Field Crops Research ELSEVIER www.elsevier.com/locate/fer Principles underlying genetic improvement for high and stable crop yield potential Vasilia A.Fasoula*Dionysia A.Fasoula Received 3 February 2002:accepted 7 February 2002 Abstract This pape netic basis of high and stable crop yield,and delineates the condition vide the link be through intralocus recombination,of two heterotic nonadditive allelic loci into a singl is due to the improved additive allelic complementation,that at the same time enhances epistatic interactions.The ful the abo to maxi atic crop yield and d can be lly achiev selection becc at the crtical distanceere the range of oypicp nb breedingu transition from s ingle-trait evaluation to whole-genome phenotypic evaluation.The concept of whole-genome phe a whole and belong to three (1)gene limit of th nt d esponsiveness to inputs.The outcome of selection.based on whole-genome pheno ypic valuation during all generations of oreeding program,is high yielding.stable,and density-independent cultivars.C2002 Elsevier Science B.V.All rights reserved Kewwords:Inbred vigor:Heterosis:Polyploidy:Resp nse to selection:Competition:Honeycomb breeding:Crop yield 1.Introduction Plant breeding aims at developing cultivars with high and stable crop yield potential,a process asso- ciated with both genetic and environmental factors. ee front atter2002 Elsevier Science B.V.All rights reserved
! "#$"$" % & '(' ) "" ! *! + ! " # $ % & ' ( & & ($ # )* + )* + )* & $ + + + # # $ $* # ( , -./ - / ( $ & $* & # & & $ ( ) + &+ # &+ * & , -./ $ & ! - / $ ! - / # &+ + 0 1 2 ,- . 3 ! ) ! ! ! 4! ) ! 4 & + & 4 "5 - / .6.7 6 4 #, .+" 8+59 + 6 ! $, .+" 8+5: +:. /# . ; ; - / ":+9 6 < <= 7 0 1 2 33, 1 ":+9 6 - / 8+8
192 VA Fasou D.A.Fasoul/Field Crops Research 75(02)191-209 The obiective of the r sent paper is to examine these tion is the development of high yielding.stable,and factors and discuss the principles underlying selection density indenendent cultivars These cultivars are efficiency in breeding programs.Our focus is on three favored by farmers because their yield is optimized main areas:(1)elucidation of the genetic basis of three over a wide range of plant densities. important phenomena related to heritability,i.e. To effectively select for the three components of vigor,degeneration,anc response to selection;(2) crop yield potential,a number of requirements need to that optimize response (3)tra -trait evaluation fulfilled:(1)comparisons on the basis genome phenotypi c evaluation sho not ch have the insights progeny the ha the thre oonents should be applied acr ross the plant-to-plant interference with the equal sharing of target area of adantation from the earliest generation growth resources caused by genetic and acquired and during all stages of the breeding program:(3) differences.and the clarification of its negative role honeycomb designs should be utilized to sample on two important aspects:(1)crop yield and (2) effectively for environmental diversity and exploit efficiency of sel he genotype-by-environment interaction for the xplains the ons that led to th ce the or h line hybrids.or The scientific reaso in t is essential to ark that“g otential underlyins the superiority of inbred lines is discussed concems the yielding capacity of a cultivar under extensively.The negative effect of competition on production environments that may range from stress- selection efficiency explains why the honeycomb ful to optimal. breeding advocates selection at the level of the indi where the nd the The phenotypic iority on which selection is link between single-plant performance and crop yield, based.may be due to hybrid vigor,inbred vigor,or through the partition of crop yield potential into both.Hybrid vigor is the superiority of the F over the genetic components that can be accurately assessed two inbred parents and results from heterozygosity at the single plant level,from the earliest gene rations whereas inbred vigor is the superiority of the recom of a breeding program(Fasoula and Fasoula,2000) binant inbreds over the two parents s and the Fi anc The firs ent of con P These gene、 e gene as sugges 74),and refer h plant dens ra The second compo cenes con 'inbred vigor" .will be used pr ntially.as it p trolling tolerance to the biotic and abiotic stresses to the cause of the phenomenon whereas "transgres These genes are assessed by the genotype's progeny sive vigor"is a descriptive term.Inbred vigor is due to standardized mean and expand the upper limit of the recombinants that outperform the hybrid,and by plant density range for optimal productivity.The third definition is superior to hybrid vigor.Inbred vigor represents the condition where the homozygote out forms the heterozygote and generates response to shows a plex locus nsible for the rotein.The two closely
# > $ ( ? , -./ ! - / * ! - / + &+ 4 ( # # - 0 - 1 0 ( & , -./ - / ( # + $ @ # ( $ # ( $ & + & & $* $ ( -4/ + * # &+ % & + - / # ( # A $ & # + # A * $ # + # + A * $ & # &+ + + # * & # ' ( , -./ &+ $ & ! - / ! - / * $ + + 1 + & + ( % BB AA + # & ) . & * & + & . * # BB AA & -.6" .6"9/ BB AA 1 -.65 / )& BB AA & & BB + AA 3 ( 3 & * + * . & $ CD # & .6 2' ' 3 ' ' 3 4 3 % $* 5" "6 78" 7
VA.Fasoula.D.A.Fasoula/Field Crops Research 75 (2002)191-209 193 HYBRID VIGOR REPULSION COUPLING ve 0 may con 0 409%Yz 40%Y 80%6 YZ RATION RES TO NSE 0 SELECTION 80%YZ Fig.1.The self-fe (() nonadditive ation that on-pha that generate and re s inbred vigo linked genes are partially dominant and com Dared zygous state of partial dominant genes may produce with the dominant alleles that produce 100%of the the dimeric YZ protein either through trans-comple- protein,they produce only 40%of the required mentation of the linked genes (trans-acting DNA amount.At the molecular level,these events translate sequences)or through cis-complementation (cis to changes either in the coding or in the regulatory acting DNA sequences).Trans-complementation gen- regions of a gene.As shown in Fig.1.the hetero- erates the repulsion-phase vigor(or repulsion vigor)
% & . E 9 E ' & . + * CD ++ % -+ F ' / + -+ F ' / 9+ + + - / . # ' +* #& $ CD & , -./ + ! - / + ! - / F -9/ F 2' ' 3 ' ' 3 4 3 % $* 5" "6 78" 7 .6
194 VA.Fasoula,D.A.Fasoula/Field Crops Research 75(202)191-09 which is nonfixable.as the action of the two linked inbred vigor the three coexistine genes is interdenendent or nonadditive However kinds of vigor may oceur at any locus in a nonulation intralocus recombination may convert an interdepen- and the predominance of one kind over the other at the dent into an independent allelic action that generates genomic level depends mainly on the reproductive coupling-phase vigor (or coupling vigor).Coupling system. Thus,the asexual reproduction and cross vigor is based on independent allelic action and ger fertilization favor the predominance of repulsion vigor erate vigor through gen nxation.In vigo variation,whereas self-te the gene coupling and to selection due to superior to bothr epulsion and co is onsible for advance through selection.Im natural selection chose distinct pathways.Artificial allelic action and interaction have also an indirect selection overcame the lack of response via the route favorable effect on vigor.by optimizing epistatic of coupling and inbred vigor.This explains why during domestication,there is a clear transition from and response to election are generatec autogamy (Schwanitz. nd/or to ecin having of inb the han 80 yea n日main d tha laced in this lastc nside preponderance of additive genetic variation in this partial dominance can hardly be denied.As shown in crop (Hallauer and Miranda,1981).Conversely.nat- Fig.1.the fixation of loci having an independent or ural selection.apart from the route of inbred vigor additive allelic action (partial dominant alleles)is a overcame the lack of response by exploiting the prerequisite to selection.Indeed.the advantages o Polyploidy overcomes the amount pro duc under n ba r repu 1)1 n inbre (2 80 ount o g do ee that the euolution of the of the additive gen ariation:(3)fixing b superiority upon which selectionand e self-pollinated allopo the based.starts with the nonfixable repulsion vigor,that levels of heterosis in cross-pollinated autopolyploids. converts through intralocus recombination into the The increased levels of nonadditive genetic varia- phenotypically equivalent,but fixable coupling vigor. tion in autopolyploids have been docume ted by This in turn converts,through gene fixation,into a Peloquin (1995)in autotetraploid potato transgressed form of vigor;i.e.,the inbred vigor: L)andB1 ngham（9 )in autote recombination a (Me an inbred vigor coupli ectio d fo The first two kinds of vigor are those that gave birth to Their difference is that the for the overdominance and the dominance theories of and constant progress through selection.whereas the heterosis,respectively.Here,we would like to empha latter is accomplished at a single step.The superiority size that a reliable explanation of heterosis should of inbred vigor is due to the systematic removal of consider the two kinds of vigor as being mutually defe ctive genes and their replacement by favorable additiv presupposes the The thhat citing data fro mFA0(198 g of the 15 volution of hybrid vigo into s and only74%bong ed op ed to five
& ($ & % )& + + - / 4 + ($ 3 + # 3 * 3 ( ! 3 % G -.6."/ ( ( : ) + & . ($ - / ' 3 CD -9 E/ -: E/ H & & ($ ' ($ # ($ ! , $ # ( & % ) & & % + * $ & % $ # % $ # $ % % # $ + * & ++ * # % ( & ( % # $ & -1&* .68"/ < + - / @* -: I/ -) @ .6:./ 4 + % $ , -./ F! - / % & + & $ ! - / ($ + ! -9/ + # + ' -.665/ - 1 I/ 2 -.66:/ - I/ 3 & # & # -.6::/ ? -.6: / : 8E & + .5 ."9E ( .69 2' ' 3 ' ' 3 4 3 % $* 5" "6 78" 7
VA.Fasoula.D.A.Fasoula/Field Crops Research 75 (2002)191-209 195 und ion wei ageneaofi 240 1150 ,1998).Bi 10081 Extensive efforts throughou it the 20th centur ome blocks in the ex improve the vield potential of crons were generally sion of heterosis.confirms the conclusion reached b successful.Research that assessed the yield gains lones (1917)that dominance of linked factors attributed to genetic improvement,revealed an annual accounts successfully for heterosis.He furthermore genetic gain ranging from 0.5 to 1.5%depending on attributed the yield gains in autotetraploid alfalfa to the crop and the conditions ofexp ation(Wych pseudo-overdominance.excluding the presence of nd K and 198 984 (1988 ased the CV of single-plan 1984:Waddington et al..1986:Feil.1992:Tollenaar of two otato cultiv et al..1994:Cox et al..1988).The crops under research were harley (Hordew vulgare l)oats (Avena sativa L.).wheat (Triticum aestivum L.).sor oe e ghum (Sorghum bicolor L.Moench).soybean (Gly CV values of two tomato (Lycopersicon ax (L. Mill.)F2's,whose CV of single- m oved crop com 087 gene Chri 001 The of additive genetic variation clonally ated tato cultivars and the inbre In contrast in crons characterized by nonadditive potato exploits mainly nonadditive allelic variation genetic variation,like the clonally propagated potato whereas that of tomato.additive allelic variation.This (Peloquin,1995).selection for productivity has failed explains why in inbreeders,the genetic advance to esta h ger ic gain.Results reported by Douche through selection is realized even when the genetic bas very pre Phil that a(2000 principal uggested s of the ploid p tato al the difficulty to combine uality and re vide ime to rt re pulsion-in with vield potential and the narrow ge netic has of cultivated potato in North America.We believe deleterious genes This can be accomplished by though.that the principal cause for the reported yield exploiting the advantages of both inbred and hybrid stasis has to be explored in the high load of deleterious vigor.as it is presently done with maize (Duvick. on-pha nkages. 1997 strategy cannot degree of degeneration wher up ne p to the loss of vigor follo heransnong allelic conver loitatio to he Alfalfa is another crop that has failed to Hooker (1968)in maize and Flor (1965)in flax (Lim genetic gain through selection because of the high load usitatissimn i)screened large numbers of progen of deleterious genes.Indeed,the genetic contribution and were able to convert codominant nonfixable to yield in alfalfa over an 80-year period f 0m1898t effects into dominant effects that were fixed upon 1985 was only3%(Hill et al 1988 Holland an inbreeding.These results have far-re hing breeding Bingham,199).Elimination of deleterious genes and implications,because they suggest that the yield o
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
196 VA Fasou D.A.Fasoul/Field Crops Research 75(02)191-209 inbreds in maize and other nay be raised at inbred line perfon ticed effectively higher levels by converting ulsion into coupling deleterious alleles will be replaced by additive alleles vigor and coupling into inbred vigor through intralo- increasing the productivity of inbred lines(Fasoulas, cus recombination and fixation(Fasoula and Fasoula, 1997:Fasoula and Fasoula,1997b).Fixation of het. 1997b).In fact,Genter (1967)identified two inbred sis and improvement of the productivity of inbrec lines in maize that showed little inbreeding depres- ines have already been accomplished in maiz sion (Tokatlidis t al. 98 999),nce (Ntar aing the genetic basis of 20 ,1998 by the and Eas and mato(Christ 20011R pla nt by fo inbred lines at the other,arrived at the following wo advantages:(D the amelioration of the additive conclusion "True overdominance has been shown allelic action responsible for inbred vigor and genetic to have no demonstrable role in hybrid vigor.although progress through selection.and(2)the steady transmis spurious overdominance arising from a linkage dis sion of the accomplished amelioration from generation equilibrium with or without superimposed nonallelic to generation through sexual reproduction. 3.Conditions optimizing response to selection ge Genetic imn vement is closely linked with the Response to selection.or genetic gain.is the elimination of the undesirable action of deleterious improvement of the population mean as a result of genes,first at the allelic and,by extension,at the selection.The simplest form of selection is to choose epistatic level.Elimination of the undesirable action is individuals on the basis of their own phenotypic accomplished by two means:either through the con favorab 品 Ihe parameter nse (R)in one terious eOnt9breame.culivan.Thedevelor ment of modern hybrids involves the difficult task of discover- R=aphi ing inbred lines that cover mutually their defects,a fact that renders the whole process a numbers game. uth n r henpuy the coefficient of heritability,h=/ Conversely,the development of inbred line cultivars and i the standardized selection differential, should aim at a constant cement of less favorable ap,with eing the nean yield of tr by more favorable addit e allele proces of the beaaiaceentO the ie the of the selected point is reached where hybrid developr ent is no tion to be used as ts Therefore the sm ner the more cost effective (Fasoulas,1997).When this has proportion of the selected parental plants.the large been accomplished,cultivar production should be the standardized selection differential and,by exten targeted not toward developing hybrid cultivars,but ion,the higher the expected response through selec- toward boosting the yield of inbred lines beyond that ton is. of hybrids. The conditions that optimize response toselectio of vigorous inbred li hat approa hybri ar心，in gencof E bilit for inbred line anc ection (2)he e pe of high
* + ($ - .66"/ 3 L -.68"/ ( & * & + G% -.6:./ ( + . $ & , BB# & % + ' & & # & + + AA L % & ( $ 0 & , + + # ( $( & # ( % + & 4 # + & & - .66"/ H & & & ? & - .66"! .66"/ $ + * -#% .66: .666/ -F .66:! F % ./ -4 % ./ & , -./ - / + $ # & & $ & # $ -%/ + - .6:6/, % -./ & + ( * ) ) & ) # * & + # * $+ $ + # * 0' -./! , -./ + -/! - / - /! - / - / .68 2' ' 3 ' ' 3 4 3 % $* 5" "6 78" 7
VA.Fasoula.D.A.Fasoula/Field Crops Research 75 (2002)191-209 197 Phenotypic variance is partitioned into three main that attributed large and heritable components: phenotypic varation 3.1.Maximizing phenotypic variance whereis the phenotypic variance.the genetic the viro The mean ()the phenotypic standard deviation and ype- (p).and the CV of single-plant yields are affected differently in the absence vs.the presence of competi- gful if the la tion(Kyriakou and Fasoulas,1985).When the same the ed to the environme rye (Secale cereale L.)population was compared at and the genotype-by-environment interaction var- two dif ent pl 1.4 plant iances.When and are significantly minimized, heritability approaches the maximal value of 1Eq.(3). e vas reduced by 70 and the Cy wa ased by 03 61%(Fig.2).In other words,the competition envir 2= ++0x 3) onment reduced the mean and standard deviation 7.5 Evidently.the conditions that allow to maximize CV application of high Similar results were obtained by Hamblin et al ction pressures are toog as compared cm.The mear NIL-COMPETITION COMPETITION (1.4 plants/m2) (51.3 plants/m2) 112.4g 14.9g 37.5g 0, （7.9g 33%4 CV →53% oes the range of yield ex n.In the of etition the range of yield expression is maximized and
, - / & + + 0' - / + + + H ( * $ . 0' - / - / 0 & $* & ! '' )= # ) -/ 4 + + -J % .6:5/ H - I/ & & .9 < -+ / 5. < -/ & :"E & "6E 4 & 8.E - / 3 & + "5 9" & 4 + .8 1 & ) -.6":/ & & ++ 9 9 # @ ) -/ 4 + & & ! .9 < -+/ 5. < -/ # $ 3 $ $* 4 + - J % .6:5/ 2' ' 3 ' ' 3 4 3 % $* 5" "6 78" 7 .6"
198 VA.Fasoula,D.A.Fasoula/Field Crops Research 75(202)191-09 C(1969 CV(1970) 29% 41% 51% 32% 46% 56% 110 ↓ 1107 100 100 90 70 70 60 50 6 40 30 30 20 20 10- 0 0 9.3 12.4 62 9.3 124 Plants/m2 Plants/m2 Fig.3.The CVof single-plant yields of two maize hybrids inc with the ine se of plant density.This is due to the fact that the mean() is reduced aa faster rate compared to the standard deviation((data from Daynard and). and standard deviation (2)increasing ctively, g the undesirable plant-to- (increased CV).Co 'alupbatvanation of s The increased CV under competition is due to the higher reduction rate of compared to the rate of two important things:(1)it magnifies the yield differ This conclusion is further supported by results ences among means (large x)and (2)it increases the depicted in Fig.3 that show the mean grain yield reliability by which the means are estimated(small per plant ()the phenotypic standard deviation() CV). Apart from cing response to selection. Muldo among de 997a. st defin r rate mpetition:i the plant--pan inerfer than The rapid reduction of the mean with the the equal sharing of growth resources caused by increase in plant density compared to the relatively genetic and acquired differences and measured by slow reduction of the standard deviation,indicates that the CV of single-plant yields.Competition reduces the mean is a much more sensitive parameter for crop yield because competitive gains do not counter measuring how much the competiti reduces the alance competitive losses.Theoretically,crop yield xpre when comp tally of ced plant-to o CV.Results of a
& 9. 8 & 4 & .8 # 4 ) # & ) -/ 4 + & * & 8 6 . 9 < - @ .6: / & 4 & ) # & & & $ ) + + , -./ $ - ) / - / ++ - 4/ 4 & , -./ ( + - )/ - / & - 4/ & - .66"/ # & ( ( ! # # - 0 - 1 0 1 2 # 4 + # & $* & * D + & ' & * ++ * 4 # 4 + & * & # ) -/ - @ .6: / .6: 2' ' 3 ' ' 3 4 3 % $* 5" "6 78" 7
VA.Fasoula.D.A.Fasoula/Field Crops Researeh 75(2002)191-209 199 a newer s Tolles hybr on eran Wu and d by rtad th ahpationh ording duced grain yield per unit area due to the lodging.heat and drough insect and disea increased interplant competition.Yield reduction was greater yield stability to the new maize hybrids.and larger in the double-cross hybrid due to the enhanced allow them to withstand the greater demands put on genetic competition within double-cross hybrids. individual plants by dense planting.The evidence in compared to the absence of genetic competition within maize,therefore,suggests that selection applied over a single-cross hyb approaching the to stres lized equal ocal an h plan dent inheritance and select jointly for the two considered essential.One is the use of monogenotyp ances,the result would be density-dependent hybrids cultivars,to minimize the unequal sharing of resources that are at a disadvantage because their productivity among plants caused by genetic differences.The other depends on a specific plant density. is the development of monogenotypic cultivars with The possibility to develop density-independen high individual buffering.to minimize the unequa hybrids o great advantages y acquired and ac ns of the (1)the prids with the inal The ad ave for the b eder is that it and (2)the extensiv al tests of ngle vides further eviden ce that evaluatio n and selectio hybrids for improving the tolerance stresses.including tolerance to higher plant densities. yield expression,and allows one to select effectively The latter has been confirmed by Duvick(1984)and for reduced plant-to-plant variation or stand unifor Tollenaar (1992). who shov ed that modern hybrids mity.In fact,Tollenaar and Wu(1999)have shown tha idlanthpopuationdtensitic tic impro nt in malze has been the newe plar on hicnpnie et al n 1007 ide The supe and envire nts the on for and e and Rosenbrook (1983)suggested selection under environmental stresses. higher plant densities as a means of improving crop Fasoula and Fasoula (2000)emphasized that the yield.This is a logical and interesting suggestion.but real need of the armer is the development of density independent cultivars:i.e..cultivars that yield opti ce to ghe ange o plant de hig e on due to the shed rest tolerance to str rde e of co vield pe high nlant density the cause or the effect?if tolerance plant expands the lower limit of the plant density ra to high plant density is the cause and tolerance to for optimal productivity.whereas selection for toler stresses is the effect,then we should be able to ance to stresses,expands the upper limit.Duvick incorporate genes for tolerance to stresses by selecting sis the cause and has neither increased nor decreased during the pas
+ * & & + # H -.666/ C & + & + & + # + * ' * + & ? * ' # & * ' ' 3 * & & , -./ + + & + - / $ + # ( % -.6:9/ # -.66 / & & 3 & & -# .66"/ # * % # % -.6: / # & ( & & ' , H M 3 & M 3 & 4 + & % -.6:8/ & * & & # * & & + & F & + > & + & + ( # + & # # $* $ & ++ + 3 # H -.666/ & * ++ & + & # & - / * + ! + & # 1 $ & & + $ % -.66"/ & * & & . < 2' ' 3 ' ' 3 4 3 % $* 5" "6 78" 7 .66
200 VA.Fasoula,D.A.Fasoula/Field Crops Research 75(202)191-09 has increased dramatically as plant density increased from 3 to 7.9 plants/m2.Evidently,maize breeders have successfully expanded the upper limit of the opti mum plant density range by selecting for increasec tolerance to r,by not selecting to increased f the POP UL AT IONS hybrids. Tokatlidis et al.(2001)showed that combined selection in the absence of competition for high yield ASEXUAL HYBRIDS potential per plant and tolerance to stresses accelerates the development of density-independent maize Fig.4.The four basi categories of cultivars are depicted a the odf s bo on progeny tion.The ines are c by ther standardized mean CTokatlidis et al (adapted from Fasoulas.198). 1998.1999).Twelve inbred lines were chosen to develop six hybrids that were in essence "recycled" recombination (Thuriaux.1977:Hulbert.1998) since they were derived from PR-3183.The recycled Regarding the removal of deleterious genes and the hybrids were found to be les lensity-dependent than ixation of desirable genes,the ideal system of sexua the original hybrid PR-318 because they reproduction is self-fertilization.Under cross-fertili ow plar caton,gen I through lar equa at ine trolle (2)application ve stre lants 4 and stability of performance of maize hybrids in the nlant-to.plant acings that exclude inte future is the development of density-independent tition and facilitate self-pollination:(5)selection for hybrids.A prerequisite for developing density reduced anthesis-to-silking interval to facilitate self independent cultivars is the application of whole fertilization (only maize). genome phenotypic evaluation,1.e.,joint selection The evolutionary course for the thr nts of crop yield sible only domes ton is depicte when applie id othing:(1 3.2.Maximizing genetic variance the population requred to lish a c tain level The genetic variation,essential for both inbred of success.For example,the size of the sampled vigor and advance through selection,is enhanced by population for an autogamous crop (i.e..tomato) sexual reproduction.Sexual reproduction allows the differs greatly from the one required for an asexua repulsion into coupling vigo hybrid (i.e..potato).Usually,a tomato F2 populatior recombination.and of coupling gh gene hxation (Fig.1). ng t hybrid to ha th fre of the ared to the freque ency of interge imcrease to 1.000.000 et al 1992).This
" C "6 & '"' )= # $ 1$ & ($ - ./ # % ' ' -# $ .6""! ) .66:/ ($ $ +* N ++ * ($ , -./ +! - / ! - / - $ /! -9/ ++ $ + +! -5/ ++ % + * - */ # 9 # & , -./ ! - / * ' $ * - / ' $ - / N -H .656/ )& * . -# .66 / # 9 # # # * - .6::/ 2' ' 3 ' ' 3 4 3 % $* 5" "6 78" 7