SOX9, a high mobility group (HMG) box transcription factor, plays critical roles during embryogenesis and its activity is required for development, differentiation and lineage commitment in various tissues including the intestinal epithelium. Here, we present functional and clinical data of a broadly important role for SOX9 in tumorigenesis. SOX9 was overexpressed in a wide range of human cancers, where its expression correlated with malignant character and progression. Gain of SOX9 copy number is detected in some primary colorectal cancers. SOX9 exhibited several pro-oncogenic properties, including the ability to promote proliferation, inhibit senescence and collaborate with other oncogenes in neoplastic transformation. In primary MEFs and colorectal cancer cells, SOX9 expression facilitated tumor growth and progression whilst its inactivation reduced tumorigenicity. Mechanistically, we have found that Sox9 directly binds and activates the promoter of the polycomb protein Bmi1, whose upregulation represses the tumor suppressor Ink4a/Arf locus. In agreement with this, human colorectal cancers showed a positive correlation between expression levels of SOX9 and BMI1 and a negative correlation between SOX9 and ARF in clinical samples. Taken together, our findings provide direct mechanistic evidence of the involvement of SOX9 in neoplastic pathobiology, particularly in colorectal cancer.

The development of the inner ear involves complex processes of morphological changes, patterning and cell fate specification that are under strict molecular control. SOX2 and SOX9 are SOX family transcription factors that are involved in the regulation of one or more of these processes. Previous findings have shown early expression of SOX9 in the otic placode and vesicle at E8.5–E9.5. Here we describe in detail, the expression pattern of SOX9 in the developing mouse inner ear beyond the otocyst stage and compare it with that of SOX2 from E9.5 to E18.5 using double fluorescence immunohistochemistry. We found that SOX9 was widely expressed in the otic epithelium, periotic mesenchyme and cartilaginous otic capsule. SOX2 persistently marked the prosensory and sensory epithelia. During the development of the sensory epithelia, SOX2 was initially expressed in all prosensory regions and later in both the supporting and hair cells up to E15.5, when its expression in hair cells gradually diminished. SOX9 expression overlapped with that of SOX2 in the prosensory and sensory region until E14.5 when its expression was restricted to supporting cells. This initial overlap but subsequent differential expression of SOX2 and SOX9 in the sensory epithelia, suggest that SOX2 and SOX9 may have distinct roles in molecular pathways that direct cells towards different cell fates.

The voltage-gated sodium channel NaV1.8 is expressed exclusively in nociceptive sensory neurons and plays an important role in pain pathways. NaV1.8 cannot be functionally expressed in non-neuronal cells even in the presence of β-subunits. We have previously identified Pdzd2, a multi PDZ-domain protein, as a potential interactor for NaV1.8. Here we report that Pdzd2 binds directly to the intracellular loops of NaV1.8 and NaV1.7. The endogenous NaV1.8 current in sensory neurons is inhibited by antisense- and siRNA-mediated downregulation of Pdzd2. However, no marked change in pain behaviours is observed in Pdzd2-decificent mice. This may be due to compensatory upregulation of p11, another regulatory factor for NaV1.8, in dorsal root ganglia of Pdzd2-deficient mice. These findings reveal that Pdzd2 and p11 play collaborative roles in regulation of NaV1.8 expression in sensory neurons.

To investigate the feasibility of conducting a genomic-scale protein labeling and localization study in Escherichia coli, a representative subset of 23 coding DNA sequences (CDSs) was selected for chromosomal tagging with one or more fluorescent protein genes (EGFP, EYFP, mRFP1, DsRed2). We used λ-Red recombination to precisely and efficiently position PCR-generated DNA targeting cassettes containing a fluorescent protein gene and an antibiotic resistance marker, at the C-termini of the CDSs of interest, creating in-frame fusions under the control of their native promoters. We incorporated cre/loxP and flpe/frt technology to enable multiple rounds of chromosomal tagging events to be performed sequentially with minimal disruption to the target locus, thus allowing sets of proteins to be co-localized within the cell. The visualization of labeled proteins in live E. coli cells using fluorescence microscopy revealed a striking variety of distributions including: membrane and nucleoid association, polar foci and diffuse cytoplasmic localization. Fifty of the fifty-two independent targeting experiments performed were successful, and 21 of the 23 selected CDSs could be fluorescently visualized. Our results show that E. coli has an organized and dynamic proteome, and demonstrate that this approach is applicable for tagging and (co-) localizing CDSs on a genome-wide scale.

Expression of the type II collagen gene (human COL2A1, mouse Col2a1) heralds the differentiation of chondrocytes. It is also expressed in progenitor cells of some nonchondrogenic tissues during embryogenesis. DNA sequences in the 5′ flanking region and intron 1 are known to control tissue-specific expression in vitro, but the regulation of COL2A1 expression in vivo is not clearly understood. We have tested the regulatory activity of DNA sequences from COL2A1 on the expression of a lacZ reporter gene in transgenic mice. We have found that type II collagen characteristic expression of the transgene requires the enhancer activity of a 309-bp fragment (+2,388 to +2,696) in intron 1 in conjunction with 6.1-kb 5′ sequences. Different regulatory elements were found in the 1.6-kb region (+701 to +2,387) of intron 1 which only needs 90-bp 5′ sequences for tissue-specific expression in different components of the developing cartilaginous skeleton. Distinct positive and negative regulatory elements act together to control tissue-specific transgene expression in the developing midbrain neuroepithelium. Positive elements affecting expression in the midbrain were found in the region from −90 to −1,500 and from +701 to +2,387, whereas negatively acting elements were detected in the regions from −1,500 to −6,100 and +2,388 to +2,855.

There are conflicting views on whether collagen X is a purely structural molecule, or regulates bone mineralization during endochondral ossification. Mutations in the human collagen α1(X) gene (COL10A1) in Schmid metaphyseal chondrodysplasia (SMCD) suggest a supportive role. But mouse collagen α1(X) gene (Col10a1) null mutants were previously reported to show no obvious phenotypic change. We have generated collagen X deficient mice, which shows that deficiency does have phenotypic consequences which partly resemble SMCD, such as abnormal trabecular bone architecture. In particular, the mutant mice develop coxa vara, a phenotypic change common in human SMCD. Other consequences of the mutation are reduction in thickness of growth plate resting zone and articular cartilage, altered bone content, and atypical distribution of matrix components within growth plate cartilage. We propose that collagen X plays a role in the normal distribution of matrix vesicles and proteoglycans within the growth plate matrix. Collagen X deficiency impacts on the supporting properties of the growth plate and the mineralization process, resulting in abnormal trabecular bone. This hypothesis would accommodate the previously conflicting views of the function of collagen X and of the molecular pathogenesis of SMCD.

Despite the high prevalence of intervertebral disc disease, little is known about changes in intervertebral disc cells and their regenerative potential with ageing and intervertebral disc degeneration. Here we identify populations of progenitor cells that are Tie2 positive (Tie2+) and disialoganglioside 2 positive (GD2+), in the nucleus pulposus from mice and humans. These cells form spheroid colonies that express type II collagen and aggrecan. They are clonally multipotent and differentiated into mesenchymal lineages and induced reorganization of nucleus pulposus tissue when transplanted into non-obese diabetic/severe combined immunodeficient mice. The frequency of Tie2+ cells in tissues from patients decreases markedly with age and degeneration of the intervertebral disc, suggesting exhaustion of their capacity for regeneration. However, progenitor cells (Tie2+GD2+) can be induced from their precursor cells (Tie2+GD2−) under simple culture conditions. Moreover, angiopoietin-1, a ligand of Tie2, is crucial for the survival of nucleus pulposus cells. Our results offer insights for regenerative therapy and a new diagnostic standard.

Back pain and sciatica are often caused by intervertebral disc degeneration. Sakai and colleagues identify a subset of nucleus pulposus progenitor cells from the intervertebral disc and show that loss of these progenitor cells correlates with ageing and intervertebral disc degeneration.

Cartilage and endochondral bone development require SOX9 activity to regulate chondrogenesis, chondrocyte proliferation, and transition to a non-mitotic hypertrophic state. The restricted and reciprocal expression of the collagen X gene, Col10a1, in hypertrophic chondrocytes and Sox9 in immature chondrocytes epitomise the precise spatiotemporal control of gene expression as chondrocytes progress through phases of differentiation, but how this is achieved is not clear. Here, we have identified a regulatory element upstream of Col10a1 that enhances its expression in hypertrophic chondrocytes in vivo. In immature chondrocytes, where Col10a1 is not expressed, SOX9 interacts with a conserved sequence within this element that is analogous to that within the intronic enhancer of the collagen II gene Col2a1, the known transactivation target of SOX9. By analysing a series of Col10a1 reporter genes in transgenic mice, we show that the SOX9 binding consensus in this element is required to repress expression of the transgene in non-hypertrophic chondrocytes. Forced ectopic Sox9 expression in hypertrophic chondrocytes in vitro and in mice resulted in down-regulation of Col10a1. Mutation of a binding consensus motif for GLI transcription factors, which are the effectors of Indian hedgehog signaling, close to the SOX9 site in the Col10a1 regulatory element, also derepressed transgene expression in non-hypertrophic chondrocytes. GLI2 and GLI3 bound to the Col10a1 regulatory element but not to the enhancer of Col2a1. In addition to Col10a1, paired SOX9–GLI binding motifs are present in the conserved non-coding regions of several genes that are preferentially expressed in hypertrophic chondrocytes and the occurrence of pairing is unlikely to be by chance. We propose a regulatory paradigm whereby direct concomitant positive and negative transcriptional control by SOX9 ensures differentiation phase-specific gene expression in chondrocytes. Discrimination between these opposing modes of transcriptional control by SOX9 may be mediated by cooperation with different partners such as GLI factors.

Author Summary

Chondrogenic differentiation is a key process in the formation of endochondral bone. Despite the wealth of information about gene expression patterns and signaling pathways important for this process, it is not clear how differentiation state-specificity of transcription is controlled. The transcription factor SOX9 regulates chondrocyte differentiation, proliferation, and entry into hypertrophy and is highly expressed in immature/proliferating chondrocytes. It directly transactivates Col2a1, enhancing this gene's expression in immature/proliferating chondrocytes. The Col10a1 gene is specifically expressed in hypertrophic chondrocytes in which Sox9 is downregulated. How is differentiation phase-specific transcription of genes controlled in chondrocytes, particularly during hypertrophy? We found that SOX9 directly represses Col10a1 expression in immature/proliferating chondrocytes of the growth plate, so that its expression is restricted to hypertrophic chondrocytes. Discrimination of this concomitant opposing transcriptional control may involve cooperation between SOX9 and different partners such as GLI factors (effectors of hedgehog signaling). SOX9 control of chondrocyte maturation therefore may be integrated with hedgehog signaling. Mutations in human SOX9 cause the skeletal malformation syndrome campomelic dysplasia, which is attributed to the disruption of the chondrogenic differentiation program because of failure to express SOX9 target genes. This interpretation should be revised to include inappropriate expression of genes normally repressed by SOX9.

Developmental pathways have been shown to be important in the initiation and progression of cancer in various tissues. We showed that the transcription factor SOX9 is expressed in the epithelia of the mouse embryonic prostate and is required for proper prostate development. We have performed an in vivo investigation into the role of SOX9 in prostate cancer in mouse and human. Studies on Pten and Nkx3.1 mutant mice show that cells with an increased level of SOX9 appear within the epithelia at the early stages of prostate neoplasia and this high expression correlates with all stages of neoplastic progression. Using genetically modified mice we show that overexpression of SOX9 in prostate epithelia leads to an increase in cell proliferation without inducing hyperplasia. In mice that were heterozygous for the conditional mutant allele of Pten, overexpression of SOX9 gave rise to an earlier induction of high-grade prostate intraepithelial neoplasia. Consistent with this role, loss of Sox9 in prostate epithelia led to a decrease in proliferating cells in normal and in homozygous Pten mutant mice with prostate neoplasia. Analysis of a cohort of 880 human prostate cancer samples showed that SOX9 expression is associated with increasing Gleason grades and higher Ki67 staining. These studies identify SOX9 as part of a developmental pathway that is reactivated in prostate neoplasia where it is involved in regulating proliferation and suggests it can contribute to carcinogenesis in specific genetic contexts.

In protein folding and secretion disorders, activation of endoplasmic reticulum (ER) stress signaling (ERSS) protects cells, alleviating stress that would otherwise trigger apoptosis. Whether the stress-surviving cells resume normal function is not known. We studied the in vivo impact of ER stress in terminally differentiating hypertrophic chondrocytes (HCs) during endochondral bone formation. In transgenic mice expressing mutant collagen X as a consequence of a 13-base pair deletion in Col10a1 (13del), misfolded α1(X) chains accumulate in HCs and elicit ERSS. Histological and gene expression analyses showed that these chondrocytes survived ER stress, but terminal differentiation is interrupted, and endochondral bone formation is delayed, producing a chondrodysplasia phenotype. This altered differentiation involves cell-cycle re-entry, the re-expression of genes characteristic of a prehypertrophic-like state, and is cell-autonomous. Concomitantly, expression of Col10a1 and 13del mRNAs are reduced, and ER stress is alleviated. ERSS, abnormal chondrocyte differentiation, and altered growth plate architecture also occur in mice expressing mutant collagen II and aggrecan. Alteration of the differentiation program in chondrocytes expressing unfolded or misfolded proteins may be part of an adaptive response that facilitates survival and recovery from the ensuing ER stress. However, the altered differentiation disrupts the highly coordinated events of endochondral ossification culminating in chondrodysplasia.

Author Summary

The assembly and folding of secreted proteins in the endoplasmic reticulum (ER) is exquisitely regulated by a complex mechanism that maintains an equilibrium between folded and unfolded proteins. Perturbation of this homeostasis induces ER stress, which, if not alleviated through ER stress signaling (ERSS), ultimately triggers cell death. Normal bone growth occurs through a highly coordinated differentiation program that yields specialized cartilage cells (chondrocytes); when this program is disrupted, chondrodysplasia, or malformed skeletons, can result. Chondrodysplasias caused by mutations that affect protein assembly and secretion are characterized by a disorganization of bony growth plates and distension of the ER. We tested whether these chondrodysplasia characteristics were linked to ERSS. By investigating the impact of ER stress on the cell fate of hypertrophic chondrocytes (HCs) in transgenic mice expressing mutations in collagen that prevent proper folding, we revealed a novel adaptive mechanism that helps alleviate the unfolded protein load. Instead of undergoing apoptosis, the HCs undergoing ER stress adapt, re-enter the cell cycle, and revert to a less-mature state in which expression of the mutant collagen is reduced. Our findings have broad implications for adaptive mechanisms to ER stress in vivo and for the pathophysiology underlying chondrodysplasias caused by mutations that impact on protein assembly and secretion.

When subjected to ER stress (by expression of misfolded or unfolded proteins), hypertrophic chondrocytes undergo alterations to their developmental program that may be part of an adaptive response.

Recombinogenic engineering methodology, also known as recombineering, utilizes homologous recombination to create targeted changes in cellular DNA with great specificity and flexibility. In Escherichia coli, the Red recombination system from bacteriophage lambda has been used successfully to modify both plasmid and chromosomal DNA in a highly efficient manner, using either a linear double-stranded DNA fragment or a synthetic single-stranded oligonucleotide (SSO). The current model for Red/SSO-mediated recombination involves the SSO first annealing to a transient, single-stranded region of DNA before being incorporated into the chromosome or plasmid target. It has been observed previously, in both eukaryotes and prokaryotes, that mutations in the two strands of the DNA double helix are ‘corrected’ by complementary SSOs with differing efficiencies. Here we investigate further the factors that influence the strand bias as well as the overall efficiency of Red/SSO-mediated recombination in E.coli. We show that the direction of DNA replication and the nature of the SSO-encoded mismatch are the main factors dictating the recombinational strand bias. However, the influence that the SSO-encoded mismatch exerts upon the recombinational strand bias is abolished in E.coli strains that are defective in mismatch repair (MMR). This reflects the fact that different base–base mispairs are corrected by the mutS/H/L-dependent MMR pathway with differing efficiencies. Furthermore, our data indicate that transcription has negligible influence on the strand bias. These results demonstrate for the first time that the interplay between DNA replication and MMR has a major effect on the efficiency and strand bias of Red/SSO-mediated recombination in E.coli.

SOX proteins bind similar DNA motifs through their high-mobility-group (HMG) domains, but their action is highly specific with respect to target genes and cell type. We investigated the mechanism of target selection by comparing SOX1/2/3, which activate δ-crystallin minimal enhancer DC5, with SOX9, which activates Col2a1 minimal enhancer COL2C2. These enhancers depend on both the SOX binding site and the binding site of a putative partner factor. The DC5 site was equally bound and bent by the HMG domains of SOX1/2 and SOX9. The activation domains of these SOX proteins mapped at the distal portions of the C-terminal domains were not cell specific and were independent of the partner factor. Chimeric proteins produced between SOX1 and SOX9 showed that to activate the DC5 enhancer, the C-terminal domain must be that of SOX1, although the HMG domains were replaceable. The SOX2-VP16 fusion protein, in which the activation domain of SOX2 was replaced by that of VP16, activated the DC5 enhancer still in a partner factor-dependent manner. The results argue that the proximal portion of the C-terminal domain of SOX1/2 specifically interacts with the partner factor, and this interaction determines the specificity of the SOX1/2 action. Essentially the same results were obtained in the converse experiments in which COL2C2 activation by SOX9 was analyzed, except that specificity of SOX9-partner factor interaction also involved the SOX9 HMG domain. The highly selective SOX-partner factor interactions presumably stabilize the DNA binding of the SOX proteins and provide the mechanism for regulatory target selection.