REVIEW ARTICLE
Edua do Villalobo áPila Pe
 ez-Rome o
RocõÂoSa
Ânchez-Sil a áAn onio To es
Unusual cha ac e is ics o cilia e ac ins
Recei ed: 30 Ap il 2001 / Accep ed: 15 June 2001 / Published online: 10 No embe 2001
ÓSp inge -Ve lag and SEM 2001
Abs ac Ac in is a cy oskele al p o ein ha is ubiqui-
ous in euka yo es, hence he co esponding genes and
p o eins ha e been isola ed om nume ous o ganisms as
die en as animals, plan s, ungi and p o ozoa. Se e al
a omic models a e a ailable o he monome ic as well
as he ®lamen ous o m, and mo e han 70 p o eins ha
bind ac in and con ol ®lamen dynamics ha e been
isola ed om di e se euka yo es. Mo eo e , he unc-
ion and dynamics o he ac in cy oskele on in se e al
euka yo ic sys ems ha e been depic ed in dep h. Unlike
o he p o ozoa, such as amoeba, ac in is no an abun-
dan p o ein in cilia es, whose cy oskele on is mainly
composed o mic o ubula a ays. Cilia e ac in has been
s udied in se e al species, and i was es ablished ea ly on
ha his cilia e p o ein is e y die en om ha o
o he euka yo es. Simila ly, he ac in-binding p o eins
s udied in cilia es display g ea die ences wi h hose o
o he euka yo es. Consequen ly, cilia e ac in has been
conside ed as ``uncon en ional,'' and his e iew ocuses
on molecula da a leading o his conclusion.
Keywo ds Ac in áAc in-binding p o ein áCilia e á
Cy oskele on
In oduc ion
Ac in is one o he mos highly conse ed and bes
s udied euka yo ic p o eins. I may be p esen as
monome ic G-ac in in he absence o sal , o as mic o-
®lamen s (F-ac in) in he p esence o physiological
concen a ions o sal . Monome ic ac in binds 1 mol o
ATP and one di alen ca ion pe mol o p o ein. I is a
Mg
2+
-s imula ed ATPase, bu a low sal concen a-
ions, in which i does no polyme ise, his ATPase
ac i i y is low. Unde polyme isa ion condi ions,
monome s a e inco po a ed in o ®lamen s in he ATP
s a e and he nucleo ide is hyd olysed o ADP, al hough
®lamen assembly is no dependen on his hyd olysis.
Ac in p o ein is a majo cellula componen o bo h
he cy oskele on and he muscle sa come e. In muscle
cells, ac in is in ol ed in myo®b illa cons uc ion. In
non-muscle cells, he ac in cy oskele on plays an essen-
ial ole in mul iple cellula p ocesses, including cell
elonga ion and shape de e mina ion, cy oplasmic
s eaming, cell mo ili y, di ision-plane localisa ion,
ch omosome seg ega ion, sec e ion, endocy osis, and
o ganelle anspo [45]. All o hese p ocesses ely on
he capaci y o he ac in cy oskele on o espond o
cellula signals and eo ganise spa ially and empo ally
in o a a ie y o speci®c s uc u es.
The 42-kDa G-ac in monome ypically consis s o
375 amino acid esidues. P ima y s uc u es o ac ins
om die en euka yo ic species exhibi excep ionally
high sequence iden i ies. The e a e a leas six die en
iso o ms o ac in in mammals; wo cy oplasmic ac ins, b
and c, and ou muscle ac ins. The la e include wo
s ia ed muscle (a-skele al and a-ca diac) and wo
smoo h muscle (a-ao ic and c-en e ic) ac ins. In mos
euka yo es, ac in is encoded by a mul igene amily [7],
al hough o ganisms such as yeas s [19], Gia dia [13], and
se e al ¯agella ed p o is a [2] con ain only a single ac in
gene. The mul igene amilies a e likely o ha e a isen
om gene duplica ions ollowed by di e gence o he
duplica ed copies du ing e olu ion. The die en ac in
iso o ms p oduced by he membe s o he ac in gene
amilies a e ypically highly conse ed a he amino acid
sequence le el, appa en ly due o hei c i ical unc ional
oles in cells.
C ys allog aphic da a [5, 29,48] indica e ha ac in
monome is a globula p o ein consis ing o wo do-
mains connec ed by a hinge egion. These wo domains,
o iginally e med la ge and small (al hough now hey a e
In Mic obiol (2001) 4: 167±174
DOI 10.1007/s10123-001-0032-1
E. Villalobo áP. Pe
 ez-Rome o
R. Sa
Ânchez-Sil a áA. To es (&)
Depa amen o de Mic obiologõÂa,
Facul ad de BiologõÂa. Uni e sidad de Se illa.
Apdo 1095. 41080 Se ille, Spain
E-mail: [email p o ec ed]
Tel.: +34-954557115
Fax: +34-954557830
known o ha e almos iden ical sizes), a e u he di-
ided in o subdomains. The small domain is composed
o subdomain 1 (amino acid esidues 1±32, 70±144, 338±
375, abbi a-ac in numbe ing) and subdomain 2 (amino
acid esidues 33±69), whe eas he la ge domain com-
p ises subdomain 3 (amino acid esidues 145±180 and
270±337) and subdomain 4 (amino acid esidues 181±
269). The la ge subdomain has a s uc u al co e made up
o esidues om bo h subdomains 3 and 4, whe eas he
small-domain co e is made up exclusi ely o esidues
om subdomain 1. Subdomains 1 and 3 a e hough o
ha e e ol ed by duplica ion o an ances al gene coding
o a polypep ide o abou 150 amino acids ha appea s
o be an ancien nucleo ide-binding pocke also ound in
hea shock p o eins, suga kinases and se e al poly-
pep ides known o egula e he p oka yo e cell cycle [3].
Subdomains 2 and 4 we e p obably inse ed subse-
quen ly in o subdomains 1 and 3, espec i ely.
The nucleo ide/ca ion complex is bound a he bo -
om o he cle be ween subdomains 1 and 3, wi h he
adenosine base es ing in a hyd ophobic pocke o med
be ween subdomains 3 and 4. The e a e 15 esidues in-
ol ed in nucleo ide-binding si es, o which 13 belong o
subdomains 1 and 3, and wo a e wi hin subdomain 4.
The egion o con ac wi h DNase I consis s o hyd ogen
bonds, elec os a ic and hyd ophobic in e ac ions in-
ol ing subdomain 2 (amino acid esidues 39±46; 60±64)
and subdomain 4 (amino acid esidues 202±204 and
207). Residues 40±50 wi hin subdomain 2 a e highly
diso de ed and o m he DNase I loop. Ac in-ac in in-
e ac ions in he ®lamen in ol e abou 45 esidues.
Subdomains 2 and 4 o he ac in monome ha e ac in-
ac in in e aces in e ac ing wi h subdomains 1 and 3 o
o he ac in monome s. Se e al amino acid esidues in
subdomain 2, including a egion o he sequence ha is
nea ly iden ical o ha o he DNase I con ac , ha e
been shown o be in con ac wi h o he ac in monome s.
The ac in helix is s abilised h ough a loop o 11 amino
acids (262±272) ha includes a ou - esidue hyd ophobic
plug. This loop inse s in o a hyd ophobic pocke
o med by subdomains 2 and 3 o wo adjacen mono-
me s on he opposing s and.
In he cell, many ac in-binding p o eins (ABPs)
egula e he nuclea ion and assembly-disassembly o
ac in in o ®lamen . Mos amilies o ABPs ha e been
widely conse ed o e phylogeny, in bo h p ima y
s uc u e and biochemical p ope ies, and hey can be
ound in o ganisms as di e se as humans and yeas .
This sugges s ha hese p o eins al eady exis ed in a
common euka yo ic ances o and ha he basic
mechanisms egula ing he dynamics o he ac in cy-
oskele on a e conse ed among di e se o ganisms and
cell ypes. Mo e han 70 ABPs ha can con ol ®la-
men dynamics ha e been iden i®ed [35,37]. Some a -
ec monome s by con olling seques a ion o
nucleo ide exchange; o he s con ol ®lamen o ma ion
and s abili y by egula ing capping, nuclea ing, c oss-
linking, bundling, and se e ing. Finally, cellula sig-
nalling molecules and small GTPases con ol ac in
emodelling by egula ing he ac i i ies o ABPs and
nume ous di ec and indi ec eec o s [36,44].
Cilia e ac in
This cy oskele al p o ein has been s udied in se e al
species, al hough esul s a e sca ce, some imes con a-
dic o y, and gene ally es ic ed o Te ahymena and
Pa amecium. The ollowing sec ions a e dedica ed o
e iewing he mos ele an knowledge on cilia e ac in.
The p o ein
The ® s a emp s o demons a e he p esence o ac in
in cilia es we e ca ied ou by indi ec echniques, such
as hea y me omyosin (HMM) deco a ion o immun-
ode ec ion. Thus, ac in was localised in he co ex o
Discoph ya [23], Pa amecium [49] and Te ahymena [31],
and in he o al appa a us o Pseudomic o ho ax [25].
La ely, Pa amecium ac in was ound in ood acuoles,
a ound con ac ile acuoles [9], and in he epiplasm [8].
Howe e , i was no de ec ed deepe in he co ex, as
p e iously sugges ed. In Te ahymena, ac in was also
ound in he ood and con ac ile acuole, in he di ision
u ow [38], and in he basal body-cage complex [28].
The link be ween cilia e ac in and basal bodies is no an
excep ion, since ac in has also been epo ed o be as-
socia ed wi h axonemes in algae [43] and bi ds [47], and
wi h cen osomes in e eb a es [6]. E idence o he as-
socia ion o cilia e ac in wi h basal bodies has been
ob ained by us using an an ibody aised agains cen -
ac in, a cen osome-associa ed ac in. In Pa amecium,
his an i-cen ac in an ibody labelled basal bodies no
only in he co ical ows bu also in he o al appa a us
(ou unpublished da a).
Ac in has been pu i®ed in Climacos omum [18], Pa -
amecium [49] and Te ahymena [27]. In hese cilia es, he
pu i®ed ac in co esponds o a p o ein o abou 43 kDa,
i some con using epo s on Te ahymena ac in a e no
conside ed [39,41]. Con en ional me hods we e used o
pu i y ac in in Pa amecium and Te ahymena, whe eas
DNase I-ani y ch oma og aphy was used in Climaco-
s omum. Using his las me hod, a 43-kDa p o ein was
also isola ed in Pa amecium [8], bu he au ho s ques-
ioned whe he his p o ein was ac ually ac in. In ac ,
al hough i s pep ide mapping did no coincide wi h ha
o abbi muscle ac in, an an ibody aised agains his
43-kDa p o ein [8] deco a ed he epiplasm, he same
s uc u e deco a ed by a monoclonal an ibody agains
a skele al ac in [34].
Fa om shedding ligh on cilia e ac in, hese esul s
pose new ques ions as o i s biochemical na u e. Ne -
e heless, o da e a de ailed biochemical cha ac e isa ion
o ac in is a ailable only in Te ahymena [27]. In
some espec s, Te ahymena ac in shows biochemical
p ope ies simila o ha o abbi muscle ac in, namely:
(1) i polyme ises in i o in o mic o®lamen s in an
168
ion-dependen -manne ; (2) i o ms a owhead s uc-
u es wi h HMM; and (3) i ac i a es he Mg
+2
-ATPase
o myosin S-1. Howe e , Te ahymena ac in shows un-
usual p ope ies in wo espec s: ® s , i does no bind o
phalloidin; second, i does no inhibi DNase I ac i i y.
This la e ea u e o Te ahymena ac in aises he pos-
sibili y ha he pu i®ca ion o Pa amecium and Cli-
macos omum ac in by DNase-I ani y ch oma og aphy
is a e ac ual. This is suppo ed by he ac ha he
DNase-I-pu i®ed ac in o Climacos omun o ms mic o-
®lamen s in i o bu independen ly o sal concen a-
ion. All hese unusual p ope ies allow us o assume
ha , o e all, cilia e ac in can be conside ed uncon en-
ional. This would explain in pa why many an ibodies
aised agains con en ional ac in ail o ecognise any
p o ein in he 43 kDa ange in o al p o ein ex ac s o
many cilia e species. Fo ins ance, in o al p o ein ex-
ac s, he an ibody o Lin [34] did no de ec any
polypep ide in Pa amecium and Te ahymena, al hough
he an ibody ecognised a polypep ide o abou 43 kDa
in Climacos omum [18] and, su p isingly, a se o poly-
pep ides anging om abou 110 o 97 kDa in Euplo es
(ou unpublished esul s).
The genes
Ac in genes ha e been sequenced in many species, de-
spi e he ac ha he p o ein has been isola ed in only a
ew cilia e species. As in yeas s, he numbe o mac o-
nuclea ac in genes in cilia es is low, anging om one in
His iculus [42] and Te ahymena [10,26] o h ee in
Oxy icha [12,30] and S ylonychia [24]. The size o he
ac in mac onuclea molecules in hypo ichs, in which he
size o he whole gene uni can be ob ained, anges om
abou 1,200 bp o abou 1,600 bp, excluding elome e
epe i ions. The mic onuclea coun e pa s o he mac-
onuclea ac in genes ha e been sequenced only in
Oxy icha and U os yla. In e es ingly, in he mic o-
nuclea DNA o Oxy icha i allax, en mac onuclea -
des ined sequences (MDSs) co esponding o ac in I a e
dispe sed and diso de ed in he ch omosome, a phe-
nomenon called ``sc ambling'' [14]. Sc ambled MDSs
mus be ea anged du ing mac onuclea de elopmen
o o m a unc ional mac onuclea molecule.
Mac onuclea ac in genes a e ela i ely A+T- ich
(mean o abou 56%), al hough his ichness depends on
he egion conside ed and he species analysed. In gen-
e al, he A+T con en in non-coding egions (5¢- and 3¢-
un ansla ed egion and in ons) is highe han in he
coding egions. This A+T con en is especially high (75±
91%) in he wo sho in ons p esen in Pa amecium
[11], Vo icella and Opis honec a ac ins [46], he only
a ailable cilia e ac in genes known o be in e up ed by
in ons. No e ha , in Pa amecium, one in on is loca ed
wi hin subdomain 1 and he o he one wi hin subdomain
3, which suppo s he idea ha bo h domains a ose by
duplica ion o an ancien polypep ide and/o by exon
shuing.
A p omo e sequence mo i (TATA-box) can be
ound in mos bu no all he a ailable ac in sequences,
bu whe he his mo i is di ec ly implied in ansc ip ion
is no known. In His iculus ca icola [42], he e a e wo
po en ial TATA boxes (40 and 65 bp ups eam o he
ini ia ion codon) in i s unique ac in gene, and he ini i-
a ion o ansc ip ion occu s in he ou h nucleo ide (A)
o he ® s mo i (posi ion ±40). The e o e, i is assumed
ha he second mo i (posi ion ±65) migh be he ac ual
p omo e . This second mo i (posi ion ±65) is 22 bp
ups eam o he ansc ip ion ini ia ion, esembling ha
desc ibed in Euplo es c assus ac in (a po en ial TATA-
box is si ua ed 21 bp ups eam o he ansc ip ion
ini ia ion, which is also an A [22]). No e ha he 64
nucleo ides ups eam o he ini ia ion codon a e a he
simila in H. ca icola,O. allax, and O. i allax ac in
sequences, and ha a pe ec alignmen can be p oduced
a ound he TATA box [42].
Rela ed o he polyadenyla ion consensus sequence
(AATAAA), i is uncommon o ®nd pe ec ma ches in
cilia e ac in genes, bu his is no su p ising, as yeas and
plan s would no equi e his mo i o add he poly(A)
ac [50]. In ac , in E. c assus, whe e no pe ec
polyadenyla ion signal can be ecognised in ac in, wo
die en polyadenyla ion si es ha e been desc ibed.
These co espond o wo die en de elopmen al s ages,
al hough whe he polyadenyla ion addi ion o a pa ic-
ula si e is ela ed o a pa icula s age o de elopmen
emains o be de e mined [22].
The exp ession le el o ac in has been examined in a
ew cases. In Te ahymena, o ins ance, ac in seems o
be ac i ely exp essed, bu wi h li le ¯uc ua ion
h oughou he cell cycle [54]. In S e kiella his iomus-
co um, ou p e ious esul s indica ed ha ac in is also
exp essed ac i ely, al hough o a lesse ex en han
a- ubulin. Mo eo e , ac in is down- egula ed du ing
s a a ion ( eaching a minimum in he cys ) and up-
egula ed du ing excys men , p obably un il he le el o
he ege a i e cell is eached.
The p ima y amino acid sequence
A his le el, he p ima y amino acid sequences o cilia e
ac ins shows a ela i ely low le el o iden i y wi h hose
o ac ins deposi ed in da abases. In some cases, cilia e
ac ins a e no mo e simila o each o he han o hose o
o he o ganisms. Fo example, Pa amecium ac in is ap-
p oxima ely 71% and 61.3% iden ical o A abidopsis
haliana and Oxy icha i allax ac ins, espec i ely. In
he same manne , Pa amecium ac in is mo e simila o
ac in o an ea ly di e ging p o ozoan (58.3% iden i y
wi h Gia dia lamblia ac in) han o he ac in o a membe
o he cilia es (56.5% iden i y wi h Euplo es c assus ac-
in). This si ua ion is accen ua ed in Oxy icha no a,
which has h ee highly di e gen ac in genes (ac in II
and ac in III a e only 50% iden ical). These s iking
die ences among cilia e ac in sequences show he high
e olu iona y a e o ac in wi hin he phylum Ciliopho a.
169
As a consequence o his unusually apid a e o e olu-
ion, cilia es appea o be polyphyle ic, eme ging qui e
low in he ac in phylogene ic ee [13]. Howe e , a
weal h o mo phological e idences and phylogene ic
ees o RNA and ubulins [1] unques ionably demon-
s a es ha cilia es cons i u e a solid monophyle ic
g oup. Maybe ac in, in con as o ubulin, is poo ly
used in cilia es. I his is ue, ac in would ha e less
unc ional cons ain s and, he e o e, i could e ol e a a
high mu a ion a e. This assump ion could explain, in
pa , why cilia e ubulins a e so simila in hei amino
acid sequences, whe eas cilia e ac ins a e no . I could
also explain he unusual cha ac e is ics o cilia e ac in,
which migh ha e e ol ed new unc ions.
The e ia y s uc u e
Since he X- ay s uc u es o monome ic ac in ha e
become a ailable [5, 29,48], i has been possible o de-
e mine p ecisely which amino acids a e signi®can o
any gi en ac in p ope y, such as binding o di alen
ca ion, o nucleo ide, o o DNase I. Supe imposi ion o
cilia e ac in sequences on he h ee-dimensional s uc-
u e o abbi ac in has allowed us o analyse he deg ee
o conse a ion o se e al ligand-binding si es in he
ac in o cilia es. Fu he mo e, we ha e analysed in mo e
de ail he s uc u e o H. ca icola ac in by in silico
modelling.
Fi s , we aligned he ac in amino acid sequences o
se e al cilia es wi h ha o abbi a-ac in and calcula ed
he pe cen age o iden i y by subdomain (Table 1). The
highes deg ee o conse a ion is ound in subdomain 1,
especially in he egion o amino acids 70±144, and in he
egion o amino acids 145±180 o subdomain 3. These
zones cons i u e he inne hyd ophobic co e o ac in and
hey o m he hyd ophobic pocke in ol ed in nucleo-
ide- and ca ion-binding. Acco dingly, he nucleo ide-
binding si es a e well-conse ed mo i es in cilia e ac ins
(Table 2). Subdomains 2 and 4, and he egion o amino
acids 270±337 o subdomain 3 a e less conse ed
(Table 2), wi h he g ea es di e gence mainly loca ed on
he su ace o he molecule (Fig. 1); he e o e, he
DNase loop (amino acids 40±50) o subdomain 2 and
he pe iphe al egions o amino acids 194±203 and 223±
242 inside o subdomain 4 a e no well-conse ed
(Table 2, Fig. 1). The DNase-I-binding si es o cilia e
ac ins sha e 35.3±70.6% iden i y wi h he co esponding
sequence o abbi a-ac in (Table 2). Since T. py i o mis
ac in, wi h 58.8% iden i y o he DNase-I-binding si es,
lacks DNase-I-binding ac i i y, i is likely ha hypo-
ich, oligo ich and pe i ich ac ins, wi h a DNase-
binding mo i less conse ed han in Te ahymena, also
ail o bind DNase I. The ac in-ac in in e ac ions in he
mic o®lamen include a egion ha is almos iden ical o
ha associa ed wi h DNase I±ac in con ac , he DNase I
loop (di e gen in cilia e ac ins) being o p ima y im-
po ance o he s abilisa ion o he ac in ®lamen .
Mo eo e , in abbi a-ac in, a loop (amino acids 262±
272) ha includes a ou - esidue hyd ophobic plug
inse ed in o a hyd ophobic pocke o med by wo
adjacen monome s on he opposing s and is also c u-
cial o s abilisa ion o he ac in helix. In a con en ional
ac in, such as ha o yeas , a mu a ion (L
266
D) in he
egion o amino acids 262±272 p oduces dis up ions in
hyd ophobic in e ac ions, which esul s in he inhibi ion
o ac in polyme isa ion a low empe a u e [4]. The 262±
272 loop is no well-conse ed in cilia e ac ins (Table 2).
In his egion, hypo ich and Hal e ia ac ins sha e only
27.3±45.5% iden i y wi h abbi a-ac in, and hyd o-
phobic esidues wi hin he plug a e eplaced by hyd o-
philic ones. These modi®ca ions in cilia e ac in
Table 1 Pe cen ages o iden i y be ween abbi muscle ac in (whole molecule and subdomains) and ac ins om die en cilia es
Whole
molecule
Subdomain 1 Subdomain
2
Subdomain 3 Subdomain 4
1a 1b 1c 3a 3b
Amino
acids
1±375
Amino
acids
1±32
Amino
acids
70±144
Amino
acids
338±375
Amino
acids
33±69
Amino
acids
145±180
Amino
acids
270±337
Amino
acids
181±269
Hypo ichs
His iculus ca icola 65.5 72.4 85.3 67.5 48.6 69.4 44.1 65.2
Oxy icha no a I 65.8 64.5 85.3 67.5 45.9 69.4 52.9 64.0
Oxy icha no a II 67.2 71.9 75.7 85.0 58.3 80.0 56.1 58.0
Oxy icha allax 66.2 73.3 69.3 67.5 45.9 69.4 50.0 65.2
Oxy icha i allax 66.8 73.3 85.3 67.5 48.6 69.4 50.0 66.3
Euplo es c assus 61.9 59.4 65.3 67.5 54.1 72.2 60.3 58.4
Pa amecium
e au elia
73.6 78.9 83.1 83.7 70.2 83.3 73.1 61.3
Te ahymena
py i o mis
74.9 53.1 82.0 89.4 75.6 86.4 69.5 66.3
Pe i ichs
Vo icella mic os oma 69.8 68.4 78.6 76.3 62.2 83.7 69.1 58.4
Opis honec a
ma iensis
64.2 61.4 77.3 73.3 62.2 67.0 58.8 55.4
Olygo ichs
Hal e ia sp. 66.2 68.7 81.3 65.7 48.6 81.3 48.5 67.4
170
sequences may indica e unusual polyme isa ion p ope -
ies o an inabili y o polyme ise.
Second, we aligned he ac in amino acid sequence o
H. ca icola wi h ha o abbi a-muscle ac in and hen
ca ied ou in silico modelling, aking in o accoun he
e ia y s uc u e o he complex ac in±DNase I a ail-
able om he P o ein Da a Bank (accession numbe
1ATN). These in silico s uc u es ha e also se ed o
calcula e a molecula dynamics model and can be used
o sea ch o con o ma ional die ences among hem.
The in silico s uc u es o H. ca icola and abbi
a-muscle ac in a e simila , al hough small die ences can
be seen in subdomains 2 and 4. These die ences can
also be seen by compa ing he a e age s uc u es c ea ed
wi h molecula dynamic simula ions (Fig. 2), as well as
by compa ing he pe cen age o iden i y o hese sub-
domains a he amino acid le el (see he p eceden pa-
ag aph). Mo eo e , du ing he simula ion, subdomain
2 mo ed owa d subdomain 4, he displacemen being
mo e p onounced in H. ca icola ac in han in abbi
a-muscle ac in. The igh e in e ac ion be ween subdo-
mains 2 and 4 in H. ca icola ac in caused an allos e ic
hind ance ha could inhibi DNase I binding.
The ac in-binding p o eins
As can be deduced om he p eceding sec ions, knowl-
edge on cilia e ac in is s ill e y sca ce and ABPs a e no
an excep ion bu a he an ex eme case. In ac , da a a e
Fig. 1 The h ee-dimensional
s uc u e o ac in±DNaseI
complex acco ding o Kabsch
e al. [30]. The ® s and las
amino acids esidues in helices
and shee s ands a e speci®ed.
Regions o subdomains 2 and 4
ha a e highly di e gen in
cilia e ac ins a e shown in blue
(amino acids 40±50, 194±203,
223±242 and 262±272)
Table 2 Pe cen ages o iden i ies be ween nucleo ide-binding si es, DNase-I-binding si es, DNase loop (amino acids 40±50), and amino
acid 194±203, 223±242 and 262±272 loops o abbi muscle ac in and he co esponding egions in cilia e ac in sequences
Nucleo ide binding DNase I binding DNase I loop 194±203 loop 223±242 loop 262±272 loop
His iculus ca icola 73.3 35.3 9.1 60 26.6 36.4
Oxy icha no a I 86.7 35.3 9.1 40 26.6 36.4
Oxy icha no a II 73.3 52.9 45.5 40 26.6 27.3
Oxy icha allax 86.7 35.3 9.1 40 26.6 36.4
Oxy icha i allax 93.3 35.3 9.1 40 26.6 36.4
Euplo es c assus 86.7 58.9 45.5 30 26.6 45.5
Pa amecium e au elia 86.7 70.6 72.7 40 20.0 63.6
Te ahymena py i o mis 80.0 58.9 63.6 30 26.6 54.4
Vo icella mic os oma 86.7 41.2 45.5 30 20.0 45.5
Opis honec a ma iensis 86.7 47.0 45.5 30 13.3 54.5
Hal e ia sp. 86.7 35.3 9.1 60 20.0 27.3
171

es ic ed o Te ahymena. I is gene ally belie ed ha all
euka yo ic cells con ain ac in and i s ®lamen -binding
mo o p o ein myosin. This is ce ainly ue, bu in cil-
ia es he demons a ion o myosin has been dicul ,
maybe because o he high di e gence o cilia e myosin.
A epo on he p esence o Te ahymena myosin ap-
pea ed in 1995 [21], i.e. mo e han a decade a e he ® s
epo s on Te ahymena ac in. The au ho s desc ibed he
p esence o wo polypep ides o 180 and 15 kDa in he
basal-body cage complex o Te ahymena ha could
co espond, based on biochemical e idence, o he hea y
and ligh chains o myosin. Fu he mo e, au ho s in he
same labo a o y desc ibed he cloning o a myosin hea y
chain gene [20] which is indeed exp essed in g owing
cells. The p edic ed amino acid sequence o his gene
shows ha all he signa u e mo i es o he head domain
o known myosins a e conse ed. None heless, a phy-
logene ic analysis shows ha Te ahymena myosin hea y
chain belongs o a new myosin amily. Mo eo e , dis-
up ion o his gene aec s endocy osis and mac o-
nuclea elonga ion [53]. P o®lin, an ac in-seques e ing
p o ein ha modula es ac in polyme isa ion, has also
been desc ibed in Te ahymena. A con en ional p o®lin
o 12.8 kDa was isola ed in T. py i o mis by poly
(L-p oline) ani y column [16]. The same p ocedu e was
used o isola e p o®lin in T. he mophila, and he pu i®ed
p o ein was used o ob ain i s co esponding cDNA [52].
This cDNA sequence p edic s a polypep ide o 16.7 kDa
wi h li le homology wi h he p e iously epo ed
T. py i o mis p o®lin gene [15]. This las Te ahymena
p o®lin gene is also di e gen when compa ed o mam-
malian p o®lins bu i s N- and C- e minal egions a e
ela i ely conse ed. By immuno¯uo escence, p o®lin
was de ec ed in he di ision u ow o Te ahymena [17].
P o®lin seemed no o be he only ABP ha co-localised
wi h ac in in he di ision u ow, since h ee o he Te -
ahymena co-localising p o eins ha e been desc ibed:
®mb im, elonga ion ac o 1a(EF-1a) and calmodulin.
Te ahymena ®mb in was isola ed as a 61-kDa poly-
pep ide ha was shown o be he pa ial deg ada ion
p oduc o a 71-kDa polypep ide. A cDNA co e-
sponding o his ®mb in was sequenced and p edic ed o
be a p o ein o abou 65.1 kDa wi h wo ac in-binding
domains, bu lacking he EF-hand-binding domain. This
las ea u e and he low homology wi h o he known
®mb ins sugges s ha his p o ein is a new membe o
he ®mb in/plas in amily. Te ahymena ®mb in co-loc-
alises wi h ac in in di iding cells, bu in in e phase cells
also a he o al appa a us and acuole po es [51]. EF-1a
is in ol ed in p o ein syn hesis in euka yo es, al hough
some epo s show ha i can pe o m o he unc ions,
usually ela ed o cy oskele on egula ion. In ac , Te -
ahymena EF-1aco-p ecipi a es wi h F-ac in and has
F-ac in bundling ac i i y, as shown by elec on mic os-
copy [33]. Fu he mo e, i has been shown ha Ca
+2
/
calmodulin di ec ly in e ac s wi h EF-1ainhibi ing i s F-
ac in bundling ac i i y [32]. Bo h EF-1aand calmodulin
ha e been de ec ed in he o al appa a us and apical e-
gion o con ac ile acuoles in in e phase Te ahymena
cells [40], as has also been obse ed wi h ®mb in.
Summa ising, he esul s desc ibed he e show ha he
cilia e ABPs iden i®ed hus a a e somehow die en
om hose o o he euka yo es. This is simila o he
case o cilia e ac in when compa ed o con en ional
ac ins, hus ea ming he uncon en ional cha ac e o
he cilia e ac in cy oskele on.
Fu u e di ec ions
I is clea ha he unc ion o cilia e ac in is s ill un-
known. We can assume ha cilia e ac in plays he same
o a simila ole ha con en ional ac in does. This ex-
apola ion, howe e , could be inexac , since he p i-
ma y amino acid sequence and biochemical p ope ies o
cilia e ac ins die om hose o con en ional ac ins.
Indica ions on cilia e ac in unc ion come om he
localisa ion o his p o ein in he cell. These da a sugges
ha ac in may be in ol ed in some s eps o phagocy-
osis, he con ac ili y o he co ex and he di ision
u ow. Ne e heless, implica ion o cilia e ac in in hese
o o he p ocesses needs mo e expe imen al e idence,
which can be ob ained using cu en molecula ap-
p oaches, such as ans o ma ion. This echnique has
been success ully applied o Te ahymena,Pa amecium,
S ylonychia and Euplo es. T ans o ma ion would allow,
o example, obse a ion o ac in dynamics in i o
h ough usion wi h g een ¯uo escen p o ein. I would
Fig. 2 Compu e -simula ed
h ee-dimensional s uc u e o
abbi a-ac in (A) and His icu-
lus ca icola ac in (B). These
ep esen a ions co espond o
he a e age s uc u es o mo-
lecula dynamic simula ions.
Whi e ba s Dis ance be ween
subdomains 2 and 4, which is
sho e in H. ca icola ac in han
in abbi a-ac in
172
also be possible o use ``gene silencing'' o educe
ac in exp ession, i his we e no le hal o cells. This
pheno ype can be ob ained whene e he ole o ac in
becomes less essen ial o o ganelle mo emen s, cell
shape o ch omosome ea angemen s in cilia es han in
o he euka yo es. P elimina y expe imen s de eloped in
collabo a ion wi h A. Fleu y (Uni e si e
ÂPa is-XI, O -
say, F ance) ha e allowed us o ob ain a non-le hal
pheno ype in ac in- ans o med Pa amecium. Howe e ,
as ac in exp ession has no been analysed ye , we do no
know whe he he gene has ac ually been silenced.
The ole o cilia e ac in as bo h egula o o cellula
signalling pa hways and co-o dina o o cellula beha-
iou should be exploi ed. A en ion o ABPs, kinases,
phospha ases, calmodulin, o small GTPases and o
mo phogene ic p ocesses, such as conjuga ion, encys -
men /excys men , egene a ion and polymo phic ans-
o ma ion, will no doub lead o e y p omising lines o
esea ch.
Acknowledgemen s This esea ch was suppo ed by DGICYT
g an PB97±0710-C02±01. We hank M. A. de la Rosa and A.
DõÂaz-Quin ana o hei help wi h modelling analysis. We also
hank B. Pe
 ez Uz o c i ically eading he manusc ip .
Re e ences
1. Ba oin Tou ancheau A, Villalobo E, Tsao N, To es A,
Pea lman E (1998) P o ein coding ees in cilia es: compa ison
wi h RNA-based phylogenies. Mol Phylogene E ol 10:299±
309
2. Bha acha ya D, S ickel S, Sogin M (1991) Molecula phylo-
gene ic analysis o ac in genic egion om A. bisexualis and
C. cos a a. J Mol E ol 33:525±536
3. Bo k P, Sande C, Valencia A (1992) An ATPase domain
common o p oka yo ic cell cycle p o ein, suga kinases, ac in,
and hsp70 hea shock p o eins. P oc Na l Acad Sci USA
89:7290±7294
4. Chen X, Cook RK, Rubens ein PA (1993) Yeas ac in wi h a
mu a ion in he ``hyd ophobic plug'' be ween subdomains 3
and 4 (L
266
D) displays a cold-sensi i e polyme iza ion de ec . J
Cell Biol 123:1185±1195
5. Chik JK, Lindbe g U, Schu CE (1996) The s uc u e o an open
s a e o b-ac in a 2.65A esolu ion. J Mol Biol 263:607±623
6. Cla k SW, Meye DI (1992) Cen ac in is an ac in homologue
associa ed wi h he cen osome. Na u e 359:246±250
7. Cle eland DW, Lopa a MA, MacDonald RJ, Ru e WJ,
Ki schne MW (1980) Numbe and e olu iona y conse a ion
o a- and b- ubulin and b- and c-ac in genes using speci®c
cloned cDNA p obes. Cell 20:95±105
8. Cohen J, Beisson J (1988) The cy oskele on. In Go
È z H-D (ed)
Pa amecium. Sp inge , Be lin, Heidelbe g New Yo k, pp 363±
392
9. Cohen J, Ga eau de Loub esse N, Beisson J (1984) Ac in
mic o®lamen s in Pa amecium: localiza ion and ole in in a-
cellula mo emen s. Cell Mo il Cy oskele on 4:443±468
10. Cupples CG, Pea lman RE (1986) Isola ion and cha ac e iza-
ion o he ac in gene om Te ahymena he mophila. P oc Na l
Acad Sci USA 83:5160±5164
11. Diaz-Ramos C, Villalobo E, Pe ez-Rome o P, To es A (1998)
Pa amecium e au elia encodes uncon en ional ac in con ain-
ing sho in ons. J Euka yo Mic obiol 45:507±511
12. Dizick SJ, P esco DM (1999) Th ee mac onuclea molecules
encoding h ee highly di e gen ac in genes in Oxy icha no a.
Eu J P o is ol 35:375±377
13. D ouin G, Moniz de Sa
ÂM, Zuke M (1995) The Gia dia
lamblia ac in gene and he phylogeny o euka yo es. J Mol E ol
41:841±849
14. DuBois M, P esco DM (1995) Sc ambling o he ac in I gene
in wo Oxy icha species. P oc Na l Acad Sci USA 92:3888±
3892
15. Edama su M, Hi ono M, Takemasa T, Wa anabe Y (1991) The
p ima y s uc u e o Te ahymena p o®lin. Biochem Biophys
Res Commun 175:543±550
16. Edama su M, Hi ono M, Wa anabe Y (1990) Pu i®ca ion and
cha ac e iza ion o Te ahymena p o®lin. Biochem Biophys Res
Commun 170:957±962
17. Edama su M, Hi ono M, Wa anabe Y (1992) Te ahymena
p o®lin is localized in he di ision u ow. J Biochem 112:637±
642
18. Fah ni JF (1992) Ac in in he cilia ed p o ozoan Climacos o-
mum i ens: pu i®ca ion by DNase I ani y ch oma og aphy,
elec opho e ic cha ac e iza ion, and immunological analysis.
Cell Mo il Cy oskele on 22:62±71
19. Gallwi z D, Su es I (1990) S uc u e o a spli yeas gene:
comple e nucleo ide sequence o he ac in gene in Saccha -
omyces ce e isiae. P oc Na l Acad Sci USA 77:2546±2550
20. Ga ces J, Ga in RH (1998) A PCR sc een iden i®es a no el,
uncon en ional myosin hea y chain gene (MYO1)inTe a-
hymena he mophila. J Euka yo Mic obiol 45:252±259
21. Ga ces JA, Hoey JG, Ga in RH (1995) Pu a i e myosin
hea y and ligh chains in Te ahymena: co-localiza ion o he
basal body-cage complex and associa ion o he hea y chain
wi h skele al muscle ac in ®lamen s in i o. J Cell Sci
108:869±881
22. Ghosh S, Ja aczewski JW, Klobu che LA, Jahn CL (1994)
Cha ac e iza ion o ansc ip ion ini ia ion, ansla ion ini ia-
ion, and poly (A) addi ion si es in he gene-sized mac onulcea
DNA molecules o Euplo es. Nucleic Acid Res 22:214±221
23. Hackney CM, Bu le RD (1981) Ten acle con ac ion in gly-
ce ina ed Discoph ya collini and he localiza ion o HMM-
binding ®lamen s. J Cell Sci 47:65±75
24. Ha pe DS, Jahn CL (1989) Ac in, ubulin and H4 his one
genes in h ee species o hypo ichous cilia ed p o ozoa. Gene
75:93±107
25. Hause M, Hausman K, Jockusch BM (1980) Demons a ion
o ubulin, ac in and alpha-ac inin by immuno¯uo escence in
he mic o ubule-mic o®lamen complex o Pseudomic o ho ax
dubius. Exp Cell Res 125:265±274
26. Hi ono M, Endoh H, Okada N, Numa a O, Wa anabe Y
(1987) Cloning and sequencing o he Te ahymena ac in gene
and iden i®ca ion o i s gene p oduc . J Mol Biol 194:181±192
27. Hi ono M, Kumagai Y, Numa a O, Wa anabe Y (1989) Pu i-
®ca ion o Te ahymena ac in e eals some unusual p ope ies.
P oc Na l Acad Sci USA 86:75±79
28. Hoey JG, Ga in RH (1992) Localiza ion o ac in in he Te -
ahymena basal body-cage complex. J Cell Sci 103:629±641
29. Kabsch W, Mannhe z HG, Suck D, Pai EF, Holmes KC (1990)
A omic s uc u e o he ac in:DNase I complex. Na u e 347:37±
44
30. Kaine BP, Spea BB (1980) Pu a i e ac in genes in he mac-
onucleus o Oxy icha allax. P oc Na l Acad Sci USA
77:5336±5340
31. Ka suma u H, Fukui Y (1982) In i o iden i®ca ion o Te a-
hymena ac in p obed by DMSO induc ion o nuclea bundles.
Exp Cell Res 137:353±363
32. Ku asawa Y, Hanyu K, Wa anabe Y, Numa a O (1996) F-
ac in bundling ac i i y o Te ahymena elonga ion ac o 1ais
egula ed by Ca
+2
/calmodulin. J Biochem 119:791±798
33. Ku asawa Y, Wa anabe Y, Numa a O (1996) Cha ac e iza ion
o F-ac in bundling ac i i y o Te ahymena elonga ion ac o
1ain es iga ed wi h abbi skele al muscle ac in. Zool Sci
13:371±375
34. Lin JJC (1981) Monoclonal an ibodies agains myo®b illa
componen s o a skele al muscle deco a e he in e media e
®lamen s o cul u ed cells. P oc Na l Acad Sci USA 78:2335±
2339
173
35. Maci e SK (1998) How ADF/co®lin depolyme izes ac in ®l-
amen s. Cu Opin Cell Biol 10:140±144
36. Mackay DJ, Hall A (1998) Rho GTPases. J Biol Chem
273:20685±20688
37. McGough A (1998) F-ac in-binding p o eins. Cu Opin S uc
Biol 8:166±176
38. Me enie G (1984) Ac in in Te ahymena pa a o ax: ul a-
s uc u al localiza ion o HMM-binding ®lamen s in glyce i-
na ed cells. J P o ozool 31:205±215
39. Mi chell EJ, Zimme man AM (1985) Biochemical e idence o
he p esence o an ac in p o ein in Te ahymena py i o mis.
J Cell Sci 73:279±297
40. Numa a O, Ku asawa Y, Gonda K, Wa anabe Y (2000) Te -
ahymena elonga ion ac o -1ais localized wi h calmodulin in
he di ision u ow. J Biochem 127:51±56
41. Numa a O, Wa anabe Y (1982) In i o assembly and disas-
sembly o 14-nm ®lamen om Te ahymena py i o mis.The
p o ein componen o 14-nm ®lamen is a 49,000-dal on p o-
ein. J Biochem 91:1563±1573
42. Pe ez-Rome o P, Villalobo E, Diaz-Ramos C, Cal o P, To es
A (1999) Ac in o His iculus ca icola: Cha ac e is ic o he
highly di e gen hypo ich cilia e ac ins. J Euka yo Mic obiol
46:469±472
43. Pipe no G, Luck D (1979) An ac in-like p o ein is a componen
o axonemes om Chlamydomonas ¯agella. J Biol Chem
254:2187±2190
44. Polla d TD, Blanchoin L, Mullis RD (2000) Molecula mech-
anisms con olling ac in ®lamen dynamics in nonmuscle cells.
Annu Re Biophys Biomol S uc 29:545±76
45. Reisle E (1993) Ac in molecula s uc u e and unc ion. Cu
Opin Cell Biol 5:41±47
46. Sa
Ânchez-Sil a R, Fe na
Ândez-Aliseda MC, Pe ez-Rome o P,
Cal o P, To es A (2000) Ac in o pe i ich cilia es. J Euka yo
Mic obiol (Abs ac s om Sec ion Mee ings 2000)
47. Sandoz D, Gounon P, Ka sen i E, Sau on E (1982) Immuno-
cy ochemical localiza ion o ubulin, ac in and myosin in axo-
nemes o cilia ed cells om quail o iduc . P oc Na l Acad Sci
USA 79:3198±3202
48. Schu CE, Myslik JC, Rozycki MD, Gooneseke e NCW,
Lindbe g U (1993) The s uc u e o c ys alline p o®lin-b-ac in.
Na u e 365:810±816
49. Tiggemann R, Pla ne H (1981) Localiza ion o ac in in he
co ex o Pa amecium e au elia by immuno and ani y ¯u-
o escence mic oscopy. Eu J Cell Biol 24:184±190
50. Wahle E, Kelle W (1992) The biochemis y o 3¢-end clea age
and polyadenyla ion o messenge RNA p ecu so s. Annu Re
Biochem 61:419±440
51. Wa anabe A, Yonemu a I, Gonda K, Numa a O (2000)
Cloning and sequencing o he gene o a Te ahymena ®m-
b im-like p o ein. J Biochem 127:85±94
52. Wilkes DE, O o J (2000) Molecula cloning o p o®ling om
Te ahymena he mophila. Gene 246:295±301
53. Williams SA, Hosein RE, Ga ces JA, Ga in RH (2000) MYO1,
a no el, uncon en ional myosin gene aec s endocy osis and
mac onuclea elonga ion in Te ahymena he mophila.J
Euka yo Mic obiol 47:561±568
54. Zimme man AM, Zimme man S, Thomas J, Ginzbu g I (1983)
Con ol o ubulin and ac in gene exp ession in Te ahymena
py i o mis du ing cell cycle. FEBS Le 164:318±321
174