Function of the ATP-dependent Metalloprotease FtsH during sporulation in Bacillus subtilis

Func ion o he ATP-dependen me allop o ease F sH du ing spo ula ion in Bacillus sub ilis Disse a ion zu E langung des G ades eines Dok o s de Na u wissenscha en -D . e . na .- de Fakul ä ü Biologie, Chemie und Geowissenscha en de Uni e si ä Bay eu h o geleg on Ai Thi Thuy Le Bay eu h 2008 Die o liegende A bei wu de in de Zei on Janua 2005 bis Ok obe 2007 an de Uni e si ä Bay eu h am Leh s uhl ü Gene ik un e de Be euung on P o . D . Wol gang Schumann ange e ig . Volls ändige Abd uck de on de Fakul ä ü Biologie, Chemie und Geowissenscha en de Uni e si ä Bay eu h genehmig en Disse a ion zu E langung des akademischen G ades eines Dok o s de Na u wissenscha en (D . e . na .). P omo ionsgesuch einge eich am: 09.01.2008 Tag des wissenscha lichen Kolloquiums: 14.03.2008 1. Gu ach e : P o . D . Wol gang Schumann 2. Gu ach e : P o . D . F anz X. Schmid i Acknowledgemen s This hesis was ca ied ou in he g oup o P o . D . Wol gang Schumann a he Depa men o Gene ics, Uni e si y o Bay eu h. I would like o exp ess my special g a i ude o my supe iso P o . D . Wol gang Schumann who sha ed wi h me a lo o his expe ise and esea ch insigh . I app ecia e his excellen di ec ion and use ul commen s on his wo k. I would like o hank PD. D . Thomas Wiege o his help ul commen s h oughou his p ocess. Thanks o Ka in Ange mann o he echnical discussions and he kindness. I wish o hank all my iends and my coun y-ma es in Bay eu h o ines imable iendship. Special hanks o Hue Nguyen, one o my g ea iends, o he assis ance on he minis e ial compu e . My mos since e hanks o he inancial suppo o he Deu sche Fo schungs- gemeinscha (Schu 414/20-2) and he EU (LSGH-CT2004-503468). Las bu no leas , I am o e e indeb ed o my pa en s, my husband and my daugh e o hei unde s anding, endless pa ience and encou agemen when i was mos equi ed. ii Index 1. Summa y 1 Zusammen assung 3 2. In oduc ion 5 2.1. F sH - an ATP-dependen memb ane p o ease 5 2.1.1. S uc u e and unc ion 5 2.1.2. Al e a ion o F sH has pleio opic e ec s on cell physiology 6 2.1.3. Mechanism o deg ada ion by F sH 10 2.1.4. Cy oplasmic p o ein deg ada ion by F sH 11 2.1.5. Memb ane p o ein deg ada ion by F sH 12 2.1.6. Recogni ion s a egy o F sH 13 2.2. The spo ula ion in Bacillus sub ilis 13 2.2.1. Mo phological s ages 14 2.2.2. Spo0A 15 2.2.2.1. Spo0A is a mas e egula o o spo ula ion ini ia ion 15 2.2.2.2. Ac i a ion o Spo0A - The phospho elay 17 2.2.2.3. The au o-s imula ion o Spo0A 18 2.2.3. Regula ion o he phospha e low 19 2.2.3.1. Regula ion o kinase 19 2.2.3.2. Regula ion o esponse egula o s 20 2.2.3.2.1. The Rap amily o phospha ases 20 2.2.3.2.2. The Spo0E amily o phospha ases 21 2.2.3.3. Con ol o he Rap phospha ase 22 2.2.4. Bis able ou coming in spo ula ion 23 2.2.5. SpoVM - an essen ial mo phogene ic p o ein 25 2.3. Cons uc ion o a cold-inducible exp ession sys em o ecombinan p o eins in he B. sub ilis 26 iii 3. Resul s and Discussion 28 Pa A 28 3.1. In luence o F sH on he syn hesis and ac i a ion o he mas e egula o Spo0A 28 3.1.1. Syn hesis o Spo0A is pa ly impai ed in he sH knockou 28 3.1.2. Ac i a ion o Spo0A does no occu in he sH knockou 29 3.1.3. A e a i icial induc ion o ac i e Spo0A, cells a e able o spo ula e in he absence o F sH 30 3.2. The le el o Spo0A is es o ed in he sH spo0E knockou , bu no in he sH ap knockou 30 3.3. The ac i a ion o Spo0A in he phospho elay phospha ase dele ion s ains 31 3.4. Spo ula ion equency inc easing in he phospho elay phospha ase dele ion s ains 31 3.5. Spo0E is a a ge o F sH 32 3.6. The C- e minal end o Spo0E is esponsible o deg ada ion by F sH 33 3.7. The 25 C- e minal amino acids o Spo0E is conside ed as he p o eoly ic ag o F sH-media ed deg ada ion 34 Pa B 35 3.1. Assump ion ha F sH in e ac s wi h he small spo ula ion pep ide SpoVM 35 3.2. Sc eening po en ial sub a es o F sH la e du ing spo ula ion 36 3.3. T ansc ip ion o spoVM du ing spo ula ion 36 3.4. The 5´un ansla ed egion o spoVM ac s as a nega i e egula o o i s own ansla ion 37 Pa C 38 4. Re e ences 39 i 5. Own Con ibu ion 53 Appendix : Own publica ions Pa A 54 Pa B 85 Pa C 101 6. Abb e ia ions 108 E klä ung I. Summa y 1 I. Summa y The analysis o he unc ion o he sH gene o Bacillus sub ilis s a ed abou en yea s ago. I was shown a ha ime ha an sH knockou was iable, bu exhibi ed a pleio opic pheno ype. Cells a e sensi i e o sal and hea shock, exhibi ilamen ous g ow h, a e di icul o be ans o med and a e almos unable o spo ula e. Despi e he se e e pheno ype caused by he absence o he sH gene, he p ecise unc ions o his p o ein emained unclea . This PhD hesis p esen s da a o elucida e he unc ion o sH du ing spo ula ion. Fu he mo e, i desc ibes he cons uc ion o a cold-inducible exp ession sys em. The majo inding o his hesis is ha he F sH p o ease in e e es wi h he syn hesis and/o phospho yla ion o Spo0A, he mas e egula o du ing ini ia ion o spo ula ion called phase 0. In he sH knockou , he amoun o Spo0A is g ea ly educed, and he small amoun s p esen a e inac i e. When he wild- ype spo0A allele was eplaced by an IPTG-inducible allele coding o mu an Spo0A p o ein being ully ac i e in he absence o phospho yla ion (Spo0A-Sad67), spo es we e o med a a no mal a e in an sH knockou . Again, his esul indica es ha F sH is clea ly in ol ed in he o ma ion o ac i e Spo0A and ha his p o ease is only essen ial du ing s age 0 o spo ula ion. To become ac i e, Spo0A needs o be phospho yla ed by he mul i-componen sys em called phospho elay. Since no ac i e Spo0A is p esen in an sH knockou , i was hypo hesized ha F sH has o deg ade one o mo e nega i e egula o (s) ei he p e en ing he phospho yla ion o Spo0A o /and being in ol ing in i s apid dephospho yla ion. The u he analysis ocused on ou an agonis s o he phospho elay, h ee Rap phospha ases being in ol ed in he dephospho yla ion o Spo0F~P, and Spo0E which a ge s Spo0A~P. When a null allele in any one o hem was combined wi h he sH knockou , he wild- ype amoun o Spo0A was es o ed only in he case o he sH spo0E knockou and he spo ula ion equency was inc eased by wo o h ee o de s o magni ude in all double knockou s, bu emained below 1%. Since o e exp ession o Spo0E educes he spo ula ion equency and emo al o he gene om he genome has an opposi e e ec , a di ec in e ac ion be ween F sH and Spo0E was en isaged. In i o p o eolysis assays wi h pu i ied GST-F sH and GST-Spo0E showed ha Spo0E is indeed a a ge o F sH. In con as , he wo homologs o Spo0E, YisI and YnzD, emained s able upon incuba ion wi h F sH. Since all h ee p o eins a e dis inguished by a C- e minal ex ension o abou 25 amino acids p esen in Spo0E, bu no in he wo o he phospha ases, hese addi ional amino acids could se e as a a ge o F sH. When wo mu an e sions o Spo0E, Spo0E94 and Spo0E11, wi h unca ed C- e minal ends we e I. Summa y 2 analyzed, hey u ned ou o be s able in he p esence o F sH. When he C- e minal 25 amino acids was ans e ed o YnzD, his usion p o ein became uns able when incuba ed wi h F sH. In conclusion, he C- e minal end o Spo0E con e s ins abili y o his enzyme. Since a spo0E knockou in a wild- ype backg ound does no esul in a spo ula ion equency close o 100% and a combina ion o a spo0E and an sH knockou aises he spo ula ion equency only close o 1%, i can be concluded ha he e a e addi ional a ge s o F sH in e e ing wi h he syn hesis o ac i e Spo0A. Mo eo e , i is likely ha F sH also exe s a unc ion la e du ing spo ula ion. I could be shown ha SpoVM, a small pep ide essen ial o spo e mo phogenesis, inhibi s he p o eoly ic ac i i y o he B. sub ilis F sH p o ease in i o. I can be in e ed ha SpoVM also inhibi s ac i i y o F sH du ing spo ula ion, and in he absence o SpoVM, F sH will deg ade a leas one p o ein essen ial o success ul comple ion o spo ula ion. When he in acellula p o eomes o spoVM+ and spoVM- cells we e compa ed, a o al o 83 p o eins we e iden i ied being ei he comple ely absen o p esen in educed amoun s in he absence o he pep ide. Analysis o he exp ession o he spoVM gene e ealed ha cells s a ed o syn hesize he spoVM ansc ip a s age 2 while he SpoVM pep ide accumula ed a s age 4. The 5´ un ansla ed egion o he spoVM ansc ip has been iden i ied o ac as a nega i e egula o o i s own ansc ip ion o ansla ion. Fu he mo e, a cold-inducible exp ession sys em has been cons uc ed allowing in a- and ex acellula p oduc ion o ecombinan p o eins. This exp ession sys em makes use o a wo-componen signal ansduc ion sys em, which senses changes in he luidi y o he cy oplasmic memb ane. I. Zusammen assung 3 I. Zusammen assung Die Analyse de Funk ion des sH-Gens on Bacillus sub ilis begann o e wa 10 Jah en. Damals konn e gezeig we den, dass eine sH Knockou -Mu an e lebens ähig is , abe übe einen pleio open Phäno yp e üg . Die Zellen sind Salz- und Hi ze- sensi i , wachsen ilamen ös, sind schwie ig zu ans o mie en und zeigen eine s a k eduzie e Spo ula ions equenz. T o z diese g a ie enden Phäno ypen blieb die Funk ion on sH bislang im Dunkeln. Diese Dok o a bei p äsen ie Da en, die einige Funk ionen on sH wäh end de Spo ula ion au decken. Auße dem wi d die Kons uk ion eines Käl e-induzie ba en Exp essionssys ems besch ieben. Das besonde e E gebnis diese Disse a ion is de Be und, dass die F sH P o ease mi de Syn hese und/ode de Phospho ylie ung on Spo0A, dem Mas e - Regula o wäh end de Ini ia ion, de Phase 0, in e e ie . In eine sH-Knockou is die Menge an Spo0A signi ikan eduzie und die ge ingen Mengen sind inak i . Wenn das Wild yp-Allel on spo0A du ch ein IPTG-induzie ba es Allel e se z wu de, welches ü ein mu an es P o ein codie , das auch in Abwesenhei on Phospho ylie ung oll ak i is (Spo0A-Sad67), dann wu de eine Spo ula ions equenz gemessen, die de on Wild yp-Zellen en sp ach. Aus diesem E gebnis is zu olge n, dass sH nu wäh end de Phase 0 essen iell is . Um Ak i i ä zu e langen, muss Spo0A phospho ylie we de, und dies geschieh du ch ein Phospho elay. Da in eine sH-Mu an e kein ak i es Spo0A nachweisba is , is zu e mu en, dass F sH einen ode meh e e nega i e Regula o en abbauen muss, die en wede die Phospho ylie ung on Spo0A e hinde n ode an eine schnellen Dephospho ylie ung be eilig sind. Die wei e e Analyse konzen ie e sich au ie e schiedene An agonis en des Phospho elays, d ei Rap Phospha asen, die Spo0F~P dephospho ylie en und Spo0E, welche Spo0A~P dephospho ylie . Wenn Null- Allele de ie Phospha asen mi eine sH-Knockou kombinie wu den, dann wu de nu im Fall on Δspo0E Wild yp-Mengen an Spo0A de ek ie . Die Spo ula ions equenz wu de in allen ie S ämmen um 2-3 G ößeno dnungen e höh gegenübe de sH- Knockou , blieb abe in allen Fällen un e 1%. Da eine Übe exp ession on Spo0E die Spo ula ions equenz eduzie und ein spo0E-Knockou den gegen eiligen E ek ha , wu de eine di ek e In e ak ion zwischen F sH und Spo0E in Be ach gezogen. In i o P o eolyse es s mi ge einig em GST-F sH und GST-Spo0E e gaben, dass Spo0E abgebau wi d. Im Gegensa z dazu e wiesen sich zwei Homologe on Spo0E, YisI und YnzD, als s abil. Da alle d ei Phospha asen sich nu in ih en N-Te mini un e scheiden und nu Spo0E einen um e wa 25 Aminosäu e es e e länge en C-Te minus en häl , wa zu e mu en, dass diese Anhang on F sH als Ta ge e kann wi d. Wenn zwei 2. In oduc ion 10 Spo0A is esponsible o he ailu e o spo ula e in a sH null mu an s ain (Deue ling, e al., 1997). Hence, he i s unc ion o F sH p oposed is ela ed o he ini ia ion s ages o spo ula ion. Ano he s udy could show ha a spoVM mu an is blocked in exp ession o sigmaK-dependen genes (as expec ed o a mu an blocked a s age II-III). Ex agenic supp esso s o spoVM mu an s mapped in he sH gene and i was e iden ha he 26- amino-acid SpoVM pep ide could inhibi he p o eoly ic ac i i y o F sH (Cu ing e al., 1997). 2.1.3. Mechanism o subs a e deg ada ion by F sH All cha ac e ized ATP-dependen p o eases exhibi mul ime ic ing-like s uc u es in which he p o eoly ic ac i e si e is bu ied wi hin a cen al ca i y. A consequence o his con igu a ion is ha he ac i e si e is accessible only o un olded and ex ended polypep ides. Acco ding o he p e ailing model, p o eases u ilize speci ic deg ada ion signals o ecognize and bind subs a es. Then, powe ed by ATP hyd olysis, hey un old hese subs a es and ansloca e hem in o he p o eoly ic chambe (Saue e al., 2004). Wi hin his chambe , he p o ein is hyd olyzed o small pep ides, which a e eleased in o he cy oplasm. The e idence suppo ing his model is de i ed p ima ily om s udies o he bi-pa i e cy oplasmic ClpAP and ClpXP p o eases (Webe -Ban e al., 1999; Kim e al., 2000; Hoskins e al., 2000). These Clp- amily p o eases exhibi obus un oldase ac i i y, enabling hem o deg ade subs a e p o eins wi h in insic he mos abili ies (Kennis on e al., 2003). A ecen wo k indica es ha F sH ope a es h ough a di e en mode o deg ada ion han hese cy oplasmic p o eases (He man e al., 2003). In con as o he Clp- amily p o eases, F sH lacks a obus un oldase ac i i y which would be necessa y o h ead a subs a e in o he p o eoly ic ca i y. I was p oposed ha he weak un oldase ac i i y allows he cell o con ol he a e a which F sH deg ades egula o y a ge s like σ32 and LpxC (He man e al., 2003). Unde app op ia e condi ions, i is likely ha F sH employs co- ac o s ha accele a e he un olding and he deg ada ion o a ge s. Thus, F sH can sense he olding s a e o egula o y subs a es wi hin he physiological ange o he cell. Apa om i s ole in egula ed p o eolysis, i was specula ed ha he weak un oldase ac i i y employed by F sH may be a cen al ea u e o memb ane p o ein deg ada ion in gene al (He man e al., 2003). F sH migh be able o sense he olded s a e o p o eins, and deg ade only hose ha exhibi low in insic he mos abili y (e.g. unassembled o mis olded memb ane p o eins), independen ly o a speci ic deg ada ion signal. 2. In oduc ion 11 2.1.4. Cy oplasmic p o ein deg ada ion by F sH Mos o he F sH subs a es hus a cha ac e ized a e om E. coli in which soluble subs a es a e he mos pa na u ally sho -li ed. The bac e iophage lambda p o ein CII is sho -li ed ansc ip ion ac o o genes equi ed o he es ablishmen o lysogeniza ion. I is apidly deg aded by F sH in i o and in i o (Kiha a e al., 1997; Sho land e al., 1997; Sho land e al., 2000). The cIII gene p oduc , a small memb ane- in e ac ing p o ein wi h an amphiphilic α-helix egion, is slowly deg aded by F sH in i o (He man e al., 1997), and i an agonizes he F sH-ca alyzed in i o p o eolysis o CII (Sho land e al., 2000). Thus, he balance among hese gene p oduc s is an impo an ac o o he decision be ween he ly ic g ow h and lysogeniza ion/in eg a ion o he in ec ing λ genome (Hoy e al., 1982). The hea shock sigma ac o σ32 is apidly deg aded in he absence o a hea shock o o he s ess, wi h an hal -li e ha is a ec ed by he le el o F sH (He man e al., 1995b; Tomoyasu e al., 1995) as well as by o he ATP-dependen p o eases (Kanemo i e al., 1997). In a pu i ied eac ion sys em, F sH deg ades σ32 in Zn2+- and ATP- dependen manne s (Okuno e al., 2004, Toyomasu e al., 1995). F sH may also con ibu e o he deg ada ion o ano he sho -li ed ansc ip ion ac o , SoxS (G i i h e al., 2004). LpxC is he mos impo an F sH subs a e because deg ada ion o LpxC ende s F sH essen ial. Since he same eac ion p ecu so (R-3-hyd oxymy is oyl-ACP) is used by he lpxC (en A) gene-encoded deace ylase o he biosyn hesis o lipid A, a LPS componen , and by he abA gene-encoded dehyd ase o a y acid biosyn hesis, he balance o hese enzymes is impo an o main ain a p ope LPS/phospholipids a io in E. coli cells. The LpxC deace ylase is sho -li ed (hal -li e o 4 min) owing o F sH-ca alyzed deg ada ion, dys unc ion o which esul s in he le hal o e -accumula ion o LPS (Ogu a e al., 1999). LpxC is a globula p o ein. Basing on he s uc u e o LpxC om Aqui ex aeolicus has been sol ed (Coggins e al., 2003; Whi ing on e al., 2003) sugges ing ha he inal 14 esidues a e no s uc u ed. As he E. coli p o ein con ains an ex ension o ele en esidues, an uns uc u ed C- e minus o 25 amino acids was p edic ed. In i o deg ada ion LpxC by F sH was e iden and a de ailed mu a ional analysis e ealed six non-pola esidues in he C- e minus o LpxC ha a e c i ical o deg ada ion (Füh e e al., 2007). The Ss A- ag is a sho sequence ha is appended o he C- e minus o unca ed p o eins on s alled ibosomes o p omo e hei p o eolysis (Keile e al., 1996). In E. coli, deg ada ion occu s mainly h ough he ClpAP/XP p o eases (Keile e al., 1996; Go esman e al., 1998), which ecognize speci ic esidues wi hin he Ss A- ag (Flynn e 2. In oduc ion 12 al., 2001). In addi ion, F sH also ecognizes he Ss A- ag and e icien ly deg ades he λCI-Ss A model subs a e (He man e al., 1998) hough he λCI N- e minal domain is a s able cy osolic p o ein. I appea s ha he Ss A- ag con e s a ela i ely unspeci ic deg ada ion signal o he p o ein a which i is a ached (He man e al., 1998). I emains o be elucida ed how he di e en p o eases, ClpAP/XP and F sH, ecognize he same deg ada ion ag. 2.1.5. Memb ane p o ein deg ada ion by F sH The majo housekeeping unc ion p oposed o F sH is he apid emo al o ha m ul memb ane p o ein subuni s when hey ail o in eg a e in o unc ional complexes (I o e al., 2005). A s udy o memb ane deg ada ion has been ca ied ou wi h wo memb ane subs a es, he anslocase subuni SecY, and YccA, a memb ane p o ein o unknown unc ion. SecY is one o he majo componen s o he ansloca ion appa a us. I becomes a subs a e o F sH media ed deg ada ion when i ails o assemble in o a complex wi h i s pa ne , SecE and SecG (Kiha a e al., 1995). This can occu ei he when SecY is o e exp essed, o when he SecY-SecE in e ac ion is weakened h ough a mu a ion (Chiba e al., 2000). YccA is inhe en ly uns able, and i is unknown whe he YccA o ms any highe o de complexes (Kiha a e al., 1998). YccA is deg aded by F sH wi h p o ease ecogni ion h ough i s cy oplasmic ail. Howe e , YccA associa es wi h F sH e en when i s deg ada ion signal is absen (Kiha a e al., 1998). F sH can also deg ade in eg al memb ane p o eins s a ing om hei cy oplasmic ends, p o ided ha he e is an un olded cy oplasmic ail, o any sequence composi ion, ha is a leas 20 amino acids in leng h (Chiba e al., 2002). Such deg ada ion can begin a he N- o C- e minus and p oceeds sequen ially o he o he end, appa en ly pulling he p o ein h ough he memb ane, as he pe iplasmic domains a e deg aded in an F sH-dependen manne (Chiba e al., 2002). Because he p o eoly ic ac i e si e o F sH esides in he cy oplasm, subs a es mus be ex ac ed om he lipid bilaye be o e en e ing he p o eoly ic ca i y, a p ocess called disloca ion (Kiha a e al., 1999). Disloca ion o memb ane p o eins aises he in iguing ques ion o how hyd ophilic domains a e se he plasma memb ane. Nume ous s udies on p o ein anspo ac oss cellula memb anes ha e demons a ed he equi emen o p o einaceous hyd ophilic po es o he ansloca ion p ocess (P akash e al., 2004). Mo e speci ically, he u no e o memb ane p o eins in he yeas endoplasmic e iculum, media ed by he 26S p o eosome, in ol es he e og ade ansloca ion o subs a es in o he cy osol ia he Sec61 anslocase machine y (S i ling e al., 2006; G oll e al., 2005). Sec61 is he yeas homolog o E. coli SecY. A ecen in i o s udy, using pu i ied F sH and YccA in econs i u ed p o eoliposomes, showed ha 2. In oduc ion 13 F sH is able o disloca e and deg ade YccA in he absence o o he componen s (Akiyama e al., 2003). 2.1.6. Recogni ion s a egy o F sH The abili y o F sH o disc imina e be ween co ec and inco ec p o ein subs a es is c i ical o bo h i s housekeeping and egula o y unc ions. F sH ecognizes and deg ades cy oplasmic p o eins ha con ain C- e minal non-pola ails (He man e al., 1998). This ail-speci ic ecogni ion is physiologically ele an , and is used o deg ading he λCII ac i a o , Ss A- agged p o eins and he LpxC deace ylase (He man e al., 1998; Kobile e al., 2002). The Ss A- ag consis s o 11 esidues ha is appended o he C- e minus o unca ed p o eins on s alled ibosomes o p omo e hei p o eolysis (Keile e al., 1996). The a o emen ioned ail-speci ic ecogni ion, howe e , he C- e minal egion is no impo an o deg ada ion in case o he hea shock ac o σ32 (Toyomasu e al., 2001); Ins ead, an in e nal egion may be impo an o he F sH-media ed deg ada ion o his p o ein (Be ani e al., 2001). A mode o σ32 p o eolysis is also desc ibed, in which i is deg aded by F sH in he N- o C- e minus di ec ion (Okuno e al., 2004; Ob is e al., 2007). I is clea ha F sH eadily deg ades unc ional na i e p o ein subs a es ha a e suscep ible o he p o eoly ic eac ion o F sH in ini ia ion-signal-dependen manne s. This mode o p o eolysis likely in ol es sequen ial subs a e un olding ha could be ini ia ed a an ini ia ion signal and hen p opaga ed along he polypep ide chain (I o e al., 2005). In addi ion o speci ic sequence ecogni ion, he subs a e p o ein he mos abili y plays an impo an ole in he decision o deg ade i ; he mo e he mos able he p o ein, he less likely ha F sH will deg ade i , e en when i ca ies a good ecogni ion ag. This is due o he ac ha F sH does no possess a obus un oldase ac i i y, which would be necessa y o h ead a he mos able subs a e in o he p o eoly ic ca i y. Lacking o a obus un oldase allows F sH o ca y ou a second egula o y s ep in he decision o deg ade a p o ein by enabling F sH o sense he olding s a e o p o eins wi hin he physiological ange o he cell, and deg ade only hose ha display low he mos abili y (He man e al., 2003). This second egula o y s ep in subs a e selec ion may be c ucial o he deg ada ion o egula o y p o eins and memb ane p o eins. σ32, a na u al subs a e o F sH, indeed con ains egion o low he modynamic s abili y (I o e al., 2005). 2.2. The spo ula ion in Bacillus sub ilis Unde condi ions o nu ien dep i a ion, cells o B. sub ilis can unde go a p ocess o de elopmen ha leads o he o ma ion o do man , en i onmen ally esis an spo es. Spo ula ion akes app oxima ely six o eigh hou s and in ol es ex ensi e changes in 2. In oduc ion 14 gene exp ession and mo phology (E ing on, 2003; Hilbe e al., 2004; Piggo e al., 2004). The hallma k o endospo e o ma ion is an asymme ic cell di ision ha p oduces wo cell ypes: a la ge cell called he mo he cell, and a smalle cell called he o espo e. 2.2.1. Mo phological s ages The se ies o complex mo phological changes ha occu du ing he spo ula ion p ocess in B. sub ilis ha e been ex ensi ely s udied (Fig. 4). En y in o spo ula ion is cha ac e ized by he o ma ion o a so-called axial ilamen in which wo ch omosomes om he las ound o DNA eplica ion become aligned ac oss he long axis o he cell. Nex , a sep um is o med a an ex eme pola posi ion. This pa i ions he de eloping cell (he ea e e e ed o as he spo angium) in o la ge and small compa men s known as he mo he cell and he o espo e (o p espo e), espec i ely, whe e each ecei es a ch omosome (S agie and Losick, 1996). Ini ially, he la ge and he small compa men s lie side-by-side, bu in he nex s age o de elopmen he o espo e becomes engul ed by he mo he cell. Du ing engul men , he memb ane on he mo he cell ace o he pola sep um mig a es a ound he memb ane su ounding he o espo e and e en ually he o espo e is comple ely pinched o as a ee p o oplas wi hin he mo he cell, such ha he spo angium becomes a cell- wi hin-a-cell ( o his eason spo es o Bacillus and ela ed gene a a e mo e p ope ly known as endospo es) (S agie and Losick, 1996). In subsequen mo phogenesis, he o espo e p oduces la ge amoun s o a amily o small acid-soluble p o eins known as SASP. Some o hese p o eins bind o and coa he o espo e ch omosome, packaging i in o a doughnu -like s uc u e and con e ing on i esis ance o ul a iole adia ion. Meanwhile, in he in e memb ane space be ween he o espo e and mo he cell, a hin laye o pep idoglycan known as he ge m cell wall is p oduced on he su ace o he o espo e memb ane. This is ollowed by he syn hesis o a hick laye o pep idoglycan known as he co ex, which is hough o be in ol ed in a aining o main aining he dehyd a ed and hea - esis an s a e o he spo e. The mo he cell p oduces a p o einaceous coa ha assembles on he ou side su ace o he mo he - cell memb ane a ound he o espo e. The coa consis s o a lamella inne laye and an elec on-dense ou e laye and p o ides a hick, p o ec i e ba ie ha encases he ma u e spo e. E en ually, a e abou 6–8 hou s o de elopmen , when ma u a ion is comple e, he ully ipened spo e is libe a ed by lysis o he mo he cell. Thus, he mo he cell is mo al in ha i unde goes p og ammed cell dea h, whe eas he o espo e is immo al in ha i becomes he spo e and gi es ise o subsequen p ogeny (S agie and Losick, 1996). 2. In oduc ion 15 FIG. 4. Schema ic ep esen a ion o he s ages o spo ula ion in Bacillus sub ilis This sequence o mo phological e en s is di ided in o di e en s ages: S age 0 ep esen s cells ha ha e no en e ed he spo ula ion pa hway. S age I ep esen s cells ha ha e en e ed he pa hway and ha e o med an axial ilamen ; S age II and III e e o spo angia ha ha e eached he s ages o pola sep a ion and engul men , espec i ely. Syn hesis o a dis inc i e o m o pep idoglycan be ween he memb anes su ounding he p espo e is de ined as s age IV. Deposi ion o spo e coa a ound he p espo e is de ined as s age V. S age VI is ma u a ion, when he spo e acqui es i s ull esis an p ope ies spo e. S age VII ep esen s lysis o he mo he cell and elease o he ma u e spo e (Hilbe and Piggo , 2004) 0- G ow h I- Axial Filamen a ion II Asymme ic Sep um III- Engul men IV- Co ex Syn hesis V- Coa Syn hesis VII- Mo he Cell Lysis V I- Ma u a ion 2.2.2. Spo0A 2.2.2.1. Spo0A is a mas e egula o o spo ula ion ini ia ion The mas e egula o o en y in o spo ula ion in B. sub ilis is he DNA-binding p o ein Spo0A, which is a membe o he esponse egula o amily o ansc ip ion ac o s (Pe ego and Hoch, 2002). The ac i a ion o his key ansc ip ional egula o y p o ein occu s h ough en i onmen al and physiological signals, igge ed by nu ien deple ion and cell densi y. Ac i a ion o Spo0A p oceeds h ough se e al phases. Ini ial ac i a ion a he end o exponen ial g ow h leads o he ‘ ansi ion s a e’, which is associa ed wi h such phenomena as p o ease p oduc ion, mo ili y, compe ence o ans o ma ion (Sonenshein, 2000), bio ilm o ma ion (B anda e al., 2001; Hamon and 2. In oduc ion 16 Lazazze a, 2001) and e en cannibalism (Gonzalez-Pas o e al., 2003) and spo e o ma ion which is hough o equi e inc eased he phospho yla ion o Spo0A. The molecula de ails o he in e ac ion o Spo0A wi h i s a ge DNA, he ‘Spo0A box’, a consensus 7-bp sequence (5′-TGNCGAA-3′, wi h a p e e ence o N = T), ha e now been analyzed wi h a c ys al s uc u e (Zhao e al., 2002). Spo0A has been ound o in luence, di ec ly o indi ec ly, he exp ession o o e 500 genes du ing he ea ly s ages o de elopmen . An app oach in combina ion wi h ansc ip ional p o iling using gene mic oa ays, gel elec opho e ic mobili y shi assays, using he DNA-binding domain o Spo0A, and bioin o ma ics enabled o assign a o al 121 genes, which a e o ganized as 30 single-gene uni s and 24 ope ons, a e likely o be unde he di ec con ol o Spo0A. Abou one- hi d o hese genes a e ac i a ed and he emainde a e ep essed (Molle e al., 2003). Among he iden i ied membe s o he egulon whe e ansc ip ion was s imula ed by Spo0A a e genes o me abolic enzymes and genes o e lux pumps (Molle e al., 2003). Among he membe s whe e ansc ip ion ha was inhibi ed by Spo0A a e genes encoding componen s o he DNA eplica ion machine y and genes ha go e n lagellum biosyn hesis and chemo axis. Du ing s age 0 o spo ula ion, he ac i e o m o Spo0A (Spo0A~P) ac s as a ep esso o ce ain ege a i ely exp essed genes (e.g. ab B) (Pe ego e al., 1988; S auch e al., 1989; S auch and Hoch, 1993; Fuji a and Sadaie, 1998) and an ac i a o o genes di ec ly in ol ed in spo ula ion (Piggo and Losick, 2002). So a , a o al o 10 ansc ip ion uni s which a e o ganized in six single-gene uni s (ab B, kinA, kinC, spo0A, spo0F and spoIIE) and ou ope ons (dl , sin, spoIIA and spoIIG) a e con olled by Spo0A~P. Among he genes ac i a ed by Spo0A~P a e hose in ol ed in emodeling he sis e ch omosomes o he spo ula ing cell in o an ‘axial ilamen ’ (Pogliano e al., 2002; Ben-Yehuda e al., 2003) and in he o ma ion o a pola sep um ha di ides he de eloping cell in o a small o espo e compa men and a la ge mo he cell compa men (Le in and Losick, 1996; Ben-Yehuda and Losick, 2002). Spo0A~P is also esponsible o ac i a ing genes ha lead o he appea ance o he cell-speci ic egula o y p o eins σF and σE which ac in he o espo e and he mo he cell, espec i ely (S agie and Losick, 1996; Piggo and Losick, 2002). Recen wo k indica es ha Spo0A~P con inues o unc ion a e he pola sep um is o med, when i accumula es o high le els and di ec s ansc ip ion in he mo he cell (Fuji a and Losick, 2003). Clea ly, Spo0A has a p o ound e ec on he global pa e n o gene exp ession (Molle e al., 2003). Impo an ly, cells equi e a high h eshold o ac i e Spo0A o ini ia e spo ula ion (Fuji a e al., 2005). Mu a ions wi hin he phospho elay, leading o lowe concen a ions o in acellula Spo0A~P, caused a smalle popula ion o cells ini ia ing spo ula ion 2. In oduc ion 17 (Molle, e al., 2003). The e a e ou ca ego ies o genes wi hin he Spo0A egulon ha espond o di e en h esholds o Spo0A as ollows: (i) hose ha equi e a high le el o Spo0A o become ac i a ed, (ii) hose ha equi ed a high le el o Spo0A o be ep essed, (iii) hose ha we e ac i a ed a a low le el o he egula o , and (i ) hose ha we e ep essed a a low dose o he egula o . Genes ha equi ed a high dose o Spo0A o be ac i a ed we e ound o ha e low binding cons an s o Spo0A~P. Some genes ha we e u ned on a a low dose o Spo0A ei he had a high binding cons an o he egula o y p o ein o we e ac i a ed by an indi ec mechanism in ol ing Spo0A-media ed elie o ep ession by ep esso p o ein Ab B (Fuji a e al., 2005). Mo eo e , Spo0A~P also se s in mo ion se e al posi i e and nega i e egula o y loops ha go e n he a e o exp ession o spo0A and o he elay and phospha ase genes. De ailed accoun s o he unc ioning o he phospho elay, he na u e o hese egula o y loops, and he modula ion o phospha e low by speci ic phospha ase a e indica ed below. 2.2.2.2. Ac i a ion o Spo0A - The phospho elay The ac i i y o Spo0A is go e ned by a mul icomponen phospho elay - an ex ended e sion o he ypical wo-componen sys em, which consis s o i e his idine au okinases (KinA, KinB, KinC, KinD and KinE) and wo phospho elay p o eins (Spo0F and Spo0B) (Jiang e al., 2000) (Fig. 5). Di e en ial signals ac i a e mul iple his idine kinases o au ophospho yla e and hen ans e hei phospho yl g oup o he in e media e esponse egula o Spo0F (Bu bulys e al., 1991; Jiang e al., 2000). Spo0F∼P is he subs a e o a phospho ans e ase, Spo0B, which ans e s he phospho yl g oup o he Spo0A esponse egula o and ansc ip ion ac o (Bu bulys e al., 1991). Signal in eg a ion is he esponsibili y o he phospho elay, whose s uc u al complexi y e lec s he equi emen o p ecise coo dina ion o nume ous cellula e en s. The mul icomponen s uc u e o he phospho elay p o ides mul iple en ies o egula o y signals a ec ing he inal goal o p oducing he app op ia e le el o Spo0A~P. These egula o y mechanisms a e exe ed bo h on he le el o ansc ip ion o he phospho elay componen s and on hei enzyma ic ac i i y. KinA is he p ima y kinase in he phospho elay and i has he majo ole a he onse o spo e o ma ion (S ephenson and Hoch, 2001). I has been demons a ed ha he ac ion o cells ha ini ia e spo ula ion is dec eased in a kinA mu an backg ound (Chung e al., 1994). The mos amino- e minal end o he domains in KinA is impo an o spo e o ma ion and has been shown o bind ATP, bu is unlikely o be egula ed di ec ly by ATP le els (S ephenson and Hoch, 2001). 2. In oduc ion 18 2.2.2.3. The au o-s imula ion o Spo0A The ac i i y o Spo0A is subjec o se e al au o-s imula o y loops (S auch e al., 1992; S auch e al., 1993; Fuji a and Sadaie, 1998). These loops in ol e ansc ip ion o spo0A and phospho yla ion o Spo0A. T ansc ip ion o spo0A is di ec ly ac i a ed by Spo0A~P (S auch e al., 1992) and indi ec ly ac i a ed by induced exp ession o spo0H. Fi s o all, Spo0A~P ep esses he exp ession o ab B, a gene encoding a ansc ip ional FIG. 5. The phospho elay signal ansduc ion sys em o spo ula ion ini ia ion. In he phospho elay, wo cy oplasmic kinases (KinA and KinE) and h ee memb ane- bound kinases (KinB, KinC, KinD) phospho yla e he Spo0F (0F) esponse egula o in esponse o di e en ial signals. Spo0F∼P ans e s he phospho yl g oup o he Spo0B (0B) phospho ans e ase ha , in u n, ans e s i o he Spo0A (0A) esponse egula o and ansc ip ion ac o o spo ula ion ini ia ion. KapB is a lipop o ein essen ial o KinB ac i i y. The Rap phospha ases dephospho yla e he Spo0F∼P in e media e while Spo0E, YisI and YnzD dephospho yla e Spo0A∼P. T ansc ip ion o he phospha ase coding genes is ac i a ed by physiological condi ions an i he ical o spo ula ion such as g ow h and compe ence o DNA ans o ma ion (Pe ego, 2001). 2. In oduc ion 19 egula o ha inhibi s a ious s a iona y phase p ocesses (Robe son e al., 1989). This esul s in an indi ec au os imula o y loop ac i ed ia he ansc ip ional egula o Ab B. Du ing exponen ial g ow h, Ab B ep esses a ious s a iona y phase p ocesses, including he ansc ip ion o genes equi ed o spo ula ion (e.g. kinA) (S auch e al., 1989). Impo an ly, Ab B ep esses gene exp ession o he al e na i e RNA polyme ase sigma ac o σH ha ecognizes an al e na i e p omo e ups eam o spo0A, and in addi ion, ac i a es genes equi ed o phospho yla ion o Spo0A such as kinA and spo0F (P edich e al., 1992). Thus, when Spo0A is phospho yla ed, alle ia ion o Ab B ep ession by Spo0A~P s imula es bo h ansc ip ion o spo0A and indi ec ly phospho yla ion o Spo0A. A simpli ied scheme o he au os imula ion o Spo0A is desc ibed in Fig. 6. FIG. 6. Simpli ied schema ic ep esen a ion o he au os imula o y loop in ol ing ansc ip ion and ac i a ion o Spo0A. Pe pendicula s and a ows ep esen he nega i e and posi i e egula ions, espec i ely (Smi s e al., 2006). 2.2.3. Regula ion o he phospha e low 2.2.3.1. Regula ion o kinase The disco e y ha he le el o Spo0A~P is c ucial in de e mining he cell a e led o he disco e y o a se ies o mechanisms ha modula e he lux o phospha e in he phospho elay in esponse o speci ic signals. The i s le el o con ol is on he his idine kinases, KinA and KinB. Al hough mechanisms o ac i a ion o hese kinases a e p edic able, a nega i e egula o o phospha e inpu has been desc ibed as an inhibi o o he kinase ac i i y o KinA. KipI is a po en inhibi o o he au ophospho yla ion eac ion o kinase A bu does no inhibi phospha e ans e o he Spo0F esponse egula o once 2. In oduc ion 26 assembly o he coa a ound he o espo e. GFP-SpoIVA has been shown o su ound he o espo e in a shell-like s uc u e ha is belie ed o se e as a basemen laye o he coa ( an Ooij and Losick, 2003; Ramamu hi e al., 2006). Gene ic, biochemical and cy ological e idence indica es ha his mu ual dependence is media ed in pa by con ac be ween an amino acid side-chain loca ed nea he ex eme C- e minus o SpoIVA and an amino acid side-chain on he hyd ophilic ace o he SpoVM helix. SpoVM se ing as a memb ane ancho , i adhe es o he ou e o espo e memb ane ia he hyd ophobic ace o he helix i s and hen e he s SpoIVA (Ramamu hi e al., 2006). 2.3. Cons uc ion o a cold-inducible exp ession sys em o ecombinan p o eins in B. sub ilis One o he majo d awbacks du ing high-le el p oduc ion o ecombinan p o eins in bac e ia is he inabili y o many p o eins o each hei na i e con o ma ion. Unde condi ions o o e p oduc ion, p o eins end o accumula e wi hin e ac ile agg ega es designa ed inclusion bodies (Mogk e al., 2002). Se e al s a egies ha e been desc ibed o educe he o ma ion o inclusion bodies including cul i a ion o he cells a low empe a u es (Thomas and Baneyx, 1996). Besides educing o ma ion o inclusion bodies, low- empe a u e exp ession lowe s he deg ada ion o p o eoly ically sensi i e p o eins (Eme ick e al., 1984;Chesshy e and Hipkiss, 1989). To ensu e high le el p oduc ion o ecombinan p o eins a low empe a u e, wo di e en s a egies can be used: (i) Fusion o he coding egion o he p o ein o in e es o an inducible p omo e ollowed by g ow h a a low empe a u e, e.g. 20°C. (ii) Fusion o he gene o in e es o a cold-inducible p omo e , g ow h o he exp ession s ain a he physiological empe a u e i s ollowed by induc ion a he app op ia e low empe a u e. Such cold-inducible exp ession sys ems ha e al eady been de eloped o E. coli (Mujacic e al., 1999;Qing e al., 2004). He e, he p omo e egion o he cold-inducible cspA ( o cold-shock p o eins A) gene has been used. This gene is exp essed a all empe a u es, bu he ansc ip is ex emely uns able a physiological empe a u es and g ea ly s abilized a e a empe a u e downshi o 20°C (Fang e al., 1997). A sudden dec ease in empe a u e a ec s memb ane luidi y, and o es o e i s luidi y B. sub ilis cells inc ease he le el o a memb ane-bound desa u ase (Aguila e al., 1998). This enzyme (called Δ5-Des) is encoded by he des gene and ca alyzes he in oduc ion o a cis double bond a he Δ5 posi ion o a wide a ie y o a y acids (Aguila e al., 1998). While he des ansc ip is ba ely de ec able a 37°C, i s syn hesis is ansien ly induced upon a empe a u e downshi (Aguila e al., 1999). Exp ession o he des gene does no depend on de no o p o ein syn hesis, bu on a wo-componen signal ansduc ion sys em which consis s o he senso kinase DesK and he esponse 2. In oduc ion 27 egula o DesR (Aguila e al., 2001). I is assumed ha he ansmemb ane domain o he kinase senses a empe a u e downshi h ough changes in he physical s a e o he cy oplasmic memb ane (Hunge e al., 2004). The C- e minal kinase domain o DesK unde goes au ophospho yla ion, and he phospho yl g oup is hen ans e ed o he esponse egula o DesR. Phospho yla ed DesR binds o wo adjacen DNA-binding si es leading o he ec ui men o RNA polyme ase o he des p omo e and ac i a ion o ansc ip ion (Cybulski e al., 2004). The Δ5-desa u ase di ec ly in oduces double bonds in o memb ane lipids leading o a e u n o he o iginal luidi y o he memb ane. This is sensed by DesK which changes om a kinase o phospha ase ac i i y leading o a dephospho yla ion o DesR wi h a concomi an u n o o he des gene (Mansilla and De Mendoza, 2005). Based on hese da a, a cold-inducible exp ession sys em o B. sub ilis was de eloped. 3. Resul s and Discussion 28 3. Resul s and Discussion Pa A: Func ion o F sH du ing ini ia ion o spo ula ion 3.1. In luence o F sH on he syn hesis and ac i a ion o he mas e egula o Spo0A 3.1.1. Syn hesis o Spo0A is pa ly impai ed in he sH knockou Based on p e ious s udies, sH inse ion mu an s we e comple ely de icien in spo ula ion whe e he spo ula ion equency o he mu an was less han 10-8 as compa ed o he wild- ype s ain (Deue ling e al., 1997). The s age o he spo ula ion p og am, which was impai ed in he sH mu an s had been de e mined based on ansc ip ional usions o he lacZ epo e o genes exp essed du ing wo di e en spo ula ion s ages: spoIIA-lacZ and spoIID-lacZ ep esen ing s age 0 and s age II, espec i ely. Bo h ansc ip ional usions we e induced du ing he ansi ion phase in he wild- ype s ain, bu hei induc ion was comple ely abolished in he sH mu an s (Deue ling e al., 1997). These esul s clea ly demons a ed ha sH is equi ed a an ea ly s age o he spo ula ion p ocess. Spo0A is a mas e egula o o he ini ia ion o spo ula ion (E ing on, 1993). He e, I asked whe he sH in e e es wi h he syn hesis o ac i i y o Spo0A. An sH null mu an , in which he sH gene was comple ely eplaced by an e m-casse e (Weh l e al., 2000) was used h oughou his s udy. The amoun o Spo0A was measu ed by Wes e n blo ing (Fig. 1, Pa A). In he wild- ype s ain, Spo0A s a ed o be p esen om s age 0 on and con inued o be p oduced o a leas s age 3. On he con a y, Spo0A was p esen in g ea ly educed amoun s in he sH knockou . The spo0A gene is ansc ibed om he wo p omo e s P ( o ege a i e) and Ps ( o s a iona y). Du ing exponen ial g ow h, he ansc ip ion o spo0A is unde he σA- dependen P p omo e and occu s a a low le el. Unde spo ula ion condi ions, so a unknown me abolic signals igge he phospho yla ion o Spo0A h ough he phospho elay. The phospho yla ed Spo0A (Spo0A~P) di ec ly ac i a es he Ps p omo e h ough binding o a Spo0A box loca ed adjacen o his p omo e which is ecognized by σH. I indi ec ly s imula es i s own ansc ip ion by inhibi ing he syn hesis o Ab B, which is a ep esso o he gene coding o σH (Hoch, 1991). Fu he mo e, kinA (gene coding o he senso kinase A o he phospho elay) and spo0A belong o he ca ego y o low- h eshold ac i a ed Spo0A genes, and bo h genes a e ansc ibed by he σH-con aining RNA polyme ase holoenzyme (P edich e al., 1992; Fuji a and Sadaie, 1998). Low- 3. Resul s and Discussion 29 h eshold ac i a ion o kinA and spo0A is likely o be media ed by he posi i e eedback loop as indica ed abo e (see In oduc ion - 2.2.3) in ol ing syn hesis and ac i a ion o Spo0A. Thus, he ansc ip ion o spo0A du ing s a iona y phase is s imula ed by a ce ain amoun o phospho yla ed Spo0A (Spo0A~P), and in u n, he phospho yla ion o Spo0A is accele a ed h ough he Spo0A au o-s imula o y loop. 3.1.2. Ac i a ion o Spo0A does no occu in he sH knockou The e is likely o be a leas a pa ial co ela ion be ween he le el o Spo0A and he le el o i s phospho yla ion as Spo0A is pa o a posi i e eedback loop, in which he esponse egula o di ec ly and indi ec ly s imula es he exp ession o genes in ol ed i s phospho yla ion (Hoch, 1991). We asked whe he he small amoun o Spo0A in he sH knockou is p esen in i s ac i e o inac i e o m. Recen ly, i was epo ed ha many o he genes o he Spo0A egulon espond o he ansc ip ion ac o in a dose-dependen manne . Fou di e en ca ego ies o esponses o ac i e Spo0A we e dis inguished: (i) genes ha equi e a high le el o Spo0A~P o be ac i a ed (e.g. spoIIA), (ii) hose ha equi e a high le el o Spo0A~P o be ep essed (e.g. apA), (iii) hose ha a e ac i a ed a a low le el o he egula o (e.g. sk ), and (i ) hose ha a e ep essed a a low dose o he egula o (e.g. ab B) (Fuji a e al., 2005). Indeed, I could show p og essi e inc eases in he le el o Spo0A~P in he wild- ype s ain leading o he ansc ip ion o sk -lacZ and spoIIA-lacZ om he onse o he ansi ion phase on de ined as 0. While u he accumula ion o ab B-lacZ was success ully ep essed in he wild- ype s ain, he sH knockou ailed o ac i a e ansc ip ion o sk -lacZ and spoIIA-lacZ o o ep ess exp ession o ab B-lacZ (Fig. 8). These esul s ag ee wi h p e ious indings ha exp ession o spoIIE-lacZ was p e en ed in he sH::spc mu an (Lysenko e al., 1997) and sugges ha Spo0A is p esen in i s inac i e o m in he sH mu an . The absence o su icien amoun s o ac i e Spo0A a e esponsible o he ailu e o spo ula e. I can be hypo hesized ha ei he Spo0A is no phospho yla ed a all o immedia ely dephospho yla ed a e phospho yla ion. The e o e, i has o be assumed ha F sH has o deg ade one o mo e p o eins which ac , di ec ly o indi ec ly, as an nega i e egula o s o he syn hesis o /and ac i a ion o Spo0A. 3. Resul s and Discussion 30 Fig.8. T ansc ip ion om he Spo0A- con olled p omo e s a a ious le el o Spo0A in spo ula ing g ow h condi ion. β-gala osidase ac i i y was measu ed in s ains con aining he usions in eg a ed a amyE locus. Symbols (■) he wild- ype sH s ain, (Ο) he sH::e m s ain Be a-galac osidase ac i i y [uni s] 0 5 10 15 20 25 -2 -1 0 1 2 3 4 spoIIA -lacZ -2 -1 0 1 2 3 4 Time [h] 0 10 20 30 -2 -1 0 1 2 3 4 ab B-lacZ -2 -1 0 1 2 3 4 Time [h] 0 20 40 60 80 100 120 -1 0 1 2 3 4 sk -lacZ -1 0 1 2 3 4 3.1.3. A e a i icial induc ion o ac i e Spo0A, cells a e able o spo ula e in he absence o F sH To examine whe he F sH is essen ial o spo ula ion only because o i s a ec on he syn hesis o /and phospho yla ion o Spo0A, he wild- ype spo0A allele was eplaced by an IPTG-inducible a ian o Spo0A (spo0A-sad67D56N), which is ac i e in he comple e absence o phospho yla ion (I e on e al., 1993). The spo ula ion equencies upon induc ion o he cons i u i ely ac i e o m o Spo0A in he p esence o absence o sH we e compa able and eached 37.2% and 33.5%, espec i ely (Table 2, Pa A), while he spo ula ion equency was low in he absence o IPTG. These esul s s ongly sugges ha sH is only essen ial du ing s age 0 o spo ula ion. 3.2. The le el o Spo0A is es o ed in he sH spo0E knockou , bu no in he sH ap knockou Since F sH has been clea ly shown o be in ol ed in he syn hesis and ac i a ion o Spo0A, i can be assumed ha F sH can in luence one o he di e en componen s o he phospho elay, he eby p e en ing phospho yla ion o Spo0A. I ocused on a se ies o an agonis s o he phospho elay, each o which esponds o a pa icula en i onmen al 3. Resul s and Discussion 31 signal he eby al e ing he a e o accumula ion o Spo0A~P. O pa icula ly in e es a e he phospho elay-associa ed phospha ases Rap (RapA, RapB and RapE) and he Spo0E amily (Spo0E, YisI and YnzD), which dephospho yla e Spo0F~P and Spo0A~P, espec i ely. They we e i s examined whe he hey a e in ol ed in he p oduc ion o ac i e Spo0A in he p esence and absence o F sH. Eigh s ains we e analysed o he p oduc ion o Spo0A by Wes e n blo ing. While in he absence o apA, apB and spo0E he amoun o Spo0A was inc eased a 0 as compa ed o he wild- ype si ua ion, i s amoun was educed a bo h s age 0 and s age 1 in he apE knockou (Fig. 1, Pa A). When he sH null allele was added, he amoun o Spo0A d opped in all s ains ca ying ap dis up an alleles as al eady obse ed o he wild- ype s ain in he absence o sH, bu no in he case o he Δspo0E. In he Δspo0E Δ sH s ain, he amoun o Spo0A was educed a s age 0 and u he inc eased o le els compa able o hose p esen in he sH+ s ain. This esul sugges s an in e ac ion be ween F sH and Spo0E, ei he di ec ly o indi ec ly, he eby in luencing on he exp ession o spo0A. 3.3.The ac i a ion o Spo0A in he phospho elay phospha ase dele ion s ains The ansc ip ional sk -lacZ usion, which esponds o a low h eshold o Spo0A∼P o ac i a ion, was used as a epo e sys em o s udy he ac i i y o Spo0A in s ains wi h mu an ap and spo0E alleles. While he exp ession s a ed in all s ains a s age 0 and eached i s pla eau alue a s age 2 in he wild- ype and in he Δ apB s ain, i accele a ed in he Δspo0E ea lie and u he inc eased in he Δ apA and he Δ apE s ains (Fig. 2, Pa A). Howe e , in he absence sH, he ailu e o ac i a e he sk p omo e was obse ed in all s ains. In conclusion, he absence o ap o spo0E alleles in an sH knockou leads o an exp ession o he spo0A gene, bu he p o ein emained inac i e. These da a indica e ha sH in luences p oduc ion o ac i e Spo0A by ei he allowing i s phospho yla ion o p e en ing i s apid dephospho yla ion. 3.4. Spo ula ion equency inc easing in he phospho elay phospha ase dele ion s ains In o de o assess he ela i e con ibu ion o each Rap and o he Spo0E phospha ase in modula ing he ac i i y o Spo0A, he spo ula ion e iciency in each phospha ase mu an in he p esence o absence o F sH was measu ed. The spo ula ion equencies in all ap and spo0E knockou s we e highe han ha o he wild- ype s ain and anged om 67% o 75%, whe e he spo ula ion equency o wild- ype cells was de e mined o be 58%. A simila obse a ion has been published o RapA, RapE (Jiang e al., 2000a) and Spo0E (Pe ego and Hoch, 1991). I a null allele in any one o hem 3. Resul s and Discussion 32 was combined wi h an sH knockou , he spo ula ion equency was inc eased by abou wo o h ee o de s o magni ude as compa ed o a single sH mu an , bu always emained below 1% (Table 1, Pa A). The spo ula ion equency in he phospha ase and sH double knockou s was only pa ly es o ed and s ill abou 100- old lowe han he wild- ype le el. I can be concluded ha sH is somehow in ol ed in he syn hesis o ac i e Spo0A. This esul ag ees wi h he p e ious s udies ha dele ion o he ap genes o spo0E gene esul s in inc eased spo ula ion equency (Jiang e al., 2000a; Pe ego, 2001), while o e -p oduc ion o hese genes esul s in inhibi ion o spo ula ion. I is likely ha he phospha ases ac as nega i e egula o s o he de elopmen al p ocess and espond o in- and ex e nal signals o in luence he amoun o ac i e Spo0A. Membe s o he Rap amily o phospha ases a e known o be di e en ially ac i a ed by physiological p ocesses al e na i e o spo ula ion, e.g. compe ence de elopmen induces RapA and RapE, while ege a i e g ow h condi ions induce RapB (Pe ego e al., 1994; Jiang e al., 2000). Since ege a i e g ow h and compe ence a e p ocesses ha canno occu in a spo ula ing cell, he induc ion o ap phospha ases p e en s spo ula ion om in e e ing wi h hese p ocesses. T ansc ip ion o he Spo0E phospha ase is a ec ed by signals ha a e s ill unknown (Pe ego and Hoch, 1991; Ohlsen e al., 1994), bu i is induced a he end o he exponen ial g ow h phase as ep ession by Ab B is elie ed owing o accumula ion o Spo0A P (S auch e al., 1989; Pe ego, 2001). Appa en ly, cells ha ini ia e spo0A au o- ac i a ion also induce he Spo0E le els. This co-exp ession sugges s ha Spo0E se es as some kind o a ‘sa e y lid’ o p e en o e -s imula ion o spo0A au oac i a ion. 3.5. Spo0E is a a ge o F sH in i o The obse a ion ha F sH in e e es wi h he syn hesis and phospho yla ion o Spo0A esul ed in he hypo hesis ha F sH has o deg ade one o mo e nega i e egula o s ei he p e en ing he phospho yla ion o Spo0A o being in ol ed in i s apid dephospho yla ion. Since Spo0E di ec ly a ge s Spo0A∼P and ep esses spo ula ion when o e p oduced and inc eases spo ula ion as i s absence, one possibili y o explain hese da a is a di ec in e ac ion be ween F sH and Spo0E esul ing in i s deg ada ion. To es his hypo hesis, F sH was pu i ied wi h a GST- ag as epo ed be o e (Ko schwa e al., 2005). This pu i ica ion ag keeps he p o ein soluble in he absence o any added de e gen . Fu he mo e, GST has a low he modynamic s abili y, which is e icien ly deg aded by F sH (Okuno e al., 2003). The C- e minus o GST con ains an alpha-helix as a las s uc u ed elemen (Anduja -Sanchez e al., 2005), which is ollowed by an uns uc u ed space ; his ag was expec ed o acili a e exposu e o used 3. Resul s and Discussion 33 polypep ides. Indeed, GST is gene ally s able in E. coli bu when he p o eoly ic Ss A- ag is added, he usion p o ein is apidly deg aded (Okuno e al., 2004). F om ou da a, GST-F sH is s able in B. sub ilis (da a no shown). In he in i o deg ada ion assay wi h bo h pu i ied GST-F sH and GST-Spo0E, i could be shown ha Spo0E was deg aded by F sH (Fig 3, Pa A). Since no an ibodies agains Spo0E we e a ailable, an ibodies agains GST we e used o con i m he ins abili y o Spo0E. As men ioned abo e, wo homologs o Spo0E, YisI and YnzD, a e also able o dephospho yla e Spo0A in i o (Pe ego, 2001). These wo phospha ases a e dis inguished om Spo0E by wo cha ac e is ics: Fi s , hei genes a e exp essed du ing he ege a i e g ow h phase and second, hey lack a C- e minal ex ension o abou 25 amino acid esidues. I could be shown ha bo h GST-YisI and GST-YnzD emained s able upon incuba ion wi h F sH up o a leas 5 h (Fig. 4, Pa A). These da a clea ly demons a ed ha nei he YisI no YnzD a e subs a es o F sH. They u he sugges ha he C- e minal ex ension o Spo0E ende s his phospha ase uns able. Does F sH ully deg ade Spo0E in all cells o does i modula e i s s eady-s a e le el? Based on my da a, I would like o sugges ha F sH egula es he s eady-s a e le el o Spo0E a he han comple ely deg ading i . Al e na i ely, hough less likely, i migh ully deg ade Spo0E is some cells and no a ack i a all in o he s. The i s assump ion is based on wo obse a ions:(i) a spo0E knockou leads o an inc ease in he spo ula ion equency, which ne e eaches 100%; (ii) he cellula amoun o Spo0E is low as he p o ein is ba ely de ec able in cellula ex ac s (da a no shown). A simila obse a ion has been published o LpxC o E. coli, whe e only a ew hund ed molecules pe cell a e p esen (Füh e e al., 2006). He e, a igh con ol o he amoun o his enzyme by F sH is essen ial o p e en he accumula ion o abno mal memb anes in he pe iplasm (Ogu a e al., 1999) leading o cell dea h (Sulli an and Donachie, 1984). This inding u he indica es ha F sH has o deg ade o egula e he s eady-s a e le el o one o mo e p o eins in e e ing nega i ely wi h success ul spo ula ion. 3.6. The C- e minal end o Spo0E is esponsible o deg ada ion by F sH A ema kable ea u e o he F sH p o ease is i s speci ici y o a ge ecogni ion. To ini ia e cy oplasmic p o ein deg ada ion, F sH ecognizes a ail, no mally loca ed a C- e minus, which con ains a c i ical signal o deg ada ion ini ia ion. The published obse a ion ha he wo Spo0E homologs, he p o eins YisI and YnzD, lack abou 25 amino acid esidues a hei C- e minus and ha he wo unca ed e sions o Spo0E, Spo0E94 and Spo0E11, a e s ill able o dephospho yla e Spo0A∼P s ongly sugges ha 3. Resul s and Discussion 34 he C- e minus o Spo0E may ha e an inhibi o y ole a he han being equi ed o i s enzyma ic ac i i y. Spo0E11 and Spo0E94 we e agged wi h GST, o e p oduced in E. coli and pu i ied. When hese wo pu i ied p o eins we e incuba ed wi h F sH, bo h emained s able o a leas 5 h, while β-casein as a con ol was deg aded unde hese condi ions (Fig. 4, Pa A). Since he ull-leng h Spo0E p o ein is uns able when incuba ed wi h F sH, I in e ha he 25 C- e minal amino acid esidues is esponsible o his ins abili y. 3.7. The 25 C- e minal amino acids o Spo0E is conside ed as he p o eoly ic ag o F sH-media ed deg ada ion Since Spo0E se es as a a ge o F sH, bu wo i s homologs YisI and YnzD no , I used he coding egion o he Spo0E C- e minal 25 amino acids o ynzD designa ed YnzD-0E. The GST- agged hyb id p o ein was o e p oduced in E. coli, pu i ied by a ini y ch oma og aphy and incuba ed wi h GST-F sH. As can be seen om Fig. 6, Pa A (lane 6), he YnzD-0E is la gely deg aded o e ime. In conclusion, he C- e minal 25 amino acids o Spo0E con ain he esidues ecognized by F sH and i was conside ed as p o eoly ic ag o B. sub ilis F sH-media ed deg ada ion. Simila obse a ions we e made wi h wo di e en E. coli F sH subs a es, he phage λ CII p o ein, whose C- e minus is equi ed o deg ada ion by F sH (Kobile e al., 2002), and he Ss A-deg ada ion- ag (Keile e al., 1996), in which he λ CI p o ein has been con e ed in o an F sH subs a e by a achmen o he Ss A- ag o i s C- e minus (He man e al., 1998). Which amino acids a e ecognized by he F sH p o ease? I has been sugges ed ha he F sH p o ease ecognizes lexible ail speci ici y o e en he leng h o he a ge p o ein (He man e al., 1998; Füh e e al., 2007). The ail-speci ic ecogni ion is physiologically ele an , and is used o deg ading he λ CII ac i a o , Ss A- agged p o eins (He man e al., 1998; Kobile e al., 2002) and LpxC, he key enzyme in lipopolysacha ide o ma ion by con olling he a io be ween LPS and phospholipids (So ensen e al., 1996). These sequences a e en iched in non-pola amino acids a hei e y C- e minus. Compa ison o he C- e mini om LpxC and λ CII e ealed no simila i ies. In con as , he Ss A- ag exhibi s a ema kable simila i y o he inal ele en esidues o LpxC. The exac LpxC deg ada ion- ag was de e mined by mu a ional analysis. Six non-pola amino acids wi hin he C- e minal ele en esidues o LpxC u ned ou o be equi ed o deg ada ion (Füh e e al., 2007). F om se e al s udies wi h F sH and i s euka yo ic homologs, i was concluded ha he p o ease p e e s hyd ophobic and non-pola esidues a clea age si es whe eas acidic esidues abolish deg ada ion (I o and Akiyama, 2005; Koppen and Lange , 2007). In he case o Spo0E, he e is no simila i y o he B. sub ilis Ss A- ag (Wiege and Schumann, 2001). The e o e, he amino 3. Resul s and Discussion 35 acid sequence ecognized by he F sH p o ease is di e en om ha o he Ss A- ag. Expe imen s a e in p og ess o iden i y he amino acid esidues o Spo0E ecognized by F sH. Pa B: Func ion o F sH la e du ing spo ula ion 3.1. Assump ion ha F sH in e ac s wi h he small spo ula ion pep ide SpoVM spoVM is a de elopmen al gene essen ial o spo ula ion. Some spoVM mu an s a es spo ula ion a s age IV-V and allow he o ma ion o he o espo e bu impai syn hesis and assembly o he spo e co ex (Le in e al., 1993). SpoVM was conside ed as a mo phogene ic p o ein since i is syn hesized in he mo he cell compa men and almos quan i a i ely localized o he engul ing memb ane (Le in e al., 1993; an Ooij and Losick, 2003). Fu he mo e, spo e mo phogenesis is dependen on he p ope localiza ion o SpoVM ( an Ooij and Losick, 2003). A ansposon inse ion wi hin spoVM leading o spo ula ion-de icien cells was used o selec o ex agenic supp esso s; such ex agenic supp esso s we e mapped wi hin sH (Cu ing e al., 1997). Fu he mo e, i could be shown ha chemically syn hesized SpoVM was able o inhibi deg ada ion o σ32 by pu i ied E. coli F sH (Cu ing e al., 1997). I could show in an in i o deg ada ion assay wi h pu i ied B. sub ilis F sH ha β- casein, an uns uc u ed p o ein s ongly deg aded by F sH, u ned ou o be s abilized in he p esence o SpoVM (Fig. 1, Pa B). These indings s ongly sugges ha , i s , F sH and SpoVM in e ac unc ionally and ha , second, SpoVM inhibi s he B. sub ilis F sH p o ease la e du ing spo ula ion. This assump ion is sus ained by wo obse a ions: Fi s , F sH-GFP has been shown o accumula e wi hin he asymme ic sep um (Weh l e al., 2000) and, second, SpoVM-GFP colocalized wi h he pola sep um, oo ( an Ooij and Losick, 2003). I can be in e ed ha , in he absence o SpoVM, F sH will deg ade a leas one p o ein essen ial o comple e success ul spo ula ion o , al e na i ely, egula e he s eady-s a e le el o SpoVM; bo h possibili ies a e no mu ually exclusi e. In some deg ada ion assays, he pa ial disappea ance o SpoVM was obse ed (da a no shown) as desc ibed o he bac e iophage λ CIII pep ide, which is known o inhibi λ CII p o ein deg ada ion by F sH, bu being uns able when F sH is o e p oduced (He man e al., 1997). A sho domain ( esidues 16-37) o CIII may o m an amphipa hic α-helix which is essen ial o i s ac i i y (Ko ni ze e al., 1991). In e es ingly, SpoVM was also p edic ed o o m such an amphipa hic α-helix, hough i displays no sequence simila i y wi h λ CIII (P ajapa i, e al., 2000; Ramamu hi e al., 2006). We in e om hese da a ha he essen ial SpoVM 4. Re e ences 42 Fleischmann,R.D., Adams,M.D., Whi e,O., Clay on,R.A., Ki kness,E.F., Ke la age,A.R. e al. (1995) Whole-genome andom sequencing and assembly o Haemophilus in luenzae Rd. Science 269: 496-512. Flynn,J.M., Le chenko,I., Seidel,M., Wickne ,S.H., Saue ,R.T., and Bake ,T.A. (2001) O e lapping ecogni ion de e minan s wi hin he ss A deg ada ion ag allow modula ion o p o eolysis. P oc Na l Acad Sci USA 98: 10584-10589. F ase ,C.M., Gocayne,J.D., Whi e,O., Adams,M.D., Clay on,R.A., Fleischmann,R.D. e al. (1995) The minimal gene complemen o Mycoplasma geni alium. Science 270: 397-403. Füh e ,F., Langklo z,S., and Na be haus,F. (2006) The C- e minal end o LpxC is equi ed o deg ada ion by he F sH p o ease. Mol Mic obiol 59: 1025-1036. Füh e ,F., Mülle ,A., Baumann,H, Langklo z,S., Ku sche ,B., and Na be haus,F. (2007) Sequence and leng h ecogni ion o he C- e minal u no e elemen o LpxC, a soluble subs a e o he memb ane-bound F sH p o ease. J Mol Biol 372: 485-496 Fuji a,M., Amemu a,A., and A amaki,H. (1998) T ansc ip ion o he g oESL ope on in Pseudomonas ae uginosa PAO1. FEMS Mic obiol Le 163: 237-242. Fuji a,M., and Losick,R. (2003) The mas e egula o o en y in o spo ula ion in Bacillus sub ilis becomes a cell-speci ic ansc ip ion ac o a e asymme ic di ision. Genes De 17: 1166-1174. Fuji a,M., and Sadaie,Y. (1998) Feedback loops in ol ing Spo0A and Ab B in in i o ansc ip ion o he genes in ol ed in he ini ia ion o spo ula ion in Bacillus sub ilis. J Biochem 124: 98-104. Fuji a,M., Gonzalez-Pas o ,J.E., and Losick,R. (2005) High- and low- h eshold genes in he Spo0A egulon o Bacillus sub ilis. J Bac e iol 187: 1357-1368. Geisle ,U., and Schumann,W. (1993) Isola ion o s ess mu an s o Bacillus sub ilis by a no el gene ic me hod. FEMS Mic obiol Le 108: 251-254. Gonzalez-Pas o ,J.E., Hobbs,E.C., and Losick,R. (2003) Cannibalism by spo ula ing bac e ia. Science 301: 510-513. Go esman,S., Roche,E., Zhou,Y.N., and Saue ,R.T. (1998) The ClpXP and ClpAP p o eases deg ade p o eins wi h ca boxy- e minal pep ide ails added by he Ss A- agging sys em. Genes De 12: 1338-1347. G i i h,K.L., Shah,I.M., and Wol ,R.E. (2004) P o eoly ic deg ada ion o Esche ichia coli ansc ip ion ac i a o s SoxS and Ma A as he mechanism o e e sing he 4. Re e ences 43 induc ion o he supe oxide (SoxRS) and mul iple an ibio ic esis ance (Ma ) egulons. Mol Mic obiol 51: 1801-1816. G oll,M., Boch le ,M., B ands e e ,H., Clausen,T., and Hube .R. (2005) Molecula machines o p o ein deg ada ion. Chembiochem 6: 222-256. G ossman,A.D. (1995) Gene ic ne wo ks con olling he ini ia ion o spo ula ion and he de elopmen o gene ic compe ence in Bacillus sub ilis. Annu Re Gene 29: 477- 508. Guex,N., and Pei sch,M.C. (1997) SWISS-MODEL and he Swiss-PdbViewe : an en i onmen o compa a i e p o ein modeling. Elec opho esis 18: 2714-2723. Halde ,S., Da a,A.B., and Pa ack,P. (2007) P obing he an ip o ease ac i i y o λCIII, an inhibi o o he Esche ichia coli me allop o ease H lB (F sH). J Bac e iol 189: 8130-8138. Hamon,M.A., and Lazazze a,B.A. (2001) The spo ula ion ansc ip ion ac o Spo0A is equi ed o bio ilm de elopmen in Bacillus sub ilis. Mol Mic obiol 42: 1199-1209. Has y,J., P adines,J., Dolnik,M., and Collins.J.J. (2000) Noise-based swi ches and ampli ie s o gene exp ession. PNAS 9: 2075-2080. He man,C., and D'A i,R. (1998) P o eolysis and chape ones: The des uc ion/ econs uc ion dilemma. Cu Opin Mic obiol 1: 204-209. He man,C., Ogu a,T., Tomoyasu,T., Hi aga,S., Akiyama,Y., I o,K. e al. (1993) Cell g ow h and lambda phage de elopmen con olled by he same essen ial Esche ichia coli gene, sH/h lB. P oc Na l Acad Sci USA 90: 10861-10865. He man,C., P akash,S., Lu,C.Z., Ma ouschek,A., and G oss,C.A. (2003) Lack o a obus un oldase ac i i y con e s a unique le el o subs a e speci ici y o he uni e sal AAA p o ease F sH. Mol Cell 11: 659-669. He man,C., Thé ene ,D., Bouloc,P., Walke ,G.C., and D'A i,R. (1998) Deg ada ion o ca boxy- e minal- agged cy oplasmic p o eins by he Esche ichia coli p o ease H lB (F sH). Genes De 12: 1348-1355. He man,C., Thé ene ,D., D'A i,R., and Bouloc,P. (1995) Deg ada ion o σ32, he hea shock egula o in Esche ichia coli, is go e ned by H lB. P oc Na l Acad Sci USA 92: 3516-3520. He man,C., Thé ene ,D., D'A i,R., and Bouloc,P. (1997) The H lB p o ease o Esche ichia coli deg ades i s inhibi o λCIII. J Bac e iol 179: 358-363. 4. Re e ences 44 Hilbe ,D.W., and Piggo ,P.J. (2004) Compa men aliza ion o gene exp ession du ing Bacillus sub ilis spo e o ma ion. Mic obiol Mol Biol Re 68: 234-262. Hoskins,J.R., Kim,S.Y., and Wickne ,S. (2000) Subs a e ecogni ion by he ClpA chape one componen o ClpAP p o ease. J Biol Chem 275: 35361-35367. Hoy ,M.A., Knigh ,D.M., Das,A., Mille ,H.I., and Echols,H. (1982) Con ol o phage λ de elopmen by s abili y and syn hesis o cII p o ein: Role o he i al cIII and hos h lA, himA, and himD genes. Cell 31: 565-573. Hunge ,K., Becke ing,C.L., and Ma ahiel,M.A. (2004) Gene ic e idence o he empe a u e-sensing abili y o he memb ane domain o he Bacillus sub ilis his idine kinase DesK. FEMS Mic obiol Le 230: 41-46. I e on,K., Rudne ,D.Z., Si anosian,K.J., and G ossman,A.D. (1993) In eg a ion o mul iple de elopmen al signals in Bacillus sub ilis h ough he Spo0A ansc ip ion ac o . Genes De 7: 283-294. Isaacs,F.J., Has y,J., Can o ,C.R., and Collins,J.J. (2003) P edic ion and measu emen o an au o egula o y gene ic module. P oc Na l Acad Sci USA 100: 7714-7719. I o,K., and Akiyama,Y. (2005) Cellula unc ions, mechanism o ac ion, and egula ion o sH p o ease. Annu Re Mic obiol 59: 211-231. Ja osch,E., Taxis,C., Volkwein,C., Bo dallo,J., Finley,D., Wol ,D.H., and Somme ,T. (2002) P o ein disloca ion om he ER equi es polyubiqui ina ion and he AAAATPase Cdc48. Na Cell Biol 4:134-139. Jiang,M., G au,R., and Pe ego,M. (2000a) Di e en ial p ocessing o p opep ide inhibi o s o Rap phospha ases in Bacillus sub ilis. J Bac e iol 182: 303-310. Jiang,M., Shao,W., Pe ego,M., and Hoch,J.A. (2000b) Mul iple his idine kinases egula e en y in o s a iona y phase and spo ula ion in Bacillus sub ilis. Mol Mic obiol 38: 535-542. Kanemo i,M., Nishiha a,K., Yanagi,H., and Yu a,T. (1997) Syne gis ic oles o Hs1VU and o he ATP-dependen p o eases in con olling in i o u no e o σ32 and abno mal p o eins in Esche ichia coli. J Bac e iol 179: 7219-7225. Ka zai,A.W., Roche,E.D., and Saue ,R.T. (2000) The Ss A-SmpB sys em o p o ein agging, di ec ed deg ada ion and ibosome escue. Na S uc Biol 7: 449-455. Keile ,K.C., and Saue ,R.T. (1996) Sequence de e minan s o C- e minal subs a e ecogni ion by he Tsp p o ease. J Biol Chem 271: 2589-2593. 4. Re e ences 45 Kennis on,J.A., Bake ,T.A., Fe nandez,J.M., and Saue ,R.T. (2003) Linkage be ween ATP consump ion and mechanical un olding du ing he p o ein p ocessing eac ions o an AAA+ deg ada ion machine. Cell 114: 511-520. Kiha a,A., Akiyama,Y., and I o,K. (1995) F sH is equi ed o p o eoly ic elimina ion o uncomplexed o ms o SecY, an essen ial p o ein anslocase subuni . P oc Na l Acad Sci USA 92: 4532-4536. Kiha a,A., Akiyama,Y., and I o,K. (1997) Hos egula ion o lysogenic decision in bac e iophage lambda: T ansmemb ane modula ion o F sH (H lB), he cII deg ading p o ease, by H lKC (H lA). P oc Na l Acad Sci USA 94: 5544-5549. Kiha a,A., Akiyama,Y., and I o,K. (1998) Di e en pa hways o p o ein deg ada ion by he F sH/H lKC memb ane-embedded p o ease complex: An implica ion om he in e e ence by a mu an o m o a new subs a e p o ein, YccA. J Mol Biol 279: 175-188. Kiha a,A., Akiyama,Y., and I o,K. (1999) Disloca ion o memb ane p o eins in F sH- media ed p o eolysis. EMBO J 18: 2970-2981. Kim,Y.I., Bu on,R.E., Bu on,B.M., Saue ,R.T., and Bake .T.A. (2000) Dynamics o subs a e dena u a ion and ansloca ion by he ClpXP deg ada ion machine. Mol Cell 5: 639-648. Kim,K.I., Cheong,G.W., Pa k,S.C., Ha,J.S., Woo,K.M., Choi,S.J., and Chung,C.H. (2000) Hep ame ic ing s uc u e o he hea -shock p o ein ClpB, a p o ein- ac i a ed ATPase in Esche ichia coli. J Mol Biol 303: 655-666. Kobile ,O., Koby,S., Te ,D., Cou ,D., and Oppenheim,A.B. (2002) The phage lambda CII ansc ip ional ac i a o ca ies a C- e minal domain signaling o apid p o eolysis. P oc Na l Acad Sci USA 99: 14964-14969. Ko a ,B., Mo l,H., and Keck,W. (1991) Penicillin-binding p o ein 4 o Esche ichia coli: molecula cloning o he dacB gene, con olled o e exp ession, and al e a ions in mu ein composi ion. Mol Mic obiol 5: 675-684. Ko ni ze ,D., Al u ia,S., and Oppenheim,A.B. (1991) The ac i i y o he CIII egula o o lambdoid bac e iophages esides wi hin a 24-amino acid p o ein domain. P oc Na l Acad Sci USA 88: 5217-5221. Ko schwa ,M., Ha s,E., Ohanjan,T., and Schumann,W. (2005) Cons uc ion and analyses o mu an sH alleles o Bacillus sub ilis in ol ing he ATPase- and Zn- binding domains. Cu Mic obiol 49: 180-185. 4. Re e ences 46 K zywda,S., B zozowski,A.M., Ve ma,C., Ka a a,K., Ogu a,T., and Wilkinson,A.J. (2002) The c ys al s uc u e o he AAA domain o he ATP-dependen p o ease F sH o Esche ichia coli a 1.5 Å esolu ion. S uc u e 10: 1073-1083. Le in,M.E., Hend ix,R.W., and Casjens,S.R. (1993a) A p og ammed ansla ional ameshi is equi ed o he syn hesis o a bac e iophage lambda ail assembly p o ein. J Mol Biol 234: 124-139. Le in,P.A., and Losick,R. (1996) T ansc ip ion ac o Spo0A swi ches he localiza ion o he cell di ision p o ein F sZ om a medial o a bipola pa e n in Bacillus sub ilis. Genes De 10: 478-488. Le in,P.A., Fan,N., Ricca,E., D iks,A., Losick,R., and Cu ing,S. (1993b) An unusually small gene equi ed o spo ula ion by Bacillus sub ilis. Mol Mic obiol 9: 761-771. Lysenko,E., Ogu a,T., and Cu ing,S.M. (1997) Cha ac e iza ion o he sH gene o Bacillus sub ilis. Mic obiology 143: 971-978. Ob is ,M., Milek,S., Klauck,E., Hengge,R., and Na be haus,F. (2007) Region 2.1 o he Esche ichia coli hea -shock sigma ac o RpoH (sigma32) is necessa y bu no su icien o deg ada ion by he F sH p o ease. Mic obiology 153: 2560-2571 Makino,S., Qu,J.N., Uemo i,K., Ichikawa,H., Ogu a,T., and Ma suzawa,H. (1997) A silen mu a ion in he sH gene o Esche ichia coli ha a ec s F sH p o ein p oduc ion and colicin ole ance. Mol Gen Gene 254: 578-583. Mansilla,M.C., and De Mendoza,D. (2005) The Bacillus sub ilis desa u ase: a model o unde s and phospholipid modi ica ion and empe a u e sensing. A ch Mic obiol 183: 229-235. Michaels,M.L., C uz,C., G ollman,A.P., and Mille ,J.H. (1992) E idence ha Mu Y and Mu M combine o p e en mu a ions by an oxida i ely damaged o m o guanine in DNA. P oc Na l Acad Sci USA 89: 7022-7025. Mogk,A., Maye ,M.P., and Deue ling,E. (2002) Mechanisms o p o ein olding: molecula chape ones and hei applica ion in bio echnology. Chembiochem 3: 807-814. Molle,V., Fuji a,M., Jensen,S.T., Eichenbe ge ,P., Gonzalez-Pas o ,J.E., Liu,J.S., and Losick,R. (2003) The Spo0A egulon o Bacillus sub ilis. Mol Mic obiol 50: 1683-1701. Muelle ,J.P., and Sonenshein,A.L. (1992) Role o he Bacillus sub ilis gsiA gene in egula ion o ea ly spo ula ion gene exp ession. J Bac e iol 174: 4374-4383. 4. Re e ences 47 Muelle ,J.P., Bukusoglu,G., and Sonenshein,A.L. (1992) T ansc ip ional egula ion o Bacillus sub ilis glucose s a a ion-inducible genes: Con ol o gsiA by he ComP- ComA signal ansduc ion sys em. J Bac e iol 174: 4361-4373. Mujacic,M., Coope ,K.W., and Baneyx,F. (1999) Cold-inducible cloning ec o s o low- empe a u e p o ein exp ession in Esche ichia coli: applica ion o he p oduc ion o a oxic and p o eoly ically sensi i e usion p o ein. Gene 238: 325-332. Nilsson,D., Lau idsen,A.A., Tomoyasu,T., and Ogu a,T. (1994) A Lac ococcus lac is gene encodes a memb ane p o ein wi h pu a i e ATPase ac i i y ha is homologous o he essen ial Esche ichia coli sH gene p oduc . Mic obiology 140: 2601-2610. Niwa,H., Tsuchiya,D., Makyio,H., Yoshida,M., and Mo ikawa,K. (2002) Hexame ic ing s uc u e o he ATPase domain o he memb ane-in eg a ed me allop o ease F sH om The mus he mophilus HB8. S uc u e 10: 1415-1423. Ogu a,M., Hi ao,S., Ohshi o,Y., and Tanaka,T. (1999a) Posi i e egula ion o Bacillus sub ilis sigD by C- e minal unca ed LacR a ansla ional le el. FEBS Le 457: 112-116. Ogu a,M., Liu,L., Lacelle,M., Nakano,M., and Zube ,P. (1999b) Mu a ional analysis o ComS: e idence o he in e ac ion o ComS and MecA in he egula ion o compe ence de elopmen in Bacillus sub ilis. Mol Mic obiol 32: 799-812. Ogu a,T., and Wilkinson,A.J. (2001) AAA+ supe amily ATPases: common s uc u e- di e se unc ion. Genes o Cells 6: 575-597. Ohlsen,K.L., G imsley,J.K., and Hoch,J.A. (1994) Deac i a ion o he spo ula ion ansc ip ion ac o Spo0A by he Spo0E p o ein phospha ase. P oc Na l Acad Sci USA 91: 1756-1760. Okuno,T., Yamada-Inagawa,T., Ka a a,K., Yamanaka,K., and Ogu a,T. (2004) Spec ome ic analysis o deg ada ion o a physiological subs a e σ32 by Esche ichia coli AAA p o ease F sH. J S uc Biol 146: 148-154. Okuno,T., Yamanaka,K., and Ogu a,T. (2006) An AAA p o ease F sH can ini ia e p o eolysis om in e nal si es o a model subs a e, apo- la odoxin. Genes Cells 11: 261-268. Pa ke ,G.F., Daniel,R.A., and E ing on,J. (1996) Timing and gene ic egula ion o commi men o spo ula ion in Bacillus sub ilis. Mic obiology 142: 3445-3452. Pe ego,M. and HochJ.A. (2002) Two-componen sys ems, phospho elays and egula ion o hei ac i i ies by phospha ases, pp.473-481. In A.L. Sonenshein, e . 4. Re e ences 48 al. (ed.) Bacillus sub ilis and I s Closes Rela i es: F om Genes o Cells. ASM P ess, Washing on, D.C. Pe ego,M. (2001) A new amily o aspa yl phospha e phospha ases a ge ing he spo ula ion ansc ip ion ac o Spo0A o Bacillus sub ilis. Mol Mic obiol 42: 133- 143. Pe ego,M., and Hoch,J.A. (1991) Nega i e egula ion o Bacillus sub ilis spo ula ion by he spo0E gene p oduc . J Bac e iol 173: 2514-2520. Pe ego,M., and Hoch,J.A. (1996) Cell-cell communica ion egula es he e ec s o p o ein aspa a e phospha ases on he phospho elay con olling de elopmen in Bacillus sub ilis. P oc Na l Acad Sci USA 93: 1549-1553. Pe ego,M., Hans ein,C., Welsh,K.M., Dja akhish ili,T., Glase ,P., and Hoch,J.A. (1994) Mul iple p o ein-aspa a e phospha ases p o ide a mechanism o he in eg a ion o di e se signals in he con ol o de elopmen in B. sub ilis. Cell 79: 1047-1055. Pe ego,M., Spiegelman,G.B., and Hoch,J.A. (1988) S uc u e o he gene o he ansi ion s a e egula o , ab B: egula o syn hesis is con olled by he spo0A spo ula ion gene in Bacillus sub ilis. Mol Mic obiol 2: 689-699 Piggo ,P.J., and Losick,R. (2002) Spo ula ion genes and in e compa men al egula ion. In Bacillus sub ilis and i s Closes Rela i es: om Genes o Cells. Sonenshein, A.L., Hoch, J.A., and Losick, R. (eds). Washing on, DC: Ame ican Socie y o Mic obiology P ess: 483-517. Piggo ,P.J., and Hilbe ,D.W. (2004) Spo ula ion o Bacillus sub ilis. Cu Opin Mic obiol 7: 579-586. Pogliano,J., Sha p,M.D, and Pogliano,K. (2002) Pa i ioning o ch omosomal DNA du ing es ablishmen o cellula asymme y in Bacillus sub ilis. J Bac e iol 184: 1743-1749. P ajapa i,R.S., Ogu a,T., and Cu ing,S.M. (2000) S uc u al and unc ional s udies on an F sH inhibi o om Bacillus sub ilis. Biochim Biophys Ac a Gen Subj 1475: 353-359. P akash, S. and Ma ouschek,A. (2004) P o ein un olding in he cell. T ends Biochem Sci 29: 593-600. P edich,M., Nai ,G., and Smi h,I. (1992) Bacillus sub ilis ea ly spo ula ion genes kinA, spo0F, and spo0A a e ansc ibed by he RNA polyme ase con aining σH. J Bac e iol 174: 2771-2778. 4. Re e ences 49 Qing,G.L., Ma,L.C., Kho chid,A., Swapna,G.V.T., Mal,T.K., Takayama,M.M. e al. (2004) Cold-shock induced high-yield p o ein p oduc ion in Esche ichia coli. Na Bio echnol 22: 877-882. Qu,J.N., Makino,S., Adachi,H., Koyama,Y., Akiyama,Y., I o,K. e al. (1996) The olZ gene o Esche ichia coli is iden i ied as he sH gene. J Bac e iol 178: 3457- 3461. Ramamu hi,K.S., Clapham,K.R., and Losick,R. (2006) Pep ide ancho ing spo e coa assembly o he ou e o espo e memb ane in Bacillus sub ilis. Mol Mic obiol 62: 1547-1557. Robe son,J.B., Goch ,M., Ma ahiel,M.A., and Zube ,P. (1989) Ab B, a egula o o gene exp ession in Bacillus sub ilis, in e ac s wi h he ansc ip ion ini ia ion egions o a spo ula ion gene and an an ibio ic biosyn hesis gene. P oc Na l Acad Sci USA 86: 8457-8461. San os,D., and Almeida,D.F. (1975) Isola ion and cha ac e iza ion o a new empe a u e-sensi i e cell di ision mu an o Esche ichia coli K-12. J Bac e iol 124: 1502-1507. Saue ,R.T., Bolon,D.N., Bu on,B.M., Bu on,R.E., Flynn,J.M., G an ,R.A. e al. (2004) Sculp ing he p o eome wi h AAA plus p o eases and disassembly machines. Cell 119: 9-18. Sho land,Y., Koby,S., Te ,D., Mansu ,N., O en,D.A., Ta ema su,K. e al. (1997) P o eolysis o he phage lambda CII egula o y p o ein by F sH (H lB) o Esche ichia coli. Mol Mic obiol 24: 1303-1310. Sho land,Y., Shi in,A., Zi ,T., Te ,D., Koby,S., Kobile ,O., and Oppenheim,A.B. (2000a) P o eolysis o bac e iophage lambda CII by Esche ichia coli F sH (H lB). J Bac e iol 182: 3111-3116. Sho land,Y., Te ,D., Koby,S., Kobile ,O., and Oppenheim,A.B. (2000b) Cha ac e iza ion o a conse ed λ-helical, coiled-coil mo i a he C- e minal domain o he ATP-dependen F sH (H IB) p o ease o Esche ichia coli. J Mol Biol 299: 953-964. Sil aggi,J.M., Pe kins,J.B., and Losick,R. (2006) Genes o small, noncoding RNAs unde spo ula ion con ol in Bacillus sub ilis. J Bac e iol 188: 532-541. Smi s,W.K., Kuipe s,O.P., and Veening,J.W. (2006) Pheno ypic a ia ion in bac e ia: he ole o eedback egula ion. Na Re Mic obiol 4: 259-271. 4. Re e ences 50 Sonenshein,A.L. (2000) Con ol o spo ula ion ini ia ion in Bacillus sub ilis. Cu Opin Mic obiol 3: 561-566. So ensen,P.G., Lu kenhaus,J., Young,K., E eland,S.S., Ande son,M.S., and Rae z,C.R. (1996) Regula ion o UDP-3-O-[R-3-hyd oxymy is oyl]-N- ace ylglucosamine deace ylase in Esche ichia coli. The second enzyma ic s ep o lipid a biosyn hesis. J Biol Chem 271: 25898-25905. Sowell,M.O., and Buchanan,C.E. (1983) Changes in he penicillin binding p o eins du ing spo ula ion o Bacillus sub ilis. J Bac e iol 153: 1331-1337. S ephenson,K., and Hoch,J.A. (2001) PAS-A domain o phospho elay senso kinase A: A ca aly ic ATP-binding domain in ol ed in he ini ia ion o de elopmen in Bacillus sub ilis. P oc Na l Acad Sci USA 98: 15251-15256. S i ling,C.J. and Lo d,J.M. (2006) Quali y con ol: linking e o ansloca ion and deg ada ion. Cu Biol 16: 1035-1047. S agie ,P., and Losick,R. (1996) Molecula gene ics o spo ula ion in Bacillus sub ilis. Annu Re Gene 30: 297-341. S auch,M.A., and Hoch,J.A. (1993a) Signal ansduc ion in Bacillus sub ilis spo ula ion. Cu Opin Gene De 3: 203-212. S auch,M.A., and Hoch,J.A. (1993) T ansi ion-s a e egula o s: sen inals o Bacillus sub ilis pos -exponen ial gene exp ession. Mol Mic obiol 7: 337-342. S auch,M.A., Spiegelman,G.B., Pe ego,M., Johnson,W.C., Bu bulys,D., and Hoch,J.A. (1989) The ansi ion s a e ansc ip ion egula o ab B o Bacillus sub ilis is a DNA binding p o ein. EMBO J 8: 1615-1621. S auch,M.A., T ach,K.A., and Hoch,J.A. (1992) Spo0A ac i a es and ep esses i s own syn hesis by binding a i s dual p omo e s. Biochimie 74: 619-626. S auch,M.A., Wu,J.-J., Jonas,R.H., and Hoch,J.A. (1993) A posi i e loop con ols ansc ip ion o he spo0F gene, a componen o he spo ula ion phospho elay in Bacillus sub ilis. Mol Mic obiol 7: 967-974. Sulli an,N.F., and Donachie,W.D. (1984) T ansc ip ional o ganiza ion wi hin an Esche ichia coli cell di ision gene clus e : di ec ion o ansc ip ion o he cell sepa a ion gene en A. J Bac e iol 160: 724-732. Te ,D., Koby,S., Sho land,Y., Ogu a,T., and Oppenheim,A.B. (2000) A colicin- ole an Esche ichia coli mu an ha con e s H l pheno ype ca ies wo mu a ions in he egion coding o he C- e minal domain o F sH (H lB). FEMS Mic obiol Le 183: 115-117. 4. Re e ences 51 Thomas,J.G., and Baneyx,F. (1996) P o ein mis olding and inclusion body o ma ion in ecombinan Esche ichia coli cells o e exp essing hea -shock p o eins. J Biol Chem 271: 11141-11147. Tomoyasu,T., A sène,F., Ogu a,T., and Bukau,B. (2001) The C e minus o σ32 is no essen ial o deg ada ion by F sH. J Bac e iol 183: 5911-5917. Tomoyasu,T., Game ,J., Bukau,B., Kanemo i,M., Mo i,H., Ru man,A.J. e al. (1995) Esche ichia coli F sH is a memb ane-bound, ATP-dependen p o ease which deg ades he hea -shock ansc ip ion ac o σ32. EMBO J 14: 2551-2560. Tomoyasu,T., Yamanaka,K., Mu a a,K., Suzaki,T., Bouloc,P., Ka o,A. e al. (1993a) Topology and subcellula localiza ion o F sH p o ein in Esche ichia coli. J Bac e iol 175: 1352-1357. Tomoyasu,T., Yu a,T., Mo imu a,S., Mo i,H., Yamanaka,K., Niki,H. e al. (1993b) The Esche ichia coli F sH p o ein is a p oka yo ic membe o a p o ein amily o pu a i e ATPases in ol ed in memb ane unc ions, cell cycle con ol, and gene exp ession. J Bac e iol 175: 1344-1351. Van Ooij,C., and Losick,R. (2003) Subcellula localiza ion o a small spo ula ion p o ein in Bacillus sub ilis. J Bac e iol 185: 1391-1398. Veening,J.-W., Hamoen,L.W., and Kuipe s,O.P. (2005) Phospha ases modula e he bis able spo ula ion gene exp ession pa e n in Bacillus sub ilis. Mol Mic obiol 56: 1481-1494. Wang,C.C., and Tsou,C.L. (1998) Enzymes as chape ones and chape ones as enzymes. FEBS Le 425: 382-384. Webe -Ban,E.U., Reid,B.G., Mi anke ,A.D., and Ho wich,A.L. (1999) Global un olding o a subs a e p o ein by he Hsp100 chape one ClpA. Na u e 401: 90-93. Weh l,W., Niede weis,M., and Schumann,W. (2000) The F sH p o ein accumula es a he sep um o Bacillus sub ilis du ing cell di ision and spo ula ion. J Bac e iol 182: 3870-3873. Whi ing on,D.A., Rusche,K.M., Shin,H., Fie ke,C.A., and Ch is ianson,D.W. (2003) C ys al s uc u e o LpxC, a zinc-dependen deace ylase essen ial o endo oxin biosyn hesis. P oc. Na l. Acad. Sci. USA 100: 8146-8150. Wiege ,T., and Schumann,W. (2001) Ss A-media ed agging in Bacillus sub ilis. J Bac e iol 183: 3885-3889. Woodman,P.G. (2003) A p o ein coping wi h mul iple iden i ies. J Cell Sci 116: 4283- 4290. p o ein is egula ed by phospho yla ion h ough he phospho elay signal ansduc ion sys em (Bu bulys e al., 1991;Hoch, 1993). The ans e o he phospha e o Spo0A in ol es a complex ne wo k consis ing o se e al kinases (KinA, KinB, KinC, KinD and KinE), whe e each p obably esponds o a di e en s imulus (Jiang e al., 2000b). Upon au ophospho yla ion, he phospha e is ans e ed by wo in e media es, Spo0F and Spo0B and inally o Spo0A (Bu bulys e al., 1991). The phospho ans e eac ions o he phosphop o eins a e subjec o egula ion by phospha ases, whe e one g oup, he Rap phospha ases, a e egula ed by pen apep ides (Pe ego, 1998). The Rap phospha ases speci ically dephospho yla e Spo0A~P, while ano he h ee phospha ases (Spo0E, YisI and YnzD) a ack Spo0A~P. The sH gene coding o a memb ane-ancho ed me allop o ease is p esen in mos i no all bac e ial species (Schumann, 1999;Ogu a and Wilkinson, 2001). The F sH p o ein and i s biochemical and biological unc ions ha e been s udied in de ail in E. coli. I ca ies wo ansmemb ane segmen s close o i s N- e minal end which ancho his p o ein in o he cy oplasmic memb ane in such a way ha bo h i s sho N- and i s long C- e minus a e exposed in o he cy oplasm (Tomoyasu e al., 1993a). The C- e minal pa con ains a Walke A and B box, in ol ed in binding and hyd olysis o ATP (Tomoyasu e al., 1993b), and a binding si e o Zn2+. The Aqui ex aeolicus F sH p o ein de oid o i s ansmemb ane segmen s has been c ys allized and shown o o m a ing- like hexame ic s uc u e (Suno e al., 2006). While he sH gene in E. coli is essen ial (Ogu a e al., 1999), a B. sub ilis sH knockou is iable, bu displays a pleio opic pheno ype (Deue ling e al., 1997). Cells wi h an sH null allele a e sensi i e o hea - and osmo ic s ess, g ow la gely as ilamen s and las , bu no leas , exhibi a signi ican ly educed spo ula ion equency. He e, we s a ed o analyze he ole o sH du ing spo ula ion. So a , we could show ha sH in e e es wi h he syn hesis o /and phospho yla ion o Spo0A. Based on his esul we hypo hesize ha he F sH p o ease has o deg ade one o mo e p o eins in ol ed, di ec ly o indi ec ly, in he p oduc ion o a su icien amoun o ac i e Spo0A. To his end, we ha e iden i ied he Spo0E phospha ase as one o he a ge s o F sH, and ou obse a ions indica e ha he C- e minus o Spo0E is necessa y o deg ada ion. Fu he mo e, we show ha F sH is needed only du ing s age 0. Resul s In he absence o he F sH me allop o ease only small amoun s o inac i e Spo0A a e p esen du ing he onse o spo ula ion Based on he analysis o ansc ip ional usions, we concluded ha sH in e e es wi h he syn hesis o ac i i y o Spo0A (Deue ling e al., 1997). Nex , we a emp ed o 2 iden i y he gene(s) esponsible o his e ec . We i s measu ed he spo ula ion equencies in he wild- ype and he sH knockou s ain (Table 1). While abou 59% o he cells in ou wild- ype s ain we e able o o m hea - esis an spo es, he spo ula ion equency d opped by i e o de s o magni ude in he absence o he sH allele con i ming ea lie da a (Deue ling e al., 1997) and u he unde lining he impo ance o he me allop o ease o he spo ula ion p ocess. Nex , we measu ed he amoun o Spo0A in bo h s ains by Wes e n blo ing. As can be seen om Fig. 1, Spo0A s a ed o be p esen om s age 0 on and con inued o be p oduced o a leas s age 3. On he con a y, Spo0A is p esen in g ea ly educed amoun s in his knockou (Fig. 1). A e he small amoun s o Spo0A p esen in he sH knockou ac i e ha means p esen in he phospho yla ed o m? To answe his ques ion, we cons uc ed a ansc ip ional usion be ween he p omo e o he sk ope on and he lacZ epo e gene and in eg a ed his usion ec opically a he amyE locus. I has been epo ed ha small amoun s o ac i e Spo0A (Spo0A~P) a e su icien o ac i a e he sk ope on (Fuji a e al., 2005). When his usion was analyzed in he wild- ype backg ound, he β-galac osidase ac i i y s a ed o inc ease om s age 0 on (Fig. 2A). When he same ope on usion was es ed in he sH knockou , only a e y low backg ound ac i i y was measu ed wi hou any inc ease a leas up o 3 (Fig. 2A). We conclude om hese esul s ha sH in e e es wi h he syn hesis o /and ac i a ion o Spo0A, whe e bo h a e in e wo en (S auch e al., 1992). We u he assume ha F sH has o deg ade one o mo e p o eins which ac , di ec ly o indi ec ly, as nega i e egula o s o he syn hesis o /and ac i a ion o Spo0A. The sH in e e es wi h he exp ession o ac i i y o h ee Rap phospha ases Exp ession and ac i a ion o Spo0A is embedded in a sophis ica ed ne wo k in ol ing a ple ho a o egula o s among hem h ee phospha ases e med RapA, RapB and RapE which speci ically dephospho yla e Spo0F~P, he second componen o he phospho elay (Pe ego, 1998). Fi s , we asked whe he sH in luences he spo ula ion equencies in he p esence o absence o one o he h ee phospha ases. We cons uc ed knockou s in all h ee genes as desc ibed in he Expe imen al p ocedu e sec ion. Then, hese null alleles we e combined wi h an sH knockou each, and all six s ains we e analyzed o hei spo ula ion equencies, o he amoun o Spo0A p esen and o i s ac i i y s a us. As can be seen om Table 1, he spo ula ion equencies in all h ee ap knockou s a e highe han ha o he wild- ype s ain and ange om 67% o 72%, whe e he spo ula ion equency o wild- ype cells was de e mined o be 59%. A simila obse a ion has been published o RapA and RapE (Jiang e al., 2000a). When he spo ula ion equencies in he double knockou s we e measu ed, i u ned ou o be inc eased by wo o h ee o de s o magni ude as compa ed o a single sH null mu an , 3 bu emained below 1% (Table 1). These da a clea ly indica e an in luence o he sH allele on all h ee Rap phospha ases. Nex , we analyzed all six s ains o he p oduc ion o Spo0A by Wes e n blo ing. Fig. 1 shows he esul s om a ep esen a i e assay. The absence o any o he h ee phospha ases in he o he wise wild- ype backg ound exhibi ed a di e en ou come. While in he absence o bo h apA and apB he amoun o Spo0A was inc eased a 0 as compa ed o he wild- ype si ua ion, i s amoun was educed a bo h 0 and 1 in he apE knockou (Fig. 1). When he sH null allele was added, he amoun o Spo0A d opped as al eady obse ed o he wild- ype s ain in he absence o sH (Fig. 1). Is he Spo0A p o ein p esen in he double knockou s ac i e? To answe his ques ion, he Psk -lacZ usion was in oduced in all six s ains ollowed by measu emen o he β-galac osidase ac i i ies o he s ains g own in spo ula ion medium. Exp ession o he Psk -lacZ usions is somewha di e en wi hin he six s ains. While exp ession s a ed in all s ains a 0 and eached i s pla eau alue a 2 in he wild- ype and in he Δ apB s ains, i u he inc eased in he Δ apA and he Δ apE s ains (Fig. 2). In he p esence o Δ sH, he exp ession o he ope on usion did no inc ease o e he basal le el wi h he excep ion o Δ apE whe e a sligh inc ease o abou 20 uni s was obse ed (Fig. 2D). We conclude om hese esul s ha bo h he apA and he apE genes a e in ol ed in shu ing o Psk - lacZ a 2. The sH gene in e e es wi h he phospho yla ion s a us o Spo0A h ough Spo0E Besides he Rap phospha ases, ano he se o h ee phospha es is in ol ed in he speci ic dephospho yla ion o Spo0A~P designa ed Spo0E, Yis and YnzD, whe e only he i s is ac i e du ing spo ula ion (Pe ego, 2001). While o e p oduc ion o Spo0E educed he spo ula ion equency, dele ion o spo0E esul ed in an inc ease (Pe ego and Hoch, 1991). We cons uc ed a spo0E knockou , combined i wi h he sH null allele and measu ed he spo ula ion equencies in bo h s ains. As o be expec ed he spo ula ion equency aised in he absence o he spo0E gene abo e he le el obse ed in he wild- ype s ain (Table 1). I combined wi h an sH knockou , he spo ula ion equency was inc eased 1000- old o e he le el measu ed in he Δ sH s ain, bu was s ill abou 100- old lowe han he wild- ype le el (Table 1). Nex , we analyzed o he p oduc ion o Spo0A in bo h mu an s ains. The Wes e n-blo analysis e ealed ha Spo0A is p esen in la ge amoun s al eady a 0 in he Δspo0E s ain ollowed by no signi ican u he inc ease when cells en e ed he spo ula ion pa hway (Fig. 1). When he Δ sH allele was added, he amoun o Spo0A was educed a 0 and u he inc eased o le els compa able o hose p esen in he sH+ s ain (Fig. 1). This esul sugges s an 4 in e ac ion be ween bo h p o eins, ei he di ec ly o indi ec ly, he eby in luencing exp ession o spo0A. When we es ed o he ac i i y o Spo0A in bo h mu an s ains, i u ned ou o esul in a highe ac i a ion o he sk p omo e , bu comple ely ailed o ac i a e his p omo e in he absence o sH (Fig. 2E). In conclusion, he absence o an ac i e spo0E allele in an sH knockou leads o an exp ession o he spo0A gene, bu he p o ein emained inac i e. These da a indica e ha sH in luences p oduc ion o ac i e Spo0A by ei he allowing i s phospho yla ion o p e en ing o i s apid dephospho yla ion. Spo0E is a a ge p o ein o F sH One possibili y o explain he in e ac ion be ween F sH and Spo0E is a di ec one whe eby F sH deg ades Spo0E. To es ha possibili y, we decided o pu i y bo h p o eins and o incuba e hem unde condi ions whe e F sH is able o deg ade β-casein (Ko schwa e al., 2005). F sH was pu i ied wi h a GST- ag as epo ed be o e whe e he pu i ica ion ag keeps he p o ein soluble in he absence o any de e gen (Ko schwa e al., 2005). Since we ailed o o e p oduce Spo0E equipped wi h a His- ag (unpublished da a), we decided o add he GST- ag as well. Nex , bo h p o eins we e incuba ed in he p esence and absence o ATP. While in he absence o ATP, he GST-Spo0E emained s able du ing a 4 h incuba ion ime, i was la gely deg aded in he p esence o he nucleo ide (Fig. 3A). This could be e i ied by p obing some lanes wi h αGST (Fig. 3B). To ule ou he possibili y ha clea age occu s a o wi hin he GST ag a he han wi hin Spo0E, his ag was pu i ied and incuba ed wi h GST-F sH. I could be shown ha GST emained s able o a leas 5 h (da a no shown). As men ioned abo e, wo homologues o Spo0E, YisI and YnzD, a e also able o phospho yla e Spo0E (Pe ego, 2001). A e hese wo phospha ases also a subs a e o F sH? While β-casein was comple ely deg aded wi hin 5 h o incuba ion, bo h GST-YisI and GST-YnzD emained s able unde hese condi ions (Fig. 4). These da a clea ly demons a e ha nei he YisI no YnzD a e subs a es o F sH. I u he con i ms ha he GST- ag is no ecognized by F sH. The mu an p o eins Spo0E11 and Spo0E94 a e no deg aded by F sH The spo0E11 and spo0E94 gain-o - unc ion mu a ions encode o e ac i e phospha ases ha inhibi spo ula ion by speci ically dephospho yla ing Spo0A~P (Pe ego and Hoch, 1991;Ohlsen e al., 1994). Bo h mu a ions esul ed in a s op codon educing he leng h o he p o eins om 85 o 71 (spo0E11) and 59 amino acids (spo0E94) (Pe ego and Hoch, 1987;Ohlsen e al., 1994). To ind ou whe he hese wo sho ened e sions o he Spo0E p o ein a e s ill a a ge o F sH, bo h we e agged wi h 5 GST, o e p oduced in E. coli and pu i ied. When hese wo pu i ied p o eins we e incuba ed wi h F sH, bo h emained s able o a leas 5 h, while β-casein as a con ol was deg aded unde hese condi ions (Fig. 4). Since he ull-leng h Spo0E p o ein is uns able when incuba ed wi h F sH, we in e ha he C- e minal 25 amino acid esidues a e esponsible o his ins abili y. The C- e minal end o Spo0E con e a ge speci ici y o F sH As al eady men ioned he Spo0E phospha ase is dis inguished om he YisI and YnzD phospha ases by a C- e minal ex ension o abou 25 amino acid esidues (Pe ego, 2001). Since Spo0E se es as a a ge o F sH, bu YisI and YnzD no , we asked whe he he C- e minal ex ension o Spo0E is esponsible o ecogni ion by F sH. To answe his ques ion, we used he coding egion o he C- e minal 25 amino acids o ynzD (YnzD-0E). The GST- agged hyb id p o ein was o e p oduced in E. coli, pu i ied by a ini y ch oma og aphy and incuba ed wi h GST-F sH. As can be seen om Fig. 6 (lane 6), he YnzD-0E is la gely deg aded o e ime. We conclude om his expe imen ha indeed he C- e minal end o Spo0E con ains he ecogni ion sequence o he F sH p o ease. Does he absence o spo0E in luence exp ession o yisI o /and ynzD? The yisI and ynzD genes ha e been epo ed o be exp essed du ing he ege a i e g ow h phase while spo0E is induced a a ound 0 (Pe ego and Hoch, 1987). We asked whe he he e is a c oss alk be ween hese genes conce ning hei exp ession le el. Is he e inc eased exp ession o ei he yisI o /and ynzD in a spo0E knockou ? Fi s , we used he p omo e s o he wo genes o lacZ and in eg a ed bo h ansc ip ional usions a he amyE locus. Nex , he spo0E knockou was in oduced in o bo h s ains. Then, all ou s ains (see Table 3) we e g own in DSM, samples we e aken om 0 up o 3, and he β-galac osidase ac i i y was de e mined. While no di e ence was measu ed o he ynzD p omo e independen o he p esence o absence o he spo0E allele (Fig. 7B), he e was a sligh inc ease in he ansc ip ion o he yisI gene (Fig. 7A) in he absence o spo0E. To conclude emo al o he spo0E gene did no in luence exp ession o he wo o he genes signi ican ly excluding a c oss alk a he ansc ip ional le el. Does he spo0A-sad67 allele allow success ul spo ula ion in he sH knockou ? Se e al spo0A mu a ions ha e been isola ed and analyzed among hem hose which a e ac i e in he absence o phospho yla ion. One o hese mu a ions, spo0A- sad67D56N, ca ies an in e nal in- ame dele ion emo ing amino acids 63 h ough 81 and a poin mu a ion exchanging he aspa a e o an aspa agine (I e on e al., 1993). The 6 aspa a e esidue a posi ion 56 o Spo0A ac s as he phospho yla ion si e (Bu bulys e al., 1991) and is dispensable in he spo0A-sad67 allele (I e on e al., 1993). We asked whe he an sH knockou s ain is able o o m spo es in he p esence o he spo0A- sad67D56N allele. We measu ed he spo ula ion equencies in s ain SIK190 which ca ies he spo0A-sad67D56N allele used o an IPTG-inducible p omo e . While a spo ula ion equency o 0.06% was measu ed in he absence o IPTG (Table 2), induc ion o he mu an allele a 0 esul ed in 37% hea - esis an cells. Then, he sH:: e knockou was in oduced in o SIK190 (SIK190F) and he spo ula ion equency was de e mined. While he spo ula ion equency was low in he absence o IPTG, i was high a e IPTG-induc ion and bo h alues we e compa able o hose measu ed in he sH wild- ype s ain (Table 2). These esul s s ongly sugges ha sH is needed only du ing phase 0 and ha he spo0A-sad67D56N allele can be exp essed in he absence o sH. Measu e ac i a ion o he wo p omo e s (Ps and P ) p eceding he spo0A gene T ansc ip ion o he spo0A gene is ini ia ed a wo di e en p omo e s e med P and Ps (Fe a i e al., 1985;Kudoh e al., 1985). While P is ecognized by he housekeeping sigma ac o σA and unc ions as a low-le el p omo e o p oduce a main enance le el o he Spo0A p o ein du ing exponen ial g ow h (Yamashi a e al., 1989), he second p omo e , Ps, is ecognized by he s a iona y sigma ac o σH. This p omo e is equi ed o induc ion o he p o ein a he end o exponen ial g ow h and du ing s age 0 (Fe a i e al., 1985;Yamashi a e al., 1989) and is u he ac i a ed indi ec ly by Spo0A~P, which ep esses exp ession o ab B, a nega i e egula o o sigH (Pe ego e al., 1988). The e o e, phospho yla ion o Spo0A a he onse o spo ula ion ac i a es an au o egula o y loop leading o an inc ease in ac i a ed Spo0A. We asked whe he sH o /and spo0E in luences ansc ip ion a ei he p omo e . Bo h p omo e s we e sepa a ely used o lacZ and he ansc ip ional usions we e ec opically in eg a ed a he amyE locus. Then, ei he he sH o he spo0E knockou s o bo h we e added and he β-galac osidase ac i i ies we e de e mined in all eigh s ains. While he β- galac osidase ac i i y ini ia ed a P inc eased sligh ly up o 1 ollowed by a modes dec ease in he wild- ype s ain, i s ac i i y was educed o abou 50% in he sH knockou (Fig. 8A). While he enzyma ic ac i i y in he spo0E null mu an was compa able o ha measu ed in he wild- ype s ain, addi ion o he spo0E null allele o ha o sH esul ed in a sligh inc ease in he β-galac osidase ac i i y (Fig. 8A). In summa y, he in luence o bo h spo0E and sH on he P p omo e is mino . Nex , we measu ed he β-galac osidase ac i i y o lacZ used o he Ps p omo e . In he wild- ype backg ound, his p omo e is induced abou 7- old be ween -1 and 1 (Fig. 8B). In he absence o spo0E, i is induced abou 10- old, while an only 3- old 7 induc ion was measu ed in he sH knockou which was no inc eased in he double knockou (Fig. 8B). To conclude ansc ip ion a Ps is s ongly educed in Δ sH which is no compensa ed by Δspo0E. In o al, sH in luences only ansc ip ion a Ps mos p obably h ough he s ongly educed le el o ac i e Spo0A which is needed as pa o he au o egula o y loop. Discussion When B. sub ilis cells en e he ansi ion phase, se e al di e en gene ic p og ams a e ac i a ed including he p oduc ion o ex acellula enzymes and pep ide an ibio ics, cells become mo ile and compe en , and, as he esponse o las eso , ini ia e he p ocess o spo e o ma ion. I has been shown ha cells ei he become compe en o spo ula e, ne e bo h oge he (E ing on, 1993). I is also known ha ne e 100% o he cells spo ula e. The decision o spo ula e o no o spo ula e is dependen on he amoun o ac i e Spo0A a he end o s age 0 which akes abou 2 h. Spo ula ing cells sense a mul i ude o mos ly unknown signals including he me abolic s a e, he a e o he ch omosomes, he cell densi y and o he s, in eg a e and p ocess hese signals by he phospho elay which con ols he le el o phospho yla ed Spo0A. Cells able o syn hesize ac i e Spo0A abo e a h eshold alue ha e been called Spo0A-ON and hose which ail o do so Spo0A-OFF (Chung e al., 1994). The o ma ion o wo subpopula ions o o he wise isogenic cells is designa ed bis abili y (Smi s e al., 2006;Dubnau and Losick, 2006). Bu i has o be ques ioned whe he cells in he ansi ion phase exhibi mul i- a he han bis abili y based on he obse a ion ha so di e en gene ic p og ams a e ac i a ed. This can be es ed by double and iple labelling using ansc ip ional usions be ween p og am-speci ic p omo e s and g p and i s de i a i es (Ma golin, 2000). Se e al yea s ago, we disco e ed ha he spo ula ion equency in an sH knockou is educed by abou i e o de s o magni ude (Deue ling e al., 1997). The sH gene codes o a memb ane-ancho ed ATP-dependen me allop o ease which seems o be p esen in all bac e ial species (Schumann, 1999;Ogu a and Wilkinson, 2001). The objec i e o his ongoing esea ch p ojec is o elucida e he ole o he F sH p o ease du ing spo ula ion o B. sub ilis cells. I is based on he assump ion ha F sH has o deg ade o o egula e he s eady-s a e le el o one o mo e p o eins nega i ely in e e ing wi h he spo ula ion p og am. To his end we could show ha he ac i i y o F sH is needed only du ing phase 0 which culmina es in he Spo0A-ON s a us. Cells ca ying an sH knockou syn hesize a signi ican ly educed amoun o Spo0A which, based on gene ic da a, is inac i e. This obse a ion explains why sH null mu an s exhibi a d ama ically educed spo ula ion equency. Bu sH could also play a ole 8 du ing subsequen spo ula ion s ages. This possibili y could be uled ou by in oduc ion o an sH null allele in a s ain ca ying a mu an spo0A allele which is ac i e in he absence o phospho yla ion. Upon exp ession o he spo0A-sad67D56N allele, cells exhibi ed a no mal spo ula ion equency. The e o e, he ole o sH is exclusi ely con ined o he syn hesis o /and ac i a ion o wild- ype Spo0A. The nex ques ion o be aised is why he amoun o Spo0A is signi ican ly educed and why his educed amoun is inac i e? So a , we used he candida e app oach o iden i y pu a i e subs a e p o eins o F sH. Fi s , we cons uc ed knockou s o ou di e en phospha ases whe e i has al eady been published ha he spo ula ion equencies is sligh ly, bu ep oducibly enhanced in hei absence o wo o hem (Jiang e al., 2000a); hese esul s could be con i med and ex ended o wo addi ional phospha ases. Upon in oduc ion o an sH null allele in o hese ou mu an s ains, he spo ula ion equencies we e inc eased by wo o h ee o de s o magni ude, bu s ill emained e y low. No su p isingly, no inc ease in he amoun o ac i e Spo0A could be measu ed. We would like o conclude ha sH in e e es di ec ly o indi ec ly wi h exp ession o ac i i y o hese ou phospha ases. To explain he inc eases in he spo ula ion equencies, we u he sugges ha a ew cells wi hin he whole popula ion a e in he Spo0A-ON s a e. To in es iga e his possibili y, we will use he p omo e o he sk and he spoIIA ope ons o g p and analyze single cells unde he luo escence mic oscope. Whe eas he sk A p omo e needs a low amoun o ac i e Spo0A, he spoIIA one equi es a high amoun o become ac i a ed (Fuji a e al., 2005). M. Pe ego iden i ied wo homologs o he Spo0E phospha ase, YisI and YnzD (Pe ego, 2001). She could show ha bo h phospha ases a e able o dephospho yla e Spo0A~P in i o. These wo phospha ases a e dis inguished om Spo0E by wo cha ac e is ics: Fi s , hei genes a e exp essed du ing he ege a i e g ow h phase and second, hey lack a C- e minal ex ension o abou 25 amino acid esidues. She sugges ed ha his C- e minal ex ension could be ecognized by a p o ease. Based on he assump ion and ou inding ha an sH spo0E double knockou exhibi ed he highes spo ula ion equency o all es ed phospha ase null alleles, we asked he ques ion whe he he Spo0E p o ein is a subs a e o F sH. Bo h p o eins we e o e p oduced in E. coli wi h a GST immobiliza ion ag and incuba ed unde app op ia e condi ions. We could show ha indeed Spo0E is deg aded by F sH. When GST- agged YisI o YnzD we e incuba ed wi h F sH, hese p o eins u ned ou o be s able unde he same condi ions. In he las expe imen , we used he C- e minal ex ension o Spo0E o YnzD. This usion p o ein was shown o be uns able in he p esence o F sH. To conclude, he C- e minal 25 amino acids o Spo0E con ain he esidues ecognized by F sH. 9 Which amino acids a e ecognized by he F sH p o ease? One o he subs a e p o eins o he E. coli F sH p o ease is LpxC enzyme (Ogu a e al., 1999). This enzyme ep esen s he key enzyme in lipopolysacha ide (LPS) o ma ion and con ols he a ion be ween LPS and phospholipids (So ensen e al., 1996). Since o e p oduc ion o LpxC causes accumula ion o abno mal memb anes in he pe iplasm (Ogu a e al., 1999) leading o cell dea h (Sulli an and Donachie, 1984), he amoun o LpxC mus be ca e ully egula ed which is done by F sH. He e, he C- e minus has been iden i ied, oo, o be esponsible o being ecognized and deg aded by F sH (Füh e e al., 2006). The au ho s poin ed o he abou en amino acids being p esen a he immedia e C- e minus which esemble he Ss A- ag. In he case o Spo0E, he e is no simila i y o he B. sub ilis Ss A- ag (Wiege and Schumann, 2001). The e o e, he amino acid sequence ecognized by he F sH p o ease is di e en om ha o he Ss A- ag. Expe imen s a e in p og ess o iden i y he amino acid esidues o Spo0E ecognized by F sH. Does F sH ully deg ade Spo0E in all cells o does i modula e i s s eady-s a e le el? Base on ou da a, we would like o sugges ha F sH egula es he s eady-s a e le el o Spo0E a he han comple ely deg ading i . This assump ion is based on he obse a ion ha a knockou o spo0E leads o in inc ease in he spo ula ion equency which should no occu when Spo0E is comple ely deg aded by F sH. In conclusion, ou esul s s ongly sugges ha egula ion o s abili y o se e al p o eins in ol ed di ec ly o indi ec ly in he syn hesis o ac i e Spo0A exe s a new le el o pos ansla ional egula ion h ough he F sH p o ease. Ano he p o ease has been iden i ied yielding a compa able pheno ype. Inac i a ion o clpP esul ed in cells de icien in spo ula ion ini ia ion and in compe ence and in a highly ilamen ous mo phology (Msadek e al., 1998;Ge h e al., 1998). In such a mu an , he exp ession o spo0A and spo0H, coding o he s a iona y sigma ac o σH, was signi ican ly dec eased (Nanamiya e al., 2000). In oduc ion o a mu an spo0E allele in o he clpP knockou es o ed he exp ession o spo0A, bu no spo ula ion. Based on ou esul s, addi ional genes ha e o be iden i ied o in luence he syn hesis o ac i e Spo0A. These genes will be iden i ied wi h h ee di e en expe imen al s a egies: Fi s , he candida e s a egy; second, sa u a ed ansposon mu agenesis using pMa A (Le B e on Y. e al., 2006); and hi d, cons uc ion o an sH ap mu an (Flynn e al., 2003). Iden i ica ion o hese addi ional a ge s migh shed some ligh on he molecula mechanism o bis abili y. Expe imen al p ocedu es Bac e ial s ains, plasmids, media and g ow h condi ions 10 All s ains used in his s udy a e lis ed in Table 3. E. coli DH10B was used o plasmid cons uc ion and p opaga ion. E. coli A8926 is a de i a i e o W3110 used o he exp ession o GST- agged p o eins. The B. sub ilis s ain 1012 was used in mos o he expe imen s. All s ains we e ei he g own in Lu ia-Be ani (LB) o in Di co Spo ula ion medium (DSM). An ibio ics we e added when app op ia e a he ollowing concen a ions: ampicillin, 100 μg ml-1; chlo amphenicol, 10 μg ml-1; e y h omycin, 50 μg ml–1; neomycin, 10 μg ml–1; kanamycin, 20 μg ml –1; spec inomycin, 100 μg ml–1. Cons uc ion o plasmids and ecombinan s ains All ansc ip ional usions we e cons uc ed using he in eg a ion ec o pDG1728 (Gué ou -Fleu y e al., 1996). This ec o con ains a p omo e -less lacZ and allows inse ion o he ope on usions ec opically a he amyE locus. Th ee di e en p omo e s we e used o lacZ gene a ed by PCR using ch omosomal DNA o s ain 1012 DNA as empla e. These p omo e s a e Psk (ampli ied by p ime s ON1 and ON2; see Table 4) p eceding he sk ope on which is ac i a ed by a low amoun o ac i e Spo0A (Fuji a e al., 2005), he ege a i e and he s a iona y phase induced p omo e s P (ON3/ON4) and Ps (ON5/ON6), espec i ely, o he spo0A gene (Chibazaku a e al., 1991). While Psk was inse ed be ween he EcoRI and HindIII si es o pDG1728, P and Ps we e liga ed in o he EcoRI and BamHI si es. Knockou s in he ou genes apA, apB, apE and spo0E we e cons uc ed as ollows. Fi s , he wo lanking egions o each gene (abou 300 bp) we e ampli ied (see Table 4 o he p ime sequences) and inse ed in o pBluesc ip SKII+. Nex , he chlo amphenicol esis ance casse e was ampli ied using pDG364 as empla e and inse ed be ween he lanking egions o he h ee ap genes. In he case o he spo0E gene, a phleomycin esis ance ma ke gene a ed plasmid pBlueSKII+-phleo was liga ed be ween he wo lanking egions. In he las s eps, PCR agmen s con aining he esis ance ma ke and he lanking egions we e ans o med in o B. sub ilis 1012 ollowed by selec ion on LB pla es con aining ei he chlo amphenicol o phleomycin. Ch omosomal DNA was p epa ed om se e al ans o man s each and checked by Sou he n blo ing o eplacemen o he wild- ype alleles. One knockou mu an each was kep o u he s udies. Recombinan ec o s allowing o e exp ession and pu i ica ion o GST- agged p o eins we e p epa ed using pGEX-2T. The genes sH (ON21/22), spo0E (ON23/ON24), yisI (ON25/26) and ynzD (ON29/ON30) we e ampli ied using ch omosomal DNA o s ain 1012. The mu an s spo0E11 and spo0E94 ca y s op codons a posi ions 72 and 60, espec i ely (Pe ego and Hoch, 1987). The wo unca ed e sions we e gene a ed by ampli ica ion o he app op ia e coding egion (spo0E11: 11 Table 2. Spo ula ion equencies o s ains exp essing spo0A ac i e in he absence o phospho yla iona S ain sH IPTGb Viable Spo e % o geno ype added cell coun coun spo ula ion SIK190 + - 6.0 x 106 3.7 x 103 0.06 SIK190 + + 2.5 x 1010 9.3 x 109 37.2 SIK190F - - 4.8 x 106 2.4 x 103 0.05 SIK190F - + 1.7 x 1010 5.7 x 109 33.5 See legend o Table 1 o echnical de ails. a Rep esen a i e o h ee di e en expe imen s b IPTG was added a a inal concen a ion o 1 mM 18 Table 3. Bac e ial s ains and plasmids used in his s udy Plasmid o s ain Rele an geno ype Sou ce S ains E. coli DH10B mc A Δ(m hsdRMS mc BC) φ80d lacZM15 ΔlacX74 deoR ecA1 a aD139 Δ(a a leu)7697 galU galK psL endA1 nupG Be hesda Resea ch Labo a o ies, Inc. A8296 s hC zad-220::Tn10 Δ sH3::kan (Ta su a e al., 1998) B. sub ilis 1012 leuA8 me B5 pC2 hs M1 (Sai o e al., 1979) WW01 1012 Δ sH::e m (Weh l e al., 2000) ED04 1012 Δ sH:: e (Deue ling e al., 1997) AL31 Δ apA::ca This s udy AL32 Δ apA::ca Δ sH::e m This s udy AL33 Δ apB::ca This s udy AL34 Δ apB::ca Δ sH::e m This s udy AL35 Δ apE::ca This s udy AL36 Δ apE::ca Δ sH::e m This s udy AB07 Δspo0E::bleo A. B andl AB08 Δspo0E::bleo Δ sH::e m A. B andl AL37 amyE::Psk –lacZ spc This s udy AL38 amyE::Psk –lacZ spc Δ sH::e m This s udy AL39 amyE::Psk –lacZ spc Δ apA::ca This s udy AL40 amyE::Psk –lacZ spc Δ apA::ca Δ sH::e m This s udy AL41 amyE::Psk –lacZ spc Δ apB::ca This s udy AL42 amyE::Psk –lacZ spc Δ apB::ca Δ sH::e m This s udy AL43 amyE::Psk –lacZ spc Δ apE::ca This s udy AL44 amyE::Psk –lacZ spc Δ apE::ca Δ sH::e m This s udy AL45 amyE::Psk –lacZ spc Δspo0E::bleo This s udy AL46 amyE::Psk –lacZ spc Δspo0E::bleo Δ sH::e m This s udy 19 AL47 amyE::P (spo0A)-lacZ spc This s udy AL48 amyE::PB (spo0A)-lacZ spc Δ sH::e m This s udy AL49 amyE::PB (spo0A)-lacZ spc Δspo0E::bleo This s udy AL50 amyE::Ps(spo0A)-lacZ spc This s udy AL51 amyE::Ps(spo0A)-lacZ spc Δ sH::e m This s udy AL52 amyE::Ps(spo0A)-lacZ spc Δspo0E::bleo This s udy AL53 amyE::PyisI-lacZ spc This s udy AL54 amyE::PyisI-lacZ Δ spo0E::bleo This s udy AL55 amyE::PynzD-lacZ spc This s udy AL56 amyE::PynzD-lacZ Δ spo0E::bleo This s udy AL57 amyE::P (spo0A)-lacZ spc Δspo0E::bleo Δ sH::e m This s udy AL58 amyE::Ps(spo0A)-lacZ spc Δspo0E::bleo Δ sH::e m This s udy SIK190 amyE::(Pspac-spo0A-sad67D56N ca ), spo0A::e m, Em , Cm (I e on e al., 1993) SIK190F sH:: e in SIK190 This s udy Plasmids pDG1728 Pe mi s ansc ip ional usion o lacZ (Gué ou -Fleu y e al., 1996) p1728-Psk amyE::Psk –lacZ This s udy p1728-P amyE::P (spo0A)-lacZ This s udy p1728-Ps amyE::Ps(spo0A)-lacZ This s udy pGex-2 Exp ession ec o Pha macia pGST- sH P ac-GST- sH This s udy pGST-spo0E P ac-GST-spo0E This s udy pGST-yisI P ac-GST-yisI This s udy pGST-ynzD P ac-GST-ynzD This s udy pGST-spo0E94 P ac GST-spo0E94 This s udy pGST-spo0E11 P ac-GST-spo0E11 This s udy pGST-ynzD-C0E P ac-GST-ynzD-C0E This s udy pBluec ip SKII+Cloning ec o S a agene p apA-ca pBluesc ip SKII+ wi h 300 bp up- and downs eam lanking egions o apA gene and ca casse e This s udy p apB-ca pBluesc ip SKII+ wi h 300 bp up- and This s udy 20 downs eam lanking egions o apB gene and ca casse e p apE-ca pBluesc ip SKII+ wi h 300 bp up- and downs eam lanking egions o apE gene and ca casse e This s udy 21 Table 4. Oligonucleo ides used in his s udy 0ligonucleo ide p ime SequencesaDesc ip ion ON1 ggcca GAATTC acaggagac ca ca (EcoRI) 5´ sk A p omo e ON2 ggcca AAGCTTaag aaacc cc c caa (HindIII) 3´ sk A p omo e ON3 ggcca GAATTCgaaaag ga cgg gc g cac (EcoRI) 5´ spo0A ege a i e p omo e ON4 ggcca GGATCCa c c g a a accg a (BamHI) 3´ spo0A ege a i e p omo e ON5 ggcca GAATTCa cacg cc g g caaa (EcoRI) 5´ spo0A spo ula ion p omo e ON6 ggcca GGATCCg c cc ccccaaa g ag (BamHI) 3´ spo0A spo ula ion p omo e ON7 ggcca AAGCTT gagga gaagcagacga ccg (HindIII) 5´ apA ups eam ON8 ggcca GAATTC cgagaagccc g cagc g a (EcoRI) 3´ apA ups eam ON9 ggcca GGATCCcgaagcgcaaaaaaag a cg ga (BamHI) 5´ apA downs eam ON10 ggcca TCTAGAa ca a aaacaa c cc c c (XbaI) 3´ apA downs eam ON11 ggcca AAGCTTa ggccgcg acgaga cccg ca (HindIII) 5´ apB ups eam ON12 ggcca GAATTCa ac caga aa ccggaga gc (EcoRI) 3´ apB ups eam ON13 ggcca GGATCCgccgcga acgg ac a gaaaa (BamHI) 5´ apB downs eam ON14 ggcca CCGCGG ac ca a aaacaa cg cc (SacII) 3´ apB downs eam ON15 ggcca AAGCTT ga a caa caca cagc gaa (HindIII) 5´ apE ups eam ON16 ggcca GAATTCgaag a aa aa a gcccgca (EcoRI) 3´ apE ups eam ON17 ggcca GGATCCcaagcaa gga gc ccgcaaa (BamHI) 5´ apE downs eam ON18 ggcca TCTAGAga ca aca gca cccc cg (XbaI) 3´ apE downs eam ON19 ggcca GAATTCcgga cgc acgc caaa cc (EcoRI) 5´ ca casse e ON20 ggcca GGATCCa c caac aacggggcagg a (BamHI) 3´ ca casse e ON21 ggcca GGATCCaa cggg c gcg aa acc (BamHI) 5´ sH 22 ON22 ggcca AGATCT ac c cg a cg c c c (BglII) 3´ sH ON23 ggcca GGATCCggcgg c c gaacaagaaa (BamHI) 5´ spo0E ON24 ggcca CCCGGGa a a gca ca a gc ggc (SmaI) 3´ spo0E ON25 ggcca GGATCCaacag aaaa gaagaaa ga (BamHI) 5´ yisI ON26 ggcca CCCGGGa aca acgggag caaga (SmaI) 3´ yisI ON27 ggcca CCCGGGa a gggaa g cg c gca a (SmaI) 3´ 75-nucleo ides-sho en spo0E ON28 ggcca CCCGGGa accacaagcc aa c acaaggc (SmaI) 3´ 39-nucleo ides-sho en spo0E ON29 ggcca GGATCCa agagagca c a aaaag (BamHI) 5´ ynzD ON30 ggcca CCCGGG ca cacccgc ac gc cga (SmaI) 3´ ynzD ON31 ggcca GAATTCaa caaggcc g aaagaaa ag (EcoRI) 5´ C- eminus o spo0E ON32 ggcca GAATTCa a a gca ca a gc ggc (EcoRI) 3´ C- eminus o spo0E a es ic ion endonuclease si es a e shown in capi al le e s 23 Fig. 1. Amoun o Spo0A in di e en B. sub ilis s ains. The s ains we e g own in DSM a 37°C, and aliquo s we e aken a e en y in o he ansi ion phase ( 0) and up o 3 h la e ( 1 o 3). Cells we e lysed by sonica ion and equal amoun s o p o eins we e applied pe lane (5 µg). Fi s line: wild- ype 1012 and WW01 (Δ sH); second line: AL31 (Δ apA) and AL32 (Δ apA Δ sH); hi d line: AL33 (Δ apB) and AL34 (Δ apB Δ sH); ou h line: AL35 (Δ apE) and AL36 (Δ apE Δ sH); i h line: AB07 (Δspo0E) and AB08 (Δspo0E Δ sH). 24 Fig. 2. T ansc ip ion om he Spo0A~P-ac i a ed p omo e sk . Cells con aining he Psk -lacZ usion in eg a ed ec opically a he amyE locus we e g own in DSM a 37°C, and aliquo s (5 OD578 uni s) we e wi hd awn o measu emen o β-galac osidase ac i i ies. (A) AL37 ( sH+) and AL38 (Δ sH); (B) AL39 (Δ apA sH+) and AL40 (Δ apA Δ sH); (C) AL41 (Δ apB sH+) and AL42 (Δ apB Δ sH); (D) AL43 (Δ apE sH+) and AL44 (Δ apE Δ sH); (E) AL45 (Δspo0E sH+) and AL46 (Δspo0E Δ sH). ■ sH+, ○ Δ sH. 25 A B Fig. 3. Spo0E ac s as a subs a e o F sH. Pu i ied GST-F sH was incuba ed wi h GST- Spo0E unde condi ions desc ibed in he Expe imen al p ocedu es. (A) Aliquo s o he eac ion mix u es we e sepa a ed by SDS-PAGE and s ained wi h Coomassie b illian blue. (B) Wes e n blo using αGST and showing GST-Spo0E. The incuba ion ime o all h ee samples was 4 h. 26 A B Fig. 4. The phospha ases YisI and YnzD a e s able in he p esence o F sH. GST-F sH was incuba ed wi h (A) GST-YisI and (B) GST-YnzD. β-casein se ed as a con ol. Fig. 5. Two mu an Spo0E p o eins a e s able in he p esence o F sH. GST- agged Spo0E94 and Spo0E11 p o eins we e incuba ed wi h F sH. The eac ion p oduc s we e esol ed by SDS-PAGE and s ained wi h Coomassie blue. 27 Besides se ing as a memb ane ancho , a second unc ion has been sugges ed o SpoVM. When a ansposon inse ion wi hin spoVM which is biologically inac i e was used o iden i y possible in e ac ing p o eins, he memb ane-bound F sH me allop o ease was iden i ied (3). Mu a ions in sH supp essed he spo ula ion de ec o ce ain spoVM mu an s bu no o he s, ano he case o allele-speci ic ex agenic supp esso s. Fu he mo e, i could be shown ha chemically syn hesized SpoVM is able o inhibi deg ada ion o σ32 by pu i ied E. coli F sH (3). Based on hese indings, one can assume ha a second unc ion o SpoVM is o inhibi he B. sub ilis F sH p o ease la e du ing spo ula ion. This assump ion is sus ained by wo obse a ions: Fi s , F sH-GFP has been shown o accumula e wi hin he asymme ic sep um (28) and, second, SpoVM-GFP colocalizes wi h he pola sep um, oo (26). I can be in e ed ha , in he absence o SpoVM, F sH will deg ade a leas one p o ein essen ial o comple e success ul spo ula ion o , al e na i ely, egula e he s eady-s a e le el o SpoVM. The objec i e o he cu en wo k was o analyze whe he SpoVM inhibi s he p o ease ac i i y and o s udy exp ession o he spoVM gene. Pu i ied GST-F sH was incuba ed wi h o wi hou he SpoVM pep ide o up o 4 h (Fig. 1). While he band o β-casein pa ly disappea ed a e 3 h o incuba ion in he p esence o GST-F sH (lane 2), i was almos comple ely absen 4 h a e o incuba ion (lane 4). Mos in e es ingly, he β-casein was no comple ed deg aded, bu con e ed in o dis inc deg ada ion p oduc s. I he SpoVM pep ide was p esen , he β-casein u ned ou o be s abilized (lanes 3 and 5). In he absence o GST-F sH, β-casein emained s able (lane 6). In conclusion, GST-F sH is able o deg ade he subs a e p o ein β-casein in o speci ic agmen s, and his p o eoly ic ac i i y can be inhibi ed by he SpoVM pep ide. In some expe imen s, we obse ed he pa ial disappea ance o SpoVM (da a no shown) as desc ibed o he bac e iophage λ encoded CIII pep ide (10, 12), whe e a sho domain ( esidues 16-37) may o m an amphipa hic α-helix which is essen ial o CIII ac i i y (13). In e es ingly, SpoVM was also p edic ed o o m such an amphipa hic α- helix, hough i displays no sequence simila i y wi h CIII (16, 17). We in e om hese da a ha he essen ial SpoVM pep ide, by in e ac ion wi h F sH, ei he p e en s deg ada ion o one o mo e p o eins essen ial o success ul spo e o ma ion o ha F sH in luences he s eady-s a e le el o SpoVM. Bo h possibili ies a e no mu ually exclusi e. To cons uc a spoVM knockou , abou 300 bp each o i s up- and downs eam egion we e ampli ied by PCR and liga ed in o pBR322 (2). The ups eam egion was gene a ed using he p ime pai ON01 and ON02 (Table 1), he downs eam egion ON03 and ON04 and ch omosomal DNA o s ain 1012 (18) as empla e. While he amplicon ep esen ing he ups eam egion was lanked by EcoRI and SmaI si es, he 2 downs eam amplicon was lanked by SmaI and HindIII si es allowing hei inse ion in o EcoRI and HindIII clea ed pBR322 esul ing in o pMB02. In he nex s ep, a spec inomycin esis ance casse e was ampli ied using he p ime pai ON05 and ON06 and pK2-spec (9) as empla e and liga ed in o SmaI linea ized pMB02 esul ing in pMB03. Then, pMB03 was ans o med in o B. sub ilis 1012 whe e he plasmid is unable o eplica e. T ans o med cells we e pla ed on LB aga pla es con aining spec inomycin o selec o hose cells whe e he spoVM wild- ype allele has been eplaced by he knockou allele. Ch omosomal DNA om se e al candida es was p epa ed and checked by Sou he n blo ing o success ul eplacemen (da a no shown). S ain MB03 was kep o u he s udies. Nex , we de e mined he spo ula ion equency o he knockou s ain by g owing cells in DSM, a spo ula ion medium (19), a 37°C o 36 h, hea ed hem o 20 min o 80°C and hen pla ed o su i o s as desc ibed (5). Whe eas abou 82% o he cells o he wild- ype s ain 1012 we e able o spo ula e, cells o s ain MB03 u ned ou o be comple ely de icien in spo ula ion (less han 10-6). These da a a e in ag eemen wi h p e iously published esul s whe e B. sub ilis s ains wi h mu an spoVM alleles we e unable o p oduce hea - esis an spo es (3, 14). Since SpoVM in e e es wi h he p o eoly ic ac i i y o F sH, we examined he possibili y ha F sH will deg ade one o mo e p o eins essen ial o comple e spo e o ma ion. This p o ein(s) could be loca ed ei he in he ou e spo e memb ane o in he cy oplasm o he mo he cell. In a i s a emp o iden i y his p o ein(s), s ains 1012 and MB03 we e g own in DSM o 5. In one expe imen , he memb anes o he mo he cell and o he p espo e we e isola ed as desc ibed (1) and he memb ane p o eins esol ed by SDS-PAGE. A ew p o ein bands could be iden i ied which a e ei he absen o p esen in he spoVM knockou (da a no shown). Expe imen s a e in p og ess o iden i y hese p o eins in collabo a ion wi h he g oup o D . M. Hecke , Uni e si y o G ei swald. In a second expe imen , we compa ed he cy oplasmic p o eomes o s ains 1012 and MB03 using he 2D-gel elec opho esis echnique. He e, a o al o 83 p o ein spo s we e p esen in he wild- ype, bu comple ely absen o p esen in educed amoun s in he spoVM null mu an . One o hese p o eins u ned ou o be SpoIVA which seems o be comple ely absen in he spoVM knockou (Fig. 2). This aises he in e es ing ques ion whe he SpoIVA is uns able in he absence o SpoVM and which ATP-dependen p o ease is esponsible o i s deg ada ion. I SpoIVA is a subs a e o F sH, i will be in e es ing o ind ou whe he e he ing o SpoIVA o memb ane-bound SpoVM will be su icien o p e en i s deg ada ion o whe he SpoVM has o di ec ly in e ac wi h he p o ease ac ing as an an ip o ease as desc ibed o he phage λ CIII p o ein (8). 3 The spoVM gene o ms a monocis onic ope on, and i s ansc ip ion is con olled by he spo ula ion-speci ic sigma ac o σE which ac ing in conjunc ion wi h he DNA- binding p o ein SpoIIID, a 93-amino-acid p o ein (14). To ind ou when ansc ip ion o spoVM is ini ia ed, cells o s ain 1012 we e g own in DSM a 37°C in o s a iona y phase. Aliquo s we e aken immedia ely upon en e ing he s a iona y phase ( 0) and up o 5 h la e ( 1 o 5). As can be seen om Fig. 3, he spoVM ansc ip (abou 200 nucleo ides in leng h) s a ed o appea a e 2 and con inued o be p oduced in la ge quan i ies a leas un il 5. Using a spoVM-lacZ ansc ip ional usion, a simila exp ession pa e n was desc ibed (14). When o al RNA isola ed om he spoVM knockou a 5 was analyzed, no signal was ob ained as expec ed (Fig. 3). Nex , we wan ed o ind ou when he SpoVM pep ide can be de ec ed in spo ula ing B. sub ilis cells. Cells o s ain 1012 we e g own again as desc ibed be o e and aliquo s whe e p epa ed o a Wes e n blo . As shown in Fig. 4, he SpoVM pep ide s a ed o accumula e a 4 (lane 4) and u he inc eased du ing he nex 2 h (lane 5 and 6). No SpoVM pep ide could be de ec ed in he null mu an s ain a 6 (lane 7). Chemically syn hesized SpoVM se ed as a posi i e con ol (lane 8). All aliquo s we e also checked o he p esence o he hea shock p o ein H pG (20) which se ed as a con ol o a p o ein no subjec o spo ula ion egula ion (Fig. 4). In summa y, hese esul s clea ly demons a e ha while he spoVM ansc ip is p esen abou 3 h a e en y in o he ansi ion phase, syn hesis o he SpoVM pep ide is delayed by 2 h sugges ing pos ansc ip ional egula ion o spoVM (see below). The lag-phase o abou 2 h be ween he onse o spoVM ansc ip ion and ansla ion sugges s a so a unknown mechanism ha p e en s ea ly appea ance o he pep ide. This could in ol e s abiliza ion o he ansc ip , delayed ansla ion ini ia ion o s abili y o he pep ide. A close inspec ion o he spoVM ansc ip e eals an 87 nucleo ide - un ansla ed egion (5' UTR) p eceding he coding egion (Fig. 5A). We asked whe he his 5' UTR is in ol ed in exp ession o spoVM. Two di e en ansla ional usions using lacZ as a epo e gene we e cons uc ed one wi h and he o he wi hou he un ansla ed egion. In bo h cases, he 9 h codon o spoVM was used o he 8 h codon o lacZ. To disc imina e be ween cis- and ans-ac ing ac o s, we i s analyzed he wo usions in E. coli. This decision was based on he assump ion ha ans-ac ing ac o s a e comple ely absen om his hos . When bo h usions we e analyzed in E. coli, be ween 30 and 50 uni s o β-galac osidase ac i i y we e measu ed du ing he exponen ial g ow h phase wi h no di e ence be ween he wo usions (Fig. 5B). This esul indica es ha he 5' UTR does no ac as a cis-ac ing ac o in E. coli, e.g. by in luencing he s abili y o he ansc ip . 4 Nex , we in es iga ed he in luence o he 5' UTR in B. sub ilis g own in DSM. In he p esence o he un ansla ed egion he β-galac osidase ac i i y s a ed o ise a 2 and inc eased om abou 7 o 20 uni s a 5 and d opped he ea e (Fig. 5C). When he ansla ional usion wi hou he 5' UTR was analyzed, exp ession o he epo e gene s a ed a 2, oo, bu a a signi ican ly highe le el (abou 5- old) and u he inc eased a la e spo ula ion imes (Fig. 5C). In summa y, he 5' UTR o spoVM nega i ely in luences i s own ansc ip ion o ansla ion in B. sub ilis. Since his e ec was no obse ed in E. coli, we conclude ha a ans-ac i e egula o is in ol ed in ansc ip ion o ansla ion o spoVM which is no p esen in E. coli. Ye ano he possibili y is ha he 5' UTR a ec s he s abili y o he ansc ip . We ega d his mechanism as a he unlikely because such an e ec could no be obse ed in E. coli. The ac o in luencing egula ion a he 5' UTR could be a ansla ional ep esso p o ein binding wi hin he un ansla ed egion, e.g., a he in e ed epea (Fig. 5A). Al e na i ely, a non-coding (nc) RNA may in e ac wi h he spoVM ansc ip he eby educing i s ansla ion. Recen ly, se e al ncRNAs ha e been desc ibed which a e unde spo ula ion con ol (21). In conclusion, he SpoVM pep ide exe s a leas wo unc ions. Fi s , as shown by R. Losick and cowo ke s, i adhe es o he ou e o espo e memb ane ia hyd ophilic amino acid side-chains on he hyd ophobic ace o he helix (17). Then, i will ec ui he SpoIVA p o ein, a mo phogene ic p o ein ha o ms he basemen laye o he spo e coa . Whe he eally SpoVM i s adhe es o he memb ane and hen e he s SpoIVA is no clea . Al e na i ely, bo h componen s could in e ac in he cy oplasm and hen binds o he memb ane. Second, SpoVM, a leas in i o, inhibi s he p o eoly ic ac i i y o F sH, and we can assume ha i does he same in i o. Since we could show ha F sH is essen ial o cells o en e he spo ula ion p og am (5), an ea ly syn hesis o SpoVM could p e en cells go beyond s age 0. Why F sH has o be inhibi ed by SpoVM? A leas wo possibili ies can be conside ed, which a e no mu ually exclusi e. Fi s , F sH migh ine- une he amoun o SpoVM and he eby p e en accumula ion o inc eased amoun s which migh be dele e ious o he cells as has been shown o he LpxC p o ein (15). Second, by binding o F sH SpoVM may p e en deg ada ion o a leas one p o ein needed o comple e he spo ula ion p og am. Since SpoIVA is absen in a spoVM knockou as e ealed by a 2D-gel analysis, his mo phogene ic p o ein migh be a subs a e o F sH and i s in e ac ion wi h SpoVM will p o ec i om deg ada ion. We would like o hank P o . M. Hecke and his g oup o hei gene ous help wi h he 2D-gel elec opho esis and he iden i ica ion o p o ein spo s by mass spec ome y. We also app ecia e he help o Monika Ba ze wi h some o he expe imen s. This p ojec was suppo ed by he Deu sche Fo schungsgemeinscha (Schu 414/20-2). 5 Re e ences 1. Bagyan, I., M. Noback, S. B on, M. Paidhunga , and P. Se low. 1998. Cha ac e iza ion o yhcN, a new o espo e-speci ic gene o Bacillus sub ilis. Gene 212:179-188. 2. Boli a , F., R. Rod iquez, M. Be lach, and H. Boye . 1977. Cons uc ion and cha ac e iza ion o new cloning ehicles I: Ampicillin- esis an de i a i es o he plasmid pMB9. Gene 2:75-93. 3. Cu ing, S., M. Ande son, E. Lysenko, A. Page, T. Tomoyasu, K. Ta ema su, T. Ta su a, L. K oos, and T. Ogu a. 1997. SpoVM, a small p o ein essen ial o de elopmen in Bacillus sub ilis, in e ac s wi h he ATP-dependen p o ease F sH. J. Bac e iol. 179:5534-5542. 4. Cu ing, S. M. and P. B. Vande Ho n. 1990. Gene ic analysis, p. 27-60. In C. R. Ha wood and S. M. Cu ing (ed.), Molecula biological me hods o Bacillus. John Wiley & Sons, Chiches e . 5. Deue ling, E., A. Mogk, C. Rich e , M. Pu ucke , and W. Schumann. 1997. The sH gene o Bacillus sub ilis is in ol ed in majo cellula p ocesses such as spo ula ion, s ess adap a ion and sec e ion. Mol. Mic obiol. 23:921-933. 6. D iks, A. 1999. Bacillus sub ilis spo e coa . Mic obiol. Mol. Biol. Re . 63:1-20. 7. Eymann, C., A. D eisbach, D. Alb ech , J. Be nha d , D. Beche , S. Gen ne , l. T. Tam, K. Bu ne , G. Buu man, C. Scha , S. Venz, U. Volke , and M. Hecke . 2004. A comp ehensi e p o eome map o g owing Bacillus sub ilis cells. P o eomics. 4:2849-2876. 8. Halde , S., A. B. Da a, and P. Pa ack. 2007. P obing he an ip o ease ac i i y o CIII, an inhibi o o he Esche ichia coli me allop o ease H lB (F sH). J. Bac e iol. 189:8130-8138. 9. Hä l, B., W. Weh l, T. Wiege , G. Homu h, and W. Schumann. 2001. De elopmen o a new in eg a ion si e wi hin he Bacillus sub ilis ch omosome and cons uc ion o compa ible exp ession casse es. J. Bac e iol. 183:2696-2699. 10. He man, C., D. Thé ene , R. D'A i, and P. Bouloc. 1997. The H lB p o ease o Esche ichia coli deg ades i s inhibi o lambdacIII. J. Bac e iol. 179:358-363. 11. Homu h, G., S. Masuda, A. Mogk, Y. Kobayashi, and W. Schumann. 1997. The dnaK ope on o Bacillus sub ilis is hep acis onic. J. Bac e iol. 179:1153-1164. 12. Kobile , O., S. Koby, D. Te , D. Cou , and A. B. Oppenheim. 2002. The phage lambda CII ansc ip ional ac i a o ca ies a C- e minal domain signaling o apid p o eolysis. P oc. Na l. Acad. Sci. USA 99:14964-14969. 13. Ko ni ze , D., S. Al u ia, and A. B. Oppenheim. 1991. The ac i i y o he CIII egula o o lambdoid bac e iophages esides wi hin a 24-amino acid p o ein domain. P oc. Na l. Acad. Sci. USA 88:5217-5221. 6 14. Le in, P. A., N. Fan, E. Ricca, A. D iks, R. Losick, and S. Cu ing. 1993. An unusually small gene equi ed o spo ula ion by Bacillus sub ilis. Mol. Mic obiol. 9:761-771. 15. Ogu a, T., K. Inoue, T. Ta su a, T. Suzaki, K. Ka a a, K. Young, L.-H. Su, C. A. Fie ke, J. E. Jackman, C. R. H. Rea z, J. Coleman, T. Tomoyasu, and H. Ma suzawa. 1999. Balanced biosyn hesis o majo memb ane componen s h ough egula ed deg ada ion o he commi ed enzyme o lipid A biosyn hesis by he AAA p o ease F sH (H lB) in Esche ichia coli. Mol. Mic obiol. 31:833-844. 16. P ajapa i, R. S., T. Ogu a, and S. M. Cu ing. 2000. S uc u al and unc ional s udies on an F sH inhibi o om Bacillus sub ilis. Biochim. Biophys. Ac a Gen. Subj. 1475:353-359. 17. Ramamu hi, K. S., K. R. Clapham, and R. Losick. 2006. Pep ide ancho ing spo e coa assembly o he ou e o espo e memb ane in Bacillus sub ilis. Mol. Mic obiol. 62:1547-1557. 18. Sai o, H., T. Shiba a, and T. Ando. 1979. Mapping o genes de e mining nonpe missi eness and hos -speci ic es ic ion o bac e iophages in Bacillus sub ilis Ma bu g. Mol. Gen. Gene . 170:117-122. 19. Schae e , P., J. Mille , and J. P. Aube . 1965. Ca abolic ep ession o bac e ial spo ula ion. P oc. Na l. Acad. Sci. USA 54:704-711. 20. Schulz, A., S. Schwab, S. Ve s eeg, and W. Schumann. 1997. The h pG gene o Bacillus sub ilis belongs o class III hea shock genes and is unde nega i e con ol. J. Bac e iol. 10:3103-3109. 21. Sil aggi, J. M., J. B. Pe kins, and R. Losick. 2006. Genes o small, noncoding RNAs unde spo ula ion con ol in Bacillus sub ilis. J. Bac e iol. 188:532-541. 22. S agie , P. and R. Losick. 1996. Molecula gene ics o spo ula ion in Bacillus sub ilis. Annu. Re . Gene . 30:297-341. 23. Tam, T., H. An elmann, C. Eymann, D. Alb ech , J. Be nha d , and M. Hecke . 2006. P o eome signa u es o s ess and s a a ion in Bacillus sub ilis as e ealed by a 2-D gel image colo coding app oach. P o eomics. 6:4565-4585. 24. Te , D., S. Koby, Y. Sho land, T. Ogu a, and A. B. Oppenheim. 2000. A colicin- ole an Esche ichia coli mu an ha con e s H l pheno ype ca ies wo mu a ions in he egion coding o he C- e minal domain o F sH (H lB). FEMS Mic obiol. Le . 183:115-117. 25. Tomoyasu, T., J. Game , B. Bukau, M. Kanemo i, H. Mo i, A. J. Ru man, A. B. Oppenheim, T. Yu a, K. Yamanaka, H. Niki, S. Hi aga, and T. Ogu a. 1995. Esche ichia coli F sH is a memb ane-bound, ATP-dependen p o ease which deg ades he hea -shock ansc ip ion ac o 32. EMBO J. 14:2551-2560. 26. Van Ooij, C. and R. Losick. 2003. Subcellula localiza ion o a small spo ula ion p o ein in Bacillus sub ilis. J. Bac e iol. 185:1391-1398. 27. Wa ens, A. N., M. D. Jones, and R. I. Lechle . 1997. Splicing by o e lap ex ension by PCR using asymme ic ampli ica ion: an imp o ed echnique o he gene a ion o hyb id p o eins o immunological in e es . Gene 186:29-35. 7 28. Weh l, W., M. Niede weis, and W. Schumann. 2000. The F sH p o ein accumula es a he sep um o Bacillus sub ilis du ing cell di ision and spo ula ion. J. Bac e iol. 182:3870-3873. 8 TABLE 1. Oligonucleo ides used P ime Sequence (5' o 3')a ON01 GGCCAT GAATTC GAGCTGATCATTTTTTAGGAAAC; EcoRI ON02 GGCCAT CCCGGG AAACGAAAAAGTACCTCGTGAAT; SmaI ON03 GGCCAT CCCGGG TTCAAAGCCCTCTTTCACCACAT; SmaI ON04 GGCCAT AAGCTT TGAAAGATGATGAAACAATAGTTGC; HindIII ON05 GGCCAT CCCGGG CGATTTGACATTTTTCTTGTG, SmaI ON06 GGCCAT CCCGGG ATCAATAGTTACAAATTCTTTCA; SmaI ON07 GGCCAT GGATCC CTGGCCGTCGTTTTACAACGT; BamHIII ON08 GGCCAT GGATCC TTATTTTTGACACCAGACCAACTGGTAAT; BamHI ON09 GGCCAT GGATCC AATATCCTCTAAATAATTGTCATAT; BamHIII ON10 GGCCAT AAGCTT CGGCAATTTAATGGTGTAAAATTT; HindIII ON11 AAAGCCATATTAATAATGATAAGTATAGGAGGGGACAAAAATG ON12 CTTATCATTATTAATATGCCTTTT a G/C clamps a e shown in bold ace; es ic ion si es a e unde lined 9 FIG. 1. The SpoVM pep ide e a ds deg ada ion o β-casein by he F sH p o ease. The SpoVM pep ide has been p epa ed by chemical syn hesis (KLH; Pep ide Speciali y Labo a o ies, Heidelbe g, Ge many) and F sH has been pu i ied as a GST- agged p o ein as desc ibed (24). Incuba ion o he di e en componen s ollowed a published me hod (25). The comple e eac ion mix u e (30 µl) consis ed o he ollowing componen s: 50 mM T is-ace a e (pH 8.0), 5 mM magnesium ace a e, 12.5 µM zinc ace a e, 80 mM NaCl, 1.4 mM β- me cap oe hanol, 5 mM ATP, 100 µg/ml bo ine se um albumin (BSA), 200 µg/ml o pu i ied SpoVM pep ide, 50 µg/ml β-casein and 50 µg/ml o pu i ied B. sub ilis GST- F sH. SpoVM pep ide, β-casein, and GST-F sH we e p esen a a mola a io o 132:4:1. Reac ions we e pe o med a 40°C o he ime poin s indica ed. Aliquo s o he eac ion mix u es we e analyzed by 15% SDS-PAGE ollowed by s aining wi h Coomassie blue. 10 FIG. 2. The cy oplasmic p o eomes o B. sub ilis wild- ype (A) and ΔspoVM::spec (B) s ains. S ains 1012 and MB03 we e g own in DSM a 37°C in o s a iona y phase. Cy oplasmic p o eins we e sepa a ed by wo-dimensional (2D) gel elec opho esis using immobilized pH g adien s (IPG) in he ange 4-7 as desc ibed (23). Fo iden i ica ion o he p o eins by mass spec ome y, he 2D gels we e s ained wi h Colloidal Coomassie b illian blue (Ame sham Biosciences). Spo cu ing, yp ic diges ion o he p o eins and spo ing o he esul ing pep ides on o he MALDI a ge s we e pe o med as desc ibed (7). He e, only a small pa o he gel is shown. The whi e a ow indica es he posi ion o he SpoIVA p o ein. 11 leading o a dephospho yla ion o DesR wi h a concomi- an u n off o he des gene [13]. Based on hese da a, we de eloped a cold-inducible exp ession sys em o B. sub ilis making use o he des p o- mo e . We show he e ha cold-induc ion esul s in a signi - ican induc ion o epo e genes la gely p e en ing o ma ion o agg ega es o an agg ega ion-p one p o ein. Cold-inducible exp ession sys ems ha e also been de el- oped o Esche ichia coli which a e based on a diffe en p inciple [14,15]. Cold-inducible exp ession sys ems p o- ide an inexpensi e al e na i e echnology especially o indus ial p oduc ion o ecombinan p o eins comple- men ing he widely used IPTG- and xylose-inducible sys ems. Ma e ials and me hods Ma e ials Bac e ia, plasmids and g ow h condi ions Bac e ial s ains and plasmids used a e lis ed in Table 1. Cells we e g own in Lu ia B o h (LB) medium a 37 o 25 °C unde ae a ion. An ibio ics we e added whe e app o- p ia e (ampicillin a 100 lg/ml, neomycin a 10 lg/ml and chlo amphenicol a 10 lg/ml). Me hods Cons uc ion o a ansc ip ional usion be ween he des p omo e and he lacZ epo e gene The p omo e egion o he des gene was used o he lacZ epo e gene using he in eg a ion ec o pDG1728 [16]. The esul ing ansc ip ional usion is sandwiched be ween amyE- on and amyE-back allowing i s in eg a- ion in o he B. sub ilis ch omosome a he amyE locus. The des p omo e egion was amplified using oligonucleo- ides (ON) ON1 (GGCCATGAATTCTCCGGCATCCC GATCATCGC; es ic ion si e unde lined) and ON2 (GGCCATAAGCTTTCTCATTGTGTGTCTCGGTTC AG). The amplicon was clea ed wi h EcoRI and HindIII and inse ed in o pDG1728 cu wi h he same enzymes esul ing in pDG1728-des. This ecombinan plasmid was ans o med in o s ain WW02, and ans o man s we e selec ed on LB pla es con aining chlo amphenicol and sc eened o he loss o he neomycin esis ance ma ke , and one posi i e ans o man (AL03) was kep o u he s udies. Cons uc ion o a des null mu an To cons uc a des knockou , he gene including flanking egions was amplified using he p ime pai s ON3/ON4 (GGCCATGTCGACTGAACCGAGACACACAATG; GGCCATGAGCTCATAGTTGAGCACCTTTGG), and he amplicon was clea ed wi h SalI and SacI and cloned in o pBluesc ip SKII + ea ed wi h he same enzymes. Nex , he ecombinan plasmid was ea ed wi h HindIII and BclI o emo e a 61-bp in e nal agmen o des which was eplaced by he neo ma ke using pBgaB as empla e and he p ime pai ON5/ON6 (GGCCATAAGCTT AGGTCGAGATCAGGGAATGAGTT; GGCCATTGA TCAGATCAATTCTGACAGCCATG). Using he p ime pai ON3/ON4, he modified gene was amplified and ans- o med in o B. sub ilis 1012. Neomycin- esis an ans o - man s we e selec ed and checked by Sou he n-blo o ch omosomal eplacemen o he des by he neo gene (da a no shown). One s ain (AL02) was kep o u he s udies. Table 1 S ains and plasmids used in his s udy S ains Geno ype Re e ence/sou ce E. coli DH10B F  mc A D(m hsdRMS mc BC)u80d lacZ DM15 deoR ecA1 a aD139 D(a a leu) 7697 galU galK k  psL endA1 nupG BRL B. sub ilis 1012 leuA8 me B5 pC2 hs M1 [26] WW02 1012 amyE::neo [27] AL02 1012 des::neo This wo k AL03 1012 amyE::Pdes-lacZ This wo k AL04 AL02 des::neo amyE::Pdes-lacZ This wo k AL05 AL02 h pG::e m This wo k AL06 AL02 pbpE::e m This wo k Plasmids pBluesc ip SK + Cloning ec o S a agene pBgaB In eg a ion ec o con aining he bgaB gene [28] pDG1728 Vec o allowing in eg a ion o DNA sequences a he amyE locus [16] pHT01 De i a i e o pNDH33 wi hou a di ec epea [22] pKTH10 Recombinan ec o con aining he amyQ gene [17] pAL10 Exp ession ec o allowing cold-inducible in acellula p oduc ion o ecombinan p o eins This wo k pAL12 Exp ession ec o allowing cold-inducible sec e ion o ecombinan p o eins This wo k pNDH33-h pG h pG used o an IPTG-inducible p omo e [22] A.T. Thuy Le, W. Schumann / P o ein Exp ession and Pu ifica ion 53 (2007) 264–269 265 The Pdes–lacZ usion was in oduced in o AL02 by ans- o ma ion esul ing in AL04. Cons uc ion o he wo exp ession ec o s pAL10 and pAL12 We s a ed om he ec o pHT01, whe e lacI,Pg ac and bgaB we e emo ed by SacI/BamHI diges ion ollowed by eliga ion wi h he des p omo e egion including he binding si es o DesR gene a ed by PCR using ON7/ON8 (GGCCATGAGCTCTCCGGCATCCCGAT CATCGC; GGCCATGGATCCTCTTGATCGCCTCCT CATTGTGTGTCTCGG) and esul ing in he new exp es- sion ec o pAL10 (Fig. 1a). This ec o allows in acellu- la p oduc ion o ecombinan p o eins. A second ec o allowing sec e ion o p o eins was ob ained by using he signal sequence o he amyQ gene [17] o he des p omo e (ON9/ON10 (GGGCCCATGGATCCATGATTCAA AAACGAAAGCGGACAG; GGCCATTCTAGATTTT TCTGAACATAAATGGAGACG) and pKTH10 as em- pla e) esul ing in he exp ession-sec e ion ec o pAL12 (Fig. 1b). To es sui abili y and efficacy o he new exp es- sion ec o s, diffe en genes we e used o he des p omo e and he syn hesis was moni o ed a e empe a u e down- shock o 25 °C. pAL10 was es ed by inse ion o h pG, coding o a hea shock p o ein o unknown unc ion [18],andpbpE encoding he penicillin-binding p o ein Pbp4 * [19]. The h pG gene was amplified using ON11/12 (GGCCATGGATCCATGGCGAAAAAAGAGTTTAAA GC; GGCCATTCTAGATTACACCATGACCTTGCAA ATATTGTTCG), pbpE ON13/14 (GGCCATGGATCCA TGAAGCAGAATAAAAGAAAGC; GGCCATGGATC CTTACTACTTCGTACGGACCGCTTCT) and ch omo- somal DNA o B. sub ilis 1012 as empla e. To analyse o he e sa ili y o pAL12, he coding egion o amyQ [17] was inse ed (ON15/16 (GGCCATTCTAGAGTAA ATGGCACGCTGATGCAGT; GGCCATCCCGGGTT ATTTCTGAACATAAATGGAGACG) and pKTH10 as empla e). De e mina ion o enzyma ic ac i i ies and Wes e n blo analysis The b-galac osidase ac i i ies encoded by lacZ was de e mined as desc ibed elsewhe e [20], wi h he excep ion ha LacZ ac i i y was measu ed kine ically in a mic opla e eade (Ve saMax, Molecula De ices) a 405 nm a 28 °C. One uni was defined as DE405 min1OD1 578 103,in which OD 578 is he op ical densi y o he g ow h cul u e when samples we e d awn. Wes e n blo analyses we e ca - Fig. 1. Gene ic and es ic ion map o he wo ec o s pAL10 and pAL12 allowing in a- and ex acellula exp ession o ecombinan p o eins, espec i ely. (a) pAL10 and he DNA sequence o he Pdes p omo e (in capi al le e s); (b) pAL12 and he DNA sequence o Pdes, he ibosome-binding si e (unde lined) and he coding egion o he signal sequence (highligh ed in g ey). Unique es ic ion si es which can be used o inse ion o ecombinan genes a e also p esen ed. 266 A.T. Thuy Le, W. Schumann / P o ein Exp ession and Pu ifica ion 53 (2007) 264–269 ied ou as published [21]. Blo s we e de eloped wi h poly- clonal an ibodies agains H pG and Pbp4 * used a a dilu- ion o 1:10,000. Resul s and discussion Cons uc ion and analysis o an ope on usion be ween he des p omo e and lacZ Based on published da a [11], we de ised a cold-induc- ible exp ession sys em consis ing o he egula o y egion o he des gene consis ing o he des p omo e and i s ups eam egion se ing as binding si es o DesR P. This egion was cloned in o he in eg a ion ec o pDG1728 [16] ollowed by inse ion a he amyE locus. S ain AL03 was fi s g own a 37 °C o he ea ly exponen- ial g ow h phase. Then, he cul u e was di ided in o wo subcul u es whe e one was u he kep a 37 °C while he second was challenged wi h a cold shock o 25 °C. Aliquo s we e wi hd awn jus be o e di iding he cul u e ( = 0) and up o 12 h pos -induc ion o de e mina ion o he b-galac- osidase ac i i ies. As shown in Fig. 2, he enzyma ic ac i - i y o he unshocked cul u e s a ed wi h abou 4 uni s and inc eased o abou 15 uni s o e ime. In con as , he cold- shocked cul u e exhibi ed an inc ease o abou 105 uni s a e 1.5 h ollowed by a dec ease o 50 uni s 5 h a e induc ion (Fig. 2). The dec ease can be explained by induc- ion o he des gene om he ch omosome es o ing he flu- idi y o he memb ane ollowed by u ning off exp ession o des gene [11]. I should be possible o p e en u ning off exp ession o he des gene by dele ing his gene om he ch omosome as published [11]. This has been done as desc ibed unde Ma e ials and me hods, and he ope on usion was in e- g a ed a he amyE locus o s ain AL03 whe e he des gene has been eplaced by a neomycin esis ance ma ke . When his s ain AL04 was analysed, he b-galac osidase ac i i y was e en lowe du ing g ow h a 37 °C, while he ac i i y inc eased om 7 o abou 100 uni s wi hin he fi s 12 h a e cold challenge (Fig. 2). This esul clea ly demon- s a es ha no u n off o he lacZ exp ession occu s in he absence o he desa u ase. Based on his finding we asked whe he a u he inc ease in he enzyma ic ac i i y can be ob ained upon p olonged incuba ion. As shown in Fig. 3, while he OD 578 con inued o inc ease s eadily o a leas 58 h, he b-galac osidase ac i i y inc eased up o abou 12 h and dec eased he ea e . This esul indica es ha ei he he hal -li e o he enzyme o /and he syn hesis capaci y o he cells dec ease du ing p olonged incuba ion a 25 °C. We also measu ed he b-galac osidase ac i i y a e empe a u e downshi o 20 and 15 °C. While abou 100 uni s we e measu ed a e 5 h o g ow h a 25 °C (Fig. 2), 60 uni s and 5 uni s we e de e mined a 20 and 15 °C, espec i ely (da a no shown). This esul s sugges s ha g ow h a 20 °C educes he exp ession le el o 60%, while exp ession o he lacZ gene is comple ely abolished a 15 °C. The exp ession ec o pAL10 allows p oduc ion o ecombinan p o eins o a significan le el Nex , we a emp ed o di ec ly isualize he amoun o ecombinan p o eins p oduced. To accomplish his goal, wo diffe en genes we e used o Pdes in he exp ession ec- o pAL10, namely he h pG and he pbpE gene coding o a hea shock p o ein o unknown unc ion and a penicillin- binding p o ein, espec i ely [18,19]. Bo h s ains ( he ch o- mosomal copies o h pG and pbpE ha e been dele ed) we e g own in LB medium o he mid-exponen ial g ow h phase, di ided in o wo subcul u es whe e one was u he incu- ba ed a 37 °C, while he second was cold-shocked o 25 °C. Aliquo s we e wi hd awn a diffe en ime poin s o he analysis o he p esence o he H pG o Pbp4 * p o- ein as indica ed. While no H pG p o ein was isible when he s ain AL05 con aining he plasmid pAL10-h pG was incuba ed a 37 °C, his p o ein became appa en al eady a e 3 h and inc eased in i s amoun up o 9 h (Fig. 3a) Fig. 2. Induc ion o b-galac osidase ac i i y in wo B. sub ilis s ains g own a wo diffe en empe a u es. (a) B. sub ilis s ains AL03 and AL04 (Ddes::neo) we e g own in LB medium a 37 °C o he ea ly loga i hmic g ow h phase. Then, he cul u es we e di ided in o wo subcul u es (a = 0) whe e one was u he g own a 37 °C, while he second was challenged wi h 25 °C. Aliquo s we e emo ed o de e mina ion o b-galac osidase ac i i ies a he ime poin s indica ed. The comple e expe imen s we e epea ed h ee imes and yielded compa able esul s. Da a om one o hese expe imen s a e p esen ed. AL03 g own a 37 °C (d)o 25°C(s); AL04 g own a 37 °C(j)o 25°C(h). (b) B. sub ilis s ain AL04 was g own up o 58 h a e he empe a u e downshock. The OD 578 was measu ed du ing g ow h ( ) and he b-galac osidase ac i i ies as indica ed (whi e columns). A.T. Thuy Le, W. Schumann / P o ein Exp ession and Pu ifica ion 53 (2007) 264–269 267 whe e i ep esen ed abou 10% o he o al cellula p o ein. When he h pG gene was exp essed om an IPTG-induc- ible p omo e o 6 h a 25 °C, only iny amoun s o he H pG p o ein became isible (Fig. 3a, lane 10). We con- clude om his esul ha he exp ession le el is a he low a 25 °C om he IPTG-inducible p omo e and can be compensa ed om a cold-inducible p omo e . In pa al- lel, we isualized H pG by Wes e n blo . While small amoun s we e p esen a 37 °C mos p obably due o he leakiness o he p omo e , i inc eased d ama ically up o 9 h a e cold-shock (Fig. 3b). The Pbp4 * p o ein has been epo ed o be memb ane- a ached due o one o mo e hyd ophobic pa ches [19]. We could al eady show ha o e p oduc ion o his p o ein a 37 °C leads mainly o insoluble Pbp4 * [22]. The e o e, we wonde ed whe he o e p oduc ion a low empe a u e will influence he olding o Pbp4 * leading o mainly soluble p o ein. We analysed he amoun o Pbp4 * by Wes e n blo om h ee diffe en ac ions: o al cellula con en , soluble and insoluble ac ion ob ained a e a cen i uga ion s ep. While some Pbp4 * p o ein was p esen al eady a 37 °C incuba ion as epo ed o H pG, i s amoun inc eased sig- nifican ly 6 h a e incuba ion o he cells a 25 °C(Fig. 4). As can be seen, mos o he ecombinan p o ein s ayed sol- uble indica ing ha he lowe empe a u e a ou s o ma- ion o olded polypep ides as desc ibed o he agg ega ion-p one usion p o ein p eS2-S0-b-galac osidase in E. coli [23]. A compa able esul has been ob ained du - ing cons i u i e high le el p oduc ion o he DnaK and G oE chape one sys ems [22]. We conclude ha p oduc- ion o agg ega ion-p one ecombinan p o eins a low empe a u es is al e na i e way o la gely p e en o ma- ion o agg ega es. The exp ession-sec e ion ec o pAL12 allows egula ed sec e ion o exop o eins To es he sec e ion capabili y a low empe a u e, he amyQ gene coding o an a-amylase [17] was inse ed in o pAL12 esul ing in pAL12-amyQ. S ain AL02 ca ying pAL12-amyQ was g own in LB medium a 37 and 25 °C, and aliquo s we e aken a he ime poin s indica ed in Fig. 5. I he amoun o a-amylase p esen a 37 and 25 °C we e compa ed, significan ly mo e enzyme was p es- en a 25 °C as compa ed o 37 °C(Fig. 5). We also mea- Fig. 4. Immunoblo analysis o Pbp4 * .B. sub ilis s ain 1012 ca ying pAL10-pbpE was g own as desc ibed in he legend o Fig. 3. Cells we e lysed by sonifica ion and he cellula lysa e was applied di ec ly (T) o a e a cen i uga ion s ep o sepa a e soluble (S) om insoluble (P) p o ein. 0.3 lg o p o ein we e applied pe lane. Fig. 5. De ec ion o ex acellula a-amylase by SDS–PAGE. S ain AL02 ca ying pAL12-amyQ was g own as desc ibed in he legend o Fig. 3. Aliquo s we e aken om he supe na an o bo h cul u es a he ime poin s indica ed. Pu ified a-amylase was added o one lane. 11 lg o p o ein we e applied pe lane. Fig. 3. Iden ifica ion o he h pG gene p oduc . Cells o s ain AL05 ca ying he plasmid pAL10-h pG we e g own in LB medium a 37 °C o mid-log ( = 0), di ided in o wo cul u es, whe e one was u he incuba ed a 37 °C and he second cold-shocked o 25 °C. As a con ol, cells o s ain 1012 ca ying pNDH33-h pG we e g own a 37 °C o mid-log and hen cold-shocked and induced by addi ion o 1 mM IPTG o 6 h. Cells we e lysed by sonifica ion and 0.5 lg o p o ein was loaded pe lane on an 10% SDS–PAGE. (a) A e gel elec opho esis, he p o eins we e s ained wi h Coomassie blue. 37 °C cul u e: lanes 1, 2, 4, 6 and 8; 25 °C cul u e: lanes 3, 5, 7 and 9; lane 10, IPTG- ea ed cells g own a 25 °C. (b) Immunoblo analysis o H pG. Cells we e g own and ea ed as desc ibed. A e sepa a ion o he p o eins and Wes e n blo , H pG was de ec ed using an ibodies aised agains his p o ein. Molecula weigh ma ke s a e indica ed. 268 A.T. Thuy Le, W. Schumann / P o ein Exp ession and Pu ifica ion 53 (2007) 264–269 su ed he a-amylase ac i i ies wi hin supe na an and com- pa ed i o hose p oduced a 37 °C. While he ac i i ies we e compa able du ing he fi s 5 h, highe ac i i ies we e measu ed a la e imes in acco dance wi h he esul s ob ained by gel analysis (da a no shown). A 20 °C, we ha e been unable o de ec any a-amylase indica ing ha sec e ion o his enzyme and mos p obably many o he s is se e ely impai ed unde hese g ow h condi ions. I has been epo ed ha he SecA abundance in E. coli was 3- old highe a 20 °C han a 37 °C[24], in acco dance wi h he no ion ha he E. coli p o ein expo includes some in insically cold-sensi i e elemen [25]. Based on hese obse a ions we can only specula e ha a leas one componen o he Sec pa hway does no unc ion p ope ly a 20 °C. This could be he SecA mo o p o ein o /and he SecYEG anslocons o /and a so a unknown componen . Acknowledgmen s This wo k was financially suppo ed by he Deu sche Fo schungsgemeinscha (Schu 414/20-2). The wo exp es- sion ec o s pAL10 and pAL12 can be o de ed om MoBiTec (www.mobi ec.com). Re e ences [1] A. Mogk, M.P. Maye , E. Deue ling, Mechanisms o p o ein olding: molecula chape ones and hei applica ion in bio echnology, Chem- biochem 3 (2002) 807–814. [2] A. Mi aki, J. King, P o ein olding in e media es and inclusion body o ma ion, Bio/Technology 7 (1989) 690–697. [3] J.G. Thomas, F. Baneyx, P o ein mis olding and inclusion body o ma ion in ecombinan Esche ichia coli cells o e exp essing hea - shock p o eins, J. Biol. Chem. 271 (1996) 11141–11147. [4] C.H. Schein, M.H.M. No ebo n, Fo ma ion o soluble ecombinan p o eins in Esche ichia coli is a o ed by lowe g ow h empe a u e, Bio/Technology 6 (1988) 291–294. [5] A.W. Eme ick, B.L. Be olani, A. Ben-Bassa , T.J. Whi e, M.W. Kon ad, Exp ession o a b-lac amase p ep oinsulin usion p o ein in Esche ichia coli, Bio/Technology 2 (1984) 165–168. [6] J.A. Chesshy e, A.R. Hipkiss, Low empe a u es s abilize in e e on a-2 agains p o eolysis in Me hylophilus me hylo ophus and Esche- ichia coli, Appl. Mic obiol. Bio echnol. 31 (1989) 158–162. [7] B. Ha ¨ l, W. Weh l, T. Wiege , G. Homu h, W. Schumann, De elopmen o a new in eg a ion si e wi hin he Bacillus sub ilis ch omosome and cons uc ion o compa ible exp ession casse es, J. Bac e iol. 183 (2001) 2696–2699. [8] M.H.W. Webe , M.A. Ma ahiel, Bac e ial cold shock esponses, Science P og ess 86 (2003) 9–75. [9] P.S. Aguila , J.E. C onan J ., D. De Mendoza, A Bacillus sub ilis gene induced by cold shock encodes a memb ane phospholipid desa u ase, J. Bac e iol. 180 (1998) 2194–2200. [10] P.S. Aguila , P. Lopez, D. De Mendoza, T ansc ip ional con ol o he low- empe a u e-inducible des gene, encoding he Del a5 desa - u ase o Bacillus sub ilis, J. Bac e iol. 181 (1999) 7028–7033. [11] P.S. Aguila , A.M. He nandez-A iaga, L.E. Cybulski, A.C. E azo, D. De Mendoza, Molecula basis o he mosensing: a wo-componen signal ansduc ion he mome e in Bacillus sub ilis,EMBOJ.20 (2001) 1681–1691. [12] L.E. Cybulski, G. Del Sola , P.O. C aig, M. Espinosa, D. De Mendoza, Bacillus sub ilis DesR unc ions as a phospho yla ion- ac i a ed swi ch o con ol memb ane lipid fluidi y, J. Biol. Chem. 279 (2004) 39340–39347. [13] M.C. Mansilla, D. De Mendoza, The Bacillus sub ilis desa u ase: a model o unde s and phospholipid modifica ion and empe a u e sensing, A ch. Mic obiol. 183 (2005) 229–235. [14] M. Mujacic, K.W. Coope , F. Baneyx, Cold-inducible cloning ec o s o low- empe a u e p o ein exp ession in Esche ichia coli: applica- ion o he p oduc ion o a oxic and p o eoly ically sensi i e usion p o ein, Gene 238 (1999) 325–332. [15] G.L. Qing, L.C. Ma, A. Kho chid, G.V.T. Swapna, T.K. Mal, M.M. Takayama, B. Xia, S. Phad a e, H.P. Ke, T. Ac on, G.T. Mon elione, M. Iku a, M. Inouye, Cold-shock induced high-yield p o ein p oduc ion in Esche ichia coli, Na . Bio echnol. 22 (2004) 877–882. [16] A.M. Gue ´ ou -Fleu y, N. F andsen, P. S agie , Plasmids o ec opic in eg a ion in Bacillus sub ilis, Gene 180 (1996) 57–61. [17] I. Pal a, Molecula cloning o alpha-amylase gene om Bacillus amylolique aciens and i s exp ession in B. sub ilis, Gene 19 (1982) 81–87. [18] A. Schulz, S. Schwab, S. Ve s eeg, W. Schumann, The h pG gene o Bacillus sub ilis belongs o class III hea shock genes and is unde nega i e con ol, J. Bac e iol. 10 (1997) 3103–3109. [19] D.L. Popham, P. Se low, Cloning, nucleo ide sequence, and egula- ion o he Bacillus sub ilis pbpE ope on, which codes o penicillin- binding p o ein 4 * and an appa en amino acid acemase, J. Bac e iol. 175 (1993) 2917–2925. [20] T. Wiege , G. Homu h, S. Ve s eeg, W. Schumann, Alkaline shock induces he Bacillus sub ilis W egulon, Mol. Mic obiol. 41 (2001) 59–71. [21] G. Homu h, S. Masuda, A. Mogk, Y. Kobayashi, W. Schumann, The dnaK ope on o Bacillus sub ilis is hep acis onic, J. Bac e iol. 179 (1997) 1153–1164. [22] T.T.P. Phan, H.D. Nguyen, W. Schumann, No el plasmid-based exp ession ec o s o in a- and ex acellula p oduc ion o ecombinan p o eins in Bacillus sub ilis, P o ein Exp . Pu i . 46 (2006) 189–195. [23] J.A. Vasina, F. Baneyx, Exp ession o agg ega ion-p one ecombi- nan p o eins a low empe a u es: a compa a i e s udy o he Esche ichia coli cspA and ac p omo e sys ems, P o ein Exp ess. Pu i . 9 (1997) 211–218. [24] A. Mu akami, H. Naka ogawa, K. I o, T ansla ion a es o SecM is essen ial o he basal and egula ed exp ession o SecA, P oc. Na l. Acad. Sci. USA 101 (2004) 12330–12335. [25] K.J. Pogliano, J. Beckwi h, The Cs sec mu an s o Esche ichia coli eflec he cold sensi i i y o p o ein expo i sel , Gene ics 133 (1993) 763–773. [26] H. Sai o, T. Shiba a, T. Ando, Mapping o genes de e mining nonpe missi eness and hos -specific es ic ion o bac e iophages in Bacillus sub ilis Ma bu g, Mol. Gen. Gene . 170 (1979) 117–122. [27] W. Weh l, M. Niede weis, W. Schumann, The F sH p o ein accumula es a he sep um o Bacillus sub ilis du ing cell di ision and spo ula ion, J. Bac e iol. 182 (2000) 3870–3873. [28] A. Mogk, R. Haywa d, W. Schumann, In eg a i e ec o s o cons uc ing single-copy ansc ip ional usions be ween Bacillus sub ilis p omo e s and a ious epo e genes encoding hea -s able enzymes, Gene 182 (1996) 33–36. A.T. Thuy Le, W. Schumann / P o ein Exp ession and Pu ifica ion 53 (2007) 264–269 269 6. Abb e ia ions 108 6. Abb e ia ions σ Sigma ac o ∆ dele ion µg mic og am µl mic oli e 0C deg ee cen ig ade 2D-Gel wo-dimensional gel 5' UTR 5' UnT ansla ed Region aa amino acid(s) ADP Adenosine-5’-diphospha e ATP Adenosine-5’- iphospha e B. sub ilis Bacillus sub ilis bp base pai s BSA Bo ine Se um Albumin ca gene coding o chlo amphenicol-ace y ans e ase c u colony o ming uni s DNA Deoxy ibonucleic acid DSM Di co Spo ula ion medium E. coli Esche ichia coli EDTA E hylene diamine e aace ic acid e m Gene coding o e y h omycine esis ance e al. e al e i G g am GFP G een Fluo escen P o ein GSH Glu a hione GST Glu a hione-S- ans e ase H hou (s) IPG Immobilized pH g adien IPTG Isop opyl-β-D- hiogalac o py anoside 6. Abb e ia ions 109 kDa kilo-Dal on l li e lacZ be a-galac osidase gene LB Lu ia-Be ani (g ow h medium) LPS Lipopolysacha ide M Mola MALDI-TOF Ma ix-assis ed Lase Deso p ion-ioniza ion Time-o - ligh MS Mass Spec ome y mg millig am min minu e(s) ml milili e mM milimole mRNA messenge RNA OD Op ical Densi y PBP Penicillin binding p o ein PBS Phospha e-bu e saline PCR Polyme ase Chain Reac ion RBS Ribosome Binding Si e RNA Ribonucleic acid SDS-PAGE Sodium Dodecyl Sul a e Polyac ylamide Gel Elec opho esis sec second spec Gene coding o Spec inomycin xs age x o spo ula ion p og am e Gene coding o e acycline esis ance WT Wild- ype E klä ung Hie mi e siche e ich, die o liegende A bei selbs s ändig e ass und keine ande en als die on mi angegebenen Quellen und Hil smi el benu z zu haben. Fe ne e klä e ich, dass ich wede an de Uni e si ä Bay eu h, noch an eine ande en Hochschule e such habe, eine Disse a ion einzu eichen, ode mich eine P omo ionsp ü ung zu un e ziehen. Ai Thi Thuy Le Bay eu h, Decembe 2007