Novel pattern of DNA methylation in Neurospora crassa transgenic for the foreign gene hph

 1997 Ox o d Uni e si y P ess 2409–2416 Nucleic Acids Resea ch, 1997, Vol. 25, No. 12 No el pa e n o DNA me hyla ion in Neu ospo a c assa ansgenic o he o eign gene hph An onio C. Codón, Yeon-Su Lee and Vincenzo E. A. Russo* MPI Molekula e Gene ik, Ihnes asse 73, D-14195 Be lin, Ge many Recei ed Feb ua y 25, 1997; Re ised and Accep ed Ap il 21, 1997 ABSTRACT I has p e iously been epo ed ha mul iple copies o he hph gene in eg a ed in o he genome o Neu ospo- a c assa a e me hyla ed a Hpa II si es (CCGG) du ing he ege a i e li e cycle o he ungus, while hph genes in eg a ed as single copies a e no me hyla ed. Fu - he mo e, me hyla ion is co ela ed wi h silencing o he gene. We epo he e he me hyla ion s a e o cy osine esidues o he majo pa o he p omo e egion o he hph gene in eg a ed in o he genome o he mul iple copy s ain HTA5.7 du ing he ege a i e s age o he li e cycle. Cy osine me hyla ion is sequence dependen , bu he sequence speci ici y is complex and is di e en om he sequence speci ici y known o mammals and plan s (CpG and CpNpG). The pa e n o DNA me hyla ion epo ed he e is e y di e en om ha measu ed a e meiosis in Neu o- spo a o in Ascobulus . A e he sexual cycle in hose wo ungi all he cy osines o mul iple s e ches o DNA a e hea ily me hyla ed. This indica es ha he s ill unknown me hyl ans e ase in Neu ospo a has a di e en speci ici y in he sexual and he ege a i e s ages o he li e cycle o ha he e a e di e en me hyl ans e ases. The pa e n o me hyla ion e- po ed he e is also di e en om he pa e n o cy osine me hyla ion o ansgenes o Pe unia , he only pa e n published un il now in plan s ha has DNA me hyla ion a cy osines which a e no in he canonical sequences CpG and CpNpG. INTRODUCTION In p oka yo es he majo me hyla ed bases a e 4-me hylcy osine (m4C), 5-me hylcy osine (m5C), 5-hyd oxyme hylcy osine (hm5C) and 6-me hyladenine (m6A) (1). Hund eds o me hyl ans e ases ha e been iden i ied which me hyla e ei he cy osine o adenine a speci ic ecogni ion si es. The ecogni ion si e o me hyla ion can be as simple as he double base CpG o a much mo e complex sequence (up o 15 bases). On a e age i is 4–6 bases long (1). O e 320 es ic ion endonucleases sensi i e o si e-speci ic me hyla ion a e known oday (1). Many biological unc ions ha e been a ibu ed o m6A gene a ed by he Dam me hyl ans e ase in Esche ichia coli: induc ion o pos - eplica i e misma ch epai , con ol o E.coli ch omosome eplica ion and seg ega ion, con ol o plasmid seg ega ion, egula ion o ansposi ion, gene exp ess- ion and con ol o ini a ion o phage P1 DNA packaging (2). In con as , in euka yo es he base modi ica ion mos ly s udied is m5C. The modi ied cy osine is only in he symme ical sequence CpG o CpNpG in bo h mammals and plan s (3). The equency o m5C is e y high, om 4% in humans (4) o 36% in some highe plan s (5). Recen ly i was ound ha in ansgenic ungi a e he meio ic phase cy osine me hyla ion occu s a e y high equency (>50%) and a any cy osine wi h equal p obabili y in Neu ospo a (6) and in Ascobolus (7). In Neu ospo a, DNA me hyla ion was s udied in co ela ion wi h RIP ( epea -induced poin ) mu a ions, which ac s on sequence duplica ions and occu s only in special p e-meio ic cells con aining haploid nuclei om bo h pa en s (6). In he ascomyce e Ascobolus imme sus, duplica ed DNA segmen s a e subjec o he me hyla ion-induced p e-meio ic (MIP) p ocess (7). The only pa e n o m5C in plan s ha is di e en om he simple one quo ed abo e was ound in ansgenic Pe unia (8). An indica ion ha some non-symme ical sequences could be me hyla ed in plan s was published in he same yea by wo di e en g oups (9,10). Tashe a and Rou a (11,12) epo ed densely me hyla ed DNA islands in mammalian ch omosomal eplica ion o igins ega dless o hei dinucleo ide composi ion. Howe e , hose esul s could no be ep oduced by Rein e al. (13,14). In euka yo es, m5C has a ole in gene silencing, gene ic imp in ing, X-ch omosome inac i a ion in mammals and pa a- mu a ion in plan s (15). In Neu ospo a, he only o eign gene ha is known o be silenced is hph, a gene ha con e s esis ance o he an ibio ic hyg omycin (16). T ansgenic Neu ospo a wi h a single copy o he hph gene had conidia ha we e 100% esis an o he an ibio ic, while in many ansgenics wi h mul iple copies he hph gene was silenced, so ha only a small pe cen age o he conidia we e esis an o hyg omycin. The hph gene in he ansgenic wi h a single copy was no me hyla ed, while he genes in he mul iple copies ansgenic we e hea ily me hyla ed (16). The deg ee o me hyla ion o he hph gene was measu ed as esis ance o es ic ion by he enzymes HpaII and MspI ( es ic ion si e CCGG). Fu he e idence ha me hyla ion could be he cause o silencing o he hph gene came om expe imen s wi h he deme hyla ing agen 5-azacy idine, which con e ed silenced sensi i e s ains in o esis an s ains (16,17). All he s udies on silencing and me hyla ion o he hph gene we e done * To whom co espondence should be add essed. Tel: +49 30 8413 1264; Fax: +49 30 8413 1385; Email: [email p o ec ed] Nucleic Acids Resea ch, 1997, Vol. 25, No. 12 2410 on mycelia in he ege a i e s age o he li e cycle, which ne e wen h ough meiosis a e ans o ma ion. He e we p esen e idence ha he hph genes o a ansgenic s ain o Neu ospo a wi h mul iple copies ha e a pa e n o cy osine me hyla ion ha is di e en om ha o animals (3), highe plan s (5) and ansgenic Pe unia (8). MATERIALS AND METHODS Neu ospo a c assa s ains and cul u e condi ions The s ains used (wild- ype, HTA5.7 and HTA23) ha e been desc ibed p e iously (16). The cul u e condi ions we e s anda d ones (16). S ain HTA23 has only one copy o he hph gene inse ed in o he genome, while s ain HTA5.7 has a leas ou copies (18). Plasmid pCSN44 Plasmid pCSN44 (19), con aining he hph gene, was a gene ous gi o C.Yano sky (S an o d Uni e si y). Fo he Sou he n blo analysis he plasmid was g own in E.coli s ain JM110 (Biolabs), which is dam– and dcm–. Sou he n blo s The Sou he n blo s we e hyb idized using as p obe a s e ch o DNA ha was labelled wi h digoxigenin acco ding o he manu ac u e (Boeh inge Mannheim) (16). Bisulphi e eac ion The bisulphi e eac ion was pe o med acco ding o Olek e al. (20), wi h he ollowing modi ica ions. The ch omosomal DNA o Neu ospo a was diges ed wi h he es ic ion enzyme EcoRV, boiled o 10 min and quickly chilled on ice. A o al o 5 µg DNA we e mixed wi h 2 ol 2% LMP aga ose (FMC Bio-p oduc s, Rockland, ME) dissol ed in wa e o gi e a inal olume o 70 µl. This mix u e was di ec ly pipe ed in o chilled mine al oil o o m se en beads o 10 µl. In he eac ion ubes he se en beads we e co e ed wi h 1 ml bisulphi e solu ion and he p o ocol o Olek e al. (20) was s ic ly ollowed. Choice o oligonucleo ide p ime s We no iced ha he oligonucleo ides o ampli ica ion o he DNA a e bisulphi e eac ion mus be chosen a e a e y s ingen and ime consuming es in o de o a oid ampli ica ion o non- ep esen a i e DNA s ands. When we ampli ied a s e ch o he p omo e egion o he hph gene o he mul iple copies inse HTA5.7 s ain a e bisulphi e ea men we ound ha all he C esidues mus ha e been me hyla ed. This esul was ound also in he case whe e he bisulphi e- ea ed DNA came om s ain HTA23, which con ains only a single in eg a ed copy. This was a a iance wi h he esul s o he Sou he n blo s, which indica ed no me hyla ion o he hph gene in his s ain (da a no shown). I seems ha ou oligonucleo ide p ime s we e ampli y- ing a e DNA s ands ha did no ep esen he bulk o he bisulphi e- ea ed DNA. We de ised he ollowing es o check i a pai o p ime s ampli ied DNA s ands in a biased way (bias es ). The pCSN44 plasmid, con aining he hph gene, was ea ed wi h bisulphi e. A pai o degene a ed p ime s we e used o PCR ampli y a s e ch o DNA con aining he p omo e o he hph gene, using oligonucleo ides O1 and O2 and as empla e bo h he plasmid be o e and a e bisulphi e ea men (Fig. 1). A e - wa ds, we cloned he PCR p oduc s in o he TA ec o (In i ogen) and sequenced hem. We ob ained in his way wo di e en ypes o TA plasmids. One had he sub egion o he hph gene wi h he o iginal sequence (pTAC), he o he had he same sequence bu wi h all he C esidues changed o T (pTAT). These wo plasmids we e used as empla e, mixed in equal amoun , in a PCR eac ion wi h he pai o p ime s o be es ed, he a ionale being ha i he e was no selec ion a e PCR, cloning and sequencing, he numbe o clones wi h he pTAC sequence ob ained in he second ound o PCR should be equal o he numbe o clones wi h he pTAT sequence. Using in his bias es he pai o p ime s ha ga e con lic ing esul s be ween he deg ee o me hyla ion measu ed wi h he bisulphi e eac ion and ha measu ed wi h he Sou he n blo , we ob ained 30 clones wi h he pTAC sequence and ze o wi h he pTAT sequence. This is a clea indica ion ha his pai o p ime s made a s ong selec ion among he wo di e en plasmids. We hen es ed ou mo e pai s o p ime s chosen a andom and ound in o al wo pai s o oligonucleo ides ha , in he bias es explained abo e, did no gi e any bias, while he o he h ee did. One o hese pai s o p ime s was used he e and hei sequences a e shown in Figu e 2. PCR expe imen s The PCR cock ail was assembled acco ding o he ecomenda- ions o he manu ac u e o he ki (Ampli aq DNA polyme ase, S o el agmen ; Pe kin Elme Ce us). Dena u a ion was a 96C o 30 s and annealing a 50C o 30 s. The ex ension s ep was a 72C o 1 min. A he end o 40 cycles a delay o 4 min a 72C was added. Cloning and sequencing The PCR p oduc s we e liga ed and cloned ollowing he ins uc ions o he TA cloning ki (In i ogen). Sequence analysis o he cloned DNA agmen s was pe o med by cycle sequencing using 25 ng DNA, 2 µl d/ddNTP eac ion mix u e (P omega Sil e sequencing ki ), 1.25 µl special bu e (50:1 mix u e o 5× P omega Sil e sequencing bu e and T i on X-100), 2.5 pmol luo escen ly ma ked p ime , 2 µl s e ile wa e , 0.25 µl Taq polyme ase (P omega Sil e sequencing ki ). The DNA was dena u ed a 93C o 2 min. Ampli ica ion was o 36 cycles o 20 s dena u a ion a 94C, 20 s annealing a 55C and 40 s elonga ion a 72C. The inal elonga ion s ep p oceeded o 10 min and he eac ion mix u e was hen cooled and s o ed a 4C a e 3 µl sequencing s op bu e (Pha macia) had been added. Sequences we e ead on an ALF au oma ic sequence (Pha macia). The ollowing luo escen ly ma ked p ime s we e used o he sequencing: uni e sal p ime and e e se p ime (Pha macia). RESULTS The wo me hods mos widely used oday o de e mine he s a e o me hyla ion o cy osine esidues in euka yo ic genomes a e: (i) Sou he n blo s analysis o DNA es ic ed wi h me hyla ion- sensi i e enzymes (1); (ii) he bisulphi e me hod (20). The i s me hod akes ad an age o he ac ha many enzymes do no diges DNA i one cy osine o he ecogni ion/ es ic ion si e is me hyla ed. The me hod is simple and eliable, bu i has he disad an age ha only ew cy osines can be p obed, namely 2411 Nucleic Acids Resea ch, 1994, Vol. 22, No. 1 Nucleic Acids Resea ch, 1997, Vol. 25, No. 12 2411 Figu e 1. Expe imen al scheme o he bias es . In he op plasmids a e shown he p omo e and he coding egion o he hph gene wi h he posi ion o he O1 and O2 oligonucleo ides used o ampli y a s e ch o DNA con aining he p omo e egion, as well as he posi ion o he p ime s es ed in he bias es . In he i s PCR a s e ch o DNA con aining he p omo e egion was ampli ied wi h he O1 and O2 oligonucleo ides om a pCSN44 plasmid and cloned in o a TA ec o . The plasmid ob ained (pTAC) con ains he o iginal sequence (whi e ba ). Wi h he same p o ocol he same egion was ampli ied and cloned om p e iously bisulphi e- ea ed pCSN44. The plasmid ob ained (pTAT) con ains he same inse as ha in pTAC wi h he di e ence ha all he C esidues ha e been changed o T (ha ched ba ). All plasmids we e checked by sequencing. The wo plasmids we e used o he ac ual bias es . They we e used as empla e, in equal amoun , in a second PCR using he pai o oligonucleo ides unde es . The ampli ied DNA was cloned in o he TA ec o and se e al clones we e sequenced. We assumed ha a pai o oligonucleo ides display no p e e ence, nei he in PCR no du ing cloning o he esul ing PCR p oduc , only when he numbe o clones wi h he whi e sequence was equal o he numbe o clones wi h he ha ched sequence. In he case o he oligonucleo ides used in he expe imen s epo ed in Figu e 2 his numbe was se en o eigh . only hose which a e in he ecogni ion/ es ic ion si e o he me hyla ion-sensi i e enzymes. The bisulphi e me hod akes ad an age o he ac ha bisulphi e, unde app op ia e expe imen al condi ions, will change in i o any C o a single s and in o U. m5C is esis an o he ac ion o bisulphi e unde he same expe imen al condi- ions. Bisulphi e does no wo k on cy osine esidues o double- s anded DNA. The bisulphi e- ea ed DNA is subsequen ly ampli ied by PCR, cloned and sequenced. Du ing PCR ampli ica- ion he U is subs i u ed by T, wi h he consequence ha in he DNA sequence o he ampli ied s e ch o DNA he e will be a T o any C o he o iginal sequence and a C o any m5C. The ad an age o his echnique is ha i can de e mine he me hyla ion s a e o each single cy osine and he me hyla ion pa e n on each single ilamen o DNA. I in a issue he e a e se e al iden ical genes wi h di e en pa e ns o me hyla ion, ha can be picked up. The d awback is ha i is di icul o use because he e a e many po en ial a i ac s ha can gi e alse esul s. Se e al pape s ha e been published a e he o iginal one in o de o imp o e his powe ul echnique (22–24). We used he e ano he p o ocol o bisulphi e ea men , de eloped in ou ins i u e, ha minimizes eannealing o he DNA du ing he bisulphi e ea men and pe mi s ampli ica ion o long s e ches o DNA (20). We ha e also ound ha a e y impo an poin is he choice o p ime s o he PCR, as discussed a leng h in Ma e ials and Me hods. Fo unknown easons many pai s o p ime s can selec a e DNA s ands wi h a sequence ha is no ep esen a i e o he sequence o he bulk o he DNA s ands p esen a e bisulphi e ea men . DNA me hyla ion o he ansgene hph acco ding o he bisulphi e me hod In Neu ospo a, mul iple copies o he same s e ch o DNA a e meiosis con ain RIP mu a ions and a e hea ily me hyla ed, wi h me hyla ion a any cy osine ha ing equal p obabili y (6). I was sugges ed ha RIP mu a ions c ea ed a signal o cy osine me hyla ion (6). We we e in e es ed o know i he me hyla ion pa e n is he same in he case o ans o man s which ne e wen h ough he meio ic cycle. The e o e, we ha e analysed he pa e n o me hyla ion o he p omo e egion o he hph gene in he s ain HTA5.7, which has a leas ou inse s o he hph gene, and in he s ain HTA23, which has only one inse (18). I was epo ed p e iously ha he hph ansgene was me hyla ed a he HpaII si es in he HTA5.7 s ain, while i was no me hyla ed in he HTA23 s ain (16,18). The s e ch o DNA be ween posi ions –359 and –137 o he p omo e egion o he gene in he HTA5.7 s ain was analysed. The pa e n o me hyla ion in he HTA5.7 Nucleic Acids Resea ch, 1997, Vol. 25, No. 12 2412 Figu e 2. Me hyla ed C esidues de e mined by he bisulphi e me hod. The –359 o –137 egion ups eam o he s a codon (+1 is he i s posi ion o he ATG codon), con aining mos o he p omo e egion o he hph gene, was analyzed. Da a om 12 DNA ilamen s cloned om he op s and and 14 clones om he bo om s and, om h ee di e en bisulphi e eac ions, ha e been compiled. Me hyla ed C esidues a e labelled wi h illed symbols. The sequence o wo oligonucleo ides used o cloning o he PCR p oduc a e bisulphi e ea men a e highligh ed. The es ic ion si es o enzymes used in he Sou he n blo analysis p esen ed in Figu e 3 a e unde lined. Y and R in he sequences o he wo oligonucleo ides indica e ha in his posi ion he bases C and T (Y) o A and G (R) ha e been inco po a ed wi h equal p obabili y. The i s se en s ands o he op s and and he i s nine o he bo om s and (coun ing om he op) we e ob ained in he i s bisulphi e ea men , he nex h ee s ands we e ob ained om he second bisulphi e ea men and he las wo s ands we e ob ained om he hi d bisulphi e ea men . s ain is qui e complex, as can be seen in Figu e 2. Twel e ampli ied DNA s ands o he op s and and 14 o he bo om s and ha e been analysed, om a o al o h ee independen bisulphi e eac ions. I can be no iced ha each DNA s and analysed has a di e en me hyla ion pa e n, bu he e is s ill a gene al end. The gene al le el o me hyla ion is e y high, bo h in he op s and (35% o all cy osines a e me hyla ed) as well as in he bo om s and (61%). This le el is much highe han he a e age le el o na u al m5C in Neu ospo a, which is 1.5% o all cy osines (25). The same egion o he hph gene shown in Figu e 2 was analysed om he single inse s ain HTA 23. Eigh clones (se en om he bo om and one om he op s and) we e sequenced om one bisulphi e eac ion. None o he cy osines analysed appea ed o be me hyla ed (da a no shown). An in e nal con ol was pe o med, o es he alidi y o ou p o ocol o he bisulphi e me hod. We also ampli ied a s e ch o he p omo e egion o he cpc gene om he same bisulphi e eac ions used o he analysis in Figu e 2. This gene is no expec ed o be me hyla ed and was ampli ied using he same p ime s as in Selke e al. (6). Fo each bisulphi e eac ion, a s e ch o he p omo e egion o he cpc gene was PCR ampli ied, cloned and i e clones we e sequenced. The sequence had all he C esidues changed o T, e idence ha he bisulphi e eac ed wi h 100% e iciency (da a no shown). The PCR sequencing was e y ai h ul, wi h only h ee mis akes in >5000 bases sequenced. Con i ma ion o he me hyla ion s a e o some cy osines by Sou he n blo analysis Knowing he pa e n o me hyla ion o he p omo e egion o he hph gene in he HTA5.7 s ain om he da a o Figu e 2, we we e able o ob ain a second and independen e alua ion o he me hyla ion s a e o some o he cy osines by Sou he n blo analysis. We diges ed he DNA wi h se e al es ic ion enzymes: wo enzymes whose si es we e no me hyla ed (AluI and C oI), wo enzymes whose si es we e me hyla ed (HphI and Sau3A), one enzyme ha is insensi i e o me hyla ion (MboI) and one enzyme ha cu s a a si e ha does no con ain cy osines (AcsI). The las enzyme was used as a con ol o he quali y o he DNA. The posi ion o he es ic ion si es o hese enzymes in he sequence o Figu e 2 a e: C oI, –186; Sau3A and MboI, –215; HphI, –269; AluI, –320; AcsI, –258. Table 1 showns he es ic ion si es o hese enzymes, he me hylcy osines ha inhibi hei ac i i y and he me hyla ion s a e o hose si es in he s e ch o DNA analysed in Figu e 2. The DNAs used in he Sou he n blo s we e hose o HTA 5.7, he DNA used o he bisulphi e analysis o Figu e 2, HTA23, he single copy ans o man which did no show any me hyla ion acco ding o he bisulphi e me hod, and plasmid pCSN44. The plasmid DNA was used as a posi i e con ol o he ac i i y o he enzymes. The esul s shown in Figu e 3 indica e ha he single copy inse in s ain HTA23 had no me hyla ion. The enzymes HphI and Sau3A did no cu he DNA o he mul icopy inse s ain HTA 5.7, while he o he enzymes did cu i , as expec ed om he da a o Table 1. This is con i ma ion ha o a leas i e di e en cy osines he me hyla ion le el was ha measu ed by he bisulphi e me hod: a posi ions –218 (Sau3A) and –270 (HphI) he cy osines a e me hyla ed, while a posi ions –187 and –189 (C oI) and –222 (AluI) he cy osines a e no me hyla ed. 2413 Nucleic Acids Resea ch, 1994, Vol. 22, No. 1 Nucleic Acids Resea ch, 1997, Vol. 25, No. 12 2413 Table 1. Me hyla ion s a e o he cys osines p esen in he es ic ion si es o he enzymes used in Figu e 3 Res ic ion enzyme Si es cu Si es no cu Me hyla ion s a e o he es ic ion si e in he p omo o egion o he hph gene acco ding o sequence da aa HTA5.7 HTA23 pCSN44 AcsIAATT – – – – AluI AGCT AGm5CT AGCT AGCT AGCT C oI GCGC Gm5CGC GCGC GCGC GCGC GCGm5C HphI TCACC Tm5CACC Tm5CACm5C TCACC TCACC TCACm5C TCAm5CC MboI GATC – GATm5C GATC GATC GATm5C Sau3A GATC GATm5C GATm5C GATC GATC aHTA5.7 acco ding o he sequence o Figu e 2, HTA23 and pCSN44 om sequences no shown. Table 2. F equency o me hyla ion in he cen al C o all possible iple s o sequence 5′-NCN T iple Neu ospo a c assa Pe unia hyb ida Top s and Bo om s and Top s and Bo om s and ACA 58, 67, 50, 67 43, 50, 93, 78 60 24 TCA 58, 75, 67, 50 78, 100, 100, 100 16 0, 0, 0, 92 CCA 50, 42 93, 93, 93, 93, 64, 71, 100 50, 0, 50, 100, 50, 0, 100, 0 76, 16, 76, 68 GCA 67, 17, 25, 25 43 50, 16, 0, 24 76, 60 ACT 0, 17, 0 14, 86, 14, 57 40 16 TCT 42 62, 93, 64 100, 0 100, 84, 8, 92 CCT 58 100, 100, 78 0, 76 0, 24, 76, 76 GCT 0, 41, 8, 0 62, 43, 64, 21, 21, 14 – 0, 16 ACC – 21, 29, 21, 21, 43 0, 0, 0 76 TCC 58 86, 78 50, 16, 0, 0, 0 0, 0, 8, 60, 8, 68 CCC 25, 33 100 0, 0, 0, 16, 0, 8 0 GCC 50, 0 14, 21, 64, 78, 14, 100 16, 40 68 ACG – 86, 64 100, 100, 100, 100, 100 100, 100, 100, 100, 92 TCG 50, 17, 58 86 100, 100, 100 92, 92, 86 CCG – 43, 71, 93 100 – GCG 17, 33, 0, 0 21 92 100, 92 Each en y ep esen s he equency o me hyla ion o one iple . The o de o en ies is 5′→3′ o each s and. The equency o me hyla ion is aken om he da a p esen ed in Figu e 2 o Neu ospo a c assa and om Meye e al. (8) o Pe unia hyb ida. Fo example, he i s ACA iple o he op s and in Neu ospo a is a posi ion –307 and has a equency o me hyla ion o 58%, he nex ACA iple is a posi ion –265 and has a equency o me hyla ion o 67%, e c. The a e age pe cen age o m5C is 35% in he op s and and 61% in he bo om s and o Neu ospo a; i is 25% in he op s and and 41% in he bo om s and o Pe unia. A e cy osine esidues in he p omo e egion o he hph gene me hyla ed a andom? In o de o answe his ques ion he o e all le el o m5C in he op s and and in he bo om s and o he s e ch o DNA shown in Figu e 2 was analysed. In he op s and he e we e, on a e age, 4.2 m5C a each cy osine posi ion o he 12 DNA s ands analysed ( his numbe was ob ained by coun ing he o al numbe o m5C in he 12 op s ands di ided by he numbe o cy osines be ween he wo p ime s in he op s and, 35). I me hyla ion we e andom, independen o he DNA sequence, hen we would expec he dis ibu ion o m5C o ollow a Poisson dis ibu ion. A compa ison be ween he dis ibu ion o he expe imen al da a and he Poisson dis ibu ion is shown in Figu e 4. I is clea ha he e is a disc epancy be ween he wo cu es a bo h low and high pe cen ages o m5C. Simila esul s we e ob ained on analysis o he dis ibu ion o m5C in he bo om s and, which con ains on a e age 8.5 m5C a each cy osine posi ion o he 14 DNA ilamen s analysed (Fig. 4). In bo h s ands he e a e mo e cy osines wi h e y high o e y low le els o me hyla ion han expec ed om a andom p ocess. The e o e, me hyla ion mus ha e some sequence p e e ence, since cy osines a e no me hyla ed a andom. Pa e n o DNA me hyla ion o he p omo e egion o he hph gene in he HTA5.7 ans o man We did no expec he ecogni ion si e o DNA me hyla ion ac i i y o be e y complex, because ∼50% o he cy osines a e me hyla ed. We i s asked whe he in Neu ospo a he con igu - a ion CpG and/o CpNpG is a ecogni ion si e o DNA me hyl ans e ase ac i i y, as i is o mammals and plan s. The Nucleic Acids Resea ch, 1997, Vol. 25, No. 12 2414 Figu e 3. Me hyla ion s a e o he p omo e o he hph gene measu ed by Sou he n blo analysis. The DNAs ha e been cu wi h he ollowing es ic ion enzymes: (i) MboI, which cu s a GATC independen ly o me hyla ion o he C esidue, gi ing DNA agmen s o expec ed sizes 195 and 262 bp; (ii) Sau3A, which cu s he same sequence, GATC, i he C is no me hyla ed; (iii) HphI, which cu s TCACC i nei he o he wo in e nal C esidues a e me hyla ed, gi ing DNA agmen s o expec ed sizes 289 and 1318 bp; (i ) AluI, which cu s AGCT i he C is no me hyla ed, gi ing DNA agmen s o expec ed sizes 60 and 402 bp; C oI cu s GCGC i none o he C esidues a e me hyla ed, gi ing DNA agmen s o expec ed sizes 345 and 440 bp; AcsI, which cu s AATT, gi ing DNA agmen s o expec ed sizes 133 and 506 bp. In each lane labelled HTA5.7, 30 µg HTA5.7 DNA was loaded; in he lanes labelled HTA 23, 30 µg HTA23 was loaded; in he lanes labelled pCSN44, 30 µg wild- ype DNA + 3 ng DNA om plasmid pCSN44 we e loaded. Figu e 4. Dis ibu ion o he me hyla ed C esidues o he op and bo om s and compa ed wi h he Poisson dis ibu ion. The abscissa indica es he me hyla ion deg ee o a gi en C esidue, namely he numbe o DNA ilamen s ha ha e a me hyla ed C esidue a a gi en posi ion o he DNA sequence analysed. The me hyla ion deg ee can be be ween 0 and 12 o he op s and and be ween 0 and 14 o he bo om s and. The o dina e indica es he numbe o C esidues ha ing a gi en me hyla ion deg ee. The a e age alue o he me hyla ion deg ee is 4.2 o he op s and and 8.5 o he bo om s and. Wi h hese wo alues we calcula ed he Poisson dis ibu ion o each s and (ha ched his og am). The his og am wi h a con inuous line shows he expe imen al da a ob ained om he da a p esen ed in Figu e 2. dis ibu ions shown in Figu e 5 indica e ha hey a e no p e e ably me hyla ed. I was o in e es o compa e ou da a wi h he da a on Pe unia hyb ida, he only da a published un il now on he me hyla ion pa e n o a ansgenic gene in plan s (8). We analysed he da a o Meye e al. (8) in he same way as ou own da a and his analysis is shown in Figu e 5. I is e iden ha he con igu a ions CpG and/o CpNpG a e s ong signals o almos 100% me hyla ion o cy osine, as known in plan s o endogenous genes (5). We de e mined he ea e whe he he immedia e neighbou - hood o a cy osine in luences he le el o me hyla ion. The da a on he le el o cy osine me yla ion o a s e ch o he p omo e egion o he hph gene (Fig. 2) we e classi ied in o 16 di e en g oups, depending on he bases ha a e 5′ and 3′ o he cy osine. The da a a e shown in Table 2. All he cy osine esidues in he ollowing iple con ex we e highly me hyla ed (42–100%): ACA (63 ± 16), TCA (78 ± 20), CCA (77 ± 21), TCT (65 ± 21), CCT (84 ± 20), TCC (74 ± 14), ACG (75 ± 15), CCG (69 ± 25); he a e age pe cen age cy osine me hyla ion ± SD is epo ed in 2415 Nucleic Acids Resea ch, 1994, Vol. 22, No. 1 Nucleic Acids Resea ch, 1997, Vol. 25, No. 12 2415 Figu e 5. Dis ibu ion o he me hyla ed C esidues p esen in he CpG and CpNpG sequences. On he le a e he da a o Neu ospo a ob ained by analysis o he da a o Figu e 2. On he igh a e he da a o Pe unia ob ained om he same ype o analysis using he published da a o Meye e al. (8). The abscissa shows he pe cen age o me hyla ion o a gi en C esidue and he o dina e indica es how many di e en DNA ilamen s ha e a any C esidue wi h ha pa icula pe cen age o me hyla ion. The ligh boxes ep esen cy osines om he op s and, da k boxes cy osines om he bo om s and. pa en heses a e each iple . In con as , he cy osines in he middle o he iple s ACC (27 ± 9) and GCG (14 ± 14) ha e <43% me hyla ion, while he cy osines in he o he six iple s ha e bo h high and low le els o me hyla ion. In he case o Pe unia i is no possible o ind a iple wi h a consis en ly high le el o m5C, excep o he ou iple s con aining CpG (ACG, TCG, CCG, GCG); only in he case o he iple CCC (3 ± 6) is he middle C almos ee o me hyla ion (Table 2). This is e idence ha me hyla ion du ing he ege a i e s age o he li e cycle o Neu ospo a has some sequence speci ici y and ha his is di e en om ha o Pe unia. DISCUSSION We ha e p esen ed da a on he pa e n o cy osine me hyla ion in he s ain HTA5.7 con aining a leas ou copies o he p omo e o he hph ansgene (16). Two di e en echniques ha e been used, he bisulphi e me hod and Sou he n blo analysis o DNA diges ed by ou me hyla ion-sensi i e es ic ion enzymes. The le el o me hyla ion is e y high: 35% o all cy osines a e me hyla ed in he op s and and 61% in he bo om s and (Fig. 2). This le el is much highe han he a e age le el o cy osine me hyla ion in wild- ype Neu ospo a (1.5%) (25). We ha e epo ed ha he s ain HTA23 which has, a e ans o ma ion, a single copy inse o he hph gene, has no m5C, in acco d wi h he Sou he n blo da a published p e iously (16,18; da a no shown). We do no know why he e is a di e ence in he me hyla ion le el o he wo s ands. We can only no e ha he same di e ence exis s in he da a published on he pa e n o me hyla ion o a ansgene in Pe unia (8; Table 2). One possibili y is ha he bo om s and is mo e hea ily me hyla ed because i is ansc ibed. The pa e n o DNA me hyla ion o HTA5.7 is complex. I is clea , howe e , ha he me hyla ion p ocess is no a andom one (Fig. 4). The cy osines in he middle o he iple s ACA, TCA, CCA, TCT, CCT, TCC, ACG and CCG ha e consis en ly high le els o me hyla ion (Table 2). The cy osines in he middle o he iple s ACC and GCG ha e low le els o me hyla ion (Table 2). The cy osines in he o he six iple s ha e a wide ange o me hyla ion (Table 2). I is clea om ou da a ha cy osines in he CpG and CpNpG con igu a ions a e no a p e e ed a ge o cy osine me hyla ion, as is he case in plan s (5; Fig. 5). The pa e n o me hyla ion epo ed he e is e y di e en om ha epo ed ea lie in Neu ospo a o he am gene a e induc ion o RIP mu a ions, whe e ‘mos o he molecules assayed showed me hyla ion a mo e han 80% o he cy osines and nea ly hal showed me hyla ion a 95 o 100% o he cy osines’ (6). I seems ha he hypo he ical me hyl ans e ase o Neu ospo a has a di e en speci ici y in he sexual cycle han in he asexual one o ha he e is mo e han one me hyl ans e ase, as sugges ed o Ascobulus (27). Wi h de e mina ion o he in i o pa e n o DNA me hyla ion o he p omo e egion o he o eign gene hph in he mul iple copy s ain HTA5.7 we a e in he posi ion o cha ac e ize he hypo he ical me hyl ans e ase(s), once i is isola ed. Only an isola ed me hyl ans e ase(s) ha p oduces an in i o me hyla ion pa e n simila o ha epo ed he e can be in ol ed in silencing o he hph gene. The ques ion o whe he me hyla ion is essen ial o silencing o he hph gene is s ill open. In a simila sys em, quelling o silencing o he al-1 gene in Neu ospo a, i has been epo ed ha DNA me hyla ion is no essen ial o silencing (28). Howe e , he wo silencing p ocesses a e qui e di e en om each o he (17). Only a comple e analysis o he sys em, a e isola ion and cha ac e iza ion o he DNA me hyl ans e ase(s) and i s gene(s), can elucida e he s ill obscu e bu ascina ing phenomenon o silencing o o eign genes in Neu ospo a. ACKNOWLEDGEMENTS We a e g ea ul o J.Alonso, R.Hakenbeck, S.Meie -Ewe , W.Messe , T.A.T au ne and Jö n Wal e o e y help ul discussions and c i ical eading o he manusc ip . Dedica ed o T.A.T au ne on he occasion o his 65 h bi hday. REFERENCES 1 McClelland,M., Nelson,M. and Raschke,E. (1994) Nucleic Acids Res., 22, 3640–3659. 2 Noye -Weidne ,M. and T au ne ,T.A. (1993) In Jos ,J.P. and Saluz,H.P. (eds), DNA Me hyla ion: Molecula Biology and Biological Signi icance. Bi käuse , Basel, Swi ze land, pp. 39–108. 3 Holliday,R. (1996) In Russo,V.E.A., Ma ienssen,R.A. and Riggs,A.D. (eds), Epigene ics Mechanisms o Gene Regula ion. Cold Sp ing Ha bo Labo a o y P ess, Cold Sp ing Ha bo , NY, pp. 5–28. 4 Yang,A.S., Jones,P.A. and Shiba a,A. (1996) In Russo,V.E.A., Ma ienssen,R.A. and Riggs,A.D. (eds), Epigene ics Mechanisms o Gene Regula ion. Cold Sp ing Ha bo Labo a o y P ess, Cold Sp ing Ha bo , NY, pp. 77–94. 5 Finnegan,E.J. (1996) In Russo,V.E.A., Ma ienssen,R.A. and Riggs,A.D. (eds), Epigene ics Mechanisms o Gene Regula ion. Cold Sp ing Ha bo Labo a o y P ess, Cold Sp ing Ha bo , NY, pp. 127–140. 6 Selke ,E.U., F i z,D.Y. and Singe ,M.J. (1993) Science, 262, 1724–1728. 7 Guyon,C., Noguei a,T.I.V. and Fauge on,G. (1994) J. Mol. Biol., 240, 42–51. 8 Meye ,P., Niedenho ,I. and en Lohuis,M. (1994) EMBO J., 13, 2084–2088. Nucleic Acids Resea ch, 1997, Vol. 25, No. 12 2416 9 Ingelb ech ,I., Van Houd ,H., Van Mon agu,M. and Depicke ,A. (1994) P oc. Na l. Acad. Sci. USA, 91, 10502–10506. 10 Ma ienssen,R. and Ba on,A. (1994) Gene ics, 136, 1157–1170. 11 Tashe a,E.S. and Rou a,D.J. (1994) Mol. Cell. Biol., 14, 5636–5644. 12 Tashe a,E.S. and Rou a,D.J. (1995) Cell. Mol. Gene ., 19, 369–383. 13 Rein,T., Zo bas,H. and DePamphilis,M.L. (1997) Mol. Cell. Biol., 17, 416–426. 14 Rein,T., Na ale,D.A., Gä ne ,U., Niggemann,M., DePamphilis,M.L. and Zo bas,H. (1997) J. Biol. Chem,, 272, 10021–10029. 15 Russo,V.E.A., Ma ienssen,R.A. and Riggs,A.D. (eds) (1996) Epigene ics Mechanisms o Gene Regula ion. Cold Sp ing Ha bo Labo a o y P ess, Cold Sp ing Ha bo , NY. 16 Pandi ,N.N. and Russo,V.E.A. (1992) Mol. Gen. Gene ., 234, 412–422. 17 Russo,V.E.A., Lee,Y.-S. and Codón,A.C. (1996) In Russo,V.E.A., Ma ienssen,R.A. and Riggs,A.D. (eds), Epigene ics Mechanisms o Gene Regula ion. Cold Sp ing Ha bo Labo a o y P ess, Cold Sp ing Ha bo , NY, pp. 345–360. 18 Lee,Y.-S. (1996) Dok o a bei (PhD hesis), F ee Uni e si y Be lin. 19 S aben,C., Jensen,B., Singe ,M., Pollock,J., Schech man,M., Kinsey,J. and Selke ,E. (1989) Fungal Gene . Newsl., 36, 79–81. 20 Olek,A., Oswald,J. and Wal e ,J. (1996) Nucleic Acids Res., 24, 5064–5066. 21 F omme ,M., McDonald,L.E., Milla ,D.S., Collis,C.M., Wa ,F., G igg,G.W., Molloy,P.L. and Paul,C.L. (1992) P oc. Na l. Acad. Sci. USA, 89, 1927–1831. 22 Cla k,S.J., Ha ison,J., Paul,C.L. and F omme ,M. (1994) Nucleic Acids Res., 22, 2990–2997. 23 Feil,R., Cha l on,J., Bi d,A.P., Wal e ,J. and Reik,W. (1994) Nucleic Acids Res., 22, 695–696. 24 Raizis,A.M., Schmi ,F. and Jos ,J.P. (1995) Anal. Biochem., 226, 161–166. 25 Russell,P.J., Wagne ,S., Rodland,K.D., Feinbaum,R.L., Russell,J.P., B e -Ha e,M.S., F ee,S.J. and Me ze be ge ,R.L. (1984) Mol. Gen. Gene ., 196, 275–282. 26 Bull,J.H. and Woo on,J.C. (1994) Na u e, 310, 701–704. 27 Goyon,C., Ba y,C., G égoi e,A., Fauge on,G. and Rossignol,J.-L. (1996) Mol. Cell. Biol., 16, 3054–3065. 28 Cogoni,C., I elan,J.T., Schumache ,M., Schmidhause ,T.J., Selke ,E.U. and Macino,G. (1996) EMBO J., 15, 3153–3163.