675
When is homology not homology?
Gregory A Wray* and Ehab Abouheif
Although genes have specific phenotypic consequences in a
given species, this functional relationship can clearly change
during the course of evolution. Many cases of evolutionary
dissociations between homologous genes and homologous
morphological features are now known. These dissociations
have interesting and important implications for understanding
the genetic basis for evolutionary change in morphology.
Addresses
Department of Ecology & Evolution, State University of New York,
Stony Brook, New York 11794-5245, USA
*e-mail: [email protected]
ie-mail: [email protected]
Current Opinion in Genetics & Development 1998, 8:675-680
http://biomednet.com/elecref/O959437XO0800675
,o Current Biology Ltd ISSN 0959-437X
It is w o r t h pat,sin K bricflv to cnnsider what is meant by the
term qmmologv' bcfore procccding to a discussion of these
issues. Although most biologists seem to have a good intuitive feel for the concept of homolog T, the literature on
homology is fimlouslv fldl of philosophical and methodnlogical dcbates. "Fhc clearest, most practical, and most widely
accepted definition of the term homology is simply the prese l ] c e o f a f e a t , i r e in the n]()st r e c e n t c o r n n ' l o n ancestor of t ~ o
species [8,9]. This definition has the added xirtucs of being
applicable to a n \ feature of biological organization (molecular, behavioral, dcvelopmcntal, etc.) and of forcing one to be
explicit about the phylngenetic history of the features of
in,crest 17",10]. As ancestors are rarely available for direct
examination, homology is usually a hypothesis about evolutionary history rather than a direct observation [6,7°°,9,10].
Wc will usc the term 'homology' in this formal sense, to
mean a hypothesis that a particular similarity in two extant
spccies prcdates thcir evolutionary divergence.
Introduction
()ne of the fascinating realizations to e m c r g e during the
past two decades of d e v e l o p m e n t a l genetics is the extraordinary c o m p l e x i t y o f the rclationship b e t w e e n gennt3 pc
and p h c n o t v p e . T h i s cnmplcxity has important, and quite
intcrcsting, implications for u n d e r s t a n d i n g hnmologv,
which is the central c o n c e p t of comparative binlog T.
(;avin de Bccr was among the first to recognize the evolutionary i m p l i c a t i o n s of the c o m p l e x i t y of the
g c n o t y p c - p h e n o t y p c relationship: in an insightful essay
p u b l i s h e d in 1971 [1], he noted that homologous genes do
not necessarily e n c o d e hon]ologous structures and that
homnlogous structures nccd not be c n c o d c d by homolngnus gcnes, T h c s e werc rcmarkabl.v prescicnt inferences,
as they were m a d e at a time when relatively fcw p e r t i n e n t
data wcre available, and before nmlecular tcchniques
transformed d e v c l o p m e n t a l biologx.
In thc q u a r t c r - c c n t u r y since de Becr p u b l i s h e d his cssa%
many additional e x a m p l e s havc corroborated his conch,sions. It is now clear that scveral distinct kinds of
dissociations can evolxc b e t w e e n h o m o l o g o u s genes and
honlolognus aspects of mnrphoh)gy [2,3",4",5] ( F i g u r e 1).
In such cases~ homolngx at one I t \ e l of biological organJzatinn does not reflect homoh)gy at a n o t h e r [6,7"].
l : u r t h e r m o r e , cvolutionarv dissociations of this kind may
bc more c o m m o n than is generall T a p p r e c i a t e d , particularly w h c n c o m p a r i s o n s arc m a d c across d e e p
phyh,gcnctic
divides.
Ew)lutinnary
dissociations
b e t w e e n g c n n t y p e and p h e n n t y p c limit to somc c x t e n t
the usefulness nf g e n c e x p r e s s i o n d o m a i n s for m a k i n g
inferences a b o u t the evolutionarx h i s , o r \ of morphoh)gical structures [4",5,6]. ()n thc o t h e r hand, thcsc samc
dissociations provide an c x t r a o r d i n a r i h valuable windm~
into u n d e r s t a n d i n g the g c n c t i c basis for morphnh)gical
evnh,tion [2,3",4",5].
Homologous genes, non-homologous
morphology
T h e more that is learncd about regulatory genes, the clearer it bccnmes that t'ex~q if any, are dedicated to a single
dcvclopmental task [3",4°,5]. For instance, thc Notch signalling system is utilized on many separate occasions
during the d c v e l o p m e n t of Drosopld/a mdam~gz/ster. T h e s e
include the production of structures that are clearly not
homologous, such as wings, n m m a t i d i a , and bristles [3",11,12]. In nthcr animals, homologous e l e m e n t s of
this signalling p a t l m a y are also used repeatcdly during
dcvclnpment, again in structures that are not homologous,
such as feathers and T-lymphocytcs ('lhble 1) [3",13,14].
Given thc diversity of uses to which this signalling system
has been put during the course nfanimal evolution, it is difficult (at least from existing data) to guess what its ancestral
talc may have been. (Note that "lhble 1 is a only partial list
of the known devclopmcntal rolcs of the Notch signalling
system.) T h e same conclusions emergc from a consideration of other intercellular signalling systems, such as those
mediated by hedgehog, T G F - ~ , and W n t family m e m b c r s
[3",5]. T h e r e arc ahnost certainly many more intercellular
signalling events than thcrc are intercellular signalling svs[Gills ill 111()St l ] l e t a z ( ) a n s , i n l p i y i l l g l l t l n l e r ( ) / l S c a s e s where a
homologous gcnc has become involved in the d c v e l o p m c n t
of a nnn-homolo]zous structure (lqgurc I a).
This situation is not unique to signalling proteins. For
example, ~uzgrai/ed--which encodes a tmmeodomain transcription f a c t n r - - r e g u l a t e s e m b r y o n i c patterning, gut
diffcrentiatinn, and ncurogcnesis (among other things) in
Drosop/fi/a [15,16]; in Jlus mu.uu/us it is involved in patterning thc brain and somite diffcrentiation (among nthcr
things) [17]: and in the echinodcma Amphipho/is.waamata, its
expression is assnciatcd with skeletogcncsis and neuronal
676
Genomes and evolution
Figure 1
(a)
dissociations
homologous
genes and homologous
structures.
R’
G’
Evolutionary
The evolutionary
developmental
developmental
S’
between
histones
of
regulatory genes, their
roles, and the structures
to
which they give rise are not always congruent.
Most developmental
R’+
G’
regulatory
genes of
metazoans are clearly more ancient than some
of their current developmental
roles. For
S’
instance,
homeodomaln
transcription
factors
predate the origin of the metazoans [47] but
are involved in patterning many structural
features
unique to particular
[I 4,15,18*,22**,45,46*].
groups
roles
(a) and lost (b) on many
have been gained
occasions
metazoan
Developmental
(for examples,
see text and
Figure 2). The acquisition of new
developmental
roles may be Important
origin of evolutionary
(b)
noveltles,
In the
but can
confound the use of gene expresslon to
identify homologous
structures. G, gene;
S’
R, developmental
role; S, structure.
G’
s2
S’
G’
s2
’n
G
i
Current
R
Gene
Op~mon in Geneks
encoded by /I~ud/kk
and
role
Structure
Loss of role
& Development
[ 1X’]. Similarly,
diffcrcntiation
niarion
Developmental
the
is in\d~ui
dcxlopnwit
of
transcription
in embryonic
central
the
/)tv.sop~j/u. \vhcrcas its cxprcssion
ncr~wis
in the
factor
pattern
for-
system
Iccch
in
Hay genes
insects
howe\w,
only
set of roles [l’,].
Hoth transcrip-
For
tion
dcvelopnental
processes
;Ippendages
factors
participate
in
that arc ccrkrinly
again. iI is no1 immcdiatciy
thcsc pies
cniergc
tars
ma)-
iia\r
clear 11hat the ancestral
hxn.
‘I’hc
l\s midi
which
dctailcd
signalling
sttlriies
over and o\-cr again, not just
single orginisni
Ililt
ha~c
twwription
been
transcriprion
sysmm,
during
roics of
carricd
hctors
the dc~~ciopnient
thro~~ghout e\miution
(diereforc
of a
each
rabic
It is worth
cniphasizing
roles arc often
role
\vith
observation
conser~al
conscr\.ution
\\,idcly [3’.ZO].
that none of the esamplcs
the
‘l’hc
are
thar
in c\x)lurion.
kno\vn
niosI
and
f~iiiioiis
have
example
ahow
is
examples
ken
to
that
it
bctnccn
rhonibomcrcs.
‘I’hc
is
so
and
arc uniclue
In addi-
but
art: niorc
much amzntion
roles, nowx~nser~ui
u) \.crtcbr;ltcs
or
no\ cl.
role is n1cnlo-
roles
that
nun~crolls
has
rcpi-odiw
lio~~iologo~~s
recruited.
one conser\wi
both dc\.clol7mentaii~
not.
patterning
the
that arc not
striking
phyla
to hu\,c
the classic ncsrcd domains
representing
are. Although
conscrvcd
nificant,
de\ elopnicntal
hlany
kcausc
conser\.cd
in
of
3s this.
arc clcirly
but not in arthropods.
solne of u,hich
roles).
appc;irs
roics
in\x)l\.cd
31. structures
ncccssarily
dc~~elopnicn~;il
OIICS
inconip~itii~lc
and
role
arc
;IXS
in cases sllch
orher
they cshibit-
in soinites.
other
to
[4*..5].
in \xrtehrates
tracx [21.22”.2.
tk-
l<\en
se\wal
Ho.1genes
of esprcssion
0111.
arc used
whiic
Con, in \~ertelxitcs
ti\x
general conclusions
s31iie
from :i conip~~rison of many other
for
that
not hon~oio,go~~s and.
insrance,
the 3nteropostcrior
[Zl].
dcvclopmental
one
conser\wI
Hdohddl~~ Lxen
t/i.wrk/i.~
impiics a different
prodi~cc structures
in patterning
11la)
and \wtebratcs
been
are
not
than
rhc
dc\~otcd
roles arc clearly also si,qand
c\.olution;lril!.
of
discussed
is rhc role that
As de Beer \fmw
troilcd
by idcnticd
nearly 30 years
:igo
11 1,
genes arc not ncccssariiy
“characters
con-
f~omoio~ous“.
When
is homology
not homology?
677
Wray and Abouhelf
Table 1
Other casts ha\x been documenred mithin the insects. LS~,x--
A partial list of the many known roles of the Notch signalling
pathway.
l&/l
(Sk/l is ;I ‘imstcr regulator!; gene thar controls sex
determination in /hv.s(//)hi/~~ ///~INNI?,“/I.C~(throiigh 3 wcllcharxwrized
path\vq of altcmxive
splicing [Zc)]. ‘I’his
Phylum
Developmental
Arthropoda
Nematoda
Chordata
(a)From [3’,1 l-l
‘I‘he
Patterning
wing lmaglnal
Specifying
Specifying
distinct
dlstlnct
disks.
Specifying
Speclfylng
dlstlnct AB lineage
vulva1 cell fates.
bristle organ cell fates.
ommatldial cell fates.
Patterning
feather
Specifying
distinct
primordla.
T-lymphocyte
Specifying
various
neuronal
between
a homologous
of I>hcno~~pt2
can
gene
and
be conser\xxi
but
3 hotnoloften
1’1cn Icaving aside e\wlutionar):
coni~xirisons.
repfxted
IISC of ;I gene during
the dexlopincnt
of
not.
the
to build
rule
only
rather
non-homologolIs
than
the
Ivmvecn
of
arc
often
complex
the
jicnotye
strucmres
esccption
of conscr\3tion
part
basis of the phylogenctic disrribution of thcsc data. the sexdererniinarion role of .S.v/ in Ih.sf~phi/~~ is almost certainI>
the cleri\ cd cast (Figure 21,). ‘I’his is the rekersc simation
cell fates.
cell fates.
from the a’i‘(’
cwniple: here 2 ,qcnc has l~~~mie in\dved in
3 de\elolmcnral process after that process first c~wl\wl.
association
instmccs
almost cerrainl\- not invol\,ed in scs determination:
although rhc gent is present, it is not dternati\ ell; spliced
and is not expressed at the corrcc~ time [31’,33’]. On the
cell fates.
31
ORIOLES3slxxx
or~misin
pathuxy appears to bc present in at least t\vo other
Dmsop/lilcl spccics, based on altcrnatc splicing
of
transcripts [?A)]. In se~wal other dilxcrans - including
C0utifi.s utpiiclto and . J/N.su~ hnu.siiru - howtxw, cCd is
processe&
it is
In thcsc exan~ples, a homologous gent and a honiologous
thr
phcnotypic
a singlr
is probal)l~
[.3’,4’].
Although
impressive,
e\~olutionary
they
are
rcl3tionship
and phenotyx.
Non-homologous
genes, homologous
morphology
.A gro\\ing niimbcr of cases demonstrate that the in\u-sc
situation. \bhere gcncs that are not honiologous encode a
honiologot~s morl~hological feature, can also occur. One of
the first cases to Ix recognized involws
e\mlutionrirv
changes in the dc\,clopmental roles of PLYL-skip/we/(KC),
which cncodcs a homcodomain
transcription
factor.
‘I’hc eponyniowi role ofe~,
identified
sion
occurs
ccgnicnt
is pattern
in a ‘pair-rule’
and is Irequired
tional
in I~mrf~~/li/~~, where
and characterized.
[Z-I].
patrern
for die corrwt
I,ikc
other
~l~\~elol’nicntal
during
roles
in
cxpres-
cnibryojicncsis
de~~elopnicnt
regulator)-
it was first
formarion:
of e\xry
gcncs.
ncriro~enesis [2.5] 2nd less b\ cll-characterized
pad
and dorsal niesoderni [Xl.
including
roles
in the
anal
Surprisingly
iii the
the
me has lost its pair-rule patterning role \vith-
insects:
wasp
pattern
in the
locust
.Ip/riclic~c P/Y,; thet-e
of esprcssion
within
AS~~is~ocwr~/ ~~M~~I+U/IIN and
is no scgnentall~
the
cctodcrm,
in
rcitcrareci
although
the
role is present (IJigure Za) [27,2X’]. In
both C;ISCS.honiologous snuxurcs
(scgnients) arc prcxnt
but at IcmI one homologous gene no longcr contributes to
their dc~~clopnient. Indeed,
in the case of the ~rxsp. the
latei-
neiirogcnic
relativclv
closely
relared
spwics
/Ium~~
feamrc
but
in
arc present
only
some
in all of the
of them
species
does the
nien-
gent
contribute I0 the development of the feature. Other cases
include apparent losses of scgnientation and honieotic
patterning
roles for,fi/.s/li-tcl/n~:l/ [3.3,34] and .wu [.%I during arthropod CLolution. In none of the c:iscs discussed
here is the genetic basis for de\~elopn~cnt lmo\z.n in
species other than I)msop/IiI~~ INP/NII/?,“N.s~(,‘I;
rctlecCng the
significant technical difficiiltics
intrinsic in making such
wxssnicnts.
It also points to an important, but rarely
acknowledged, hias in our knowledge of conilxtrative
de\,elol”iicnral genetics: most of the molecular nicthods
used to stud!- de\elopnicnt in non-model organisms rel)
on sequcncc similarity to w)rk. ‘I’his nicans that finding
differences is inherently much niore difficult than finding
siniilaritics,
and inhcrcntly
more difficult to interpret.
‘l’his rcchnical bias is winforced b)- 3n apparent bias in
inrercst, in that many molecular biologists seem more
cxcitcd by similarities
e\~olutionary biologisu
than t)!- differences (while man)
would
have the opposite bias).
other
crc’(’ has addi-
I~msofhi/c/.
tioncd.
/ru~~~,w~-has
Once again. \\e cun sunlnlarizc thcsc exaniplcs u.ith :I
cluorarion from de Ikcr [ 11, “honiologous structllres need
not Ix controlled by honiologo~~s ,qmes”. Kclativelv fenexamples of this phcnoinenon ha\x been dcscrilxd but
die inherent difficulty of detecting this kind of rwlutionarv dissociation, combined v irh some clear cxaniples of its
csistence, suggests thar it is not sufficicnrly rare rhat it can
Ix safely ignored. Although this kind of c\wlutionary dissociation between genotype kind phenotylx
ii-u> be
rclati\.cly iinconiimon when coniparing closely related
spccics, both die co and kS.d csanil~les dcnlonstratc that
the gcnctic basis for an iinporranr daclopmentd
process
can change c\xn among quite closely related taxa.
the
usiial pair-rule lxlttcrn of cupression bvithin the ectodcrm
[2X*]. ‘I’hc loss of a segnienration role for 62 in A4p/ric/i~~.r
is
prol)al~ly ;I rcsulr of its highI> modified early dc~~clopnient
2s an cndolxmsitc
rather than with any niodification in
adult morphology (28’1.
Implications for understanding
of morphology
the evolution
‘1%~ cxaniptcs discwssed above. along with many other
similar cases, xc beginning to pro\.ide 3 clcarcr iinderstanding of the coniplcx, and ofwn swprising. cwlritionar)-
678
Genomes
Figure
and evolution
2
Evolutionary
(a)
Megase//a
Sxl gene
scalaris
history of (a) Sex-lethal
(b) even-skipped
acqulsltion
\
Chrysomya
Unknown
Musca
domestfca
they encode.
a developmental
Ceratftis
I
sexes
/ -+role
Sex-determination
gene’ In the
Drosophila
virilis
Drosophia
species [29,301. Sxl IS present in
other flies, all of which have separate sexes and
Drosophila
subobscura
sexual dimorphism, but does not appear to be
involved in sex determlnatlon
[31’,32’1. As a
Drosophila
melanogaster
result of the relatively derived phylogenetic
position of the genus Drosophfla, a role In sex
eve gene
IS almost certainly a new function
for Sx/[31’,32’].
Neurogenlcrole
\
I
\
(b) Loss of a developmental
role. The gene eve IS involved in segmentation,
Ai-
CNS
Schistocerca
americana
Dermestes
neurogenesis, and other developmental
processes in Drosoph//a and some other
Insects [24-261. In the locust Sch/sfocerca
frischf
americana
Callosobrucus
Tribo//um
maculatus
certainly
melanogaster
represents
role [28’].
\
Segments
‘/I
I
?/>
Loss of segmentation
1
Bracon
hebetor
Aphidius
ervl
K\.olutionar):
the absence
can only be determined
expression
arthropods
hct\vcen
genotype
and
bet\f.cen
gcnoryI,e
in particular
rypc
~xo\ide
two
ini~xxtanc
phenot)pc.
roles within
and
absurd.
pheno-
single
by examining
for
mow
expression
‘I’hc
first
ttndcr
is 2 clcmx
which
understanding
gcnc
infcrcnces
expression
vbottt
rcswrch
program
Ixcn
:tlq>lied
\vith
ttnra\xling
of
oft&s
gre3t
rhe e~~olttrionary
comparisons
[.37,.3X’].
‘I’he
relati\,c
SC\ cd
transcription
ccnttxl
ncrwtts
ccl’h~tlocliorutcs
inferring
hotnologot~s
body
[ZO,X~].
for
example.
of rhe \wtel~rate
espression
of csprcssion
is concordant
ot’ \ ertebratcs,
Co
brain
among
taxa
domains
in the
for
and
among
morphological
regions
for
of the
ncll for identifying
in crttst;tcc;tns
[39’] and hontol-
h:ts also worked
regions
fins
:~nd limbs
[.5.6,7**.10.11,42].
c-qmxsion.
gtridc
for deuxniining
uould
inrerpret
‘Il~ble
I I0 nxxi
l~oniolo~o~~s
hunians.
plc
‘l’hc
cspres~ion
howe\w,
the
the honiolog~~
North
that
the
tlic
e)r
siniplc
tkr
to
is certainly
[ 72”.10].
domtins
md
that
gcncs
regttlatory
play
niuttiple
to
to
trsinx
coni~xlinfi
In general,
thi5 apptmtrh
to
\.iding
landmarks
rel:tri\xly
will
rcalisric
Ix
dissociations
and honiologot~s
morphological
u ill Ix
closcl~
broadly
\jill
lxtuxxn
lx
rcl:trccl
similar.
of
progcnc
Icss likelihood
hon~olo~otts
strttcCttrcs
of
I,> SC\ cT;tI
inrcrprerarions
data: in st~ch c~scs. there
of c\olutionary
gcnc
h) lwlhcscs
prcscntcd
In such cases, atxttoniy
cxprcssion
infcrcnccs
ha\ c txxn
:tpplicd
for
facitc
\\ irh rccon~nm-
;t1>1)roachc\
spwics.
‘I’hc second,
more
ic
evolutionary
basis
ot-igin
for
of
sotnites
infallible
tcs
No one
gcnc?
[S.-IL].
in
dqy~~.~ is
‘Ikells
of
h:t\.c nittlti-
de~elopmcn~:tl
lishcd
tmor~~holo,~~~. 'I‘llC
no\,cltics
-
sttch
has long
hcritablc
rypically
to the
~~lleles (or
in\~ol\w~
in sc\ctxI
single
rcsoltttion
spwic
of thi\
pt-oblcni
illl
prodttcx
th:tr
of ho\\
Im_x,111c csrab
:t fttncrionall\
rcgul:ttory
distinct
gcnc\
de\ cIopnICnt:tl
pro\.idcs
apptrcnl
mttst
of no\ cl
;t scI of ncm allc-
llic
fitct
chordatc
phcnot\Ixzs
could
thq
‘I‘hc
3s
c\.olttlion;tr>
the origin
rcqttirc
posts
p211cs)
heforc
phcnotqx.
within
p~~uletl
ncu
will
‘I’his
jienes.
the gcnct-
id
it is ~tss~~nied th;tt
in ;I population
processw
clue
As
lxtsis,
new
nc\v
ad\2ntagcotts
arc
-
mcss;~g?;c concern\
change
of any complexity
c\cn
sc\,cr;tl
\vings
[43.4-l].
a genetic
or
exciting,
morphological
or insccr
biologists
havr
data sttmmarized
;I of (.i/f,///,/./rc/l~liti~
of /ho.co/~/l~/~
md
an
of smtctttres.
expression
\III\
not
\r,hcn
such
along
among
honiolog~
reliable
strttcxttrcs
(ictic
rigorous
role; S, structure.
untcrior
urochordates,
‘landmat-ks’
of ~wwl~rate
authors
most
It has
inolccular
uppro:tch
ogott~ regions
to make
strttctttrcs.
1x0\ iding
homologies
‘I’his
gene
fwtors
-
sttcccss,
17ositions
systems
be used
~~roniisc
history
of
c:tn
makes
difficult):
to ciiscriniinatc
morp~iologic~tl
tit-cttnistxtccs
morphological
considerable
through
brain.
dat;t
honiolo~y
‘I‘hic
of the
of this
eve
in other, more distantly-related
(hence the questlon marks). G,
org;tnisms
IXscLlssion
dations
messages.
of pair-rule
expresslon in Schistocerca also represents a
loss, or is the ancestral condition for Insects,
role
dissociations
role IS
the loss of a developmental
Whether
gene; R, developmental
rclarionship
ervi, both of
the segmenatlon
absent [27,28-I. As a result of the derived
phylogenetlc posltlon of Aphldius, the absence
of pair-rule eve expression In this species almost
castaneum
Drosophila
role?
and the wasp Aphid&s
which have segments,
/
Segmentation
of
sex-determination
pathway of Drosophila
melanogaster and at least some other
capftata
determination
(b)
(a) Acquisition
role. The gene Sxl is the
‘master regulatory
/
role
Separate
and loss of developmental roles can
simultaneous gains and losses of
occur without
ruflfacies
the phenotypes
++
(SxO and
(eve) In Insects. The
an
p;tr;iclox.
imporf;tnI
(:hdng:c\
When is homology
in
inter:lctions
,qcts
-
btlt\vccn
than
r:tthcr
of gcncs
ma!.
-
morphology.
regulatory
changes
in the
underlit
many
includin,g
Regulatory
genes
genes
biochemical
novel
of
take
old ones during
the course
le;ds
hypothesis
the
to the
relics,
dccclopniental
casts
related
been
such
\vith
origin
by
gcncs
of new
directly.
111 c I1 t
of
the
cctopic
cnprcssion
practical.
r
disruptiry,
in non-niodcl
it should
hasi4 for thcsc
Ix
c\wlutionary
ogv to get incre:lsingly
fruitful
Hall BK (Ed): Homology: The Merarchlcal Basis of Comparative
Bioiogy. San Francisco: Academic Press; 1994.
10.
Abouhelf E: Developmental
genetics and homology:
approach. Trends Ecol Evol 1997, 12:405-408.
11.
Bang AG, Bailey AM, Posakony JW: Hairless promotes stable
commitment
to the sensory organ precursor cell fate by
negatively regulating the activity of the Notch signaling pathway.
Dev Biol 1995, 172:479-494.
12.
Kim J, Irvine KD, Carroll SB: Cell recognition,
signal induction, and
symmetrical
gene activation at the dorsal-ventral
boundary of the
developing
Drosophila wing. Cell 1995, 82:795-802.
13.
Robey E: Notch in vertebrates.
14.
Crowe R, Henrlque D, Ish-Horowlcz D, Nlswander L: A new role for
Notch and Delta in cell fate decisions: patterning the feather
array. Development 1998, 125:767-775.
15.
Pate1 NH, Martin-Blanco E, Coleman KG, Poole SJ. Ellis MC,
Kornberg TB, Goodman CS: Expression of engrailed proteins
arthropods,
annelids, and chordates. Cell 1989, 58:955-968.
and
into
changes
in
for
inducing
genetic
Regulator);
gcncs
and
fying
aspwts
gcncs
pro\idc
diversity
results.
morl’hological
thcsc
proniisc
horndog);
pholog>.
3 prows5
:Zs analysts
niorc
that
to light.
‘I’hese
of biolo+d
honiolog~
at another.
as ‘noise’
morl~hological
\\indow
paid
genes
casts
that
difficult
mea11
Such uses
honiologics.
should
with
‘I’hcy
also
to study.
to more
not
not bc \icwcd
the
ha\ e
homology
always
at
provide
sini-
and recommended
l
1.
2.
of special Interest
* of outstandlng interest
de Beer GR: Homology:
University Press: 1971.
an Unsoived
a hierarchical
Curr Opm Genet Dev 1997, 7:551-557.
in
1 7.
Davis CA, Holmyard DP, Mlllen KJ, Joyner AL: Examining pattern
formation in mouse, chicken and frog embryos with an ENspecific antiserum. Development 199 1, 111:287-298.
of
18. Lowe CJ, Wray GA: Radical alterations
in the roles of homeobox
.
genes during echinoderm
evolution. Nature 1997, 389:718-721.
This paper provides several examples of probable gains and losses of developmental roles for three transcription factors (distal-less, engrailed, and orthodentlcle). These factors play some developmental
roles that are distinct
from those In either arthropods and chordates, and play different developmental roles from class to class within echinoderms.
19.
Savage RM, Shankland M: Identification
and characterization
of a
hunchback
orthologue,
Lzf2, and its expression
during leech
embryogenesis.
Dev Bioll996,
175:205-217.
20.
DeRobertls EM, Sasal Y: A common plan for dorsoventral
patterning in Bilateria. Nature 1996, 380:37-40.
21.
McGinnis W, Krumlauf R: Homeobox
Cell 1992, 68:283-302.
reading
Papers of patilcular interest, publlshed within the annual period of review,
have been hIghlIghted as:
l
Belknap
Rogers BT, Kaufman TC: Structure of the insect head as revealed
by the EN protein pattern in developing
embryos. Development
1996, 122:341 Q-3432.
an exiting
e\wlution.
Acknowledgements
References
Boston:
16.
imply
identification
basis fol- ~morphological
Thought
and
dissociation
that
dots
of imor-
structures
of B/o/og/ca/
of
has clear
to the role that
are extended
interferes
into the genetic
indicts
of c\dutionag
orpnization
the
the uni-
diversity
more
dissociations
Idi
Kegulator)~
approach
has ken
is inherently
nunieroiIs
Ic\d
ply
this
lio~iiologoi~s genes and honiologous
Ixtween
one
as molecular
in generating
of regulatory
spccics,
come
attention.
for wit
although
plrlyxl
into
‘lb date,
far nxxe
I:ar less attention
I?;cnes ha\x
insights
morpholog):
have attracted
ha\-e extrcnie
limitations.
important
of aninial
7.
..
in niorphol-
Conclusions
unit)
the homology?
9.
cases
gent
nicthods
to del\~
where’s
Mayr E: The Growth
Press; 1982.
these
become
usa
possible
and other
As
and morphology:
19-121.
0.
cases
has the
espression
gene
Dlcklnson WJ: Molecules
Trends Genet 1995, II:1
the
At present,
in none
structure.
no\,el
cxpurinicntdly
6.
incrcasingl)
ha\-e
providing
been demonstrated to play a role in the cie\,elop-
qw2ition
Wray GA: Evolutionary
dissociations
between homologous
genes
and homologous
structures. In Homology (Novartis Foundation
Symp 222). ChIchester: Wiley; 1998:in press.
Abouhetf E, Akam M, Dlcklnson WJ, Holland PWH, Meyer A, Pate1 N,
Raff RA, Roth VL, Wray GW: Homology and developmental
genes.
Trends Genef 1997, 13:432-433.
This short paper outlines some pitfalls associated with using regulatory
genes as IndIces of homology at the morphologlcal
level: failure to distingulsh genes which are orthologous (gene copies derived from speciation)
and paralogous (gene copies derived from duplication);
use of the term
‘functlonal homology’ which confuses similarity because of a common evolutlonary origin (homology) with slmllarlty due to functlonal convergence
(analogy); and confllctlng compansons across different levels of biological
organlzatlon, such as genes, their expression patterns, their developmental roles, and the structures to which they give rise. The paper provides
practical solutions to avold these.
cor-
novelty
,.?5,15,16’]).
genes
5.
Sc~ral
is phylogenetically
association:
Gerhart J, Kirschner M: Cells, Embryos, and Evolution. Boston:
Blackwell Science; 1997.
A central theme of this book is the multiple uses to which regulatory
have been put during the course of animal evolution.
679
4.
Duboule D, Wilkins AS: The evolution of ‘bricolage:
Trends Genet
.
1998, 14:54-59.
This review discusses the recruitment of existing genes to new developmental roles In terms of the Idea of evolution as a ‘tlnkenng process.
‘I’his
[4*,1X’].
Wray and Abouheif
3.
.
of 1iior1)Iiolo~ic;II
of 3 morl~~iological
hypothesis
guilt
in
de\.elop”iciital
ofc~~~lution.
origin
existing
[1X*,22”
acti\-ities
changes
on the acquisition
recruitment
(c.g.
this
in\~ol\~e
in part,
from
role
the
idcntificd
for testing
all
roles
whcrc
that
at Icast
tar-
struCturt‘s.
on nc\v
roles and lost
nmvlties
their
cvolutionar)
the origin
can clearly
and
not homology?
genes and axial patterning.
Problem. Oxford: Oxford
Zuckerkandl E: Molecular pathways to parallel evolution:
I. Gene
nexuses and their morphological
correlates. J MO/ Evol 1994,
39:661-678.
22. Shubin N, Tabln C, Carroll S: Fossils, genes, and the evolution of
..
animal limbs. Nature 1997, 388:639-648.
This revjew provides Instructive analyses of comparative gene expressIon
data during limb development In arthropods and chordates. The dIscussIons
hIghlIght the importance of a robust phylogenetlc framework and a detailed
680
Genomes
and evolution
understandmg of comparative anatomy for understanding the role that particular genes have played In the course of morphological evolution.
35.
Falclanl F, Hausdorf B, Schroder R, Akam M, Tautz D, Denell R,
Brown S: Class 3 Hox genes in insects and the origin of zen.
Proc /Vat/ Acad Sci USA 1996, 93:8479-8484.
23.
Satokata I, Benson G, Maas R: Sexually dimorphic sterility
phenotypes in Hex a-10 deficient mice. Nature 1995, 374:460-463.
36.
Slack JMW: A Rosetta Stone for pattern formation
Nature
1984, 310:364-365.
24.
Ntisslein-Volhard C, Wleschaus E: Mutations affecting segment
number and polarity in Drosophila. Nature 1980, 287:795-801.
37.
Holland ND, Panganiban G, Henyey EL, Holland LZ: Sequence and
developmental
expression
of AmphiD//, an amphioxus Distal-less
gene transcribed in the ectoderm, epidermis and nervous system:
insights into the evolution of the craniate forebrain and neural
crest. Developmenf 1996, 122:291 l-2920.
25.
Doe CCI Smouse D, Goodman CS: Control of neuronal fate by the
Drosophila segmentation gene even-skipped. Nature 1988,
333:376-388.
26.
Pate1 NH, Condron BG, Zinn K: Pair-rule expression
patterns of
even-skipped are found in both short- and long-germ beetles.
Nature 1994, 367:429-434.
2 7.
Pate1 NH, Ball EE, Goodman CS: Changing role of even-skipped
during the evolution of insect pattern formation. Nature 1992,
357:339-342.
2%. Grbic M, Strand MR: Shifts in the life history of parasitic wasps
.
correlate with pronounced alterations in early development.
Proc /Vat/ Acad Sci USA 1998, 95:i 097-l 101.
This paper provides a good example of how the developmental role of a regulatory gene can change in assoclatlon with shifts in an organism’s life history. It describes
the surprising evolutionary
loss of the well-studted
segmentation role of eve.
29.
Clone T: The Drosophila sex determination signal: how do flies
count to two? Trends Genet 1993, 9:385-390.
30.
Bopp D, Calhoun G, Horabin JI, Samuels M, Schedl P: Sex-specific
control of Sex-lethal is a conserved mechanism for sex
determination in the genus Drosophila. Developmenf 1996,
122:971-982.
31.
.
Meise M, Hllfikerklelner D, Dubendorfer A, Brunner C, Nothiger R,
Bopp D: Sex-lethal, the master sex-determining gene in
Drosophila, is not sex-specifically regulated in Musca
domesfica. Development 1998, 125:1487-l 494.
This, and the folIowIng report 132’1, demonstrate that a regulatory gene (Sx/)
can become Involved In a developmental process (sex determination) after
that process first evolved. It also demonstrates that Sxl IS probably not a
‘master regulatory gene’ for sex deteremlnatlon In most fltes.
32.
.
Saccone G, Peluso I, Artiaco D, Glordano E, Bopp D,
Pollto LC: The Ceratifis capifafa homologue of the Drosophila sex
determining gene sex-lethal is structurally conserved, but not
sex-specifically regulated. Development 1998, 125:1495-l
500.
See annotation [31’1.
33.
Dawes R, Dawson I, Faicianl F, Tear G, Akam M: Dax, a locust Hox
gene related to fushi-farazu but showing no pair-rule expression.
Development 1994, 120:1561-l
572.
34.
Brown S, Hilgenfeld RB, Denell RE: The beetle Tribolium casfaneom
has a fushi farazu homolog expressed in stripes during
segmentation.
froc Nat/ Acad Sci USA 1994, 91 :12922-l 2926.
in animals?
38.
.
Wada H, Saiga H, Satoh N, Holland PWH: Tripartite organization
of
the ancestral chordate brain and the antiquity of placodes:
insights from ascidian Pax-21518, Hox. and Ofx genes.
Development 1998, 125:1113-l
122.
The authors illustrate how the expression domalns of several genes provide
a more secure interpretation of morphologlcal homologies among species
than does a single gene. The dtscusslon provides a clear example of apply1ng expression data to understanding the history of morphology.
39.
.
Averof M, Pate1 N: Crustacean
appendage evolution associated
with changes in Hox gene expression.
Nature
1997,
388:682-686.
This paper provides another careful appllcatlon of gene expression data to
understanding the history of morphologlcal change.
40.
Sordlno P, Duboule D: A molecular approach to the evolution of
vertebrate paired appendages.
Trends Ecol Evol1996.
11 :114-l 19
41.
Balker JA, Raff RA: Developmental
genetics
homology. Bmessays
1996, 18:489-494.
42.
Holland PWH: How gene duplication
affects homology. In
Homology (Novartis Foundation Symp 222). ChIchester: Wiley;
1998:ln press.
43.
Nlteckl M (Ed): Evoluhonary
Chlcago Press; 1989.
44.
Mtiller GB, Wagner GP: Novelty in evolution: restructuring
concept. Ann Rev Ecol Syst 1991, 22:229-256.
45.
Carroll SB, Gates J, Keys DN, Paddock SW, Panganlban G,
Selegue JE, Wllllams JA: Pattern formation and eyespot
determination in butterfly wings. Science 1994, 265:109-l
InnovaOons.
and traditional
Chicago:
University of
the
14.
46.
.
Panganlban G, Irvine SM, Lowe C, Roehl H, Corley LS, Sherbon B,
Grenier JK, Fallon JF, Klmble J, Walker M et a/.: The origin and
evolution of animal appendages. Proc Nat/ Acad So USA 1997,
94:5162-5166.
Apparent cases of parallel recruitment of D&al-less
Into the development of
appendages that are not themselves homologous are documented En this
paper. The data illustrate one of the pltfalls of using regulatory gene expression to infer homology of structures.
47.
Bharathan G, Janssen B-J, Kellogg EA, Slnha A: Did homeodomain
proteins duplicate before the origin of angiosperms,
fungi, and
metazoa? Proc Nat/ Acad So USA 1997, 94:13749-l 3753.