Expansion, diversification, and expression of T-box family genes in Porifera.

Abstract

Sponges are among the earliest diverging lineage within the metazoan phyla. Although their adult morphology is distinctive, at several stages of development, they possess characteristics found in more complex animals. The T-box family of transcription factors is an evolutionarily ancient gene family known to be involved in the development of structures derived from all germ layers in the bilaterian animals. There is an incomplete understanding of the role that T-box transcription factors play in normal sponge development or whether developmental pathways using the T-box family share similarities between parazoan and eumetazoan animals. To address these questions, we present data that identify several important T-box genes in marine and freshwater sponges, place these genes in a phylogenetic context, and reveal patterns in how these genes are expressed in developing sponges. Phylogenetic analyses demonstrate that sponges have members of at least two of the five T-box subfamilies (Brachyury and Tbx2/3/4/5) and that the T-box genes expanded and diverged in the poriferan lineage. Our analysis of signature residues in the sponge T-box genes calls into question whether "true" Brachyury genes are found in the Porifera. Expression for a subset of the T-box genes was elucidated in larvae from the marine demosponge, Halichondria bowerbanki. Our results show that sponges regulate the timing and specificity of gene expression for T-box orthologs across larval developmental stages. In situ hybridization reveals distinct, yet sometimes overlapping expression of particular T-box genes in free-swimming larvae. Our results provide a comparative framework from which we can gain insights into the evolution of developmentally important pathways.

Full text

Page 1

ORIGINAL ARTICLE
Expansion, diversification, and expression of T-box family
genes in Porifera
Kay Holstien & Ajna Rivera & Pam Windsor & Siyu Ding &
Sally P. Leys & Malcolm Hill & April Hill
Received: 2 August 2010 /Accepted: 28 October 2010 /Published online: 17 November 2010
# Springer-Verlag 2010
Abstract Sponges are among the earliest diverging lineage
within the metazoan phyla. Although their adult morphology
is distinctive, at several stages of development, they possess
characteristics found in more complex animals. The T-box
family of transcription factors is an evolutionarily ancient
gene family known to be involved in the development of
structures derived from all germ layers in the bilaterian
animals. There is an incomplete understanding of the role
that T-box transcription factors play in normal sponge
development or whether developmental pathways using the
T-box family share similarities between parazoan and
eumetazoan animals. To address these questions, we
present data that identify several important T-box genes
in marine and freshwater sponges, place these genes in a
phylogenetic context, and reveal patterns in how these
genes are expressed in developing sponges. Phylogenetic
analyses demonstrate that sponges have members of at
least two of the five T-box subfamilies (Brachyury and
Tbx2/3/4/5) and that the T-box genes expanded and
diverged in the poriferan lineage. Our analysis of signature
residues in the sponge T-box genes calls into question
whether “true” Brachyury genes are found in the Porifera.
Expression for a subset of the T-box genes was elucidated
in larvae from the marine demosponge, Halichondria
bowerbanki. Our results show that sponges regulate the
timing and specificity of gene expression for T-box
orthologs across larval developmental stages. In situ
hybridization reveals distinct, yet sometimes overlapping
expressionofparticularT-boxgenes infree-swimming larvae.
Our results provide a comparative framework from which we
can gain insights into the evolution of developmentally
important pathways.
Keywords Porifera.T-box.Basal metazoa.Halichondria
Introduction
The earliest diverging metazoans are an important group of
organisms to study if we are to understand the evolutionary
history and emergence of animal body plans. Though the
relationships of the basal metazoan taxa (i.e., Porifera,
Cnidaria, Ctenophora, and Placozoa) are still uncertain
from a phylogenetic perspective (e.g., see Sperling et al.
2009), our most current understanding of metazoan
phylogenomics places sponges at the base of the animal
lineage (e.g. Srivastava et al. 2008; Sperling et al. 2009;
Pick et al. 2010). Thus, the phylogenetic position of the
Porifera as a metazoan outgroup to all other animals
makes it a crucial group for exploring hypotheses
regarding early animal history. Furthermore, sponges
Communicated by M. Martindale
Electronic supplementary material The online version of this article
(doi:10.1007/s00427-010-0344-2) contains supplementary material,
which is available to authorized users.
K. Holstien:A. Rivera:S. Ding:M. Hill:A. Hill (*)
Department of Biology, University of Richmond,
28 Westhampton Way,
Richmond, VA 23173, USA
e-mail: ahill2@richmond.edu
P. Windsor:S. P. Leys
Department of Biological Sciences, University of Alberta,
CW 405, Biological Sciences Bldg,
Edmonton, AB, Canada T6G 2E9
Present Address:
K. Holstien
Program in Developmental Biology, Baylor College of Medicine,
One Baylor Plaza T603,
Houston, TX 77030, USA
Present Address:
A. Rivera
Department of Biological Sciences, University of the Pacific,
3601 Pacific Avenue,
Stockton, CA 95211, USA
Dev Genes Evol (2010) 220:251–262
DOI 10.1007/s00427-010-0344-2

Page 2

possess important body plan features that include both
shared and derived characteristics when compared to other
animals. Among the shared metazoan characteristics, one
finds polarized body plans in larvae and adult sponges, as
well as differentiated epithelia and sensory cells. Traits
found in sponges but not the rest of the Metazoa include a
canal system for pumping water built with choanocytes in
the choanosome. Thus, sponges provide the only opportunity
for comparative studies directed at (1) understanding the
genetic regulatory components that were in place prior to the
advent of nerves, complex tissues, and complex body
plans (e.g. Simpson 1984; Larroux et al. 2006; Nichols et
al. 2006; Sempere et al. 2006) and (2) how important
components of the genetic regulatory machinery were utilized
in an animal lineage with simple tissues and body plans.
Recent analyses of the genetic toolkit present in sponges
and its regulation of development have yielded important
insights regarding the early evolution of animals (e.g.
Larroux et al. 2008; Sakarya et al. 2007). An important
family of genes known to play crucial roles in a wide
variety of developmental processes across the animal
kingdom is the T-box family. This evolutionarily ancient
class of transcription factors contains a large DNA-binding
domain, and T-box proteins exhibit a strong conservation of
DNA binding functions among family members (Naiche et
al. 2005). The eponymous T-box family member, T
(Brachyury), is instrumental in vertebrate mesodermal
formation and notochord differentiation (Herrmann et al.
1990; Holland et al. 1995). Other members of the T-box
family play roles in endodermal and ectodermal specification
invertebrates,functioninextraembryonictissues,gastrulation
and patterning of embryonic mesoderm, cardiogenesis, limb
patterning, craniofacial development, pituitary cell fate,
and T-cell differentiation (reviewed in Naiche et al.
2005). Given the expansive roles of T-box genes, it is
not surprising that the gene family is quite large with 18
members in mammals (Wilson and Conlon 2002). The
biologically important roles for T-box family members
have also been demonstrated in a variety of triploblastic
model systems other than the vertebrates. Some genes,
likeBrachyury, have homologous functions across bilaterians
in processes involving morphogenetic movements, while
other family members demonstrate species-specific roles
(reviewed in Wilson and Conlon 2002; Showell et al. 2004).
Comparative studies indicate that the gene family
includes five evolutionarily related subfamilies designated as
Brachyury/T, Tbx1 (including Tbx1/10, 15/18/22, 20), Tbx2
(including Tbx2/3, 4/5), Tbx6, and Tbr1 (including Tbr1,
Tbx21 and Eomes/Tbr2) (Papaioannou 2001).
T-box orthologs have been found in all of the basal
metazoans including cnidarians (Technau and Bode 1999;
Spring et al. 2002; Scholz and Technau 2003), ctenophores
(MartinelliandSpring2005; Yamada et al. 2007), placozoans
(Martinelli and Spring 2003), and sponges (Adell et al. 2003;
Manuel et al. 2004; Larroux et al. 2006; Larroux et al. 2008).
In the Porifera, T-box family members have been found in all
majorspongelineagesincludingthedemosponges,calcareous
sponges, and hexactinellids. T-box genes were not found,
however, in the genome of the choanoflagellate Monosiga
brevicollis (King et al. 2008), a unicellular protist belonging
to a lineage that shares ancestry with the animal lineage. This
finding supported a long held view that T-box genes arose in
the common ancestor of metazoans. Recently, however,
putative T-box genes were detected in sequences from the
genomes of the unicellular amoeba opisthokont, Capsaspora
owczarzaki which is a close relative of multicellular animals
and fungi (Broad Institute Sequence Database), and the
mesomycetozoean Amoebidium parasiticum (Mikhailov et
al. 2009). These findings reveal that T-box-like genes were
lost in the choanoflagellate lineage and evolved early in the
opisthokont lineage long before multicellularity evolved.
Among the basal metazoans, roles for T-box genes have
been elucidated only partially. The most in depth study of
T-box function in a diploblast is from the ctenophore
Mnemiopsis leidyi. Expression analysis of T-box genes
during gastrulation and early organogenesis in this organism
showed that all five ctenophore T-box family members
exhibited distinct expression patterns during gastrulation and
thatsomemembers are also expressed duringthe formation of
the mouth, presumptive mesendoderm, sensory organs, and
the tentacular system (Yamada et al. 2007). Further studies
demonstrated that the M. leidyi Brachyury ortholog (MlBra)
is expressed in ectodermal cells around the site of gastrulation
and in cells derived from the blastopore; it is also involved in
regulating morphogenetic movements involved with gastru-
lation as determined by morpholino oligonucleotide
knockdown (Yamada et al. 2010). Using Xenopus
embryos, it was shown that Xbra and MlBra are
functionally interchangeable, thus showing that the primitive
role of Brachyury is to regulate morphogenetic movements
involved in the blastopore (Yamada et al. 2010). In the
Hydrozoa, the Hydra Brachyury ortholog, HyBra1, is
expressed in endodermal cells of the head and plays a role
in head formation (Technau and Bode 1999). In the
anthozoan Nematostella vectensis, the Brachyury ortholog
is expressed around the blastopore (Scholz and Technau
2003), and another Brachyury gene is expressed at the site of
ingression in early jellyfish gastrula (Spring et al. 2002). In
placozoans, two T-box genes show distinct expression
patterns with the Brachyury-like ortholog expressed in
potential outgrowth zones while the Tbx2/3 ortholog is
expressed at the periphery of attached animals (Martinelli
and Spring 2003).
Adult expression patterns of T-box proteins have been
reported for only one species of demosponge, Suberites
domuncula (Adell and Müller 2005). In S. domuncula, a
252 Dev Genes Evol (2010) 220:251–262

Page 3

purported Brachyury ortholog is expressed in early sponge
cell cultures (“pre-primmorphs”), when cell–cell and cell–
matrix interactions are being established, and also in
adherent primmorphs during a stage of tissue reorganization
(Adell and Müller 2005). The latter finding may suggest a
role for this Brachyury-like gene in morphogenetic
movementsthroughregulationofcelladhesionandmigration.
A second Suberites Tbx gene (SdTbx2), a likely Tbx4/5
ortholog, is expressed during the first day of sponge cell
culture and in isolated cells of the mesohyl of adult
sponges suggesting possible roles in cell identity determina-
tion (Adell and Müller 2005). Expression of T-box proteins
has so far not been studied during sponge embryogenesis or
metamorphosis of larvae into adults, when ganstrulation-like
movements may occur (Leys 2004).
The relatively limited body of data regarding the
structure and function of the T-box class of transcription
factors among poriferan and diploblast animals underscores
why it is important that additional studies be conducted to
elucidate the early evolution of this gene family. Here, we
report T-box family members from degenerate PCR surveys
in both a freshwater and marine sponge (Halichondria sp.
and Ephydatia muelleri). We compare these sequences
with T-box genes from the genomes of Amphimedon
queenslandica, the homoscleromorph sponge Oscarella
carmela, and other known sponge T-box genes, to
illustrate that T-box gene duplication in Porifera may be
more extensive than was previously believed. We show
that some T-box family members appear to have been lost
in certain sponge lineages. We also present in situ
hybridization data on expression patterns during larval
development in a marine sponge (Halichondria sp.).
These findings show that T-box family members have
distinct yet overlapping expression profiles, and they
reveal patterns of expression at the anterior and posterior
ends, as well as along the larval midline in swimming
sponge larvae for two T-box family members.
Methods
Collection and rearing of sponges
Halichondria sponges were collected from the Chesapeake
Bay at Virginia Institute of Marine Science, Gloucester
Point, Virginia. In the laboratory, sponges were reared in
re-circulating seawater aquaria (all water was replaced
weekly) or were used immediately. For larval collection,
individual sponges were placed in beakers containing
filtered, sterile seawater. The mother sponges were
allowed to naturally release larvae into the water column.
Newly released, mature larvae of Halichondria were
easily collected as they swim toward the surface in slow
spirals and congregate at the air–water interface. Larvae
were washed several times in filtered, sterile seawater and
were grown in 24-well plates (≈50 larvae per well) in
filtered, sterile seawater, which was replaced daily.
For the purposes of this experiment, we defined four
basic developmental stages: free-swimming larvae, skating
larvae, attached larvae, and settled/spreading tissue. Free-
swimming larvae were periodically collected from the
mother sponge over the course of a day; since no larvae
were present when the mother sponge was initially placed
in the beaker, we considered all larvae collected during the
day to belong to the 0–8 h post-release age cohort. Before
settlement, the larvae of this species stop moving, sink to
the bottom of the container, lose the characteristic free-
swimming morphology, and skate or crawl along the
substrate. We call this stage “skating” since larvae look as
if they are gliding along the bottom surface. After the skating
stage,larvaeformanattachmenttothesubstrateandcannotbe
removed without force; we define this stage “attachment.”
After attachment, the larval cells begin “spreading” across the
bottom of the well, differentiate, and will eventually develop
into juvenile sponges (i.e., rhagons).
E. muelleri gemmules were collected and harvested from
sponges collected from a dam outflow near Griswold,
Connecticut, USA (41°35′4″ N, 71°55′15″ W). Gemmules
were washed in 3% hydrogen peroxide and re-washed several
times in sterile, cold, ×1 Strekal’s (Strekal and McDiffett
1974) media, and stored at 4°C in the dark until use.
Gemmules were hatched in ×1 Strekal’s media and grown to
developmental stages 0–6 as per Funayama et al. (2005).
Isolation of Tbx sequences
Halichondria bowerbanki and E. muelleri RNA was
isolated from either larvae, reaggregated adult tissue, or
hatched gemmules all of which were washed several times
in sterile media before RNA was isolated using the RNeasy
mini kit (Qiagen) according to manufacture’s protocol.
cDNA was made using Thermoscript Reverse Transcriptase
(Invitrogen) using oligo(dT) and/or random hexamer
primers.
Degenerate PCRs were performed using primer pairs
designed from an alignment of the conserved T-domain
from a collection of T-box genes isolated in other animals.
The following primer combinations were used in all
possible combinations including nesting when appropriate:
GRRMFP, NP(Y/F)AKAF, TAYQNE, NEMIVT or (T/N)
EMI(V/I)TK, FG(S/A)HWM. cDNA from Halichondria sp.
from pooled larval stages and from reaggregated adult
tissue or cDNA from Ephydatia isolated from gemmules
hatched and harvested across several developmental stages
served as template. Annealing temperatures ranged from
40°C to 55°C. All resulting PCR products of expected sizes
Dev Genes Evol (2010) 220:251–262253

Page 4

were excised and cloned using the TOPO TA Cloning Kit
(Invitrogen). Clones were sequenced using the SequiTherm
EXCEL II kit (Epicenter) on a LiCor DNA Sequencing
System or using an ABI sequencer. Sequences for each gene
wereextended(whenpossible)inthe5′and3′directionsusing
RACE (SMART RACE, Clontech). Genbank accession
numbers are given in Supplemental Fig. 1 where sponge
sequences are highlighted according to classes.
Phylogenetic analyses
Bilaterian species were chosen based on the availability
of whole-genome protein models. Sequences from two
deuterostomes, Homo sapiens and Strongylocentrotus
purpuratus, and two ecdysozoans, Caenorhabditis elegans
and Drosophila melanogaster, were obtained by using
BLAST searching of the NCBI genome databases.
Sequences from two lophotrochozoans, Capitella telata
and Lottia gigantea, were obtained by using BLAST
searching of the JGI genome databases (JGI, unpublished
data) as was placozoan sequence from Trichoplax adhaerans
(Srivastava et al. 2008), choanoflagellate sequence from M.
brevicollis (King et al. 2008), and cnidarian sequence from
N. vectensis (Putnam et al. 2007). Sequence from a second
cnidarian species, Hydra magnipapillata, was obtained from
NCBI. A single sponge genome, A. queenslandica, was
searched using BLAST at the Trace Archive draft genome
downloaded from NCBI. Four opisthokont genomes,
Fig. 1 Unrooted ML phylogenetic tree of the T-box gene family. The
T-box phylogeny was evaluated with both ML and Bayesian
methodologies. Bootstrap support values and posterior probabilities
are shown in the numerator and denominator, respectively. The
topology shown includes clades with >60% support. Major families
are identified with blue bars; sponge genes are identified in red.
Accession numbers for each of the entries are provided in the
supplementary files
254Dev Genes Evol (2010) 220:251–262

Page 5

Allomyces macrogynus, C. owczarzaki, Spizellomyces
punctatus, and Proterospongia sp., were searched using
BLAST at the Broad Institute website (http://www.broad
institute.org/annotation/genome/multicellularity_project/
MultiHome.html). Additional cnidarian, ctenophore, and
poriferan sequences were obtained from the published
literature (Bielen et al. 2007; Yamada et al. 2007; Martinelli
and Spring 2005; Adell et al. 2003; Spring et al. 2002) or
using the methods described above. EST sequences from
Acropora mellifera, a cnidarian coral, were also used and
were obtained by BLAST searches of a larval EST database
(http://sequoia.ucmerced.edu/SymBioSys/) and genomic
sequences from O. carmela were obtained via BLAST
searches by Dr. Scott Nichols (unpublished data).
Alignments were performed using Muscle (Edgar 2004)
implemented inSeaviewusing the default settings. Sequences
were trimmed to contain only the Tbox regions. Gblocks
was run under a variety of conditions and yielded subsets
of the sites we used in our phylogenetic analysis
(Dereeper et al. 2008). We employed both Maximum
Likelihood(ML)andBayesianapproachesinourphylogenetic
analyses.Amodelofsequenceevolutionwas determinedfrom
alignedsequences using ProtTest (v1.2.6, Abascaletal. 2005).
PhyML (Guindon and Gascuel 2003) was used for ML
analysis with 500 bootstrap replicates; in PhyML, gaps are
treated as ambiguous characters. All parameters were
optimized based on empirical data. Bayesian analysis, as
implemented in Mr. Bayes 3.1.2 (Huelsenbeck and Ronquist
2001), was performed until the average standard deviation of
split frequencies achieved stationarity (n=2,000,000
generations). We used four independent chains in our
analyses. The gamma distribution for among site substitution
rates was approximated using four rate categories with a
proportion of invariable sites. The first 25% of the samples
were discarded as burn-in. All branches with less than 60%
support were collapsed in the ML tree shown (Fig. 1).
Expression analyses
Tissues from the various developmental stages were either
stored in RNAlater (Ambion) overnight and then placed at
−80°C for subsequent RNA isolation or fixed for in situ
hybridization. RNA was isolated using the RNAeasy Mini
Kit (Qiagen) and treated on column with DNase I to limit
contaminating genomic DNA. For RT reactions, 200 ng of
RNAwas reverse transcribed using the Thermoscript RT kit
(Invitrogen) and subsequent PCRs were first carried out
using Platinum Taq DNA polymerase (Invitrogen) to test
gene specific primers and RT reactions. SYBR Green
chemistry and the Chromo4 (BioRad) were used for
qRT-PCR using cycling conditions of: 94°C for 3 min
followed by 30 s 94°C, 30 s 55–61°C, 1 min 72°C for
35cycles.Gene-specificprimerswereusedtoamplifyisolated
Halichondria Tbx genes for profiling expression by qRT-
PCR across developmental stages. In each case, controls
were performed to ensure expression levels were from
cDNA and not contaminating genomic DNA and that each
primer pair only amplified the target Tbx gene (as
determined by testing each primer pair on plasmids for
each Tbx gene). Actin was picked to standardize the
amount of expression calculated for Tbx genes at each
developmental stage because it has often been utilized in
other systems (including cnidarians) for this purpose
(McCurley and Callard 2008; Rodriguez-Lanetty et al.
2008; Yüzbaşıoğlu et al. 2010). Nonetheless, we used
qRT-PCR to compare actin mRNA levels at each devel-
opmental stage compared to total RNA amounts and
though some degree of variability was observed, the
expression stability across larval developmental stages
was high. Further, we have also used Ef1α (Siah et al.
2008; Curtis et al. 2010) as a standardization control for
some of these genes as well and do not see differences in
the relative expression profiles. For each gene, data from
two different batches of RNA were assessed, and all PCRs
were performed with two replicates.
To determine mRNA distribution in larvae, we used a
protocol similar to that described in Hill et al. (2010) but
adapted for the larvae. Briefly, sponge tissues were fixed
overnight in 4% paraformaldehyde, 0.02% glutaraldehyde
in sterile seawater and then transferred into ascending
concentrations of methanol, and stored in 100% methanol
at −80°C. Fixed tissues were rehydrated through a methanol
and PTw (×1 PBS containing 0.1% Tween-20) series.
Tissues were prepared for prehybridization through three
washes in ×1 PTw and one wash of ×1 PTw containing
Proteinase K (1 μg/mL) at 37°C (alternatively, in some
cases, tissues were washed for 30 min in detergent solution
(1% SDS, 0.5% Tween, 50 mM Tris–HCl pH 7.5, 1 mM
EDTA, and 150 mM NaCl) followed by six washes in PTw
with no differences observed). The tissues were then re-
fixed in 4% paraformaldehyde in PBS, washed twice
with 0.1 M TEA buffer (pH 8), and then treated once in
0.1 M TEA containing 0.25% acetic anhydride, followed
by two washes in ×1 PTw. Tissue was processed to a 1:1
solution of hybridization buffer (50% formamide, ×5
SSC, 50 mg/ml heparin, 0.25% Tween-20, 1% SDS,
100 μg/ml sheared salmon sperm DNA; pH 5) and PTw.
Tissue was then prehybridized in hybridization solution
at 60°C for at least 3 h. All probes were labeled using
the Dig RNA labeling kit (Roche). After overnight
hybridization at 45–60°C, tissue was washed seven times
in hybridization solution at 60°C and gradually processed
to room temperature through half washes in TBST and
hybridization solution. Alternatively, after overnight
hybridization, tissue was washed three times in ×2
SSC, twice in ×1 SSC, and once in TBST or NTE
Dev Genes Evol (2010) 220:251–262255

Page 6

buffer containing 20 μg/mL RNAse A at 37°C, followed
by two washes of ×0.1 SSC at 37°C. After several
washesinTBST,tissuewasincubatedinTBSTcontaining1%
BSA to block nonspecific binding of antibody. Anti-Dig
alkaline phosphatase antibody (Roche) was diluted 1:3,000 in
TBST with BSA and larvae were incubated overnight in this
solution at 4°C. Larvae were washed five times in TBSTand
then processed for staining using NBT and BCIP in AP
reaction buffer. After staining, embryos were then cleared in
80% glycerol/PBS before imaging. For further validation of
the in situ staining patterns, for some genes (TbxA and Tbx),
two different probes were utilized that were directed to
different portions of the genes. In these cases, no differences
in staining patterns were observed. For TbxA, one probe was
labeled for position 697–928 nt which included the 3′ portion
of the T-domain. The alternate TbxA probe was 1,527 nt in
length and corresponded to the entire TbxA mRNA sequence
including 3′ UTR. For Tbx4/5, one probe included a 428 nt
coding region containing the majority of the T-domain and
the alternate probe was a 3′ RACE product that included a
195-bp overlap with the first probe (from forward 5′ primer:
GCGGTATGGGAGAAGCAGCTGAT) and extending into
the 3′ end of the clone.
Results and discussion
T-box gene families in basal metazoans: divergent
evolution
While the relationships between many T-box sub-families
remain ambiguous, several families, along with new
members of those families, have high support in our
phylogenetic analysis (Fig. 1). The clear sub-families,
supported by both Bayesian and Maximum Likelihood
analysis, include bilaterian-specific groups, Cnidaria
+Bilateria-specific groups, Porifera+Cnidaria+Bilateria-
groups, and Porifera-specific groups (Fig. 1). The sponge-
specific groups include a demosponge-specific TbxPor
clade and a demosponge-specific clade within the Tbx4/5
group (a homoscleromorph Tbx4/5 sequence is not included
in that clade). Also supported are a demosponge-specific
lineage we call TbxA and a demosponge-specific lineage
designated as TbxC/D. Finally, there is another group
identified in Fig. 1, TbxA/B/C/D/E, that is supported by
the ML tree, but not by the Bayesian tree. It is intriguing
that this group includes protist, poriferan, and placozoan
Tbx genes. These sequences in the TbxA/B/C/D/E group
may represent ancestral Tbx genes. However, this is a
region of the tree that could suffer from long-branch
attraction issues (Felsenstein 2004). Additional sequences
from other protists and basally branching animals would be
necessary to address this issue and the hypothesis that these
TbxA/B/C/D/E genes are ancestral and not lineage-specific
duplications.
Given that some T-box family clades do not contain
poriferan representatives and that the sponges appear to
contain unique T-box families, we propose that the
Urmetazoan had at least one T-box gene that underwent
several rounds of independent duplication and divergence
after the sponges split from other metazoans. While synteny
studies remain to be performed, our phylogenetic analyses
suggest that a proto-T-box gene evolved before the advent
of multicellularity since we find three T-boxes in the
genome of a unicellular opisthokont amoeba, C. owczarzaki
(50–60% similarity to H. sapiens T-brain T-box). This kind
of large-scale duplication and divergence has been observed
repeatedly for many eumetazoan “toolkit” genes (e.g.
Larroux et al. 2008; Yamada et al. 2007). However, the
presence of T-box genes in the protistan genome could be
the result of lateral gene transfer, though we believe
common ancestry is the most parsimonious explanation of
the data.
Our analysis recovered three demosponge-specific Tbx
clades. The TbxPor, TbxA, and TbxC/D clades have strong
support in both Bayesian and Maximum Likelihood
analysis. They are comprised of T-box sequences from
four, three, and two sponge species respectively, and
each clade contains two of the orders in the Demospongiae
(Haploclerida and Halichondrida). Additionally, our analysis
indicates that there are a number of divergent T-box genes in
Porifera (e.g., AqTbxE and several putative O. carmela Tbox
sequences) that are not associated with any specific T-box
clades. Another sponge sequence, AqTbxB groups with the
amoeboid C. owcsarzaki with weak support.
While the sponge-specific groups contain only demo-
sponge sequences (i.e., they lack representatives from the
Calcarea or Hexactinellida), it isunclear whether this is due to
demosponge-specificduplicationsorbecauseoftheextremely
small amount of sampling that has been done in Porifera.
However, with two poriferan genomes (A. queenslandica and
O. carmela (S. Nichols, unpublished data)), and our
exploration using degenerate PCR in H. bowerbanki and E.
muelleri, we can make inferences about the evolution of Tbx
genes in sponges. For example, it is clear that the non-
bilaterian Tbx clades contain a large number of sponge
members. This, and the presence of several sponge-specific
groups, is suggestive of extensive duplication and divergence
of Tbx genes specific in the sponge lineage.
Tbx1 subfamily (Tbx1/10, 15/18/22, 20)
The Tbx1 subfamily contains bilaterian, cnidarian, and
ctenophore members. This subfamily has no sponge
members, making it likely that it arose after sponges split
from the rest of Metazoa. The Tbx-1/10 clade groups with
256Dev Genes Evol (2010) 220:251–262

Page 7

the Tbx-15/18/20/22 clade, a clade with Bilateria+Cnidaria
genes. These groups were likely present in the bilaterian/
cnidarian ancestor, but not in the Urmetazoan unless
Tbx1 was lost in the sponge lineage. The evolutionary
history of the ctenophore (M. leidyi) Tbx1 gene is
unclear and since there is no ctenophore genome
sequenced and the position of ctenophores relative to
bilaterians and cnidarians is debated, there may be
additional ctenophore Tbx1 subfamily members. Our
Tbx1/10 group is consistent with results from previous
studies (Larroux et al. 2008; Yamada et al. 2007). In those
studies, sponge sequences AvTbx1/15/20 and AqTbx1/15/
20 were placed in the Tbx1 subfamily. In our analysis, a
separate poriferan-specific clade, that we call TbxPor, has
strong support. Whether this clade is part of the Tbx1
subfamily or a distinct Tbx lineage will require greater
resolution.
Tbx2 subfamily (Tbx2/3, Tbx4/5)
The Tbx2/3/4/5 subfamily genes are most often known for
their demonstrated roles in heart and eye development and
in the evolution of developmental programs involved in
appendage outgrowth and patterning across the vertebrates
(reviewed in Papaioannou 2001; Horton et al. 2008). While
our ML analysis (but not Bayesian) supports the grouping
of Tbx2/3 with Tbx1, though with low bootstrap support,
synteny analyses across Metazoa suggest that Tbx2/3 and
Tbx4/5 are more closely related (see below). However, a
fuller understanding of the true evolutionary history of the
origins and order of appearance of these gene families await
additional data from more species. By our analysis Tbx2/3
clade has high support and comprises genes from bilaterian,
cnidarian, placozoan, and ctenophore species (Fig. 1).
Sponges are not represented in the Tbx2/3 group. The
Tbx-4/5 sub-family is one of only two large clades in our
phylogeny with representatives from Porifera, Placozoa,
Cnidaria, and Bilateria (Fig. 1). We recovered a single clade
containing all known sponge Tbx4/5 members (Fig. 1).
Interestingly, while A. queenslandica and O. carmela have
single Tbx4/5 representatives, there are two E. muelleri
Tbx4/5 genes, possibly representing a duplication within the
spongillid sponge lineage. The lack of a ctenophore Tbx4/5
may be due to incomplete sampling or gene loss in this
lineage (it is known that Tbx4/5 has also been lost in two
major bilaterian lineages (Horton et al. 2008)).
Close linkage of the Tbx2/3 to the Tbx4/5 genes has been
reported across chordates (except ascidians) and cephalochor-
dates (Horton etal. 2008), and recently, it was shown that the
Nematostella genome contains a Tbx2/3 and Tbx4/5 gene in
the same orientation within 20 kb of each other (Yamada et
al. 2007). It has thus been proposed that a duplication of an
ancestral Tbx2/3/4/5 locus pre-dated the divergence of
modern diploblasts and triploblasts (Horton et al. 2008;
Yamada et al. 2007). There are two possible historical
explanations for the distribution of Tbx2 subfamily genes as
indicated by our data. First, the Tbx2/3 gene may have been
lost in the poriferan lineage. This would imply that
duplication of the Tbx2/3/4/5 ancestral gene occurred before
the sponges and Metazoa diverged; alternatively, that
duplication event occurred after sponges and Metazoa split.
The problem with this interpretation of the data is that there
is strong support for sponge Tbx4/5 genes in our phylogeny.
Either convergent evolution pushed the sponge lineage
toward the diploblast/triploblast Tbx4/5-like gene sequence,
or the ancestral Tbx2/3/4/5 gene was Tbx4/5-like.
Brachyury/T subfamily
The Brachyury clade is the only large group with represen-
tatives from all phyla sampled, including an amoeba
sequence from C. owczarzaki. As Brachyury is typically
associated with mesoderm development, gastruation, and
morphogenic movements, we were surprised to recover
Brachyury-like sequences from a unicellular organism.
Placozoan, ctenophore, and cnidarian sequences also fell
into this clade as well as sequences from three sponges—
Oopsacas minuta (OmBra—a hexactinellid), S. domuncula
(SdBra—a demosponge), and Sycon raphanus (SyBra—
calcareous sponge). Of particular relevance to this point, it
should be noted that neither the A. queenslandica genome or
the O. carmela genome possess a gene in the Brachyury
clade, nor did we recover a Brachyury-like gene from
Halichondria sp. or E. muelleri despite numerous attempts
with Tbx and Brachyury-specific primers. Several explana-
tions of this pattern are possible. It is possible that the
Brachyury-like gene was lost in some sponge lineages.
Alternatively, sponges may lack a true Brachyury gene, and
those sponges that do fall in this group may have converged
on similar signature sequences.
To further distinguish between these possibilities, we
examined key residues in the T-box regions of proteins
falling within the Brachyury clade (See Supplemental
Figure 2). While the placozoan, ctenophore, and cnidarian
sequences appear to represent true Brachyury proteins,
inspection of the sequence alignments reveals that sponge
and protist sequences are likely not true Brachyurys.
Brachyury is the most well studied Tbx gene family and,
as such, functionally and phylogenetically important
residues have been elucidated. Among putative DNA
and protein binding residues (∼30 total (Müller and
Herrmann 1997)), there are two Bra-specific residues that
differentiate Brachyury genes from other closely related T-
box family members (except S. purpuratus). These are a
diagnostic Lys106, involved in DNA-binding specificity
(rev. in Wilson and Conlon 2002), and Met-42, potentially
Dev Genes Evol (2010) 220:251–262257

Page 8

involved in dimerization (Müller and Herrmann 1997) (all
numbering is based on the Drosophila Byn T-box protein
sequence starting at LDDRELW). A third residue, Asn-85,
is found in all canonical Brachyury and Eomes/Tbx genes
recovered in our analysis, excepting S. purpuratus. This
residue is a potential synapomorphy of the Brachyury +
Eomes clade and is not found in protist or Sy-Bra.
Both the protist and SyBra lack Lys106 and have the
canonical T-box Arginine at this position (note that the
Eomes group has an Asparagine at this position, unlike
most other Tbox groups). Of the two sponges included in
the Brachyury clade we recovered (Fig. 1), the known
OmBra sequence is extremely short and lacks much of the
T-box domain. Of 30 residues directly involved in DNA
binding and protein–protein interactions, 11 have not yet
been elucidated from the OmBra sequence. The third
sponge sequence, SdBra does not have the Met-42.
Interestingly, the functionally important Alanine at position
171 in Brachyury genes is also found in the TbxPor group.
This residue is known to be involved in DNA-binding
(Müller and Herrmann 1997) and is typically replaced with
a Guanine in other T-box genes. Since Brachyury and
TbxPor are not sister groups, it seems likely that they
converged on this Alanine, especially due to the fact that it
has functional significance. The potential convergence in T-
box sequence is likely due to strong selective pressures on
these sites given the constraints of DNA-binding and
dimerization. This may also be responsible for the difficulties
encountered in resolving branching orders among the T-box
families(i.e.,thehighdegreeofpolytomy)andthepresenceof
non-Bra genes in the Brachyury sub-family. The presence of
T-box genes but absence of a Brachyury homolog in some
sponges may be important to our understanding how sponges
form tissues without undergoing the typical or “convention-
al” morphogenetic movements seen in gastrulation in other
animals (e.g., Stern 2004). Whereas sponge embryos show
cellular differentiation and form layers made of distinct cell
types during early development, a feeding epithelium
(equivalent of a gut) is only formed at metamorphosis (Leys
2004). T-box’s may, therefore, be involved in directing
morphogenetic events involved in the formation of polarity
during larval development and at metamorphosis.
Distinct larval expression profiles of Poriferan T-box genes
We used real time RT-PCR to assess relative levels of
expression for four Halichondria Tbx genes. For each gene,
we determined expression levels during settlement and
attachment relative to free-swimming larvae, across four
larval developmental stages (Fig. 2). The first stage
examined was free-swimming larvae, which consisted of a
pool of larvae that were collected from the top of the water
column between 0 and 24 h post-larval release. Though
these larvae are positively phototactic throughout the free-
swimming period (which typically lasts 48 h but can
continue past 72 h), they enter a stage where they swim at
or near the bottom of the dish where they may temporarily
settle. This behavior has been described in a variety of
sponges (see Simpson 1984) as “creeping,” “crawling,” or
“preattachment” and may involve cilia cell-substratum
interactions. We call this stage “skating” since the larval
behavior looks more like a gliding movement than a creep
or a crawl in Halichondria. In fact, the larvae are often
observed to be spinning on their axis as they glide along the
bottom of the dish. During the skating stage, larvae may
resume swimming near the bottom of the dish if a pulse of
water current is applied near their site of contact, but they
resume skating quickly after the disturbance. The next stage
we collected at included larvae that were attached to the
dish and clearly had basopinacocyte formation that adhered
to the larvae to the substrate. Finally, we collected larvae
that had begun the metamorphosis process with proliferative
(archaeocyte) cells that were “spreading” across the surface of
the dish. In this study, we did not follow development
through metamorphosis to the juvenile rhagon stage
because of a parasite that preys on the juveniles that
we could not eliminate from the cultures without also
compromising the sponge’s development. Future studies
using another species of demosponge (E. muelleri) will
be aimed at examining the role of orthologs to these Tbx
genes during metamorphosis and adult sponge function.
The TbxA gene showed the highest expression levels
during the “skating” stage relative to free-swimming and
had very low expression in spreading larvae. TbxC/D,
Tbx4/5, and TbxPor all showed the greatest expression
Fig. 2 qRT-PCR analysis of T-box gene expression during larval
developmental stages of Halichondria. Gene expression levels are
plotted relative to free-swimming larvae and normalized to actin
expression levels at each developmental stage. Y-axis denotes relative
levels of expression
258Dev Genes Evol (2010) 220:251–262

Page 9

levels at larval attachment to the substrate. For Tbx4/5,
expression during larval attachment is more than two-fold
higher than at the spreading stage and is more than three-
fold higher than free-swimming or skating stages. TbxC/D
expression at attachment is up to six-fold higher than all
other developmental stages. The overall expression
profile for TbxPor and Tbx4/5 are similar with attachment
as the highest expression level, spreading as the next
highest, followed by free-swimming larval expression and
skating as the lowest level of expression. We did not assay
expression levels of the TbxA2 gene since it was
discovered much later in our study (during RACE of
TbxA), and we would not have been able to compare
qRT-PCR data directly with the other genes since each
gene was analyzed from multiple matched sets of RNA/
cDNA at each developmental stage. Nonetheless, it is
clear from the expression profiles for TbxA, TbxC/D,
Tbx4/5, and TbxPor that there has been some level of
functional divergence of these Tbx genes. The different
levels of expression over several developmental stages
Fig. 3 Expression of T-box genes in early stage free-swimming
Halichondria larvae with whole-mount in situ hybridization. Anterior
poles of larva are oriented at the top left of each panel. The sense
probe for HbTbx 4/5 is shown, other sense probes also exhibited no
staining. HbTbxPor is highly expressed at the poster pole as indicated
by black arrow. HbTbxC/D is expressed in the columnar epithelial
layer (CE, black arrow) as well as throughout the subepithelial layer
(SE) that is directly beneath the CE and also in the ICM. HbTbx4/5 is
expressed throughout the ICM and SE, but not in the CE. HbTbxA2 is
expressed at the anterior end of the larvae and in cells around the inner
cell mass (black arrow). HbTbxA reveals an asymmetric expression
pattern on one side of the larvae as indicated by black arrow
Fig. 4 Expression of HbTbx4/5
and HbTbxPor in late stage free-
swimming Halichondria larvae
by whole-mount in situ hybrid-
ization. Anterior poles of larvae
are oriented at the top left. b and
e are higher magnifications of
the anterior poles and c and f are
higher magnifications of the
posterior poles. a–c HbTbxPor
expression is most concentrated
at the anterior and posterior pole
of the larvae with expression
extending along the midline,
mostly in the sub-epithelial cell
layer (black arrow). d–f
HbTbx4/5 expression is also
seen at the anterior and posterior
poles with expression extending
along the midline in the sub-
epithelial layer (black arrow)
Dev Genes Evol (2010) 220:251–262259

Page 10

suggests that the sponge T-box genes have undergone
divergence in their cis-regulatory regions, at least. This is
likely tied to some functional divergence, which might be
as simple as a split in the timing of deployment or as
complex as a completely novel function for one of the
duplicates.
To identify patterns of expression and possibly suggest
functions, the expression of sponge T-box genes were also
assayed by whole mount in situ hybridization to newly
released sponge free-swimming larvae (Fig. 3). These
larvae are about 250–300 μm long, and have a ciliated
columnar epithelium (CE). Immediately inside the CE is
a sub-epithelial layer of cells that surrounds a large
spicule-containing inner cell mass (ICM). At maturity,
the posterior pole of the larvae will have longer cilia
(∼36 μm in length compared to ∼12 μm around the rest
of the surface (see Fell and Jacob 1979)). Since all five
Halichondria T-box genes were expressed in 0–24 h free-
swimming larvae, we chose this stage for initial analysis.
Given that expression profiles for genes in this species of
larvae have not been reported, we include a supplemental
figure (Supp. Fig. 2) showing positive controls for
expression of actin which hybridizes in all cells and the
BarBsh gene which has previously been reported to be
expressed in the inner cell mass of A. queenslandica larvae
(Larroux et al. 2006). We also observe expression of BarBsh
exclusively in the inner cell mass of Halichondria larvae
thus illustrating that the expression patterns we observe in
this study have been validated with positive and negative
controls (Supp. Fig. 2).
Each T-box gene exhibited a distinct pattern of expression,
though some of the expression patterns (e.g., Tbx4/5 and
TbxC/D) were less pronounced. The TbxPor gene seemed
to be most concentrated at the posterior end and spread
toward the anterior within the inner cell mass. Tbx4/5
staining is nearly ubiquitous throughout the inner cell
mass and subepithelial layer at this stage; however, it does
not seem to be expressed in the columnar epithelium.
TbxC/D staining is evident in all cells (though most
concentrated in the inner cell mass), including the ciliated
columnar epithelium and it is the only T-box gene
identified that is evidently expressed in these cells (see
arrow). The TbxA gene has an interesting expression
domain that is concentrated on one lateral side of the
larvae (see arrow, Fig. 3). This apparently asymmetrical
pattern of expression in symmetrical larvae is enigmatic.
Finally, TbxA2 has an expression pattern that includes a
concentration of staining at the anterior end with several
small foci (see arrow) of cells around the outside of the
inner cell mass (Fig. 3). These data support the hypothesis
that the T-box genes have diverged in the poriferan lineage
and may perform sub-functionalized their roles in larval
development.
To further investigate the potential polarity of expression
observed for the TbxPor gene, we examined expression in
later stage free-swimming larvae. These larvae have a
pronounced ciliated posterior pole and can often be seen
making connections with the substrate at this end. Interest-
ingly, we see distinct expression of TbxPor (Fig. 4a–c) at
both the anterior and posterior poles of the larvae. The
expression at the posterior pole is most concentrated at the
far posterior and less concentrated in more anterior cells as
was seen in the earlier larvae (Fig. 3). At the anterior pole,
there are distinct cells, mostly along the subepithelial layer
that express TbxPor. The expression in the subepithelial
layer extends around the lateral sides of the larvae as well.
We also examined expression for Tbx4/5 in late free-
swimming larvae. Like TbxPor, expression is most concen-
trated at both the posterior highly ciliated pole and also at
the anterior pole. Furthermore, expression is also observed
around the lateral sides of the larvae in the subepithelial
layer. These interesting patterns of expression in free-
swimming larvae may suggest some involvement of
TbxPor and Tbx4/5 in establishing or maintaining axial
polarity in these early metazoans. Additionally, the
location of staining along the lateral sides of the larvae
is in cells that look quite similar to the “flask” cells
observed in the subepithelial layer of A. queenslandica
larvae. These cells have been shown to express a variety
of post-synaptic orthologs and have been suggested to be
evolutionary intermediates to neurons (Sakarya et al.
2007). We are currently developing gain and loss of
function methods in our lab that will help us test whether or
not either of these Tbx genes play roles in larval axis
formation, settlement, metamorphosis, or other aspects of
development. Given the conserved roles that some T-box
family members have played over the course of evolution,
investigating the roles of poriferan T-box genes during early
development and during metamorphosis or formation of the
adult body plan is particularly relevant.
Acknowledgements
Thomas F. and Kate Miller Jeffress Memorial Trust to A.H. and a
Beckman Scholars Program award to the University of Richmond.
Special thanks to Scott Nichols for providing T-box sequences for O.
carmela.
This work was supported by a grant from the
References
Abascal F, Zardoya R, Posada D (2005) ProtTest: selection of best-fit
models of protein evolution. Bioinformatics 21:2104–2105
Adell T, Müller WEG (2005) Expression pattern of the Brachyury and
Tbx2 homologues from the sponge Suberites domuncula. Biol
Cell 97:641–650
Adell T, Grebenjuk VA, Wiens M, Müller WEG (2003) Isolation and
characterization of two T-box genes from sponges, the phyloge-
netically oldest metazoan taxon. Dev Genes Evol 213:421–434
260Dev Genes Evol (2010) 220:251–262

Page 11

Bielen H, Oberleitner S, Marcellini S, Gee L, Lemaire P, Bode HR, Rupp
R, Technau U (2007) Divergent functions of two ancient Hydra
Brachyury paralogues suggest specific roles for their C-terminal
domains in tissue fate induction. Development 134:4187–4197
Curtis KM, Gomez LA, Rios C, Garbayo E, Raval AP, Perez-Pinzon
MA, Schiller PC (2010) EF1a and RPL13a represent normaliza-
tion genes suitable for RT-qPCR analysis of bone marrow derived
mesenchymal stem cells. BMC Mol Biol 11:61
Dereeper A, Guignon V, Blanc G, Audic S, Buffet S, Chevenet F,
Dufayard JF, Guindon S, Lefort V, Lescot M, Claverie JM,
Gascuel O (2008) Phylogeny.fr: robust phylogenetic analysis for
the non-specialist. Nuc Acid Res 36:465–469
Edgar RC (2004) MUSCLE: multiple sequence alignment with high
accuracy and high throughput. Nucleic Acids Res 32:1792–
1797
Fell P, Jacob W (1979) Reproduction and development of Halichondria
sp. in the Mystic estuary, Connecticut. Biol Bull 155:62–75
Felsenstein J (2004) Inferring phylogenies. Sinauer, Sunderland
FunayamaN,NakatsukasaM,HayashiT,AgataK(2005) Isolationofthe
choanocyte in the fresh water sponge, Ephydatia fluviatilis and its
lineage marker, Ef annexin. Dev Growth Differ 47:243–253
Guindon S, Gascuel O (2003) A simple, fast, and accurate algorithm
to estimate large phylogenies by Maximum Likelihood. Syst Biol
52:696–704
Herrmann BG, Labeit S, Poustka A, King TR, Lehrach H (1990)
Cloning of the T gene required for mesodermal formation in the
mouse. Nature 343:617–622
Hill A, Boll W, Ries C, Warner L, Osswalt M, Hill M, Noll M
(2010) Origin of Pax and Six gene families in sponges: single
PaxB and Six1/2 orthologs in Chalinula loosanoffi. Dev Biol
343:106–123
Holland PW, Koschorz B, Holland LZ, Herrmann BG (1995)
Conservation of Brachyury (T) genes in amphioxus and
vertebrates: developmental and evolutionary implications.
Development 121:4283–4291
Horton N, Mahadevan NR, Minguillon C, Osoegawa K, Rokhsar
DS, Ruvinsky I, de Jong PJ, Logan MP, Gibson-Brown JJ
(2008) Conservation of linkage and evolution of developmental
function within the Tbx2/3/4/5 subfamily of T-box genes:
implications for the origin of vertebrate limbs. Dev Genes
Evol 218:613–628
Huelsenbeck JP, Ronquist F (2001) MRBAYES: Bayesian inference
of phylogeny. Bioinformatics 17:754–755
King N, Westbrook MJ, Young SL, Kuo A, Abedin M, Chapman J,
Fairclough S, Hellsten U, Isogai Y, Letunic I, Marr M, Pincus D,
Putnam N, Rokas A, Wright KJ, Zuzow R, Dirks W, Good M,
Goodstein D, Lemons D, Li W, Lyons JB, Morris A, Nichols S,
RichterDJ,SalamovA,SequencingJGI,BorkP,LimWA,Manning
G, Miller WT, McGinnis W, Shapiro H, Tjian R, Grigoriev IV,
Rokhsar D (2008) The genome of the choanoflagellate Monosiga
brevicollis and the origin of metazoans. Nature 451:783–788
Larroux C, Fahey B, Liubicich D, Hinman VF, Gauthier M, Gongora
M, Green K, Worheide G, Leys S, Degnan BM (2006)
Developmental expression of transcription factor genes in a
demosponge: insights into the origin of metazoan multicellularity.
Evol Dev 8(2):150–173
Larroux C, Luke GN, Koopman P, Rokhsar D, Shimeld SM, Degnan
BM (2008) Genesis and expansion of metazoan transcription
factor gene classes. Mol Biol Evol 25(5):980–996
Leys SP (2004) Gastrulation in sponges. In: Stern C (ed) Gastrulation.
Cold Spring Harbor Laboratory Press, Cold Spring Harbor, pp
23–31
Manuel M, Le Parco Y, Borchiellini C (2004) Comparative analysis of
Brachyury T-domains, with the characterization of two new
sponge sequences from a hexactinellid and a calcisponge. Gene
340:291–301
Martinelli C, Spring J (2003) Distinct expression patterns of the two
T-box homologues Brachyury and Tbx2/3 in the placozoan
Trichoplax adhaerens. Dev Genes Evol 231:492–499
Martinelli C, Spring J (2005) T-box and homeobox genes from
ctenophore Pleurobrachia pileus: comparison of Brachyury,
Tbx2/3 and Tlx in basal metazoans and bilaterians. FEBS Lett
579:5024–5028
McCurley AT, Callard GV (2008) Characterization of housekeeping
genes in zebrafish: male–female differences and effects of tissue
type, developmental stage and chemical treatment. BMC Mol
Biol 9:102
Mikhailov KV, Konstantinova AV, Nikitin MA, Troshin PV, Rusin
LY, Lyubetsky VA, Panchin YV, Mylnikov AP, Moroz LL,
Kuman S, Aleoshin VV (2009) The origin of Metazoa: a
transition from temporal to spatial cell differentiation. Bioessays
31:758–768
Müller CW, Herrmann BG (1997) Crystallographic structure of the T
domain-DNA complex of the Brachyury transcription factor.
Nature 289:884–888
Naiche LA, Harrelson Z, Kelly RG, Papaioannou VE (2005) T-box
genes in vertebrate development. Annu Rev Genet 39:219–239
Nichols SA, Dirks W, Pearse JS, King N (2006) Early evolution of
animal cell signaling and adhesion genes. Proc Natl Acad Sci
USA 103:12451–12456
Papaioannou VE (2001) T-box genes in development: from hydra to
humans. Int Rev Cytol 207:1–70
Pick KS, Philippe H, Schreiber F, Erpenbeck D, Jackson DJ, Wrede P,
Wiens M, Alié A, Morgenstern B, Manuel M, Wörheide G (2010)
Improved phylogenomic taxon sampling noticeably affects non-
bilaterian relationships. Mol Biol Evol 27(9):1983–1987
Putnam NH, Srivastava M, Hellsten U, Dirks B, Chapman J, Salamov
A, Terry A, Shapiro H, Lindquist E, Kapitonov VV, Jurka J,
Genikhovick G, Grigoriev IV, Lucas SM, Steele RE, Finnerty JR,
Technau U, Martindale MQ, Rokhsar DS (2007) Sea anemone
genome reveals ancestral eumetazoan gene repertoire and
genomic organization. Science 317:86–94
Rodriguez-Lanetty M, Phillips WS, Dove S, Hoegh-Guldberg O, Weis
VM (2008) Analytical approach for selecting normalizing genes
from a cDNA microarray platform to be used in q-RT-PCR
assays: a cnidarian case study. J Biochem Biophys Methods
70:985–991
Sakarya O, Armstrong KA, Adamska M, Adamski M, Wang IF, Tidor
B, Degnan BM, Oakley KKS (2007) A post-synaptic scaffold at
the origin of the animal kingdom. PLoS ONE 2:e506
Scholz CB, Technau U (2003) The ancestral role of Brachyury:
expression of NemBra1 in the basal cnidarian Nematostella
vectensis (Anthozoa). Dev Genes Evol 212:563–570
Sempere LF, Cole CN, McPeek MA, Peterson KJ (2006) The
phylogenetic distribution of metazoan microRNAs: insights into
evolutionary complexity and constraint. J Exp Zool B Mol Dev
Evol 306(6):575–88
Showell C, Binder O, Conlon F (2004) T-box genes in early
embryogenesis. Dev Dyn 229:201–218
Siah A, Dohoo C, McKenna P, Delaporte M, Berthe FCJ (2008)
Selecting a set of housekeeping genes for quantitative real-time
PCR in normal and tetraploid haemocytes of soft-shell clams,
Mya arenaria. Fish Shellfish Immunol 25:202–207
Simpson TL (1984) The cell biology of sponges. Springer, New York
Sperling EA, Peterson KJ, Pisani D (2009) Phylogenetic-signal
dissection of nuclear housekeeping genes supports the paraphyly
of sponges and the monophyly of Eumetazoa. Mol Biol Evol 26
(10):2261–2274
Spring J, Yanze N, Josch C, Middel AM, Winninger B, Schmid V
(2002) Conservation of Brachyury, Mef2 and snail in the
myogenic lineage of jellyfish: a connection to the mesoderm of
Bilateria. Dev Biol 244:372–384
Dev Genes Evol (2010) 220:251–262261

Page 12

Srivastava M, Begovic E, Chapman J et al (2008) The Trichoplax
genome and the nature of placozoans. Nature 454:955–960
Stern C (2004) Gastrulation: from cells to embryo. Cold Spring
Harbor Press, Cold Spring Harbor
Strekal TA, McDiffett W (1974) Factors affecting germination,
growth, and distribution of the freshwater sponge, Spongilla
fragilis Leidy (Porifera). Biol Bull 146:267–278
Technau U, Bode HR (1999) HyBra1, a Brachyury homologue, acts
during head formation in Hydra. Development 126:999–1010
Wilson V, Conlon F (2002) The T-box family. Genome Biol 3:3008.1–
3008.7
Yamada A, Pang K, Martindale MQ, Tochinai S (2007) Surprisingly
complex T-box gene complement in diploblastic metazoans. Evol
Dev 9(3):220–230
Yamada A, Martindale MQ, Fukui A, Tochinai S (2010) Highly
conserved functions of the Brachyury gene on morphogenetic
movements: insight from the early-diverging phylum Ctenophora.
Dev Biol 339:212–222
Yüzbaşıoğlu A, Onbaşılar İ, Kocaefe Ç, Özgüç M (2010) Assessment
of housekeeping genes for use in normalization of real time PCR
in skeletal muscle with chronic degenerative changes. Exp Mol
Pathol 88:326–329
262Dev Genes Evol (2010) 220:251–262