| Human Molecular Genetics | Pages |
Functional analysis of Gscl in the pathogenesis of the DiGeorge and velocardiofacial syndromes
Introduction
Results
Generation of Gscl null mice
Gscl null mice are viable and fertile
Altered Es2 expression in the brains of Gscl null mice
Mice that lack both Gsc and Gscl
Discussion
Materials And Methods
Deletion of Gscl by gene targeting in mouse ES cells
Generation of chimeric mice and germline transmission of the Gscl mutant allele
RNA in situ hybridization and immunostaining
Acknowledgements
References
Functional analysis of Gscl in the pathogenesis of the DiGeorge and velocardiofacial syndromes
INTRODUCTION
Heterozygous deletions within 22q11.2 (del22q11) are associated with at least two syndromic presentations: DiGeorge syndrome (DGS) (1) and velocardiofacial syndrome (VCFS) (1). The main clinical findings associated with del22q11 are craniofacial anomalies, congenital heart disease, T cell immunodeficiency and hypocalcemia. The deleted region is estimated to span 2-3 Mb of DNA. Several genes have been identified in this region; however, the role of any of these genes in the pathogenesis of DGS/VCFS remains to be defined (2).
The mouse Gscl gene was recently discovered by DNA sequence analysis of a region of mouse chromosome 16 that is syntenic to the DGS/VCFS deleted region (3). The human GSCL gene has also been isolated and maps to this region (4-6). Gscl encodes a homeodomain protein that is related to Goosecoid (Gsc). Gsc was originally identified in a screen for homeobox genes expressed in the dorsal blastopore lip of Xenopus embryos (7). Injection of Gsc mRNA into Xenopus embryos led to formation of secondary body axes, suggesting that Gsc regulated vertebrate axis formation (8). In mice, Gsc is expressed in the anterior visceral endoderm and the anterior primitive streak of the gastrula, further suggesting a role in body axis formation (9-11). At later stages, Gsc is expressed in craniofacial regions, ventral body wall and limbs (12,13). Surprisingly, Gsc null mice were normal with respect to axis formation and also limb development but developed craniofacial abnormalities and died soon after birth (13-16). The lack of axis formation abnormalities in Gsc null mice led to the speculation that another Gsc gene existed that could compensate for Gsc during gastrulation (13-15). Because Gscl is structurally related to Gsc, Gscl became a candidate compensating gene for Gsc and an exciting candidate gene for DGS/VCFS.
Gscl transcripts are detected in the mouse from gastrulation [6.5 days post-coitum (d.p.c.)] to birth (3,4,13,17,18). In situ hybridization and immunohistochemistry have localized Gscl/Gscl expression to primordial germ cells and the developing pons region (17,18). Interestingly, Gscl expression in the pons region overlaps with expression of Es2 that resides just downstream of Gscl, suggesting that these two genes may share common cis-acting regulatory sequences (18). In addition, Gscl/GSCL is expressed in multiple adult tissues, including brain, eye, pituitary, thymus, thyroid region, gut, bladder and testis (4,5,13). Thus, Gscl may contribute to DGS/VCFS because it is expressed in some of the same organs that are abnormal in DGS/VCFS patients.
To determine the role of Gscl during mouse embryogenesis and its role in the pathogenesis of DGS/VCFS, we have deleted Gscl in mouse embryonic stem (ES) cells. Although Gscl expression is detected throughout mouse embryogenesis and in the adult predominantly in the testis, Gscl heterozygous and Gscl null mice were found to be viable and fertile. Interestingly, expression of the Es2 gene in the pons was absent in Gscl null embryos, suggesting that mutations within the DGS/VCFS deletion region can influence expression of adjacent genes. Furthermore, Gscl/Gsc double homozygous mutants appeared normal at 9.5 d.p.c. These results demonstrate that in mice Gscl is dispensible for the normal development of structures and organs affected in DGS/VCFS. These mice should be useful for crosses with other mice with targeted mutations that are being generated in DGS/VCFS candidate genes.
RESULTS
Generation of Gscl null mice
To determine the role of Gscl in the development of DGS/VCFS, a 1.7 kb fragment containing the three protein coding exons of Gscl was deleted by gene targeting in mouse ES cells (Fig.
A
 B, C  |
Figure 1. Targeted deletion of the Gscl gene. (A) Strategy for targeted mutation of the Gscl locus. (Top) Structure of the Gscl wild-type allele. Boxes, exons; shaded boxes, the Gscl homologous regions used in the targeting vector. The unique Southern probes are shown below the locus. The sizes of the restriction fragments detected by the indicated probes in wild-type DNA are shown above the locus. (Middle) The gene targeting vector. NEO, PGK-neo expression cassette that introduces novel EcoRI and EcoRV restriction sites; TK, MC1-tk expression cassette for negative selection. The arrows for NEO and TK indicate the direction of transcription of each cassette. (Bottom) Structure of the mutant allele. The sizes of the restriction fragments detected by the Southern probes in the targeted locus are shown above the locus. E, EcoRI; K, KpnI; N, NotI; P, PstI; S, SmaI; V, EcoRV. (B) Southern analysis of genomic DNA isolated from ES cell lines. Correct targeting events are shown for three different ES clones (58, 64 and 90). (Left) EcoRI/BamHI co-digested ES cell DNA hybridized with the 5[prime] probe. The 6.4 kb wild-type (wt) and 5.0 kb mutant (m) bands are shown. (Right) EcoRV-digested ES cell DNA hybridized with the 3[prime] probe. The 6.2 kb wild-type and 5.1 kb mutant bands are shown. AB1, wild-type ES cell line. (C) Southern blot showing genotypes of mice from a Gscl heterozygote mating. (Left) EcoRV-digested tail DNA hybridized with the 3[prime] probe. The 6.2 kb wild-type and 5.1 kb mutant bands are shown. (Right) EcoRV-digested tail DNA hybridized with a Gscl genomic fragment containing exons 1 and 2. Hybridization is only detected in the wild-type and heterozygote samples. +/+, wild-type; +/-, heterozygous mutant; -/-, homozygous mutant.
When the targeting vector recombines with the endogenous Gscl locus, novel EcoRI and EcoRV sites are introduced. Correctly targeted ES cell clones can therefore be detected by the presence of an additional 5.0 kb DNA fragment when co-digested with EcoRI and BamHI and hybridized with a 5[prime] probe external to the region of vector homology or by the presence of a 5.1 kb fragment when digested with EcoRV and hybridized with a 3[prime] probe external to the region of vector homology. Correct targeting deletes the entire Gscl coding region, suggesting that this mutation will be a null allele. Three correctly targeted ES cell clones were found to contribute to the germlines of mouse chimeras generated by blastocyst injection (Fig.
Gscl null mice are viable and fertile
Although patients heterozygous for del22q11 develop DGS/VCFS pathologies, Gscl heterozygous mice were normal and fertile. Furthermore, mice homozygous for the Gscl deletion were recovered at weaning from matings between heterozygotes at the predicted Mendelian ratios (Fig.
Table 1.
| Number (%) | |||
| Male | Female | Total | |
| Wild-type | 17 (24) | 13 (21) | 30 (23) |
| Heterozygote | 39 (55) | 31 (51) | 70 (53) |
| Mutant | 15 (21) | 17 (28) | 32 (24) |
| Total | 71 | 61 | 132 |
Gscl has been reported to be highly expressed in migrating primordial germ cells and germ cells within the fetal gonads (17), suggesting a role for Gscl in germ cell development. However, both male and female Gscl null mice proved to be fertile. Indeed, when male and female Gscl null mice were bred together they yielded normal sized litters containing only Gscl null progeny (data not shown). Thus, Gscl is not essential for germ cell development and fertility.
Expression studies of adult mouse and human tissues suggest that Gscl may have a functional role in brain, eye, pituitary, thymus, thyroid, stomach, bladder and testis (4,5,13). The Gscl null mice appeared normal in behavior without any obvious neurological deficits. Furthermore, none of the Gscl-expresssing organs listed above were grossly or histologically abnormal in Gscl null mice. We also examined skeleton preparations of Gscl null neonates stained with alizarin red for bone and alcian blue for cartilage and found no abnormalities (data not shown). These results indicate that Gscl is not a major contributing factor in DGS/VCFS.
Altered Es2 expression in the brains of Gscl null mice
We have previously shown that Gscl shares an expression domain in the developing pons with the gene Es2, located immediately downstream of Gscl (18). We hypothesized a possible biological relationship between the two genes, including a shared cis-acting regulatory element. In order to understand whether the targeted mutation of Gscl could affect expression of Es2, we performed RNA in situ hybridization on Gscl null and wild-type 14.5 d.p.c. embryos using a previously reported probe. Results showed that Es2 expression was not detectable in the developing pons of Gscl null embryos (Fig.
Figure 2. Es2 expression in Gscl null embryos. RNA in situ hybridization with an MES600 antisense probe on sagittal sections through the brain of 14.5 d.p.c. mouse embryos. (A and C) Wild-type embryo; (B and D) Gscl null embryo. mv, mesencephalic vesicle; c, cerebellum; cp, choroid plexus; 4th vent, 4th ventricle. We have reported a partial overlap between the Es2/Gscl expression domain in the pons and serotonin-positive cells of the raphe nuclei. In order to test whether Gscl loss of function could affect the distribution of serotonin-positive cells, we performed immunostaining of Gscl null 14.5 d.p.c. embryos with an anti-serotonin antibody. Results showed no detectable difference between Gscl null and wild-type embryos (data not shown). Hence, the Gscl mutation does not affect the serotonin immunohistochemical phenotype.
Mice that lack both Gsc and Gscl
RT-PCR studies demonstrate that Gscl is expressed during gastrulation (6.5-7.5 d.p.c.) and, therefore, temporally overlaps with Gsc expression (13,18). The lack of axis formation abnormalities in Gsc null mice suggests that Gscl may compensate for Gsc in its absence. To test this hypothesis, we bred Gscl heterozygotes with Gsc heterozygotes to generate Gscl/Gsc double heterozygotes. These Gscl/Gsc double heterozygotes were then backcrossed with Gscl null mice to generate mice that were both null for Gscl and heterozygous for Gsc. These Gscl null/Gsc heterozygous mice were normal and fertile. Finally, Gscl null/Gsc heterozygous mice were interbred to generate Gscl/Gsc double homozygous mutants that would lack both Gscl and Gsc. Gscl/Gsc double homozygous mutant embryos were recovered at E9.5 that were normal with respect to body axis formation (Fig.
Figure 3. Axis formation in Gsc/Gscl double homozygous mutant embryos. (A-F) 9.5 d.p.c. mouse embryos. (A and B) Gscl null/Gsc null; (C) Gscl heterozygous/Gsc null; (D) Gscl null/Gsc heterozygous; (E) Gscl wild-type/Gsc null; (F) Gscl null/Gsc wild-type. The tails of the embryos in (A) and (E) were removed for photographic purposes. 22q11 deletions constitute an important cause of developmental defects in human. Progress has been made in the identification of the gene content of the deleted region but the causative gene(s) has not been identified. Furthermore, the precise developmental pathway(s) affected by the deletion are unknown. Comparative mapping data have shown conservation of the del22q11 locus in mouse (22,23), indicating that the mouse can be used as a model for functional analysis of the region. The lack of any obvious abnormalities associated with DGS/VCFS in Gscl null mice strongly suggests that Gscl is not a major contributing factor in the pathogenesis of DGS/VCFS. Thus, there must be another gene or genes associated with del22q11 that is responsible for DGS/VCFS. Therefore, the results of our functional study refocus attention on other genes within del22q11. The phenotypic interpretation of a targeted mutation requires knowledge of the expression pattern of the mutated gene. However, the precise expression pattern of Gscl during embryogenesis in mouse and human is not clear, perhaps because of the high GC content of Gscl/GSCL (3-5,13,17,18). Most studies agree that Gscl is expressed throughout mouse embryogenesis and at least in the 9-10-week-old human fetus. Gscl protein has been localized using a polyclonal antibody to the primordial germ cells of 9.5 d.p.c. embryos to at least 14.5 d.p.c. and the developing pons region beginning at 12.5 to at least 14.5 d.p.c. (17). However, earlier Gscl expression has been detected by RT-PCR and northern analysis between 6.5 and 8.5 d.p.c., suggesting that Gscl may have roles beyond primordial germ cells and the central nervous system (4,13,18). Expression of Gscl during gastrulation at 6.5 d.p.c. suggests a role in axis formation that might compensate for an absence of Gsc (14,15). However, mice null for both Gscl and Gsc have no abnormalities in axis formation. One simple explanation would be that Gsc and Gscl are expressed in different regions of the mouse gastrula. We have been unable to detect Gscl transcripts at 6.5 d.p.c. by in situ hybridization. The insertion of a reporter gene such as lacZ into the endogenous Gscl locus by gene targeting will facilitate the detection of Gscl-expressing cells during development. Recently, another Gsc-related gene, GSX, was isolated in chicken whose expression intially overlaps with GSC expression during gastrulation (24). Whether a mouse Gsx gene exists that can compensate for Gsc during gastrulation remains to be determined. Gscl/GSCL expression in adult mouse and human tissues is also unclear (4,5,13,17). Northern analysis has only detected GSCL in the adult testis (5). Gscl/GSCL expression in adult testis has been readily detected by RT-PCR (4,5,13). However, in situ hybridization and immunohistochemistry apparently did not detect Gscl expression in adult testis (17). RT-PCR analysis also detected Gscl/GSCL in adult pituitary, brain, eye, thymus, thyroid region, stomach and bladder (5,13). Expression of Gscl in the thymus and the thyroid region, organs affected in DGS/VCFS patients, suggests that a deficiency of Gscl may contribute to a subset of DGS/VCFS pathologies. However, the normal thymus and thyroid of Gscl null mice suggests that lack of Gscl alone is not sufficient to generate these abnormalities. One clear abnormality associated with Gscl deletion is the absence of Es2 expression in the pons of Gscl null fetuses. The lack of expression of Es2 in the pons (but not in other tissues) could be due to several factors. First, the PGKneo cassette might inactivate transcription of Es2. This possibility is difficult to reconcile with the tissue-specific nature of Es2 inactivation. A second possibility is that the Gscl null mutation may cause defective development/migration/proliferation of Es2-positive cells in the pons leading to an absence of Es2 signal in this region. Using standard histological staining, we could detect no difference between Gscl null and wild-type embryos. However, this method may not be sufficient to detect the absence or inappropriate development of a relatively small population of cells. A third possibility is that the targeted 1.7 kb deletion eliminates or inactivates a cis-acting, tissue-specific regulatory element important for transcription of Es2 in these cells. Finally, the Gscl gene product may function as an Es2 transcription activator in these cells. Hence, although at this time we cannot completely rule out the possibility that Es2 inactivation is an artifact produced by the targeting cassette, it seems more likely that our expression data reveal either a defective development of a population of cells in the brain caused by Gscl loss of function or a biological interaction between Gscl and Es2. The nature and function of Es2- and Gscl-expressing cells in the developing pons are unknown. Galili et al. (17) reported a partial overlap with the expression domain of Pax6 and proposed that these were relatively undifferentiated, migrating cells. We noticed a spatial correlation between these cells and serotonergic neurons of the median raphe nuclei, including partial overlap with serotonin-positive cells (18). Anti-serotonin immunostaining, however, could not detect any abnormality in Gscl null embryos. The only available marker for these cells is Gscl (or Es2) expression. The insertion of a reporter gene like lacZ into the Gscl locus will provide a powerful tool to further analyze these cells and to establish their fate. These data constitute the first report of a functional test for a DGS/VCFS candidate gene in vivo. Our studies demonstrate, at least in mice, that Gscl is not a primary factor in the pathogenesis of DGS/VCFS. Because it is not known if DGS/VCFS is caused by a reduction in the expression levels of one or multiple genes, it is still possible that Gscl may interact with other genes within or adjacent to the del22q11 locus to generate the spectrum of abnormalities found in DGS/VCFS patients. Perhaps the most important and fundamental mutation to generate in mice with regard to DGS/VCFS will be a targeted deletion that contains the entire mouse del22q11 syntenic region (25). It will be of interest to determine if heterozygosity for such a deletion will result in DGS/VCFS in mice. It should be noted that a number of human autosomal dominant haploinsufficiency syndromes have been recapitulated in mice. However, in a subset of these situations, mice heterozygous for these mutations are normal, and only homozygous mutants exhibit phenotypic abnormalities (26-30). Thus, it is becoming increasingly clear that a molecular understanding of DGS/VCFS will require a series of single gene mutations and deletions within the mouse del22q11 syntenic region. Therefore, our Gscl null mice will be an important resource for crosses with other mutations being generated in or near the DGS/VCFS deletion region. A 3.3 kb PstI-KpnI 5[prime] genomic fragment and a 2.2 kb BamHI-SmaI 3[prime] genomic fragment derived from the Gscl locus of mouse strain 129/SvEv origin were used to generate a targeting vector. A PGKneobpA expression cassette (31) flanked by loxP sites was inserted in the forward orientation relative to the direction of Gscl transcription between the two Gscl regions. A MC1tkpA herpes simplex virus thymidine kinase expression cassette was added onto the 5[prime] arm of homology to enrich for homologous recombinants by negative selection (32) with 1-(2-deoxy-2-fluoro-[beta]-D-arabinofuranosyl)-S-iodouracil (FIAU). The targeting vector was linearized at a unique NotI site outside the homology and introduced into AB-1 ES cells as described (33,34). A total of 136 G418/FIAU-resistant ES cell clones were initially screened by EcoRV digestion and hybridized with a unique 3[prime] probe external to the region of vector homology and by EcoRI/BamHI co-digestion and hybridized with a unique 5[prime] probe external to the region of vector homology. Three of the 174 (2%) G418/FIAU-resistant ES cell clones screened were found to be correctly targeted. Gscl mutant ES cell clones were microinjected into B6 blastocysts and the resulting chimeric embryos were transferred to the uterine horns of day 2.5 pseudopregnant foster mothers (35). Chimeras were identified among the resulting progeny by their agouti fur (ES derived) and were subsequently bred with B6 mates. Three of the mutant ES cell clones were found to be capable of contributing to the germlines of chimeric mice. Tail DNA from the agouti pups that resulted from these matings was analyzed by Southern blot using EcoRV digestion and the 3[prime] probe to identify the Gscl heterozygotes. The probe MES600 identifying the 3[prime]-end of the mouse Es2 gene has been previously reported (18). Sense and antisense 35S-labeled RNA probes were produced and hybridized to paraffin embedded embryo sections according to a published protocol (36). Hybridization was overnight at 58°C, followed by post-hybridization stringency washes at 64°C. Slides were then coated in NTB2 photographic emulsion (Kodak) and exposed at 4°C for 6 days. After development and prior to mounting, slides were counterstained in Hoechst 33258 dye to stain the cell nuclei. Digital images are double exposures, one for Hoechst staining and one for silver grains in dark field. Images, obtained using a cooled CCD camera (Photometrics), were pseudocolored and merged using Adobe Photoshop software. Immunostaining of paraffin embedded embryo sections was performed according to a published protocol (37). To detect serotonin, we utilized a commercial polyclonal rabbit anti-serotonin antibody (Biomeda). Detection was performed using an FITC-conjugated goat anti-rabbit antibody (Zymed) diluted 1:200 with PBS. Slides were mounted without drying in a glycerol based anti-fade mounting solution. We thank Jian Min Deng for help with tissue culture and Allan Bradley for the AB-1 ES and SNL 76/7 STO cell lines. These studies were supported by grants from the National Institutes of Health HL51524 (to A.B.) and DE12705 (to R.R.B.).
DISCUSSION
MATERIALS AND METHODS
Deletion of Gscl by gene targeting in mouse ES cells
Generation of chimeric mice and germline transmission of the Gscl mutant allele
RNA in situ hybridization and immunostaining
ACKNOWLEDGEMENTS
REFERENCES
This article has been cited by other articles:
This page is run by Oxford University Press, Great Clarendon Street, Oxford OX2 6DP, as part of the OUP Journals
Comments and feedback: www-admin{at}oup.co.uk
Last modification: 12 Nov 1998
Copyright©Oxford University Press, 1998.
CiteULike 
Connotea 
Del.icio.us What's this?
T. Thyagarajan, S. Totey, M. J. S. Danton, and A. B. Kulkarni
GENETICALLY ALTERED MOUSE MODELS: THE GOOD, THE BAD, AND THE UGLY
Critical Reviews in Oral Biology & Medicine,
May 1, 2003;
14(3):
154 - 174.
[Abstract]
[Full Text]
[PDF]
A. Baldini
DiGeorge syndrome: the use of model organisms to dissect complex genetics
Hum. Mol. Genet.,
October 1, 2002;
11(20):
2363 - 2369.
[Abstract]
[Full Text]
[PDF]
P. J. Scambler
The 22q11 deletion syndromes
Hum. Mol. Genet.,
October 1, 2000;
9(16):
2421 - 2426.
[Abstract]
[Full Text]
[PDF]
T. Kume, K. Deng, and B. L. M. Hogan
Minimal Phenotype of Mice Homozygous for a Null Mutation in the Forkhead/Winged Helix Gene, Mf2
Mol. Cell. Biol.,
February 15, 2000;
20(4):
1419 - 1425.
[Abstract]
[Full Text]
J. Brickman, C. Jones, M Clements, J. Smith, and R. Beddington
Hex is a transcriptional repressor that contributes to anterior identity and suppresses Spemann organiser function
Development,
January 6, 2000;
127(11):
2303 - 2315.
[Abstract]
[PDF]
This Article 
Abstract
FREE Full Text (PDF)
Alert me when this article is cited
Alert me if a correction is posted
Services
Email this article to a friend
Similar articles in this journal
Similar articles in ISI Web of Science
Similar articles in PubMed
Alert me to new issues of the journal
Add to My Personal Archive
Download to citation manager
Search for citing articles in:
ISI Web of Science (19)
Request Permissions Google Scholar
Articles by Wakamiya, M.
Articles by Behringer, R. R.
Search for Related Content
PubMed
PubMed Citation
Articles by Wakamiya, M.
Articles by Behringer, R. R.
Social Bookmarking
What's this?