidelines  to  obtain  inducers  for  research  use  and  commercial  use  are  as  follows.       For  research  use  by  NARS:  The  NARS  institutions  interested  in  accessing  the  haploid  inducers  for  specific   purposes,  for  example,  for  development  of  DH  lines  for  use  in  breeding  programs,  may  send  a  letter  of   intent  or  expression  of  interest  to  CIMMYT.  For  eligible  NARS  institutions,  the  haploid  inducers  will  be   provided   free-­‐of-­‐charge   by   CIMMYT   and   UHo,   after   signing   of   a   Research   Use   MTA.   The   use   of   the   inducers  by  NARS  institutions  for  their  own  commercial  purposes  or  for  commercial  purposes  of  others   should  be  in  accordance  with  a  separate  license  agreement  for  commercial  use  (as  given  below).     For   commercial   use:   Applicants   may   access   the   inducers   for   commercial   use   pursuant   to   signing   of   a   Material  Transfer  and  License  Agreement  with  CIMMYT  and  UHo.  Each applicant  shall  pay  to  UHo  a  one-­‐ time   licence   fee   (US$   25,000)   for   provision   of   seed   of   two   haploid   inducers;   these   include   one   of   the   parents   of   a   tropicalized   haploid   inducer   hybrid   and   the   haploid   inducer   hybrid   itself.   If   the   applicant   wishes  to  access  the  other  parent  of  the  haploid  inducer  hybrid,  an  additional  one-­‐time  licence  fee  of   $10,000  will  be  payable  to  UHo.       Seed   of   the   above-­‐mentioned   haploid   inducers   will   be   provided   by   CIMMYT   to   the   applicant   normally   within   three   weeks   after   signing   of   the   MTA   (for   research   use)   or   Material   Transfer   and   License   Agreement  (for  commercial  use)  and  receipt  of  the  one-­‐time  License  fee,  as  relevant.             48   Maize  DH  service  by  CIMMYT  to  International  Maize  Improvement  Consortium  (IMIC)  partners   CIMMYT   recently   established   a   maize   DH   production   facility   at   its   experimental   station   in   Agua   Fría,   State  of  Puebla,  Mexico.  Through  this  facility,  a  DH  line  production  service  will  be  offered  to  members  of   the   International   Maize   Improvement   Consortiums   operating   in   Asia   and   Latin   America   (i.e.   IMIC-­‐Asia   and   IMIC-­‐LA)   on   a   cost-­‐recovery   basis.   For   information   as   to   how   to   become   a   Consortium   member,   please  contact  CIMMYT  Global  Maize  Program  Director.       At  the  Agua  fría  Station,   the  nursery   for   haploid   induction   in   the   source   materials   is   planted   in   late   May,   and   the   seed   is   harvested   by   September.   Haploid   seeds   are   identified   using   kernel   color   markers,   and   the   seedlings   are   subjected   to   chromosomal   doubling   immediately.   The   haploid   (D0)   nursery   will   be   raised  during  November–April.  The  D1  seed  of  the  DH  lines  will  be  processed  and  sent  back  to   partners   by  May/June,  following  the  necessary  germplasm  export  protocol.     Possible  models  for  breeding  programs  to  adopt  DH  technology   Model  1  –  full  service:  Under  this  scenario,  the  partner  sends  in  the  source  germplasm  (populations  for   developing   DH   lines)   and   CIMMYT   conducts   all   the   steps   (including inductions,   classification,   chromosome   doubling,   and   D1   seed   derivation)   that   are   needed   for   DH   line   development.   At   the   end   of   this  process,  CIMMYT  sends  all  the  seed  of  the  DH  line  (D1  seed)   that has been produced  back  to  the   client.   For   this   scenario   to   work   and   be   effective,   the   partner   should   consult   CIMMYT   in   advance   and   express  interest  in  sending  populations  for  DH  line  development.       Model  2  –  partial  service:  There  are  two  possible  scenarios  under  this  model:  (1)  the  partner  does  the   induction  of  haploid  kernels  and  sends  the  kernels  for  selection,  chromosome  doubling,  and  generation   of   DH   lines   in   CIMMYT’s   centralized   facility;   or   (2)   the   partner   does   the   haploid   induction   as   well   as   selection   of   haploid   kernels,   and   sends   only   the   haploid   kernels   for   chromosome   doubling   and   subsequent  generation  of  DH  lines  (D1  seed)  at  the  CIMMYT  facility.         How  to  indent  for  the  DH  service   At   present,   the   DH   service   facilities   at   CIMMYT's   Agua   Fría   station   can   handle   a   total   of   150   populations   per   year,   for   meeting   both   the   internal   and   external   demands   for   DH   line   production.   Interested   partners   can   submit   a   maximum   of   5   to   10   populations   for   haploid   inductions  and  DH  line  generation.     An   announcement   will   be   made   each   year   in   January   inviting   requests   for   DH   line   production.   Partners   wishing   to   utilize   this   service   need   to   sign   an   MTA,   with   CIMMYT   by   the   end   of   February.     After  signing  the  MTA  and  paying  the  necessary  charge,  as  applicable,  for  cost  recovery  (as  per   the   details   given   below),   partners   should   send   200   seeds   for   each   population   for   haploid   inductions   by   the   end   of   April.   Along   with   the   seed,   partners   need   to   provide   flowering   time   information  (especially  silking)  and  adaptation  (tropical/subtropical/highland).     If  partners  wish  to  send  only  sorted  haploid  seeds  for  partial  service  (chromosome  doubling  and   DH  line  generation),  the  haploid  kernels  should  be  sent  to  CIMMYT  (after  signing  the  MTA  and   paying  the  appropriate  service  fee)  by  01  October  at  the  latest.   CIMMYT   will   inform   the   partners   about   success   in   production   of   DH   lines   from   the   source   populations   received,   after   haploid   induction   and   DH   line   generation.   In   case   a   source   population   contains   the   kernel   color   inhibitor   gene   that   prevents   reliable   identification   of   haploid   kernels,   CIMMYT   will   inform   the   concerned   partner,   and   that   specific   source   population   will  not  be  further  continued  for  DH  line  production.  In  such  cases,  only  haploid  induction  cost   (US$  200  per  population)  will  be  charged  to  the  partner.     49   Partners  from  private  sector  institutions  need  to  pay  the  DH  service  fee  after  signing  the  MTA   and   before   initiation   of   DH   production   work.   Partners   from   public   sector   institutions   may   utilize   the  collaborator's  budget  for  DH  services  before  the  MTA  is  signed.  If  the  request  is  approved,   service   charges   will   be   deducted   internally   in   CIMMYT   from   the   collaborator’s   budget,   as   applicable.   Public   partners   without   collaboration   budgets   should   arrange   the   funding   for   DH   service  before  the  MTA  is  signed  and  work  is  initiated  by  CIMMYT.   A   cancellation   cost   will   be   charged   for   cancellation   of   any   indented   DH   service   work.   The  cost   charged  will  be  proportional  to  the  amount  of  work  already  undertaken  before  the  receipt  of  a   formal  letter  from  the  partner  requesting  cancellation  of  the  indented  DH  service.       Cost  recovery  for  DH  line  production  service   The  costs  for  complete  or  partial  service  will  be  as  indicated  below:   1) Only   haploid   induction:   US$   200   will   be   charged   for   each   source   population   subjected   for   haploid  induction.   2) Haploid  seed  identification  and  chromosome  doubling:  US$  25  will  be  charged  for  each  DH  line   supplied.   3) Only  chromosome  doubling  and  DH  (D1))  seed  production:  US$  22  will  be  charged  for  each  DH   line  supplied.   4) Complete  DH  service  (including  haploid  induction,  haploid  identification,  chromosome  doubling,   and  DH  line  production):  US$  30  will  be  charged  for  each  DH  line  supplied.     Note:  These  costs,  solely  from  the  cost  recovery  viewpoint,  may  be  reconsidered  and  possibly  revised  by   CIMMYT  each  year  depending  on  operational  costs.       For  further  details,  please  contact:   Dr  BM  Prasanna,  Director,  Global  Maize  Program,  CIMMYT  ([email protected])  or     Dr  Vijay  Chaikam,  DH  Specialist,  Global  Maize  Program,  CIMMYT  ([email protected]).       50   ISBN: 978-607-8263-00-4