Respi a o y Physiology & Neu obiology 132 (2002) 17–41
Signi icance o ROS in oxygen sensing in cell sys ems wi h
sensi i i y o physiological hypoxia
Cons ancio Gonzalez *, Glo ia Sanz-Al aya e, M. Te esa Agapi o,
Angela Gomez-Nin˜o, Asuncio´n Roche , Ana Obeso
Depa amen o de Bioquı´mica y Biologia Molecula y Fisiologı´a,Facul ad de Medicina,
Ins i u o de Biologı´a y Gene´ ica Molecula
(
IBGM
)
,Uni6e sidad de Valladolid y CSIC,
47005
Valladolid,Spain
Accep ed 28 Augus 2001
Abs ac
Reac i e oxygen species (ROS) a e oxygen-con aining molecula en i ies which a e mo e po en and e ec i e
oxidizing agen s han is molecula oxygen i sel . Wi h he excep ion o phagocy ic cells, whe e ROS play an impo an
physiological ole in de ense eac ions, ROS ha e classically been conside ed undesi able byp oduc s o cell
me abolism, exis ing se e al cellula mechanisms aimed o dispose hem. Recen ly, howe e , ROS ha e been
conside ed impo an in acellula signaling molecules, which may ac as media o s o second messenge s in many cell
unc ions. This is he p oposed ole o ROS in oxygen sensing in sys ems, such as ca o id body chemo ecep o cells,
pulmona y a e y smoo h muscle cells, and e y h opoie in-p oducing cells. These unique cells comp ise essen ial pa s
o homeos a ic loops di ec ed o main ain oxygen le els in mul icellula o ganisms in si ua ions o hypoxia. The
p esen a icle examines he possible signi icance o ROS in hese h ee cell sys ems, and p oposes a se o c i e ia ha
ROS should sa is y o hei conside a ion as media o s in hypoxic ansduc ion cascades. In none o he h ee cell
ypes do ROS sa is y hese c i e ia, and hus i appea s ha al e na i e mechanisms a e esponsible o he
ansduc ion cascades linking hypoxia o he elease o neu o ansmi e s in chemo ecep o cells, con ac ion in
pulmona y a e y smoo h muscle cells and e y h opoie in sec e ion in e y h opoie in p oducing cells. © 2002 Else ie
Science B.V. All igh s ese ed.
Keywo ds
:
Ca o id body, signal ansduc ion; Con ol o b ea hing, O
2
sensing; Media o s, e y h opoie in; Media o s, ROS; Oxygen,
cellula sensing; Signal ansduc ion, ca o id body, e y h opoie in cell, pulmona y ascula u e
www.else ie .com/loca e/ esphysiol
1. Cell sys ems wi h sensi i i y o physiological
hypoxia
The appea ance o O
2
on a p e iously anae o-
bic ea h, o ced li ing o ganisms o de elop an-
ioxidan de ense sys ems. O ganisms ha e ol ed
ole ance o O
2
also de eloped mechanisms o use
his new elemen o me abolic ans o ma ions
* Co esponding au ho . Tel.: +34-983-423-089; ax: +34-
983-423-588
E-mail add ess
:
[email p o ec ed] (C. Gonzalez).
1569-9048/02/$ - see on ma e © 2002 Else ie Science B.V. All igh s ese ed.
PII: S1569-9048(02)00047-2
C.Gonzalez e al.
/
Respi a o y Physiology & Neu obiology
132 (2002) 17–41
18
and e icien ene gy p oduc ion by using elec on
anspo chains wi h O
2
as he inal elec on
accep o . The de elopmen o e icien and secu e
ene gy p oduc ion sys ems allowed he appea ance
o mul icellula o ganisms. The acquisi ion o O
2
and nu ien s by di usion om he su ounding
aqueous medium was succeeded in mul icellula
o ganisms by he appea ance o a ci cula o y
sys em, a sys em o ubes o ca y and o dis ibu e
O
2
and nu ien s among all cells o he li ing
o ganism. E olu ion la e p o ided an oxygena ing
o espi a o y sys em capable o supplying O
2
o he
luid mo ing in he ci cula o y sys em, and he
mo ing luid i sel e ol ed o blood o gain in
capaci y o anspo O
2
. All hese acquisi ions
p o ided li ing o ganisms wi h he independence
necessa y o abandon wa e and o mo e eely on
land. Ne e heless, ull acquisi ion o e es ial
eedom was only possible by he simul aneous
appea ance o homeos a ic mechanisms. Some o
hose homeos a ic mechanisms ma ch O
2
supply o
whole-body ene ge ic needs, by moni o ing O
2
le els in he body, and al e ing he o e all unc ion
o he espi a o y and ci cula o y sys ems, as well
as he capaci y o blood o anspo O
2
.
The e a e wo main sys ems ega ding O
2
homeos asis. They con o m wo unc ional loops
wi h hei o igin in he O
2
-moni o ing cells, he
chemo ecep o cells o he ca o id body (CB) and
he e y h opoie in (EPO)-p oducing cells (Richale ,
1997; Gonzalez, 1998). These wo cell ypes a e
ecep o –e ec o uni s, which de ec a e ial blood
O
2
le els and espond wi h a sec e o y esponse.
The unc ional loop o igina ed in he CB wo ks as
ollows: Chemo ecep o s cells de ec a e ial blood
P
O
2
and become ac i a ed when a e ial P
O
2
de-
c eases; hey elease neu o ansmi e s which ac i-
a e he senso y ne e endings o he ca o id sinus
ne e, whose cen al p ojec ions end in he nucleus
o he ac us soli a ius in he medulla. In eg a ion
o he ca o id sinus ne e inpu in he espi a o y
con ol cen e s o he b ain s em esul s in an
inc eased ac i i y o he espi a o y muscles, wi h
inc eased en ila ion and inc eased al eola and
a e ial blood P
O
2
(Gonzalez e al., 1992, 1994).
EPO-p oducing cells, which in adul animals a e
mainly loca ed in he kidney, also de ec a e ial
blood O
2
le els and a e ac i a ed by hypoxia o
elease EPO; his glycop o eic ho mone eaches he
bone ma ow and ac i a es e y h opoiesis. An
inc ease in he ed blood cell mass and ul ima ely
in he capaci y o he blood o anspo O
2
ollow
(Jelkmann, 1992). These wo sys ems p obably a e
aided in main aining he homeos asis o O
2
by he
pulmona y a e y smoo h muscle cells (PASMC),
which, as he o he cell wo ypes, moni o oxygen
le els and eac o hypoxia wi h a con ac ile
esponse leading o asocons ic ion and lung
blood low edis ibu ion which would amelio a e
he oxygena ion o blood (Ma shall e al., 1994).
We ha e p e iously de ined physiological hypoxia
and hese h ee cell ypes as endowed wi h sensi i -
i y o physiological hypoxia (Gonzalez, 1998).
These h ee cell sys ems, con ibu ing o he
homeos asis o O
2
, sha e o he p ope ies. They
ha e a e y low h eshold o hypoxia, becoming
ac i a ed a an a e ial P
O
2
o :70 mmHg and
inc easing hei e ec o ac i i y wi h he in ensi y
o hypoxia. In con as , he esponse o hypoxia in
mos cells is a dec ease in hei ac i i y (Hochachka
e al., 1997) and as he hypoxic h eshold o his
beha io is highe , a mo e in ense hypoxia is needed
o elici i . The di e en beha io o he cell ypes
in ol ed in homeos a ic loops e sus gene al cells
o he o ganism allows o p edic ha he de ec ion
o O
2
le els (i.e. (O
2
-sensing), i linked o O
2
-con-
sump ion a e, should be associa ed o an inc ease
o O
2
-consump ion in he case o he cells in ol ed
in homeos a ic loops and o a dec ease in he es
o he cells. The inc ease in O
2
-consump ion du ing
hypoxia in he o me cells is an absolu e equi e-
men o suppo he inc eased ac i i y hey exhibi
in hese ci cums ances; he dec ease in O
2
-con-
sump ion du ing hypoxia in he es o he cells is
in e ed om he nea shu down o all ene gy- e-
qui ing p ocesses. Al e na i ely, O
2
-sensing could
be independen o O
2
-consump ion a e in bo h
ca ego ies o cells. Wha e e he case, he e a e
expe imen al obse a ions suppo ing he con-
en ion ha he cell ypes in ol ed in homeos a ic
loops, as exempli ied by he CB chemo ecep o
cells, exhibi an inc eased me abolic a e and O
2
-
consump ion du ing hypoxia. Thus, ATP le els
a e no signi ican ly di e en in no moxia and
mode a e hypoxia (Ve na e al., 1990), and a he
same ime, hypoxia o compa able in ensi y in-
C.Gonzalez e al.
/
Respi a o y Physiology & Neu obiology
132 (2002) 17–41
19
Fig. 1. Redox models o oxygen sensing in he h ee cells sys ems loca ed a he o igin o oxygen homeos a ic loops. The op la e al
a ows wi h ques ion ma ks ep esen al e na i e oxygen sensing models (e.g. hemop o ein sensing) no discussed in his a icle.
c eases glucose oxida ion o CO
2
in an ouabain-
dependen manne (Obeso e al., 1989b, 1993).
An al e na i e way o s a ing he no ions ex-
p essed in p e ious pa ag aphs is o say ha in
highe o ganisms he e a e a ew cell sys ems
de o ed o main aining O
2
le els in he en i e
o ganism which appea o ha e a unique mecha-
nism o O
2
-sensing. The uniqueness o his O
2
-
sensing mechanism e e s o i s h eshold and o
i s abili y o become coupled o and o ac i a e
cell p ocesses demanding an ene gy supply. This
a icle will conside he edox models o O
2
-sens-
ing. We shall discuss he a ailable in o ma ion in
a o and agains he pa icipa ion o oxygen
eac i e species (ROS) in O
2
-sensing and he pos-
sible signi icance o ROS as coupling ac o s be-
ween hypoxia and he e ec o machine y o
he cells in ol ed in he homeos asis o oxygen
(Fig. 1).
2. F ee adicals, oxygen eac i e species (ROS)
and allied molecules
2
.
1
.P ima y ee adicals
F ee adicals a e chemically de ined as a oms o
molecules ha ha e one o mo e unpai ed elec-
on; hey usually a e e y eac i e (Puncha d and
Kelly, 1996; Halliwell and Gu e idge, 1999). The
p ima y p ocesses ha gi e ise o ee adicals in
biological sys ems a e di ided in o wo main ca e-
go ies. Due o he high wa e con en o li ing
o ganisms, he clea age o he chemical bonds o
wa e by high-ene gy adia ion (wa e adiolysis)
cons i u es an impo an sou ce o ee adicals
acco ding o he eac ion:
H
2
OH(hyd ogen adical)
+OH(hyd oxyl adical)
C.Gonzalez e al.
/
Respi a o y Physiology & Neu obiology
132 (2002) 17–41
20
Ul a iole ligh and e en isible ligh in he
p esence o some sensi ize molecules can simi-
la ly p oduce homoly ic scission o hyd ogen pe -
oxide in he skin o gene a e hyd oxyl adicals
(H
2
O
2
2OH). Mo e impo an , howe e , (a
leas quan i a i ely) is he genesis o ee adicals
by eac ions in ol ing he ans e ence o elec-
ons, and among hem he mos impo an a e
hose eac ions in which a single elec on is ans-
e ed o a molecule o O
2
o o m he supe oxide
adical, O
2
−
(Da ies and Dean, 1997). Phagocy es
possess an enzyma ic complex con aining FAD
and a speci ic b- ype cy och ome capable o ak-
ing an elec on om NADPH and ans e ing i
o O
2
o o m O
2
−
. This is p obably he only
eac ion in which O
2
−
is a bene icial p oduc ; his
adical and o he ROS o med om i a e used by
phagocy ic cells o kill bac e ia. The e a e, how-
e e , many unwan ed p ocesses p oducing O
2
−
as
a byp oduc , some o which a e b ie ly com-
men ed he e. The e a e au o-oxida ion eac ions
in which se e al biologically impo an molecules
such as y osine, ca echolamines and yp ophan
slowly educe O
2
o o m O
2
−
.O
2
−
in u n is
capable o u he oxidizing hose molecules and a
cascade o eac ions migh be s a ed wi h po en-
ially dele e ious e ec s. Mic osomal enzymes in-
ol ed in eac ions o oxida ion and
hyd oxyla ion o many ypes o molecules (alco-
hols, ba bi u a es, an ibio ics, s e oid ho mone
syn hesis and me abolism, e c.), which use cy-
och ome P450 as an elec on anspo sys em,
can also gene a e O
2
−
. Supe oxide adical is also
o med in he anspo o O
2
by hemoglobin.
Al hough essen ially all O
2
binding o hemoglobin
(and myoglobin) does no imply he oxida ion o
Fe
2+
in he heme moie y, he e is, howe e , a
ce ain deg ee o delocaliza ion o an elec on
when O
2
is bound and he eby he appea ance o
some kind o in e media e s uc u e ha can be
w i en as an appa en esonan ‘equilib ium’:
hemo-Fe
2+
–O
2
lhemo-Fe
3+
–O
2
−
. Usually, he
Fe
2+
o heme g oup keeps he delocalized elec-
on and hemoglobin eleases O
2
, bu occasionally
(3%/day) he elec on is kep by O
2
and
hemoglobin hen eleases O
2
−
, he Fe
2+
o he
heme g oup is ans o med in o Fe
3+
and
me hemoglobin is o med. In si ua ions o issue
inju y, especially in he p ocesses o ischemia-
epe usion, he enzyme xan hine dehyd ogenase
which oxidizes hypoxan hine and xan hine o u ic
acid using NAD
+
as accep o o elec ons, is
modi ied by oxida i e o p o eoly ic p ocesses and
con e ed in o xan hine oxidase, which is capable
o ans e ing he elec ons o he pu ines o O
2
o o m O
2
−
. Howe e , he e is no doub ha
quan i a i ely speaking he mos impo an sou ce
o O
2
−
is he mi ochond ial elec on anspo
chain, whe e uni alen educ ion o O
2
may ake
place in some s eps o he o e all anspo sys em
(Fig. 2). In he wo places whe e O
2
−
can be
p oduced (complex I and ubiquinone pool) he e
a e elec on shu le molecules ( la in adenin
mononucleo ide and ubiquinone), which can ac-
cep one o wo elec ons. When hey accep a
single elec on hey become ee adical molecules
hemsel es capable o slipping he elec on o
molecula O
2
. I is es ima ed ha a small pe cen -
age (B5%) o he o al O
2
consumed by mi o-
chond ia a no mal P
O
2
is con e ed o O
2
−
.
To comple e his o e iew on adicals i e-
mains o say some wo ds on eac i e ni ogen
species. Ni ic oxide o ni ogen monoxide (NO)
is a ee adical as i has one unpai ed elec on. I
is a gas molecule capable o eely di using
h ough cell memb ane sys ems. NOis physiolog-
ically syn hesized by a amily o di e en ni ic
oxide syn hases om he amino acid
L
-a ginine. I
binds o Fe
2+
o heme g oups, and in ac many
o i s physiological ac ions a e media ed by he
soluble guanyla e cyclase, which is a hemop o ein
ha syn hesizes cGMP, al hough cyclase indepen-
den physiological e ec s ha e been desc ibed
(A che e al., 1994; Bolo ina e al., 1994;
Ohkuma and Ka su a, 2001). Due o i s abili y o
eac wi h heme g oups, NO can inhibi cy-
och ome oxidase compe i i ely wi h O
2
(Bo e is
e al., 2000). In body luids NO eac s wi h O
2
o
gi e he mo e eac i e adical ni ogen dioxide
(2NO+O
2
2NO
2
), bu la gely i is oxidized o
ni i e ion (4NO+O
2
+2H
2
O4NO
2
−
) o eac s
wi h he heme g oups o oxyhemoglobin o o m
me ahemoglobin and NO
3
−
(ni a e). In ac , he
e alua ion o NOsyn hesis in i o is qui e e-
quen ly assessed by he measu emen o ni i e
and ni a e le els.
C.Gonzalez e al.
/
Respi a o y Physiology & Neu obiology
132 (2002) 17–41
21
Fig. 2. Gene al schema o he mi ochond ial espi a o y chain illus a ing he loci whe e O
2
−
can be gene a ed. The schema also
shows he mos commonly used inhibi o s o he espi a o y chain, which when ac ing downs eam o he loci o p oduc ion o O
2
−

would inc ease, while hose ac ing ups eam will inhibi O
2
−
p oduc ion.
2
.
2
.Seconda y ee adicals and o he eac i6e
species
Once gene a ed, hese h ee main ee adicals
(OH,O
2
−
and NO) can unde go an eno mous
a ie y o eac ions wi h o he molecules gi ing
ise o new adicals o o species, which a e
oxidized o educed. When new adicals a e
o med hey a e capable o unde going u he
eac ions o igina ing chain eac ions un il one o
he adicals eac s wi h a molecule (sca enge ) o
gene a e an un eac i e compound. This wide-
sp ead eac i i y is he basis o he oxic e ec s
o ee adicals. Fo example, in hei eac ions
wi h memb ane lipids, OHabs ac an a om o
hyd ogen om polyunsa u a ed a y acids o m-
ing H
2
O plus a modi ied a y acid which is a
ca bon adical species eadily eac ing wi h molec-
ula O
2
. The inco po a ion o O
2
by he a y acid
ca bon adical species ans o ms i in o a pe oxyl
adical (RCOOCOOH) which in u n is capa-
ble o abs ac ing ano he hyd ogen a om om a
nea by polyunsa u a ed a y acid, he eby
sp eading ou he al e a ion o lipids; al e na-
i ely, i migh a ack ano he double bond in he
same molecule o o m a cyclic endope oxide
compound which inally would b eak o gene a e
malonaldehyde and o he byp oduc s whose le els
a e usually de e mined o es ablish he magni ude
o lipid pe oxida ion in ci cums ances o in e es .
The inal poin is ha he modi ica ions and
b eakdown o he lipids o he memb anes change
hei luidi y and many o hei basic p ope ies.
Adi ionally, pe oxyl adicals, malonaldehyde and
o he aldehyde byp oduc s o lipid pe oxida ion,
as well as OHi sel and ca bon cen e ed adicals,
can eac wi h hiol and wi h amino g oups o
p o eins o o a ack many si es in essen ially
e e y amino acid o gene a e all kinds o dis o ed
molecules (b eakdown, in e molecula o in-
amolecula c oss-linking o p o eins wi h he
o ma ion o agg ega es o memb ane p o eins,
e c.) which ul ima ely imply he loss o he mem-
b ane unc ion and i s e en ual des uc ion. The
same kind o c oss-linking, b eakdowns and addi-
ional modi ica ions can be made by OHin DNA
molecules and be ween DNA and nuclea p o eins
leading o al e a ions in he epai o DNA o in

C.Gonzalez e al.
/
Respi a o y Physiology & Neu obiology
132 (2002) 17–41
22
i s eplica ion o ansc ip ion. In sum, OHis so
eac i e ha i will eac in he e y same place
whe e i is p oduced and disappea immedia ely
as such en i y, bu a he same ime new eac i e
molecules would appea which would sp ead he
e ec s. The ques ion in he con ex o he p esen
a icle is i he e exis s any mechanism in he cells
capable o ci cumsc ibing he ac ion o OH o i
o ac as he signaling molecule du ing hypoxia as
Kie zmann e al. (2000) sugges . Wha e e he
mechanism(s) i should be based in wo p emises:
i s , he p oduc ion o OHshould be loca ed in
a e y es ic ed a ea o he cell; his es ic ed
a ea mus be coinciden wi h he loca ion o he
cell machine y ha ini ia es he hypoxic ansduc-
ion cascade ( o example, he O
2
-sensi i e K
+
channels o hypoxia inducible ac o 1a, HIF-1a),
and second, he ‘OH- ecep o molecule’i sel
(e.g. he O
2
-sensi i e K
+
channels o HIF-1a)
should ac as a e y e ec i e OHsca enge o
a oid he widesp ead ac ion o seconda y eac i e
molecules wi h he ine i able loss o capaci y o
ocus he hypoxic signal.
The O
2
−
is conside ably much less eac i e han
OH. I can be p o onized o o m HO
2
(hy-
d ope oxyl adical) which has a g ea e educing
ac i i y and which in addi ion is eely pe meable
o memb anes owing o is lack o cha ge; bo h
cha ac e is ics make HO
2
an impo an ee adi-
cal in spi e o he ac ha i s concen a ion in
biological sys ems is 1/100–400 ha o O
2
−
.In
addi ion, hyd ope oxyl adical is in ol ed in he
spon aneous dismu a ion o O
2
−
 o H
2
O
2
and
molecula O
2
. This eac ion, which in ol es he
educ ion o one O
2
−
and he oxida ion o an-
o he o molecula O
2
, occu s h ough he o ma-
ion o hyd ope oxyl adical acco ding o he
equa ions:
H
−
+O
2
−
HO
2
;
HO
2
+O
2
−
+H
+
H
2
O
2
+O
2
Supe oxide adical can dona e one elec on o
ansi ion me als ans o ming Fe
3+
,Cu
2+
and
Mn
3+
in o Fe
2+
,Cu
+
and Mn
2+
, espec i ely,
and hen i ac s as a educing agen , bu i can
also ac as an oxidizing agen e e sing he eac-
ion. These educing/oxidizing eac ions a e e y
impo an because he enzymes esponsible o he
cellula elimina ion o O
2
−
, he supe oxide dismu-
ases (cy osolic o CuZn supe oxide dismu ase
and mi ochond ial o manganese supe oxide dis-
mu ase), a e me allop o eins con aining hose
me als and h ough cycles o educ ion and oxida-
ion o hei me als accele a e eno mously he
disappea ance o O
2
−
. O he impo an eac ions
o O
2
−
include oxida ion o asco ba e and he
ca echol ing o o m asco ba e and diphenol
adicals and H
2
O
2
. Ano he eac ion o in e es is
ha o O
2
−
wi h hypoclo ous acid ( o med in
u n by he ac ion o myelope oxidase on H
2
O
2
+
Cl
−
HOCl+OH
−
) o yield O
2
,Cl
−
and OH.
Summa izing, we can s a e ha up o his poin
mos o O
2
−
ends up as H
2
O
2
due o spon aneous
and enzyme ca alized dismu a ion. To some ex-
en , we may say ha H
2
O
2
is he inal common
pa h o mos ee adicals in he o ganism.
Howe e , H
2
O
2
is ha m ul o he cells. I be-
longs o a ca ego y o molecules collec i ely called
eac i e oxygen species (ROS), which include ee
adicals. H
2
O
2
and o he ROS such as hypoclo -
ous acid o pe oxyni i e (NO+O
2
−
ONOO
−
;
see below) ha e in common wi h ee adicals
being mo e eac i e han molecula O
2
, bu unlike
ee adicals hey do no ha e unpai ed elec ons.
The oxici y o H
2
O
2
would explain he exis ence
o speci ic enzyma ic sys ems o des oy i , includ-
ing ca alases (2H
2
O
2
2H
2
O+O
2
) and pe oxi-
dases, he mos impo an o which is glu a hione
pe oxidase (H
2
O
2
+2GSHGSSG+2H
2
O;
GSSG is back educed o GSH by he ac ion o
he NADPH based glu a hione educ ase). The
oxici y o H
2
O
2
is no exe ed di ec ly unless i is
applied a high concen a ions ( 10 mm o
many cells), bu by he OHi can gene a e upon
eac ion wi h ansi ion me als by he Fen on
eac ion (Fe
2+
+H
2
O
2
Fe
3+
+
OH+OH
−
) o upon homoly ic ission by he
ac ion o ul a iole ligh . We al eady ha e exam-
ined he widesp ead damaging e ec s o OH
(Fig. 3).
Ni ic oxide (NO) is also eadily eac i e wi h
many o he adicals. Fo example, i can eac
apidly wi h supe oxide o yield pe oxyni i e
(NO+O
2
−
ONOO
−
) and wi h hyd oxyl adi-
cal o o m ni ous acid (NO+OHHNO
2
).
C.Gonzalez e al.
/
Respi a o y Physiology & Neu obiology
132 (2002) 17–41
23
Fig. 3. Gene al schema o ROS gene a ion in he o ganisms. H
2
O
2
, a he cen e o he scheme, ep esen s a common inal p oduc
whose le els a e kep low by speci ic deg ading enzyma ic sys ems.
Pe oxyni i es and ni ous acid a e ni ogen eac-
i e species capable o eac ing wi h SH g oups o
p o eins and wi h i on/sul u elec on anspo -
ing p o eins o mi ochond ia. As we ha e al eady
men ioned, NOcan eac in a e e sible manne
wi h cy och ome oxidase, he eby compe ing wi h
O
2
; his migh be a mechanism o con ol he
oxida ion a e o he ene gy-yielding compounds
(nu ien s). Howe e , mi ochond ia mus ha mo-
nize his po en ially e e sible egula ing mecha-
nism o NOwi h i s abili y o eac wi h O
2
−
 o
o m pe oxyni i e and h ough i o p oduce
i e e sible damaging e ec s on elec on ans-
po ing p o eins. Pe oxyni i e can in e ac wi h
y osine esidues in a eac ion ca alyzed by supe -
oxide dismu ase o o m he e y s able 3-ni o y-
oxine, changing he o e all cha ge o he esidues
o y osine p esen in p o eins and making i
di icul o 3-ni o y osine con aining p o eins o
assemble in o mul ime ic uni s o o in e ac wi h
o he p o eins (Beckman and Koppenol, 1996)
and o pa icipa e in signaling cascades, by al e -
ing he e iciency o y osine phospho yla ion. The
g ea s abili y o 3-ni o y osine makes he mea-
su emen o i s le el a good index o ni ic oxide/
pe oxyni i e in biological sys ems. Pe oxyni i e
and o he ni ogen eac i e species can eac wi h
hiols, including ha in glu a hione, o o m ni-
oso hiols. Special in e es has ecen ly been
gi en o he ni osyla ion o one cys eine esidue
in he yanodine ecep o /channel, which seems o
occu in a P
O
2
-dependen manne (Eu e al.,
2000). When sa coplasmic esicles a e exposed o
ambien P
O
2
(:150 mmHg) six o eigh hiol
esidues pe yanodine ecep o become oxidized,
he oxida ion e e ing upon exposu e o P
O
2
o
10 mmHg, which is in he ange o ha physiolog-
ically encoun e ed in he muscle; on swi ching
back o ambien P
O
2
, eoxida ion o hose hiol
esidues akes place. These cys eine esidues will
cons i u e, acco ding o (Eu e al., 2000), an O
2
-
senso . I is concep ually impo an o ealize ha
his po en ial O
2
-senso needs o be coupled o a
edox sys em, which mus be p esen in he sa -
coplasmic e iculum esicles, and which mus be
esponsible o swi ching back and o h om he
oxidized o he educed s a e. Only he educed
o m o he channel is suscep ible o physiologi-
cal modula ion by NO, which does so by ni osy-
la ing one o he educed cys eine esidues, o in
o he wo ds, he edox s a e o he channel se by
P
O
2
de e mines he capaci y o espond o (o he
abili y o be egula ed by) NO.
C.Gonzalez e al.
/
Respi a o y Physiology & Neu obiology
132 (2002) 17–41
24
3. ROS as media o s o he hypoxic ansduc ion
cascade
3
.
1
.C i e ia o conside ROS media o s o a
physiological cellula esponse
Following classical s udies on messenge s such
as neu o ansmi e s (McLennan, 1963) and he
mos ecen a icle by Lande (1997) on ROS in
signal ansduc ion, i is possible o pu o wa d a
se o c i e ia as equi emen s o conside ROS as
ue second messenge s wi h he capaci y o con-
ol cellula unc ions unde physiological condi-
ions. A i s c i e ion should be ha he pu a i e
ROS molecule ac ing as messenge o media o o
he hypoxic esponse mus inc ease o dec ease
du ing hypoxia; a gi en ela ionship (modula ion
can modi y ha pu a i e ela ionship) should exis
be ween he ROS le el change, he in ensi y o
hypoxia and he s eng h o he hypoxic esponse.
The second c i e ion would be ha he cellula
mechanism esponsible o he change in he le el
o ROS molecule, pu a i e messenge , needs o be
iden i ied. The hi d c i e ion would be he de ini-
ion o he p ima y a ge o he ROS as well as
he de ini ion o he change p oduced by he ROS
molecule on i s a ge . The possibili y exis s ha
he p ima y a ge o he pu a i e messenge ROS
molecule is i sel he e ec o (s) o he cell e-
sponse o on he con a y, i is possible ha he
p ima y a ge o ROS is a i s link o a chain o
cascade o e en s coupling he change in he ROS
le els o he e ec o (s) o he cell esponse; in his
la e case he en i e cascade should be de ined.
The ou h c i e ion would be o de ine he cellu-
la esponses elici ed by he e ec o (s). The i h
c i e ion would be he de ini ion o mechanisms
esponsible o he inac i a ion o ROS signal.
Ob iously i he ROS species inc ease du ing hy-
poxia we should be looking o ROS deg ading
mechanisms, bu i hey dec ease we should be
looking o he enzyme o cells p ocess leading o
he es o a ion o no mal ROS le els. In any case,
i a gi en s imulus ac s ia a change in ROS
le els, he mechanisms o he elimina ion o
hose ROS should be induced by p olonged p e-
sen a ion o he s imulus. The six h c i e ion
(pha macological) would be ha he manipula ion
o he ROS le els, whe he by eagen s ha mod-
ula e glu a hione le els o ha di ec ly sca enge
ROS, should al e he cell esponse in a p e-
dic able and ep oducible manne ; a six h bo
se en h c i e ion would be ha o e exp ession o
silencing o he exp ession o he ROS sca enging
enzymes should al e he cell esponse in a p e-
dic able and ep oducible manne .
3
.
1
.
1
.Fi s c i e ion
:
he signal ROS le6els du ing
hypoxia
The e y i s ques ion ha should be add essed
is i hypoxia inc eases o dec eases ROS in gen-
e al o i he e is a unique ROS species whose
p oduc ion ollows a pa icula beha io du ing
hypoxia. The classical iew is ha he a e o
p oduc ion o ROS changes in di ec ela ion o
he issue P
O
2
(Chance e al., 1979; A che e al.,
1986, 1989). In ac , his iew is gene ally ac-
cep ed a p esen (A che e al., 1993; Go lach e
al., 1994; Fand ey e al., 1994, 1997; Halliwell and
Gu e idge, 1999; Kie zmann e al., 2000; Gnaige
e al., 2000). Howe e , qui e ecen ly Schumacke
and cowo ke s (Chandel e al., 1998; Du an eau
e al., 1998; Chandel and Schumacke , 2000;
Chandel e al., 2000) ha e ound he opposi e:
hypoxia p oduces an inc ease in he cellula a e
o ROS p oduc ion which akes place a he le el
o mi ochond ia and which is p opo ional o he
le el o hypoxia. Chandel e al. (1998) claim ha
hei measu emen s using dichlo o luo escein
luo escence, which appa en ly measu es p e e en-
ially H
2
O
2
, a e he co ec ones. Howe e he e
a e many measu emen s using di e en me hods
(see Chance e al., 1979) in which i has been
shown ha hypoxia dec eases he a e o p oduc-
ion o ROS, he dec ease being a ibu ed o a
dec ease o p oduc ion in mi ochond ia as well as
in mic osomal and pe oxisomal enzymes. A che
e al. (1993) and Paky e al. (1993) using lucigenin
chemiluminescence, which appa en ly measu es
p e e en ially O
2
−
(Halliwell and Gu e idge,
1999), also ound ha hypoxia dec eases ROS
p oduc ion a e. Simila ly con adic o y a e he
obse a ions ega ding o CoCl
2
(an induce o
EPO p oduc ion), which dec eases he p oduc ion
o ROS i hey a e measu ed wi h dihyd o ho-
damine 123 (Fand ey e al., 1997), o inc eases
C.Gonzalez e al.
/
Respi a o y Physiology & Neu obiology
132 (2002) 17–41
25
hei p oduc ion i ROS a e measu ed wi h
dichlo o luo escein (Chandel e al., 1998). In e -
es ingly, bo h dyes yielded an inhibi ion o ROS
gene a ion unde o enone and diphenileneiodo-
nium (DPI) (C oss e al., 1990; Ehleben e al.,
1997; Chandel e al., 1998; Du an eau e al.,
1998); o enone and DPI also dec eased lucigenin
chemiluminescence (A che e al., 1993, 1999).
Howe e , while an imycin A inc eased dichlo -
o luo escein luo escence (Chandel e al., 1998) i
dec eased lucigenin chemiluminescence (A che e
al., 1993).
Wi h his pano ama i is di icul o make a
meaning ul o mula ion o he pu a i e ROS
molecule(s), which a e signals o he hypoxic
ansduc ion cascade. We do no know i he
signal should be a pa icula ROS species ha
dec eases du ing hypoxia, a gene al dec ease in
he ROS le els, o i on he con a y, we should
be sea ching o some hing ha inc eases du ing
hypoxia. We would like o quo e (Semenza, 1999,
2000) when he w i es ha ‘ he di ec measu e-
men s o ROS a e so demanding ha hey gene -
a e da a suppo ing opposing iews’. Howe e , we
wan o make explici some hough s on he p ob-
lem, being awa e ha expe imen al da a migh
con adic he opinions exp essed he e. When we
conside he gene al mechanisms o ROS p oduc-
ion p esen ed in p e ious pa ag aphs, he in u-
i i e no ion would be ha he a e o p oduc ion
o ROS in cells should be di ec ly p opo ional o
he P
O
2
o he cells. Reac ions o au o-oxida ion
as well as oxida ions d i en by enzymes and he
accep ance by molecula O
2
o elec ons leaking a
he le el o he quinone pool in mi ochond ia (see
Fig. 2) would dec ease du ing hypoxia, when P
O
2
is diminished. A he same ime, howe e , i
would appea ha he gene a ion o O
2
−
a he
quinone pool le el would depend on he a e o
low o elec ons h ough he espi a o y chain,
i.e. on he espi a o y a e. As s a ed in he i s
sec ion o his a icle, cell sys ems in ol ed in he
gene al homeos asis o O
2
a e ac i a ed by hy-
poxia, his ac i a ion equi ing an inc ease in cell
espi a ion o suppo cell and homeos a ic loop
unc ion (Obeso e al., 1993, 1997). Thus, i could
be expec ed ha in hese specialized cells hypoxia
p oduces an inc ease in elec on low du ing an
ample ange o hypoxia ( om :70 o 20 mmHg
o a e ial P
O
2
), and he eby, an inc ease in he
elec on leakage and mi ochond ial o ma ion o
O
2
−
is concei able in hose si ua ions (see Fig. 3
in Hoshi and Heinemann, 2001) when hypoxia is
in ense enough a comp omise will occu and in
spi e o he high a ini y o cy och ome oxidase
o O
2
, i will be unable o ans e enough elec-
ons o O
2
and he espi a o y a e and he
genesis o ATP will d op. In he case o he CB in
i o his me abolic comp omise occu s a a e ial
P
O
2
:20 mmHg; below his P
O
2
he CB is unable
o adequa ely signal he le el o hypoxia and
chemo ecep o ac ion po en ial equency de-
c eases, ins ead o inc easing (see Fidone and
Gonzalez, 1996). The majo i y o he cells o he
o ganism will beha e du ing hypoxia as he CB
below he comp omise P
O
2
: all O
2
−
-gene a ing
p ocesses including ha o mi ochond ia would
dec ease as he in ensi y o hypoxia inc eases. To
s a e his in a di e en manne , i would appea
ha a di e en ial esponse could be expec ed: in
mos cells o he o ganism hypoxia would end o
dec ease ROS p oduc ion while in CB chemo e-
cep o cells, in EPO-p oducing cells and in
smoo h muscle cells o he pulmona y a e ies
hypoxia would end o inc ease ROS p oduc ion
a leas in hose anges o hypoxia compa ible
wi h li e. In con as , Schumacke and co-wo ke s
measu ed an inc ease in ROS p oduc ion in ca -
diomyocy es and hepa oma cells (EPO-p oducing)
which is compa able in magni ude and in P
O
2
dependence in bo h cell ypes (Chandel e al.,
1998; Du an eau e al., 1998) and Fand ey e al.
(1994) measu ed a dec ease in he a e o p oduc-
ion and elease o H
2
O
2
o he incuba ing solu-
ion by EPO-p oducing hepa oma cells p ecisely
in he ange o P
O
2
whe e hey sec e e EPO.
In sum, he i s c i e ion is no se led, we do
no know whe he hypoxia inc eases o dec eases
ROS le els.
3
.
1
.
2
.Second c i e ion
:
cell mechanisms
esponsible o he change in ROS p oduc ion
du ing hypoxia should be de ined
Those p oposing ha hypoxia would dec ease
ROS le els (C oss e al., 1990; Youngson e al.,
1993; Acke , 1994a,b; Acke and Xue, 1995; Fan-
C.Gonzalez e al.
/
Respi a o y Physiology & Neu obiology
132 (2002) 17–41
32
iden i ica ion o HIF-1 (Wang and Semenza,
1993a,b) as a ac o induced by hypoxia, capable
o binding o he 3’enhance egion o EPO and
o he hypoxia- egula ed genes and capable o
causing an inc ease in hei a e o ansc ip ion.
The ea e , i was demons a ed ha in ac HIF-
1 was a he e odime composed o HIF-1a, whose
le els a e egula ed by low P
O
2
, and by he cons i-
u i ely exp essed HIF-1b[also known as ARNT
(a yl hyd oca bon ecep o nuclea ansloca o ),
as AhR (a yl hyd oca bon ecep o ) and as dioxin
ecep o ], he dime iza ion was shown o be neces-
sa y, no o he ansloca ion o bo h compo-
nen s o he nucleus, bu o he s able associa ion
o HIF-1 wi hin he nuclea compa men (see
Gassmann e al., 2000). Addi ional s udies
demons a ed ha HIF-1aas well as HIF-1b
mRNA le els, ansc ip ion and ansla ion a es
we e no modi ied by hypoxia. Since HIF-1a
p o ein inc eased du ing hypoxia i implied ha in
no moxia he p o ein is deg aded and ha hy-
poxia p e en s he deg ada ion (Huang e al.,
1996).
The e o e, he ques ion we should make a his
poin is how he change in ROS le els p oduces
such s abiliza ion o HIF-1a o p e en i s deg a-
da ion. Huang e al. (1996) showed ha he le els
o HIF-1a, which a e inc eased by hypoxia, de-
c eased soon i in ac cells we e e-exposed o
no moxia (21% O
2
), bu exposu e o cell ex ac s
o he same no moxic le els o O
2
did no al e
he binding o HIF-1a o DNA. In o he wo ds,
o high P
O
2
o des abilize HIF-1a he in eg i y o
he cells is necessa y. In addi ional expe imen s i
was shown ha H
2
O
2
(0.1–1.0 mM) esul ed in a
dose-dependen dec ease in he EPO mRNA ex-
p ession induced by hypoxia which uns pa allel
o a dec ease in HIF-1abinding o DNA and o a
dec ease in HIF-1ap o ein, wi hou modi ica ion
o HIF-1amRNA le els o ansla ion a e. I
was also obse ed ha ex ac s om hypoxic cells
ea ed wi h H
2
O
2
showed a no mally high bind-
ing capaci y o DNA. Those indings appea o
indica e ha he binding o HIF-1a o DNA is
con olled p ima ily a he le el o HIF-1a
p o ein. In addi ion, Huang e al. (1996) also
s udied some o he modi ica ions o HIF-1a
which migh al e i s binding o DNA. They
ound ha oxidizing sul hyd yl eagen s added o
cell ex ac s inhibi ed he binding in a dose-depen-
den manne , his inhibi ion was e e sible wi h
educing hiol eagen s excep when sul hyd yl
alkyla ing agen s we e used. In addi ion, in a
co ans ec ion expe imen using a cons uc ex-
p essing hio edoxin unde he con ol o he con-
s i u i e cy omegalo i al p omo e and enhance ,
along wi h ano he cons uc exp essing luci e ase
unde he con ol o he SV40 p omo e and
EPO-enhance , hey obse ed ha o e exp ession
o hio edoxin esul ed in a po en ia ion o he
exp ession o luci e ase bo h in no moxic and
hypoxic condi ions, such po en ia ion appea ed o
be unde he con ol o he HIF-1 binding because
ans ec ion wi h a mu a ed EPO enhance , which
was unable o bind HIF-1a, caused he loss o he
esponse. As a whole he esul s would indica e
ha H
2
O
2
, like no moxia, p oduces modi ica ions
in he HIF-1amolecule which o ce i o be
di ec ed o deg ada ion, bu which does no im-
pede i s binding o DNA. Al e na i ely, hypoxia
p omo es changes in HIF-1a, which p e en i s
deg ada ion, leading o accumula ion o he an-
sc ip ion ac o ; he changes induced by hypoxia
appea o be edox dependen because hey can be
mimicked by hio edoxin and coun e ac ed by
no moxia and H
2
O
2
; a simila conclusion was
a ained by S ini as e al. (1999). An al e na i e
in e p e a ion o he indings o Huang e al., is
ha hypoxia (o he dec ease o ROS) o he
o e exp ession o hio edoxin p oduced changes
in he deg ading enzymes o he p o eosoma
which make hem incapable o deg ading a no -
mal (o educed) HIF 1a.
Howe e , in a se ies o expe imen s Chandel e
al. (2000) ound essen ially he opposi e: hypoxia
and CoCl
2
, bu no des e iexamine, inc eased
ROS le els, and he h ee si ua ions augmen ed
he exp ession o a epo e gene unde he con-
ol o he EPO-p omo e as well as he amoun
o HIF-1ap o ein. Inhibi o s o he espi a o y
chain a he complex I, he hiol educ an pi o-
lidine di hioca bama e, and o e exp ession o
ca alase, dec eased hypoxic ROS le els and p e-
en ed he augmen ed exp ession o HIF-1a.The
hiol educ an and he o e exp ession o ca alase
dec eased he ROS le els inc eased by CoCl
2
and

C.Gonzalez e al.
/
Respi a o y Physiology & Neu obiology
132 (2002) 17–41
33
also he exp ession o HIF-1a. Nei he maneu e
dec eased he exp ession o he epo e gene o
dec eased he le els o HIF-1ap omo ed by des-
e ioxamine. Addi ion o H
2
O
2
and e -bu ylhy-
d ope oxide o no moxic cells p omo ed an
inc ease o HIF-la p o ein as well as an inc eased
exp ession o he epo e gene; hese maneu e s
we e ine ec i e in cells o e exp essing ca alase. A
phospha ase inhibi o and an inhibi o o phos-
pha idyl inosi ol 3-kinase abolished he s abiliza-
ion o HIF-1ainduced by hypoxia, CoCl
2
and
des e ioxamine and a p o easome inhibi o was
able o e e se hose e ec s; because ea men o
he cells wi h H
2
O
2
was unable o p e en he
e ec s o he phospha ase and kinase inhibi o s,
he au ho s concluded ha ROS (including H
2
O
2
)
ac o s abilize HIF-1aa a s ep p io o he
ac ion o he phospha ases and kinases. Des e -
ioxamine would p oduce he s abiliza ion o
HIF-1aby a mechanism independen o ROS and
would also ac a a s ep p io o he phospha ases
and kinases. Cu iously, an imycin A, which in-
c eased ROS, did no induce he exp ession o he
epo e gene o accumula ion o HIF-1a. Chan-
del e al. (2000) s a e ha i is because his mi o-
chond ial inhibi o causes a smalle inc ease in
ROS le els han hypoxia. Howe e , i is di icul
o unde s and how hypoxia can p oduce a mo e
educed s a e a he quinone pool han an imycin
(see Fig. 2). In addi ion, an imycin A was applied
in no moxia implying ha he concen a ions o
O
2
o accep he elec ons om semiquinone a e
much highe han in hypoxia (20% O
2
s. 1.5% in
hei expe imen s) and, he e o e, he o ma ion
o O
2
−
should be acili a ed (Chance e al., 1979).
The con o e sy is ob ious: o Huang e al.
(1996) i is he dec ease o ROS le els which
causes he s abiliza ion o HIF-1aand o Chandel
e al. (2000) i is he inc ease o ROS le els ha
does he same. The e o e, and in keeping wi h he
s a emen o he hi d c i e ion, we do no know i
we should be looking o a educ ion o an oxida-
ion o HIF-1aas esponsible o i s s abiliza ion.
An addi ional aspec ha we eel o ced o es a e
a his poin (which is seldom aken in o accoun ;
see abo e) is ha in he p oposed signalling by
ROS he subcellula localiza ion o he ROS sig-
nal a he a ge place should be aken in o con-
side a ion, o he wise a ‘ ue’signaling would no
exis , Finally, we ag ee wi h Halliwell and Gu e-
idge (2000) when hey s a e ega ding signaling
by ROS: ‘‘one mus be wa y: cells in cul u e a e
usually hype oxic and hey lack many o he
an ioxidan s hey use in i o…Cells adap o li e
in cul u e. I is no impossible ha , wi h umo
cell lines in pa icula , adap a ions o a o
g ow h may occu by using adicals o igge
pa hways ha a e igge ed by o he means in
i o’’.
3
.
1
.
4
.Fou h c i e ion.De ini ion o he cellula
esponses elici ed by he e ec o ac i6a ed by he
changes in ROS le6els
A his poin in he a icle i is ob ious ha we
can de ine he cellula esponses elici ed by he
s imulus, i.e. hypoxia. Howe e , we ha e se ious
di icul ies in de ining he cellula e ec s p oduced
by he ROS le el changes occu ing du ing hy-
poxia because we do no know wha hose
changes a e. And e en i we knew he changes in
ROS le els p oduced du ing hypoxia we do no
know i hey a e cause o consequence o ac i a-
ion o he e ec o and cellula esponses.
Hypoxia in CB chemo ecep o cells and
PASMC educes he opening p obabili y o he
O
2
egula ed K
+
channels (see c i e ion 3) leading
o cell depola iza ion, ac i a ion o ol age ope -
a ed Ca
2+
channels and elease o neu o ansmi -
e s in chemo ecep o cells o con ac ion in
smoo h muscle cells (see Fig. 1). In c i e ion 3 we
ha e al eady indica ed ha sul hyd yl eagen s
cause changes in he opening p obabili y o am-
pli ude o he cu en s ha can be compa ible
wi h a dec ease o an inc ease in ROS le els.
Re e ing speci ically o ROS, he e a e many
s udies in which H
2
O
2
has been applied di ec ly
and i s e ec s on se e al ypes o K
+
channels
es ablished. Fo example, Vega-Saenz de Mie a
and Rudy (1992) showed ha H
2
O
2
augmen ed
he cu en ca ied by ansien ype K
+
channels
(like hose O
2
-sensi i e channels in abbi
chemo ecep o cells) wi hou a ec ing o he K
+
channels. Simila ly he K 3.3 p esen in ai way
chemo ecep o s a e ac i a ed by H
2
O
2
(Wang e
al., 1996; Fu e al., 2000). And he e a e many
epo s showing ha H
2
0
2
o e -bu ylhyd ope -
C.Gonzalez e al.
/
Respi a o y Physiology & Neu obiology
132 (2002) 17–41
34
oxide ac i a e maxi-K
+
(Hayabuchi e al. 1998;
Shin e al., 2000; Guo e al., 2000; Ba low e al.,
2000). All hese da a would a o he no ion ha
hypoxia dec eases ROS, and he eby hypoxia
would dec ease he ac i i y o K
+
channels and
depola iza ion would occu . Howe e , he e a e
s udies in which i has been ound ha H
2
O
2
inhibi s K
+
channels. Tha is he case o maxi-
K
+
(DiChia a and Reinha , 1997; Tang e al.,
2001). These s udies would be in ag eemen wi h
he no ion ha hypoxia inc eases ROS le els and
he eby causes inhibi ion o K
+
channels, cell
depola iza ion and inc eased cell ac i i y.
When we conside he e ec s o H
2
O
2
on he
o e all esponse o he sys em, he s udy o Os-
anai e al. (1997) in he in ac CB is e y illus a-
i e as hey concluded ha maneu e s aimed a
al e ing endogenous H
2
O
2
, including exogenous
H
2
O
2
applica ion, did no p o ide e idence o i s
supposedly c i ical ole in O
2
chemo ecep ion. On
he con a y, in he pulmona y ci cula ion e -
bu ylhyd ope oxide, H
2
O
2
and ROS gene a ing
sys ems (i.e. pe usion o he lungs wi h xan hine
plus xan hine oxidase) clea ly inhibi ed hypoxic
pulmona y asocons ic ion (A che e al., 1986;
Mohazzab and Wolin, 2000). In he case o EPO-
p oducing cells we ha e commen ed in p e ious
pa ag aphs on he appa en ly dual ac ions o
H
2
O
2
since on he one hand i inhibi ed EPO-p o-
duc ion induced by hypoxia (Fand ey e al., 1994)
and on he o he hand i augmen ed he s abiliza-
ion o HIF-1aand p omo ed he exp ession o a
epo e gene unde he con ol o he EPO en-
hance (Chandel e al., 2000). I should also be
men ioned ha Ueno e al. (1988) wo king wi h
enal ca cinoma EPO-p oducing cells and Ndele
e al. (1996)) wo king wi h HEP 3B cells ound
ha H
2
O
2
and ROS gene a ing sys ems inc eased
EPO p oduc ion while ROS sca enge s ha e he
opposi e e ec s.
3
.
1
.
5
.Fi h c i e ion.The mechanism
(
s
)
esponsible o he inac i6a ion o he ROS signal
should be de ined
Ob iously we should conside he e bo h possi-
bili ies: ha he signal is a dec ease and ha he
signal is an inc ease in ROS le els. In he i s
case, i a dec ease in ROS le els is conside ed o
be he esul o a dec ease in he ac i i y o
NADPH oxidase, he signal would disappea , i.e.
no mal ROS le els will be eco e ed, when P
O
2
e u ns o no moxic le els due o an inc ease in
he ac i i y o he enzyme by inc eased le els o
subs a e (i migh also equi e a eassembly o
he NADPH subuni s p omo ed by unknown
mechanisms; see second c i e ion). I is concei -
able ha ch onic hypoxia would p oduce a down-
egula ion o NADPH oxidase. A compa able
mechanism o inac i a e he ROS signal would be
alid whe he he dec ease o ROS le els is con-
side ed o occu a he mi ochond ial le el, cy-
och ome P450, e c. I on he con a y, he signal
is an inc ease in ROS le els, a gi en le el o ROS
should co espond o a gi en le el o s imula ion
in o de o ob ain a esponse p opo ional o he
in ensi y o he s imulus. This would imply ha
simul aneously o he ac i a ion o he mecha-
nisms gene a ing ROS he e is a pa allel ac i a-
ion o he mechanisms deg ading hem. I he
s imula ion is long las ing (e.g. ch onic hypoxia)
an up egula ion o he mechanisms deg ading
ROS should be expec ed (Halliwell and Gu e-
idge, 1999). Ob iously we a e e e ing he e o
GSH, glu a hione pe oxidase, ca alase and supe -
oxide dismu ase as he main sys ems o dispose
ROS. The e a e no , o ou knowledge, di ec
measu emen s o hose ROS sca enge sys ems in
he cell ypes we a e discussing, howe e , he e a e
some agmen a y da a ob ained in in ac animals
and humans exposed o high al i ude. Fo exam-
ple, Cos a e al. (1993) ob ained e idence indica -
ing ha in he li e o a s exposed o a simula ed
al i ude o 4400 m he e was a dec ease in he a e
o ROS p oduc ion wi h a concomi an dec ease
in supe oxide dismu ase ac i i y. Howe e ,
Nakanishi e al. (1995) a a simula ed al i ude o
5500 m ound a complex pa e n o changes in
hose enzymes which was o gan speci ic, and
which was accompanied by an inc ease in lipid
pe oxida ion and by a ma ked dec ease in body
weigh o he animals a ibu ed o an inc eased
me abolic a e due o he hypoba ic s ess. Al-
hough i is di icul om hese and simila da a
o each sound conclusions, some expe s in he
a ea would sugges ha he e a e isks o oxida-
i e s ess, independen o UV adia ion, a high
al i ude (Simon-Schnass, 2000).
C.Gonzalez e al.
/
Respi a o y Physiology & Neu obiology
132 (2002) 17–41
35
3
.
1
.
6
.Six h c i e ion.Manipula ion o ROS le6els
by al e a ion o glu a hione le6els o by he di ec
ac ion o ROS sca6enge s should p oduce he
p edic ed change in he cell esponse
O6e exp ession and elimina ion o he enzymes
sca6enging ROS should p oduce he p edic ed
change in he cell esponses.
The in e p e a ion o he da a esul ing om he
manipula ion o ROS and/o glu a hione le els is
no always s aigh o wa d due o he mul iple
a ge s o ROS. Fo example, he ac i a ion o
mimicking o glu a hione pe oxidase (e.g. by ad-
minis a ion o ebselen) would dec ease he GSH/
GSSG a io, jus as a si ua ion in which he e is an
inc ease in he a e o ROS p oduc ion, bu con-
a y o ha si ua ion ebselen will p oduce a
dec ease in he le els o H
2
O
2
. Since he dec ease
in he GSH/GSSG a io would end o inc ease
he opening p obabili y o K
+
channels, and
he eby o cause memb ane hype pola iza ion
(and asodila ion, in he case o pulmona y ci cu-
la ion), and he dec ease o H
2
O
2
would end o
educe he ac i i y o guanyla e cyclase, and
he eby o inhibi Ca
+
dependen K
+
channels
(and o p oduce asocons ic ion, in he case o
pulmona y ci cula ion), he o e all e ec could be
e y complex and di icul o de ine (see Wolin e
al., 1999; Mohazzab and Wolin, 2000). In addi-
ion, and a leas in he case o he lung, ROS can
modi y he me abolism o a achidonic acid-
de i ed me abolism in such a manne ha an
inc ease in ROS le els would augmen he p oduc-
ion o asocons ic o me aboli es (Seege e al.,
1986; Sande ud e al., 1993; see Wolin e al., 1999).
Expe imen s in he CB di ec ly ocused on his
las c i e ion ha e been pe o med in ou labo a-
o y e y ecen ly. We ha e measu ed GSH and
GSSG le els in cal CB in no moxia, hypoxia and
a e ea men o he o gans wi h N-ace ylcys-
eine, a p ecu so o glu a hione and di ec sca -
enge o ROS and measu ed he elease o
ca echolamines om chemo ecep o cells in hose
expe imen al si ua ions (Obeso e al., 2000b). N-
ace ylcys eine would mimic a si ua ion in which
he e is a dec eased p oduc ion o ROS and
would oppose a si ua ion o inc eased a e o
ROS p oduc ion. In o he wo ds, N-ace ylcys eine
would mimic hypoxia, i we accep ha hypoxia
dec eases ROS le els, and would oppose hypoxia
i we accep ha i augmen s ROS le els. We
ound ha hypoxia (P
O
2
:46 mmHg) did no
al e he absolu e le els o he GSH/GSSG quo-
ien s. N-ace ylcys eine inc eased absolu e glu-
a hione le els by p ima ily inc easing GSH;
he e o e, i inc eased he GSH/GSSG a io. Ye
N-ace ylcys eine did no modi y he basal no -
moxic o hypoxia-induced elease o ca e-
cholamines sugges ing ha ROS a e no c i ical in
se ing he hypoxic esponse.
In he case o he pulmona y ci cula ion,
A che e al. (1986) showed ha diamide (an
agen ha eac s wi h GSH and oxidizes i o
GSSG) educed he p esso esponse o hypoxia.
Diamide would mimic a si ua ion in which he e
is an inc eased p oduc ion o ROS: i will cause a
dec ease in GSH/GSSG a io and he e o e he
cellula powe o sca enging o ROS leading o
an inc ease in ROS le els. Bo h he dec ease in
GSH/GSSG a io and he inc ease in ROS le els
would end o oppose he asocons ic ion p o-
oked by hypoxia (see abo e). In ag eemen wi h
he expec ed ac ions, p e ea men o he animals
wi h N-ace ylcys eine p e en ed he e ec s o di-
amide (A che e al., 1986). In isola ed pe used
lungs, suplemen a ion o he supe usa es wi h
ca alase and supe oxide dismu ase neu alized he
inhibi ion on hypoxic pulmona y asocons ic ion
p oduced by ROS gene a ing sys ems (xan hine/
xan hine oxidase) (A che e al., 1986). Bu in
ano he a icle i was epo ed ha nei he ca a-
lase no supe oxide dismu ase modi ied he aso-
cons ic ion elici ed by hypoxia (Kjae e e al.,
1990), and mo e ecen ly Weissmann e al. (1998)
also ound ha supe oxide dismu ase was simi-
la ly ine ec i e. In his la e s udy se e al po en-
ial ROS sca enge s we e used and he esul s
we e inconclusi e as om all he agen s es ed
[supe oxide dismu ase, amino iazole (a ca alase
inhibi o ), i on (a p esumed supe oxide sca -
enge ) and ni o blue e azolium (ano he supe -
oxide sca enge )] only ni o blue e azolium in-
hibi ed hypoxic pulmona y asons ic ion. In
sum, he expe imen s aimed o al e ROS le els by
manipula ion o ROS sca enging sys ems in he
lung ha e yielded con lic ing esul s, and a e he e-
o e no conclusi e wi h ega d o he possible
C.Gonzalez e al.
/
Respi a o y Physiology & Neu obiology
132 (2002) 17–41
36
signi icance o ROS le els on hypoxic pulmona y
asocons ic ion.
The da a ela ed o his c i e ion in EPO-p o-
ducing cells a e likewise inconclusi e. In dis-
cussing he s abiliza ion o HIF-1a(c i e ion 3)
we ha e ocused on some disc epancies ha migh
ela e o his c i e ion. Fo example, we ha e
men ioned ha acco ding o Huang e al. (1996)
o e exp ession o hio edoxin po en ia ed he ex-
p ession o a epo e gene unde he con ol o
EPO enhance implying s abiliza ion o HIF-1a.
On he o he hand, Chandel e al. (2000) epo ed
ha o e exp ession o ca alase p e en ed s abi-
liza ion o HIF-1a. A simila con o e sy exis s
ega ding H
2
O
2
. Thus Fand ey e al. (1994) e-
po ed ha H
2
O
2
inhibi ed he p oduc ion o EPO
induced by hypoxia and ha ca alase e e sed he
inhibi ion p oduced by H
2
O
2
. On he con a y,
Ueno e al. (1988), Ndele e al. (1996) (see also
Chandel e al., 2000) epo ed ha ca alase inhib-
i ed he EPO p oduc ion s imula ed by H
2
O
2
and
by ROS-gene a ing sys ems glucose/glucose ox-
idase and xan hine/xan hine oxidase.
4. Conclusion
Wi h he a ailable in o ma ion we do no know
i hypoxia inc eases o dec eases ROS le els, and
he e o e we do no know i he pu a i e ole o
ROS, i.e. o inhibi K
+
channels in he CB
chemo ecep o cells and in PASMC and o s abi-
lize HIF-1ain EPO-p oducing cells, is he esul
o he educ ion o he oxida ion o he K
+
channel p o eins o HIF-1a. The unce ain ies
emain when we conside he e ec s o agen s ha
inc ease o dec ease ROS le els on he esponses
o CB chemo ecep o cells, PASMC and EPO-
p oducing cells. The e a e se e al mi ochond ial
poisons, which ha e opposi e e ec s on he a e o
ROS p oduc ion and le els, and ye all ac i a e
chemo ecep o cells and PASMC in no moxia
and po en ia e hei hypoxic esponse. Simila ly,
he e a e mi ochond ial agen s ha can ei he
dec ease o inc ease ROS le els, ye hey ail o
al e no moxic EPO p oduc ion, implying ha
nei he he dec ease no he inc ease in ROS le els
mimic hypoxia. The pano ama is no cla i ied by
he use o agen s o expe imen al maneu e s ha
al e he a e o ROS sca enging. The e a e e-
po s indica ing ha ROS sca enging enzymes o
d ugs end o mimic hypoxia and epo s indica -
ing he opposi e, bo h, in PASMC and in EPO-
p oducing cells. In he CB chemo ecep o cells he
sca enging o ROS did no al e he no moxic
ou pu o he cells no hei hypoxic esponse. In
sum, he a ailable in o ma ion on O
2
-sensing/hy-
poxic ansduc ion does no suppo an unequi o-
cal ole o ROS in hose cellula p ocesses. The
possibili y exis s ha many o he con adic o y
indings, which ha e been ob ained in umo cell
lines, ep esen peculia i ies o pa icula clones
used in di e en labo a o ies. In any case, i
should always be kep in mind ha he me abolic
p ope ies o umo cells migh di e subs an ially
om he pa en al no mal cells. In designing ex-
pe imen s, i should also be emembe ed ha
maneu e s aimed a manipula ing ROS can do
mo e han simply al e ROS le els: special ca e
should be aken in selec ing he po en ially
pleio opic ROS sca enge s and ROS mime ics.
Acknowledgemen s
Suppo ed by Spanish DGICYT G an PB97/
0400. Thanks o P o esso Beni o He e os o
c i ical eading.
Re e ences
Acke , H., 1994a. Cellula oxygen senso s. Ann. New Yo k
Acad. Sci. 718, 3–10.
Acke , H., 1994b. Mechanisms and meaning o cellula oxygen
sensing in he o ganism. Respi . Physiol. 95, 1–10.
Acke , H., Xue, D., 1995. Mechanisms o O
2
sensing in he
ca o id body in compa ison wi h o he O
2
-sensing cells.
News Physiol. Sci. 10, 211–215.
Alma az, L., Gonzalez, C., Obeso, A., 1986. E ec s o high
po assium on he elease o H-dopamine om he ca
ca o id body in i o. J. Physiol. London 379, 293–307.
Anichko , S.V., Belen’kii, M.L., 1963. Pha amacology o he
Ca o id Body Chemo ecep o s. MacMillan, New Yo k,
USA.
A che , S.L., Will, J.A., Wei , E.K., 1986. Redox s a us in he
con ol o pulmona y ascula one. He z 11, 127–141.
C.Gonzalez e al.
/
Respi a o y Physiology & Neu obiology
132 (2002) 17–41
37
A che , S.L., Nelson, D.P., Wei , E.K., 1989. Simul aneous
measu emen o O
2
adicals and pulmona y ascula eac-
i i y in a lung. J. Appl. Physiol. 67, 1903–1911.
A che , S.L., Huang, J., Hen y, T., Pe e son, D., Wei , E.K.,
1993. A edox-based O
2
senso in a pulmona y ascula-
u e. Ci c. Res. 73, 1100–1112.
A che , S.L., Huang, J.M., Hampl, V., Nelson, D.P., Shul z,
P.J., Wei , E.K., 1994. Ni ic oxide and cGMP cause
aso elaxa ion by ac i a ion o a cha ybdo oxin-sensi i e
K channel by cGMP -dependen p o ein kinase. P oc.
Na l. Acad. Sci. USA 91, 7583–7597.
A che , S.L., Souil, E., Dinh-Xuan, A.T., Sch emme , B.,
Me cie , J.C., El Yaagoubi, A., Nguyen-Huu, L., Ree e,
H.L., Hampl, V., 1998. Molecula iden i ica ion o he ole
o ol age-ga ed K
+
channels, K l.5 and K 2.1, in hypoxic
pulmona y asocons ic ion and con ol o es ing mem-
b ane po en ial in a pulmona y a e y myocy es. J. Clin.
In es . 101, 2319–2330.
A che , S.L., Ree e, H.L., Michelakis, E., Pu agun a, L.,
Wai e, R., Nelson, D.P., Dinaue , M.C., Wei , E.K., 1999.
O
2
sensing is p ese ed in mice lacking he gp91 phox
subuni o NADPH oxidase. P oc. Na l. Acad. Sci. USA
96, 7944–7949.
A che , S.L., Wei , E.K., Ree e, H.L., Michelakis, E., 2000.
Molecula iden i ica ion o O
2
senso s and O
2
-sensi i e
po assium channels in he pulmona y ci cula ion. Ad .
Exp. Med. Biol. 475, 219–240.
Ba low, R.S., El-Mowa y, A.M., Whi e, R.E., 2000. H
2
O
2
opens BK(Ca) channels ia he PLA
2
-a achidonic acid
signaling cascade in co ona y a e y smoo h muscle. Am.
J. Physiol. 279, H475–H483.
Beckman, J.S., Koppenol, W.H., 1996. Ni ic oxide, supe ox-
ide, and pe oxyni i e: he good, he bad, and ugly. Am. J.
Physiol. 271, C1424–C1437.
Bolo ina, V.M., Najibi, S., Palacino, J.J., Pagano, P.J., Cohen,
R.A., 1994. Ni ic oxide di ec ly ac i a es calcium-depen-
den po assium channels in ascula smoo h muscle. Na-
u e 368, 850–853.
Bo e is, A., Cos a, L.E., Pode oso, J.J., Ca e as, M.C.,
Cadenas, E., 2000. Regula ion o mi ochond ial espi a ion
by oxygen and ni ic oxide. Ann. New Yo k Acad. Sci.
USA 899, 121–135.
Buckle , K.J., 1997. A no el oxygen-sensi i e po assium cu -
en in a ca o id body ype I cells. J. Physiol. London
498, 649–662.
Buckle , K.J., 1999. Backg ound leak K
+
-cu en s and oxygen
sensing in ca o id body ype I cells. Respi . Physiol. 115,
179–187.
Buckle , K.J., Williams, B.A., Hono e, E., 2000. An oxygen-,
acid- and anaes he ic-sensi i e TASK-like backg ound
po assium channel in a a e ial chemo ecep o cells. J.
Physiol. London 525, 135–142.
Chance, B., Sies, H., Bo e is, A., 1979. Hyd ope oxide
me abolism in mammalian o gans. Physiol. Re . 59, 527–
605.
Chandel, N.S., Schumacke , P.T., 1999. Cells deple ed o
mi ochond ial DNA ( ho0) yield insigh in o physiological
mechanisms. FEBS Le . 454, 173–176.
Chandel, N.S., Schumacke , P.T., 2000. Cellula oxygen sens-
ing by mi ochond ia: old ques ions, new insigh . J. Appl.
Physiol. 88, 1880–1889.
Chandel, N.S., Mal epe, E., Goldwasse , E., Ma hieu, C.E.,
Simon, M.C., Schumacke , P.T., 1998. Mi ochond ial eac-
i e oxygen species igge hypoxia-induced ansc ip ion.
P oc. Na l. Acad. Sci. USA 95, 11715–11720.
Chandel, N.S., McClin ock, D.S., Feliciano, C.E., Wood,
T.M., Melendez, J.A., Rod iguez, A.M., Schumacke , P.T.,
2000. Reac i e oxygen species gene a ed a mi ochond ial
complex III s abilize hypoxia-inducible ac o -1adu ing
hypoxia: a mechanism o O
2
sensing. J. Biol. Chem. 275,
25130–25138.
Chou, C.L., Shi aha a, M., 1996. Two ypes o ol age-ga ed
K
+
channels in ca o id body cells o adul ca s. B ain Res.
742, 34–42.
Cos a, L.E., Llesuy, S., Bo e is, A., 1993. Ac i e oxygen
species in he li e o a s submi ed o ch onic hypoba ic
hypoxia. Am. J. Physiol. 264, C1395–C1400.
C oss, A.R., Hende son, L., Jones, O.T.G., Delpiano, M.A.,
Hen schel, J., Acke , H., 1990. In ol emen on NAD(P)H
oxidase as a P
O
2
senso p o ein in he a ca o id body.
Biochem. J. 272, 743–747.
Dang, P.M., C oss, A.R., Babio , B.M., 2001. Assembly o he
neu ophil espi a o y bu s oxidase: a di ec in e ac ion
be ween p67PHOX and cy och ome b558. P oc. Na l.
Acad. Sci. USA 98, 3001–3005.
Da ies, M.J., Dean, R.T., 1997. Radical-Media ed P o ein
Oxida ion. F om Chemis y o Medicine. Ox o d Science
Publica ions, Ox o d, UK, pp. 25–120.
DiChia a, T.J., Reinha , P.H., 1997. Redox modula ion o
hslo Ca
2−
ac i a ed K
+
channels. J. Neu osci. 17, 4942–
4955.
Duchen, M.R., Biscoe, T.J., 1992. Rela i e mi ochond ial
memb ane po en ial and [Ca
2+
]
i
in ype I cells isola ed
om he abbi ca o id body. J. Physiol. London 450,
33–61.
Du an eau, J., Chandel, N.S., Kulisz, A., Shao, Z., Schu-
macke , P.T., 1998. In acellula signaling by eac i e oxy-
gen species du ing hypoxia in ca diomyocy es. J. Biol.
Chem. 273, 11619–11624.
D o ako a, M., Hohle , B., Volle hun, R., Fischbach, T.,
Kumme , W., 2000. Mac ophages: a majo sou ce o cy-
och ome b558 in he a ca o id body. B ain Res. 852,
349–354.
Ehleben, W., Po wol, T., Fand ey, J., Kumme , W., Acke ,
H., 1997. Cobal and des e ioxamine e eal c ucial mem-
be s o he oxygen sensing pa hway in HepG2 cells. Kidney
In . 51, 483–491.
Eu, J.P., Sun, J., Xu, L., S amle , J.S., Meissne , G., 2000. The
skele al muscle calcium elease channel: coupled O
2
senso
and NO signaling unc ions. Cell 102, 499–509.
Fand ey, J., F ede, S., Jelkmann, W., 1994. Role o hyd ogen
pe oxide in hypoxia-induced e y h opoie in p oduc ion.
Biochem. J. 303, 507–510.
Fand ey, J., F ede, S., Ehleben, W., Po wol, T., Acke , H.,
Jelkmann, W., 1997. Cobal chlo ide and des e ioxamine

C.Gonzalez e al.
/
Respi a o y Physiology & Neu obiology
132 (2002) 17–41
38
an agonize he inhibi ion o e y h opoie in p oduc ion by
eac i e oxygen species. Kidney In . 51, 492–496.
Fidone, S., Gonzalez, C, 1996. Ini ia ion and con ol o
chemo ecep o ac i i y in he ca o id body. In: Fishman,
A.P. (Ed.), Handbook o Physiology. The Respi a o y
Sys em. Ame ican Physiology Socie y, Be hesda, MD, pp.
247–312.
Fi h, J.D., Ebe , B.L., Pugh, C.W., Ra cli e, P.J., 1994.
Oxygen- egula ed con ol elemen s in he phosphogly ce -
a e kinase 1 and lac a e dehyd ogenase A genes: simila i ies
wi h he e y h opoie in 3%enhance . P oc. Na l. Acad. Sci.
USA 91, 6496–6500.
Fu, X.W., Nu se, C.A., Wang, Y.T., Cu z, E., 1999. Selec i e
modula ion o memb ane cu en s by hypoxia in in ac
ai way chemo ecep o s om neona al abbi . J. Physiol.
London 514, 139–150.
Fu, X.W., Wang, D., Nu se, C.A., Dinaue , M.C., Cu z, E.,
2000. NADPH oxidase is an O2 senso in ai way chemo e-
cep o s: e idence om K
+
cu en modula ion in wild- ype
and oxidase-de icien mice. P oc. Na l. Acad. Sci. USA 97,
4374–4379.
Gabig, T.G., Sche ish, E.W., San inga, J.T., 1982. Func ional
ela ionship o he cy och ome b o he supe oxide-gene a -
ing oxidase o human neu ophils. J. Biol. Chem. 257,
4114–4119.
Gan o nina, M.D., Lo´pez-Ba neo, J., 1991. Single K
+
chan-
nels in memb ane pa ches o a e ial chemo ecep o cells
a e modula ed by O
2
ension. P oc. Na l. Acad. Sci. USA
88, 2927–2930.
Gassmann, M., Chilo , D., Wenge , R.W., 2000. Regula ion
o he hypoxia-inducible ac o -1a. ARNT is no necessa y
o hypoxic induc ion o HIF-1ain he nucleus. Ad . Exp.
Med. Biol. 475, 87–99.
Geisz , M., Kopp, J.B., Va nai, P., Le o, T.L., 2000. Iden i i-
ca ion o enox, an NAD(P)H oxidase in kidney. P oc.
Na l. Acad. Sci. USA 97, 8010–8014.
Gleadle, J.M., Ebe , B.L., Ra cli e, P.J., 1995.
Diphenyleneiodonium inhibi s he induc ion o e y h opoi-
e in and o he mammalian genes by hypoxia. Eu . J.
Biochem. 234, 92–99.
Gnaige , E., Mendez, G., Hand, S.C., 2000. High phospho y-
la ion e iciency and dep ession o uncoupled espi a ion in
mi ochond ia unde hypoxia. P oc. Na l. Acad. Sci. USA
97, 11080–11085.
Goldbe g, M.A., Glass, G.A., Cunningham, J.M., Bunn, H.F.,
1987. The egula ed exp ession o e y h opoie in by wo
human hepa oma cell lines. P oc. Na l. Acad. Sci. USA 84,
7972–7976.
Goldbe g, M.A., Dunning, S.P., Bunn, H.F., 1988. Regula ion
o he e y h opoie in gene: e idence ha he oxygen senso
is a heme p o ein. Science 242, 1412–1415.
Gonzalez, C., 1998. Sensi i i y o physiologic hypoxia. In:
Lopez-Ba neo, J., Wei , E.K. (Eds.), Oxygen egula ion o
Ion Channels and Gene Exp ession. Fu u a Publishing Co
Inc, A monk, NY, pp. 321–336.
Gonzalez, C., Alma az, L., Obeso, A., Rigual, R., 1992.
Oxygen and acid chemo ecep ion in he ca o id body
chemo ecep o s. T ends Neu osci. 15, 146–153.
Gonzalez, C., Alma az, L., Obeso, A., Rigual, R., 1994.
Ca o id body chemo ecep o s: om na u al s imuli o sen-
so y discha ges. Physiol. Re . 74, 829–898.
Go lach, A., Hol e mann, G., Jelkmann, W., Hancock, J.T.,
Jones, S.A., Jones, O.T., Acke , H., 1993. Pho ome ic
cha ac e is ics o haem p o eins in e y h opoie in-p oduc-
ing hepa oma cells (HepG2). Biochem. J. 290, 771–776.
Go lach, A., Fand ey, J., Hol e mann, G., Acke , H., 1994.
E ec s o cobal on haem p o eins o e y h opoie in-p o-
ducing HepG2 cells in mul icellula sphe oid cul u e. FEBS
Le . 348, 216–218.
G imminge , F., Weissmann, N., Sp ies e sbach, R., Becke ,
E., Rosseau, S., Seege , W., 1995. E ec s o NADPH
oxidase inhibi o s on hypoxic asocons ic ion in bu e -
pe used abbi lungs. Am. J. Physiol. 268, L747–L752.
Guo, J., Giles, W.R., Wa d, C.A., 2000. E ec o hyd ogen
pe oxide on he memb ane cu en s o sinoa ial node cells
om abbi hea . Am. J. Physiol. 279, H992–H999.
Halliwell, B., Gu e idge, J.M.C., 1999. F ee Radicals in Biol-
ogy and Medicine, hi d ed. Ox o d Uni e si y P ess, Ox-
o d UK.
Halliwell, B., Gu e idge, J.M.C., 2000. F ee adicals and
an ioxidan in he yea , a his o ical look o he u u e.
Ann. NewYo k Acad. Sci. 899, 136–147.
Hayabuchi, Y., Nakaya, Y., Ma suoka, S., Ku oda, Y., 1998.
Hyd ogen pe oxide-induced ascula elaxa ion in po cine
co ona y a e ies is media ed by Ca
2+
-ac i a ed K
+
chan-
nels. Hea Vessels 13, 9–17.
Hochachka, P.W., Land, S.C., Buck, L.T., 1997. Oxygen
sensing and signal ansduc ion in me abolic de ense
agains hypoxia: lessons om e eb a e acul a i e anae -
obes. Comp. Biochem. Physiol. A Comp. Physiol. 118,
23–29.
Hoshi, T., Heinemann, S.H., 2001. Regula ion o cell unc ion
by me hionine oxida ion and educ ion. J. Physiol. London
531, 1–11.
Huang, L.E., A any, Z., Li ings on, D.M., Bunn, H.F., 1996.
Ac i a ion o hypoxia-inducible ansc ip ion ac o de-
pends p ima ily upon edox-sensi i e s abiliza ion o i s
alpha subuni . J. Biol. Chem. 271, 32253–32259.
Iizuka, T., Kanegasaki, S., Makino, R., Tanaka, T., Ishimu a,
Y., 1985. S udies on neu ophil b- ype cy och ome in si u
by low empe a u e abso p ion spec oscopy. J. Biol.
Chem. 260, 12049–12053.
Jelkmann, W., 1992. E y h opoie in: s uc u e, con ol o p o-
duc ion, and unc ion. Physiol. Re . 72, 449–489.
Joels, N., Neil, E., 1968. The idea o a senso y ansmi e . In:
To ance, R.W. (Ed.), A e ial Chemo ecep o s. Blackwell,
Ox o d, pp. 153–178.
Kie zmann, T., Fand ey, J., Acke , H., 2000. Oxygen adicals
as messenge s in oxygen-dependen gene exp ession. News
Physiol. Sci. 15, 202–208.
Kjae e, J., Veel, T., Bje naes, L., 1990. Allopu inol inhibi s
hypoxic pulmona y asocons ic ion. Role o oxic oxygen
me aboli es. Ac a Anaes hesiol. Scand. 34, 384–388.
Kumme , W., Acke , H., 1995. Immunohis ochemical demon-
s a ion o ou subuni s o neu ophil NAD(P)H oxidase
C.Gonzalez e al.
/
Respi a o y Physiology & Neu obiology
132 (2002) 17–41
39
in ype I cells o ca o id body. J. Appl. Physiol. 78,
1904–1909.
Lahi i, S., I u iaga, R., Mokashi, A., Ray, D.K., Chugh, D.,
1993. CO e eals dual mechanisms o O
2
chemo ecep ion
in he ca ca o id body. Respi . Physiol. 94, 227–240.
Lande , H.M., 1997. An essen ial ole o ee adicals and
de i ed species in signal ansduc ion. FASEB J. 11, 118–
124.
Liu, H., Moczydlowski, E., Haddad, G.G., 1999. O
2
dep i a-
ion inhibi s Ca
2+
-ac i a ed K
+
channels ia cy osolic
ac o s in mice neoco ical neu ons. J. Clin. In es . 104,
577–588.
Lo´pez-Ba neo, J., Lo´pez-Lo´pez, J.R., U en˜a, J., Gonzalez, C.,
1988. Chemo ansduc ion in he ca o id body: K
+
cu en
modula ed by P
O
2
in ype I chemo ecep o cells. Science
241, 580–582.
Lo´pez-Ba neo, J., Pa dal, R., Mon o o, R.J., Smani, T., Ga -
cia-Hi sch eld, J., U ena, J., 1999. K
+
and Ca
2+
channel
ac i i y and cy osolic [Ca
2+
] in oxygen-sen sing issues.
Respi . Physiol. 115, 215–227.
Lo´pez-Lo´pez, J.R., Gonzalez, C., 1992. Time cou se o K
+
cu en inhibi ion by low oxygen in chemo ecep o s cells o
adul abbi ca o id body. E ec s o ca bon monoxide.
FEBS Le . 299, 251–254.
Lo´pez-Lo´pez, J., Gonzalez, C., U en˜a, J., Lo´pez-Ba neo, J.,
1989. Low P
O
2
selec i ely inhibi s K
+
channel ac i i y in
chemo ecep o cells o he mammalian ca o id body. J.
Gen. Physiol. 93, 1001–1015.
Lo´pez-Lo´pez, J.R., Gonzalez, C., Pe ez-Ga cia, M.T., 1997.
P ope ies o ionic cu en s om isola ed adul a ca o id
body chemo ecep o cells. E ec s o hypoxia. J. Physiol.
London. 499, 429–441.
Ma shall, B.E., Ma shall, C., F asch, F., Hanson, C.W., 1994.
Role o hypoxic pulmona y asocons ic ion in pulmona y
gas exchange and blood low dis ibu ion. 1. Physiologic
concep s. In ensi e Ca e Med. 20, 291–297.
Ma shall, C., Mama y, A.J., Ve hoe en, A.J., Ma shall, B.E.,
1996. Pulmona y a e y NADPH-oxidase is ac i a ed in
hypoxic pulmona y asocons ic ion. Am. J. Respi . Cell
Mol. Biol. 15, 633–644.
Ma sumo o, S., Nagamine, T., Yaka a, H., Kohno, M., 1987.
Al e a ion o chemo e lex ph enic esponses by oligomycin
in he abbi . A ch. In . Pha macodyn. The . 286, 136–
144.
McLennan, H., 1963, Synap ic ansmission. Saunde s.
Philadelphia, USA.
Mohazzab, K.M., Wolin, M.S., 2000. Oxidan signalling and
ascula oxygen sensing. Role o H
2
O
2
in esponses o he
bo ine pulmona y a e y o changes in P
O
2
. Ad . Exp.
Med. Biol. 475, 249–258.
Mulligan, E., Lahi i, S., 1982. Sepa a ion o ca o id body
chemo ecep o esponses o O
2
and CO
2
by oligomycin
and by an imycin A. Am. J. Physiol. 242, C200–C206.
Nakanishi, K., Tajima, F., Nakamu a, A., Yagu a, S.,
Ookawa a, T., Yamashi a, H., Suzuki, K., Taniguchi, N.,
Ohno, H., 1995. E ec s o hypoba ic hypoxia on an ioxi-
dan enzymes in a s. J. Physiol. London 489, 869–876.
Ndele, J.K., Yoshioka, K., Fishe , J.W., 1996. Hyd ogen
pe oxide in he egula ion o e y h opoie in (Epo) gene
exp ession in hepa ocellula ca cinoma cells. Eas A .
Med. J. 73, 143–146.
Obeso, A., 1984. Hipo esis me abolica de quimio ansduccion:
hechos expe imen ales. (Ph. D Thesis). Uni e sidad de
Valladolid, Valladolid, Spain.
Obeso, A., Alma az, L., Gonzalez, C., 1989a. E ec s o cya-
nide and uncouple s on chemo ecep o ac i i y and ATP
con en o he ca ca o id body. B ain Res. 481, 250–257.
Obeso, A., Gonzalez, C., Dinge , B., Fidone, S., 1989b.
Me abolic ac i a ion o ca o id body glomus cells by hy-
poxia. J. Appl. Physiol. 67, 484–487.
Obeso, A., Gonzalez, C., Rigual, R., Dinge , B., Fidone, S.,
1993. E ec o low O
2
on glucose up ake in abbi ca o id
body. J. Appl. Physiol. 74, 2387–2393.
Obeso, A., Roche , A., He e os, B., Gonzalez, C., 1997.
Oxygen consump ion and ene gy me abolism o he ca o id
body. In: Gonzalez, C. (Ed.), The Ca o id Body Chemo e-
cep o s. Sp inge , NY, pp. 31–45.
Obeso, A., Gomez-Nin˜o, A., Gonzalez, C., 1999. Inhibi ion o
NADPH oxidase does and no in e e e wi h low P
O
2
ansduc ion in chemo ecep o cells. Am. J. Physiol. 276,
C593–C601.
Obeso, A., Sanz-Al aya e, G., Agapi o, M.T., Gonzalez, C.,
2000a. Signi icance o ROS in oxygen chemo ecep ion in
he ca o id body chemo ecep ion. Appa en lack o a ole
o NADPH oxidase. Ad . Exp. Med. Biol. 475, 425–434.
Obeso, A., Sanz-Al aya e, G., Agapi o, M.T., Gonzalez, C.,
2000b. Glu a hione le els and GSH-GSSG a ios in he
ca o id body and oxygen sensing in chemo ecep o cells.
High Al i ude Med. Biol. 1, 255.
Ohkuma, S., Ka su a, M., 2001. Ni ic oxide and pe oxyni i e
as ac o s o s imula e neu o ansmi e elease in he CNS.
P og. Neu obiol. 64, 97–108.
Osanai, S., Mokashi, A., Rozano , C., Bue k, D.G., Lahi i, S.,
1997. Po en ial ole o H
2
O
2
in chemo ecep ion in he ca
ca o id body. J. Au on. Ne . Sys . 63, 39–45.
Osipenko, O.N., Ta e, R.J., Gu ney, A.M., 2000. Po en ial
ole o k 3.1b channels as oxygen senso s. Ci c. Res. 86,
534–540.
Paky, A., Michael, J.R., Bu ke-Wolin, T.M., Wolin, M.S.,
Gu ne , G.H., 1993. Endogenous p oduc ion o supe ox-
ide by abbi lungs: e ec s o hypoxia o me abolic in-
hibi o s. J. Appl. Physiol. 74, 2868–2874.
Pa el, A.J., Lazdunski, M., Hono e, E., 1997. K 2.1/K 9.3, a
no el ATP-dependen delayed- ec i ie K
+
channel in oxy-
gen-sensi i e pulmona y a e y myocy es. EMBO J. 16,
6615–6625.
Pee s, C., 1990. Hypoxic supp ession o K
+
cu en s in ype I
ca o id body cells: selec i e e ec on he Ca
2+
-ac i a ed
K
+
cu en . Neu osci. Le . 119, 253–256.
Pe´ ez-Ga cı´a, M.T., Lo´pez-Lo´pez, J.R., 2000. A e K chan-
nels he essence o O
2
sensing. Ci c. Res. 86, 490–491.
Pe ez-Ga cı´a, M.T., Lo´pez-Lo´pez, J.R., Gonzalez, C., 1999.
K b1.2 subuni i modula es he sensi i i y o hypoxia o
cloned ol age-ga ed K
+
channels in a sub amily-speci ic
manne . J. Gen. Physiol. 113, 897–907.
C.Gonzalez e al.
/
Respi a o y Physiology & Neu obiology
132 (2002) 17–41
40
Pe´ ez-Ga cı´a, M.T., Lo´pez-Lo´pez, J.R., Riesco, A.M., Hoppe,
U., Gonzalez, C., Ma ban, E., Johns, D.C., 2000. Sup es-
sion o ansien ou wa d K
+
cu en s in chemo ecep o
cells o he abbi ca o id body by i al gene ans e o
inducible dominan nega i e K 4.3 cons uc s. J. Neu osci.
20, 5689–5695.
Pos , J.M., Hume, J.R., A che , S.L., Wei , E.K., 1992. Di ec
ole o po assium channel inhibi ion in hypoxic pul-
mona y asocons ic ion. Am. J. Physiol. 262, C882–C890.
Pugh, C.W., Tan, C.C., Jones, R.W., Ra cli e, P.J., 1991.
Func ional analysis o an oxygen- egula ed ansc ip ional
enhance lying 3’ o he mouse e y h opoie in gene. P oc.
Na l. Acad. Sci. USA 88, 10553–10557.
Puncha d, N.A., Kelly, F.J., 1996. F ee Radicals. A P ac ical
App oach. IRL P ess-Ox o d Uni e si y P ess, Ox o d,
UK, pp. 1–8.
Ra cli e, P.J., Ebe , B.L., Fe guson, D.J., Fi h, J.D.,
Gleadle, J.M., Maxwell, P.H., Pugh, C.W., 1995. Regula-
ion o he e y h opoie in gene. Neph ol. Dial. T ansplan .
10 (Suppl 29), 18–27.
Richale , J.P., 1997. Oxygen senso s in he o ganism: examples
o egula ion unde al i ude hypoxia in mammals. Comp.
Biochem. Physiol. A Physiol. 118, 9–14.
Riesco, A., Pe e´z-Ga cia, M.T., Gonzalez, C., Lo´pez-Lo´pez,
J.R., 2001. Ca
2+
-dependen K
+
channels a e modula ed
by hypoxia and CO in isola ed memb ane pa ches o a
chemo ecep o cells. Biophys. J. 80, 206a.
Roche , A., Obeso, A., He e os, B., Gonzalez, C., 1988.
Ac i a ion o he elease o dopamine in he ca o id body
by e a idine. E idence o he p esence o ol age-depen-
den Na
+
channels in ype I cells. Neu osci. Le . 94,
274–278.
Roy, A., Rozano , C., Mokashi, A., Daudu, P., Al-mehdi,
A.B., Shams, H., Lahi i, S., 2000. Mice lacking in gp91
phox subuni o NAD(P)H oxidase showed glomus cell
[Ca
2+
]
i
and espi a o y esponses o hypoxia. B ain Res.
872, 188–193.
Ruppe sbe g, J.P., S ocke , M., Pongs, O., Heinemann, S.H.,
F ank, R., Koenen, M., 1991. Regula ion o as inac i a-
ion o cloned mammalian IK(A) channels by cys eine
oxida ion. Na u e 352, 711–714.
Sande ud, J., Bjo o, K., Saugs ad, O.D., 1993. Oxygen adi-
cals s imula e h omboxane and p os acyclin syn hesis and
induce asocons ic ion in pig lungs. Scand. J. Clin. Lab.
In es . 53, 447–455.
Seege , W., Su o p, N., Schmid , F., Neuho , H., 1986. The
glu a hione edox cycle as a de ense sys em agains hyd o-
gen-pe oxide-induced p os anoid o ma ion and asocon-
s ic ion in abbi lungs. Am. Re . Respi . Dis. 133,
1029–1036.
Semenza, G.L., 1994. Regula ion o e y h opoie in p oduc-
ion. New insigh s in o molecula mechanisms o oxygen
homeos asis. Hema ol. Oncol. Clin. No h Am. 8, 863–
884.
Semenza, G.L., 1999. Pe spec i es on oxygen sensing. Cell 98,
281–284.
Semenza, G.L., 2000. Chai man’s summa y: mechanisms o
oxygen homeos asis, ci ca 1999. Ad . Exp. Med. Biol. 475,
303–310.
Shin, J.H., Lee, J.E., Pa k, J.M., Suh, C.K., 2000. T-bu yl
hyd ogen pe oxide inc eases he ac i i ies o he Maxi-K
channels o a b ain. Li e Sci. 67, 2485–2491.
Simon-Schnass, I.M., 2000. Risk o oxida i e s ess a high
al i ude and possible bene i o an ioxidan supplemen a-
ion. In: Sen, C.K., Packe , L., Hannimen, O. (Eds.),
Handbook o Oxidan s and An ioxidan s in Exe cise. El-
se ie , Ams e dam, The Ne he lands, pp. 555–577.
Soh, R., Jung, W., Uhm, D.Y., Chung, S., 2001. Modula ion
o la ge conduc ance calcium-ac i a ed po assium channels
om a hippocampal neu ons by glu a hione. Neu osci.
Le . 298, 115–118.
S ini as, V., Zhang, L.P., Zhu, X.H., Ca o, J., 1999. Cha ac-
e iza ion o an oxygen/ edox-dependen deg ada ion do-
main o hypoxia-inducible ac o alpha (HIF-a) p o eins.
Biochem. Biophys. Res. Commun. 260, 557–561.
Tamayo, L., Lo´pez-Lo´pez, J.R., Cas an˜eda, J., Gonza´lez, C.,
1997. Ca bon monoxide inhibi s hypoxic pulmona y aso-
cons ic ion by a cGMP-independen mechanism. P lu¨ge s.
A ch. Eu . J. Physiol. 434, 698–704.
Tan, C.C., Ra cli e, P.J., 1991. E ec o inhibi o s o oxida-
i e phospho yla ion on e y h opoie in mRNA in isola ed
pe used a kidneys. Am. J. Physiol. 261, F982–F987.
Tang, X.D., Dagge , H., Hanne , M., Ga cia, M.L., Mc-
Manus, O.B., B o , N., Weissbach, H., Heinemann, S.H.,
Hoshi, T., 2001. Oxida i e egula ion o la ge conduc ance
calcium-ac i a ed po assium channels. J. Gen. Physiol.
117, 253–274.
Thomas, H.M., Ca son, R.C., F ied, E.D., No i ch, R.S.,
1991. Inhibi ion o hypoxic pulmona y asocons ic ion by
diphenyleneiodonium. Biochem. Pha macol. 42, R9–R12.
Thompson, J.S., Jones, R.D., Roge s, T.K., Hancock, J.,
Mo ice, A.H., 1998. Inhibi ion o hypoxic pulmona y
asocons ic ion in isola ed a pulmona y a e ies by
diphenyleneiodonium (DPI). Pulm. Pha macol. The . 11,
71–75.
Thu inge , D., Findlay, I., 1997. Con as ing e ec s o in a-
cellula edox couples on he egula ion o maxi-K
+
chan-
nels in isola ed myocy es om abbi pulmona y a e y. J.
Physiol. London 500, 583–592.
Ueno, M., B ookins, J., Beckman, B.S., Fishe , J.W., 1988.
E ec s o eac i e oxygen me aboli es on e y h opoie in
p oduc ion in enal ca cinoma cells. Biochem. Biophys.
Res. Commun. 154, 773–780.
Vega-Saenz de Mie a, E., Rudy, B., 1992. Modula ion o K
+
channels by hyd ogen pe oxide. Biochem. Biophys. Res.
Commun. 186, 1681–1687.
Ve na, A., Talib, N., Roumy, M., P ade , A., 1990. E ec s o
me abolic inhibi o s and hypoxia on he ATP, ADP and
AMP con en o he abbi ca o id body in i o: he
me abolic hypo hesis in ques ion. Neu osci. Le . 116,
156–161.
Wang, G.L., Semenza, G.L., 1993a. Cha ac e iza ion o hy-
poxia-inducible ac o 1 and egula ion o DNA binding
ac i i y by hypoxia. J. Biol. Chem. 268, 21513–21558.
C.Gonzalez e al.
/
Respi a o y Physiology & Neu obiology
132 (2002) 17–41
41
Wang, G.L., Semenza, G.L., 1993b. Gene al in ol emen o
hypoxia-inducible ac o 1 in ansc ip ional esponse o
hypoxia. P oc. Na l. Acad. Sci. USA 90, 4304–4308.
Wang, D., Youngson, C., Wong, V., Yege , H., Dinaue ,
M.C., Vega-Saenz Mie a, E., Rudy, B., Cu z, E., 1996.
NADPH-oxidase and a hyd ogen pe oxide-sensi i e K
+
channel may unc ion as an oxygen senso complex in
ai way chemo ecep o s and small cell lung ca cinoma cell
lines. P oc. Na l. Acad. Sci. USA 93, 13182–13187.
Wei , E.K., A che , S.L., 1995. The mechanism o acu e
hypoxic pulmona y asocons ic ion: he ale o wo chan-
nels. FASEB J. 9, 183–189.
Wei , E.K., Wya , C.N., Ree e, H.L., A che , S.L., Pee s, C.,
1994. Diphenyleneiodonium inhibi s bo h K
+
and Ca
2+
cu en s in isola ed pulmona y a e y smoo h muscle cells.
J. Appl. Physiol. 76, 2611–2615.
Weissmann, N., G imminge , F., Voswinckel, R., Conzen, J.,
Seege , W., 1998. Ni o blue e azolium inhibi s bu does
no mimic hypoxic asocons ic ion in isola ed abbi
lungs. Am. J. Physiol. 274, L721–L727.
Wenge , R.H., Ma i, H.H., Schue e -Maly, C.C., K ie iko a,
I., Baue , C., Gassmann, M., Maly, F.E., 1996. Hypoxic
induc ion o gene exp ession in ch onic g anuloma ous
disease-de i ed B-cell lines: oxygen sensing is independen
o he cy och ome b558-con aining nico inamide adenine
dinucleo ide phospha e oxidase. Blood 87, 756–761.
Wolin, M.S., Bu ke-Wolin, T.M., Mohazzab-H, K.M., 1999.
Roles o NAD(P)H oxidases and eac i e oxygen species
in ascula oxygen sensing mechanisms. Respi . Physiol.
115, 229–238.
Wya , C.N., Pee s, C., 1995. Ca
2+
-ac i a ed K
+
channels in
isola ed ype I cells o he neona al a ca o id body. J.
Physiol. London 483, 559–565.
Wya , C.N., Wei , E.K., Pee s, C., 1994. Diphenyleneiodo-
nium blocks K
+
and Ca
2+
cu en s in ype I cells isola ed
om neona al a ca o id body. Neu osci. Le . 172, 63–
66.
Youngson, C., Nu se, C., Yege , H., Cu z, E., 1993. Oxygen
sensing in ai way chemo ecep o s. Na u e 365, 153–155.
Yuan, X.J., Goldman, W.F., Tod, M.L., Rubin, L.J.,
Blaus ein, M.P., 1993. Hypoxia educes po assium cu en s
in cul u ed a pulmona y bu no mesen e ic a e ial my-
ocy es. Am. J. Physiol. 264, L116–L123.
Yuan, X.J., Wang, J., Juhaszo a, M., Golo ina, V.A., Rubin,
L.J., 1998. Molecula basis and unc ion o ol age-ga ed
K
+
channels in pulmona y a e ial smoo h muscle cells.
Am. J. Physiol. 274, L621–L635.
Zhang, F., Ca son, R.C., Zhang, H., Gibson, G., Thomas,
H.M., 1997. Pulmona y a e y smoo h muscle cell [Ca
2+
]
i
,
and con ac ion; esponses o diphenyleneiodonium and
hypoxia. Am. J. Physiol. 273, L603–L611.
Zhu, H., Qiu, H., Yoon, H.W., Huang, S., Bunn, H.F., 1999.
Iden i ica ion o a cy och ome b- ype NAD(P)H oxido e-
duc ase ubiqui ously exp essed in human cells. P oc. Na l.
Acad. Sci. USA 96, 14742–14747.