- Email: [email protected]
Lhx9 gene expression during early limb development in mice requires the FGF signalling pathway
Gene Expression Patterns xxx (2015) 1e7
Contents lists available at ScienceDirect
Gene Expression Patterns journal homepage: http://www.elsevier.com/locate/gep
Lhx9 gene expression during early limb development in mice requires the FGF signalling pathway Yisheng Yang, Megan J. Wilson* Developmental Biology and Genomics Laboratory, Department of Anatomy, Otago School of Medical Sciences, University of Otago, P.O. Box 56, Dunedin 9054, New Zealand
a r t i c l e i n f o
a b s t r a c t
Article history: Received 28 May 2015 Received in revised form 11 July 2015 Accepted 21 July 2015 Available online xxx
Lhx9 is a member of the LIM-homeodomain gene family necessary for the correct development of many organs including gonads, limbs, heart and the nervous system. In the context of limb development, Lhx9 has been implicated as an integrator for Fibroblast growth factor (FGF) and Sonic hedgehog (Shh) signalling required for proximaledistal (PD) and anterioreposterior (AP) development of the limb. Three splice variants of the Lhx9 transcript are expressed during development, two of which are predicted to act in a dominant negative fashion, competing with the DNA binding version of Lhx9 for binding to cofactors via the LIM-domain. We examined the expression pattern for the three alternative splice forms of Lhx9; Lhx9a, Lhx9b and Lhx9c during early limb development. We have found that of the three Lhx9 isoforms, only Lhx9a and Lhx9c (intact homeodomain) are expressed during early limb development, each with their own distinct expression pattern. Additionally we determined that Lhx9 expression overlaps with FGF10 expression in the developing limb bud mesenchyme. Limb bud explant cultures, in the presence of signalling pathway inhibitors, also indicated that Lhx9 mRNA expression in the limb bud was dependent on FGF signalling. © 2015 Published by Elsevier B.V.
Keywords: LIM-homeodomain Lhx9 Isoforms Limb development Regulation
1. Introduction LIM-homeodomain gene 9 (Lhx9) is a member of the Apterous group in the LIM-homeodomain (LIM-HD) family of evolutionary conserved transcription factors crucial for development. Lhx9 is expressed in a number of areas such as the gonads (Birk and et al., 2000), limbs (Retaux et al., 1999), heart (Smagulova et al., 2008) and many regions of the nervous system including the spinal cord neurons, diencephalon, telencephalic vesicles, dorsal mesencephalon, hypothalamus and pineal gland (Bertuzzi and et al., 1999; Wilson et al., 2008; Yamazaki and et al., 2014). Generation of murine Lhx9 mutant lines indicated that Lhx9 expression is essential for some aspects of neuronal development and absolutely essential for the normal formation of the bipotential gonad (Birk and et al., 2000; Wilson et al., 2008). Gross deficits in mouse gonad development are caused by the loss of only a single Lhx9 allele, other aspects of development remained unchanged (Birk and et al., 2000; Wilson et al., 2008). This appears to be due to be a result of functional redundancy with Lhx9 and the paralogue of Lhx2 (Wilson
* Corresponding author. . E-mail address: [email protected] (M.J. Wilson).
et al., 2008; Tzchori and et al., 2009) as both genes show marked overlap in expressed regions during development. Together, Lhx2 and Lhx9 make up the mammalian Apterous group of LIM-HD transcription factors and both have been implicated in human development and disease (Hou and et al., 2013; Kuzmanov and et al., 2014; Vladimirova and et al., 2009). Lhx9 and Lhx2 are required for proximaledistal (PD) and anterioreposterior (AP) limb development as a double knockout mouse strain for both genes displays gross defects along these axes; such as shortened zeugopod (radius and ulna/tibia and fibula) and autopod (carpals/tarsals, metacarpals/metatarsals and digits) regions, with only two or three digits formed at 15.5 days post coitum (dpc) (Tzchori and et al., 2009). Tzchori and colleagues deduced that Lhx9 and Lhx2, together with their cofactor, Ldb1 integrate the signalling interactions for both PD and AP outgrowth and patterning through mediating FGF10 and Shh expression in response to FGF8 signalling, and Grem1 expression in response to Shh signalling. This proposed function of Lhx9 in the context of limb development is comparable to its Drosophila homologue, apterous (ap), that is critical for wing development (Cohen et al., 1992). The ap gene is required for the specification of dorsal cell fates, acting as a signalling centre at
http://dx.doi.org/10.1016/j.gep.2015.07.002 1567-133X/© 2015 Published by Elsevier B.V.
Please cite this article in press as: Yang, Y., Wilson, M.J., Lhx9 gene expression during early limb development in mice requires the FGF signalling pathway, Gene Expression Patterns (2015), http://dx.doi.org/10.1016/j.gep.2015.07.002
2
Y. Yang, M.J. Wilson / Gene Expression Patterns xxx (2015) 1e7
the boundary between dorsal and central cell to direct limb outgrowth (Diaz-Benjumea and Cohen, 1993; Ng et al., 1996). However, whether FGF or Shh signalling may control Lhx9 or Lhx2 expression in the early phases of limb development was not determined. Additionally, it was previously suggested that Lhx2/9 together with Ldb1 may act to maintain the limb mesenchymal cells in a progenitor state in response to FGF and Shh signals from the apical ectodermal ridge (AER) (Tzchori and et al., 2009), lending support to the notion that FGF and/or Shh signals may influence Lhx2/9 expression. Evidence from Xenopus laevis has shown that FGF15 is able to up-regulate expression of the Xenopus orthologue for Lhx9 (XLhx9) expression during nervous system development (Atkinson-Leadbeater et al., 2009). Additionally, while ectopic Shh expression during chick brain development was shown to decrease Lhx9 expression (Sun et al., 2008), the effect FGF and Shh signalling has on Lhx9 expression have not been investigated in regards to the mouse model or limb development. Lhx9 has three different protein isoforms derived from mRNA transcript variants (Smagulova et al., 2008; Failli et al., 2000). Of the three isoforms, transcript variant 3 (Lhx9 isoform c, also called Lhx9-HD) is the only one to retain a full HD domain and thus directly bind to DNA. Whereas transcript variant 1 (Lhx9 isoform a) and 2 (Lhx9 isoform b) have truncated HD domains, along with different 50 and 30 untranslated regions but intact LIM domains, making them LIM-only factors (LMOs) as they are predicted not to bind to DNA. However Lhx9a/b proteins are still able to interact with protein partners or cofactors (Molle et al., 2004) and it has been proposed that the LMOs could act competitively against Lhx9-HD for co-factors (Failli et al., 2000; Molle et al., 2004), given the LIM domain is the proteineprotein interaction motif to interact with co-factors (Agulnick and et al., 1996; Jurata et al., 1996). Interestingly, Lhx9a protein distribution is limited to the nuclear compartment, as opposed to Lhx9c, which located in both the cytoplasm and the nucleus (Molle et al., 2004). In addition, Lhx9a/b is the major splice form expressed during mouse heart development, as opposed to Lhx9c, which is not expressed at all (Smagulova et al., 2008). This indicates that there are differences in the regulation of Lhx9 isoform expression, in particular Lhx9a may not just be a regulator of Lhx9c through cofactor binding, but may have its own role too. To date, the spatial expression of the Lhx9 isoforms during early limb development has not been investigated. Here we examine the expression patterns of Lhx9 splice isoforms during early limb development and expand on the relationship between Lhx9 and the signalling pathways critical for early limb development. Lhx9c and Lhx9a mRNA transcript variants expression show a distinct pattern of expression from one another but are still located predominantly towards the anterior region of limb buds. We find that Lhx9 protein is associated with proliferative cells in the progress zone, in addition to Lhx9 mRNA expression overlaps markedly with the FGF10 expression. We further show that Lhx9 is dependent on FGF signalling for its expression. 2. Materials and methods 2.1. Embryo collection Embryos at stages 10.5 and 11.5 dpc were collected from timed mated inbred BALB/c mice. Whole embryos used for in situ hybridisation were fixed overnight in 4% paraformaldehyde (PFA) and dehydrated in methanol prior to use. For explant culture, forelimb and hindlimb buds from 10.5 dpc embryos were dissected in Leibovitz's L-15 media (Life Technologies) immediately prior to cell culture. All animal work was approved by the University of
Otago Animal Ethics Committee (ET 18/11). 2.2. RNA probe design and in situ hybridisation Primers were designed to target each of the three described Lhx9 isoforms; a b and c (HD) based on mRNA sequences available € in GenBank (accession numbers NM_001025565.2, NM_010714.3 and NM_001042577.1 respectively). Primers used are described in Table 1. Complimentary DNA (cDNA) derived from total mouse RNA was used as template for PCR amplification of each target transcript. Amplified products were cloned into pBluescript II KS (±) vector (Fermentas Molecular Biology Tools) and sequenced prior to in vitro transcription. DIG-labelled probe synthesis and in situ hybridisation were performed as described in Hargrave et al. (2006). A RNA probe to target the coding regions of the Lhx9 mRNA sequence (referred to as Lhx9E1-5 from this point forward), cloned in to (pGEM®-T) was gifted to us from Prof. Peter Koopman from the Institute of Molecular Bioscience, University of Queensland. The colour reaction was performed with nitroblue tetrazolium chloride (NBT) and 5-bromo-4-chloro-30 -indolyphosphate p-toluidine salt (BCIP) (Roche). Stained embryos were fixed with 4% paraformaldehyde (PFA) and mount in embryo wells with 100% glycerol. Images were taken using Olympus SZX10 bright field microscope with Olympus DP digital camera and software. 2.3. Immunohistochemistry Intact 10.5 dpc embryos were fixed overnight in 4% PFA, dehydrated with methanol before being paraffin embedded and sectioned to a thickness of 12 mm. Sections were deparaffinised and rehydrated with 100% xylene (2, 3 min each) and ethanol water dilutions (2 100%, 90%, 70%, 50%, 3 min each). Antigen retrieval was performed by boiling samples for a total of 30 min, immersed in sodium citrate (10 mM, 0.05% Tween 20, pH 6.0) using a conventional microwave oven on high power. Immunohistochemistry detection for Lhx9, E-cadherin (E-cad) and Ki-67 was achieved using indirect horseradish peroxidase (HRP)-linked secondary antibody. Slides were washes three times with PBTx (PBS containing 0.025% Triton X-100) and blocked with 10% heat inactivated donkey serum and 1% bovine serum albumin in PBS. Slides were incubated overnight at 4 C with the primary antibody diluted in blocking solution. After three washes with PBTx, slides were incubated with 3% H2O2 for 5 min, and washed twice with PBTx. Slides were then incubated with secondary antibody diluted in blocking solution at room temperature for 2 h, washed, mounted and imaged with an Olympus AX70 light microscope. Antibodies used were: rabbit polyclonal anti-Lhx9 (Abcam Ab28737, 1: 100 dilution), rabbit polyclonal anti-E-cadherin (Cell Signalling Technologies, 1: 400 dilution), rabbit polyclonal anti-Ki67 (Novus Biologicals, 1: 50
Table 1 Oligonucleotide primers used in this study. Target
Forward primer 50 e 30
Reverse primer 50 e 30
Lhx9a Lhx9b Lhx9c FGF8 FGF10 Shh Lhx9 qPCR Gapdh qPCR
TCTGGCCACCTCCTCTCT AA ACGGTAGCTTTGCTTGTAGGA GGGGGTGTTGATAAAGCTGA GGGACACGAGGACCGACCCTT CCCGCTGACCTTGCCGTTCTT CAAGGAGGAGCGCACACGCA CGTCTCTACGCTTCTGCATC CATGGCCTTCCGTGTTCCTA
TCCTTGCAGTAAATGCTACCG AAATCTATCGAGCAGGAGGGT CGCCCATCGTAATACCATTT CTCCGGGGCCCAAGTCCTCT CCCGCTGACCTTGCCGTTCTT ACGCGGTCTCCGGGACGTAA GGCGGAAAGGACACGAAT CCTGCTTCACCACCTTCTTGAT
Please cite this article in press as: Yang, Y., Wilson, M.J., Lhx9 gene expression during early limb development in mice requires the FGF signalling pathway, Gene Expression Patterns (2015), http://dx.doi.org/10.1016/j.gep.2015.07.002
Y. Yang, M.J. Wilson / Gene Expression Patterns xxx (2015) 1e7
dilution) and goat anti-rabbit HRP conjugated secondary (Santa Cruz, 1 in 400 dilution).
2.4. Explant tissue culture Embryos were dissected from timed-mated female mice in Leibovitz's L-15 media (Life Technologies). Excised limb buds were transferred to DMEM-agar mould consisting of DMEM (Life Technologies), 10% fetal bovine serum (Peribio Science), 1% L-glutamine (Thermo Scientific), 0.1% ampicillin (50 mg/mL, Thermo Scientific) and 1.5% agar. Moulds were made by setting mould mixture on top of riveted stainless steel discs specified previously (Capel and Batchvarov, 2008). Limb buds placed in DMEM-agar moulds were cultured for 23e24 h (37 C, 5% CO2) in DMEM culture media (DMEM, 10% fetal bovine serum, 1% L-glutamine and 0.1% ampicillin) containing 100 nM of either PD-166285 hydrate (Sigma), or cyclopamine hydrate (Sigma), or vehicle controls consisting of DMSO or 100% ethanol as matched to controls for the respective inhibitors. Following culture, limb buds were either fixed with 4% PFA overnight and dehydrated in methanol for in situ hybridisation, or washed in PBS prior to RNA extraction using PureLink® RNA Mini kit (life Technologies). A total of three limb buds per culture, per treatment were pooled for RNA extraction.
3
2.5. Reverse transcriptase quantitative PCR (RT-qPCR) Following RNA extraction, limb bud RNA samples were treated with DNAseI prior to conversion to cDNA using iScript™ cDNA Synthesis kit (Bio-Rad). Quantitative PCR (qPCR) was performed using specific oligonucleotides for Lhx9 (that would amplify all splice forms of Lhx9) and a reference control gene, Gapdh using SYBR select mastermix (Life Technologies). Sequences for oligonucleotide primers used for RT-qPCR are listed in Table 1. Two microliters of cDNA was used for each reaction. qPCR assays were performed on an Alignment Mx3000P system. A total of three technical replicates were performed for each cDNA sample along with two biological replicates. Each biological replicate was a pooled sample of six limb buds. To determine if differences between controls and inhibitor treated samples were statistically significant, a non-parametric ManneWhitney U test was performed.
3. Results and discussion 3.1. Two Lhx9 transcript variants are expressed in the E11.5 mouse limb Previous studies have shown that Lhx9 mRNA to be strongly
Fig. 1. In situ hybridisation to detect expression of Lhx9 transcript variants during early limb development (A) Genome browser image of the Lhx9 gene locus including the exonintron structure for the Lhx9 alternative splice variants. MultiZ alignment indicating the orthologous sequences in the named vertebrate genomes. Exons 1 to 5 of the Lhx9c splice form are highly conserved throughout vertebrates, exon 5 encodes for the DNA binding domain. Transcript variant named Lhx9b (NM_010714.3) differs in both first and last exons to that of Lhx9c. Whereas Lhx9a, (NM_001025565.2) only differs with respect to the final exon. Coloured boxes indicate the exon regions used to design probes targeted to specific Lhx9 variant mRNAs. (BeE) Whole-mount staining of 11.5 dpc fore- (left) and hind- (right) limb buds with probes specifying the different transcript variants of Lhx9. (B) Lhx9E1-5 staining in the distal regions of the limb bud extending from the anterior most limb margin to the posterior but was not detected near the body wall on the posterior side of the limb bud (asterisk). (C) Weak staining for Lhx9c was distributed throughout the entire limb bud. (D) Lhx9a/b staining is concentrated to the anterior of the limb bud only. As the Lhx9bspecific probe (E) produces no significant staining in either the limb bud signal observed with Lhx9a/b (D) is likely to be due to Lhx9a expression. Scale bars represent 200 mm. Abbreviations: Forelimb (FL), Hindlimb (HL).
Please cite this article in press as: Yang, Y., Wilson, M.J., Lhx9 gene expression during early limb development in mice requires the FGF signalling pathway, Gene Expression Patterns (2015), http://dx.doi.org/10.1016/j.gep.2015.07.002
4
Y. Yang, M.J. Wilson / Gene Expression Patterns xxx (2015) 1e7
expressed in both fore and hindlimb buds (Retaux et al., 1999; Tzchori and et al., 2009). Initially we used an in situ hybridisation probe against the full-length coding region of Lhx9 exons 1 to 5 (Lhx9E1-5; Fig. 1A). At 11.5 dpc, the Lhx9E1-5 probe detected expression in the distal regions of both the fore and hindlimb (Fig. 1B). We also found that posterior staining for Lhx9E1-5 did not completely extend back towards the body wall in both limbs (Fig. 1B, asterisks). While the Lhx9E-15 probe includes exon 5 encoding for an intact C-terminus HD domain, it also substantially overlaps with the other two Lhx9 transcript splice forms (exons 1e4) and there is likely to hybridise to all splice forms of Lhx9 mRNA (Fig. 1A). To further distinguish between the three major Lhx9 transcripts, we amplified separate regions of mRNA from cDNA; Lhx9c probe was specific for the exon and 30 untranslated region (UTR) for the intact HD domain (Fig. 1A, exon shaded grey). Lhx9a/b probe specified an alternative 30 end of the transcript, exon and 30 UTR (Fig. 1A, exon shaded green). To distinguish between Lhx9a and Lhx9b transcripts a third probe Lhx9b was used, that is predicted to hybridise to 50 UTR and exon specific for the Lhx9b transcript (Fig. 1A, exons shaded red). Staining with the Lhx9E1-5 probe produced strong staining at the anterior and distal regions of the both fore and hindlimb buds (Fig. 1B) but did not extend to the posterior most limb margin (Fig. 1B, asterisk). The Lhx9c probe targeted to the region of Lhx9 mRNA encoding for the HD domain, produced weak staining
throughout the limb bud (Fig. 1C). Staining from the Lhx9c probe was more diffuse compared to Lhx9E1-5 (Fig. 1BeC). Lhx9a expression was detected at the distal anterior regions of both limb buds (Fig. 1D), unlike Lhx9c mRNA, which was detected throughout the entire limb buds (Fig. 1C). In situ hybridisation with a RNA probe targeting the Lhx9b transcript did not detect any significant expression within 11.5 dpc limb buds (Fig. 1E), suggesting with the Lhx9a/b probe may reflect expression of the Lhx9a isoform only. Collectively this shows that while there was some overlap in the expression patterns of Lhx9 transcripts; Lhx9a and Lhx9c also have their own distinct expression domains. As Lhx9c and Lhx9a transcripts, but not the Lhx9b-specific transcript, were detected, these two splice forms are likely to be the dominant transcript variants expressed at this stage of limb development. This could be a result the functional role that Lhx9 plays during limb development or following on from a previous finding where it was shown that Lhx9a and b isoforms were expressed during heart development but not Lhx9c (Smagulova et al., 2008), highlighting differential regulatory control over the isoforms. Given previous suggestions of a competitive role that Lhx9a might have over Lhx9c by way of cofactor binding competition (Failli et al., 2000; Molle et al., 2004), this may explain the more anteriorly located expression of Lhx9a which might serve as a spatial controller of Lhx9 function over the AP axis, potentially focusing Lhx9 function distally. As Lhx9E1-5 and Lhx9a transcripts were expressed proximally
Fig. 2. Immunohistochemistry of fore- and hindlimb sections from 11.5 dpc mouse embryos. (A) Lhx9 protein was localised predominantly to the anterior and distal regions in the limb buds. (B) E-cadherin marks the epithelial covering and AER of the limb bud. (C) Ki67 staining was detected in cells that are widely distributed through the limb buds, particularly in the distal most regions of the mesenchyme. Scale bars represent 200 mm.
Please cite this article in press as: Yang, Y., Wilson, M.J., Lhx9 gene expression during early limb development in mice requires the FGF signalling pathway, Gene Expression Patterns (2015), http://dx.doi.org/10.1016/j.gep.2015.07.002
Y. Yang, M.J. Wilson / Gene Expression Patterns xxx (2015) 1e7
5
Fig. 3. Comparative whole-mount in situ hybridisation for known signalling centres with Lhx9 expression in the forelimb bud at 10.5 dpc. (A) Lhx9E1-5 mRNA is expressed in the anterior and distal regions of the limb bud. (B) FGF10 mRNA is expressed within the progress zone, in the limb mesenchyme. (C) FGF8 staining marks the AER located at the distal end of the limb bud. (D) Shh mRNA is detected in the ZPA, located at the posterior margin. Scale bars represent 100 mm.
underneath the AER, we investigated whether Lhx9 protein overlapped with proliferative mesenchymal cells in the area under the AER known as the progress zone (Summerbell and Lewis, 1975). Immunohistochemistry was used to detect Lhx9 (LHX9) protein
using 11.5 dpc limb bud sections. The antibody used was raised against the C-terminus polypeptides 289e338 of the human LHX9 protein and is predicted to recognise both isoforms of Lhx9 (with and without the homeodomain). Lhx9 protein was detected in the
Fig. 4. Lhx9 expression in limb buds following in vitro inhibition of FGF and Shh signalling pathways. (AeD) In situ hybridisation to detect changes to expression of Lhx9 following in vitro culture in the presence of either the FGF inhibitor (C), Shh inhibitor (D) or their respective vehicle controls (A, B). Lhx9E1-5 mRNA expression was no longer detectable via in situ hybridisation following culture with the FGF signalling pathway inhibitor. Whereas, strong Lhx9 signal was still detectable following culturing with the Shh signalling pathway inhibitor cyclopamine (D). Quantification of Lhx9 expression was performed using RT-qPCR (E) and revealed a significant decrease in Lhx9 expression after FGF signalling inhibition (P ¼ 0.0079; ManneWhitney U test), but Lhx9 mRNA levels were not significantly altered after Shh signalling inhibition. Scale bars represent 200 mm.
Please cite this article in press as: Yang, Y., Wilson, M.J., Lhx9 gene expression during early limb development in mice requires the FGF signalling pathway, Gene Expression Patterns (2015), http://dx.doi.org/10.1016/j.gep.2015.07.002
6
Y. Yang, M.J. Wilson / Gene Expression Patterns xxx (2015) 1e7
nucleus and enriched in an asymmetrical pattern towards the anterior mesenchyme compared to the posterior (Fig. 2A). Lhx9 protein was limited to the mesenchymal cells only and was not found in the outer epithelial layer including the AER, highlighted by the absence of staining when compared to E-cadherin (Fig. 2B). In comparison to Ki-67 (Fig. 2C), a marker of proliferative cells (Scholzen and Gerdes, 2000), Lhx9 staining showed marked overlap (Fig. 2A and C), indicating that many of the proliferative cells in the early limb bud also co-express Lhx9 protein. 3.2. Lhx9 in the early limb bud expression overlaps with FGF10 With Lhx9 expression showing a very distinct pattern in the anterior and distal regions, this suggested that there is spatial regulation of expression. As Lhx9E1-E5 gene expression is strongest within the anterior most regions of the limb bud, expanding distally and down the posterior regions but not completely to the posterior limb bud boundary at 10.5 dpc (Fig. 3A). This expression pattern prompted us to investigate Lhx9 expression with respect to both FGF and Shh signalling pathways, as they are the key signalling factors acting as morphogens in limb development (Riddle et al., 1993; LopezMartinez and et al., 1995; Sun et al., 2002; Mariani et al., 2008), to establish spatial gene expression patterns required for correct patterning of the limb. Previously, Lhx9 (along with Lhx2) was shown to be required for PD and AP limb development by integrating FGF and Shh signals (Tzchori and et al., 2009), however, the possibility of Lhx9 expression being dependent on either of the signals was not determined, but could be regulated by one or both signals resulting in the asymmetrical expression domains. Given that expression of the DNA binding form of Lhx9 (Lhx9E1-5) was reduced within the posterior most border near the region of the zone of polarizing activity (ZPA), we compared Lhx9 expression to that of Shh (Fig. 3). Shh expression in the ZPA is required for correct patterning of the anterioreposterior axes of the bud (Riddle et al., 1993), acting as a morphogen to control limb development with high to low concentrations across the AP axis. The absence of expression of Lhx9 in the ZPA could indicate that high levels of Shh signal may inhibit Lhx9 expression in this region at 10.5 dpc. FGF10 and FGF8 mRNA was detected within the progress zone (Fig. 3B) and AER respectively (Fig. 3C) In comparison with Lhx9 expression, FGF10 mRNA overlaps with Lhx9 mRNA expressing cells in the progress zone in the anterior and proximal regions of the limb bud (Fig. 3A and B). FGF8 transcript was found in the AER, located beside the limb bud region of strongest Lhx9 mRNA detection (Fig. 3A and C). Unlike the FGF factors, Shh mRNA did not overlap with expression of the full length Lhx9 transcript (Fig. 3D). In order to determine if FGF and Shh signalling could influence Lhx9 limb expression, we performed explant limb bud culture in the presence of specific FGF and Shh small molecular signalling inhibitors, PD-116285 and cyclopamine hydrate respectively. In situ hybridisation for the Lhx9 transcript in post cultured limb buds showed that Lhx9 mRNA was no longer able to be detected in the absence of FGF signals (Fig. 4A and C), indicating loss of both Lhx9c and Lhx9a isoforms as the probe detects for both variants (Fig. 1A; Lhx9E1-5). As in situ hybridisation is not quantitative Lhx9 gene expression levels were also determined by RT-quantitative PCR (RTqPCR) (Fig. 4E). Incubation of limb bud tissue in the presence of the FGF inhibitor resulted in a significant reduction of Lhx9 mRNA levels by 80% relative to the vehicle control. In situ hybridisation of Lhx9 mRNA on limb buds incubated in the presence of cyclopamine still detected high levels of Lhx9 expression (Fig. 4B and D). RT-qPCR did not detect any significant differences in Lhx9 expression levels between vehicle and Shh inhibitor samples (Fig. 4E). This suggests
Fig. 5. Model of Lhx9 regulation in the early limb bud. FGF10 (green) is produced by cells within the limb mesenchyme and, along with FGF8 (blue) secretion from the AER, acts to maintain outgrowth of the developing limb (Mariani et al., 2008). Lhx9 expression (shaded grey) is positively regulated by FGF signalling and is likely to be part of a gene network involving both FGF10 and FGF8. Shh is produced in the ZPA (red), acts as a morphogen to regulate correct anterioreposterior patterning during limb development. Our results show that Shh signalling has no effect on Lhx9 expression however earlier work indicated that Lhx9 is required for correct Shh expression (Tzchori and et al., 2009).
that FGF signalling is required for correct Lhx9 expression (Fig. 5) in the early limb bud, whereas Shh play no significant role in controlling the expression pattern of Lhx9. These results are consistent with previous evidence indicating FGF-dependent regulation of XLhx9 seen during Xenopus laevis nervous system development (Atkinson-Leadbeater et al., 2009). However, as we used a broad-spectrum tyrosine kinase inhibitor, PD-166285 (Panek and et al., 1997) to reduce FGF signalling in vitro, we are unable to pin point the exact FGF signalling factor or factors that are directly responsible for inducing Lhx9 expression. Furthermore, we cannot rule out the possibility that this FGF regulation maybe indirectly acting upon Lhx9 gene expression, as there is no evidence of the FGF signalling cascade being able to directly regulate Lhx9 expression. However, it is likely that the FGF signalling does play a role in the induction of Lhx9 given the given the marked overlap with FGF10 expression in the limb bud mesenchyme (Fig. 3). Additionally, should Lhx9 expression be under the regulation of FGF8/10, it would fit in with previous suggestions that Lhx2, Lhx9 and Ldb1 mediate the positive feedback loop between FGF8 and FGF10 (Tzchori and et al., 2009). With regards to the relationship between Lhx9 and Shh, we found no effect on Lhx9 expression with the loss of Shh signalling, indicating the Shh signal pathway is not necessary for Lhx9 gene regulation. Previous work suggested that Lhx2, Lhx9 and Ldb1 are required for Shh signal transduction as there was a reduction of Shh mRNA expression in Lhx2eLhx9 double mutants (Tzchori and et al., 2009). Our data and this previous study would indicate that Lhx9 acts upstream of Shh in early limb patterning (model shown in Fig. 5). In summary, we showed that the dominant Lhx9 isoforms expressed during limb development are Lhx9c and Lhx9a, each with their own distinct expression domain in early limb development. Lhx9 expression overlapped markedly with FGF10 expression cells in the limb mesenchyme and was associated with proliferative cells of the progress zone. In vitro inhibition of FGF signalling significantly reduced Lhx9 expression implying that correct Lhx9 gene expression during limb development requires the FGF signalling pathway.
Please cite this article in press as: Yang, Y., Wilson, M.J., Lhx9 gene expression during early limb development in mice requires the FGF signalling pathway, Gene Expression Patterns (2015), http://dx.doi.org/10.1016/j.gep.2015.07.002
Y. Yang, M.J. Wilson / Gene Expression Patterns xxx (2015) 1e7
Acknowledgements We would like to thank Phoebe Heenan, Simon Blanchoud and Mitchell Weston for helpful comments on earlier versions of this manuscript. This project was supported by a University of Otago Research Grant to MJW. References Agulnick, A.D., et al., 1996. Interactions of the LIM-domain-binding factor Ldb1 with LIM homeodomain proteins. Nature 384, 270e272. http://dx.doi.org/10.1038/ 384270a0. Atkinson-Leadbeater, K., Bertolesi, G.E., Johnston, J.A., Hehr, C.L., McFarlane, S., 2009. FGF receptor dependent regulation of Lhx9 expression in the developing nervous system. Dev. Dyn. Off. Publ. Am. Assoc. Anat. 238, 367e375. http:// dx.doi.org/10.1002/dvdy.21850. Bertuzzi, S., et al., 1999. Characterization of Lhx9, a novel LIM/homeobox gene expressed by the pioneer neurons in the mouse cerebral cortex. Mech. Dev. 81, 193e198. Birk, O.S., et al., 2000. The LIM homeobox gene Lhx9 is essential for mouse gonad formation. Nature 403, 909e913. http://dx.doi.org/10.1038/35002622. Capel, B., Batchvarov, J., 2008. Preparing recombinant gonad organ cultures. CSH Protoc. http://dx.doi.org/10.1101/pdb.prot5078, 2008, pdb prot5078. Cohen, B., McGuffin, M.E., Pfeifle, C., Segal, D., Cohen, S.M., 1992. apterous, a gene required for imaginal disc development in Drosophila encodes a member of the LIM family of developmental regulatory proteins. Genes Dev. 6, 715e729. Diaz-Benjumea, F.J., Cohen, S.M., 1993. Interaction between dorsal and ventral cells in the imaginal disc directs wing development in Drosophila. Cell 75, 741e752. Failli, V., Rogard, M., Mattei, M.G., Vernier, P., Retaux, S., 2000. Lhx9 and Lhx9alpha LIM-homeodomain factors: genomic structure, expression patterns, chromosomal localization, and phylogenetic analysis. Genomics 64, 307e317. http:// dx.doi.org/10.1006/geno.2000.6123. Hargrave, M., Bowles, J., Koopman, P., 2006. In situ hybridization of whole-mount embryos. Methods Mol. Biol. 326, 103e113. http://dx.doi.org/10.1385/1-59745007-3:103. Hou, P.S., et al., 2013. LHX2 regulates the neural differentiation of human embryonic stem cells via transcriptional modulation of PAX6 and CER1. Nucleic Acids Res. 41, 7753e7770. http://dx.doi.org/10.1093/nar/gkt567. Jurata, L.W., Kenny, D.A., Gill, G.N., 1996. Nuclear LIM interactor, a rhombotin and LIM homeodomain interacting protein, is expressed early in neuronal development. Proc. Natl. Acad. Sci. U. S. A. 93, 11693e11698. Kuzmanov, A., et al., 2014. LIM-homeobox gene 2 promotes tumor growth and metastasis by inducing autocrine and paracrine PDGF-B signaling. Mol. Oncol. 8, 401e416. http://dx.doi.org/10.1016/j.molonc.2013.12.009.
7
Lopez-Martinez, A., et al., 1995. Limb-patterning activity and restricted posterior localization of the amino-terminal product of Sonic hedgehog cleavage. Curr. Biol. CB 5, 791e796. Mariani, F.V., Ahn, C.P., Martin, G.R., 2008. Genetic evidence that FGFs have an instructive role in limb proximal-distal patterning. Nature 453, 401e405. http:// dx.doi.org/10.1038/nature06876. Molle, B., Pere, S., Failli, V., Bach, I., Retaux, S., 2004. Lhx9 and lhx9alpha: differential biochemical properties and effects on neuronal differentiation. DNA Cell Biol. 23, 761e768. http://dx.doi.org/10.1089/1044549042531422. Ng, M., Diaz-Benjumea, F.J., Vincent, J.P., Wu, J., Cohen, S.M., 1996. Specification of the wing by localized expression of wingless protein. Nature 381, 316e318. http://dx.doi.org/10.1038/381316a0. Panek, R.L., et al., 1997. In vitro pharmacological characterization of PD 166285, a new nanomolar potent and broadly active protein tyrosine kinase inhibitor. J. Pharmacol. Exp. Ther. 283, 1433e1444. Retaux, S., Rogard, M., Bach, I., Failli, V., Besson, M.J., 1999. Lhx9: a novel LIMhomeodomain gene expressed in the developing forebrain. J. Neurosci. Off. J. Soc. Neurosci. 19, 783e793. Riddle, R.D., Johnson, R.L., Laufer, E., Tabin, C., 1993. Sonic hedgehog mediates the polarizing activity of the ZPA. Cell 75, 1401e1416. Scholzen, T., Gerdes, J., 2000. The Ki-67 protein: from the known and the unknown. J. Cell. Physiol. 182, 311e322. http://dx.doi.org/10.1002/(SICI)10974652(200003)182, 3<311::AID-JCP1>3.0.CO;2-9. Smagulova, F.O., Manuylov, N.L., Leach, L.L., Tevosian, S.G., 2008. GATA4/FOG2 transcriptional complex regulates Lhx9 gene expression in murine heart development. BMC Dev. Biol. 8, 67,. http://dx.doi.org/10.1186/1471-213X-8-67. Summerbell, D., Lewis, J.H., 1975. Time, place and positional value in the chick limbbud. J. Embryol. Exp. Morphol. 33, 621e643. Sun, X., Mariani, F.V., Martin, G.R., 2002. Functions of FGF signalling from the apical ectodermal ridge in limb development. Nature 418, 501e508. http://dx.doi.org/ 10.1038/nature00902. Sun, X., Saitsu, H., Shiota, K., Ishibashi, M., 2008. Expression dynamics of the LIMhomeobox genes, Lhx1 and Lhx9, in the diencephalon during chick development. Int. J. Dev. Biol. 52, 33e41. http://dx.doi.org/10.1387/ijdb.072386xs. Tzchori, I., et al., 2009. LIM homeobox transcription factors integrate signaling events that control three-dimensional limb patterning and growth. Development 136, 1375e1385. http://dx.doi.org/10.1242/dev.026476. Vladimirova, V., et al., 2009. Aberrant methylation and reduced expression of LHX9 in malignant gliomas of childhood. Neoplasia 11, 700e711. Wilson, S.I., Shafer, B., Lee, K.J., Dodd, J., 2008. A molecular program for contralateral trajectory: Rig-1 control by LIM homeodomain transcription factors. Neuron 59, 413e424. http://dx.doi.org/10.1016/j.neuron.2008.07.020. Yamazaki, F., et al., 2014. The Lhx9 homeobox gene controls pineal gland development and prevents postnatal hydrocephalus. Brain Struct. Funct. http:// dx.doi.org/10.1007/s00429-014-0740-x.
Please cite this article in press as: Yang, Y., Wilson, M.J., Lhx9 gene expression during early limb development in mice requires the FGF signalling pathway, Gene Expression Patterns (2015), http://dx.doi.org/10.1016/j.gep.2015.07.002
Contents lists available at ScienceDirect
Gene Expression Patterns journal homepage: http://www.elsevier.com/locate/gep
Lhx9 gene expression during early limb development in mice requires the FGF signalling pathway Yisheng Yang, Megan J. Wilson* Developmental Biology and Genomics Laboratory, Department of Anatomy, Otago School of Medical Sciences, University of Otago, P.O. Box 56, Dunedin 9054, New Zealand
a r t i c l e i n f o
a b s t r a c t
Article history: Received 28 May 2015 Received in revised form 11 July 2015 Accepted 21 July 2015 Available online xxx
Lhx9 is a member of the LIM-homeodomain gene family necessary for the correct development of many organs including gonads, limbs, heart and the nervous system. In the context of limb development, Lhx9 has been implicated as an integrator for Fibroblast growth factor (FGF) and Sonic hedgehog (Shh) signalling required for proximaledistal (PD) and anterioreposterior (AP) development of the limb. Three splice variants of the Lhx9 transcript are expressed during development, two of which are predicted to act in a dominant negative fashion, competing with the DNA binding version of Lhx9 for binding to cofactors via the LIM-domain. We examined the expression pattern for the three alternative splice forms of Lhx9; Lhx9a, Lhx9b and Lhx9c during early limb development. We have found that of the three Lhx9 isoforms, only Lhx9a and Lhx9c (intact homeodomain) are expressed during early limb development, each with their own distinct expression pattern. Additionally we determined that Lhx9 expression overlaps with FGF10 expression in the developing limb bud mesenchyme. Limb bud explant cultures, in the presence of signalling pathway inhibitors, also indicated that Lhx9 mRNA expression in the limb bud was dependent on FGF signalling. © 2015 Published by Elsevier B.V.
Keywords: LIM-homeodomain Lhx9 Isoforms Limb development Regulation
1. Introduction LIM-homeodomain gene 9 (Lhx9) is a member of the Apterous group in the LIM-homeodomain (LIM-HD) family of evolutionary conserved transcription factors crucial for development. Lhx9 is expressed in a number of areas such as the gonads (Birk and et al., 2000), limbs (Retaux et al., 1999), heart (Smagulova et al., 2008) and many regions of the nervous system including the spinal cord neurons, diencephalon, telencephalic vesicles, dorsal mesencephalon, hypothalamus and pineal gland (Bertuzzi and et al., 1999; Wilson et al., 2008; Yamazaki and et al., 2014). Generation of murine Lhx9 mutant lines indicated that Lhx9 expression is essential for some aspects of neuronal development and absolutely essential for the normal formation of the bipotential gonad (Birk and et al., 2000; Wilson et al., 2008). Gross deficits in mouse gonad development are caused by the loss of only a single Lhx9 allele, other aspects of development remained unchanged (Birk and et al., 2000; Wilson et al., 2008). This appears to be due to be a result of functional redundancy with Lhx9 and the paralogue of Lhx2 (Wilson
* Corresponding author. . E-mail address: [email protected] (M.J. Wilson).
et al., 2008; Tzchori and et al., 2009) as both genes show marked overlap in expressed regions during development. Together, Lhx2 and Lhx9 make up the mammalian Apterous group of LIM-HD transcription factors and both have been implicated in human development and disease (Hou and et al., 2013; Kuzmanov and et al., 2014; Vladimirova and et al., 2009). Lhx9 and Lhx2 are required for proximaledistal (PD) and anterioreposterior (AP) limb development as a double knockout mouse strain for both genes displays gross defects along these axes; such as shortened zeugopod (radius and ulna/tibia and fibula) and autopod (carpals/tarsals, metacarpals/metatarsals and digits) regions, with only two or three digits formed at 15.5 days post coitum (dpc) (Tzchori and et al., 2009). Tzchori and colleagues deduced that Lhx9 and Lhx2, together with their cofactor, Ldb1 integrate the signalling interactions for both PD and AP outgrowth and patterning through mediating FGF10 and Shh expression in response to FGF8 signalling, and Grem1 expression in response to Shh signalling. This proposed function of Lhx9 in the context of limb development is comparable to its Drosophila homologue, apterous (ap), that is critical for wing development (Cohen et al., 1992). The ap gene is required for the specification of dorsal cell fates, acting as a signalling centre at
http://dx.doi.org/10.1016/j.gep.2015.07.002 1567-133X/© 2015 Published by Elsevier B.V.
Please cite this article in press as: Yang, Y., Wilson, M.J., Lhx9 gene expression during early limb development in mice requires the FGF signalling pathway, Gene Expression Patterns (2015), http://dx.doi.org/10.1016/j.gep.2015.07.002
2
Y. Yang, M.J. Wilson / Gene Expression Patterns xxx (2015) 1e7
the boundary between dorsal and central cell to direct limb outgrowth (Diaz-Benjumea and Cohen, 1993; Ng et al., 1996). However, whether FGF or Shh signalling may control Lhx9 or Lhx2 expression in the early phases of limb development was not determined. Additionally, it was previously suggested that Lhx2/9 together with Ldb1 may act to maintain the limb mesenchymal cells in a progenitor state in response to FGF and Shh signals from the apical ectodermal ridge (AER) (Tzchori and et al., 2009), lending support to the notion that FGF and/or Shh signals may influence Lhx2/9 expression. Evidence from Xenopus laevis has shown that FGF15 is able to up-regulate expression of the Xenopus orthologue for Lhx9 (XLhx9) expression during nervous system development (Atkinson-Leadbeater et al., 2009). Additionally, while ectopic Shh expression during chick brain development was shown to decrease Lhx9 expression (Sun et al., 2008), the effect FGF and Shh signalling has on Lhx9 expression have not been investigated in regards to the mouse model or limb development. Lhx9 has three different protein isoforms derived from mRNA transcript variants (Smagulova et al., 2008; Failli et al., 2000). Of the three isoforms, transcript variant 3 (Lhx9 isoform c, also called Lhx9-HD) is the only one to retain a full HD domain and thus directly bind to DNA. Whereas transcript variant 1 (Lhx9 isoform a) and 2 (Lhx9 isoform b) have truncated HD domains, along with different 50 and 30 untranslated regions but intact LIM domains, making them LIM-only factors (LMOs) as they are predicted not to bind to DNA. However Lhx9a/b proteins are still able to interact with protein partners or cofactors (Molle et al., 2004) and it has been proposed that the LMOs could act competitively against Lhx9-HD for co-factors (Failli et al., 2000; Molle et al., 2004), given the LIM domain is the proteineprotein interaction motif to interact with co-factors (Agulnick and et al., 1996; Jurata et al., 1996). Interestingly, Lhx9a protein distribution is limited to the nuclear compartment, as opposed to Lhx9c, which located in both the cytoplasm and the nucleus (Molle et al., 2004). In addition, Lhx9a/b is the major splice form expressed during mouse heart development, as opposed to Lhx9c, which is not expressed at all (Smagulova et al., 2008). This indicates that there are differences in the regulation of Lhx9 isoform expression, in particular Lhx9a may not just be a regulator of Lhx9c through cofactor binding, but may have its own role too. To date, the spatial expression of the Lhx9 isoforms during early limb development has not been investigated. Here we examine the expression patterns of Lhx9 splice isoforms during early limb development and expand on the relationship between Lhx9 and the signalling pathways critical for early limb development. Lhx9c and Lhx9a mRNA transcript variants expression show a distinct pattern of expression from one another but are still located predominantly towards the anterior region of limb buds. We find that Lhx9 protein is associated with proliferative cells in the progress zone, in addition to Lhx9 mRNA expression overlaps markedly with the FGF10 expression. We further show that Lhx9 is dependent on FGF signalling for its expression. 2. Materials and methods 2.1. Embryo collection Embryos at stages 10.5 and 11.5 dpc were collected from timed mated inbred BALB/c mice. Whole embryos used for in situ hybridisation were fixed overnight in 4% paraformaldehyde (PFA) and dehydrated in methanol prior to use. For explant culture, forelimb and hindlimb buds from 10.5 dpc embryos were dissected in Leibovitz's L-15 media (Life Technologies) immediately prior to cell culture. All animal work was approved by the University of
Otago Animal Ethics Committee (ET 18/11). 2.2. RNA probe design and in situ hybridisation Primers were designed to target each of the three described Lhx9 isoforms; a b and c (HD) based on mRNA sequences available € in GenBank (accession numbers NM_001025565.2, NM_010714.3 and NM_001042577.1 respectively). Primers used are described in Table 1. Complimentary DNA (cDNA) derived from total mouse RNA was used as template for PCR amplification of each target transcript. Amplified products were cloned into pBluescript II KS (±) vector (Fermentas Molecular Biology Tools) and sequenced prior to in vitro transcription. DIG-labelled probe synthesis and in situ hybridisation were performed as described in Hargrave et al. (2006). A RNA probe to target the coding regions of the Lhx9 mRNA sequence (referred to as Lhx9E1-5 from this point forward), cloned in to (pGEM®-T) was gifted to us from Prof. Peter Koopman from the Institute of Molecular Bioscience, University of Queensland. The colour reaction was performed with nitroblue tetrazolium chloride (NBT) and 5-bromo-4-chloro-30 -indolyphosphate p-toluidine salt (BCIP) (Roche). Stained embryos were fixed with 4% paraformaldehyde (PFA) and mount in embryo wells with 100% glycerol. Images were taken using Olympus SZX10 bright field microscope with Olympus DP digital camera and software. 2.3. Immunohistochemistry Intact 10.5 dpc embryos were fixed overnight in 4% PFA, dehydrated with methanol before being paraffin embedded and sectioned to a thickness of 12 mm. Sections were deparaffinised and rehydrated with 100% xylene (2, 3 min each) and ethanol water dilutions (2 100%, 90%, 70%, 50%, 3 min each). Antigen retrieval was performed by boiling samples for a total of 30 min, immersed in sodium citrate (10 mM, 0.05% Tween 20, pH 6.0) using a conventional microwave oven on high power. Immunohistochemistry detection for Lhx9, E-cadherin (E-cad) and Ki-67 was achieved using indirect horseradish peroxidase (HRP)-linked secondary antibody. Slides were washes three times with PBTx (PBS containing 0.025% Triton X-100) and blocked with 10% heat inactivated donkey serum and 1% bovine serum albumin in PBS. Slides were incubated overnight at 4 C with the primary antibody diluted in blocking solution. After three washes with PBTx, slides were incubated with 3% H2O2 for 5 min, and washed twice with PBTx. Slides were then incubated with secondary antibody diluted in blocking solution at room temperature for 2 h, washed, mounted and imaged with an Olympus AX70 light microscope. Antibodies used were: rabbit polyclonal anti-Lhx9 (Abcam Ab28737, 1: 100 dilution), rabbit polyclonal anti-E-cadherin (Cell Signalling Technologies, 1: 400 dilution), rabbit polyclonal anti-Ki67 (Novus Biologicals, 1: 50
Table 1 Oligonucleotide primers used in this study. Target
Forward primer 50 e 30
Reverse primer 50 e 30
Lhx9a Lhx9b Lhx9c FGF8 FGF10 Shh Lhx9 qPCR Gapdh qPCR
TCTGGCCACCTCCTCTCT AA ACGGTAGCTTTGCTTGTAGGA GGGGGTGTTGATAAAGCTGA GGGACACGAGGACCGACCCTT CCCGCTGACCTTGCCGTTCTT CAAGGAGGAGCGCACACGCA CGTCTCTACGCTTCTGCATC CATGGCCTTCCGTGTTCCTA
TCCTTGCAGTAAATGCTACCG AAATCTATCGAGCAGGAGGGT CGCCCATCGTAATACCATTT CTCCGGGGCCCAAGTCCTCT CCCGCTGACCTTGCCGTTCTT ACGCGGTCTCCGGGACGTAA GGCGGAAAGGACACGAAT CCTGCTTCACCACCTTCTTGAT
Please cite this article in press as: Yang, Y., Wilson, M.J., Lhx9 gene expression during early limb development in mice requires the FGF signalling pathway, Gene Expression Patterns (2015), http://dx.doi.org/10.1016/j.gep.2015.07.002
Y. Yang, M.J. Wilson / Gene Expression Patterns xxx (2015) 1e7
dilution) and goat anti-rabbit HRP conjugated secondary (Santa Cruz, 1 in 400 dilution).
2.4. Explant tissue culture Embryos were dissected from timed-mated female mice in Leibovitz's L-15 media (Life Technologies). Excised limb buds were transferred to DMEM-agar mould consisting of DMEM (Life Technologies), 10% fetal bovine serum (Peribio Science), 1% L-glutamine (Thermo Scientific), 0.1% ampicillin (50 mg/mL, Thermo Scientific) and 1.5% agar. Moulds were made by setting mould mixture on top of riveted stainless steel discs specified previously (Capel and Batchvarov, 2008). Limb buds placed in DMEM-agar moulds were cultured for 23e24 h (37 C, 5% CO2) in DMEM culture media (DMEM, 10% fetal bovine serum, 1% L-glutamine and 0.1% ampicillin) containing 100 nM of either PD-166285 hydrate (Sigma), or cyclopamine hydrate (Sigma), or vehicle controls consisting of DMSO or 100% ethanol as matched to controls for the respective inhibitors. Following culture, limb buds were either fixed with 4% PFA overnight and dehydrated in methanol for in situ hybridisation, or washed in PBS prior to RNA extraction using PureLink® RNA Mini kit (life Technologies). A total of three limb buds per culture, per treatment were pooled for RNA extraction.
3
2.5. Reverse transcriptase quantitative PCR (RT-qPCR) Following RNA extraction, limb bud RNA samples were treated with DNAseI prior to conversion to cDNA using iScript™ cDNA Synthesis kit (Bio-Rad). Quantitative PCR (qPCR) was performed using specific oligonucleotides for Lhx9 (that would amplify all splice forms of Lhx9) and a reference control gene, Gapdh using SYBR select mastermix (Life Technologies). Sequences for oligonucleotide primers used for RT-qPCR are listed in Table 1. Two microliters of cDNA was used for each reaction. qPCR assays were performed on an Alignment Mx3000P system. A total of three technical replicates were performed for each cDNA sample along with two biological replicates. Each biological replicate was a pooled sample of six limb buds. To determine if differences between controls and inhibitor treated samples were statistically significant, a non-parametric ManneWhitney U test was performed.
3. Results and discussion 3.1. Two Lhx9 transcript variants are expressed in the E11.5 mouse limb Previous studies have shown that Lhx9 mRNA to be strongly
Fig. 1. In situ hybridisation to detect expression of Lhx9 transcript variants during early limb development (A) Genome browser image of the Lhx9 gene locus including the exonintron structure for the Lhx9 alternative splice variants. MultiZ alignment indicating the orthologous sequences in the named vertebrate genomes. Exons 1 to 5 of the Lhx9c splice form are highly conserved throughout vertebrates, exon 5 encodes for the DNA binding domain. Transcript variant named Lhx9b (NM_010714.3) differs in both first and last exons to that of Lhx9c. Whereas Lhx9a, (NM_001025565.2) only differs with respect to the final exon. Coloured boxes indicate the exon regions used to design probes targeted to specific Lhx9 variant mRNAs. (BeE) Whole-mount staining of 11.5 dpc fore- (left) and hind- (right) limb buds with probes specifying the different transcript variants of Lhx9. (B) Lhx9E1-5 staining in the distal regions of the limb bud extending from the anterior most limb margin to the posterior but was not detected near the body wall on the posterior side of the limb bud (asterisk). (C) Weak staining for Lhx9c was distributed throughout the entire limb bud. (D) Lhx9a/b staining is concentrated to the anterior of the limb bud only. As the Lhx9bspecific probe (E) produces no significant staining in either the limb bud signal observed with Lhx9a/b (D) is likely to be due to Lhx9a expression. Scale bars represent 200 mm. Abbreviations: Forelimb (FL), Hindlimb (HL).
Please cite this article in press as: Yang, Y., Wilson, M.J., Lhx9 gene expression during early limb development in mice requires the FGF signalling pathway, Gene Expression Patterns (2015), http://dx.doi.org/10.1016/j.gep.2015.07.002
4
Y. Yang, M.J. Wilson / Gene Expression Patterns xxx (2015) 1e7
expressed in both fore and hindlimb buds (Retaux et al., 1999; Tzchori and et al., 2009). Initially we used an in situ hybridisation probe against the full-length coding region of Lhx9 exons 1 to 5 (Lhx9E1-5; Fig. 1A). At 11.5 dpc, the Lhx9E1-5 probe detected expression in the distal regions of both the fore and hindlimb (Fig. 1B). We also found that posterior staining for Lhx9E1-5 did not completely extend back towards the body wall in both limbs (Fig. 1B, asterisks). While the Lhx9E-15 probe includes exon 5 encoding for an intact C-terminus HD domain, it also substantially overlaps with the other two Lhx9 transcript splice forms (exons 1e4) and there is likely to hybridise to all splice forms of Lhx9 mRNA (Fig. 1A). To further distinguish between the three major Lhx9 transcripts, we amplified separate regions of mRNA from cDNA; Lhx9c probe was specific for the exon and 30 untranslated region (UTR) for the intact HD domain (Fig. 1A, exon shaded grey). Lhx9a/b probe specified an alternative 30 end of the transcript, exon and 30 UTR (Fig. 1A, exon shaded green). To distinguish between Lhx9a and Lhx9b transcripts a third probe Lhx9b was used, that is predicted to hybridise to 50 UTR and exon specific for the Lhx9b transcript (Fig. 1A, exons shaded red). Staining with the Lhx9E1-5 probe produced strong staining at the anterior and distal regions of the both fore and hindlimb buds (Fig. 1B) but did not extend to the posterior most limb margin (Fig. 1B, asterisk). The Lhx9c probe targeted to the region of Lhx9 mRNA encoding for the HD domain, produced weak staining
throughout the limb bud (Fig. 1C). Staining from the Lhx9c probe was more diffuse compared to Lhx9E1-5 (Fig. 1BeC). Lhx9a expression was detected at the distal anterior regions of both limb buds (Fig. 1D), unlike Lhx9c mRNA, which was detected throughout the entire limb buds (Fig. 1C). In situ hybridisation with a RNA probe targeting the Lhx9b transcript did not detect any significant expression within 11.5 dpc limb buds (Fig. 1E), suggesting with the Lhx9a/b probe may reflect expression of the Lhx9a isoform only. Collectively this shows that while there was some overlap in the expression patterns of Lhx9 transcripts; Lhx9a and Lhx9c also have their own distinct expression domains. As Lhx9c and Lhx9a transcripts, but not the Lhx9b-specific transcript, were detected, these two splice forms are likely to be the dominant transcript variants expressed at this stage of limb development. This could be a result the functional role that Lhx9 plays during limb development or following on from a previous finding where it was shown that Lhx9a and b isoforms were expressed during heart development but not Lhx9c (Smagulova et al., 2008), highlighting differential regulatory control over the isoforms. Given previous suggestions of a competitive role that Lhx9a might have over Lhx9c by way of cofactor binding competition (Failli et al., 2000; Molle et al., 2004), this may explain the more anteriorly located expression of Lhx9a which might serve as a spatial controller of Lhx9 function over the AP axis, potentially focusing Lhx9 function distally. As Lhx9E1-5 and Lhx9a transcripts were expressed proximally
Fig. 2. Immunohistochemistry of fore- and hindlimb sections from 11.5 dpc mouse embryos. (A) Lhx9 protein was localised predominantly to the anterior and distal regions in the limb buds. (B) E-cadherin marks the epithelial covering and AER of the limb bud. (C) Ki67 staining was detected in cells that are widely distributed through the limb buds, particularly in the distal most regions of the mesenchyme. Scale bars represent 200 mm.
Please cite this article in press as: Yang, Y., Wilson, M.J., Lhx9 gene expression during early limb development in mice requires the FGF signalling pathway, Gene Expression Patterns (2015), http://dx.doi.org/10.1016/j.gep.2015.07.002
Y. Yang, M.J. Wilson / Gene Expression Patterns xxx (2015) 1e7
5
Fig. 3. Comparative whole-mount in situ hybridisation for known signalling centres with Lhx9 expression in the forelimb bud at 10.5 dpc. (A) Lhx9E1-5 mRNA is expressed in the anterior and distal regions of the limb bud. (B) FGF10 mRNA is expressed within the progress zone, in the limb mesenchyme. (C) FGF8 staining marks the AER located at the distal end of the limb bud. (D) Shh mRNA is detected in the ZPA, located at the posterior margin. Scale bars represent 100 mm.
underneath the AER, we investigated whether Lhx9 protein overlapped with proliferative mesenchymal cells in the area under the AER known as the progress zone (Summerbell and Lewis, 1975). Immunohistochemistry was used to detect Lhx9 (LHX9) protein
using 11.5 dpc limb bud sections. The antibody used was raised against the C-terminus polypeptides 289e338 of the human LHX9 protein and is predicted to recognise both isoforms of Lhx9 (with and without the homeodomain). Lhx9 protein was detected in the
Fig. 4. Lhx9 expression in limb buds following in vitro inhibition of FGF and Shh signalling pathways. (AeD) In situ hybridisation to detect changes to expression of Lhx9 following in vitro culture in the presence of either the FGF inhibitor (C), Shh inhibitor (D) or their respective vehicle controls (A, B). Lhx9E1-5 mRNA expression was no longer detectable via in situ hybridisation following culture with the FGF signalling pathway inhibitor. Whereas, strong Lhx9 signal was still detectable following culturing with the Shh signalling pathway inhibitor cyclopamine (D). Quantification of Lhx9 expression was performed using RT-qPCR (E) and revealed a significant decrease in Lhx9 expression after FGF signalling inhibition (P ¼ 0.0079; ManneWhitney U test), but Lhx9 mRNA levels were not significantly altered after Shh signalling inhibition. Scale bars represent 200 mm.
Please cite this article in press as: Yang, Y., Wilson, M.J., Lhx9 gene expression during early limb development in mice requires the FGF signalling pathway, Gene Expression Patterns (2015), http://dx.doi.org/10.1016/j.gep.2015.07.002
6
Y. Yang, M.J. Wilson / Gene Expression Patterns xxx (2015) 1e7
nucleus and enriched in an asymmetrical pattern towards the anterior mesenchyme compared to the posterior (Fig. 2A). Lhx9 protein was limited to the mesenchymal cells only and was not found in the outer epithelial layer including the AER, highlighted by the absence of staining when compared to E-cadherin (Fig. 2B). In comparison to Ki-67 (Fig. 2C), a marker of proliferative cells (Scholzen and Gerdes, 2000), Lhx9 staining showed marked overlap (Fig. 2A and C), indicating that many of the proliferative cells in the early limb bud also co-express Lhx9 protein. 3.2. Lhx9 in the early limb bud expression overlaps with FGF10 With Lhx9 expression showing a very distinct pattern in the anterior and distal regions, this suggested that there is spatial regulation of expression. As Lhx9E1-E5 gene expression is strongest within the anterior most regions of the limb bud, expanding distally and down the posterior regions but not completely to the posterior limb bud boundary at 10.5 dpc (Fig. 3A). This expression pattern prompted us to investigate Lhx9 expression with respect to both FGF and Shh signalling pathways, as they are the key signalling factors acting as morphogens in limb development (Riddle et al., 1993; LopezMartinez and et al., 1995; Sun et al., 2002; Mariani et al., 2008), to establish spatial gene expression patterns required for correct patterning of the limb. Previously, Lhx9 (along with Lhx2) was shown to be required for PD and AP limb development by integrating FGF and Shh signals (Tzchori and et al., 2009), however, the possibility of Lhx9 expression being dependent on either of the signals was not determined, but could be regulated by one or both signals resulting in the asymmetrical expression domains. Given that expression of the DNA binding form of Lhx9 (Lhx9E1-5) was reduced within the posterior most border near the region of the zone of polarizing activity (ZPA), we compared Lhx9 expression to that of Shh (Fig. 3). Shh expression in the ZPA is required for correct patterning of the anterioreposterior axes of the bud (Riddle et al., 1993), acting as a morphogen to control limb development with high to low concentrations across the AP axis. The absence of expression of Lhx9 in the ZPA could indicate that high levels of Shh signal may inhibit Lhx9 expression in this region at 10.5 dpc. FGF10 and FGF8 mRNA was detected within the progress zone (Fig. 3B) and AER respectively (Fig. 3C) In comparison with Lhx9 expression, FGF10 mRNA overlaps with Lhx9 mRNA expressing cells in the progress zone in the anterior and proximal regions of the limb bud (Fig. 3A and B). FGF8 transcript was found in the AER, located beside the limb bud region of strongest Lhx9 mRNA detection (Fig. 3A and C). Unlike the FGF factors, Shh mRNA did not overlap with expression of the full length Lhx9 transcript (Fig. 3D). In order to determine if FGF and Shh signalling could influence Lhx9 limb expression, we performed explant limb bud culture in the presence of specific FGF and Shh small molecular signalling inhibitors, PD-116285 and cyclopamine hydrate respectively. In situ hybridisation for the Lhx9 transcript in post cultured limb buds showed that Lhx9 mRNA was no longer able to be detected in the absence of FGF signals (Fig. 4A and C), indicating loss of both Lhx9c and Lhx9a isoforms as the probe detects for both variants (Fig. 1A; Lhx9E1-5). As in situ hybridisation is not quantitative Lhx9 gene expression levels were also determined by RT-quantitative PCR (RTqPCR) (Fig. 4E). Incubation of limb bud tissue in the presence of the FGF inhibitor resulted in a significant reduction of Lhx9 mRNA levels by 80% relative to the vehicle control. In situ hybridisation of Lhx9 mRNA on limb buds incubated in the presence of cyclopamine still detected high levels of Lhx9 expression (Fig. 4B and D). RT-qPCR did not detect any significant differences in Lhx9 expression levels between vehicle and Shh inhibitor samples (Fig. 4E). This suggests
Fig. 5. Model of Lhx9 regulation in the early limb bud. FGF10 (green) is produced by cells within the limb mesenchyme and, along with FGF8 (blue) secretion from the AER, acts to maintain outgrowth of the developing limb (Mariani et al., 2008). Lhx9 expression (shaded grey) is positively regulated by FGF signalling and is likely to be part of a gene network involving both FGF10 and FGF8. Shh is produced in the ZPA (red), acts as a morphogen to regulate correct anterioreposterior patterning during limb development. Our results show that Shh signalling has no effect on Lhx9 expression however earlier work indicated that Lhx9 is required for correct Shh expression (Tzchori and et al., 2009).
that FGF signalling is required for correct Lhx9 expression (Fig. 5) in the early limb bud, whereas Shh play no significant role in controlling the expression pattern of Lhx9. These results are consistent with previous evidence indicating FGF-dependent regulation of XLhx9 seen during Xenopus laevis nervous system development (Atkinson-Leadbeater et al., 2009). However, as we used a broad-spectrum tyrosine kinase inhibitor, PD-166285 (Panek and et al., 1997) to reduce FGF signalling in vitro, we are unable to pin point the exact FGF signalling factor or factors that are directly responsible for inducing Lhx9 expression. Furthermore, we cannot rule out the possibility that this FGF regulation maybe indirectly acting upon Lhx9 gene expression, as there is no evidence of the FGF signalling cascade being able to directly regulate Lhx9 expression. However, it is likely that the FGF signalling does play a role in the induction of Lhx9 given the given the marked overlap with FGF10 expression in the limb bud mesenchyme (Fig. 3). Additionally, should Lhx9 expression be under the regulation of FGF8/10, it would fit in with previous suggestions that Lhx2, Lhx9 and Ldb1 mediate the positive feedback loop between FGF8 and FGF10 (Tzchori and et al., 2009). With regards to the relationship between Lhx9 and Shh, we found no effect on Lhx9 expression with the loss of Shh signalling, indicating the Shh signal pathway is not necessary for Lhx9 gene regulation. Previous work suggested that Lhx2, Lhx9 and Ldb1 are required for Shh signal transduction as there was a reduction of Shh mRNA expression in Lhx2eLhx9 double mutants (Tzchori and et al., 2009). Our data and this previous study would indicate that Lhx9 acts upstream of Shh in early limb patterning (model shown in Fig. 5). In summary, we showed that the dominant Lhx9 isoforms expressed during limb development are Lhx9c and Lhx9a, each with their own distinct expression domain in early limb development. Lhx9 expression overlapped markedly with FGF10 expression cells in the limb mesenchyme and was associated with proliferative cells of the progress zone. In vitro inhibition of FGF signalling significantly reduced Lhx9 expression implying that correct Lhx9 gene expression during limb development requires the FGF signalling pathway.
Please cite this article in press as: Yang, Y., Wilson, M.J., Lhx9 gene expression during early limb development in mice requires the FGF signalling pathway, Gene Expression Patterns (2015), http://dx.doi.org/10.1016/j.gep.2015.07.002
Y. Yang, M.J. Wilson / Gene Expression Patterns xxx (2015) 1e7
Acknowledgements We would like to thank Phoebe Heenan, Simon Blanchoud and Mitchell Weston for helpful comments on earlier versions of this manuscript. This project was supported by a University of Otago Research Grant to MJW. References Agulnick, A.D., et al., 1996. Interactions of the LIM-domain-binding factor Ldb1 with LIM homeodomain proteins. Nature 384, 270e272. http://dx.doi.org/10.1038/ 384270a0. Atkinson-Leadbeater, K., Bertolesi, G.E., Johnston, J.A., Hehr, C.L., McFarlane, S., 2009. FGF receptor dependent regulation of Lhx9 expression in the developing nervous system. Dev. Dyn. Off. Publ. Am. Assoc. Anat. 238, 367e375. http:// dx.doi.org/10.1002/dvdy.21850. Bertuzzi, S., et al., 1999. Characterization of Lhx9, a novel LIM/homeobox gene expressed by the pioneer neurons in the mouse cerebral cortex. Mech. Dev. 81, 193e198. Birk, O.S., et al., 2000. The LIM homeobox gene Lhx9 is essential for mouse gonad formation. Nature 403, 909e913. http://dx.doi.org/10.1038/35002622. Capel, B., Batchvarov, J., 2008. Preparing recombinant gonad organ cultures. CSH Protoc. http://dx.doi.org/10.1101/pdb.prot5078, 2008, pdb prot5078. Cohen, B., McGuffin, M.E., Pfeifle, C., Segal, D., Cohen, S.M., 1992. apterous, a gene required for imaginal disc development in Drosophila encodes a member of the LIM family of developmental regulatory proteins. Genes Dev. 6, 715e729. Diaz-Benjumea, F.J., Cohen, S.M., 1993. Interaction between dorsal and ventral cells in the imaginal disc directs wing development in Drosophila. Cell 75, 741e752. Failli, V., Rogard, M., Mattei, M.G., Vernier, P., Retaux, S., 2000. Lhx9 and Lhx9alpha LIM-homeodomain factors: genomic structure, expression patterns, chromosomal localization, and phylogenetic analysis. Genomics 64, 307e317. http:// dx.doi.org/10.1006/geno.2000.6123. Hargrave, M., Bowles, J., Koopman, P., 2006. In situ hybridization of whole-mount embryos. Methods Mol. Biol. 326, 103e113. http://dx.doi.org/10.1385/1-59745007-3:103. Hou, P.S., et al., 2013. LHX2 regulates the neural differentiation of human embryonic stem cells via transcriptional modulation of PAX6 and CER1. Nucleic Acids Res. 41, 7753e7770. http://dx.doi.org/10.1093/nar/gkt567. Jurata, L.W., Kenny, D.A., Gill, G.N., 1996. Nuclear LIM interactor, a rhombotin and LIM homeodomain interacting protein, is expressed early in neuronal development. Proc. Natl. Acad. Sci. U. S. A. 93, 11693e11698. Kuzmanov, A., et al., 2014. LIM-homeobox gene 2 promotes tumor growth and metastasis by inducing autocrine and paracrine PDGF-B signaling. Mol. Oncol. 8, 401e416. http://dx.doi.org/10.1016/j.molonc.2013.12.009.
7
Lopez-Martinez, A., et al., 1995. Limb-patterning activity and restricted posterior localization of the amino-terminal product of Sonic hedgehog cleavage. Curr. Biol. CB 5, 791e796. Mariani, F.V., Ahn, C.P., Martin, G.R., 2008. Genetic evidence that FGFs have an instructive role in limb proximal-distal patterning. Nature 453, 401e405. http:// dx.doi.org/10.1038/nature06876. Molle, B., Pere, S., Failli, V., Bach, I., Retaux, S., 2004. Lhx9 and lhx9alpha: differential biochemical properties and effects on neuronal differentiation. DNA Cell Biol. 23, 761e768. http://dx.doi.org/10.1089/1044549042531422. Ng, M., Diaz-Benjumea, F.J., Vincent, J.P., Wu, J., Cohen, S.M., 1996. Specification of the wing by localized expression of wingless protein. Nature 381, 316e318. http://dx.doi.org/10.1038/381316a0. Panek, R.L., et al., 1997. In vitro pharmacological characterization of PD 166285, a new nanomolar potent and broadly active protein tyrosine kinase inhibitor. J. Pharmacol. Exp. Ther. 283, 1433e1444. Retaux, S., Rogard, M., Bach, I., Failli, V., Besson, M.J., 1999. Lhx9: a novel LIMhomeodomain gene expressed in the developing forebrain. J. Neurosci. Off. J. Soc. Neurosci. 19, 783e793. Riddle, R.D., Johnson, R.L., Laufer, E., Tabin, C., 1993. Sonic hedgehog mediates the polarizing activity of the ZPA. Cell 75, 1401e1416. Scholzen, T., Gerdes, J., 2000. The Ki-67 protein: from the known and the unknown. J. Cell. Physiol. 182, 311e322. http://dx.doi.org/10.1002/(SICI)10974652(200003)182, 3<311::AID-JCP1>3.0.CO;2-9. Smagulova, F.O., Manuylov, N.L., Leach, L.L., Tevosian, S.G., 2008. GATA4/FOG2 transcriptional complex regulates Lhx9 gene expression in murine heart development. BMC Dev. Biol. 8, 67,. http://dx.doi.org/10.1186/1471-213X-8-67. Summerbell, D., Lewis, J.H., 1975. Time, place and positional value in the chick limbbud. J. Embryol. Exp. Morphol. 33, 621e643. Sun, X., Mariani, F.V., Martin, G.R., 2002. Functions of FGF signalling from the apical ectodermal ridge in limb development. Nature 418, 501e508. http://dx.doi.org/ 10.1038/nature00902. Sun, X., Saitsu, H., Shiota, K., Ishibashi, M., 2008. Expression dynamics of the LIMhomeobox genes, Lhx1 and Lhx9, in the diencephalon during chick development. Int. J. Dev. Biol. 52, 33e41. http://dx.doi.org/10.1387/ijdb.072386xs. Tzchori, I., et al., 2009. LIM homeobox transcription factors integrate signaling events that control three-dimensional limb patterning and growth. Development 136, 1375e1385. http://dx.doi.org/10.1242/dev.026476. Vladimirova, V., et al., 2009. Aberrant methylation and reduced expression of LHX9 in malignant gliomas of childhood. Neoplasia 11, 700e711. Wilson, S.I., Shafer, B., Lee, K.J., Dodd, J., 2008. A molecular program for contralateral trajectory: Rig-1 control by LIM homeodomain transcription factors. Neuron 59, 413e424. http://dx.doi.org/10.1016/j.neuron.2008.07.020. Yamazaki, F., et al., 2014. The Lhx9 homeobox gene controls pineal gland development and prevents postnatal hydrocephalus. Brain Struct. Funct. http:// dx.doi.org/10.1007/s00429-014-0740-x.
Please cite this article in press as: Yang, Y., Wilson, M.J., Lhx9 gene expression during early limb development in mice requires the FGF signalling pathway, Gene Expression Patterns (2015), http://dx.doi.org/10.1016/j.gep.2015.07.002