Genetic basis of trichome production in Arabidopsis lyrata

He edi as 136: 219–226 (2002) Gene ic basis o ichome p oduc ion in A abidopsis ly a a KATRI KA 8 RKKA 8 INEN 1 and JON A , GREN 2 1 Depa men o Biological and En6i onmen al Science,Uni6e si y o Jy6a¨skyla¨,Jy6a¨skyla¨,Finland 2 Depa men o Plan Ecology,E6olu iona y Biology Cen e,Uppsala Uni6e si y,Uppsala,Sweden Ka¨ kka¨inen, K. and A , g en, J. 2002. Gene ic basis o ichome p oduc ion in A abidopsis ly a a.—He edi as 136 : 219–226. Lund, Sweden. ISSN 0018-0661. Recei ed No embe 2, 2001. Accep ed May 29, 2002 Lea ichomes may p o ec plan s agains he bi o ous insec s, and may inc ease ole ance o d ough and UV- adia ion. The pe ennial he b A abidopsis ly a a (B assicaceae) is polymo phic o ichome p oduc ion and occu s in a glab ous and ichome-p oducing o m. In addi ion, he e is quan i a i e a ia ion in ichome densi y among ichome-p oducing plan s. To examine he gene ic basis o glab ousness, we conduc ed con olled c osses wi h plan s o igina ing om wo na u al popula ions in Sweden (one polymo phic o ichome-p oduc ion, and one consis ing o glab ous plan s only). In addi ion, we es ima ed he he i abili y o ichome numbe om pa en -o sp ing eg essions o plan s o igina ing om he polymo phic popula ion. C osses be ween glab ous plan s esul ed in glab ous o sp ing only, whe eas c osses be ween glab ous and ichome-p oducing plan s, and c osses be ween ichome-p oducing indi iduals, esul ed in ei he all ichome-p oducing o sp ing o bo h pheno ypes. In seg ega ing c osses be ween ichome-p oducing plan s, he a io o glab ous: ichome-p oducing o sp ing did no de ia e signi ican ly om 1:3, while in seg ega ing c osses be ween glab ous and ichome-p oducing indi iduals he a io did in mos cases no de ia e om 1:1. Wi hin- and be ween-pop- ula ion c osses ga e simila esul s. The he i abili y o ichome numbe es ima ed om eg ession o o sp ing on mid-pa en was high (h 2 9SE, 0.6590.15). The esul s sugges ha glab ousness is inhe i ed in a simple Mendelian ashion, wi h he allele coding o ichome p oduc ion being dominan o e ha o glab ousness. They u he indica e ha glab ousness is due o a mu a ion a he same locus in bo h popula ions. Ka i Ka¨ kka¨inen,Finnish Fo es Ins i u e,Van aa Resea ch Cen e,P.O.Box 18 , 01301 Van aa Finland. E-mail: [email p o ec ed] The gene ic basis o adap i e ai s de e mines how hey can be moulded by selec ion, bu is s ill poo ly known. Many adap i e cha ac e s a y con inuously and ha e wi hin a quan i a i e-gene ic amewo k been assumed o be in luenced by a la ge numbe o genes, each wi h a small e ec . Howe e , ecen s udies ha e sugges ed ha adap i e ai s can be con olled by ew genes wi h la ge e ec s (S CHEMSKE and B RADSHAW 1999; J OHANSON e al. 2000; F RARY e al. 2000; W EBER e al. 1999). On he o he hand, some ai s ha ha e been hough o ha e a simple monogenic inhe i ance ha e been ound o be much mo e complex han ea lie was belie ed (S CRIVER and W ATERS 1999). Addi ional s udies o he gene ic basis o adap i e cha ac e s a e he e o e clea ly needed. T ichomes a e uni- o mul icellula hai s ha de- elop on he lea es, sepals, and s ems o plan s, and may se e se e al unc ions. They may p o ec plan s agains UV- adia ion and d ough (E HLERINGER 1984; S KALTSA e al. 1994; E SPIGARES and P ECO 1995), and also agains damage caused by he bi o es (L EVIN 1973; A , GREN and S CHEMSKE 1993; V ALVERE e al. 2001) and pa hogens (H AGLEY e al. 1980). The p oduc ion o ichomes may hus be bene icial o se e al easons, bu can also be associa ed wi h a cos : In en i onmen s whe e ichomes do no con e an ad an age, glab ous plan s may be able o alloca e mo e esou ces o g ow h and ep oduc ion (A , GREN and S CHEMSKE 1993; B ERGELSON 1994; M AURICIO and R AUSHER 1997; M AURICIO 1998). Gene ic a ia ion in lea ichome numbe has been de ec ed in se e al membe s o he B assicaceae, in- cluding A abidopsis haliana (M AURICIO and R AUSHER 1997), B assica apa (A , GREN and S CHEMSKE 1994), and Sinapis a 6ensis (R OY e al. 1999). Mu a ions a ec ing ichome o ma ion ha e been iden i ied in A. haliana and many o hem ha e also been cloned and sequenced (M ARKS 1997). T i- chome densi y may u he be in luenced by en i on- men al condi ions (A GRAWAL 1999; R OY e al. 1999). The ou c ossing pe ennial he b A abidopsis ly a a is polymo phic o ichome p oduc ion and occu s in a glab ous (G) and a ichome-p oducing o m (T). In addi ion, he e is conside able a ia ion in i- chome densi y among ichome p oducing A.ly a a. To in es iga e he inhe i ance o glab ousness and whe he he gene ic basis o glab ousness in A.ly a a is consis en ac oss popula ions in Sweden, we con- duc ed c osses be ween glab ous and ichome-p o- ducing plan s o igina ing om wo na u al popula ions sepa a ed by abou 32 km (one polymo - phic o ichome-p oduc ion, and one consis ing o glab ous plan s only). We documen ed he seg ega- ion o glab ousness in he esul ing o sp ing. We u he es ima ed he he i abili y o ichome numbe K.Ka¨ kka¨inen and J.A , g en220 He edi as 136 (2002) om pa en -o sp ing eg essions o plan s o igina - ing om he polymo phic popula ion. MATERIAL AND METHODS The species A abidopsis ly a a (L.) O’Kane and Al-Shehbaz ssp. pe aea (B assicaceae, syn. A abis pe aea L., and Ca daminopsis pe aea (L.) Hii .) is a pe ennial, sel - incompa ible he b, closely ela ed o A. haliana L. (P RICE e al. 1994; K OCH e al. 1999; S AVOLAINEN e al. 2000). Fou geog aphically sepa a ed subspecies ha e been iden i ied in A.ly a a:A.ly a a ssp. pe- aea in Eu ope, A.ly a a ssp. ly a a in No h Ame - ica, and A.ly a a ssp. kamcha ica and ssp. kawasakiana in eas e n Asia (O’K ANE and A L -S HEH- BAZ 1997). A abidopsis ly a a has a disjunc dis ibu ion in Eu ope. I is a common species in Iceland and in he Fa oes, and is ound in he no h-wes e n pa s o he B i ish Isles, in sou he n Ge many and in he Alps (J ALAS and S UOMINEN 1994). In Scandina ia, A. ly a a occu s only in wes e n No way and in a limi ed a ea along he coas o eas e n Sweden (H ULT e´ N 1971). The 5–20 cm long in lo escence is p oduced om a lea ose e. The ose e lea es a e ypically 3–10 cm long and 0.5–1.0 cm wide. A abidopsis ly a a may p oduce ichomes on lea es and s ems. Lea i- chomes o m up o i e b anches, whe eas hose on he s ems a e p edominan ly unb anched. Among ichome-p oducing plan s, he e is conside able a i- a ion in he densi y o lea ichomes, om plan s p oducing only a ew ichomes on he lea ma gin o hose wi h lea es densely co e ed by ichomes (J ON- SELL e al. 1995). Gene ic basis o glab ousness To examine he gene ic basis o glab ousness, we conduc ed con olled c osses using plan s p oduced om seeds collec ed in wo A.ly a a popula ions in eas e n Sweden. One popula ion was polymo phic o ichome p oduc ion and loca ed a S o s ensudden (63°12%N, 18°48%E; p opo ion o glab ous plan s abou 30 %), while he o he popula ion consis ed o glab ous plan s only and was loca ed a No - a¨lls iken (62°35%N, 18°32%E; p opo ion o glab ous plan s 98 %). Seeds we e collec ed om glab ous (G) and ichome-p oducing (T) plan s and sown in a g eenhouse in he au umn 1997. The ol- lowing sp ing, he pheno ype o lowe ing indi iduals was sco ed as glab ous (no ichomes obse ed) o ichome-p oducing. Plan s we e andomly assigned as ma e nal and pa e nal pa en s o he c osses. We c ossed he wo mo phs in all combina ions (G×G, G×T, T×G, and T×T), and conduc ed c osses bo h wi hin and be ween he wo popula ions. The c osses we e epea ed wi h di e en ma e nal and pa e nal plan s. The seg ega ion o glab ousness was examined among he o sp ing om 43 c osses (38 wi hin-popu- la ion c osses and 5 be ween-popula ion c osses). Seeds we e sown in a g eenhouse in ea ly sp ing 2000 and he pheno ypes o he seedlings (al oge he 822 indi iduals) we e sco ed h ee mon hs a e sowing. Twen y o se en y indi iduals we e sco ed om each o 18 c osses, while 8–10 indi iduals we e sco ed om each o he emaining 24 c osses. C osses we e classi ied as seg ega ing when he o sp ing included bo h glab ous and ichome-p oducing plan s, and non-seg ega ing when all o sp ing we e o he same pheno ype in e ms o ichome p oduc ion. Based on he esul s o he c osses (see below), we pos ula ed a simple Mendelian model o he inhe i- ance o glab ousness (a one gene, wo-allele model in which he allele coding o ichome-p oduc ion is dominan o e he allele coding o glab ousness). The i o he model was e alua ed by es ing he ollowing p edic ions: Fi s , i he glab ous pheno- ype (G) is a ecessi e homozygo e, c osses be ween glab ous indi iduals (G×G) should p oduce only glab ous o sp ing. Second, c osses be ween i- chome-p oducing indi iduals (T×T) should p oduce ei he only ichome-p oducing o sp ing i a leas one pa en is homozygous o ichome-p oduc ion o , i bo h pa en s a e he e ozygous wi h espec o he gene coding o ichome p oduc ion, p ogenies wi h a 1:3 a io o glab ous o ichome-p oducing indi iduals. Thi d, he o sp ing om c osses be ween glab ous and ichome-p oducing indi iduals should seg ega e in a 1:1 a io o glab ous o ichome-p o- ducing plan s i he ichome-p oducing pa en is a he e ozygo e, and all o sp ing should be ichome- p oducing i he ichome-p oducing pa en is ho- mozygous o he dominan allele. Finally, i seg ega ion a ios do no di e be ween ecip ocal c osses (G×T sT×G), ichome p oduc ion is likely o be go e ned by nuclea genes a he han by cy oplasmic ac o s. C osses be ween popula ions (G×G and G×T) we e conduc ed o assess whe he he gene ic basis o glab ousness was simila in he popula ions. Fo seg ega ing c osses, de ia ions om expec ed a ios o glab ous o ichome-p oducing plan s we e examined wi h G- es s. This was done sepa a ely o each c oss wi h a la ge numbe o o sp ing (N=20– 70), and o da a pooled o e each c oss ype (T×T, G×T, T×G). Fo c osses wi h la ge numbe s o o sp ing, he da a on pheno ypes o ma e nal and pa e nal g andmo he s (de e mined in he ield when Gene ic basis o ichome p oduc ion in A abidopsis 221He edi as 136 (2002) Table 1. Summa y o he esul s o 42 con olled c osses be ween glab ous ( G ) and ichome-p oducing ( T ) A abidopsis ly a a ssp.pe aea.The plan s used in he c osses o igina ed om wo na u al popula ions in eas e n Sweden ( No a¨lls6iken ( N ) and S o s ensudden ( SU )) .In seg ega ing c osses,bo h glab ous and ichome-p oduc- ing plan s we e obse 6ed among he o sp ing O sp ing pheno ypeO igin o Numbe o c ossesType o c oss To al numbe à×ßo o sp ingin non-seg ega ingpa en al plan s Seg ega ing Non-seg ega ingà×ßc osses N×NG×G09 1) G 208 03 GSU×SU 32 N×SU 0 1 2) G42 48 3) T 250T×TSU×SU 52 TSU×SU 81G×T N×SU 3 0 93 T×GSU×SU 6 1 T 62 10SU×N 34 1) One o he 208 o sp ing s udied had p oduced wo ichomes. 2) One o he 42 o sp ing had wo ichomes 3) In one c oss, one ou o 60 o sp ing was glab ous. sampling seeds) we e used o in e he geno ypes o he pa en al plan s used in he c oss. Pheno ypic and gene ic 6a ia ion in ichome numbe To documen he equency dis ibu ion o ichome numbe pe lea , we g ew in he g eenhouse wo plan s om each o 50 ma e nal amilies collec ed in he polymo phic S o s ensudden popula ion. Th ee mon hs a e plan ing, we coun ed he numbe o ichomes on he adaxial (uppe ) su ace o one an- domly selec ed lea om each plan unde a dissec - ing mic oscope. The leng h and wid h o he sampled lea we e measu ed o he nea es mm, and lea a ea was es ima ed as he p oduc o hese wo measu es. The he i abili y o ichome numbe in he S o s ensudden popula ion was es ima ed based on pa en -o sp ing eg essions. T ichome numbe pe lea was de e mined as abo e o bo h pa en s and o sp ing in he 28 con olled c osses ha esul ed in a leas some ichome-p oducing plan s (Table 1 and 2). The mean numbe o ichomes on he uppe lea su ace in he o sp ing was eg essed on (a) he ma e nal pa en , (b) he pa e nal pa en , and (c) he mid-pa en alue (F ALCONER and M ACKAY 1996). To achie e no mali y, ichome numbe pe lea was log- ans o med (log( ichome numbe +1)) p io o analyses. RESULTS Gene ic basis o glab ousness : con olled c osses All c osses be ween glab ous indi iduals (wi hin and be ween popula ions) p oduced glab ous o sp ing only (Table 1). Fou c osses be ween ichome-p o- ducing plan s p oduced bo h glab ous and ichome- p oducing o sp ing, while eigh such c osses p oduced ichome-p oducing o sp ing only (Table 1). C osses be ween he wo mo phs p oduced ei he only ichome-p oducing indi iduals, o p ogenies wi h bo h glab ous and ichome-p oducing plan s (Table 1). In seg ega ing c osses be ween ichome-p oducing plan s, he obse ed a io o glab ous s. ichome- p oducing o sp ing did no s a is ically de ia e om he expec ed 1:3 a io in C oss 1 (N=62), o in he da a pooled om ou c osses (N=86; Table 2). The mo he s o bo h plan s used in C oss 1 we e glab ous, hus he la e we e in e ed o be he e ozygous a he locus causing glab ousness (Table 2). Bo h wi hin- (SU×SU) and be ween-popula ion c osses (N×SU o SU×N) be ween glab ous and ichome-p oducing indi iduals esul ed in seg ega - ing p ogenies. The a io o glab ous o ichome-p o- ducing plan s did no de ia e om he expec ed 1:1 a io in he wo T×G c osses wi h la ge numbe o o sp ing (one wi hin-popula ion and one be ween- popula ion c oss), o in he pooled analysis o six seg ega ing T×G c osses (40 glab ous s 48 i- chome-p oducing indi iduals, G=0.73, NS). How- e e , wo o ou seg ega ing G×T c osses p oduced G:T a ios ha de ia ed signi ican ly om he ex- pec ed 1:1 a io (bo h we e be ween-popula ion c osses; Table 2). The numbe o glab ous indi iduals was lowe han expec ed among he o sp ing o G×T C oss 1, and highe han expec ed among he o sp ing o G×T C oss 2. In he o he wo seg e- ga ing G×T c osses analysed sepa a ely (one wi hin- and one be ween-popula ion c oss), and in he pooled analysis o G×T c osses, he a io o glab ous o ichome-p oducing plan s did no de ia e om he K.Ka¨ kka¨inen and J.A , g en222 He edi as 136 (2002) Table 2. Compa ison o obse 6ed and expec ed a ios o glab ous ( G ) o ichome-p oducing o sp ing ( T ) in seg ega ing c osses wi h N 20 ,and in da a pooled o seg ega ing c osses o each ype.The plan s used in he c osses o igina ed om wo popula ions o A abidopsis ly a a ssp.pe aea in eas e n Sweden ( No a¨lls6iken ( N ) and S o s ensudden ( SU )) .The i o obse 6ed a ios o hose expec ed unde a single-locus wo-allele model wi h he allele coding o ichome p oduc ion being dominan o6e he allele coding o glab ousness was examined wi h G- es s ( d = 1) .Fo c osses ep esen ed by 20 o mo e o sp ing, he pheno ypes o he ma e nal and pa e nal g andmo he s,and he in e ed geno ypes o pa en al plan s a e gi6en O igin G- es O sp ing Expec ed Pheno ypes o Geno ypes o C oss ype pheno ypes a io o à×ßà×ßpa en al plan s 1) g andmo he s G:T in he GT ào sp ing ßà×ß T×T: 22 40 1:3 3.34 NSSU×SU G G T× TC oss 1 29 57 1:3 3.25 NSPooled da a, 4 c osses G×T: 8 22 1:1 6.68 **N×SU GC oss 1 G × T N×SUC oss 2 31 10 1:1 11.15 *** G T ×T o × T 715C oss 3 1:1 2.91 NS G T ×T o × TN×SU 21 25 1:1 0.34 NSSU×SU GC oss 4 T ×T o × T 86 88 1:1 0.02 NSPooled da a, 8 c osses T×G: 15 19 1:1 0.46 NSSU×NGC oss 1 G T× SU×SUC oss 2 9 11 1:1 0.20 NS G G T× 48 1:1 0.73 NSPooled da a, 6 c osses 40 1 The allele ecei ed om he ma e nal plan is p esen ed be o e he allele ecei ed om he pa e nal plan . expec ed 1:1 a io (eigh c osses, 86 glab ous s 88 ichome-p oducing indi iduals, G=0.02, NS; Table 2). Unde he pos ula ed model o inhe i ance o glab ousness, he ichome-p oducing pa en in seg e- ga ing c osses should be he e ozygous a he locus causing glab ousness. The obse a ion ha he i- chome-p oducing pa en had a glab ous mo he in h ee o he seg ega ing c osses is consis en wi h his iew (Table 2: G×T C oss 1; T×G C oss 1 and T×G C oss 2). Pheno ypic and gene ic 6a ia ion in ichome numbe The plan s es ablished om seeds collec ed in he S o s ensudden popula ion and g own in he g een- house exhibi ed conside able a ia ion in ichome numbe pe lea (mean=18.9, SD=32.5, N=96; Fig. 1). The dis ibu ion o ichome numbe de ia ed signi ican ly om he no mal dis ibu ion (Kol- mogo o -Smi no es o no mali y 0.302, d =96; pB0.001), wi h high alues o skewness (2.10, SE= 0.25) and ku osis (3.42, SE=0.49). This was ue also i glab ous indi iduals we e excluded om he analysis (Kolmogo o -Smi no 0.296, d =60, pB 0.001; skewness 1.46, SE=0.31, ku osis 0.87, SE= 0.61). Va ia ion in ichome densi y ( ichome numbe pe uni lea a ea) was a unc ion o a ia ion in o al numbe o ichomes p oduced, a he han o a ia ion in lea size: ichome numbe pe uni lea a ea was s ongly co ela ed wi h ichome num- be pe lea ( =0.946, N=406, pB0.001), bu was no co ela ed wi h lea a ea ( =−0.07, N=406, NS). The es ima es o he he i abili y o ichome num- be pe lea ob ained om eg essions o o sp ing on ma e nal, pa e nal and midpa en alues we e all high (h 2 9SE, 1.1790.25, 1.0390.53, and 0.659 0.15, espec i ely, N=28, Fig. 2). In c osses be ween glab ous and ichome-p oduc- ing plan s mos o sp ing p oduced ewe lea i- chomes han he ichome-p oducing pa en (Fig. 3). Fig. 1. Pheno ypic a ia ion in ichome numbe on he uppe side o he lea in A abidopsis ly a a ssp. pe aea. Plan s we e aised in he g eenhouse om seeds collec ed in he S o s ensudden popula ion, and we e sco ed o i- chome numbe h ee mon hs a e plan ing (N=100). Gene ic basis o ichome p oduc ion in A abidopsis 223He edi as 136 (2002) Fig. 2. Mean numbe o lea ichomes in he o sp ing as a unc ion o mean numbe o lea ichomes o he pa en al plan s (mid-pa en alue) in 28 c osses conduc ed wi h plan s o igina ing om he S o s ensudden popula- ion. Bo h he pa en al plan s and he o sp ing we e aised in he g eenhouse and sco ed o ichome numbe on he uppe lea su ace h ee mon hs a e plan ing. T ichome numbe was log- ans o med p io o analysis. The he i a- bili y es ima e (9SE) is indica ed in he igu e. Glab ousness may ha e a simple gene ic basis also in o he plan species. Da a om seg ega ing c osses indica e ha glab ousness is con olled by one, au o- somal ecessi e gene in Silene dioica (W ESTERBERGH 1992), and Silene la i olia (D E V RIES 1900). Mo e- o e , ichome mo phology (glandula s. non-glan- dula ichomes) can be inhe i ed in a simple Mendelian ashion as was ecen ly documen ed o Da u a w igh ii ( VAN D AM e al. 1999). In o ma ion on he gene ic basis o ichome de el- opmen sugges s ha glab ousness may ha e a simple gene ic basis also in in A abidopsis haliana. Al hough he e a e many genes in ol ed in ichome o ma ion, only a ew low in he de elopmen al pa hway (GLABROUS-1 and TTG) ha e been ound o block ichome p oduc ion comple ely (M ARKS 1997). GLABROUS-1 encodes a p o ein wi h sequence homology o a myb- ela ed ansc ip ion ac o (O P- PENHEIMER e al. 1991). Plan s homozygous o he gl1-1 allele a e i ually de oid o ichomes, al- hough ichomes a e occasionally ound a he ma - gin o ose e and cauline lea es. Plan s homozygous o he weake allele gl1-2 ha e educed ichome numbe wi h ichomes nea he ma gin o lea es (E SCH e al. 1994). Mu a ion in he gene TTG1, In con as , in c osses be ween ichome-p oducing indi iduals, a la ge p opo ion o he o sp ing p o- duced mo e lea ichomes han ei he pa en (Fig. 4). DISCUSSION Gene ic basis o glab ousness The esul s o he p esen s udy indica e ha glab ousness has a simple Mendelian inhe i ance in A abidopsis ly a a ssp. pe aea, and ha glab ousness is de e mined by allelic a ia ion a he same locus in he wo popula ions s udied. Mo eo e , hey indica e ha he e is conside able gene ic a ia ion in he numbe o ichomes o med by ichome-p oducing plan s. The ac ha c osses be ween glab ous plan s did no p oduce any ichome-p oducing o sp ing, while c osses be ween ichome-p oducing plan s e- sul ed in ei he all ichome-p oducing plan s o seg- ega ing p ogenies sugges s ha glab ousness is a ecessi e cha ac e o ichome-p oduc ion. Mo e- o e , he a io o glab ous: ichome-p oducing plan s in he seg ega ing p ogenies did no s a is ically de i- a e om he expec ed 1:3 a io. As expec ed unde a one-locus, wo-allele model, some be ween-mo ph c osses p oduced only ichome-p oducing o sp ing. The a io o glab ous o ichome-p oducing plan s did no s a is ically de ia e om he expec ed 1:1 a io in ou o he six c osses be ween glab ous and ichome-p oducing plan s, o in he pooled analysis o such c osses. Fig. 3. Va ia ion in ichome numbe on he uppe lea su ace among o sp ing om con olled c osses be ween glab ous and ichome-p oducing plan s (G×T, C osses 1–4, and T×G C osses 1–2 in Table 2). T ichome num- be s o ma e nal (solid line a ow) and pa e nal pa en s (b oken line a ow) a e indica ed. K.Ka¨ kka¨inen and J.A , g en224 He edi as 136 (2002) Fig. 4. Va ia ion in ichome numbe on he uppe lea su ace among o sp ing om con olled c osses be ween ichome-p oducing plan s (T×T). In C oss 1, some o he o sp ing we e glab ous, in he o he wo c osses all p ogeny we e ichome-p oducing. T ichome numbe s o ma e nal (solid line a ow) and pa e nal pa en s (b oken line a ow) a e indica ed. Russian popula ion (H AUSER e al. 2001) and ou own da a on Swedish and No wegian popula ions (Ka¨ RKK a¨ INEN ,K IVIM a¨ KI and A , GREN , unpubl.) sug- ges ha mu a ions ei he in he coding a ea o in he egula o y egions o GLABROUS-1 may cause glab ousness in A.ly a a. Gene ic basis o ichome numbe Pa en -o sp ing eg essions indica ed ha he he i a- bili y o ichome numbe pe lea was high in he s udied popula ion. This sugges s ha his popula ion may espond no only o selec ion on ichome o - ma ion pe se, bu also o selec ion on ichome densi y. The he i abili y es ima e based on mid-pa - en alues had a lowe s anda d e o (h 2 =0.65, SE=0.15) han he es ima es based on ma e nal (h 2 =1.17, SE=0.25) and pa e nal alues (h 2 =1.03, SE=0.53). The seg ega ion o he ecessi e allele o glab ousness may ha e con ibu ed o he high s an- da d e o s o he wo la e es ima es. The highes es ima e was ob ained when o sp ing ichome num- be was eg essed on ma e nal alue, as would be expec ed i ma e nal en i onmen al e ec s con ibu e o a ia ion in ichome numbe (R OACH and W ULFF 1987). The skewed dis ibu ion o ichome numbe ob- se ed in he S o s ensudden popula ion can pa ly be a ibu ed o he occu ence o glab ous plan s. How- e e , he dis ibu ion de ia ed om no mali y also when he comple ely glab ous indi iduals we e ex- cluded om he analysis. This is consis en wi h he hypo hesis ha he e is one o se e al genes wi h majo e ec on ichome numbe in ichome-p oduc- ing plan s. In A abidopsis haliana, a QTL-analysis o ichome numbe o wo eco ypes e ealed a majo QTL (Reduced ichome numbe , RTN) ha ex- plained 70 % o he a ia ion in ichome numbe (L ARKIN e al. 1996). In a QTL-s udy on ichome densi y o co on, one QTL wi h majo pheno ypic e ec and a ew wi h smalle e ec s we e ound (W RIGHT e al. 1999). Thus, a ew genes wi h majo e ec may in luence a ia ion in ichome numbe in se e al species. C osses be ween glab ous and ichome-p oducing plan s yielded ichome-p oducing plan s wi h a ma kedly lowe ichome numbe han hose p o- duced in c osses be ween ichome-p oducing plan s. This sugges s ha he mu a ion causing glab ousness may in luence he densi y o lea ichomes in he - e ozygous ichome-p oducing plan s. To u he cla i y he gene ic basis o ichome numbe in A a- bidopsis ly a a, we a e cu en ly employing bo h com- plex seg ega ion analysis (L YNCH and W ALSH 1998), and QTL- and candida e gene app oaches. anspa en es a glab a1, can also esul in a com- ple e loss o ichome o ma ion on mos a eal su - aces (K OORNNEEF 1981). The TTG1 gene encodes a WD40 epea p o ein ha may egula e signal ans- duc ion pa hways go e ning lea hai o ma ion and an hocyanin biosyn esis (W ALKER e al. 1999). By now, se e al mu a ions ha esul in he loss o show educ ion in ichome o ma ion ha e been desc ibed o TTG1 (W ALKER e al. 1999). The glab ous pheno ype o A abidopsis ly a a may be caused by one o he genes causing glab ousness in A abidopsis haliana. Based on sequence a ia ion in DNA, A.ly a a and A.halle i a e he closes known ou c ossing ela i es o A. haliana (O’K ANE and A L -S HEHBAZ 1997; K OCH e al. 1999). Thus, genes causing glab ousness in A. haliana a e good candi- da es o causing glab ousness also in A.ly a a.A ecen s udy which included a glab ous plan om a Gene ic basis o ichome p oduc ion in A abidopsis 225He edi as 136 (2002) ACKNOWLEDGEMENTS This s udy was suppo ed by g an s om he Academy o Finland and No FA o KK and om he Swedish Resea ch Council and Fo mas o JA , . REFERENCES Ag awal AA, (1999). Induced esponses o he bi o y in wild adish: e ec s on se e al he bi o es and plan i ness. Ecology 80: 1713–1723. A , g en J and Schemske DW, (1993). The cos o de ense agains he bi o es: an expe imen al s udy o ichome p oduc ion in B assica apa. Am. Na . 141: 338–350. A , g en J and Schemske DW, (1994). E olu ion o ichome numbe in a na u alized popula ion o B assica apa. Am. Na . 143: 1–13. Be gelson J, (1994). The e ec s o geno ype and he en i- onmen on cos s o esis ance in le uce. Am. Na . 143: 349–359. Ehle inge J, (1984). Ecology and ecophysiology o lea pubescence in No h Ame ican dese plan s. In: Biology and chemis y o plan ichomes (eds XE Rod iguez, PL Healey and I Meh a). Plenum P ess, New Yo k, p. 113–132. Esch JJ, Oppenheime DG and Ma ks MD, (1994). Cha - ac e iza ion o a weak allele o he GL1 gene o A a- bidopsis haliana. Plan Mol. Biol. 24: 203–207. Espiga es T and Peco B, (1995). Medi e anean annu e pas u e dynamics: impac o au umn d ough . J. Ecol. 83: 135–142. Falcone DS and Mackay TFC, (1996). In oduc ion o quan i a i e gene ics. Longman, Ha low, UK. F a y A, Nesbi TC, F a y A, G andillo S, an de Knaap E, Cong B, Liu J, Melle J, Elbe R, Alpe KB and Tanksley SD, (2000). w2.2: a quan i a i e ai locus key o he e olu ion o oma o ui size. Science 289: 85–88. Hagley EAC, B onskill JF and Fo d EJ, (1980). E ec o he physical na u e o lea and ui su aces by he codling mo h, Cydia pomonella (Lepidop e a: To i- cidae). Can. En omol. 112: 503–510. Hause MT, Ha B and Schlo¨ e e C, (2001). T ichome dis ibu ion in A abidopsis haliana and he close ela- i e A abidopsis ly a a: molecula analysis o he candi- da e gene GLABROUS1. Mol. Biol. E ol. 18: 1754–1763. Hul e´n E, (1971). A las o dis ibu ion o ascula plan s in NW Eu ope. Gene als ablens Li og . Ans al . S ockholm. Jalas J and Suominen J, (1994). A las Flo a Eu opaea. Dis ibu ion o ascula plan s in Eu ope. Vol. 10. C u- ci e ae (Sisimb yum o Aub ie a). Helsinki Uni . P in - ing House, Helsinki, Finland. Johanson U, Wes J, Lis e C, Michaels S, Amasino R and Dean C, (2000). Molecula analysis o FRIGIDA, a majo de e minan o na u al a ia ion in A abidopsis lowe ing ime. Science 290: 344–347. Jonsell B, Kus a˚s K and No dal I, (1995). Gene ic a ia ion in A abis pe aea, a disjunc species in no he n Eu ope. Ecog aphy 18: 321–332. Koch M, Bishop J and Mi chell-Olds T, (1999). Molecula sys ema ics and e olu ion o A abidopsis and A abis. Plan Biol. 1: 529–537. Koo nnee M, (1981). The complex o synd ome o g mu an s. A ab. In . Se . 18: 45–51. La kin JC, Young N, P igge M and Ma ks MD, (1996). The con ol o ichome spacing and numbe in A a- bidopsis. De elopmen 122: 997–1005. Le in DA, (1973). The ole o ichomes in plan de ence. Q. Re . Biol. 48: 3–15. Lynch M and Walsh B, (1998). Gene ic analysis o quan i- a i e ai s. Sinaue Associa es, Sunde land. Ma ks MD, (1997). Molecula gene ic analysis o ichome de elopmen in A abidopsis. Annu. Re . Plan . Physiol. Plan Mol. Biol. 48: 137–163. Mau icio R, (1998). Cos s o esis ance o na u al enemies in ield popula ions o he annual plan A abidopsis haliana. Am. Na . 151: 20–28. Mau icio R and Raushe MD, (1997). Expe imen al ma- nipula ion o pu a i e selec i e agen s p o ides e idence o he ole o na u al enemies in he e olu ion o plan de ense. E olu ion 51: 1435–1444. O’Kane SL and Al-Shehbaz IA, (1997). A synopsis o A abidopsis (B assicaceae). No on 7: 323–327. Oppenheime DG, He man PL, Si akuma an S, Esch J and Ma ks MD, (1991). A myb gene equi ed o lea di e - en ia ion in A abidopsis is exp essed in s ipules. Cell 67: 483–493. P ice RA, Palme JD and Al-Shehbaz IA, (1994). Sys em- a ic ela ionships o A abidopsis: a molecula and mo - phological pe spec i e. In: A abidopsis (eds EM Meye owi z and C Some ille). Cold Ha bo Labo a- o y P ess, New Yo k, p. 7–19. Roach DA and Wul RD, (1987). Ma e nal e ec s in plan s. Annu. Re . Ecol. Sys . 18: 209–235. Roy BA, S an on ML and Eppley SM, (1999). E ec s o en i onmen al s ess on lea hai densi y and conse- quences o selec ion. J. E ol. Biol. 12: 1089–1103. Sa olainen O, Langley CH, Lazza o BP and F e ille H, (2000). Con as ing pa e ns o nucleo ide polymo - phism a he alcohol dehyd ogenase locus in he ou - c ossing A abidopsis ly a a and he sel ing A. haliana. Mol. Biol. E ol. 17: 645–655. Schemske DW and B adshaw HD, (1999). Pollina o p e - e ences and he e olu ion o lo al ai s in monkey low- e s (Mimulus). P oc. Na l. Acad. Sci. USA 96: 11910–11914. Sc i e CR and Wa e s PJ, (1999). Monogenic ai s a e no simple lessons om phenylke onu ia. T ends Gene . 15: 267–272. Skal sa H, Ve ykokidou E, Ha ala C, K abou nio is G and Mane as Y, (1994). UV-p o ec i e po en ial and la anoid con en o lea hai s o Que cus ilex. Phy o- chemis y 37: 987–990. Val e e PL, Fo noni J and Nunez-Fa an J, (2001). De en- si e ole o lea ichomes in esis ance o he bi o ous insec s in Da u a s amonium. J. E ol. Biol. 14: 424– 432. an Dam NM, Ha e JD and Elle E, (1999). Inhe i ance and dis ibu ion o ichome pheno ypes in Da u a w igh ii. J. He ed. 90: 220–227. V ies H de, (1900). Das Spal ungsgese z de Bas a de. Be . D sch. Bo . Ges. 18: 83–90. Walke AR, Da idson PA, Bolognesi-Win ield AC, James CM, S ini asan N, Blundell TL, Esch JJ, Ma ks MD and G ay JC, (1999). The T anspa en Tes a Glab a1 locus, which egula es ichome di e en ia ion and an- hocyanin biosyn hesis in A abidopsis, encodes a WD40 epea p o ein. Plan Cell 11: 1337–1349. K.Ka¨ kka¨inen and J.A , g en226 He edi as 136 (2002) Webe K, Eisman R, Mo ey L, Pa y A, Spa ks J, Tausek M and Zeng ZB, (1999). An analysis o polygenes a ec ing wing shape on ch omosome 3 in D osophila melanogas e . Gene ics 153: 773–786. Wes e be gh A, (1992). The gene ic basis o hai lessness in Silene dioica (Ca yophyllaceae). He edi as 117: 287– 291. W igh RJ, Thax on PM, El-Zik KM and Pa e son AH, (1999). Molecula mapping o genes a ec ing pub- escence o co on. J. He ed. 90: 215–219.