letters to nature

letters to nature IANWOIllly.T A. & Pond. I. L. Ardusbactaial ether lipids and chtmotaxonomy Syst Ang MiroMot 7, 253-25711986/ It Ungworthy, T A.. If olzer. G, Ze&us. I. G. a Tomabent. T. G. bo. and ameiso.branchoi glycerol dietha9 of the thermophix anaerobe Themankrulitthearnon etwourrec Syne.. App! Alarattol 4. 1-17119831 If Huber, It a al Formation of ammonium from nitrate during thernobthmuttoropha growth of the exocmdy thamophilic bacterium Ammaerelcs *soon gen now. p. now. Sat Appl Morolnal 11, 10-49 (156/61. 20. Ilulut. It. a el. Aviles pyraphilm. new genus AnVspenes. repiesetLe • novel poop of wane hmenhamophilic halrogewouduing bateau Sere App! Aftonbset IS. 310-331 (19911 IL I)Rou. Al...y Id.trobatege of Farrow favnoserteenseniel m Pinrease for tfrearketeLyy wile Da Costa. M. 5..1Xuat. J. C. & Williams. R. A.11/ 167-173 (Elsevier. London 1989). 22. Vanden Mumberg.I.0 AL. 'Mama. A I. Al. & KOMI>. W. N.7he cuente of bang eureinophdie the ink el the unique achand membrane hinds. EstionapAilcs 2. 161170 (19981. U. KOMI.. R. grpionta.11. & Runt:mm.1 Alonoalkyletha plimpholipals in 'hemline. reducing lumen. Lkssallossmarta rundults and DesulforlistIslus an•vomc Anil, Altaand 176. 435-412(2001). 24. Mehta. G..v at Qum for bights laddereanesi Olisomenzation of a c)cbbutadsenedeniume. Mae.. Can,, Mr. El 11.1488-1490(19911. n. SUMS. M.. Benne:1.11. Rumen. 1G. &)nien. AI. S. M.'lhe sequencing batch macaw as • powiful cud ins the study of slowly gamin anaerobic aransoniumaudinng nueromganissur. Appt Maemeirt Binierimol SO. M9-596119981 16 llunidrup, R & Dalsgaard.7 Production of Na through anaerobic ammonium oxidation couckd to nitrate reduction in marine sediments Ang Easton. Murano' 64 1312-1318 (20021 7'7 What. /.. de Vries. S. Kuenen. I. G. & lawn. Al. S. AI. Imedairnem of a novel hydroulamine ./odoreductuve in anaerobic ammonium oxidation. Rtorfrarirray39. 5105-541.1(2000). .an [Win. A. C T a Laner. S. A computanorol chemical study of penetration and displacement of waiter films nen mineral surfaces Gmcleem Trans 0061P:01). 29. Ratner-n.11.1 C. Pooma. I P 31. van Guntartn. w. F. DiNola. A. & Hoak. I. R. Molecular dsrumics with coupling man <111O7UI bath 1. Chow. Phys. 81. 34+4-3690 (1981). SOPPlemeaLaY 1010111141100 6.0ImPimics the paper on N811014 wth2le Mtp:fiuwanalure.cOmMillut81. Acknowledgemee% We thank I. G. Kucnen. II. I licnntra. S. Schouten and W Konings for stimulating db.:M.1O114.0 Erkelens (University of Leiden) for access to the 600- and 750-MI Ix NAIR instruments.). A. Fuerst for cells of Gemmel.' obscurigiots“ and Pirellida sp. and training of LA.y.N.. A1. wolicrs.Arts for help with electrum microscopy. and K. T. van de Pas.Schoonen for help with immunolluorescence. Caimpeelna nada Material The authors declare that they have no competing financul interest, Corespondents and requests for materials should be addressed to a in : .!..risteatmunl). Jasmonate and salicylate induce expression of herbivore cytochrome P450 genes Xianchun Li' , Mary A. Schuler; 8, May R. Berenbauntt I la 1 PI °tertian. Nanjing Agrieultoral LiniVaSity. Nanjing 210095. (Atm t Department of Entomology and Department of all and Structural Biology. University of Illinois. Urbana. Illinois 61801, USA Jasmonate and salicylate are plant-produced signals that activate plant defence genes after herbivory" or pathogen' attack. Amplification of these signals, evoked by either enemy attack or experimental manipulation, leads to an increase in the syn- thesis of toxic compounds (allelochemicals)' and defence pro- teinr ° in the plants. Although the jasmonate and salicylate signal cascades activate different sets of plant defence genes10, or even act antagonistically''''', there is substantial communication between the pathways"". Jasmonate and salicylate also contrib- ute to protecting plants against herbivores by causing plants that experience insect damage to increase their production of volatile molecules that attract natural enemies of herbivorous insects". In response to plant defences, herbivores increase their pro- duction of enzymes that detoxify allelochemicals, including cytochrome P450s (refs 15, 16). But herbivores are potentially vulnerable to toxic allelochemicals in the duration between ingesting toxins and induction of detoxification systems. Here we show that the corn earworm Helicoverpa sea uses jasmonate and salicylate to activate four of its cytochrome P450 genes that are associated with detoxification either before or concomitantly with the biosynthesis of allelochemicals. This ability to 'eaves- drop' on plant defence signals protects H. sea against toxins produced by host plants. The corn earworm, H. tea, is broadly polyphagous, with over IGO known host plants including herbs, shrubs and other low-lying vegetation. We chose H. sea to address the general issue of whether herbivorous insects can activate their enzymes that metabolize a 10 CYP6828 in midgut c 8 ..og 6 S ▪ 4 tai 2 0 10 8 6 5 4 0 ts. 2 0 EF-1u C11, 6628 large cyp6B8[ band b Small d b 10 O 7.5 .? 5 LL 2.5 0 CYP689 EF-lea Ala CYP688 in mtdgu1 4 g 10 , 8 9 6 .e 4 §. 2 0 10 c 8 6 • 4 2 0 CYP6828 in fatbody D x 4 'A a CYPE88 in fatbody z ma CYP6a9 in midgut • 10 CYP6827 in midgut n ICI 5 7.5 • 5 §. 2.5 0 CYP6827 EF•tu aril Z 4 A Floret CYP68 gene egression in H. zea in response to jasmonate (JA) and salicylate (SA). a. CYP688/CYP6828. b. CYP68.0. c. CYP6827. Total midgut or lathed/ MIAs from the different diet treatments control diet. JA-L 2.9 nag "l a JA-fl. 290 pg g- r JA: SA.L. 12 mpg - ' SA: SA.H. 1.2 mgg-1 SA) were separately amplified by RT-PCR and analysed as described in Methods. The average f cid Indiction and standard deviations (error bars) for three independent RT-PCR amplificationsare shown In the histogramt an autoradlogram of a representative blot Is shown below. 712 0 2002 Nature Publishing Group NATURE IAOL 419 117 OCTOBER 2001Iwyny.nature.comauture EFTA00765611 letters to nature EMS fatelahly Tate 1 Effects of )aamonate and salltylate exposure on growth rate, welgM gain and mortality of H. rea Winton ot fourth War Marton of Win Maar Mcgtally on clay 3 Weight gain on thy 3 Pupal weight final au:orlon rate Tigatmente (c9 psi (m9) Mfg (74i (501 No earaure Contra tIbba 1.5 ±0.5 d 5.5 -z 0.5 OD 0.0b 384.4 t- 25.6 a 345.4 ± 13.5a 15.6 ± 5.1 0 84.3 ± 15.1 abc XanlhotoxIn diet 3.8 ± 0.8 b 11.7=0.3b 8.9 ± 1.9b 1022 t- 28.3 be 225.3 ± 18.5 b 30.0 ± 10.0 bed 70.0 ± 10.0 al Celery lenge 5.7 ± 0.3 a 14.7 -z 0.6 a 33.3 ± 15.3a 47.6 t- 2.7d 163.2 ± 23.4C 85.7 ± 11.8a 12.8± SS e JAL Control dials 1.5 ±0.5 d 5.5 -z 0.5e OD ± 0.0 b 3961 t- 7.8a 348.2 ± 24.3 a 5.7 ± 5.7 0 90.3 ± 5.8 a XantholoxIndlet 2.5!O.5O 9.5 -z 0.5c °At 0.0 b 78.9 t- 15.4 bad 2266 ± 2.5b Mit 11.8 bed 73.3_ 11.8 bog Celery leaves 2.5!O.5O 8.5 -z 0.5d 3.0 .±- 5.3 b 845 t- 17.9 bad 17A1 ± 39.5c 48.3 ± 16.1 be 48.7 ± 20.1 0 SAL Control Pets 1.5-10.5d 4.7 -z 0.6e &St- 5.8 b 382.1 ± 50.3 a 372.9 ± 11.5a 10.0 ± 10.0d 89.9 ± 10.0 ab Xantholoon del 2.5!O.5O 9.7 -z 0.6c OD t- 0.0 b 113.1 _9.3b 232.7 ± 5.5b 23.3 ± 5.8 ad 75.7 ± 5.8 tocl Celery leaves 2.5 ± 0.5 0 10.8 -: 0.3 b 7.0 ± 5.1 b 69.9 ± 15.5 ocl 1661 ± 15.6c 49.9 ± 10.0 b 50.0 ± 10.00 Efileds were meesurecl horn fourth Sir ta pay on cekry leaves and ce callid end 0.5% methateoin dela. The deb contained 2.9 IC rJA OM) cr 12 eg ci“ SA (SA-40We es percentages %ere weave trarefamied tetare apart. %wile are the trdralslccrnecl means e.d. In each mini. means (laved by cifferent Idlers are sisnficanly diluent < 0.05. nexilied L.SD blest). allelochemicals in response to the plant signal molecules jasmonate and salicylate before the accumulation of plant defence compounds. Transcripts of four H. sea cytochrome P450 (P450) genes are inducible by furanocoumarins, chlorogenic acid, indole-3-carbinol and flavone16•", which suggests that these genes are involved in detoxifying a range of plant allelochemicals. But these four P450 genes are not universally inducible by all allelochemicals of host plants; for example, gossypol, quercetin and rutin do not induce their transcription". Baculovirus-mediated expression of one of these proteins, CYP6B8, has shown that this protein can metabolize xanthotoxin (221.1 pmol per ml of baculovirus-expressing cell culture per min; unpublished data) —a furanocoumarin that is present in many host plants of H. sea and whose biosynthesis is stimulated by jasmonate and methyl jasmonate'. In Apium graveolem (celery)'•", a host plant of H. sea", xantho- toxin and the related furanocoumarin bergapten begin to accumu- late after 24 h and reach maximal concentrations (representing a 40-70-fold increase) 4-6d after the application of jasmonate and methyl jasmonate''. Accordingly, we fed fifth instars of H. sea for 48 h with either artificial diets supplemented with jasmonate and salicylate (at two concentrations for each chemical) or control diets, and then examined expression of four P450 genes, CYP6B8 nn 1 d NO OOOO OOQC) QUO() - IT UUUU Control MP p-HBA SA-H Chemicals Ruin 2 C1P6Bexpressmn in response to salicylate (SA) and SA.retated frhAroxybenzoic acid and methylparaben. Total FtNAs from catapillam fed on diets ambling 1.2 mg g "I SA (SA-Hr. 1.2 mg g-' methylparaben (MP) or 1.2 mg g I p-hAroxyl:enzcic acid (p-IIBA) were amplified by RT-PCR and analysed as described m Methods. The relative induction and standard denabons fix three independent RT—PCR amplifications are shown. (AF102263), CYP6B9 (AF 140278), CYP6B27 (AF285829) and CYP6B28 (AF285186), in the midgut and fatbody—the principal sites of allelochemical detoxification in this species". Among the P450 transcripts examined, CYP6B8 and CYP6B28 (99% amino acid identity), a pair of highly conserved paralogues", were simultaneously amplified by polymerase chain reaction with reverse transcription (RT—PCR), differentiated by digestion with Xmul, and quantified by gel blot analysis (Fig. la). In midguts, CYP6B28 transcripts were induced about 5.0-fold by jasmonate and salicylate irrespective of the concentration, whereas CYP6B8 tran- scripts were induced about 4.5-fold by either concentration of jasmonate, 3.3-fold by the low concentration of salicylate and 7.1- fold by the high concentration of salicylate (Fig. la). In fatbody, CYP6B28 transcripts were induced about 6.0-fold by either con- centration of salicylate, 4.2-fold by the low concentration of jasmonate, and 6.2-fold by the high concentration of jasmonate; transcripts of CYP6B8 were increased to a lesser extent by jasmonate and salicylate (Fig. la). The more divergent CYP6B9 and CYP6B27 transcripts derived from another pair of paralogous P450 genes (87% amino acid identity with CYP6B8), whose expression is restricted to midguts", were separately detected by RT—PCR gel blot analysis (Fig. lb, c). In midguts, CYP6B9 transcripts were induced 6.0-fold by low concen- trations and 8.0-fold by high concentrations of jasmonate and salicylate (Fig. Ib), whereas CYP6B27 transcripts were induced 4.8-fold and 5.8-fold by low concentrations of jasmonate and salicylate, respectively, and 6.9-fold and 7.8-fold by high concen- trations of jasmonate and salicylate, respectively (Fig. 1c). These results show clearly that expression of CYP6B is activated in the midgut and fatbody of H. sea at the low concentrations of jasmonate and salicylate that are associated with pest damage and allelochemi- cal induction in its host plaints"•". To assess the specificity of this induction response, we tested further the induction of these P450 genes in response to two salicylate-related chemicals at equivalent concentrations to high concentration of salicylate. Methylparaben, which differs from salicylate in the position of its hydroxy group and in having an additional methyl ester group, did not induce any of the CYP6B genes examined. Not surprisingly, p-hydroxybenzoic acid, which differs from salicylate only in the position of its hydroxy group, acted as a weaker inducer than salicylate and increased the amounts of CYP6B8 and CYP6B28 transcripts roughly 2.0-fold, and the amounts of CYP6B9 and CYP6B27 transcripts 4.0-5.0-fold (Fig. 2). These results indicate that the degree of activation of H. sea CYP6B genes by salicylate, jasmonate and related compounds is dependent on structural features of these signal molecules. To test whether an increase in endogenous amounts of signal substances in plants that occurs before allelochemical biosynthesis is sufficient to induce transcriptional expression, we allowed starved fourth instars of H. zea to damage celery leaves and then determined the ability of these leaves to activate transcription of CYP6B in a XATURE j VOL 419 j II OCTOBER 2002 kveve.nature.cominature 0 2002 Nature Publishing Group 713 EFTA00765612 letters to nature second set of fifth instars 2 and 4 h after damage. Compared with leaves from two undamaged control plants, which did not induce CYP6B expression, leaves from all of the plants that had been attacked for 2 and 4h induced expression of CYP6B28, CYP6B9 and CY6B27 (Fig. 3). Analysis of these plants indicated that, in a background of up to twofold constitutive differences in furanocou- marin content and composition among the test plants, no induced accumulation of furanocoumarin occurred. The differences in CYP6B expression were not correlated with variations in furano- coumarin between plants (Fig. 3), which indicated that the acti- vation of CYP6B transcription resulted from an induction of jasmonate and/or other signal substances caused by the feeding behaviours of the first set of larvae. These data provide direct evidence that H. zea can intercept the plant defence signals elicited by its own feeding activity. To determine whether activation of P450 genes in advance of exposure to furanocoumarins confers protection on H. sea, we compared the survival and growth of fourth instars that had prior exposure to low concentrations of jasmonate and salicylate for 12 h on celery leaves, 0.5% xanthotoxin diets, or control diets against that of control larvae that had not been exposed in advance to signal substances. Two-way analysis of variance (ANOVA) and multiple comparison tests on mortality, weight gain, growth rate and pupa- tion success (Table 1) indicated that caterpillars exposed to jasmo- nate and salicylate survived better on celery leaves and 0.5% xanthotoxin diets for all parameters. On control diets, however, there were no significant differences in all parameters among the three treatments. These results suggest that the 'signal-eavesdrop- ping' capability provides H. sea with prophylactic protection against plant defences at no additional cost to fitness in the absence of plant defences. a so sec s § 3 u g '5 E U. c 20 40 0 Plant b 15 Reciprocal phenotypic responses characterize many antagonistic ecological interactions; if such reciprocal phenotypic change results from adaptive plasticity in the interacting species, then coevol- utionary interactions may result in the evolution not only of fixed adaptations but also of phenotypic plasticity". The induction of P450 counterdefence genes in herbivores in response to plant signal substances that are themselves inducible by herbivore damage might be an example of such phenotypic plasticity. Although it is well known that herbivorous insects can enhance the expression of detoxification enzymes (counterdefences) in the presence of plant allelochemicals (plant defences)1" 72121, we have shown here that H. zea responses to plant damage are more sophisticated than was thought previously. By responding to plant signal molecules as well as the end-product allelochemicals, insects have the capacity to equip themselves before (or concomitant with) the accumulation of toxic concentrations of plant defence compounds. Although several examples have been found of plants using insect-derived signal substances to regulate their defence pathways'-', this represents to our knowledge the first example of the use by insects of plant signal molecules to regulate their defence systems against plant allelochemicals. The ability to use plant signal molecules as cues for activating a detoxification system may be of particular value to a broadly polyphagous herbivore such as H. sea. In contrast to oligophagous species, which encounter a relatively narrow and generally predict- able range of plant allelochemicals, generalized herbivores may encounter any of several biosynthetically distinct compounds depending on host plant choice". Few commonalities exist among the biosynthetic pathways that generate these plant defence compounds other than the fact that they share jasmonate or salicylate as initiating signals. The ability of a generalist to respond K InVeratorin • SendaF400 ▪ Isopimpinellin ▪ xanthotoxin Sphondin Ea Total 0 I A Undamaged 0 2 h of damage 4h of damage K CYP688 K CYP6828 ECYP689 IF 10 • CYP6827 ctC 2 c 0 2 5 0 Plant A Undamaged D E 2 to of damage Rgure 3 CYP6Bachvation by feeding de damaged celery leaves. a, Content of total and individual furanocoumarins for each plant. b. Relative nduchan and standard deviations for three independent RT-PCR amplifications. Total RNAs from fifth inters fed fa 48 h on F G Shot damage H leaves that were previouslyundamaged or damaged for 2 or 4 h by starved loath linters were amplified by RT-PCR and analysed as deserted in Methods. 710 0 2002 Nature Publishing Group NATURE [VOL 419111 OCTOBER 200:1 wvv.natuze.cominaturc EFTA00765613 letters to nature to these signals by upregulating several detoxification genes may maximize its ability to counter its host's response to damage, irrespective of taxon. 0 Methods Test insects An insecticide-susceptible laboratory grain of H. sea. provided by B. R. Banido (Abbott Laboratories). was used in all studies. We kept insects in an insectary maintained at 28°C in a 16:8 h light:dark cycle on a sernisynthetic control diet containing wheatgenn". Signal chemical Induction treatment Artificial diets containing 2.9 or 290 lig g jasmonate ISigmal. 12 ass or 1.2 mgg salicylate 199%. Aldrich). 1.2 mg g methylparaben I Sigma). or 1.2 mg g p. hydroxybenzoic acid (Sigma) were pecmided to 30 newly moulted fifth instars. The low concenua 'ions of iasmonate and salicylate were selected on the basis of endogenous amounts of immonate and salicylate found in the host plants of H. ear"' and the high concentrations were selected to maximize the likelihood of detecting an upper limit on the response. After 48 h. midguts and (Audits were dissected out and total RNAs were isolated from each type of tissue using guanidine-11(A extractionv and then resuspended in diethyl pyrocarbonate I DE PC: • trea ted water. Relay damage and Induction treatment We grew nine celery plants individually in pots under laboratory conditions for 3 weeks to ensure that they were free of herbivore and pathogen infestation. Eight of than were free of infestation and were assigned randomly to one of duce groups: undamaged control. 2 h of damage and 4 h of damage. For each plant. four menu with hilly expanded pain of leaflets and a terminal leaflet were choien for treatment. (hi each stem. WO fourth Instrs that had been started for 4 h were confined to the second pair of leafless by two small clip ages. with one larva per leaflet. For the undamaged controls. clip cages without larvae were placed on the second pair of leaflets on each stein for 4 h. After damage treatments. the second pair of leaflets was removed from each treated stem. We used one damaged leaflet to feed a newly moulted fifth instar that had been starved for 4 h. NE pooled another leaflet with the other three kJ nets from the same plant. oven-dried them at 50°C (or 24 h and used them for furanocoumarin determination. After 48 h of feeding on the damaged leaves. lame were killed and the mdguts and fatbmIlt, were rerniwed.The midgets from the four larvae kd the damaged leaVeN from the same plant were pooled together and total RNA was isolated as described'. RNA and turanocounatin analysis We carried out RNA isolation and RT-PCR gel blot analyses as described". For each RNA sample. three independent RT-N:R amplifications were carried out. For furanocoumarin assay. all leaf samples were weighed separately and ground to a fine powder with a plastic rod inside Eppendod tubes. Furanocoumarins were extracted, separated and quantified as described". asmonate and salicylate protection bioassay Newly moulted fourth mums 2701 from the University of Illinois laboratory colony were divided randomly into three groups (90 larvae per group) and reared individually in plastic cups with fresh control diets or supplemented diets containing 7-9 pgg jasmonate or I 2 pgg salicylate. After 12 h of exposure to plant signal molecules, each group was divided further into three subgroups (30 larvae per subgroup. three replicates of ID insects) that were transferred to plastic cups with fresh control diets. diets containing 0.5% xanthotoxin. or celery leaves. Initial weights were recorded for every individual. All larvae were weighed again on the third day after transfer to experimental diets to determine weight gain. We monitored survival and developmental stage daily until all larvae had either pupated or died. Differences in weight gain. duration of fourth and fifth instars. mortality. pupal weight and percentage of pupation among the treatments were evaluated by ANt/VA. followed by modified least significant difference test (LSD *test/. with the significance level set at P < 0.05 using the SAS statistics program. Received 19 March accepted 27 lune 2002: doi:10.1035/nature01003. 1. McCloud. F. S. & Radio* 1. T Herbivoreand catapdlar reguinums amplify the woundanduring increases in pinwale acd but not nicotine in Nintione tylnsins. Hama 103.430-435'19971. L 11.1. L.. bluiphy. 5 It it klion.G. W. Dom salighc and an asasignalincotton foe induced resistance to 1164nnespr am? 1. Owe. gent 23.1005-1818 (1997). 3. Moran. P. & Thompson.G. A. Molecular responses to aphid feeding inMeltdown in relation to plant defense pode.vays. P444I Ansa 125. 1074-1085120011. 4. Setae. M.Schulaw. V. & Raskin. I. Endogenous =M)/ salierlace in Nthoinnkoculated tobacco plants. Paine Paysial 11& M7-39211998). 5. gamine& M.Oldham. N. t & Baldwin. 1. T Rapid HPLC sateningof tomonaknwhiced increases sec tobacoaalkamds. phenolics. and deterpree glycosides in Ne4flatilt anounira.1. Apt. lined Chow. 49. 3553-3558 (20011. ▪ Tschamdm T. Thiessen. 5.. Dolch.R /4 Boland. W. Hatinon. induced resistance and manly,' *MI transfer in Arniughernow Riodinii. Sawyer. Fed. 29. 1025-1047 (NOD 7. Volker. 5. kern. P * Boland. W Biosyrithek or furonacoumoriew menbanateindependent pienttation of umbellikione in Arnow prints, (Apiacciet Phinarlinnorry 50.1141-1145 09991 & Kiddie. G. A. DoodWY. K I. & Walbgrear. R M saliejlic acid-induced accumulation or glucosinotnes in oared rap (lbw vs wpm/Ukase./ Exp Botany 45. 1343-1)46115941. 9. llialer.1. S. Mout.ht I.. Kansan. R &Duffey.5.5. Exogenous immonates nimobie insect wounding in 1011W10 plants apnyrntool onekamm) in tit< laknalogy and Inld./. Chem Ea U1767-178111990 to llsomma. B. P I. red &Wilt jaimanatedererdent arid sabeYbandrIsendan ckfewertsforne pathways in Anilddrmis are esiamial for resistance to distinct microbial pathogens. Poe. Mul Ana Ser. USA 95. 15107-15111(19981. IL Penton. C A.. LereandiensH. C. Enyedi. A I. & aalsdn 1. T Tobacco mimic nun inoculation inhibits smond.induad iasmonic acid.medimed rano/neswithin but not between plants Plana 209. 87-95 11999). 12. fetlockC. W. a d. Inverse relationship between systemic Rushlike of plants to microorganisim and to meet !tallowy. Gent But 9017-320 (19991. IS. Stout.51.1.. Eidantsef. A. L.Duffey.S. S ft Rostock P. M.Sursil intaactions in pathogen and insect muck. SySICMIC plam.inediated attractions between pathogens and herbivores of the tomato. Lfrommuten (intim:am. Hewer. 3164 Haw Rola 54. 115-110 (19991. 14. Thalet.I. 8. )aunonmeanduableplarg Mann. came manned para.:nunsof batman. Nome 399. 686-618 (1999/. IS. Schuler. M. A. The role idemoduonse P450 nxinoogrumases in plant-insect interactions. Ham Proud. 113, 1411-14191199). 16. Li. X.. Berenboum. M. R. &Schuler. NI. A Mokrube doning and e.qsreseion of CTP6BS: 4 xandicanxikinduable rd.:chrome P450 cDNA from Hottorrenn ant WC. 81100,1 Md Btel .10. 75-8418700). I7. 1.i. X, Berenbouna.M. R. de Schulte. M. A. Plant alelocbanicalidiffermtially regulate Behooving rea crtochrome P450 MO. WAY Met Bid II, 343-35112002). 18. Milusch.M.* Bolao3.W.Airborne methyl rasnxinatenimulates die biosyntheis of furanocounwrins in the leaves of celery plants (Apinia awnvolnol. Egeerientie 52, 739-743 ( I 996) 19. ear. It M. AR Mika fork Larststhd Lfi lfinents. Lady Sat*. andHanel Matiorrmitorne alter Commitment Omni Sraresentl Canal. Wt l and 2 (A. C Allyn Mut Entomology. Sarasota. Horeb. 19721. 20. Li. X. Berenbaum.51. R & Schulte. Al. A. Cytochroene P45nand actin sum opened in Ilebeerogn au and Ikea spa Amgen paresis...366.43p tdentitionton. gar osneentent. and evolution. /men &atm MA But 12, 311-320 Ono*. It, Row. 1. reel Al111.411144410.1.11441 depletion of • potato lipoxygause seduces wound induction of Nous:use inhibmin and increases weight gun of insect pots. Pew. Mid And. Sn. USA 96. 1146-1151 05991. 22. Amami. A. A Phenotypk plankity in the interactionsand evolution of species. Serowe2K 321-326 (20011. 23. Danielson. P. B.. Macintyre.R I. *Fogleman.) C. Molecular cloning of a family of senobitak- indualle drorophilid cytoduome P454* Evidence for imolwinent in host-plant alklothemkal rokunce. Peer Abel Aid Set USA 94. 10797-10802 (1997). 24. Snyder.ht L. Andersen. F ta Enenian.R. Eyre:don of cnochionw NiOgenes of the CYP4 famdy in midgut and fatbody °Diu tobaccohonnowm. Alandimi texts An*. /Winn Blithp. 321. I3-Xi 11995). 25. Beembaum. M. R. an lamb, grelegy of rke Them Roane Ids Puga.A.& Wallace. K. R1553-571 Oh*, & Francs. Philadelphia. PA, 1999). 26. Waldbauer.G. P. Cohen. R. W. & Friedman.* An improved procedure fa laboratory rearing of the corn noworm. Heliothis ma iLepadown• Nocundm). (emu Lein Enramed 17.113-114 Mb*. 27. Sambrook. ). Enoch. E. F. & Mantis. T. 31nkeubv <Seeing. A 14Invesary Manuel 7.23-716 (Cold Spring Harbor Laboratory Pane. Cold Spring limbo,. NY. 19894 28. Zanged. A R. Amu. A. XL & Rerenbaum. A R. Physsolopcal rue of an inducedchemicaldefense photosynthesis. respiration. biosynthesis, and pond. neceama 109. 413.441 119971. Acknowledgments 1* thank M.Catroll for valuable discussion. This researchwassupported by gr.tro, from USIM to M.R.B. and NIA.S.. and from the China Natural Science Foundation to X.L. Compering Interests statement The authors declare that they have no competing financial imern:s. Correspondence .Ind requests for materials should be addressed to ARR. le. mad: maybo,ustse.edu) or XL. (e-mail IxoBlife.muc.eduk A biological role for prokaryotic CIC chloride channels Ramkumar Iyer, Tina M. Iverson, Alessi° Accardi & Christopher Iii ller De porunent of Biochemistry, Howard Hughes Medical Institute, Brandeis University, 1Valthant, Massachusetts 02454, USA An unexpected finding emerging from large-scale genome ana- lyses is that prokaryotes express ion channels belonging to molecular families long studied in neurons. Bacteria and archaea are now known to carry genes for potassium channels of the voltage-gated, inward rectifier and calcium-activated classes", Clc-type chloride channels", an ionotropic glutamate receptor' and a sodium channels. For two potassium channels and a chloride channel, these homologues have provided a means to direct structure determination'". And yet the purposes of these NATURE I VOL 419117 OCTOBER 2002 1 wvevanaturecominature 0 2002 Nature Publishing Group 715 EFTA00765614