Microarray identification of novel genes downstream of Six1, a critical factor in cranial placode, somite and kidney development

Abstract

Background

Six1 plays an important role in the development of several vertebrate organs, including cranial sensory placodes, somites and kidney. Although Six1 mutations cause one form of Branchio-Otic Syndrome (BOS), the responsible gene in many patients has not been identified; genes that act downstream of Six1 are potential BOS candidates.

Results

We sought to identify novel genes expressed during placode, somite and kidney development by comparing gene expression between control and Six1-expressing ectodermal explants. The expression patterns of 19 of the significantly up-regulated and 11 of the significantly down-regulated genes were assayed from cleavage to larval stages. 28/30 genes are expressed in the otocyst, a structure that is functionally disrupted in BOS, and 26/30 genes are expressed in the nephric mesoderm, a structure that is functionally disrupted in the related Branchio-Otic-Renal (BOR) syndrome. We also identified the chick homologues of 5 genes and show that they have conserved expression patterns.

Conclusions

Of the 30 genes selected for expression analyses, all are expressed at many of the developmental times and appropriate tissues to be regulated by Six1. Many have the potential to play a role in the disruption of hearing and kidney function seen in BOS/BOR patients.

INTRODUCTION

Vertebrate Six genes are highly related to the Drosophila sine oculis gene that is critical for fly visual system development (Cheyette et al., 1994). They play major roles in the development of the eye, but also in other sense organs, cranial sensory ganglia, the central nervous system, muscle, kidney, genitalia, limb buds and lungs (Oliver et al., 1995; Kawakami et al., 1996; Ohto et al., 1998; Spitz et al., 1998; Pandur and Moody, 2000; Ghanbari et al., 2001; Fougerousse et al., 2002; Laclef et al., 2003; Bessarab et al., 2004; Xu et al., 2003; Zheng et al., 2003; Brodbeck and Englert, 2004; Ozaki et al., 2004; Zou et al., 2004; Brugmann and Moody 2005; Grifone et al., 2005; Konishi et al., 2006; Zou et al., 2006; Ikeda et al., 2007; Chen et al., 2009; Ikeda et al., 2010; Sato et al., 2010; Guo et al., 2011). Several experiments indicate that Six1 has a central role in controlling development of the cranial sensory placodes, which are patches of thickened embryonic ectoderm that will give rise to the anterior pituitary, olfactory sensory epithelium, lens, auditory and vestibular inner ear structures, and the large, distally-located neurons of the cranial sensory ganglia (reviewed in Schlosser, 2007; Schlosser, 2010; Streit, 2004; Bailey and Streit, 2006; Streit, 2007; Ladher et al., 2010; Park and Saint-Jeannet, 2010; Graham and Shimeld, 2013; Saint-Jeannet and Moody, 2014). Loss of Six1 function or repression of its downstream targets in Xenopus, chick and zebrafish results in the loss of markers specific to the pre-placodal region (PPR) and placodes (Brugmann et al., 2004; Bricaud and Collazo, 2006, 2011; Schlosser et al., 2008; Christophorou et al., 2009). In particular, otic placode defects (Christophorou et al., 2009), and loss of inner ear hair cells (Bricaud and Collazo, 2006; Bricaud and Collazo, 2011) are observed. Six1-null mice have olfactory, inner ear and cranial sensory ganglion malformations (Oliver et al., 1995; Laclef et al., 2003; Zheng et al., 2003; Ozaki et al., 2004; Zou et al., 2004; Konishi et al., 2006; Ikeda et al., 2007; Chen et al., 2009; Ikeda et al., 2010). Conversely, Six1 gain-of-function specifically in the PPR expands the domains of placode genes in frog and chick (Brugmann et al., 2004; Christophorou et al., 2009), indicating that it is a key regulator of placode fate.

In humans, SIX1 mutations can cause Branchio-Otic Syndrome (BOS, specifically BOS3; Online Mendelian Inheritance in Man [OMIM] #608389; www.omim.org), whose phenotypes include mild craniofacial defects and significant hearing loss (Ruf et al., 2004). Nine mutations in BOS3 patients from 16 unrelated families have been reported to date. Seven are missense mutations in an N-terminal domain, called the Six Domain (SD; Kawakami et al., 2000), which disrupt interactions with co-factors, and two are mutations in the homeodomain, which disrupt interactions with the DNA (Ruf et al., 2004; Ito et al., 2006; Sanggaard et al., 2007; Kochhar et al., 2008; Patrick et al., 2009; Noguchi et al., 2011; Patrick et al., 2013). In zebrafish, expression of mutant Six1 mRNA that carries a BOS patient mutation in the SD (R110W) interferes with the Six1/Eya1 interaction that promotes hair cell formation (Bricaud and Collazo, 2011). The Catweasel (Cwe) mouse mutant harbors a missense mutation in the SD (Bosman et al., 2009) that is similar to at least one BOS family (Mosrati et al., 2011). Heterozygous-Cwe mice have an ectopic row of hair cells in the cochlea and homozygous-Cwe mice have a loss of hair cells in the cochlea, semicircular canals and utricle. Mutations in the Six1 co-factor called EYA1 (Ohto et al., 1999) causes BOS1 (OMIM #602588) and the related Branchio-Otic-Renal (BOR1; OMIM #113650) syndrome, in which kidney development is additionally disrupted (Abdelhak et al., 1997; Kumar et al., 1997; Ozaki et al., 2002; Rodriguez-Soriano, 2003; Spruijt et al., 2006). However, mutations in Six1 and Eya1 only account for about half of the BOS and BOR cases, indicating that there are other genes in the Six/Eya genetic pathway that contribute to these syndromes.

To identify potential causative genes for BOS and BOR, we generated a more complete list of genes likely to be regulated by the Six1 transcription factor. We expressed Six1 in animal cap ectodermal explants (ACs), which were raised to an early neural plate stage at which time Six1 is highly expressed in the PPR, and compared their expression profile to control, uninjected ACs that will give rise to non-neural epidermis. Herein we describe the expression patterns of 30 genes that were significantly different between these two data sets, most of which have not been previously characterized in Xenopus. We also identified several chick homologues and present their expression patterns. The majority of the genes characterized are expressed at the correct times to be regulated by Six1, particularly in the otic placode and nephric mesoderm. Thus, they may play a role in the disruption of hearing and kidney functions seen in BOS/BOR patients.

RESULTS AND DISCUSSION

To identify a more complete list of genes likely to act downstream of Six1 during development, animal cap ectoderm (AC) was explanted at blastula stages from embryos microinjected with Six1 mRNA at the 2-cell stage and from control, uninjected sibling embryos. Explants were cultured to neural plate stages when the PPR is specified (Ahrens and Schlosser, 2004; Pieper et al., 2012), and prepared for microarray expression assays as previously described (Yan et al., 2010).

Six1 alters the expression of numerous genes in ectodermal explants

Of the ~14,400 genes represented on the Affymetrix GeneChip v1, 72 were expressed at a >2-fold (p<0.05, ANOVA) higher level in the Six1-injected ACs (Table 1). Gene identities were determined by BLAST (http://blast.ncbi.nlm.nih.gov/Blast.cgi) and GOs were determined by searching the Xenopus and mouse gene databases (http://www.xenbase.org; http://www.informatics.jax.org). Based on gene title or most similar match, the largest category of the up-regulated genes is “unknown” function (44.4%). Genes with known function were most commonly involved in metabolic processes (15.2%), signaling (11.1%), and transcription (9.7%); a small number of other functions also were identified (Figure 1A).

Table 1.

Genes up-regulated >2 fold by Six1 in microarray assays

Affymetrix Probe set ID	Fold Change	Accession Number	Gene Title, Unigene Cluster or EST name	GO
Xl.19919.1.A1_at	4.613	BQ400635	UniGene Xl.940; neuropilin1 (nrp1), aka A5-protein	signaling
Xl.17484.1.A1_at	4.428	BG812762	EST: daf30h01.y1; similar to arrestin beta2 (arrb2)	metabolism
Xl.24006.1.S1_at	4.300	AW640051	UniGene Xl.78601; aldehyde dehydrogenase 7 family, member A1 (aldh7a1)	metabolism
Xl.3228.1.A1_at	3.811	BG022063	EST: dg21d06.x1	unknown
Xl.26192.1.A1_at	3.561	CD324856	EST: AGENCOURT_14164067; most similar to Xenopus laevis clone CH219-20I13	unknown
Xl.20029.2.S1_a_at	3.484	M80798.1	PDGFα receptor	signaling
Xl.21032.1.S1_at	3.413	AF546707.1	Cryptic tubulin	cytoskeleton
Xl.6623.1.S1_at	3.200	BU900288	UniGene Xl.22052; heterogeneous nuclear ribonuclear protein R (hnrnpr)	nuclear
Xl.23944.1.S1_at	3.057	BG885936	UniGene Xl.37859; hypothetical protein LOC100037047	unknown
Xl.12902.1.A1_at	3.046	BJ092409	EST: BJ092409	unknown
Xl.19820.1.A1_at	3.030	BQ399495	EST: NISC_mp03h05.x1; most similar to Xenopus laevis clone CH219-33D5	unknown
Xl.2439.1.A1_at	2.996	BG023545	UniGene Xl.2439; hypoxanthine phosphoribosyltransferase1 (hrpt1)	metabolism
Xl.23880.1.A1_at	2.989	BJ083747	UniGene Xl.4536; RNA binding motif protein 42 (rbm42-b)	RNA binding
Xl.19782.1.A1_at	2.958	BQ398913	UniGene Xl.52003; hypothetical protein LOC 414689	unknown
Xl.12791.1.A1_at	2.944	BJ090916	EST: BJ090916	unknown
Xl.23533.1.S1_at	2.918	BC044278.1	Uncx4.1 homeobox	transcription
Xl.19504.1.S1_at	2.879	BQ386473	UniGene Xl.19504; xyloside xylosyltransferase 1 (xxylt1)	metabolism
Xl.11568.1.S1_a_at	2.859	AW872117	EST: db22e02.y1; matches several genes at 5′ end	unknown
Xl.21833.1.A1_at	2.839	BG345845	EST: dg42a07.y1	unknown
Xl.17880.1.A1_at	2.826	BG812152	EST: daf66c03.x1	unknown
Xl.4077.2.A1_at	2.780	BF024988	EST: dc82d05.x1; similar to mitogen-activated protein kinase binding protein 1	signaling
Xl.15434.2.A1_at	2.689	BJ079756	UniGene Xl.15434; N-acetylglucomsamine-1-phosphate transferase, gamma subunit (gnptg)	metabolism
Xl.17452.3.A1_at	2.665	BG409705	UniGene Xl.54929; cyclin L1 (ccnl1)	transcription
Xl.3765.1.A1_at	2.641	BF428468	UniGene Xl.6193; Polymerase (RNA) II polypeptide I (polr2i)	transcription
Xl.17267.1.S1_at	2.638	BM262144	UniGene Xl.17267; Predicted hamartin-like (aka tuberous sclerosis 1; tsc1)	protein binding
Xl.6714.1.S1_at	2.620	CD360857	UniGene Xl.6714; MACRO domain containing protein 2 (macrod2)	metabolism
Xl.8697.1.A1_at	2.618	AW764878	UniGene Xl.8697; empty spiracles homeobox 2 (emx2)	transcription
Xl.20538.1.S1_at	2.546	CD327089	UniGene Xl.20538; strongly similar to malignant fibrous histiocytoma amplified sequence 1 (mfhas1)	unknown
Xl.21807.1.S1_at	2.526	BC041247.1	Guanine nucleotide binding protein 13 gamma	nuclear
Xl.16468.1.A1_at	2.513	BJ048720	EST: BJ048720; most similar to Xenopus laevis clone CH219-51A4	unknown
Xl.26433.1.A1_at	2.502	BJ077709	EST: BJ077709	unknown
Xl.15912.2.A1_at	2.482	BJ055680	UniGene Xl.18703; phosphatase, orphan 2 (phospho2)	metabolism
Xl.14658.1.A1_at	2.479	BJ052130	EST: BJ052130	unknown
Xl.9485.1.A1_x_at	2.479	BG264616	UniGene Xl.10876; strongly similar to Xenopus tropicalis predicted nucleoporin 133kDA (nup133)	nuclear
Xl.11066.1.A1_at	2.449	AW764707	UniGene Xl.11066; autophagy related 4D, cysteine peptidase (atg4d)	metabolism
Xl.9896.1.A1_at	2.432	BG019991	EST: dc46e08.x1; most similar to Xenopus tropicalis clone CH216-157J16	unknown
Xl.25648.2.A1_at	2.398	BI940543	UniGene Xl.76517; strongly similar to Xenopus tropicalis predicted t-lymphoma invasion and metastasis-inducing protein 1 (tiam1)	signaling
Xl.21726.1.S1_at	2.323	CA790591	EST: AGENCOURT_10301409	unknown
Xl.18330.1.S1_at	2.311	BJ049080	UniGene Xl.64976; moderately similar to Xenopus tropicalis predicted zinc finger protein 335-like	transcription
Xl.4441.1.A1_at	2.305	BF072093	EST: db51d10.x1	unknown
Xl.11394.1.A1_at	2.297	AW639023	UniGene Xl.11394; strongly similar to JNK1/MAPK8-associated membrane protein (jkamp)	metabolism
Xl.535.2.S1_a_at	2.290	AF172400	P75 neurotrophin receptor a-2 (p75NTRa)	signaling
Xl.8063.1. S1_at	2.255	AY099224	Tumorhead (trhd-a)	intracellular transport
Xl.22358.1.A1_at	2.249	BG021407	UniGene Xl.34405; E74-like transcription factor (elf2)	transcription
Xl.14141.1.A1_at	2.192	BJ079748	EST: BJ079748; most similar to ribosomal protein L14 (rpl14)	ribosomal biogenesis
Xl.2873.1.A1_at	2.187	BG018413	UniGene Xl.18253; Uncharacterized LOC 100037177	unknown
Xl.26415.1.S1_at	2.187	CD361045	UniGene Xl.26415; testis, prostate and placenta expressed (tepp)	unknown
Xl.18588.1.S1_at	2.184	BI442945	UniGene Xl.77328; Uncharacterized LOC496297	unknown
Xl.3096.1.A1_at	2.171	BJ085327	UniGene Xl.57373; transmembrane and coiled-coiled domain family 2 (tmcc2)	membrane
Xl.13487.1.A1_at	2.164	CB564242	UniGene Xl.13487; Uncharacterized protein MGC154312	unknown
Xl.11706.1.S1_at	2.163	AY221506.1	Putative polyA-binding protein (PABPN2)	translation
Xl.11463.3.S1_a_at	2.162	BU917037	UniGene Xl.19374; G protein-coupled receptor kinase-interacting ArfGAP (git2)	intracellular transport
Xl.1095.1.S1_at	2.162	S71764.1	Sperm-specific basic nuclear protein 5 (sp5)	signaling
Xl.13738.1.A1_at	2.149	BJ075690	EST: BJ075690; most similar to WNT-inhibitory factor 1 (wif1)	signaling
Xl.16242.1.A1_at	2.145	BJ100481	UniGene Xl.16242; strongly similar to sperm associated antigen 16 protein-like (spag16)	cytoskeleton
Xl.16853.1.S1_at	2.137	BG579925	UniGene Xl.46098; weakly similar to predicted protein FAM178A	unknown
Xl.25532.1.A1_at	2.127	BJ055327	UniGene Xl.25532; strongly similar to hypothetical protein LOC100488559	unknown
Xl.18739.1.S1_at	2.119	BI445731	UniGene Xl.18739; KIAA1456	metabolism
Xl.12870.1.A1_at	2.119	BJ082868	UniGene Xl.12870; moderately similar to zinc finger protein 665-like	unknown
Xl.21416.1.S1_at	2.107	AW638131	UniGene Xl.28718; Pseudouridylate synthase 7 homolog (pus7)	unknown
Xl.12763.1.A1_at	2.104	BJ082426	EST: BJ082426	unknown
Xl.21588.1.A1_at	2.093	AB054535.1	Heparan sulfate 6-O-sulfotransferase (hs6st1)	metabolism
Xl.13831.3.S1_at	2.088	BJ075928	UniGene Xl.72612; moderately similar to predicted coiled-coiled domain-containing C6orf97	unknown
Xl.9563.1.S1_at	2.078	AF480430.1	Pbx1b	transcription
Xl.6583.1.S1_at	2.078	AW635173	UniGene Xl.6583; weakly similar to wars2 gene product	translation
Xl.15729.1.A1_at	2.066	BJ050119	EST: BJ050119	unknown
Xl.401.1.S1_at	2.058	AF032383	Metalloprotease-disintegrin (MDC11b, aka adam11)	proteolysis
Xl.3198.1.A1_at	2.050	BG021827	UniGene Xl.53283; testis expressed 26 (tex26)	unknown
Xl.19719.1.A1_at	2.049	BQ398489	EST: NISC_mo08b08.x1; moderately similar to Xenopus laevis canopy FGF signaling regulator 1 (cnpy1)	signaling
Xl.8108.1.A1_at	2.045	BJ047639	UniGene Xl.47742; Uncharacterized LOC100036828	unknown
Xl.9031.1.A1_x_at	2.019	BG579679	UniGene Xl.9031; Chromosome 5 open reading frame 22 (c5orf22)	unknown
Xl.12209.1.A1_at	2.010	BJ082981	EST: BJ082981; most similar to Xenopus laevis hypothetical protein MGC115205	unknown

Figure 1.

Gene Ontology analysis of the genes whose expression levels in animal cap explants were significantly altered >2 fold by Six1 in a microarray analysis. A. The proportions of genes up-regulated by Six1 in each of several GO functional classes. B. The proportions of genes down-regulated by Six1 in each of several GO functional classes.

Of these 72 genes, only a few have previously been implicated in cranial placode development: Nrp1, PDGFα receptor and p75 neurotrophin receptor. Nrp1-semaphorin signaling regulates neural crest migration, which affects placode contributions to cranial ganglia in Xenopus and mouse (Koestner et al., 2008; Schwarz et al. 2008). PDGF signaling can induce ophthalmic placode formation (McCabe and Bronner-Fraser, 2008) and in combination with FGF plays a role in PPR specification (Kwon et al., 2010). The p75 neurotrophin receptor is expressed on both neural crest and placode cells in developing cranial ganglia (Vazquez et al., 1994). In addition, several of the identified genes have functions that are similar to those published for Six1, supporting the idea that they may be in the same genetic pathway. For example, Six1 is known to regulate cell proliferation and is overexpressed in several cancers (Ford et al., 1998; Coletta et al., 2004; Patrick et al., 2009; Li et al., 2013; Patrick et al., 2013); genes known to be involved in cell cycle regulation (Uncx4.1; Mitogen-activated protein kinase binging protein 1, Cyclin L1; Trhd1-a) and cell movement/cancer (Tiam1, Elf2, Nrp1, p75, Pbx1; MDC11b) were found in the up-regulated category. Placode development in several vertebrates is enhanced by FGF signaling and reduced by Wnt signaling (Phillips et al., 2001; Leger and Brand, 2002; Maroon et al., 2002; Liu et al., 2003; Brugmann et al., 2004; Ahrens and Schlosser, 2005; Litsiou et al., 2005; Martin and Groves, 2005; Matsuo-Takasaki et al., 2005; Bailey et al., 2006; Hong and Saint-Jeannet, 2007; Park and Saint-Jeannet, 2008; Esterberg and Fritz, 2009; Patthey et al., 2008; Patthey et al., 2009; Kwon et al., 2010; Grocott et al., 2012); putative regulators of FGF (Cnpy1, Elf2, Ism1) and Wnt (Arrb2, Wif1) pathways were in the up-regulated category. We also searched those genes up-regulated >1.6 fold (p<0.05, ANOVA) to identify previously characterized genes that might be in the Six1 pathway based on published expression patterns (Table 2). Amongst these were genes expressed during cranial sensory placode (Ism1, Pax6, Irx1, Arnt), somite (Arnt, HoxA3) and/or kidney development (Arnt, Mdx4, Irx1).

Table 2.

Selected genes regulated >1.6 fold by Six1 in microarray assays

Affymetrix Probe set ID	Fold Change	Accession Number	Gene Title	GO	References
Xl.4323.1.A2_at	1.988	BJ078437	Ism1	signaling	Pera et al., 2002
Xl.418.1.S1_at	1.876	NM_001090739.1	Nfya	transcription	Li et al, 1998
Xl.647.7.S1_at	1.825	AF154558.1	Pax6	transcription	Li et al., 1997; Hirsch and Harris, 1997
Xl.69.1.S1_at	1.757	AJ001834.2	Irx1	transcription	Gomez-Skarmeta et al., 1998
Xl.12075.1.S1_at	1.740	AY036894.1	Arnt	transcription	Bollerot et al., 2001
Xl.7579.1.S1_at	1.717	BC077576	TBC1 domain family, member 31 (tbc1d31)	Rab GTPase activator activity
Xl.9439.1.S1_at	1.674	BC041731.1	HoxA3	transcription	Klein et al., 2002
Xl.26468.1.S1_at	1.637	AF127040.1	Mdx4	transcription	Newman and Krieg, 1999
Xl.4910.1.S1_at	1.623	BC126014	Cell division cycle associated 7-like (cdca7l)	cell cycle
Xl.1354.2.S1_at	1.615	MGC83633	PAP associated domain 4 (papd4-a)	mRNA processing	Rouhana et al., 2005
Xl.781.1.S1_at	−1.852	AY029294.1	otx1	transcription	Kablar et al., 1996
Xl.13309.1.S1_at	−1.832	X53450	snail1	transcription	Sargent and Bennett, 1990
Xl.20089.1.S1_at	−1.715	BC042303.1	foxi1 (xema)	transcription	Suri et al., 2005
Xl.283.1.A1_at	−1.642	X62053.1	hoxA1	transcription	Sive and Cheng, 1991

To assess the validity of the microarray results, we performed several experiments. First, we measured the expression of ten up-regulated genes by qPCR; 10/10 were expressed at higher levels in Six1-injected ACs compared to uninjected control ACs, half of which reached statistical significance (Table 3). Second, we knocked-down endogenous Six1 in whole embryos by injecting morpholino antisense oligonucleotides; this diminished the expression of 17/20 up-regulated candidates (Table 4). Figure 2 shows examples: loss of somite expression of Cdca7l; loss of neural crest expression of HoxA3; loss of neural tube expression of Ism1. Third, we increased Six1 expression: 12/20 up-regulated candidates showed expanded expression domains by injection of mRNAs encoding either wild-type Six1 or the activating Six1-VP16 construct, and 0/20 was expanded by the repressive EnR-Six1 construct (Table 4). As examples, Figure 2 shows examples: broader neural plate and/or neural crest expression by wild-type Six1 (BG022063, Nfya) or Six1-VP16 (Frizzled 10b, Pbx1); reduced Frizzled 10b by EnR-Six1. Although nearly all of the tested candidates require Six1, it was surprising that only about half of them were increased/expanded by wild-type or activating Six1. However, these results are consistent with SO activity in the fly eye field: overexpression rarely induces ectopic eye tissue (Jusiak et al., 2014) and it promotes eye fate mainly by repressing alternate fates (Anderson et al., 2010). Thus, many of the up-regulated genes identified by the microarray assay are likely downstream of Six1 based on published function in Xenopus, chick or mouse, published expression patterns and/or changes in expression after Six1 loss- or gain-of-function. However, Six1 ChIP analyses are required to confirm if any of these candidates are direct transcriptional targets.

Table 3.

qPCR analysis of expression levels in Six1-injected animal cap explants relative to control explants

	Fold expression over control	p value
Up-regulated
BJ02409	5.14	<0.05
BQ399495	3.49	<0.05
AF546707	3.37	<0.05
Pdgfra	2.34	<0.05
BG885936	1.85	>0.1
Hrpt1-5047264	1.83	<0.05
Aldh7a1	1.72	>0.1
Hnrnpr	1.71	>0.1
Arrb2-4740312	1.35	>0.1
XXylt1-6639045	1.35	>0.1
Down-regulated
MGC 114680	0.40	>0.1
Image 3399268	0.55	<0.05
Dnaja4.2-4930076	0.79	>0.1
Ralgds-4681426	1.08	>0.1
Cnfn1-a-6316571	1.25	>0.1

Table 4.

Percentage of embryos showing altered expression domains of up-regulated genes in response to changes in Six1 activity

Candidate	Diminished by Six1 MO knock-down (n)	Expanded by wild-type Six1 (n)	Expanded by Activating Six1-VP16 (n)	Expanded by Repressive EnR-Six1 (n)
Nrp1/BQ400635	41.2 (17)	4.8 (42)	27.8 (18)	0.0 (28)
Arrb2/BG812762	0.0 (20)	0.0 (52)	0.0 (13)	0.0 (24)
BG022063	0.0 (20)	5.9 (34)	0.0 (8)	0.0 (10)
LOC100037047	43.8 (16)	15.6 (45)	0.0 (10)	0.0 (20)
BQ399495	0.0 (16)	7.1 (28)	0.0 (7)	0.0 (20)
Hrpt1/BG023545	8.3 (24)	2.1 (49)	0.0 (12)	0.0 (16)
Rmb42-b/BJ083747	45.0 (20)	0.0 (47)	0.0 (14)	0.0 (16)
LOC414689	31.6 (19)	0.0 (45)	0.0 (12)	0.0 (12)
Uncx4.1/BC044278.1	28.6 (14)	0.0 (24)	0.0 (11)	0.0 (10)
Xxylt1/BQ386473	28.6 (14)	0.0 (40)	11.1 (18)	0.0 (14)
Pbx1b/AF480430.1	64.3 (14)	0.0 (41)	85.0 (20)	0.0 (14)
Ism1/BJ078437	50.0 (12)	0.0 (31)	0.0 (15)	0.0 (15)
Frizzled10b	57.1 (14)	0.0 (23)	72.2 (18)	0.0 (10)
Nfya/NM_0-01090739.1	75.0 (16)	0.0 (39)	47.1 (17)	0.0 (10)
Arnt/AY036894.1	54.5 (11)	0.0 (41)	0.0 (14)	0.0 (9)
Tbc1d31/BC077576	26.7 (15)	0.0 (53)	10.0 (10)	0.0 (11)
HoxA3/BC041731.1	71.4 (14)	0.0 (33)	53.8 (13)	0.0 (7)
Mdx4	38.5 (13)	0.0 (30)	7.7 (13)	0.0 (9)
Cdca7l/BC126014	66.7 (12)	0.0 (46)	0.0 (13)	0.0 (11)
Papd4/MGC83633	66.7 (12)	0.0 (21)	0.0 (14)	0.0 (8)

Figure 2.

Examples of changes in gene expression of putative Six1 targets after knock-down of endogenous Six1 levels by injection of antisense morpholino oligonucleotides (Six1 MOs), increased levels of wild type Six1 by mRNA injection (Six1 WT), activation of Six1 targets by mRNA injection of an activating construct (Six1-VP16), or repression of Six1 targets by mRNA injection of a repressive construct (EnR-Six1) (see Brugmann et al., 2004 for validation of these constructs). Six1 MOs: Expression of Cdca7l in the segmenting somites (arrows on control [ctrl] side) is lost on the injected (inj) side (side views, anterior to left). Expression of HoxA3 in the neural crest (nc) is greatly reduced on the injected side, whereas hindbrain (hb) expression is maintained (side views, anterior to left). Expression of Ism1 is lost in the midbrain (arrow) (frontal view, dorsal to the top). Expression of IMAGE 3399268 is expanded in the neural plate (line depicts width from midline to lateral border) on the injected side (dorsal view, anterior to the bottom). Expression patterns of Cnfn1-a in the hatching gland (arrows) and of Snail1 in the migrating neural crest are reduced and disrupted on the injected sides (frontal views, dorsal to the top). Six1 WT: Expression of BG022063 is broader in the neural plate (line) and neural crest (bracket) on the injected side (frontal view, dorsal to the top). Expression of Nfya is broader in the neural tube (line) on the injected side (dorsal view, anterior to the top). Expression of IMAGE 3399268 in the segmented somites (arrows, ctrl side) is lost on the injected side (side views, anterior to left). Expression of IMAGE 4057931 in the neural crest (bracket), otocyst (oto) and trigeminal ganglion (arrow) is greatly diminished on the injected side (side views, anterior to left). Six1-VP16: Expression of Frizzled10b, Pbx1 and Zic2 in the neural plate is expanded (line) on the injected side (frontal views, dorsal to the top). EnR-Six1: Expression of Frizzled10b in the neural plate is reduced (line) on the injected side (frontal view, dorsal to the top). Expression of IMAGE 4057931 is reduced in the PPR (arrows on the control side; frontal view, dorsal to the top). Expression of Zic2 and Otx1 in the neural tube is reduced (line) on the injected sides (arrows; frontal views, dorsal to the top). In MO-injected embryos, injected sides were confirmed by the use of lysamine-labeled MOs (not shown). In mRNA-injected embryos, injected sides were confirmed by the presence of nuclear β-galactosidase lineage labeling (pink dots visible in most images).

Of the genes represented on the GeneChip, 58 were expressed at a >2-fold (p<0.05, ANOVA) lower level in the Six1-expressing ACs (Table 5). The largest category of the down-regulated genes is “unknown” function (36.2%). Genes with known function were most commonly involved in metabolic processes (10.3%), signaling (19.0%), and transcription (10.3%); a small number of other functions also were identified (Figure 1B). The expression of most of the 58 genes have not been characterized across Xenopus developmental stages, but by searching a mouse expression database (http://www.informatics.jax.org) we found that many of those for which there is an identified homologue are expressed in developing tissues regulated by Six1. These include: cranial placodes (Glcci1, Rcan1, Naga, Rilpl1, Etl4, Xspr2, Jun, Phb2, Mt1, Pigc, Gap43, Tusc3), neural crest (Zic2), muscle (Rcan1, Jun, Krm2, Phb2, Rras2, Pigc, Sytl1), and nephric mesoderm (Naga, Pik3r5, Cdc42ep2, Xspr2, Rab40B, Akt, Rras2, Pigc, Vwc2, Gap43, HNF1a, Sytl1, Tusc3). Several also are transcribed in tissues in which Six1 is not normally expressed. These include: neural plate/tube (Socs3, Glcci1, Dnaj4.2, Rhot1, Pik3r5, Axin2, Xspr2, Jun, Rab40b, Akt, Krm2, Metrn, Phb2, Akap2, Ap4m1, Rras2, Mt1, Zic2; Pigc, Vwc2, Gap43, HNF1a, Sytl1, Vta1, Tusc3), epidermis (Fgk, Cnfc1, Axin2, Krm2, Phb2, Krt18-a, Pigc, Tusc3) and mesoderm (Socs3, Pik3r5, Adam13, Xspr2; Jun, Krm2, Rras2, Mt1, Pigc, HNF1a). We also searched genes down-regulated >1.6 fold (p<0.05, ANOVA) to identify previously characterized genes expected to be repressed by Six1, based on studies in chick and Xenopus (Brugmann et al., 2004; Christophorou et al., 2009). Of note are genes expressed in neural plate (Otx1), neural crest (Snail1, HoxA1) and epidermis (Foxi1/Xema) (Table 2).

Table 5.

Genes down-regulated >2 fold by Six1 in microarray assays

Affymetrix Probe set ID	Fold Change	Accession Number	Gene Title, Unigene Cluster or EST name	GO
Xl.22313.1.S1_at	−3.876	BC080322	Xenopus laevis cDNA clone IMAGE 4174145	unknown
Xl.26171.1.A1_s_at	−3.610	BC079829	Xenopus laevis cDNA clone IMAGE 3399268	unknown
Xl.14963.1.A1_x_at	−3.378	BG264797	EST: daa32g06.x1; IMAGE 4057931; most similar to 5′biphosphate nucleotidase 1 (bpnt1)	metabolism
Xl.9924.1.A1_at	−3.322	BG020210	EST: dc48g12.y1; IMAGE 3400511	unknown
Xl.2213.1.A1_at	−3.311	BG022871	UniGene Xl.2213; suppressor of cytokine signaling 3 (Socs3)	signaling
Xl.4337.1.A1_at	−3.215	BC097548	Xenopus laevis hypothetical protein MGC114680	unknown
Xl.15435.1.A1_at	−3.185	BJ081509	UniGene Xl.79199; Fin and gill-specific type-1 keratin (fgk)	cytoskeleton
Xl.16561.1.A1_at	−3.135	CF521493	EST: AGENCOURT_15529711; IMAGE 7010138	unknown
Xl.16590.1.S1_at	−3.115	BJ049628	UniGene Xl.63625; Glucocorticoid induced transcript 1 (glcci1)	unknown
Xl.23897.1.S1_at	−2.994	BC068675	Cornifelin gene1 (Cnfn1-a)	unknown
Xl.7794.1.A1_at	−2.833	BQ897357	EST: AGENCOURT_8670989; most similar to Xenopus tropicalis ral guanine nucleotide dissociation stimulator (ralgds)	signaling
Xl.2209.1.S1_at	−2.695	CB562491	UniGene Xl.55075, complement component 3 (c3)	signaling
Xl.1064.1.S1_at	−2.695	U63818.1	Ring finger protein	transcription
Xl.13604.1.A1_at	−2.646	BJ088530	EST: BJ088530; most similar to Xenopus laevis IMAGE clone 4965486	unknown
Xl.23980.1.A1_at	−2.618	AW764542	EST: da88e01.x1	unknown
Xl.2443.1.S1_at	−2.604	BC046660.1	DnaJ (Hsp40) homolog, subfamily A, member 1 (Dnaja4.2)	protein folding
Xl.9975.1.A1_at	−2.597	BJ088036	EST: BJ088036; most similar to ras homolog family member T1 (rhot1)	signaling
Xl.23110.1.A1_at	−2.584	BJ076210	UniGene Xl.83926; Regulator of calcineurin 1 (rcan1)	signaling
Xl.22434.1.A1_at	−2.525	CB560725	UniGene Xl.22434; N-acetylgalactosaminidase alpha (naga-a)	metabolism
Xl.2361.1.A1_at	−2.475	BQ399921	UniGene Xl.2361; phosphoinositide-3-kinase, regulatory subunit 5 (pik3r5)	metabolism
Xl.16796.1.S1_at	−2.463	BJ085230	EST: BJ085230; most similar to Xenopus tropicalis Rab interacting lysosomal protein-like 1 (rilpl1)	protein transport
Xl.166.1.S1_at	−2.439	U66003.1	Adam13	proteolysis
Xl.15263.1.A1_at	−2.421	BQ400687	UniGene Xl.63771; predicted sickle tail protein like (etl4)	unknown
Xl.16177.1.A1_at	−2.421	BJ055605	UniGene Xl.16177; Axin 2 (Axis inhibition protein 2)	signaling
Xl.16447.1.A1_at	−2.387	BE505669	UniGene Xl.16447; strongly similar to mouse CDC42 small effector protein 2 (cdc42ep2)	cytoskeleton
Xl.15778.1.A1_at	−2.370	BJ056356	EST: BJ056356	unknown
Xl.2755.1.S1_at	−2.336	AY062263.1	Sp1-like zinc-finger protein (XSPR-2)	transcription
Xl.541.1.S1_s_at	−2.315	BC041183.1	jun oncogene	transcription
Xl.11153.1.A1_at	−2.288	BJ057396	EST: BJ057396; most similar to Xenopus tropicalis clone ISB1-168-L6	unknown
Xl.23887.1.A1_at	−2.283	BJ090458	UniGene Xl.23887; Rab40B	signaling
Xl.738.1.S1_at	−2.268	AF317656.1	Akt	signaling
Xl.15014.1.A1_at	−2.262	BJ048100	EST: BJ048100	unknown
Xl.15701.1.S1_at	−2.247	AY150813.1	Kremen2 (krm2)	signaling
Xl.25379.1.A1_at	−2.242	BM262157	UniGene Xl.25379; weakly similar to meteorin-like protein-like (metrn)	unknown
Xl.15834.1.S1_at	−2.222	BJ050466	UniGene Xl.63885; Xenopus laevis MGC84728 protein, most similar to similar to mouse/human prohibitin 2 (phb2)	transcription
Xl.13799.1.A1_at	−2.217	BJ085895	EST: BJ085895	unknown
Xl.16517.1.A1_at	−2.217	AW199946	UniGene Xl.16517; zinc finger protein 830	cell cycle
Xl.8910.1.S1_at	−2.212	BC042269.1	Keratin type I cytoskeletal 18-A (krt18-a)	cytoskeleton
Xl.13511.1.A1_at	−2.203	BJ085865	EST: BJ085865	unknown
Xl.2612.1.A1_at	−2.188	BJ097651	UniGene Xl.2612; A-kinase anchor protein 2 (akap2)	cytoskeleton
Xl.13965.1.A1_at	−2.179	BJ081606	UniGene Xl.4228; Adaptor-related protein complex 4, mu 1 subunit (ap4m1)	protein transport
Xl.11640.1.A1_at	−2.169	BJ075244	UniGene Xl.78983; strongly similar to related RAS viral (rras2) oncogene homolog	signaling
Xl.2784.1.S1_at	−2.159	M96729.1	Metallothionein (mti)	metabolism
Xl.17176.1.A1_at	−2.141	BM261509	EST: dai45d06.x1	unknown
Xl.1085.1.S1_at	−2.132	U57453.1	Zic2	transcription
Xl.11033.1.A1_at	−2.114	BM191782	UniGene Xl.75442; phosphatidylinositol glycan anchor biosynthesis class C (pigc)	metabolism
Xl.4175.2.A1_at	−2.101	BJ081456	EST: BJ081456	unknown
Xl.15853.1.A1_at	−2.062	BJ089315	UniGene Xl.15853; strongly similar to Xenopus tropicalis von Willebrand factor C domain-containing protein 2-like, gene 2 precursor (vwc2l)	signaling
Xl.15479.1.A1_at	−2.062	CD328587	UniGene Xl.15479; Xenopus laevis uncharacterized protein MGC116516; weakly similar to mouse/human mitogen-activated protein kinase 12 (mpk12)	cell cycle
Xl.25273.1.A1_x_at	−2.058	AW782822	UniGene Xl.25885; weakly similar to Xenopus laevis zinc finger protein 33A	unknown
Xl.3057.2.S1_at	−2.053	BU901475	UniGene Xl.3057; Xenopus laevis uncharacterized protein LOC100158313; weakly similar to human ZYX gene product	unknown
Xl.9994.1.S1_s_at	−2.028	BC043990.1	Hypothetical protein MGC64520; strongly similar to Xenopus laevis growth associated protein 43 (gap43-b)	cell adhesion
Xl.24413.1.S1_at	−2.020	BQ397907	UniGene Xl.24413; moderately similar to Xenopus tropicalis acid sphingomyelinase-like phosphodiesterase 3a precursor	metabolism
Xl.428.1.S1_at	−2.012	X64762.1	HNF1 homeobox A (hnf1a-a)	transcription
Xl.13762.1.A1_at	−2.012	BJ085361	EST: BJ085361	unknown
Xl.15997.1.A1_at	−2.008	BJ043804	UniGene Xl.15997; synaptotagmin-like 1 (sytl1)	protein transport
Xl.22640.1.A1_at	−2.008	BJ086587	UniGene Xl.22640; Vps20-associated 1 homolog (vta1)	protein transport
Xl.19274.1.S1_at	−2.004	BI442114	UniGene Xl.19274; Tumor suppressor candidate 3 (tusc3)	ion transport

To assess the validity of the microarray results, we performed several experiments. First, we measured the expression of five down-regulated genes by qPCR; 3/5 were expressed at lower levels in Six1-injected ACs compared to uninjected control ACs, one of which reached statistical significance (Table 3). Second, we knocked-down endogenous Six1 in whole embryos by injecting morpholino antisense oligonucleotides; this altered the expression patterns of 10/15 down-regulated candidates (Table 6). The expression domains of two were expanded (e.g., IMAGE 3399268), whereas the remainders were reduced (e.g., Cnfn1-a, Snail1; Figure 2). Third, we increased Six1 expression: 15/15 down-regulated candidates had diminished expression domains by injection of mRNAs encoding either wild-type Six1 or the repressive EnR-Six1 construct, and only 4/15 by the activating Six1-VP16 construct (Table 6). Figure 2 shows examples: reduced somite (IMAGE 3399268) and neural crest/placode (IMAGE 4057931) expression by wild-type Six1; reduced PPR (IMAGE 4057931) and neural tube (Zic2, Otx1) expression by EnR-Six1. Interestingly, Zic2 neural tube expression was expanded by Six1-VP16 (87.5%, n=17; Figure 2). Thus, many of the down-regulated genes identified by the microarray assay are likely downstream of Six1 based on published function in Xenopus, chick or mouse, published expression patterns and/or changes in expression after Six1 loss- and gain-of-function. However, Six1 ChIP analyses are required to confirm if any of these candidates are direct transcriptional targets.

Table 6.

Percentage of embryos showing altered expression domains of down-regulated genes in response to changes in Six1 activity

Candidate	Diminished (↓) or expanded (↑) by Six1 MO knock-down (n)	Diminished by wild-type Six1 (n)	Diminished by Repressive EnR-Six1 (n)	Diminished by Activating Six1-VP16 (n)
IMAGE 4174145	0.0 (12)	66.7 (21)	92.9 (14)	0.0 (16)
IMAGE 3399268	70.0 (10) ↑	82.6 (46)	91.7 (12)	0.0 (14)
IMAGE 4057931	20.0 (10) ↓	73.0 (37)	100.0 (13)	0.0 (16)
IMAGE 3400511	70.0 (10) ↓	59.4 (32)	88.2 (17)	0.0 (10)
Socs3/BG022871	0.0 (15)	58.5 (41)	100.0 (15)	0.0 (5)
MGC114680	0.0 (14)	37.8 (37)	100.0 (12)	17.6 (17)
IMAGE 7010138	0.0 (11)	40.0 (45)	87.5 (16)	0.0 (21)
Cnfn.1a/BC068675	75.6 (14) ↓	50.0 (26)	83.3 (12)	0.0 (17)
Ralgds/BQ897357	42.9 (14) ↓	50.0 (16)	88.9 (9)	14.3 (14)
Dnaja4.2/BC046660.1	50.0 (14) ↓	45.0 (20)	86.7 (15)	5.9 (17)
Otx1	11.8 (17) ↓	73.3 (15)	77.8 (9)	5.9 (17)
Snail1	78.6 (14)↓	84.2 (19)	83.3 (15)	0.0 (10)
Foxi1 (Xema)	71.4 (14) ↓	61.1 (18)	90.9 (11)	0.0 (16)
HoxA1	0.0 (14)	nd	80.0 (15)	0.0 (16)
Zic2	42.9% (14) ↑	nd	80.0 (5)	0.0 (17)

Legend: nd, not done

Expression patterns of uncharacterized genes

The data above suggest that many of the genes whose expression patterns are altered by Six1 are likely to be involved in Six1-related developmental processes. Since a large number of these genes are of unknown function (Figure 1) and/or their developmental expression patterns have not been reported in Xenopus, we randomly picked 10 genes of unknown function up-regulated >2-fold (highlighted in Table 1) and 10 genes of unknown function down-regulated >2-fold (highlighted in Table 5) and performed expression analyses. We also chose a few uncharacterized >1.6 fold altered genes to analyze, and provide more detail for a few characterized genes (shaded in Tables 1, 2 and 5).

RT-PCR

To demonstrate expression levels over time, RT-PCR was performed for 12 of the up-regulated genes (Figure 3A). Nrp1 is strongly expressed maternally, is reduced at blastula stages, strongly expressed during neural ectoderm/neural plate stages, and is reduced at tail bud stages. Arrb2 is expressed at very low levels until late tail bud stages. BG022063 and LOC10003704 are expressed at low levels prior to gastrulation, are highly expressed from neural plate through tail bud stages, after which expression decreases. BQ399495 is strongly expressed from cleavage through gastrulation stages, followed by peaks at early and late tail bud stages. Hrpt1 is highly expressed at cleavage and blastula stags, followed by reduced expression through the remaining stages. LOC414689 is weakly detected at all stages, but more highly at neural tube through tail bud stages. Uncx4.1 is strongly expressed nearly uniformly across all stages. Xxylt1 is highly expressed at cleavage and blastula stages, decreasing at neural plate stages, and peaking again at late tail bud stages. Ism1 is expressed nearly uniformly across all developmental stages, Arnt is strongly expressed from early neural plate to late tail bud stages and Hox3A is highly expressed maternally, then uniformly at lower levels from gastrula through early tail bud, then at increased levels at later tail bud stages.

Figure 3.

Semi-quantitative RT-PCR assays illustrate the temporal expression patterns of previously uncharacterized up-regulated (A) and down-regulated (B) genes in normal, unmanipulated Xenopus embryos. RT-, minus RT step; numbers across the bottom indicate the developmental stages. H4 is an internal control. Im, IMAGE

RT-PCR was performed for 10 of the down-regulated genes (Figure 3B). IMAGE 4174145 is weakly expressed at all stages. IMAGE 3399268 is moderately expressed at all stages. IMAGE 4057931 is detected at early neural plate and early tail bud stages. IMAGE 3400511 is weakly expressed until early neural plate and late tail bud stages. Socs3 is weakly expressed from gastrulation through tail bud stages. MGC114680 and Ralgds are expressed at uniform levels from cleavage through tail bud stages, with less expression at later stages. IMAGE 7010138 is highly expressed at gastrulation and expressed at low levels after that (but see tissue expression patterns below). Cnfn1-a is highly expressed at cleavage to gastrulation stages, and again at late neural plate through early tail bud stages. Dnaja4.2 is strongly expressed at cleavage stages and then moderately expressed from gastrula though tail bud stages.

These temporal expression profiles partially overlap with that of Xenopus Six1. Six1 expression is not detected by RT-PCR or in situ hybridization at cleavage or blastula stages (Pandur and Moody, 2000). This indicates that during these early stages Six1 does not regulate the expression of the up-regulated or the down-regulated genes. However, Six1 expression is detected by RT-PCR at gastrula (st 10.5), neural plate (st 12), neural tube (st 19), tail bud (st 26/27) and larval (st 37/38; st 40) stages when placodes, somites, and kidney anlage form (Pandur and Moody, 2000). This partial overlap with the RT-PCR data for each of the tested genes (Figure 3) is consistent with the possibility that Six1 may regulate these genes after the onset of gastrulation.

In situ expression of up-regulated genes

To demonstrate spatial expression patterns, whole embryo in situ hybridization (ISH) analyses were performed for 14 up-regulated genes for which no ISH data are available, and for 5 up-regulated genes for which the data are incomplete (Nrp1, Pbx1b, Ism1, Arnt, HoxA3). At the early developmental stages, all 19 genes share very similar expression patterns (Figures 4–6, Table 7). All are expressed maternally, with enhanced localization to the animal hemisphere at cleavage stages (Figure 4). To be assured that this was not due to lack of probe penetration into the yolky vegetal cells, we also bisected embryos prior to ISH; enhanced animal hemisphere expression and lack of vegetal expression was consistently observed (Figure 7). At blastula stages, expression remained in the animal half of the embryo (Figure 4), in some cases extending into the equatorial region (Table 7; Figure 8). At gastrula stages, all genes but Ism1 are diffusely expressed throughout the ectoderm germ layer; as previously reported (Pera et al., 2002), Ism1 instead is expressed in the ventrolateral portion of the blastopore lip of the gastrula (Table 7). Some genes had enhanced expression in the animal pole ectoderm (Nrp1, Arrb2, LOC100037047, Uncx4.1); sectioning the embryos showed that others also had weak expression in the involuting mesoderm (Hprt1, Xxylt1, Trhd, Pbx1b, Nfya, Arnt, Tbc1d31, HoxA3, Cdca7l, Papd4) (Figures 4, 8; Table 7). To be assured that the diffuse staining at these early stages was not the result of non-specific background, embryos also were processed with sense directed RNA probes; very little background was detected (Figure 9).

Figure 4.

Whole mount in situ hybridization assays at early developmental stages illustrating the common expression patterns of 14 up-regulated genes. A: Nrp1, Arrb2, BG022063, LOC100037047, BQ399495, Hrpt1, Rbm42-6. B: LOC414689, Uncx4.1, Xxylt1, Nfya, Tbc1d31, Cdca7l, Papd4-a. For all genes, expression is enhanced in the animal cells (an) of cleavage and blastula embryos (side views). At gastrula stages, expression is throughout the ectoderm, occasionally enhanced in the animal cells (e.g., Arrb2) or the dorsal cells (e.g., Nrp1) (side views). At early neural plate stages, all genes are expressed in the epidermis (epi), often more intensely on the dorsal (d) side, including neural plate (np) and border zone (bz) (all except Nrp1 are side views with anterior [a] to the left; Nrp1 is an anterior view with dorsal to the top). At late neural plate and neural tube stages, expression is strongest in the neural crest (nc), preplacodal region (ppr), neural tube (nt) and occasionally cement gland (cem) (all are anterior views). For Nrp1, embryos were sectioned in the horizontal plane to demonstrate that the ppr stripe in the whole embryo is due to expression in the deep (d) layer of the ectoderm, not the superficial (s) layer of the ectoderm or the pharyngeal endoderm (phe); ph, lumen of pharynx.

Figure 6.

Whole mount in situ hybridization assays illustrating the expression patterns of previously characterized genes up-regulated (Trhd-a, Pbx1b, Ism1, Arnt, HoxA3) or down-regulated (Foxi1-Xema) genes at stages previously unpublished. Abbreviations are as in Figures 4 and 5.

Table 7.

Developmental expression patterns of up-regulated genes

Gene title/accession # OB Clone #	Cleavage	Blastula st 8–9	Gastrula st 10–11.5	Neural Plate E: st 12–13 L: st 14–17	Neural tube closure st 18–22	Tail bud st 23–31	Larva st 32–38
Nrp1/BQ400635 4969039	animal	AC	animal ecto, dorsal ecto	E: dorsal ecto L: NP, BZ, epi, stripe in midbrain, noto	1. spcd, NC, PPR	1. spcd, NC, PPR, dorsal gut; Hb, ret, olf, lens, oto, dorsal epi, epibranchial placodes, cem, BA, nephric meso	same, plus pineal, spcd, IX/Xg, BA-distal patches, heart
Arrb2/BG812762 4740312	animal	AC	animal ecto	E: dorsal ecto L: NP, BZ, epi, cem, dorsal meso	NT, NC, dorsal epi, cem, somites	Hb, ret, oto, BA, somites	same, plus dorsal spcd, lens, head epi, nephric meso
BG022063 3749434	animal	AC	ecto	E: dorsal ecto L: NP, BZ, epi, dorsal meso	NT, NC, olf, lens, oto, somites	brain, ret, spcd, olf, lens, oto, BA, somites	same, plus nephric meso
LOC100037047/BG885936 3437507	animal	AC	animal ecto	E: dorsal ecto L: NP, BZ, epi dorsal meso	NT, NC, head epi, cem, somites	brain, ret, spcd, olf, lens, oto, cem, BA, somites	same, plus nephric meso
BQ399495 4965752	animal	AC	ecto	E: dorsal ecto L: NP, BZ, epi dorsal meso	NT, NC, epi, somites	brain, ret, spcd, oto, BA, somites	same, plus lens, nephric meso, heart
Hprt1/BG023545 5047264	animal	AC, eq	ecto, weak meso	E: dorsal ecto L: NP, BZ, epi dorsal meso	NT, NC, PPR, dorsal epi, noto, somites	brain, ret, stripe in Mb, lens, oto, head epi, BA	same, plus nephric meso
Rbm42-b/BJ083747 6635289	animal	AC	ecto	E: dorsal ecto, epi L: NP, BZ, epi	NT, NC, PPR, dorsal epi	brain, ret, spcd, olf, lens, oto, head epi, BA, somites	same, plus cranial ganglia, nephric meso
LOC414689/BQ398913 6636835	animal	AC	ecto	E: dorsal ecto, epi L: NP, BZ, epi, dorsal meso	NT, NC, PPR, head epi, cem, somites	brain, ret, spcd, olf, lens, oto, head epi, cem, BA, somites	same, plus nephric meso, heart
Uncx4.1/BC044278.1 5543082	animal, eq	AC, eq	animal ecto	E: dorsal ecto, epi L: NP, BZ, epi dorsal meso	NT-lat stripes in Mb, Hb and spcd, NC, olf, head epi, somites, nephric meso	brain, ret, spcd, olf, oto, head epi, BA, somites, nephric meso	same, plus lens
Xxylt1/BQ386473 6639045	animal	AC, eq	ecto, weak meso	E: dorsal ecto, epi L: NP, BZ, epi dorsal meso	NT, NC, PPR, somites	brain, ret, spcd, lens, oto, epi, BA, somites	same, plus nephric meso, heart
Trhd-a/AY099224	2. present*	AC, eq	ecto, weak meso	E: dorsal ecto, epi L: NP, BZ, epi dorsal meso	NT, NC, PPR, epi, somites	brain, ret, spcd, oto, epi, BA, somites	same, plus nephric meso, pharyngeal pouches
Pbx1b/AF480430.1	animal	AC, eq	ecto, meso	3. NP, BZ dorsal meso	3. brain, ret, ant spcd, BA somites	3. same, plus nephric meso, lateral plate meso olf, lens, oto, somites
Ism1/BJ078437 4888159	4. animal	AC	4. ventro-lateral blastopore lip	4. E: same, plus noto 4. L: ant NP, BZ pharyngeal endoderm	4. Mb, post spcd, NC, Dlp, paraxial meso	4: Di, Mb, oto, BA, tailbud, ventral blood islands	same, plus somites
Nfya/NM_0-01090739.1 6631251	5. present* animal	AC	ecto, weak meso	E: dorsal ecto, epi L: NP, BZ, epi dorsal meso	NT, NC, olf, lens, oto, somites	brain, ret, ant spcd, olf, lens, oto, BA, somites, nephric meso	same
Arnt/AY036894.1 4203590	6. animal	6. AC	6. animal ecto, meso	6. NP, BZ, epi, dorsal meso	NT, NC, PPR, epi, somites	6. brain, ret, spcd, lens, oto, BA, somite, nephric meso olf, cranial ganglia	9. same
Tbc1d31/BC077576 5079674	animal	AC	ecto, weak meso	E: dorsal ecto, epi L: NP, BZ, epi dorsal meso	NT, NC, PPR, somites	brain, ret, spcd, olf, lens, oto, diffuse epi, BA, noto, somites	same, plus nephric meso
HoxA3/BC041731.1 4683538	animal, weak	AC	ecto, weak meso	E: dorsal ecto, epi L: NP, BZ, epi dorsal meso	NT, NC, PPR, dorsal epi, somites	7. spcd, caudal BAs, somites Strong stripe in Hb	same
Cdca7l/BC126014 6864704	animal	AC	ecto, weak meso	E: dorsal ecto, epi L: NP, BZ, epi dorsal meso	NT, NC, PPR, dorsal epi, somites	brain, ret, epi, somites	same, plus scattered epi cells, BA, nephric meso
Papd4/MGC83633 5084876	8. present* animal	AC	ecto, meso	E: dorsal ecto, epi L: NP, BZ, epi dorsal meso	NT, NC, PPR, epi, somites	brain, ret, spcd, lens, oto, epi, BA, somites	same, plus nephric meso

ISH expression patterns at different developmental stages for 14 genes for which there are no ISH expression data. ISH patterns also were completed for 5 genes for which there are available descriptions. For 3 genes (Trhd-a, Nfya, Papd4) presence of expression (*) was detected only by Northern blot, RNase protection, RT-PCR or RNA-Seq.

Abbreviations: AC, animal cap; BA, branchial arches; BZ, border zone; cem, cement gland; dlp, dorso-lateral placode; E, early; ecto, ectoderm; epi, epidermis; eq, equatorial; Hb, hindbrain; IX/Xg, glossopharyngeal/vagal ganglion; L, late; Mb, midbrain; meso, mesoderm; NC, neural crest; NP, neural plate; NT, neural tube; OB, Open Biosystems; olf, olfactory placode; oto, otocyst; PPR, pre-placodal region; ret, retina; spcd, spinal cord; st, stage; Vg, trigeminal ganglion; VIIg, facial ganglion.

Abbreviations in bold refer to previous published expression patterns; bold numbers indicate references for previously described expression data:

1. Koestner et al., (2008)
2. Wu et al. (2001)
3. Maeda et al. (2002)
4. Pera et al. (2002)
5. Li et al. (1998)
6. Bollerot et al. (2001)
7. Xenbase: http://www.xenbase.org/gene/expression.do?method=displayGenePageExpression&geneId=482266&tabId=1
8. Rouhana et al. (2005)

Figure 7.

Four examples of cleavage stage gene expression in embryos that were bisected prior to the in situ hybridization protocol to facilitate probe access to all of the cells. In each case the probe was detected in the animal pole blastomeres (a) and in marginal zone animal blastomeres (arrows), but not in vegetal blastomeres (v).

Figure 8.

Examples of embryos bisected after the in situ hybridization protocol to reveal expression patterns in internal cells. BQ399495: At blastula (left image, sagittal section), expression is primarily in the animal cap (AC) cells, and barely detected in the equatorial (eq) zone. At neural plate stages (right image, transverse section), expression is observed in the neural plate (np), somites (so) and epidermis (epi). Hprt1: At blastula (left image, sagittal section), expression is in both animal cap and equatorial cells. At gastrula stages (right image, sagittal section, anterior to the left), expression is observed in the surface ectoderm (ecto) and involuting mesoderm (meso); bl, dorsal blastoporal lip. Xxylt1: At blastula (left image, sagittal section), expression is in both animal cap and equatorial cells. At neural plate stages (right image, sagittal section, anterior to the left), expression is observed in the neural plate and underlying dorsal mesoderm (meso). IMAGE 3399268: At blastula (left image, sagittal section), expression is in both animal cap and equatorial cells. At gastrula stages (right image, sagittal section, anterior to the left), expression is observed in the animal ectoderm, but barely detected in the involuting mesoderm. Cnfn1-a: At gastrula stages (sagittal section, anterior to the left), expression is observed in the animal ectoderm, but not in the involuting mesoderm. IMAGE 7010138: At blastula (left image, sagittal section), expression is primarily in the animal cap cells, and barely detected in the equatorial zone. At gastrula stages (middle image, sagittal section, anterior to the left), there is strong expression in both the ectoderm and involuting mesoderm. At neural plate stages (right image, sagittal section, anterior to the left), weak expression is observed in both the neural plate and underlying dorsal mesoderm.

Figure 9.

Examples of Xenopus embryos hybridized with sense RNA probes to evaluate the extent of non-specific staining at the early stages when expression patterns are diffuse (cf. Figures 4, 5, 6, 10).

At neural plate stages all 19 genes are expressed in the ectoderm/epidermis, with enhanced expression on the dorsal side of the embryo, including the early neural plate and neural border zone (BZ); for nearly all of these genes sectioning the embryos also showed expression in the dorsal mesoderm (Table 7; Figure 8). At early neural plate stages Arrb2 is additionally expressed in the cement gland, and Ism1 in the pharyngeal endoderm. At neural tube closure stages, all 19 genes are expressed in the neural tube and neural crest; 15/19 genes are additionally expressed in the PPR or a subset of cranial placodes. At later tail bud and larval stages, expression patterns diverge (Figure 5, Table 7). However, each of the 19 genes is expressed in at least two tissues whose development is known to be regulated by Six1: at least one placode-derived structure (17/19), somite (18/19) or nephric mesoderm (17/19); the latter is consistent with a potential role in the BOR kidney deficit. Of those with placode expression, 17/17 are expressed in the otocyst, consistent with a potential role in the BOS3 hearing deficit. Although Six1 is thought to be primarily important in placode development, recent work also shows it is required for some neural crest derivatives (Yajima et al., 2014; Garcez et al., 2014); consistent with these reports we find that all 19 of the up-regulated genes are detected in the cranial neural crest and the branchial arch mesoderm. However, it should be noted that many of the 19 genes also have prominent expression in tissues that do not have detectable levels of Six1 at these developmental stages: neural tube/retina (19/19), heart (4/19), cement gland (3/19), pharyngeal endoderm/pouches (2/19), and blood islands (1/19). Therefore, additional transcriptional pathways must regulate these genes in these additional tissues.

Figure 5.

Whole mount in situ hybridization assays at tail bud to larval stages illustrating more divergent expression patterns of 14 up-regulated genes. A: Nrp1, Arrb2, BG022063, LOC100037047, BQ399495. B: Hrpt1, Rbm42-6, LOC414689, Uncx4.1, Xxylt1. C: Nfya, Tbc1d31, Cdca7l, Papd4-a. Side views at tail bud (left column) and larval (middle column) stages, with anterior to the left and dorsal to the top, and transverse sections at larval stages (right side) with dorsal to the top. Abbreviations: ba, branchial arches; cem, cement gland; ep, epibranchial placodes; epi, epidermis; fb, forebrain; h, heart; hb, hindbrain; IX/Xg, glossopharyngeal/vagal ganglion; l, lens; mb, midbrain; ne, nephric mesoderm; olf, olfactory placode; oto, otocyst; p, pineal; r, retina; spcd, spinal cord; so, somites; Vg, trigeminal ganglion; VIIg, facial ganglion.

Because the expression patterns of five of the genes identified in the microarray analysis were previously published, but not for all stages of development, we completed their developmental expression patterns (Figures 4–6, Table 7). Of note, Nrp1, a receptor for Sema3a, is expressed in the deep layer of the PPR (Figure 4A) and in several placodes including otic (Figure 5A). Trhda is a maternally-expressed nuclear protein that inhibits neural differentiation (Wu et al., 2001); since no ISH data are available we present the entire developmental series (Figure 6). Pbx1, a homeobox-containing transcription factor that interacts with Meis and Hox proteins to influence hindbrain and neural crest gene expression (Maeda et al., 2002), is additionally expressed in the animal ectoderm, some placodes and somites (Figure 6). Ism1 is a novel secreted factor co-expressed in FGF8-expressing domains (Pera et al., 2002); we found that in addition to the published expression pattern it is expressed in the pharyngeal endoderm and somites (Figure 6). Arnt is a bHLH/PAS factor that regulates many aspects of vertebrate development and tumorigenesis (Crews and Fan, 1999; Labrecque et al., 2013); we found that in addition to the published expression pattern (Bollerot et al., 2001), it is expressed in the neural plate, neural crest, PPR, olfactory placode, cranial ganglia and somites (Figure 6, Table 7). HoxA3 is a homeodomain transcription factor involved in axis and craniofacial patterning (Quinonez and Innis, 2014); we found that in addition to the expression pattern posted on Xenbase (http://www.xenbase.org/gene/showgene.do?method=display&geneId=482266&) it is expressed in a pattern very similar to the other up-regulated genes prior to neural tube closure, but is not subsequently expressed in the placodes (Figure 6, Table 7).

In situ expression of down-regulated genes

ISH analyses were performed for 10 down-regulated genes for which there are no ISH data available, and for 1 gene for which the data are incomplete (Foxi1/Xema) (Figures 6, 10–11, Table 8). At the early developmental stages, the 10 uncharacterized genes share very similar expression patterns with each other and with the up-regulated genes. All are expressed maternally, with enhanced localization to the animal hemisphere at cleavage and blastula stages (Figure 10); this localization was not due to poor probe penetration (Figure 7). At blastula stages, expression remained in the animal half of the embryo (Figure 10), in some cases extending into the equatorial region (Table 8; Figure 8). At gastrula stages, all 10 genes are expressed throughout the ectoderm; sectioning the embryos showed that a few also had weak expression in the involuting mesoderm (IMAGE 4057931, IMAGE 7010138, Ralgds, Dnaja4.2) (Table 8, Figure 8). The diffuse staining at early stages was not the result of non-specific background, as demonstrated by processing embryos with sense directed RNA probes (Figure 9).

Figure 10.

Whole mount in situ hybridization assays at early stages illustrating common expression patterns of 10 down-regulated genes. A: IMAGE 4174145, IMAGE 3399268, IMAGE 4057931, IMAGE 3400511, Socs3. B: MGC 114680, IMAGE 7010138, Cnfn1-a, Ralgds, Dnaja4.2. For all genes, expression is enhanced in the animal cells (an) of cleavage and blastula embryos, and is throughout the ectoderm at gastrula stages (side views). At early neural plate stages, all genes are expressed in the epidermis (epi), often more intensely on the dorsal (d) side, including neural plate (np) and border zone (bz) (all except IMAGE 3400511 are side views with anterior [a] to the left; IMAGE 3400511 is an anterior view). At later neural plate and neural tube stages, expression is strongest in the neural crest (nc), preplacodal region (ppr), neural tube (nt) and epidermis (epi) (anterior views). Cnfn1-a is uniquely expressed in the hatching gland (hg).

Figure 11.

Whole mount in situ hybridization assays at tail bud to larval stages illustrating more divergent expression patterns of 10 down-regulated genes. A: IMAGE 4174145, IMAGE 3399268, IMAGE 4057931, IMAGE 3400511, Socs3. B: MGC 114680, IMAGE 7010138, Cnfn1-a, Ralgds, Dnaja4.2. Side views at tail bud (left column) and larval (middle column) stages, with anterior to the left and dorsal to the top, and transverse sections at larval stages (right side) with dorsal to the top. Abbreviations are as in Figure 10. For Cnfn1-a: hg, hatching gland; clo, cloaca.

Table 8.

Developmental expression patterns of down-regulated genes

Gene title/ accession # OB #	Cleavage	Blastula st 8–9	Gastrula st 10–11.5	Neural Plate E: st 12–13 L: st 14–17	Neural tube closure st 18–22	Tail bud st 23–31	Larva st 32–38
IMAGE 4174145/BC080322 4174145	animal	AC	ecto	E: dorsal ecto L: NP, BZ, epi dorsal meso	NT, NC, PPR, dorsal epi, somites	brain, ret, spcd, lens, oto, head epi, BA, somites	same, plus nephric meso
IMAGE 3399268/BC079829 3399268	animal	AC, eq	ecto	E: dorsal ecto L: NP, BZ, epi dorsal meso	NT, NC, PPR, dorsal epi, somites	brain, ret, spcd, olf, lens, oto, cranial ganglia, epi, BA, somites	same, plus nephric meso
IMAGE 4057931/BG264797 4057931	animal	AC	ecto, weak meso	E: ecto L: NP, BZ, epi dorsal meso	NT, NC, PPR, dorsal epi, somites	brain, pineal, ret, spcd, lens, oto, BA, somites	same, plus olf, cranial ganglia, nephric meso
IMAGE 3400511/BG020210 3400511	animal	AC	ecto	E: dorsal ecto L: NP, BZ, epi dorsal meso	NT, NC, PPR, dorsal epi, somites	brain, ret, spcd, lens, head epi, BA, somites	same, plus oto, cranial ganglia, heart
Socs3/BG022871 4173655	animal 1. not detected*	AC 1. not detected*	ecto 1. not detected*	E: dorsal ecto L: NP, BZ, epi dorsal meso 1. strong expression*	NT, NC, PPR, dorsal epi, somites 1. strong expression*	brain, ret, spcd, lens, oto, BA, somites 1. strong expression*	same, plus cranial ganglia, nephric meso 1. strong expression*
MGC114680/BC097548 6316942	animal	AC	ecto	E: dorsal ecto L: NP, BZ, epi dorsal meso	NT, NC, PPR, somites	brain, ret, spcd, lens, oto, epi, BA, somites	same, plus cranial ganglia, nephric meso
IMAGE 7010138/CF521493 7010138	animal	AC, eq	ecto, weak meso	E: dorsal ecto L: NP, BZ, epi dorsal meso	NT, NC, PPR, dorsal epi, somites	brain, ret, spcd, lens, oto, epi, somites	same, plus BA, nephric meso
Cnfn1-a/BC068675 6316571	animal 2. present*	AC 2. present*	ecto 2. present*	E: dorsal ecto L: scattered cells in NP and BZ 2. present*	hatching gland, dorsal epi 2. present*	hatching gland, cem, scattered cells in epi 2. present*	same, plus brain, ret, spcd, oto, Vg, BA, somites, nephric meso, ventral pharynx, caudal noto 2. present*
Ralgds/BQ897357 4681426	animal	AC	ecto, meso	E: dorsal ecto L: NP, BZ, epi dorsal meso	NT, NC, PPR, dorsal epi, somites	brain, ret, spcd, olf, lens, oto, cranial ganglia, head epi, BA, somites	same, plus nephric meso
Dnaja4.2/BC046660.1 4930076	animal	AC	ecto, weak meso	E: dorsal ecto L: NP, BZ, epi dorsal meso	NT, NC, PPR, dorsal epi, somites	brain, ret, spcd, lens, oto, BA, somites	same, plus olf, cranial ganglia, nephric meso
Foxi1 (Xema)	3. not detected*	3. AC ecto	3. antero-ventral ecto	3. BZ, antero-ventral ecto	3. antero-ventral ecto	3. antero-ventral ecto	brain, ret, spcd, oto, cranial ganglia, BA, somites

ISH expression patterns at different developmental stages for 10 genes for which there are no ISH expression data. ISH pattern also was completed for 1 gene for which there are available descriptions. For 2 genes (Socs3, Cnfn1-a) presence of expression (*) was detected only by Northern blot, RNase protection, RT-PCR or RNA-Seq.

Abbreviations: AC, animal cap; BA, branchial arches; BZ, border zone; cem, cement gland; E, early; ecto, ectoderm; epi, epidermis; L, late; meso, mesoderm; NC, neural crest; noto, notochord; NP, neural plate; NT, neural tube; OB, Open Biosystems; olf, olfactory placode/pit; oto, otocyst; PPR, pre-placodal region; ret, retina; spcd, spinal cord; st, stage; Abbreviations in bold refer to previous published expression patterns; bold number indicate references for previously described expression data:

1. Kuliyev et al. (2005)
2. Xenbase: http://www.xenbase.org/gene/expression.do?method=displayGenePageExpression&tabId=1&geneId=5856331
3. Wessely et al. (2004); Suri et al. (2005)

At early neural plate stages ectodermal expression is enhanced on the dorsal side of the embryo, including the early neural plate and BZ. At later neural plate stages all genes are expressed in the epidermis, neural plate, neural crest, PPR, and epidermis, and most were also expressed in the dorsal mesoderm. Cnfn1-a is not detected in the mesoderm (Figure 8), and is uniquely expressed in scattered cells throughout the ectoderm. At neural tube closure stages, 9/10 of the genes are expressed in the neural tube, neural crest, PPR, dorsal epidermis and somites. At neural tube through tail bud stages Cnfn1-a is uniquely expressed in the hatching gland and in the epidermis near the ventral midline and surrounding the cloaca (Figures 10B, 11B). At later tail bud and larval stages, expression patterns diverge (Figure 11, Table 8). However, all 10 genes are expressed in tissues whose development is known to be regulated by Six1, including at least one placode derived structure (10/10), neural crest (9/10), somites (10/10) and nephric mesoderm (8/10). These temporal patterns of expression are mostly congruent with the RT-PCR results (Figure 3B), although in some cases the color reaction was allowed to develop for up to 48 hours due to the low abundance of the transcripts. For example, even though IMAGE 7010138 is mostly detected by RT-PCR only at the onset of gastrulation, it is clearly detectable by ISH at earlier and later stages, albeit, with less chromogenic intensity (Figure 8). Foxi1-Xema is a transcription factor involved in epidermal development (Suri et al., 2005). We found that in addition to tissues previously reported, at larval stages there is expression in neural tube-derived structures, placode-derived otocyst and cranial ganglia, neural crest-derived branchial arches and in somites (Figure 6, Table 8).

One would expect down-regulated genes to be detected in tissues that do not also express Six1. For some stages and tissues, this expectation holds. For example, 11/11 are expressed in the cleavage and blastula precursors of the ectoderm, which do not express Six1 (Pandur and Moody, 2000). In addition, at later stages 11/11 are expressed in the epidermis and neural tube; these results are consistent with gain-of-function studies in chick and frog showing that when Six1 is ectopically expressed it represses genes characteristic of these tissues (reviewed in Grocott et al., 2012; Saint-Jeannet and Moody, 2014). However, inconsistent with this expectation, most of the putative down-regulated genes are also expressed in the PPR, placodes (including otocyst), neural crest, somites, and nephric mesoderm. This overlap with Six1 expression suggests that additional transcriptional pathways upregulate these genes even in the presence of Six1.

In situ expression of homologous genes in chick

To assess whether the expression of putative Six1 target genes is conserved in amniotes we identified chick homologues using the Basic Local Alignment Search Tool. Out of 19 up-regulated and 17 down-regulated Xenopus transcripts we determined 17 and 11 chick mRNAs with high sequence identity. Of those, we selected five up-regulated genes (Pbx1, Arnt, XXYLT1, WDR67, Nrp1) and one down-regulated gene (DNAJA1) for expression analysis. Except Nrp1, all transcripts show similar expression profiles to their Xenopus homologues (Figures 12–16; Table 9). Nrp1 is not expressed in the ectoderm at any stage investigated, but found in the endoderm underlying the PPR (not shown). In contrast, Pbx1, Arnt, Wrd67 and DNAJA1 are broadly expressed in all three germ layers from primitive streak to 11–14 somite stages. However, their expression is generally enriched in the ectoderm including the future neural plate and PPR, the developing neural tube, neural crest cells and placodal tissue, in particular the otic and epibranchial placodes. Hybridization with sense-directed RNA probes indicates that these diffuse patterns of expression are not the result of non-specific background staining (Figure 17). At neural tube stages Six1 remains expressed in all placodes except for the lens (Sato et al., 2012). Interestingly, Pbx1, Arnt, XXYLT1 and WDR67 are also absent from the lens, whereas DNAJA1, a down-regulated transcript, is present. Unlike the other transcripts, XXYLT1, starts to be expressed in the posterior epiblast and primitive streak at gastrulation stages before being strongly upregulated in the neural plate and forming neural tube. At placodal stages it is present in the otic and epibranchial placodes, neural crest cells and the intermediate and posterior lateral plate mesoderm. Thus, expression patterns of selected genes in chick are generally similar to those observed in Xenopus.

Figure 12.

Whole mount in situ hybridization for Pbx1 from primitive streak to 14-somite stages. Rostral is to the top, caudal to the bottom of the images. Lines in C and E indicate the levels of sections shown in a–h; i–k are higher magnifications of the placode areal shown in c, f and g respectively. bz, border zone; eb, epibranchial placode; l, lens placode; nc, neural crest; np, neural plate; oto, otic placode; so, somite.

Figure 16.

Whole mount in situ hybridization for DNAJA1 from primitive streak to 14-somite stages. Rostral is to the top, caudal to the bottom of the images. Lines in C and E indicate the levels of sections shown in a–h; i–k are higher magnifications of the placode areal shown in c, f and g respectively. Abbreviations are as in Figure 12.

Table 9.

Developmental expression patterns of chick homologous genes

Gene Symbol/ NCBI gene ID/EST #	HH4 Gastrula	HH5 Head process	HH6/HH7 Neural plate	HH8/9 Neural folds	HH9/HH10 Neural tube	HH11/12
PBX1 / 395505/ ChEST643j12	NE, BZ, mesoderm, endoderm	NE, BZ, mesoderm, endoderm	NP, NC, PPR, mesoderm, endoderm	NP, NC, PPR, somites	NT, NC, otic, epibranchial placodes, notochord, somites, lateral plate mesoderm, endoderm	Mid- and hindbrain, NC, otic, epibranchial placode, notochord, somites, lateral plate mesoderm, endoderm
ARNT /374026 / ChEST814a24	NE, BZ, mesoderm, endoderm	NE, BZ, mesoderm, endoderm	NP, NC, PPR, ectoderm, mesoderm, endoderm	NP, NC, PPR, somites, epidermis	NT, NC, otic, epibranchial placodes, epi, notochord, somites, lateral plate mesoderm, endoderm, blood islands	Brain, notochord, NC, otic, epibranchial placode, epi, somites, lateral plate mesoderm, endoderm
XXYLT1 / 424896/ ChEST85d22	Primitive streak, posterior epiblast	NE, BZ	NP, NC	NP, NC, posterior paraxial mesoderm	Fore-, mid- and anterior hindbrain, cranial NC, posterior paraxial mesoderm	Brain, NC, otic, nephric mesoderm, posterior paraxial mesoderm
WDR67/ 420349/ ChEST325c2	NE, BZ, mesoderm, endoderm, posterior epiblast	NE, BZ, mesoderm, endoderm, posterior ectoderm	NP, NC, PPR, head fold	NP, NC, PPR, notochord, mesoderm, endoderm	NT, otic, epibranchial placodes, somites, notochord, mesoderm, endoderm	Brain, NC, lens placode, otic, epibranchial placode, epi somites, notochord, mesoderm, endoderm
DNAJA1/427376/ ChEST151p15	NE, BZ, mesoderm, endoderm, posterior epiblast	NE, BZ, mesoderm, endoderm, posterior ectoderm	NP, NC, PPR, lateral mesoderm, lateral endoderm, posterior ectoderm	NP, NC, PPR, somites, posterior ectoderm	NT, NC, lens placode, otic, epibranchial placode somites, lateral plate mesoderm, paraxial mesoderm, endoderm, blood islands	Brain, NC, lens placode, otic, epibranchial placode somites, lateral plate mesoderm, nephric mesoderm, endoderm, blood islands

Abbreviations: BZ, border zone; epi, epidermis; HH, Hamburger-Hamilton stage; NC, neural crest; NE, neural ectoderm; NP, neural plate; NT, neural tube; otic, otic placode or vesicle; PPR- Pre-placodal region

Figure 17.

Examples of chick embryos hybridized with sense RNA probes to evaluate the extent of non-specific staining (cf. Figures 12 – 16). A, D, F, I: primitive streak stages. G, head fold stage. B, E, J: 3–4 somite stage. C, 14-somite stage.

Summary

The expression patterns of the genes identified in this study as potential targets of Six1 are consistent with previous reports showing that Six1 plays an important role in the development of vertebrate cranial sensory placodes, somites and kidney. The majority of up-regulated genes are expressed in tissues known to express Six1 (PPR, placodes, somites, nephric mesoderm), which is expected for putative activated targets. However, their expression in Six1-negative tissues, including neural plate, and epidermis, indicates that they also are regulated by additional factors. The down-regulated genes are expressed in Six1-negative tissues, which is expected for putative repressed targets; their expression in some Six1-positive tissues indicates that they also are upregulated by additional factors even in the presence of Six1. Because the majority of the genes described herein are expressed in the developmental precursors of two organs whose defects are diagnostic for BOS and BOR syndromes, the ear and the kidney, they are new candidates for potential involvement in these birth defects.

EXPERIMENTAL PROCEDURES

Frog embryos, animal cap explants and microinjections

Fertilized Xenopus laevis eggs were obtained by either in vitro fertilization (for ACs) or gonadotropin-induced natural mating of adult frogs (for ISH) as described elsewhere (Moody, 2000). mRNAs were synthesized by in vitro transcription (Ambion, mMessage mMachine kit). mRNA encoding wild-type Six1 (400pg) was injected into the animal poles of both cells at the 2-cell stage. ACs were dissected from microinjected embryos and from uninjected siblings (controls) at stages 8.5–9.0, cultured in 1X Modified Barth’s solution and collected when sibling embryos reached stage 16. For assessment of Six1 levels on gene expression in whole embryos, previously characterized Six1-specific morpholino antisense oligonucleotides (Six1 MOs) or mRNAs encoding wild type Six1 (400pg), an activating Six1-VP16 fusion protein (100pg), or a repressive EnR-Six1 fusion protein (100pg) were injected into lateral-animal blastomeres of 16-cell embryos (Brugmann et al., 2004). Embryos for RT-PCR and in situ hybridization (ISH) assays were cultured in 100% Steinberg’s solution through blastula stages and in 50% Steinberg’s solution from blastula through larval stages.

Microarray and Statistical Analyses

Four independent samples of Six1 mRNA-injected ACs were collected. Each sample was derived from one different male and two different females and contained ~100 ACs. For each Six1-expressing sample, a control, uninjected sample from sibling embryos also was collected. All samples were processed in parallel for cDNA labeling and chip hybridization to reduce inter-sample variations. Total RNAs were extracted from ACs with the RNeasy mini kit (Qiagen). Integrity of RNA was assessed using the Agilent 2100 bioanalyzer (Agilent Technologies) and only samples with an integrity number > 8.0 were used. Total RNAs were labeled and fragmented with the Ovation Biotin RNA Amplification and Labeling System (NuGEN Technologies, Inc.). Briefly, 50ng total RNA was used for first- and second-strand cDNA syntheses. The synthesized cDNAs were amplified, purified with a PCR purification kit (Qiagen), then labeled and fragmented following the Ovation kit instructions. The labeled cDNAs were purified with the Dye Ex kit (Qiagen).

Chip hybridization and statistical analyses were performed by the NINDS-NIMH Microarray Consortium at The Translational Genomics Research Institute (T-GEN, Tempe, AZ). The Affymetrix GeneChip^® Xenopus laevis Genome Array (v1.0) bears 15,503 probe sets representing about 14,400 transcripts. The chips were washed and scanned as recommended by the Ovation Biotin RNA Amplification and Labeling System User Guide (version 1.0). GCOS software was used to determine signal intensities and detection calls for each gene. Five replicates of each sample were analyzed. The average replicate correlation values for the control samples and for the Six1-injected samples each were 0.94, indicating consistency between replicates despite random genetic variance between parental frogs. Using GeneSpring software, all experimental arrays were normalized to their matched control array. Genes were then filtered for at least 2 present calls out of 8 calls. The remaining list was tested for significant changes by ANOVA (p<0.05). The raw and processed data are available (GEO accession number GSE11144).

RT-PCR and qPCR

For RT-PCR analysis, whole, wild type Xenopus embryos were snap-frozen at a series of developmental stages. Total RNA was extracted the with RNeasy mini kit (Qiagen). Semi-quantitative RT-PCR within linear ranges was performed as previously described (Yan et al., 2009). Some primer sequences were obtained from published papers, whereas the others were designed with the Primer3 software (Rozen and Skaletsky, 2000; Supplemental Table 4). Reverse transcription was performed with 1.0μg of RNA using the SuperScript first-strand synthesis system (Invitrogen). Standard PCR amplifications were performed with PCR Supermix (Invitrogen) including 1.0μCi α-[³²P]dATP (Amersham) in the reaction mixture. All PCR assays were repeated at least three times with independent samples. Bands were visualized with a Storm 860 phosphorimager (Molecular Dynamics).

For qPCR analysis, three independent sets of ACs derived from wild-type Six1 mRNA-injected and sibling control embryos were obtained as above, cultured to neural plate stages, homogenized in TRIzol (Invitrogen, Carlsbad, CA), and total RNA was isolated. Residual genomic DNA was removed using DNAse (Turbo DNA-free, Invitrogen), and random-hexamer primed cDNA generated using ImPromp-II reverse transcriptase (Promega). Reactions were assembled using a EpMotion 5070 liquid handling system (Eppendorf) that combine forward and reverse gene-specific primers (0.3 μM final concentration, Integrated DNA Technologies, Supplemental Table 10), with 7.5 μl of SsoFast EvaGreen Supermix (Bio-Rad, Hercules, CA) in a 14 μl reaction. qPCR analysis was performed using a CFX-384 Real-Time PCR Detection System. Statistical differences were assessed by student’s t-test of the normalized CT values (delta-CT) for each independent analysis.

Whole mount in situ hybridization

Plasmids encoding a subset of the putative downstream genes identified in the microarray assay were either obtained from the laboratory in which they were originally reported (listed in Tables 4, 5), or purchased from Open Biosystems (Thermo Scientific) or Source Bioscience. Anti-sense and sense digoxigenin-labeled RNA probes were synthesized by in vitro transcription (Ambion, Megascript kit). Frog and chick embryos were fixed and processed for whole mount ISH according to standard protocols (Sive et al., 2000; Streit and Stern, 2001). Because some transcripts were low abundance, for some developmental stages the color reaction was allowed to develop up to 48 hours. For cleavage stage frog embryos, some were bisected after fixation to facilitate penetration of the reagents into the yolky vegetal cells during the ISH procedure. To visualize internal expression patterns Xenopus embryos already processed for whole-mount ISH were either bisected (blastula through neural plate stages) or embedded in 4% agarose and sectioned at ~70μm with a vibratome (neural tube through larval stages); chick embryos were processed for standard paraffin sectioning.

Table 10: RT-PCR and qPCR primer sequences

Figure 13.

Whole mount in situ hybridization for Arnt from primitive streak to 14-somite stages. Rostral is to the top, caudal to the bottom of the images. Lines in C and E indicate the levels of sections shown in a–h; i–k are higher magnifications of the placode area shown in c, f and g respectively. Abbreviations are as in Figure 12.

Figure 14.

Whole mount in situ hybridization for XXYLT1 from primitive streak to 14-somite stages. Rostral is to the top, caudal to the bottom of the images. Lines in B and F indicate the levels of sections shown in a–h; i–k are higher magnifications of the placode areal shown in c, f and g respectively. Abbreviations are as in Figure 12.

Figure 15.

Whole mount in situ hybridization for Wrd67 from primitive streak to 14-somite stages. Rostral is to the top, caudal to the bottom of the images. Lines in C and F indicate the levels of sections shown in a–h; i–k are higher magnifications of the placode areal shown in c, f and g respectively. Abbreviations are as in Figure12.