Current Biology, Vol. 14, R210–R211, March 9, 2004, ©2004 Elsevier Science Ltd. All rights reserved. DOI 10.1016/j.cub.2004.02.025
Hox Genes: Did the Vertebrate
Ancestor Have a Hox14?
David E.K. Ferrier
A new group of vertebrate Hox genes, Hox14, has
been discovered in two fish, the horn shark and the
coelacanth. This reveals that ancestrally all jawed
vertebrates may have had fourteen Hox paralogue
groups rather than the thirteen groups of mammals
and teleost fish.
The Hox gene cluster has served as a core paradigm
for developmental biology over the past twenty years.
The Hox genes play pivotal roles in axial patterning of
possibly all animals. Despite intensive research on
these genes, completely new Hox genes are still cropping up, particularly with the examination of a wider
range of animal taxa that goes beyond the model
organisms used by most developmental biologists.
One distinguishing feature of the Hox genes that has
intrigued many a developmental biologist is that the
genes are clustered in the genome of virtually all
animals that have been examined; although this is by
no means universally so (reviewed in [1]). The Hox clusters are not simply an aggregation of genes, but instead
form a distinctively organised series along the chromosome. Genes at the 3’-end of the cluster (Hox1 or labial
genes) pattern the anterior of the respective animal.
Progressing through the cluster, successive genes
pattern more caudal regions of the embryo, with the
genes at the 5’-end of the cluster patterning the most
posterior part of the organism. These ‘posterior’ Hox
genes comprise the Abdominal-B (Abd-B) genes of
ecdysozoans (e.g. insects and nematodes), the Post1
and Post2 genes of lophotrochozoans (e.g. annelids
and molluscs) and the genes from Hox9 onwards in
deuterostomes (e.g chordates and echinoderms) [2].
The Hox14 genes of horn sharks and coelacanths discovered by Powers and Amemiya [3] are the latest, surprising addition to the posterior Hox gene family, and
drastically alter the way we now have to think about the
evolution of this region of the Hox cluster in vertebrates.
Chordate Posterior Hox Genes
The chordate phylum encompasses the sea squirts
(Urochordata), amphioxus (Cephalochordata) and the
vertebrates. Comparisons of entire Hox gene clusters
across a phylogenetically wide spectrum of vertebrates,
including ourselves, mice, and several fish species had
led to the consensus view that vertebrates possess a
total of thirteen Hox gene paralogy groups [4,5].
Although vertebrates possess multiple Hox gene clusters, all of these have arisen by repeated duplication of
a single ancestral cluster around the time of origin of
Department of Zoology, University of Oxford, Tinbergen
Building, South Parks Road, Oxford OX1 3PS, UK.
E-mail: [email protected]
Dispatch
the vertebrates, or soon after within the vertebrate
lineage [6]. The gene clusters are thus said to be ‘paralogous’, i.e. homologues that have arisen by gene duplication within a single line of descent. The four Hox
clusters of mice (A, B, C and D) are paralogous, and
similarly the Hox1 gene of the HoxA cluster is paralogous to the Hox1 gene of the HoxB and HoxD clusters
(Hox1 has been lost from the mouse HoxC cluster).
Consequently, the genes of a paralogous group such as
the Hox1 group are more closely related to one another
than to their neighbours within each cluster. Across the
entire length of the mammal and fish clusters no more
than thirteen Hox paralogy groups (Hox1–Hox13) were
found. The ancestor of the vertebrates, prior to the
genome-wide duplication events, was thus presumed
to have possessed thirteen Hox genes [4].
This view was thrown into doubt when a fourteenth
Hox gene (AmphiHox14) was found in the cephalochordate amphioxus (Branchiostoma floridae), the invertebrate sister group to the vertebrates [7]. Amphioxus has
a single Hox cluster, which was viewed as being prototypical to the Hox clusters of vertebrates, and contains
a single representative of each of the vertebrate paralogy groups [7]. The discovery of AmphiHox14 raised
the possibility that the vertebrate ancestor had possessed fourteen paralogy groups, and that the fourteenth group had been subsequently lost from whole
swathes of vertebrate taxa.
Alternatively, AmphiHox14 might just be an evolutionary oddity of the amphioxus lineage. Perhaps
AmphiHox14 arose from a tandem duplication of
AmphiHox13, after cephalochordates and vertebrates
had diverged. This view seemed more reasonable
than postulating the loss of an entire paralogy group
from the vertebrates. Uncertainty remained, however,
because of the fact that the posterior Hox genes of
deuterostomes seem to evolve at much higher rates
than the posterior Hox genes of protostomes, or the
non-posterior Hox genes of all bilaterians (deuterostome posterior flexibility [7]). This fast rate of evolution confounds attempts to resolve the relationships
between presumptive orthologous genes of the
deuterostome phyla and subphyla.
Vertebrate Phylogeny and the Hox Cluster of the
Vertebrate Ancestor
Fish are an incredibly diverse, species-rich group of
vertebrates. This diversity is encompassed largely by
the bony fish, which includes the model species of
zebrafish and pufferfish. The genomic analysis of the
Hox clusters of fish species chosen because of their
phylogenetic position within the vertebrates is now
helping to delineate the evolutionary history of this
gene cluster. The most basal living vertebrates are the
jawless fish (Agnatha). Once a diverse, speciose group
of animals, with a heavily armoured exterior and
lengths of up to 1.5 metres in the seas of the Silurian
and Devonian, today the Agnatha are represented only
Current Biology
R211
Figure 1. Hox14 genes and vertebrate
phylogeny.
The fish clusters represent a single consensus cluster of the multiple Hox clusters
present in each animal. The Hox3–11 paralogy groups are omitted for simplicity.
The purple box encloses all of the gnathostomes and their ancestor. The gnathostome ancestor had a Hox14 gene. Hox14
genes have been lost from the lineages
leading to zebrafish and pufferfish, and to
humans. The green box encloses the
hypothetical vertebrate ancestor, prior to
the genome duplication events. It is not yet
resolved whether the Agnatha have a
Hox14 and, thus, whether the vertebrate
ancestor also had a Hox14. Amphioxus is
the outgroup to the vertebrates. Branch
lengths are of no significance.
'Anterior'
'Posterior'
Hox3-11 12 13 14
1 2
Cephalochordata
(not shown)
Amphioxus
Vertebrate
ancestor
Agnatha
Lamprey/hagfish
?
?
Gnathostome
ancestor
Chondrichthyes
Actinopterygii
Osteichthyes
Horn shark
Zebrafish/
Pufferfish
X
Coelacanth
Sarcopterygii
Human
X
Current Biology
by hagfish and lampreys. Vertebrates with jaws are
called Gnathostomata. The living gnathostomes can be
divided into two groups based on their skeleton: cartilaginous fish (Chondrichthyes) and bony fish (Osteichthyes), that includes the lobe-finned fish and the
tetrapods. Chondrichthyans are most familiar to us as
sharks and rays, including the horn shark (Heterodontus francisci). The osteichthyans are divided into the
actinopterygians (bony ray-finned fish, such as
zebrafish and pufferfish) and the sarcopterygians (the
lobe-finned fish and tetrapods). We are tetrapods, and
are basically little more than fancy fish! Among the sarcopterygians, the coelacanth (Latimeria), is renowned
for being a 'living fossil', thought to have been extinct
for 65 million years, until it was fished up and rediscovered in a South African market-place in 1938.
Up to now, it was supposed that the possession of
only 13 Hox paralogy groups in tetrapods (mice,
humans, birds) and in ray-finned fish (zebrafish, pufferfish and a basal actinopterygian [9]), was a characteristic of all vertebrates, and hence was likely to reflect the
condition of the vertebrate ancestor. The discovery of
coelacanth and horn shark Hox14 genes, which according to molecular phylogeny clearly belong to the same
paralogy group, now overturns this scenario [3]. The fact
that the coelacanth and shark Hox14 genes are paralogues is important, as it excludes an independent
origin of these genes. Thus, we can conclude that the
last common ancestor of the sarcopterygians and chondrichthyans also had a Hox14 gene (Figure 1).
In addition, the Hox14 paralogy group genes must
have been lost independently in the actinopterygian and
tetrapod lineages. This probably represents an example
of extensive gene loss, as multiple Hox clusters were
already present before the origin of the gnathostomes
[6]. This process is still in progress, judging from the
presence of a Hox14 pseudogene in the horn shark
HoxA cluster [3]. The question remains open, why these
genes are so readily lost from some lineages, whilst
being retained in others for a total of over 800–900
million years (summing the two sides of the phylogeny
after the chondrichthyan/osteichthyan divide).
It is now clear that a Hox14 gene orthologous to those
of coelacanths and sharks was present in the ancestor
of all gnathostomes (Figure 1). However, it is less clear
whether the amphioxus AmphiHox14 gene is orthologous to the Hox14 genes of gnathostomes, and hence
whether the ancestor of all vertebrates had a Hox14
gene. This uncertainty arises due to the rapid evolution
of these genes and the phenomenon of deuterostome
posterior flexibility [7]. However, the similar genomic
organisation of the gnathostome Hox14 and AmphiHox14 genes could be taken as indicative of orthology
[3]. There may be a hope of resolution. Do the Agnatha
contain a Hox14? There is some information on the
lamprey Hox clusters, but unfortunately the region
potentially containing a Hox14 has not been analysed
[10,11]. Should an agnathan Hox14 be found, its
sequence could potentially provide a phylogenetic
signal revealing orthology of all chordate Hox14 genes.
This would prove that the vertebrate ancestor did
indeed have a Hox14 after all.
References
1. Ferrier, D.E.K., and Minguillón, C. (2004). Evolution of the
Hox/ParaHox gene clusters. Int. J. Dev. Biol. 47, 605-611.
2. de Rosa, R., Grenier, J.K., Andreeva, T., Cook, C.E., Adoutte, A.,
Akam, M., Carroll, S.B., and Balavoine, G. (1999). Hox genes in brachiopods and priapulids and protostome evolution. Nature 399,
772-776.
3. Powers, T.P., and Amemiya, C.T. (2004) Evidence for a Hox14
paralog group in vertebrates. Curr. Biol., this issue.
4. Amores, A., et al. (1998). Zebrafish Hox clusters and vertebrate
genome evolution. Science 282, 1711-1714.
5. Amores, A., Suzuki, T., Yan, Y., Pomeroy, J., Singer, A., Amemiya,
C., and Postlethwait, J.H. (2004). Developmental roles of pufferfish
Hox clusters and genome evolution in ray-fin fish. Genome Res. 14,
1-10.
6. Holland, P.W.H. (1999). Gene duplication: Past, present and future.
Seminars. Cell Dev. Biol. 10, 541-547.
7. Ferrier, D.E.K., Minguillón, C., Holland, P.W.H., and Garcia-Fernàndez, J. (2000). The amphioxus Hox cluster: deuterostome posterior
flexibility and Hox14. Evol. Devel. 2, 284-293.
8. Garcia-Fernàndez, J., and Holland, P.W.H. (1994). Archetypal organization of the amphioxus Hox gene cluster. Nature 370, 563-566.
9. Chiu, C., et al. (2004). Bichir HoxA cluster sequence reveals surprising trends in ray-finned fish genomic evolution. Genome Res.
14, 11-17.
10. Force, A., Amores, A., and Postlethwait, J.H. (2002). Hox cluster
organization in the jawless vertebrate Petromyzon marinus. J. Exp.
Zool. (Mol. Dev. Evol.) 294, 30-46.
11. Irvine, S.Q., Carr. J.L., Bailey, W.J., Kawasaki, K., Shimizu, N.,
Amemiya, C.T., and Ruddle, F.H. (2002). Genomic analysis of Hox
clusters in the Sea Lamprey Petromyzon marinus. J. Exp. Zool.
(Mol. Dev. Evol.) 294, 47-62.