A new view of carcinogenesis and an alternative approach to cancer therapy

144 | LÓPEZ-LÁZARO | MOL MED 16(3-4)144-153, MARCH-APRIL 2010 CANCER AS A GENETIC DISEASE The gene ic basis o cance was dis- co e ed mo e han a cen u y ago, when Da id on Hansemann obse ed ha cells om a ious ca cinoma samples had ch omosomal al e a ions. A he be- ginning o he 20 h cen u y Theo ho Bo e y sugges ed ha cance migh e- sul as a consequence o hese ch omo- somal abe a ions, laying he ounda- ions o iewing cance as a gene ic disease (1). Bu he idea o cance as a gene ic disease was no widely accep ed a ha ime. I cance was caused by a gene ic mu a ion, i was no clea why he e was a delay o many yea s be- ween he exposu e o a mu agenic agen and he onse o cance , o why he incidence o cance inc eased so d a- ma ically wi h age. O e ime, he ob- se a ions ha no single gene de ec causes cance and ha se e al mu a- ions could be equi ed o cance o de- elop explained he long la en pe iod o cance and he age dis ibu ion o his disease (2–4). In he second hal o he 20 h cen u y, he idea ha he de elop- men o cance equi ed he acquisi ion o se e al mu a ions began o be widely accep ed, and e o s we e made o iden i y which genes we e in ol ed in ca cinogenesis (5,6). In he 1980s he i s human oncogenes and umo -sup- p esso genes we e disco e ed, and he idea ha cance was caused by mu a- ions in hese wo ypes o genes was gaining i m g ound (7–11). Mu a ions in oncogenes would inc ease he syn he- sis o p o eins ha s imula e cell p oli - e a ion, and mu a ions in umo -sup- p esso genes would esul in he loss o p o eins ha es ain cell p oli e a ion and induce apop osis. The accumula ion o se e al mu a ions in hese wo ypes o genes would allow cells o p oli e a e in an uncon olled ashion and would lead o cance . I was belie ed a ha ime ha cance would be explained by a ela i ely low numbe o mu a ions in hese genes, and ha cance would e en ually be ea able by e e sing o exploi ing hese gene ic changes. Molecula analyses o human umo s ca ied ou in he las decade ha e e- ealed ha he gene ic al e a ions o cance cells a e much mo e nume ous and uns able han p e iously hough (12). Fo ins ance, by sampling colo ec- al p emalignan polyp and ca cinoma cell genomes, S ole e al. (13) ound ha he mean numbe o genomic changes pe ca cinoma cell was app oxima ely 11,000. In addi ion, much e idence has accumula ed s a ing ha mu a ions (changes in he DNA sequence) a e no he only cause o he al e ed gene ex- p ession o cance cells. Epigene ic al e - a ions (he i able and e e sible changes o he han he DNA sequence) and ane- uploidy (nume ical and s uc u al ab- no mali ies in ch omosomes) a e com- mon al e a ions in umo cells, which modi y gene exp ession and may also play a c ucial ole in ca cinogenesis (14–18). Such is he gene ic complexi y and a iabili y o umo cells, ha he A New View o Ca cinogenesis and an Al e na i e App oach o Cance The apy Miguel López-Láza o Depa men o Pha macology, Facul y o Pha macy, Uni e si y o Se ille, Se ille, Spain Du ing he las ew decades, cance esea ch has ocused on he idea ha cance is caused by gene ic al e a ions and ha his disease can be ea ed by e e sing o a ge ing hese al e a ions. The small a ia ions in cance mo ali y obse ed du ing he p e ious 30 yea s indica e, howe e , ha he clinical applica ions o his app oach ha e been e y limi ed so a . The de el- opmen o u u e gene-based he apies ha may ha e a majo impac on cance mo ali y may be comp omised by he high numbe and a iabili y o gene ic al e a ions ecen ly ound in human umo s. This a icle e iews e idence ha umo cells, in ad- di ion o acqui ing a complex a ay o gene ic changes, de elop an al e a ion in he me abolism o oxygen. Al hough bo h changes play an essen ial ole in ca cinogenesis, he al e ed oxygen me abolism o cance cells is no subjec o he high gene ic a iabili y o umo s and may he e o e be a mo e eliable a ge o cance he apy. The u ili y o his no el app oach o he de- elopmen o he apies ha selec i ely a ge umo cells is discussed. © 2010 The Feins ein Ins i u e o Medical Resea ch, www. eins einins i u e.o g Online add ess: h p://www.molmed.o g doi: 10.2119/molmed.2009.00162 Add ess co espondence and ep in eques s o Miguel López-Láza o, Depa men o Pha macology, Facul y o Pha macy, P o eso Ga cia Gonzalez 2, 41012 Se illa, Spain. Phone: + 34 954 55 63 48; Fax: + 34 954 55 60 74; E-mail: [email p o ec ed]. Submi ed No embe 6, 2009; Accep ed o publica ion Decembe 26, 2009; Epub (www.molmed.o g) ahead o p in Decembe 28, 2009. REVIEW ARTICLE MOL MED 16(3-4)144-153, MARCH-APRIL 2010 | LÓPEZ-LÁZARO | 145 idea o unde s anding cance in e ms o changes in speci ic genes is losing g ound in a o o p oposals ha seek o a ionalize cance in e ms o a lim- i ed numbe o acqui ed pheno ypes ( he so-called hallma ks o cance ) and al e ed cellula pa hways (19–21). We a e beginning o ealize ha he high gene ic a iabili y o umo cells is a se- ious obs acle o he design o gene- based he apies ha may ha e a majo impac on cance mo ali y (12,21). Despi e he complexi y and a iabili y o he cance genome, much esea ch is de o ed o cha ac e izing he gene ic p o ile o umo s o a ionalize and pe - sonalize cance he apy (22–24). The e- cen app o al o se e al cance - a ge ed he apies ( he apies ha a e in ended o a ge he molecula de ec s o cance cells speci ically) indica e ha he al e ed genome o cance cells can be exploi ed he apeu ically (25). The landma k e en in his new ield was he de elopmen o ima inib mesyla e (Glee ec) o he ea - men o ch onic myeloid leukemia (CML). This d ug was de eloped as a se- lec i e inhibi o o he kinase BCR-ABL, he usion p o ein p oduc o a ch omo- somal ansloca ion ha is in ol ed in he pa hogenesis o CML. Clinical ials e ealed ha ima inib mesyla e induced comple e hema ological emission in a e y high pe cen age o pa ien s wi h ch onic-phase CML (26). Al hough his d ug has become he s anda d o ca e o CML, i is impo an o no e ha ima- inib mesila e does no cu e he disease (i jus keeps i unde con ol o many pa ien s o as long as hey ake he d ug) and ha some pa ien s ea ed wi h his d ug de elop esis ance o ea men (27,28). O he a ge ed he a- pies ha e been app o ed o cance ea - men in ecen yea s; howe e , none o hese he apies ha e led o majo im- p o emen s in he su i al o pa ien s wi h he mos common ypes o cance . I is impo an o be eminded ha cance mo ali y has no changed much du ing he las h ee decades (29), and ha he small declines obse ed in ecen yea s in some ypes o cance ( ha is, lung, co- lo ec al, b eas and p os a e cance ) a e no a ibu able only o be e he apies bu also o he implemen a ion o p e- en ion and ea ly de ec ion campaigns (30,31). Al hough he gene ic al e a ions o umo cells a e ex emely nume ous and uns able (12,13,21), we a e mo ing o- wa d an e a o pe sonalized ea men s based on he gene ic p o ile o each umo (22–24). I is expec ed ha much ime and e o and many esou ces will be necessa y o de elop he apies ha will be use ul in a small pe cen age o pa ien s wi h cance . In addi ion o building up a complex a ay o gene ic changes, umo cells acqui e an al e a ion in he me abolism o oxygen, a p ocess ha plays an impo an ole in ca cino- genesis and could be exploi ed o de- elop he apies o a b oad ange o pa- ien s wi h cance . KEY ROLE OF ALTERED OXYGEN METABOLISM IN CANCER Nonmalignan cells use oxygen (O2) o gene a e ene gy in he o m o ATP h ough he p ocess o oxida i e phos- pho yla ion (oxphos). Accumula ing e i- dence indica es ha , ins ead o ully cou- pling he me abolism o O2wi h he gene a ion o ene gy, cance cells ac i a e glycolysis o mee hei ene gy demands and use O2 o gene a e excessi e le els o he eac i e oxygen species (ROS) supe - oxide anion (O2 •–) and hyd ogen pe ox- ide (H2O2). This al e a ion in he me abo- lism o O2(dysoxic me abolism) is a common ea u e o cance cells and plays an impo an ole in ca cinogenesis (32–38). I is well known ha uncon olled cell p oli e a ion is he mos ele an ea u e o cance . I is also ecognized ha he gene ic de ec s o cance cells esul in an al e ed gene exp ession and in he p o- duc ion o signals ha make hese cells p oli e a e in an uncon olled ashion. Equally impo an o cell p oli e a ion is ha he di iding cell duplica es all i s cellula componen s o c ea e wo daugh e cells (Figu e 1A). To do his, p oli e a ing cells mus ake nu ien s om he blood and use hem o syn he- size all he mac omolecules and cellula componen s equi ed o he o ma ion o a new cell. The cellula up ake o glu- cose om he blood and he ac i a ion o glycolysis a e essen ial p ocesses o cell p oli e a ion, because he ac i a ion o glycolysis p o ides mos o he building blocks equi ed o he syn hesis o hese mac omolecules and cellula componen s (37). Se e al glycoly ic enzymes a e o e - exp essed in cance cells and ha e been shown o play an impo an ole in can- ce (39–41). The inc eased cellula up ake o glucose and he up egula ion o gly- colysis o cance cells can indeed be ob- se ed wi h clinical umo imaging ( luo- odeoxyglucose posi on-emission omog aphy) and is cu en ly used o ea ly diagnosis and be e managemen o oncology pa ien s (42). The ac ha he blood essels deli e bo h glucose and O2is a p oblem o he ac i a ion o glycolysis, because he p esence o O2is known o cause glycol- ysis inhibi ion (Pas eu e ec ). This si u- a ion sugges s ha o p oli e a e cance cells mus de elop he capaci y o ac i- a e glycolysis in he p esence o O2 (Figu e 1A), a cha ac e is ic ha was i s obse ed se e al decades ago by he Nobel lau ea e O o Wa bu g. Wa - bu g also p oposed ha he high gly- coly ic a es he obse ed in cance cells despi e he p esence o O2we e caused by a de ec in espi a ion (oxida i e phospho yla ion) and ha his de ec was he o igin o cance (43). Al hough he ac i a ion o glycolysis in he p esence o O2(ae obic glycolysis o he Wa bu g e ec ) has epea edly been obse ed in cance cells (37,42,44), i is no clea why and how his phenom- enon occu s. I i s p oposed ha o p o- li e a e cance cells (and no mal p oli e - a ing cells) mus ac i a e glycolysis despi e he p esence o O2(33,37). As shown in Figu e 1A, cell p oli e a ion would be comp omised i glycolysis we e always inhibi ed in he p esence o O2(33,37). The same p oposal was la e made by o he s (45). As o how his phe- nomenon occu s, e idence sugges s ha 146 | LÓPEZ-LÁZARO | MOL MED 16(3-4)144-153, MARCH-APRIL 2010 NEW VIEW OF CARCINOGENESIS AND ALTERNATIVE APPROACH TO THERAPY Figu e 1. Cance de elopmen equi es bo h he acquisi ion o DNA al e a ions and a change in he me abolism o oxygen (dysoxic me- abolism). (A) The uncon olled cell p oli e a ion ha cha ac e izes cance equi es signals o cell p oli e a ion and he syn hesis o new mac omolecules ( o example, nucleic acids, lipids, p o eins). Glycolysis p o ides building blocks ( o example, glucose 6-phospha e, dihy- d oxyace one phospha e, 3-phosphoglyce a e, phosphoenolpy u a e, py u a e) ha pa icipa e in he syn hesis o hese mac omole- cules. The p esence o O2can inhibi glycolysis (Pas eu e ec ) and, he e o e, he biosyn hesis o new mac omolecules equi ed o he uncon olled cell p oli e a ion ha cha ac e izes cance . (B) A change in he me abolism o O2(dysoxic me abolism) would allow he ac- i a ion o glycolysis in he p esence o O2and, he e o e, cell p oli e a ion and cance de elopmen . REVIEW ARTICLE MOL MED 16(3-4)144-153, MARCH-APRIL 2010 | LÓPEZ-LÁZARO | 147 he change in he me abolism o O2 (dysoxic me abolism) (Figu e 1B) is c u- cial o he ac i a ion o glycolysis in he p esence o O2(37). B ie ly, high ATP le - els ep ess glycolysis ia allos e ic inhibi- ion o phospho uc okinase, a key en- zyme in he egula ion o glycolysis. The possible basis o he Pas eu e ec is ha he p esence o O2allows ATP syn hesis h ough oxphos, which causes an al- los e ic inhibi ion o phospho uc oki- nase esul ing in he inhibi ion o glycol- ysis. This p oposed mechanism sugges s ha glycolysis is no di ec ly inhibi ed by O2, bu by ATP, and ha he p esence o O2will no cause he inhibi ion o glycol- ysis when O2is no used o gene a e enough ATP (37). The me abolic swi ch om oxphos o glycolysis commonly ob- se ed in cance cells (43,46–48), along wi h he inc eased p oduc ion o O2 •– and H2O2 ound in hese cells (49–53), suppo s he idea ha cance cells ha e his al e a ion in he me abolism o O2. This al e a ion may play a c ucial ole in ca cinogenesis by allowing he ac i a ion o glycolysis in he p esence o O2and, he e o e, he uncon olled cell p oli e a- ion ha cha ac e izes cance (Figu e 1B). A de ia ion o he me abolism o O2 om he pa hway ha gene a es ATP o he pa hway ha gene a es ROS may be necessa y o cell p oli e a ion and umo g ow h (Figu e 1B) (33,37). E idence sug- ges s ha his me abolic swi ch may also play an impo an ole in umo me as a- sis (34). Hypoxia-inducible ac o 1 (HIF-1) is a key egula o o O2homeo- s asis, and he ac i a ion o HIF-1 is known o play a i al ole in he mos ele an aspec s o ca cinogenesis, in- cluding cell su i al, angiogenesis, in a- sion, me as asis, cellula immo aliza ion and me abolic ep og amming (54–56). Because an inc eased p oduc ion o H2O2 (57) and he accumula ion o glycoly ic me aboli es (58) a e known o ac i a e HIF-1, i has been p oposed ha he dysoxic me abolism ep esen ed in Fig- u e 1B may play an impo an ole in he ac i a ion o HIF-1 (32). This dysoxic me- abolism esul s in inc eased p oduc ion o O2 •– and H2O2, and e idence sugges s ha he accumula ion o hese ROS causes oxida i e s ess and plays an im- po an ole in ca cinogenesis (35,36, 59, 60). The key ole o ROS in ca cinogene- sis is suppo ed by expe imen al da a showing ha cance cells commonly ha e inc eased le els o ROS (49–53), ha ROS can induce cell malignan ans o ma ion (61,62) and ha he ma- lignan pheno ype o cance cells can be e e sed by educing he cellula le els o ROS (63–68). O e all, e idence sug- ges s ha he al e a ion in he me abo- lism o O2 ep esen ed in Figu e 1B is a common ea u e o cance cells and may play a key ole in ca cinogenesis (32–38). A NEW MODEL OF CARCINOGENESIS The mos accep ed model o ca cino- genesis pos ula es ha umo igenesis is caused by DNA al e a ions and ha can- ce can be ea ed by e e sing o ex- ploi ing hese al e a ions (Figu e 2A). A new model o ca cinogenesis is p oposed in Figu e 2B. Acco ding o his new model, he de elopmen o any cance equi es ha he u u e umo cell bo h acqui es a complex se o DNA al e - a ions and de elops an al e a ion in he me abolism o O2. I is widely acknowl- edged ha he al e ed genome o umo cells plays a key ole in ca cinogenesis. E idence sugges s ha an al e a ion in he me abolism o O2 om he pa hway ha gene a es ene gy o he pa hway ha p oduces ROS may also play an im- po an ole in he de elopmen o cance (discussed in his e iew). This new model conside s ha bo h al e a ions mus coope a e o he o ma ion o a cance ; he acquisi ion o DNA al e - a ions leads o an al e a ion in he me ab- olism o O2and ice e sa. Indeed, he e is e idence ha he ansi ion om a no - mal o a malignan pheno ype b ough abou by cance -causing genes is associ- a ed wi h a p og essi e ene gy swi ch om oxphos o glycolysis (69). Al e - a ions in p53, one o he mos equen ly mu a ed umo -supp esso genes in can- ce , ha e also been p oposed o pa ici- pa e in he me abolic swi ch om oxphos o glycolysis (48). Mi ochond ial mu a- ions may educe he ac i i y o oxphos and ha e been associa ed wi h an in- c ease in he cellula p oduc ion o ROS (70). The ac i a ion o se e al oncogenes is also known o inc ease he cellula p oduc ion o O2 •– and H2O2(71–74). Con e sely, an al e a ion in he me abo- lism o O2( om he pa hway ha gene - a es ATP o he pa hway ha gene a es O2 •– and H2O2) can lead o he acquisi- ion o DNA al e a ions. H2O2is indeed well known o induce DNA al e a ions, Figu e 2. Models o ca cinogenesis. In addi ion o he acquisi ion a complex a ay o DNA al e a ions p oposed in he accep ed model o ca cinogenesis (A), he model discussed in his e iew (B) p oposes ha cance de elops an al e a ion in he me abolism o oxy- gen. Al hough bo h changes mus in e ac o he de elopmen o cance , he al e ed oxygen me abolism o umo cells is no subjec o he high gene ic complexi y and a i- abili y o umo s and may he e o e be a mo e eliable a ge o cance he apy. 148 | LÓPEZ-LÁZARO | MOL MED 16(3-4)144-153, MARCH-APRIL 2010 NEW VIEW OF CARCINOGENESIS AND ALTERNATIVE APPROACH TO THERAPY including DNA damage, mu a ions and gene ic ins abili y (75–78). Mos ca cinogenic agen s ha e been shown o induce DNA al e a ions. Mos ca cinogenic agen s also induce oxida i e s ess (38,79), and a de ia ion in he me- abolism o O2 owa d he pa hway ha gene a es ROS is a key ea u e o oxida- i e s ess (38,60,79). An example o how a ca cinogenic agen can induce bo h DNA al e a ions and an al e a ion in he me abolism o O2 is shown in Figu e 3. Mos chemical ca cinogens need o be enzyma ically ac i a ed o become geno- oxic, and he cy och ome P450 (P450) enzymes a e he mos p ominen en- zymes in ol ed in such ac i a ion (80). The ac i i y o P450 enzymes is associ- a ed wi h he gene a ion o O2 •– and H2O2(81), and H2O2is well known o in- duce DNA al e a ions (75–78). An in- c ease in he gene a ion o H2O2is associ- a ed wi h he ac i a ion o HIF-1 (32,57), which can lead o he ep ession o ox- phos and he ac i a ion o glycolysis (82,83). The accumula ion o glycoly ic in e media es caused by he ac i a ion o glycolysis can also inc ease he ac i i y o HIF-1 (58) (Figu e 3). Ca cinogenic agen s can induce DNA al e a ions and an al e ed O2me abolism independen ly o P450. Fo ins ance, he inc ease in he in acellula pH induced by some ca - cinogenic agen s seems o be c ucial o he de elopmen o cance (84,85). A ise in he in acellula pH can inc ease he p oduc ion o O2 •– (86) and lead o DNA al e a ions and a dysoxic me abolism h ough he pa hways ep esen ed in Figu e 3. Because exposu e o many o he ca cinogenic ac o s has been asso- cia ed wi h an inc eased p oduc ion o ROS (38,59,79), hese ca cinogenic ac o s may also lead o DNA al e a ions and a dysoxic me abolism h ough he pa h- ways ep esen ed in Figu e 3. Acco ding o he mos accep ed model o ca cinogenesis, he DNA al e a ions o umo cells a e a a ge o cance he - apy (Figu e 2A). Howe e , as discussed in abo e, he high numbe and a iabil- i y o hese DNA al e a ions is an obs a- cle o he design o gene-based he apies ha may ha e a majo impac on cance mo ali y (12,21). The impo ance o he new model o ca cinogenesis ep esen ed in Figu e 2B is ha i o e s an al e na i e a ge o he ea men o cance , which is no subjec o he high gene ic a iabil- i y o umo cells and may he e o e be mo e easily a ge ed. How he al e ed O2 me abolism o cance could be used he - apeu ically o kill umo cells selec i ely is discussed in he ollowing sec ion. TARGETING THE ALTERED OXYGEN METABOLISM OF TUMOR CELLS FOR THE TREATMENT OF CANCER Tumo cells and no mal cells me abo- lize O2di e en ly; his di e ence could be exploi ed o a ge umo cells selec- i ely (Figu e 4). No mal cells ha e ull oxphos capaci y, low p oduc ion o H2O2 and glycolysis inhibi ion in he p esence o O2. No mal cells do no need o main- ain high glycoly ic ac i i y o ensu e hei su i al. Cance cells ha e an al e - Figu e 3. Ca cinogenic agen s can induce DNA al e a ions and an al e a ion in he me- abolism o oxygen. This igu e ep esen s an example o how a ca cinogenic agen can induce bo h DNA al e a ions and a dysoxic me abolism ia P450. See ex o e e ences and u he de ails. Figu e 4. U ili y o he al e ed oxygen me abolism o cance cells o selec i ely kill hem. Cance cells and no mal cells me abolize oxygen di e en ly. Because he basal le els o H2O2a e highe in cance cells han in no mal cells, a speci ic inc ease in he concen a- ions o H2O2may lead o cy o oxic concen a ions in cance cells bu no in no mal cells. In addi ion, because he ac i a ion o glycolysis in cance cells is essen ial o p e en cell dea h induced by ATP deple ion and H2O2accumula ion, he a enua ion o glycolysis in cance cells can induce hei dea h. No mal cells would be less a ec ed by his s a egy, because hey do no need o ha e inc eased glycoly ic a es o ensu e hei su i al. See ex o u he de ails. Do ed lines indica e ha he pa hway o p ocess is ep essed. Bolded lines indica e ha he p ocess is ac i a ed o ha he le els o he molecule a e inc eased. REVIEW ARTICLE MOL MED 16(3-4)144-153, MARCH-APRIL 2010 | LÓPEZ-LÁZARO | 149 a ion in he me abolism o O2(dysoxic me abolism), which is associa ed wi h oxphos ep ession, inc eased p oduc ion o H2O2and inc eased glycoly ic ac i i y. The high glycoly ic ac i i y o cance cells is essen ial o hei su i al, be- cause i p e en s cell dea h induced by ATP deple ion and H2O2accumula ion. The dysoxic me abolism o cance cells can be exploi ed o kill cance cells selec- i ely by inc easing he cellula le els o H2O2and/o by a enua ing glycolysis. These e ec s could be achie ed by he use o p ooxidan agen s and glycolysis inhibi o s, alone o in combina ion. Selec i e Killing o Cance Cells by H2O2and P ooxidan Agen s Recen da a sugges ha oxida i e s ess may play a ole in he an icance ac i i y o many chemo he apeu ic agen s commonly used in cance ea - men , including pacli axel, cispla in, dox- o ubicin, a senic ioxide, bo ezomib, p oca bazine and e oposide (87–101). Fo ins ance, al hough i has been known o many yea s ha he mic o ubule p o ein ubulin is he he apeu ic a ge o pacli- axel ( axol), ecen expe imen s ha e shown ha H2O2plays an impo an ole in pacli axel-induced cance cell dea h (87,89). The ole o ROS in he ac i i y o many an icance agen s is inc easingly being acknowledged, and he induc ion o oxida i e s ess by p ooxidan agen s is eme ging as an a ac i e an icance s a egy (36,94,99,102–106). H2O2seems o be a key playe in ox- ida i e s ess–induced cance cell dea h. Many an icance agen s, such as pacli- axel, doxo ubicin and a senic ioxide, p oduce H2O2(87,90,92), and H2O2is known o be an e icien induce o cell dea h in cance cells (36,93,107). In e es - ingly, cance cells seem mo e suscep ible o H2O2-induced cell dea h han nonma- lignan cells (108–110). In es iga ing se - e al cance and no mal cell lines, Chen e al. (108) obse ed ha high concen a- ions o asco bic acid selec i ely killed cance cells and ha his e ec was medi- a ed by H2O2. They showed, o ins ance, ha a concen a ion o 50 μmol/L H2O2 induced a highe pe cen age o cell dea h in Bu ki lymphoma cells han 250 μmol/L in no mal lymphocy es and no mal monocy es (108). In i o and in i o da a indica e ha umo cells p o- duce highe concen a ions o H2O2 han hei no mal coun e pa s (49–53). These da a, and he ac ha he e is a h eshold o H2O2abo e which cells canno su - i e, may explain why speci ic concen a- ions o H2O2induce selec i e killing o cance cells (36). O e all, e idence sug- ges s ha inc easing he le els o H2O2in cance cells by using p ooxidan agen s may be an impo an he apeu ic s a egy. The concen a ion o a p ooxidan agen equi ed o gene a e le els o H2O2 ha kill cance cells bu no no mal cells could be de e mined in cell cul u e expe - imen s. Then, by using an app op ia e ou e o adminis a ion, such concen a- ions should be achie ed in i o o ob- se e a selec i e an i umo e ec . Selec i e Killing o Cance Cells by Glycolysis Inhibi ion The inc eased glycoly ic ac i i y o cance cells seems o be impo an o keeping adequa e ene gy le els in hese cells. Xu e al. (111) obse ed ha he in- hibi ion o glycolysis se e ely deple ed ATP in cance cells and induced cell dea h, especially in cance cells wi h mi- ochond ial espi a ion de ec s. The de- pendence o cance cells on glycoly ic en- e gy seems o inc ease as malignan ans o ma ion occu s (69). I has been p oposed ha his inc eased dependence on glycolysis o ene gy gene a ion is an impo an me abolic di e ence be ween no mal and malignan cells ha may se e o de eloping he apeu ic s a e- gies o p e e en ially kill cance cells (44,112,113). Se e al glycolysis inhibi o s ha e shown an icance e ec s ( o exam- ple, 2-deoxy-D-glucose, lonidamine, 3-b omopy u a e and dichlo oace a e) and some o hem ha e en e ed he clini- cal ial s age o in es iga ion (37,44,112, 114). Fo example, i has been shown ha dichlo oace a e, a known glycolysis in- hibi o ha has been used in humans o decades in he ea men o lac ic acidosis and inhe i ed mi ochond ial diseases, in- duced ma ked an icance e ec s in mice (115). The au ho s ound ha dichlo oac- e a e in he d inking wa e a clinically ele an doses o up o 3 mon hs p e- en ed and e e sed umo g ow h in i o, wi hou appa en oxici y and wi h- ou a ec ing hemoglobin, ansaminase o c ea inine le els. They concluded ha he ease o deli e y, selec i i y and e ec- i eness o dichlo oace a e make his agen an a ac i e candida e o cance he apy, one ha can be apidly ans- la ed in o phase II–III clinical ials (115). O he s a egies could be used o inhibi o exploi he inc eased glycoly ic ac i - i y o cance cells. Because an inc ease in he ac i i y o he Na+/K+-ATPase pump is associa ed wi h he ac i a ion o gly- colysis (116,117), he inhibi ion o his pump ( o example, by ca diac glyco- sides) may esul in he inhibi ion o gly- colysis and he selec i e killing o cance cells (113,118). The ac i a ion o glycoly- sis is known o inc ease he concen a- ion o p o ons in he cy osol. These p o- ons mus be ex uded o p e en acid-induced cell dea h. The inhibi ion o he cellula sys ems in ol ed in he ex- usion o p o ons in cance cells may also lead o he selec i e killing o cance cells (119). Combina ion o P ooxidan Agen s wi h Glycolysis Inhibi o s o An icance The apy Al hough ROS can induce cance cell dea h, umo cells a e known o de elop mechanisms ha p e en ROS om eaching cy o oxic le els. The glu- a hione and hio edoxin an ioxidan sys ems a e c ucial o de oxi ying ROS. These an ioxidan sys ems a e ac i a ed in cance cells and play an impo an ole in he de elopmen o esis ance o many an icance agen s (120–126). Like- wise, al hough he inhibi ion o glycoly- sis is an a ac i e an icance s a egy, in i o expe imen s sugges ha he inhibi- ion o glycolysis may no be su icien o induce po en an icance e ec s. Ac- co dingly, al hough he glycolysis in- hibi o 2-deoxy-D-glucose is an e icien 150 | LÓPEZ-LÁZARO | MOL MED 16(3-4)144-153, MARCH-APRIL 2010 NEW VIEW OF CARCINOGENESIS AND ALTERNATIVE APPROACH TO THERAPY induce o cell dea h in i o (127), i s an- icance in i o ac i i y is no e y high when i is used as a single agen . The an- icance ac i i y o 2-deoxy-D-glucose has been explo ed in combina ion wi h chemo he apeu ic d ugs and adia ion, and some o hese combina ions ha e en e ed clinical ials (44,112,128–132). P ooxidan agen s could be combined wi h glycolysis inhibi o s o maximize hei an icance ac i i y. E idence indi- ca es ha p ooxidan agen s can inc ease he cellula le els o H2O2and ha gly- colysis inhibi o s can educe he capaci y o cells o de oxi y H2O2. Expe imen al da a ha e shown ha malignan cells a e mo e suscep ible o glucose dep i a ion han non ans o med cells, and ha an inc ease in he le els o H2O2may medi- a e he cy o oxic e ec induced by glu- cose dep i a ion (53,133–135). Two possi- ble mechanisms may explain why he ac i a ion o glycolysis pe o ms an im- po an unc ion in p o ec ing umo cells om H2O2-induced cell dea h. Fi s , he ac i a ion o glycolysis inc eases he o - ma ion o py u a e, which is an e icien sca enge o H2O2(136–139). Second, glucose me abolism h ough he pen ose phospha e pa hway egene a es NADPH om NADP+in a eac ion in which glucose-6-phospha e is con e ed in o 6-phosphogluconolac one by he enzyme glucose-6-phospha e dehyd ogenase. The egene a ion o NADPH is equi ed o H2O2de oxi ica ion h ough he glu- a hione pe oxidase/glu a hione educ- ase sys em and h ough he hio edoxin pe oxidase/ hio edoxin educ ase sys- em (134,140,141) (Figu e 5). CONCLUSIONS Cance kills mo e han six million peo- ple wo ldwide e e y yea (142). The mo ali y a e o his disease has no changed much in he pas ew decades e en in de eloped coun ies such as he Uni ed S a es (29). The small dec eases obse ed in ecen yea s in some ypes o cance (29) a e no a ibu able only o be e he apies, bu also o he imple- men a ion o p e en ion and ea ly de ec- ion campaigns. The goal o hese cam- paigns is o make people awa e ha many cance s can be p e en ed by ol- lowing se e al guidelines, and ha can- ce he apy is e ec i e when his disease is de ec ed ea ly (30,31). Despi e hese campaigns, many cance s a e s ill diag- nosed when cells om a p ima y umo ha e al eady me as asized o o he pa s o he body. A his s age o disease, umo cells a e no longe localized and can no longe be elimina ed by su ge y o adio he apy. The main o m o ea - men a his poin is chemo he apy, which consis s o deli e ing d ugs sys- emically so ha hey can each and kill he umo cells. Bu mos o hese d ugs a e oxic o bo h umo and no mal cells, cause se e e side e ec s in pa ien s and, he e o e, need o be used a subop imal le els. The low e icacy o chemo he apy in pa ien s wi h ad anced cance s is e- lec ed in he low 5-yea su i al a es obse ed in hese pa ien s (29). Fo in- s ance, cance s a is ics show ha he mos commonly diagnosed cance in he wo ld is lung cance (142), ha app oxi- ma ely 50% o pa ien s diagnosed wi h lung cance ha e dis an me as asis (29) and ha only 3% o hese pa ien s man- age o su i e mo e han 5 yea s (29). The low e icacy o cance he apy o he ea men o pa ien s wi h me as asis makes he de elopmen o no el he a- peu ic app oaches necessa y. A no el he apeu ic app oach has eme ged s ongly in ecen yea s. This app oach seeks o a ack he umo cells selec i ely and is based on unde s and- ing o he di e ences be ween umo cells and nonmalignan cells. I has been known o many yea s ha umo cells ha e gene ic al e a ions and much e- sea ch has been done o iden i y hese al- e a ions. Recen analyses o human can- ce s ha e e ealed, howe e , ha he gene ic de ec s o umo cells a e much Figu e 5. Key ole o glycolysis in he de oxi ica ion o H2O2. Inc eased glucose me abolism helps de oxi y H2O2by inc easing he le els o he H2O2sca enge py u a e and by e- gene a ing NADPH. Glu a hione educ ase (GR) and hio edoxin educ ase (T xR) need NADPH o egene a e glu a hione (GSH) and hio edoxin [T x(SH)2], which a e used by glu- a hione pe oxidase (GPx) and hio edoxin pe oxidase (TPx) o de oxi y H2O2. Thiol (SH)- eac i e agen s can eac wi h he SH g oups o GSH and T x(SH)2and induce a p ooxi- dan e ec by dis up ing he GR/GPx and T xR/TPx an ioxidan sys ems. REVIEW ARTICLE MOL MED 16(3-4)144-153, MARCH-APRIL 2010 | LÓPEZ-LÁZARO | 151 mo e nume ous and uns able han ex- pec ed (12,13). In addi ion, he gene ic al- e a ions o umo cells a e no he same in di e en ypes o cance o e en in di e en people wi h he same ype o cance . Gi en hese ci cums ances i will p obably be e y di icul o de elop u- u e gene-based he apies ha may be use ul in a wide ange o pa ien s wi h cance (12,20,21). Despi e he complexi y o he cance genome, much esea ch is de o ed o cha ac e izing he gene ic p o ile o umo s o a ionalize and pe - sonalize cance he apy (22–24). An al e na i e app oach has been dis- cussed in his e iew. In addi ion o build- ing up a complex se o DNA changes, e - idence sugges s ha he de elopmen o any cance equi es ha umo cells ac- qui e an al e a ion in he me abolism o oxygen. In e es ingly, his al e a ion in he me abolism o oxygen can make cance cells ulne able o he apeu ic in e en- ion. Thei inc eased basal le els o H2O2 and hei highe dependence on glycoly- sis o hei su i al make cance cells mo e suscep ible han no mal cells o ea men wi h p ooxidan agen s and glycolysis inhibi o s. Because his al e - a ion in he me abolism o oxygen seems o be a common ea u e o umo cells, his he apeu ic app oach could be used o he ea men o a wide ange o pa- ien s wi h cance . Fu u e esea ch will e- eal whe he his al e na i e app oach will be su icien o inc ease he su i al o pa ien s wi h ad anced cance s, o whe he i will be necessa y o use i in combina ion wi h adi ional chemo he - apy and/o no el a ge ed he apies. DISCLOSURE The au ho s decla e ha hey ha e no compe ing in e es s as de ined by Molecu- la Medicine, o o he in e es s ha migh be pe cei ed o in luence he esul s and discussion epo ed in his pape . REFERENCES 1. Na u e miles ones in cance . (2006) Na . Re . Can- ce . 6:S8–9. 2. No dling CO. (1953) A new heo y on cance - inducing mechanism. B . J. Cance . 7:68–72. 3. Fa don JC. (1953) A econside a ion o he soma ic mu a ion heo y o cance in he ligh o some ecen de elopmen s. Science. 117:441–5. 4. A mi age P, Doll R. (1954) The age dis ibu ion o cance and a mul i-s age heo y o ca cinogene- sis. B . J. Cance . 8:1–12. 5. Nowell PC. (1976) The clonal e olu ion o umo cell popula ions. Science. 194:23–8. 6. Fea on ER, Vogels ein B. (1990) A gene ic model o colo ec al umo igenesis. Cell. 61:759–67. 7. Reddy EP, Reynolds RK, San os E, Ba bacid M. (1982) A poin mu a ion is esponsible o he ac- quisi ion o ans o ming p ope ies by he T24 human bladde ca cinoma oncogene. Na u e. 300:149–52. 8. Pa ada LF, Tabin CJ, Shih C, Weinbe g RA. (1982) Human EJ bladde ca cinoma oncogene is homo- logue o Ha ey sa coma i us as gene. Na u e. 297:474–8. 9. Gold a b M, Shimizu K, Pe ucho M, Wigle M. (1982) Isola ion and p elimina y cha ac e iza ion o a human ans o ming gene om T24 bladde ca cinoma cells. Na u e. 296:404–9. 10. Bishop JM. (1987) The molecula gene ics o can- ce . Science. 235:305–11. 11. S anb idge EJ. (1990) Iden i ying umo supp es- so genes in human colo ec al cance . Science. 247:12–3. 12. Folkman J, Hahn eld P, Hla ky L. (2000) Cance : looking ou side he genome. Na . Re . Mol. Cell Biol. 1:76–9. 13. S ole DL, e al. (1999) The onse and ex en o ge- nomic ins abili y in spo adic colo ec al umo p og ession. P oc. Na l. Acad. Sci. U. S. A. 96:15121–6. 14. Iacobuzio-Donahue CA. (2009) Epigene ic changes in cance . Annu. Re . Pa hol. 4:229–49. 15. P ehn RT. (1994) Cance s bege mu a ions e sus mu a ions bege cance s. Cance Res. 54:5296–300. 16. Ja e LF. (2003) Epigene ic heo ies o cance ini i- a ion. Ad . Cance Res. 90:209–30. 17. Rasnick D, Duesbe g PH. (1999) How aneu- ploidy a ec s me abolic con ol and causes can- ce . Biochem. J. 340:621–30. 18. Rajagopalan H, Lengaue C. (2004) Aneuploidy and cance . Na u e. 432:338–41. 19. Hanahan D, Weinbe g RA. (2000) The hallma ks o cance . Cell. 100:57–70. 20. Hahn WC, Weinbe g RA. (2002) Rules o mak- ing human umo cells. N. Engl. J. Med. 347:1593–603. 21. Vogels ein B, Kinzle KW. (2004) Cance genes and he pa hways hey con ol. Na . Med. 10:789–99. 22. Ga man KS, Ne ins JR, Po i A. (2007) Genomic s a egies o pe sonalized cance he apy. Hum. Mol. Gene . 16:R226–32. 23. an′ Vee LJ, Be na ds R. (2008) Enabling pe - sonalized cance medicine h ough analysis o gene-exp ession pa e ns. Na u e. 452:564–70. 24. Hayden EC. (2009) Pe sonalized cance he apy ge s close . Na u e. 458:131–2. 25. Ma WW, Adjei AA. (2009) No el agen s on he ho izon o cance he apy. CA Cance J. Clin. 59:111–37. 26. D uke BJ, e al. (2001) E icacy and sa e y o a speci ic inhibi o o he BCR-ABL y osine kinase in ch onic myeloid leukemia. N. Engl. J. Med. 344:1031–7. 27. Hehlmann R, Hochhaus A, Bacca ani M. (2007) Ch onic myeloid leukaemia. Lance . 370:342–50. 28. Melo JV, Ba nes DJ. (2007) Ch onic myeloid leukaemia as a model o disease e olu ion in human cance . Na . Re . Cance . 7:441–53. 29. Jemal A, e al. (2009) Cance s a is ics, 2009. CA Cance J. Clin. 59:225–49. 30. Kushi LH, e al. (2006) Ame ican Cance Socie y Guidelines on Nu i ion and Physical Ac i i y o cance p e en ion: educing he isk o cance wi h heal hy ood choices and physical ac i i y. CA Cance J. Clin. 56:254–81. 31. Smi h RA, Cokkinides V, Ey e HJ. (2006) Ame i- can Cance Socie y guidelines o he ea ly de ec- ion o cance , 2006. CA Cance J. Clin. 56:11–25. 32. Lopez-Laza o M. (2006) HIF-1: Hypoxia-in- ducible ac o o dysoxia-inducible ac o ? FASEB J. 20:828–32. 33. Lopez-Laza o M. (2006) Does hypoxia eally con- ol umo g ow h? Cell Oncol. 28:327–9. 34. Lopez-Laza o M. (2007) Why do umo s me as a- size? Cance Biol. The . 6:141–4. 35. Lopez-Laza o M. (2007) Excessi e supe oxide anion gene a ion plays a key ole in ca cinogene- sis. In . J. Cance . 120:1378–80. 36. Lopez-Laza o M. (2007) Dual ole o hyd ogen pe oxide in cance : possible ele ance o cance chemop e en ion and he apy. Cance Le . 252:1–8. 37. Lopez-Laza o M. (2008) The Wa bu g e ec : why and how do cance cells ac i a e glycolysis in he p esence o oxygen? An icance Agen s Med. Chem. 8:305–12. 38. Lopez-Laza o M. (2009) Role o oxygen in cance : looking beyond hypoxia. An icance Agen s Med. Chem. 9:517–25. 39. Ma hupala SP, Ko YH, Pede sen PL. (2009) Hex- okinase-2 bound o mi ochond ia: cance ’s s y- gian link o he “Wa bu g E ec ” and a pi o al a ge o e ec i e he apy. Semin. Cance Biol. 19:17–24. 40. Chesney J, e al. (1999) An inducible gene p od- uc o 6-phospho uc o-2-kinase wi h an AU- ich ins abili y elemen : ole in umo cell glycol- ysis and he Wa bu g e ec . P oc. Na l. Acad. Sci. U. S. A. 96:3047–52. 41. Ch is o k HR, e al. (2008) The M2 splice iso o m o py u a e kinase is impo an o cance me- abolism and umou g ow h. Na u e. 452:230–3. 42. Ga enby RA, Gillies RJ. (2004) Why do cance s ha e high ae obic glycolysis? Na . Re . Cance . 4:891–9. 43. Wa bu g O. (1956) On he o igin o cance cells. Science. 123:309–14. 44. Chen Z, Lu W, Ga cia-P ie o C, Huang P. (2007) The Wa bu g e ec and i s cance he apeu ic im- plica ions. J. Bioene g. Biomemb . 39:267–74. 45. Vande Heiden MG, Can ley LC, Thompson CB. (2009) Unde s anding he Wa bu g e ec : he 152 | LÓPEZ-LÁZARO | MOL MED 16(3-4)144-153, MARCH-APRIL 2010 NEW VIEW OF CARCINOGENESIS AND ALTERNATIVE APPROACH TO THERAPY me abolic equi emen s o cell p oli e a ion. Sci- ence. 324:1029–33. 46. Isido o A, e al. (2004) Al e a ion o he bioene - ge ic pheno ype o mi ochond ia is a hallma k o b eas , gas ic, lung and oesophageal cance . Biochem. J. 378:17–20. 47. Unwin RD, e al. (2003) P o eomic changes in enal cance and co-o dina e demons a ion o bo h he glycoly ic and mi ochond ial aspec s o he Wa bu g e ec . P o eomics. 3:1620–32. 48. Ma oba S, e al. (2006) p53 egula es mi ochond - ial espi a ion. Science. 312:1650–3. 49. Sza owski TP, Na han CF. (1991) P oduc ion o la ge amoun s o hyd ogen pe oxide by human umo cells. Cance Res. 51:794–8. 50. Bu don RH. (1995) Supe oxide and hyd ogen pe oxide in ela ion o mammalian cell p oli e a- ion. F ee Radic. Biol. Med. 18:775–94. 51. Zieba M, e al. (2000) Compa ison o hyd ogen pe oxide gene a ion and he con en o lipid pe - oxida ion p oduc s in lung cance issue and pul- mona y pa enchyma. Respi . Med. 94:800–5. 52. Lim SD, e al. (2005) Inc eased Nox1 and hyd o- gen pe oxide in p os a e cance . P os a e. 62:200–7. 53. Aykin-Bu ns N, Ahmad IM, Zhu Y, Obe ley LW, Spi z DR. (2009) Inc eased le els o supe oxide and H2O2 media e he di e en ial suscep ibili y o cance cells e sus no mal cells o glucose dep i a ion. Biochem. J. 418:29–37. 54. Semenza GL. (2003) Ta ge ing HIF-1 o cance he apy. Na . Re . Cance . 3:721–32. 55. Semenza GL. (2007) E alua ion o HIF-1 inhibi o s as an icance agen s. D ug Disco . Today. 12:853–9. 56. Lopez-Laza o M. (2006) Hypoxia-inducible ac o 1 as a possible a ge o cance chemop e en- ion. Cance Epidemiol. Bioma ke s P e . 15:2332–5. 57. Chandel NS, e al. (2000) Reac i e oxygen species gene a ed a mi ochond ial complex III s abilize hypoxia-inducible ac o -1alpha du ing hypoxia: a mechanism o O2 sensing. J. Biol. Chem. 275:25130–8. 58. Lu H, Fo bes RA, Ve ma A. (2002) Hypoxia- inducible ac o 1 ac i a ion by ae obic glycoly- sis implica es he Wa bu g e ec in ca cinogene- sis. J. Biol. Chem. 277:23111–5. 59. Ce u i PA. (1985) P ooxidan s a es and umo p omo ion. Science. 227:375–81. 60. Klaunig JE, Kamendulis LM. (2004) The ole o oxida i e s ess in ca cinogenesis. Annu. Re . Pha macol. Toxicol. 44:239–67. 61. Okamo o M, Kawai K, Rezniko CA, Oyasu R. (1996) T ans o ma ion in i o o a non umo i- genic a u o helial cell line by hyd ogen pe ox- ide. Cance Res. 56:4649–53. 62. Suh YA, e al. (1999) Cell ans o ma ion by he supe oxide-gene a ing oxidase Mox1. Na u e. 401:79–82. 63. A nold RS, e al. (2001) Hyd ogen pe oxide medi- a es he cell g ow h and ans o ma ion caused by he mi ogenic oxidase Nox1. P oc. Na l. Acad. Sci. U. S. A. 98:5550–5. 64. Chu ch SL, e al. (1993) Inc eased manganese su- pe oxide dismu ase exp ession supp esses he malignan pheno ype o human melanoma cells. P oc. Na l. Acad. Sci. U. S. A. 90:3113–7. 65. Sa o d SE, Obe ley TD, U ano M, S Clai DK. (1994) Supp ession o ib osa coma me as asis by ele a ed exp ession o manganese supe oxide dismu ase. Cance Res. 54:4261–5. 66. Yan T, Obe ley LW, Zhong W, S Clai DK. (1996) Manganese-con aining supe oxide dismu ase o e exp ession causes pheno ypic e e sion in SV40- ans o med human lung ib oblas s. Can- ce Res. 56:2864–71. 67. Zhang Y, Zhao W, Zhang HJ, Domann FE, Obe - ley LW. (2002) O e exp ession o coppe zinc su- pe oxide dismu ase supp esses human glioma cell g ow h. Cance Res. 62:1205–12. 68. Hyoudou K, e al. (2004) Inhibi ion o me as a ic umo g ow h in mouse lung by epea ed admin- is a ion o polye hylene glycol-conjuga ed ca a- lase: quan i a i e analysis wi h i e ly luci e ase- exp essing melanoma cells. Clin. Cance Res. 10:7685–91. 69. Ramana han A, Wang C, Sch eibe SL. (2005) Pe u ba ional p o iling o a cell-line model o u- mo igenesis by using me abolic measu emen s. P oc. Na l. Acad. Sci. U. S. A. 102:5992–7. 70. Ishikawa K, e al. (2008) ROS-gene a ing mi o- chond ial DNA mu a ions can egula e umo cell me as asis. Science. 320:661–4. 71. I ani K, e al. (1997) Mi ogenic signaling medi- a ed by oxidan s in Ras- ans o med ib oblas s. Science. 275:1649–52. 72. Va a O, e al. (2002) c-Myc can induce DNA dam- age, inc ease eac i e oxygen species, and mi i- ga e p53 unc ion: a mechanism o oncogene-in- duced gene ic ins abili y. Mol. Cell. 9:1031–44. 73. Sa le M, e al. (2000) The BCR/ABL y osine ki- nase induces p oduc ion o eac i e oxygen species in hema opoie ic cells. J. Biol. Chem. 275:24273–8. 74. Kopnin PB, Agapo a LS, Kopnin BP, Chumako PM. (2007) Rep ession o ses in amily genes con ibu es o oncogenic Ras-induced eac i e oxygen species up- egula ion and gene ic ins a- bili y. Cance Res. 67:4671–8. 75. Pa k S, You X, Imlay JA. (2005) Subs an ial DNA damage om submic omola in acellula hyd ogen pe oxide de ec ed in Hpx- mu an s o Esche ichia coli. P oc. Na l. Acad. Sci. U. S. A. 102:9317–22. 76. Radisky DC, e al. (2005) Rac1b and eac i e oxy- gen species media e MMP-3-induced EMT and genomic ins abili y. Na u e. 436:123–7. 77. Dayal D, Ma in SM, Limoli CL, Spi z DR. (2008) Hyd ogen pe oxide media es he adia ion-in- duced mu a o pheno ype in mammalian cells. Biochem. J. 413:185–91. 78. Jackson AL, Loeb LA. (2000) Mic osa elli e ins a- bili y induced by hyd ogen pe oxide in Es- che ichia coli. Mu a . Res. 447:187–98. 79. Ko acic P, Jacin ho JD. (2001) Mechanisms o ca - cinogenesis: ocus on oxida i e s ess and elec- on ans e . Cu . Med. Chem. 8:773–96. 80. Guenge ich FP, Shimada T. (1998) Ac i a ion o p oca cinogens by human cy och ome P450 en- zymes. Mu a . Res. 400:201–13. 81. Zanga RC, Da ydo DR, Ve ma S. (2004) Mech- anisms ha egula e p oduc ion o eac i e oxy- gen species by cy och ome P450. Toxicol. Appl. Pha macol. 199:316–31. 82. Papand eou I, Cai ns RA, Fon ana L, Lim AL, Denko NC. (2006) HIF-1 media es adap a ion o hypoxia by ac i ely down egula ing mi ochond - ial oxygen consump ion. Cell Me ab. 3:187–97. 83. Semenza GL. (2007) HIF-1 media es he Wa bu g e ec in clea cell enal ca cinoma. J. Bioene g. Biomemb . 39:231–4. 84. Reshkin SJ, e al. (2000) Na+/H+ exchange - dependen in acellula alkaliniza ion is an ea ly e en in malignan ans o ma ion and plays an essen ial ole in he de elopmen o subsequen ans o ma ion-associa ed pheno ypes. FASEB J. 14:2185–97. 85. Ha guindey S, O i e G, Luis PJ, Pa adiso A, Reshkin SJ. (2005) The ole o pH dynamics and he Na(+)/H(+) an ipo e in he e iopa hogene- sis and ea men o cance . Two aces o he same coin-one single na u e. Biochim. Biophys. Ac a. 1756:1–24. 86. Simchowi z L. (1985) In acellula pH modula es he gene a ion o supe oxide adicals by human neu ophils. J. Clin. In es . 76:1079–89. 87. Alexand e J, e al. (2006) Accumula ion o hyd o- gen pe oxide is an ea ly and c ucial s ep o pa- cli axel-induced cance cell dea h bo h in i o and in i o. In . J. Cance 119:41–8. 88. Alexand e J, e al. (2006) Imp o emen o he he apeu ic index o an icance d ugs by he su- pe oxide dismu ase mimic manga odipi . J. Na l. Cance Ins . 98:236–44. 89. Alexand e J, Hu Y, Lu W, Pelicano H, Huang P. (2007) No el ac ion o pacli axel agains cance cells: bys ande e ec media ed by eac i e oxy- gen species. Cance Res. 67:3512–7. 90. Jing Y, Dai J, Chalme s-Redman RM, Ta on WG, Waxman S. (1999) A senic ioxide selec i ely in- duces acu e p omyelocy ic leukemia cell apop o- sis ia a hyd ogen pe oxide-dependen pa hway. Blood. 94:2102–11. 91. Mizu ani H, Tada-Oikawa S, Hi aku Y, Kojima M, Kawanishi S. (2005) Mechanism o apop osis induced by doxo ubicin h ough he gene a ion o hyd ogen pe oxide. Li e Sci. 76:1439–53. 92. Ubezio P, Ci oli F. (1994) Flow cy ome ic de ec- ion o hyd ogen pe oxide p oduc ion induced by doxo ubicin in cance cells. F ee Radic. Biol. Med. 16:509–16. 93. Ikeda K, e al. (1999) In ol emen o hyd ogen pe oxide and hyd oxyl adical in chemically in- duced apop osis o HL-60 cells. Biochem. Pha ma- col. 57:1361–5. 94. Fang J, Nakamu a H, Iye AK. (2007) Tumo - a ge ed induc ion o oxys ess o cance he - apy. J. D ug Ta ge . 15:475–86. 95. Simizu S, Takada M, Umezawa K, Imo o M. (1998) Requi emen o caspase-3(-like) p o ease- media ed hyd ogen pe oxide p oduc ion o apop osis induced by a ious an icance d ugs. J. Biol. Chem. 273:26900–7.