|Home | About | Journals | Submit | Contact Us | Français|
A full understanding of transcriptional regulation requires integration of information obtained from multiple experimental datasets. These include datasets annotating gene expression within the context of an entire organism under normal and genetically perturbed conditions. Here we describe an expression dataset annotating pigment cell-expressed genes of the developing melanocyte and RPE lineages. Expression images are annotated and available at http://research.nhgri.nih.gov/manuscripts/Loftus/March2009/. Data is also summarized in a standardized manner using a universal melanoblast scoring scale that accounts for the embryonic location of cells and regional cell density. This approach allowed us to classify 14 pigment genes into 4 groupings classified by cell lineage expression, temporal-spatial context, and differential alteration in response to altered MITF and SOX10 status. Significant differences in regional populations were also observed across inbred strain backgrounds highlighting the value of this approach to identify modifier allele influences on melanoblast number and distributions. This analysis revealed novel features of in vivo expression patterns that are not measurable by in vitro-based assays, providing data that in combination with genomic analyses will allow modeling of pigment cell gene expression in development and disease.
The complete sequencing of the human and mouse genomes has provided the raw material necessary to unravel the regulation of mammalian gene transcription during development. However, we are only beginning to learn to read the genome and understand the language of transcriptional regulation, given the tremendous complexity of factors that govern gene expression. In order to begin to decode the language of gene regulation we need to identify locus-specific, DNA-intrinsic factors, such as chromatin accessibility, methylation status, and primary DNA sequence that reflects the type, number and location of specific protein binding sites (Walhout, 2006). In addition we need to generate experimentally derived data sets of high quality that describe gene expression. These data sets should not only describe gene expression within the context of an isolated cell, but should also detail expression for a single gene across multiple tissues and within the context of an entire organism. An inherent challenge is merging both the experimental data with computational predictions, in a manner that accurately describes all biological observations for a given gene (Elnitski et al., 2006).
Pigment cells are an ideal cell type for developmental transcriptional network analyses because they are readily identified in vivo, they express a subset of genes that exhibit very restricted tissue expression, and initial data with respect to the transcriptional hierarchies of these pigment cell-expressed genes has already been collected (Goding, 2000; Hou and Pavan, 2008). Added to this are large datasets that identify locations of specific protein-DNA interactions in pigment cells (Lee et al., 2008; Ozsolak et al., 2007), and datasets generated through the integration of numerous microarray-based gene expression datasets that identify correlations in gene expression between pigment cell transcription factors and potential target genes (Hoek et al., 2006). Such datasets can be used to make predicted transcription factor networks (Hawkins and Ren, 2006). However, in most cases these will require subsequent analysis and experimental validation using in vivo models, in order to annotate gene expression for defined tissues/cell types in a manner which will allow for evaluation of computationally-derived, predictive gene expression models. We here outline a systematic methodology applied to generate a gene expression dataset resource that incorporates temporal–spatial expression information and is pigment cell-lineage centric.
During embryonic development, pigment cells arise from two distinct cell lineages, the neural crest and the neural epithelium. Those that are derived from the neural crest, termed melanoblasts during development and melanocytes once differentiated, emigrate from the top of the neural tube and migrate laterally throughout the embryo, subsequently proliferating and migrating throughout the entire embryo until they reach their final destinations of the dermis/hair follicles, inner ear, Harderian gland, and iris and choroid of the eye. Neuroectoderm-derived pigment cells, which form the retinal pigmented epithelium (RPE) layer of the eye, develop from the optic cup, which arises from an out-pocketing of the presumptive forebrain region of the neural tube.
Both pigment cell lineages are characterized by a common set of genes expressed during development, many of which are known to be transcriptional targets of the “master” regulatory factor Microphthalmia-associated Transcription Factor (MITF) (Goding, 2000; Levy et al., 2006; Steingrimsson et al., 2004). Mitf mRNA expression is seen in both the developing RPE and the migratory melanoblast cell lineage (Nakayama et al., 1998; Nguyen and Arnheiter, 2000). The neuroectodermal lineage first expresses Mitf throughout the optic cup at E9.5, with Mitf expression later becoming limited to the developing RPE. Mitf-positive melanoblasts are first identifiable near the otocyst at E10.5, displaying a punctate pattern along the lateral sides of the murine embryo at E11.5 (Figure 1). Of note, the roles of MITF in regulating gene expression in the RPE and melanoblasts are similar but not identical (Nakayama et al., 1998).
Currently twenty-nine mutant alleles of Mitf have been identified in the mouse. These include both recessive and semi-dominant mutations, and display hypopigmentation that ranges from moderate spotting to complete absence of mature melanocytes (http://www.informatics.jax.org). The Mitf mutant allele Mitfmi is semidominant, with heterozygotes displaying reduced iris pigmentation and white spotting on the head, belly and tail, and homozygotes displaying microphthalmia, deafness, and a completely white coat that arise from ocular defects and a complete absence of dermal and stria vascularis melanocytes, respectively(Hodgkinson et al., 1993). Detailed analyses of the effect of the Mitfmi allele on RPE development at E13.5-14.5 have shown that the RPE of Mitfmi/mi embryos begins to develop, however it displays thickening due to cellular hypoproliferation (as does the RPE of additional Mitf mutant mutants) (Nakayama_98). Pigmentation of the RPE is never attained, and subsequently the dorsal RPE transdifferentiates into ectopic neural retina (Nakayama et al., 1998), (Bumsted and Barnstable, 2000). Consistent with this physiological characterization of the Mitfmi/mi eye, E11.5 Mitfmi/+ heterozygous embryos have visibly reduced RPE pigmentation and Mitfmi/mi homozygotes a complete lack of RPE pigmentation.
In the melanocyte lineage but not the RPE, Mitf itself is directly under the transcriptional regulation of the HMG-box transcription factor SOX10 (Bondurand et al., 2000a; Potterf et al., 2000; Verastegui et al., 2000) and mutations in both Mitf and Sox10 lead to white spotting in mice (Herbarth et al., 1998; Hodgkinson et al., 1993; Hughes et al., 1993; Southard-Smith et al., 1998) and hypopigmentation in human disease (Amiel et al., 1998; Pingault et al., 1998; Tassabehji et al., 1994). Sox10 is expressed in migratory melanoblasts, the otic vesicle, and sympathetic, enteric, cranial and dorsal root ganglia lineages (Pusch et al., 1998; Southard-Smith et al., 1998). In situ hybridization for endogenous Sox10 expression shows that Sox10-positive melanoblasts are first identifiable at E11.5 and that Sox10 is notably absent from RPE (Britsch et al., 2001; Southard-Smith et al., 1998). In addition to the critical functions that both Sox10 and Mitf perform during embryonic development, their signaling pathways have been implicated in cancer progression for melanoma, as MITF is amplified in 10–20% of melanoma metastases (Garraway et al., 2005), and mutations in either MITF or SOX10 have been observed in ~ 20% of melanoma lesions (Cronin et al., 2009). Thus the SOX10-MITF transcriptional pathways in melanocytes regulate genes that are fundamental to the processes of development, differentiation, and cell survival, having implications for disease states of the pigment cell lineage such as melanoma.
We here describe an expression dataset for 14 pigment cell-expressed genes of the developing melanocyte and RPE lineages that are known or putative downstream targets of MITF. We have established this expression data set as a resource to provide a common method of annotation in which to describe the complexities underlying differential gene expression within the context of pigment cell lineages. This methodology can be applied to other genes exhibiting expression in the pigment cell lineages, allowing for standardized description of pigment cell gene expression that will allow for controlled comparison of gene expression for analysis performed in multiple laboratories. Our results reveal novel differences of in vivo expression patterns that are not measurable by in vitro-based assays. Utilizing this approach, the data obtained from systematic description of gene expression with overlapping yet distinct developmental pigment cell expression patterns will allow the generation of developmental models that will facilitate gene identification and prediction of alterations in disease states such as melanoma.
Fourteen genes were selected for in vivo developmental expression analysis in mouse: Sox10, Mitf, endothelin receptor type B (Ednrb), kit oncogene (Kit), dopachrome tautomerase (Dct), tyrosinase related protein 1 (Tyrp1), tyrosinase (Tyr), solute carrier family 45, member 2 (Slc45a2), silver (Si), glycoprotein (transmembrane) nmb (Gpnmb), v-erb-b2 erythroblastic leukemia viral oncogene homolog 3 (avian) (Erbb3), melan-A (Mlana), transient receptor potential cation channel, subfamily M, member 1 (Trpm1/Mlsn) and RAB38, member RAS oncogene family (Rab38). These genes were selected because previous studies had shown expression in terminally differentiated melanocyte derived cell lines, that their expression patterns overlapped with those of the transcription factors Mitf and Sox10, that they were transcriptional targets of SOX10 or MITF when analyzed by in vitro assay systems, and/or that they exhibited reduced expression in mouse models harboring mutations of Sox10 and Mitf (Baxter and Pavan, 2002; Baxter and Pavan, 2003; Bentley et al., 1994; Britsch et al., 2001; Du and Fisher, 2002; Du et al., 2003; Hou et al., 2006; Keshet et al., 1991; Loftus et al., 2008; Loftus et al., 2002; Miller et al., 2004; Murisier et al., 2007; Sato-Jin et al., 2008; Steel et al., 1992; Tsujimura et al., 1996; Yasumoto et al., 1997). However, as multiple laboratories performed these studies at different stages of embryonic development and on different or undefined genetic backgrounds, it was difficult to develop models by direct comparisons among the reported gene expression patterns.
To enable a controlled comparison of gene expression patterns for this 14-gene set, we performed whole mount in situ hybridization for each gene at E11.5 on a C57BL/6J background (Supplemental Table 1). The developmental time point of E11.5 was selected as it captures the first time point when melanoblasts, prior to differentiation and melanin production, are actively migrating throughout the entire length of the embryo. Comparison of the gene expression patterns allowed grouping of the genes into lineage based categories as follows: Class 1: melanoblasts +, RPE + (Tyr, Dct, Gpnmb, Mitf, Slc45a2, Si); Class II: melanoblasts +, RPE − (Kit, Ednrb, Sox10, Erbb3), and Class III: melanoblasts -, RPE + (Mlsn, Tyrp1, Rab38, Mlana). The genes represented in Class II are then further stratified with respect to PNS expression: Class IIa: melanoblasts +, RPE −, PNS− (Kit), and Class IIb: melanoblasts +, RPE −, PNS+ (Ednrb, Sox10, Erbb3)(Figure 1). The scoring for each of the genes to be present or absent in early melanoblasts was based on the following criteria: (i) Replicable pattern of punctate, isolated cell populations; (ii) The location of the isolated cells occurs in regions consistent with the location of LacZ positive cells from tgDct-lacZ mice (Mackenzie et al., 1997); (iii) The expression is absent in these regions in melanoblast-deficient Mitfmi/mi embryos (detailed below).
Visual documentation of comprehensive melanoblast gene expression over an entire embryo can be difficult to capture in a single image, given that melanoblasts are small, independent cells found dispersed across multiple focal planes of a three-dimensional E11.5 embryo. Thus we have created a web accessible database (Supplemental Figure S1) with multiple images capturing the gene expression in the head, otic vesicle, trunk and tail regions independently, in addition to the a side views of embryos representing the whole mount in situ performed for the 14 genes, located at http://research.nhgri.nih.gov/manuscripts/Loftus/March2009/.
To summarize the image results and to represent the variation observed for the melanoblast in situ patterns in a standardized manner, we created a melanoblast scoring scale. This approach analyzes the gene expression pattern, and incorporates both the embryonic spatial location and the regional density of cells expressing a given gene. The embryo was divided into eight regions, and a 0–4 melanoblast density scale (MDS) score was used to assign a quantitative value to the density of melanoblasts expressing a given gene in each region (Figure 2). Individuals who were blinded to embryo genotype scored the left and right sides of each embryo independently. Scores were then averaged and converted to a MDS heatmap to facilitate summary and comparison (Figure 3). Of note, the MDS describes the observable melanoblast cell populations that can be detected by in situ hybridzation for a given gene. The MDS scoring is not based on whether the individual melanoblast cells observed were of lower or higher intensity, which might be affected by a number of factors including: variations in the expression level of a given gene, variations in hybridization due to probe sequences and/or variation of signal generation in the defined detection conditions.
For each of the eight embryonic regions defined at E11.5, we then compared the MDS scores for the lineage-defined Class I genes (Tyr, Dct, Gpnmb, Mitf, Slc45a2, and Si) and Class IIa gene (Kit) in E11.5 C57BL/6J embryos. We observed that MDS scores for multiple genes captured a similar trend in spatial melanoblast distribution across the embryo despite variation among these seven genes in the in situ signal intensity per melanoblast. For each of these genes the consistent trend was as follows: the highest MDS scores were found over the otocyst (region D) and at the area of the hind limb and distal to the tail (regions G and H respectively), while a reduced MDS score was consistently observed in the area of the fore limb (region E) and in the area of the trunk between the limbs (region F) (Figure 3, Supplemental Figure S2).
Given the known variation that occurs among inbred mouse strains with respect to variability in white spotting (Asher et al., 1996; Britsch et al., 2001; Pavan et al., 1995; Southard-Smith et al., 1999), we hypothesized that MDS scores may vary depending on allelic polymorphisms of genes segregating across the genetic background of the embryo analyzed. To test this, we performed standardized in situ hybridization of Si on E11.5 embryos of six inbred strains (C57BL/6J, DBA/2J, BALB/cJ, FVB/NJ, C3H/HeJ, and 129SvEv/Tac) (Supplemental Figure S3) and image results were summarized using the MDS (Figure 4, Table 1). The resulting in vivo expression patterns of Si-positive melanoblasts at E11.5 varied in a region-independent, yet strain-specific manner with several significant variations of MDS scores (Supplemental Table 2). For example, in FVB/NJ the embryos exhibited a uniform, dense population of melanoblasts across the trunk region (region F), which differed significantly (p < 0.001) from the sparse population of cells in this region observed in C57BL/6J and C3H/HeJ. However, when the melanoblast cell density in the rostral head (region A) was evaluated (figure 4A, B), it was the strains FVB/NJ, DBA/2J and C3H/HeJ that exhibited a reduction in cell density compared to the dense expression pattern observed in BALB/cJ (p < 0.001). Finally, analysis of 129SvEv/Tac embryos demonstrated a significant reduction in the number of melanoblasts in the distal tail (region H) in comparison to the strains C57BL/6J, DBA/2J, BALB/cJ, and FVB/NJ (p < 0.001). Expanding this approach to other inbred strains and specific crosses may allow the identification of polymorphic loci that may account for these differences.
In addition to variation of MDS score due to strain polymorphisms, we wanted to evaluate how the expression of early melanoblast marker genes is affected by variation in individual gene expression. Our initial analysis focused on embryos with functional variation in the transcription factors Sox10 or Mitf. Given the influence we observed of genetic background on MDS scoring, we used mice that were congenic for Sox10tm1Weg and Mitfmi on a C57BL/6J background. Gene expression patterns were evaluated for stage-matched, E11.5 littermates that were either wild type, heterozygous or homozygous for Sox10tm1Weg or Mitfmi. To facilitate visulization, the data is summarized using two MDS heatmaps: one sorted by embryo genotype (Figure 3) and one sorted by gene probe (Supplemental Figure S4).
Formatting of the summarized in situ data in this manner allowed comparisons across multiple genes and genotypes for the dataset, and revealed differences with respect to how distinct melanoblast populations and individual genes responded to alterations in Mitf and Sox10 levels. For example, evaluation of the genotype-sorted data (Figure 3) identified a unique population of Mitf-expressing melanoblasts present in both the homozygous Mitfmi/mi and Sox10tm1Weg/tm1Weg mice within the dorsal tail. Of the six other melanoblast-expressed Mitf target genes, only Gpnmb is also found expressed within this discrete population in Sox10tm1Weg/tm1Weg embryos; with no Gpnmb expression seen in these cells in Mitf mi/mi embryos.
Evaluation of the MDS dataset as a whole also allowed us to further characterize and subdivide the group of six Class I genes. Although all six genes in Class I present with a similar reduction in gene-positive melanoblast numbers is observed in Mitfmi/+embryos relative to wild-type, this is not what is observed for the same set of genes evaluated in context of Sox10tm1Weg/+ embryos. In Sox10tm1Weg/+ embryos, Mitf, Slc45A2 and Tyr exhibited a visible reduction in melanoblast numbers relative to wild-type C57BL/6J embryos, however this reduction is comparable to that which is also observed in Mitfmi/+embryos. However for Dct, the number of Dct-positive melanoblasts observed is in Sox10tm1Weg/+ embryos is less than the number of Dct+ melanoblasts obeserved in Mitfmi/+embryos and fewer than what is seen for the other Class I genes in Sox10tm1Weg/+ embryos. Distinct from both of the previous examples is the regulation of gene expression for both Gpmnb and Si. For Gpmnb and Si, the number of gene-positive melanoblasts observed in Sox10tm1Weg/+ embryos was higher than that observed for Dct, Slc45A2 and Tyr. For the Class IIa gene Kit, analysis of the MDS dataset further defines how transcriptional regulation of Kit is distinct from the group of the Class I genes. Kit is the sole gene to remain expressed in a population of punctate cells that is present along the entirety of the Mitfmi/mi embryos, while only a small, discrete number of Kit + cells could be observed over the hindlimb area in Sox10tm1Weg/tm1Weg embryos.
Although the RPE and melanoblasts are derived from distinct tissues, both are pigmented cell types in which MITF has been shown to play a fundamental role regulating gene expression. Therefore, we examined the effect of altered Mitf expression on the gene expression patterns for the ten Class I and Class III genes that were expressed in the RPE (Figure 5). When the RPE gene expression patterns were grouped based upon expression levels in wildtype, Mitfmi/+ and Mitfmi/mi embryos at E11.5, five distinct patterns of RPE expression were observed (Figure 5b). The RPE Group 1 genes, Mitf, Rab38, and Trpm1, remain expressed in the RPE even within the context of a reduction in functional MITF in Mitfmi/+and Mitfmi/mi embryos. Notably, the in situ hybridization signal for Mitf mRNA itself in the Mitfmi/mi RPE appeared more intense, although this may reflect the cellular hypoproliferation previously observed in the Mitfmi/mi RPE (Nakayama et al., 1998). This is distinct from the RPE Group 2, consisting solely of Dct, in which RPE expression is maintained in Mitfmi/mi embryos, although at diminished levels. This is consistent with previous Dct in situ analysis (Nakayama et al., 1998). The remaining six genes are not expressed in the RPE of Mitfmi/mi embryos, however their corresponding expression patterns can be further subdivided into three additional groupings. RPE Group 3 genes, consisting of Slc45a2 and Tyr, are expressed in the RPE in Mitfmi/+ and wildtype in a similar manner; RPE Group 4 genes, consisting of Gpnmb, Mlana, and Si, show decreased probe intensity in the RPE in Mitfmi/+ embryos compared to wildtype; and RPE Group 5, consisting solely of Tyrp1, shows no RPE expression in Mitfmi/+embryos. In summary, using a controlled scoring system on defined inbred and genetic backgrounds, we have been able to subdivide 14 pigment cell expressed genes into distinct classes and groupings. This approach and data is an important step towards fully developed models describing the network of gene regulation in melanocytes and understanding how they are perturbed in disease states.
Comprehensive gene expression data can provide a powerful resource for predicting and building models of gene regulatory networks. Identifying all the factors that control where, when and to what level mRNA is produced will be essential for deciphering the common language governing transcriptional regulation. Work is being done to indentify sequence elements that convey lineage, spatial and temporal expression information for a gene. However an integral part of such analyzes will be datasets containing the in vivo, temporal-spatial descriptions of gene expression in experimentally manipulated environments. We have focused on developing a method to reproducibly characterize gene expression in pigment cell lineages at E11.5 in the mouse. This captures a time when early melanoblasts are present along the entirety of the embryo and are undergoing active migration and proliferation. It is within this context that we have generated a pigment cell gene expression dataset, using systematic evaluation criteria to identify distinct melanoblast populations and categorize genes based on in situ gene expression.
An inherent requirement of using whole mount in situ hybridization data for comparative analyses is appropriate quantitation and spatial annotation of the data. To accomplish this, we have developed a defined MDS scoring system tailored to the melanoblast lineage, to rapidly, systematically and reproducibly annotate the melanoblast gene expression pattern. This system can be applied to multiple embryos and used to compare genes within backgrounds or genes in the context of genetic or environmental variations. It is important to note that the MDS system is a quantitative score of melanoblast number based on in situ hybridization detection of a specific gene. Thus if the in situ hybridization signal for a gene is detectible in all melanoblasts, than the MDS score is reflective of the total melanoblast number present in a given embryonic region. Alternatively, the MDS score may represent a sub-population of melanoblasts for several reasons. For example systemic perturbation (such as genetic background or inclusion of a genetic mutation) or natural variation of gene expression may result in a subset of melanoblasts with undetectable levels of the gene without affecting overall melanoblast cell numbers. Both indices are biologically important to identify, however one must be mindful of these parameters when incorporating appropriate controls and interpreting the data using this approach.
With this in situ data set we find examples where the MDS score captures either gene expression differences, or a reduction in overall melanoblast numbers in the embryo. An Examples of our ability to capture gene expression differences is found from the comparative analysis of all Class I genes in transcription factor-haploinsufficient conditions. Here analysis of Dct expression revealed a significant reduction in the number of Dct + melanoblasts found in Sox10tm1Weg/+ embryos in comparison to Mitfmi/+embryos. None of the other Class 1 genes exhibits a similar reduction in gene+ melanoblasts in Sox10tm1Weg/+ embryos. Taken together these results indicate that there are more melanoblasts present in Sox10tm1Weg/+ embryos than are found to be Dct+, and that an in vivo reduction in SOX10 levels results in a more dramatic reduction in Dct expression than observed due to alteration in MITF levels alone. These in vivo results are consistent with what we have observed through in vitro promoter transactivation assays indicating that both SOX10 and MITF synergistically regulate Dct expression (Jiao et al., 2004; Ludwig et al., 2004; Potterf et al., 2001), and that Mitf is also a direct target of SOX10 (Potterf et al., 2000), (Bondurand et al., 2000b) and would thus be reduced in Sox10tm1Weg/+ embryos(Britsch et al., 2001). These results also suggest that unlike Dct, the promoters for the genes Si, Gpnmb, Slc45A2 and Tyr would not be predicted to respond to MITF and SOX10 in a synergistic manner.
Another example of the capacity of the MDS score to capture variation in gene expression levels is observed for analysis of Gpnmb and Si. We observed that for both genes Gpnmb and Si there were more gene-positive melanoblasts across the entirety of the embryo in Sox10tm1Weg/+ embryos than in Mitfmi/+ embryos. Given that three other melanoblast genes, Mitf, Slc45a2 and Tyr present with similar numbers of melanoblasts in both Sox10 and Mitf heterozygous mice we suggest that regulatory elements governing Si and Gpnmb gene expression may be more sensitive to reductions in functional MITF levels than to reductions in SOX10. Therefore Gpnmb and Si can be grouped distinctly from the other Class I genes, Dct, Slc45a2 and Tyr. Future systematic analysis of the cis-regulatory regions of these genes correlated with annotation of the melanoblast and RPE lineage repertoire of available transcription factors at E11.5 will be required in order to begin to make more comprehensive predictive models that define these gene expression patterns.
The MDS score also captures gene expression differences with respect to location of gene-positive cell populations in vivo. For example, Mitf and Gpnmb gene expression patterns are distinct relative to the other genes in this data set when comparing the MDS heatmap profiles in Mitfmi/mi and Sox10tm1Weg/tm1Weg homozygous embryos. Previously, we observed Gpnmb- and Mitf-positive cells marking Sox10-independent melanoblasts in a discrete region over the hind limb and tail at E11.5 (Loftus et al., 2008). Here we have extended the analysis of this population of cells in Sox10tm1Weg/tm1Weg embryos, and determined that Dct, Si, Tyr, and Slc45a2 are not expressed in this newly defined cell population. Possibly this distinct gene expression pattern may be correlated with the observation that Gpnmb is also found to be regulated by MITF in the osteoclast cell lineage (Ripoll et al., 2008). Future analyses will be required to determine precisely what factors are contributing to the SOX10-independent expression of these two genes in the caudal melanoblast population and thus differentiate Mitf and Gpnmb from other pigment cell expressed genes.
Analysis of our dataset also provides examples of how genes can be categorized based on temporal-spatial and lineage information. For example, if expression were only examined in terminally differentiated melanocyte cell populations, most genes in this study would be grouped into one Class. However analysis at E11.5 showed that Gpnmb, Si, Slc45a2, Dct and Tyr (Class I) can be distinguished from Trpm1, Mlana, and Tyrp1 (Class II), as the latter are not expressed in melanoblasts at the E11.5 timepoint (Figure 1). This temporal difference is not due to whether or not Class III genes are MITF targets genes, as experimental evidence demonstrates that MITF contributes to the regulation of Trpm1, Mlana, and Tyrp1 in pigmented melanocytes and melanoma cells (Du et al., 2003; Yasumoto et al., 1995; Zhiqi et al., 2004). This data suggests that a more complex interaction of regulatory factors, in combination with MITF, regulate the temporal initiation of gene expression in early melanoblasts versus melanocytes.
Finally, by applying the MDS scoring system to comparison of multiple inbred strains, we demonstrate that early in melanoblast development, patterns can vary significantly depending on the strain background and embryonic location. In this case the variation observed is likely not due to modifier loci merely reducing Si expression in a region specific manner, as we have observed greater numbers of trunk melanoblasts in E11.5 FVB/NJ mice than in C56BL/6J mice for Dct, Gpnmb, and Slc45A2 expression (Baxter and Pavan, 2003), and unpublished data), similar to what is seen for Si in these two strains. Thus the strain dependant variation in numbers of Si-positive melanoblasts is likely due to regional differences in melanoblast numbers between strains and not due to variations in levels of Si gene expression within melanoblasts. Taken together this suggests that there are strain-dependant modifiers of early melanoblast numbers in the embryo. It has long been appreciated that the location and size of hypo-pigmented areas in adult mice are inherited as quantitative traits (Doolittle et al., 1975; Dunn and Charles, 1937; Pavan et al., 1995) and it has more recently been demonstrated that the background strains onto which the spotting mutation alleles Ednrb, Pax3 and Sox10 are maintained affect the severity of the spot (Asher et al., 1996; Britsch et al., 2001; Pavan et al., 1995; Southard-Smith et al., 1999). However, the total number of loci that represent strain-specific modifiers of hypopigmentation remains to be determined. Currently, 104 genes and/or loci are annotated in the MGI database that exhibit some degree of “white spotting” (http://www.informatics.jax.org/), with typical patterns being belly spots, head spots, piebald spotting, and white “belts” which encircle the caudal trunk (Baxter et al., 2004). These genes/loci are known to affect a variety of melanoblast/melanocyte cellular processes including migration, differentiation proliferation, and cell survival. There is the potential for numerous pigmentation modifier loci to segregate among the many inbred strains. Application of the MDS scoring system in the context of a consistent inbred strain background can verify that the details of gene expression differences such as those described here can be correctly attributed to either gene-intrinsic variations or independent modifiers.
All 14 genes in this data set are expressed in the terminally differentiated melanocyte lineage, with 8 of 14 genes (Si, Dct, Gpnmb, Slc45A2, Mlana Tyr, Tyrp1, and Trpm1) classified as MITF transcriptional targets. However, we here demonstrate that not all these genes respond to alteration in MITF levels in a similar manner. Through systematic evaluation of this expression dataset we have been able to sub-divide this gene set into smaller, expression-based, phenotype-correlated gene groupings. These results reinforce the fact that transcriptional regulation is a complex process. Often transcriptional regulation is approached in a simplified manner through the prism of data collected in a single isolated cell type, such as a melanocyte or melanoma derived cell line, under a single condition. Such analysis by default excludes a spatio-temporal based analysis such as observations of marked cell populations during embryonic development. Our dataset and methodology for annotation provides a framework for phenotypic correlations to be made for genes based upon in vivo expression characteristics. In the future, in depth functional analysis of genomic elements will be required to obtain detailed and precise gene expression information. More lineage-appropriate gene expression information is required and must be collected in a standardized manner if one is to develop computational models that identify parameters that define where, when, and at what level a gene is expressed. Fundamental to this type of modeling/analysis is information regarding how modulation of a single regulatory expression pathway or a single transcription factor will vary the expression of individual genes and how that variation compares when a collection of genes with defined similar patterns of expression are compared. Cataloging this variation is the first step in identifying temporal-spatial gene expression pattern correlations between genes and regulatory sequence for those genes. These correlations will provide the foundation to annotate and query the common and distinct cis-regulatory factors for each of these genes and will allow for future testing and evaluation of candidate regulatory elements, among additional genes demonstrating similar defined expression patterns, to assess the extent to which these elements are predictive of the expression patterns observed.
MitfMi/+ mice were obtained from Jackson laboratories B6C3 Fe a/a -Mitf/J strain #001573 and backcrossed to C57BL/6J mice for over 15 generations. Sox10tm1Weg mice were obtained from Dr. Michael Wegner and backcrossed to C57BL/6J for over 10 generations. Both Sox10tm1Weg/+ X Sox10tm1Weg and MitfMi/+ X MitfMi/+ mating crosses were setup, corresponding plugs were identified in the morning and established as day E 0.5. At E11.5, embryos were dissected and all embryos were developmentally staged matched based upon limb, eye and craniofacial structures as defined as E11.5 in Kaufmann (Kaufman, 2005). Genotyping of embryos was performed on DNA isolated from corresponding yolk sacs using the PureGene Kit (Gentra Systems, Inc. Minneapolis, MN). Sox10 genotyping was performed as described (Britsch et al., 2001). Mitf genotyping was performed as described (Loftus et al., 2008).
For in situ probe generation a summarized list of plasmid cDNA clones, with corresponding appropriate restriction enzyme digests and RNA polymerases that were used is provided in Supplemental Table 1. In situ hybridization probes were prepared as indicated previously for the genes Sox10 (Southard-Smith et al., 1998), Rab38 (Loftus et al., 2002), Kit (Keshet et al., 1991) Dct (Steel et al., 1992), and Ednrb (Opdecamp et al., 1998). For the following genes, probes were prepared by linearizing the indicated plasmid clone with KpnI, followed by reverse transcribing digoxigen-conjugated probes using T3 RNA polymerase (reagents from Roche Molecular Biochemical Mannheim, Germany): Mitf, Riken clone #G370008D06; Si, Riken clone #G370069C13; Gpnmb, Riken clone #G370004B14; Mlana, Riken clone #G370004L04; Slc45a2, Riken clone #G3B70045L22; Trpm1, Riken clone #G370028M02; Tyr, Riken clone #4633402C07. For Tyrp1, EST clone Riken clone #4732408D14 was used as a template for amplification using PCR Primers, TYRP1-5′_T3-F-GCGCGAATTAACCCTCACTAAAGGGTCTGAGCACCCCTGTCTTCT and TYRP1-5_T7-R-GCGCGTAATACGACTCACTATAGGGCCCAGTTGCAAAATTCCAGT) (Loftus et al., 2002). The resulting amplification product was purified using Microcon 100 column (Millipore, Bedford MA) and reverse transcribed using T7 RNA polymerase. For Erbb3, plasmid cDNA (gift of Dr. Lidia Kos) was digested with EcoRI and T7 RNA polymerase was used. Whole mount in situ hybridization was performed as previously described (Wilkinson and M. A. Nieto, 1993) with these modifications. Ribonuclease A digestion was omitted, and Tris-buffered saline was used in place of PBS. BM-purple substrate (Roche, Molecular Biochemicals) was used in place of 5-bromo-4-chloro-3-indolylphosphate nitroblue tetrazolium. In situ hybridizations were performed for each gene probe in the following manner to allow for comparisons between strains and genotypes. For comparison of Si-positive melanoblast expression patterns in the six wild-type inbred strains, four embryos of each strain were processed simultaneously, utilizing identical probe amount, and time of alkaline phosphatase reaction. For comparisons of the gene expression patterns for multiple genes in wild-type C57BL/6J embryos the following numbers of embryos were analyzed: four embryos (Si) and three embryos (Dct,) and two embryos (Gpnmb, Kit, Slc45A2, and Tyr). For comparisons of gene expression in Sox10 and Mitf deficient mice, two embryos of each genotype were hybridized and scored; and for each probe, hybridizations for all 5 genotype conditions were processed simultaneously, utilizing identical probe amount, and time of alkaline phosphatase reaction.
A Leica MZ APO microscope onto which a 3CCD MTI camera (Dage-MTI, Michigan City, IN, USA) was attached was used to capture embryo images using Adobe Photoshop 7 and Scion Image 1.63 software.
Region A: defines the melanoblasts found in the crease between the telencephalon and nasal process, extending towards the eye. Region B: defines the melanoblast population found at the crease defining the midbrain, hindbrain boundary. Region C: the area between the caudal boundary line from the eye to the mid-hindbrain in region B and a parallel line immediately anterior to the otic vesicle. Region D: defines the melanoblast population overlying the otic vesicle and extending distal to encompass the melanoblasts migrating across the area corresponding to region overlying cranial ganglia IX and X. Region E: melanoblast population lateral to the forelimb area between areas defined by region D and F. Region F: the melanoblasts present in the trunk as defined by the post axial, (caudal) side of the forelimb bud, throughout the trunk, up to the preaxial (anterior) side of the hindlimb bud. Region G: melanoblasts immediately lateral to the hind limb area. Region H: the melanoblast population caudal to the post axial hindlimb.
A nonparametric ANOVA (Kruskal-Wallis) was used to compare median values across inbred strains in each region (A-H) using commercially available software (InStat, GraphPad Software). All regions failed the null hypothesis at a 95% confidence level and a multiple comparison post-test was used to identify pairs of strains with significant differences in MDS score.
Identification of the components that delineate spatial and temporal expression for a gene throughout an organism’s lifetime is needed to understand gene regulation in both normal and disease settings. Here we generated the framework for a publicly available gene expression dataset for pigment cell genes and have developed a system to compare data across experimental conditions. We focused on gene expression in the eye and melanoblast lineage at mid-gestation in the mouse embryo, when melanoblasts are actively migrating and proliferating. Using systematic evaluation criteria, genes were subdivided based upon their expression patterns in distinct melanoblast populations that vary in time, space and genetic influence. In conjunction with genomic sequence information, this type of comprehensive, in vivo gene expression data can provide a powerful resource for building gene regulatory network models.
Images documenting melanoblast gene expression at E11.5 in wild type, Mitf and Sox10 transcription factor mutant mice for the pigment gene expression dataset have been uploaded to a database server, where all images can be queried and downloaded.
Average MDS scores for Class I and Class IIa genes from E11.5, C57BL/6J embryos for the genes Si, Dct, Gpnmb, Slc45A2, Mitf, Kit and Tyr. A single line represents Mitf, Kit and Tyr as the MDS values for these three genes were identical over all 8 regions.
Si-positive melanoblast cell distribution and number at E11.5 is not uniform among the inbred mouse strains A) BALBc/J, B) C57BL/6J, C) C3H/HeJ, D) DBA/2J, E) 129SvEv/Tac and F) FVB/NJ.
The Mb-positive cell distribution scores for all Class I, IIa and III pigment cell-expressed genes are sorted by gene probe and summarized by heatmap visualization. For each gene, 2 or more embryos were scored for both right and left sides using the MDS outlined in Figure 2. Then all MDS values for each gene were averaged and converted to a white=(0) to dark purple=(4) visualization gradient, indicating Mb cell density within a region. For Class I and IIa, expression was analyzed in C57BL/6J wildtype embryos and Mitf and Sox10 transcription factor deficient embryos (Mitfmi/+, Mitfmi/mi, Sox10tm1Weg/+, and Sox10tm1Weg/tm1Weg, respectively) at E11.5. For Class III genes, which are only expressed in the RPE, analysis was performed on wildtype, Mitfmi/+, and Mitfmi/mi embryos as Sox10 in not expressed in the RPE. The presence of gene expression in the developing eye is denoted by grey, and the absence of expression by white in the far right column.
Inbred strain pairs with significant differences were identified using ANOVA with post test.
This work was funded through the NIH intramural program, NHGRI. Experiments were performed in accordance with the NHGRI animal care and use protocol G-94-7. We would like to thank Dr. Lidia Kos for Erbb3 cDNA, Lowell Umayam and Anh-Dao Nguyen (NHGRI Bioinfomatics core) for database design and support, and Darryl Leja (NHGRI) for graphics assistance.