Skip Navigation

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (12)
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Semenova, E.
Right arrow Articles by Tilghman, S. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Semenova, E.
Right arrow Articles by Tilghman, S. M.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

Human Molecular Genetics, 2003, Vol. 12, No. 11 1301-1312
DOI: 10.1093/hmg/ddg140
© 2003 Oxford University Press

An engineered 800 kilobase deletion of Uchl3 and Lmo7 on mouse chromosome 14 causes defects in viability, postnatal growth and degeneration of muscle and retina

Ekaterina Semenova1, XiaoFei Wang2, Monica M. Jablonski2, John Levorse1 and Shirley M. Tilghman1,*

1Howard Hughes Medical Institute, Department of Molecular Biology, Princeton University, Princeton, NJ 08544, USA and 2Department of Ophthalmology, Retinal Degeneration Research Center, The University of Tennessee Health Science Center, Memphis, TN 38163, USA

Received January 26, 2003; Accepted March 30, 2003


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
The Acrg minimal region is a 1.5–1.7 Mb domain defined by genetic complementation among deletions generated around Ednrb on chromosome 14 in mice. Mice homozygous for one of the deletions, Ednrbs-1Acrg, exhibit embryonic lethality with defects associated with mesoderm development. We predicted that the region contains a single cluster of four genes that encode a TBC domain-containing protein (KIAA0603), a novel protein AK000009, the ubiquitin C-terminal hydrolase L3 (UCHL3) and an F-box/PDZ/LIM domain protein LMO7. A targeted internal deletion of Uchl3 (Uchl3{Delta}3-7) produced viable mice, eliminating this gene as a candidate for the embryonic lethality. To dissect the Acrg minimal region further, we utilized Cre–loxP-mediated chromosome engineering to generate a targeted 800 kb deletion (Lmo7{Delta}800) that removes the distal portion of the region. The deletion includes Uchl3, Lmo7 and an additional 500 kb downstream of the 3' end of Lmo7 where no genes are thought to reside. We found that ~40% of mice homozygous for this deletion die between birth and weaning, and are severely runted. The remaining homozygotes are viable, thus ruling out Lmo7 as a single gene candidate for the Ednrbs-1Acrg embryonic lethality. Both Uchl3{Delta}3-7 and Lmo7{Delta}800 mutants displayed retinal degeneration, muscular degeneration and growth retardation, but the severity of the muscular degeneration and growth retardation were enhanced in Lmo7{Delta}800 homozygotes. We suggest that the increase in severity may reflect an interaction between Uchl3 and Lmo7 in the ubiquitin-mediated protein degradation pathway.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
As the sequencing of the human and mouse genomes approaches completion (1–3), a major focus of mammalian genetics has become the analysis of gene function. One of the most powerful ways to gain understanding of human gene function is through analyses of mutant phenotypes of the cognate genes in the mouse. Despite recent advances in targeted mutagenesis techniques, the task of generating mutations one gene at a time remains labor-intensive and time-consuming.

A complementary way to approach the functional dissection of a complex genome is to generate large deletions throughout the genome to screen large domains for genes of biological importance. Deletions in the mouse can be generated by radiation and chemical mutagenesis in the germ line (4), radiation mutagenesis in ES cells (5,6) and more recently, by targeted Cre–loxP-induced site-specific recombination in ES cells (7,8). This approach has been used to model human diseases associated with duplications, deletions, inversions and translocations of chromosomal segments such as DiGeorge syndrome, Prader–Willi syndrome, Beckwith Wiedemann syndrome and leukaemia-associated translocations (9–13). Chromosome engineering has also been used to generate specifically defined deletions and inversions, which can be used in large-scale genetic screens for recessive mutations in the mouse (14).

Here we utilize Cre–loxP-mediated chromosome engineering to functionally dissect a 1.5–1.7 Mb region that was defined by induced deletions generated in the first mouse germ line mutagenesis screen called the specific locus test (SLT) (4). In the SLT multiple mutations, including overlapping deletions, were generated around seven recessive, viable loci, chosen for their easily scored phenotypes. Through genetic complementation and molecular analyses of these deletions, specific functional domains within the deletions have been defined, and a number of genes have been cloned (15,16).

The piebald (s) locus, one of the loci utilized in the SLT, encodes the endothelin B receptor (EDNRB), a G protein-coupled seven-transmembrane receptor required for migration of two neural crest derivatives, melanocytes and enteric ganglia (17,18). Mice homozygous for a null mutation in Ednrb die as juveniles of megacolon, resulting from abnormal innervation of the colon. Several larger deletions generated around Ednrb exhibited more severe phenotypes, and complementation analysis led to the mapping of regions essential for central nervous system development, skeletal patterning and early embryogenesis (19,20).

Mice homozygous for one of the SLT deletions, Ednrbs-1Acrg, arrest at embryonic day 8.5 and exhibit complex defects that include disfigured primitive streak and node, truncated posterior body axis, notochord degeneration, somite malformation and abnormal vascular development (21). It has been proposed that such defects are caused by abnormal development of streak-derived posterior mesoderm and impaired lateral morphogenesis (21). Complementation analysis identified a domain within Ednrbs-1Acrg between the proximal breakpoints of the Ednrbs-36Pub and Ednrbs-1Acrg deletions, termed the Acrg minimal region, that was responsible for the embryonic lethality (Fig. 1A) (19).



View larger version (23K):
[in this window]
[in a new window]
  		Figure 1. The generation of the Lmo7{Delta}800deletion. (A) The position of Ednrb is shown above chromosome 14, with genetic map distances in centimorgans shown below the chromosome. The extent of the two Ednrb deletions that define the Acrg minimal region is shown as thin black lines below the chromosome. The gray box depicts the Acrg minimal region, and the yellow box depicts the extent of the Lmo7{Delta}800 deletion. Arrows depict the positions of the four genes located within the Acrg minimal region, with the yellow genes representing those affected by the deletion. Small red and blue boxes depict the probes used to obtain targeting vectors from 5' Hprt and 3' Hprt libraries, respectively. (B) The 5' Hprt and 3' Hprt targeting vector backbones are depicted in red and blue, respectively, with loxP sites as yellow triangles, and 5' and 3' portions of Hprt minigene as well as the genes for the selectable markers outlined in green. The structures of the wild-type chromosome (WT chr), and double-targeted chromosomes before and after Cre-mediated deletion (deleted chr) are indicated. Relevant restriction cleavage sites are indicated by: H, HindIII; K, KpnI; S, SmaI; RI, EcoRI; X, XbaI; Hp, HpaI; and the diagnostic fragments are indicated below each chromosome. The positions of the gaps are indicated by the open and shaded boxes. (C) Southern blot analysis of DNA from wild-type (wt), single-targeted (t), double-targeted (dt) and deleted (d) ES cells. Left-most panel: HindIII digest of ES cell genomic DNA after step1; 5'-gap is used as a probe. Right-most panel: XbaI digest of DNA after step2; 3'-gap is used as a probe. Middle panel: EcoRI digest of DNA after step3; 3'-gap is used as a probe.

 
The Acrg minimal region was cloned on overlapping BACs and sequenced. Comparative sequence analysis between mouse and human using PipMaker, as well as gene prediction analysis using the GeneMachine program package, identified a single cluster of four genes in the middle of a large gene desert (22). The four genes encode a TBC domain-containing protein (KIAA0603), a novel protein AK000009, the ubiquitin C-terminal hydrolase L3 (UCHL3) and an F-box/PDZ/LIM domain protein LMO7 (22). A targeted internal deletion of Uchl3 (Uchl3{Delta}3-7) produced viable mice, eliminating the loss of this gene as an explanation for the embryonic lethality (23). Based on their expression during early embryogenesis, each of the other three genes is a plausible candidate for the Ednrbs-1Acrg embryonic lethality (22). However we considered Lmo7 to be the strongest candidate on the bases of important roles that other LIM/PDZ-containing proteins play in signal transduction, cell shape changes, motility and cell adhesion, all of which are essential for normal embryogenesis (24–27). For example, mice homozygous for a mutation in the PDZ domain-containing protein Afidin showed developmental defects during and after gastrulation, including impaired migration of mesoderm similar to that seen in Acrg deletion embryos (28).

To dissect the function of the Acrg minimal region further and to address the role of the distal half of the region, which includes Lmo7, we generated an 800 kb targeted deletion, termed Lmo7{Delta}800. The deletion removes Uchl3, Lmo7 and ~500 kb of presumed non-coding sequence downstream of Lmo7. We found that mice homozygous for this deletion were born at the expected Mendelian frequency, but a subset died between birth and weaning. Those that survived had defects in postnatal growth, as well as muscle and retinal degeneration. Comparative phenotypic analyses of mice carrying Uchl3{Delta}3-7 and Lmo7{Delta}800 mutations allowed us to assess the individual contributions of Uchl3 and Lmo7 genes towards the Lmo7{Delta}800 phenotype.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
Generation of Lmo7{Delta}800
We set out to create an 800 kb deletion that removes approximately the distal half of the Acrg minimal region, including the last eight exons of the 10-exon Uchl3 gene and all of Lmo7 (Lmo7{Delta}800, Fig. 1A). We utilized the Cre–loxP site-specific recombination system in ES cells to insert two loxP sites at the breakpoints of the intended deletion (7). To obtain targeting vectors we screened two complementary genomic 5' and 3' Hprt libraries, generated in the laboratory of Dr Alan Bradley (29). The 5' and 3' Hprt library vectors contain, in addition to mouse genomic DNA, a loxP site and 5' and 3' non-functional halves of the human Hprt minigene as well as the selectable markers for neomycin and puromycin resistance, respectively.

We screened the 5' Hprt library with a probe derived from within the second intron of Uchl3 and the 3' Hprt library with a probe derived from within a P1 clone that crosses the distal breakpoint of the Acrg minimal region (Fig. 1A). Restriction mapping of both constructs was used to orient the inserts with respect to the vector as well as with respect to the chromosome. To obtain the correct orientation, both inserts were flipped using a unique AscI restriction site in the vector backbone. Using the information from the restriction map, we generated a 1.1 kb gap within the 5' Hprt insert using KpnI and SmaI and a 1.5 kb gap within the 3' Hprt insert using HpaI (Fig. 1B). The gaps allowed for efficient insertional targeting via the gap repair machinery.

We targeted Hprt-deficient AB2.2 129/SvEv ES cells with the 5' Hprt gapped vector and obtained correctly targeted neomycin resistant (neor) clones with an efficiency of 19% (32/180). The neor cells were next targeted with the 3' Hprt gapped vector and correctly targeted puromycin resistant clones were obtained with an efficiency of 12% (21/180). To excise the 800 kb between the two loxP sites, we electroporated a Cre-expressing plasmid (pOG231) into six double-targeted ES clones, and selected for HAT resistance, indicative of recombination to restore HPRT activity (Fig. 1B). Based on Southern analysis (Fig. 1C) and on acquired sensitivity to both neomycin and puromycin, all the clones that grew in HAT medium had undergone Cre-mediated deletion. Given the high efficiency of colony formation as well as sensitivity to both drugs, we conclude that the two targeted events had occurred in cis in all six clones (7,30).

To verify that a deletion had been induced, we employed fluorescent in situ hybridization (FISH) technology. Two BACs, one within and one outside the deletion, were used as probes. The internal BAC AC079638 (D8) was located ~200 kb downstream of the proximal breakpoint of the deletion, whereas the external BAC AC074357 (2G) was located immediately upstream of the proximal breakpoint, ~300 kb away from D8. The two BACs were labeled with different fluorochromes such that D8 was visualized as a red signal and 2G as a green signal. We used double targeted ES cells before Cre excision as a control to establish that both chromosomes displayed green and red signals (Fig. 2). Following excision in the heterozygous ES cell lines, the majority of cells contained one chromosome with a double signal and another chromosome with a single green signal (Fig. 2), indicating that a deletion was generated.



View larger version (76K):
[in this window]
[in a new window]
  		Figure 2. DNA FISH analysis of ES cells carrying Lmo7{Delta}800. ES cells that were doubly targeted before (dt) and after (d) Cre excision are shown following FISH analysis. The green signal indicates hybridization of a BAC outside the deletion, and the red signal indicates hybridization of a BAC within the deleted DNA. Nuclei are stained blue with DAPI. The yellow triangle in the diagram indicates a loxP site. (A) A single ES cell nucleus, doubly targeted before Cre excision. (B) A single ES cell nucleus, doubly targeted after Cre excision. (C) A field of ES cell nuclei, doubly targeted after Cre excision. The appearance of double green spots in several cells is thought to be either due to replication, or to the large size of the BAC probe. Similar results were occasionally seen with the red probe.

 
Two independently targeted ES cell lines carrying the Lmo7{Delta}800 deletion were injected into C57BL/6 blastocysts and male chimeras were bred with C57BL/6 females to generate stable lines containing the mutation. Mice homozygous for Lmo7{Delta}800 were obtained by intercrossing F1 animals and later maintained by brother–sister mating. Lmo7{Delta}800 homozygotes were viable and were obtained at the expected frequency at birth (31/158 in an F1 intercross and 81/168 in Lmo7{Delta}800/Lmo7{Delta}800xLmo7{Delta}800/+crosses).

Impact of Lmo7{Delta}800 on the expression of genes in the cluster
The Lmo7{Delta}800 deletion removes both Uchl3 and Lmo7, the two distal genes within the four-gene cluster (Fig. 1A). Ak000009 and Kiaa0603 are located 26 and 60 kb upstream of the proximal deletion breakpoint, respectively. To ensure that the deletion did not disrupt transcription of these genes, we performed northern analysis of kidney and muscle RNAs prepared from mutant and wild-type animals. As expected, both Lmo7 and Uchl3 transcripts were absent in the homozygous deletion tissues (Fig. 3), while no change was detected in the levels of Ak000009 transcripts. Interestingly, the Kiaa0603 transcripts migrated differently in the wild-type and mutant tissues (Fig. 3). RNAs prepared from wild-type animals displayed a predominant 7 kb transcript with a weak 9 kb transcript while RNA from the homozygotes had a predominant 9 kb transcript.



View larger version (86K):
[in this window]
[in a new window]
  		Figure 3. Effects of Lmo7{Delta}800 and Uchl3{Delta}3-7 on expression of Acrg cluster genes. Total RNA was isolated from adult muscle (m) and kidney (k) of wild-type (wt), Lmo7{Delta}800 homozygotes and Uchl3{Delta}3-7 homozygotes. They were analyzed by northern blotting using hybridization probes that detect each of the genes in the cluster: Lmo7, Uchl3, Ak000009 and Kiaa0603 (from top to bottom). The small panel on the right depicts kidney RNAs prepared from C57BL/6 and 129SvEv animals, hybridized to the Kiaa0603 probe. Molecular size standards are shown to the right of each panel.

 
Based on comparisons between C57BL/6 and 129SvEv RNAs, we established that the difference reflects a strain-specific polymorphism between wild-type C57BL/6 allele and 129SvEv allele of Kiaa0603 on the Uchl3{Delta}3-7and Lmo7{Delta}800 targeted chromosomes (Fig. 3). Preliminary analysis of the two forms of the transcript suggests that the difference is likely to be in the 3'-UTR of the gene (data not shown). Based on previous analysis (22) no genes are predicted to lie 1.3 Mb proximal or 1 Mb distal to the four gene cluster, that is, ~500 kb distal to the distal Lmo7{Delta}800 deletion breakpoint. Given such distances it is unlikely that Lmo7{Delta}800 affects any additional genes, and thus the phenotypes of Lmo7{Delta}800 mutants discussed below are probably due to the removal of Uchl3 and Lmo7 alone.

Lmo7{Delta}800 affects postnatal growth and viability
Although homozygous Lmo7{Delta}800 mice were born at the expected frequency, we noticed that a subset of these mice died during the first weeks after birth, and those that survived were smaller than their littermates. Thirty-three percent (27/81) of Lmo7{Delta}800 homozygotes died before reaching adulthood, with the majority dying in either the first or third weeks after birth. In contrast only 7% (4/57) of Uchl3{Delta}3-7 homozygotes died over the same period of time.

To explore the effects of these mutations on postnatal growth we weighed Lmo7{Delta}800 and Uchl3{Delta}3-7 homozygotes on a regular basis over a period of 3 months (Fig. 4A and B). Crosses between heterozygous and wild-type mice as well as heterozygous intercrosses identified no significant difference between patterns of growth for wild-type and heterozygous animals, allowing us to use heterozygous mice as controls in crosses between heterozygotes and homozygotes (data not shown). Uchl3{Delta}3-7 homozygotes were already smaller than controls during the first week after birth (90% of normal weight, P=0.022 at 2 dpp; Fig. 4A). Beginning in the second week the homozygotes were consistently 80% of normal weight (P=0.0002 at 25 dpp), and this difference persisted thereafter. This observation is consistent with previous results showing that Uchl3{Delta}3-7 homozygous mutants were ~20% smaller than wild-type animals at 80 days of age (31).



View larger version (22K):
[in this window]
[in a new window]
  		Figure 4. Effects of Lmo7{Delta}800 and Uchl3{Delta}3-7 on viability and growth. (A) The growth rates for Uchl3{Delta}3-7 heterozygotes (n=11; blue) and Uchl3{Delta}3-7 homozygotes (n=11; green). (B) The growth rates of Lmo7{Delta}800 heterozygotes (n=30; blue), Lmo7{Delta}800 class 1 mutants (n=20; yellow) and Lmo7{Delta}800 class 2 mutants (n=7; red). Error bars are means±SE. (C) The survival of Uchl3{Delta}3-7 homozygotes (green), Lmo7{Delta}800 class 2 homozygotes (red) and Lmo7{Delta}800class 1 (yellow) are depicted.

 
Unlike Uchl3{Delta}3-7 mice, the Lmo7{Delta}800 homozygous population was not homogeneous. Rather it consisted of two distinct classes based on cluster analysis and the Wilcoxon rank sum test using either the average weight of multiple families or weights within individual families (P>0.05 for the whole period of observation). Class 1 constitutes ~60% (45/73) of all Lmo7{Delta}800 mutants and is similar to Uchl3{Delta}3-7 mutant class. Class 1 mutants are smaller than controls throughout postnatal development and reach about 75–80% normal weight in adulthood (Fig. 4B). One difference between Uchl3{Delta}3-7 mutants and Lmo7{Delta}800 class 1 mutants is that the rate of growth of Lmo7{Delta}800class 1 mutants decreases considerably during the third week of life and reaches zero for a period of 2–3 days.

Class 2 constitutes ~40% of all Lmo7{Delta}800 mutants. This class is more severely affected by the deletion, and 68% die during the first 3 weeks of life (Fig. 4C). The remaining 32% survive, but are severely runted, reaching on average ~60% normal weight in adulthood (Fig. 4B). During the third week of age these mice actually lose weight. We did not observe physiological abnormalities associated with feeding either before or after weaning. Mutant mice of both classes had milk in their stomachs and class 1 mice as well as surviving class 2 mice had a normal food intake (data not shown).

Thus although both Uchl3{Delta}3-7 and Lmo7{Delta}800 mutations cause growth retardation that is already evident at or soon after birth, the larger Lmo7{Delta}800 mutation has a more profound effect on a subset of the mice, with lethality and/or severe runting in 40% of the mutants. The periods of lethality coincide with two important transitions, immediately after birth and at weaning. These transitions are associated with high stress, as indicated by the increase in the circulating level of the stress hormone corticosterone (32,33). This observation suggests that Lmo7 gene function is required during periods of elevated stress, at least in the immediate postnatal period.

Lmo7{Delta}800 leads to muscle degeneration
LMO7 is a member of a family of proteins that are predicted to contain both PDZ and LIM protein–protein interaction domains (34). The LIM domains of several of these proteins have been shown to interact with kinases whereas PDZ domains often associate with the cytoskeleton (35,36). Targeted mutations of two LIM/PDZ proteins, actinin-associated LIM protein (ALP) and ZASP/Cypher/Oracle, resulted in cardiomyopathy and severe congenital myopathy, respectively (34,37). Given the effects of the Ednrbs-1Acrg deletion on mesoderm development, we conducted a histological analysis of a wide array of tissues in 80-day-old class 1 and 2 Lmo7{Delta}800 mutants and littermate controls (see Materials and Methods). The only defect was uncovered in skeletal muscle and was similar in the two classes.

Although the total muscle mass was proportionate to body weight, the number of nuclei in thigh muscle fibers of mutant animals was increased by a factor of 2 and the individual nuclei were elongated compared with control nuclei (Fig. 5A and B). We also observed an occasional increase in the number of muscle fibers, however it was not a consistent phenotype. We next analyzed other muscle tissues, including abdominal wall muscle, diaphragm, psoas muscle as well as cardiac muscle. The nuclear defect was most pronounced in the thigh muscle, but was present in the other tissues as well. Furthermore, the phenotype became more severe in older animals. In addition to the nuclear defect, 170 dpp muscle contained inclusions seen as pale round patches within H&E stained sections of muscle fibers (Fig. 5C, inset). A cross-section of muscle fibers revealed on average one inclusion per fiber. To analyze the inclusions further, frozen muscle biopsies were visualized by electron microscopy, which showed that the inclusions were of tubular sarcoplasmic reticulum origin (Fig. 5C). The inclusions became even more prominent in older animals (300 dpp). No abnormalities were detected in the muscle of 30 dpp animals, suggesting that the defect is progressive. Based on open-field activity studies and the rotorod test, 80-day old mutants did not reveal motor co-ordination impairments (data not shown).



View larger version (87K):
[in this window]
[in a new window]
  		Figure 5. Effects of Lmo7{Delta}800 and Uchl3{Delta}3-7 on muscle. (A) H&E staining of Lmo7{Delta}800/+ thigh muscle (sagittal section). (B) H&E staining of Lmo7{Delta}800/Lmo7{Delta}800 muscle (sagittal section). Green arrows indicate nuclei. (C) Electron micrograph of a single inclusion. Light microscope view of the inclusion is indicated with a black arrow within the inset. The inset shows the transverse section of the Lmo7{Delta}800 homozygous muscle.

 
Uchl3{Delta}3-7 mice exhibited very similar muscle defects, but with lower penetrance and later onset. While 6/6 Lmo7{Delta}800 animals showed a nuclear defect at 80 dpp, only 1/9 Uchl3{Delta}3-7 mice were abnormal at that age. At 170 dpp, 5/8 Uchl3{Delta}3-7 mice had muscle inclusions. Thus, as was the case with the growth defects, the loss of both Lmo7 and Uchl3 exacerbates a defect that is present in Uchl3{Delta}3-7 mutants.

Lmo7{Delta}800 leads to retinal degeneration
To determine the effect of the Lmo7{Delta}800 mutation on behavioral responses, class 1 mice were subjected to a battery of tests that included startle response, prepulse inhibition, habituation, learning and memory, sensory perception, nocioception and tail suspension. These tests were supplemented by alcohol screening, sociosexual behavior analysis and brain and eye histology. A small but statistically significant difference was detected in the tail suspension test, which is thought to measure depression and a response to stress, with mutants spending more time below threshold 1 (260.1 versus 225.68, P=0.01), indicating a delayed response. In the nocioception test mutants showed higher sensitivity to heat represented by shorter latency in response to 50°C hot plate stimulus (3.62 versus 4.73 for controls, P=0.03).

In addition, Lmo7{Delta}800 mutants displayed a pronounced retinal phenotype. Fundus examination of 80–90 dpp Lmo7{Delta}800 mutants revealed retinal thinning and punctuate white spots throughout the fundus that were absent in heterozygous control littermates (data not shown). Histologically, the Lmo7{Delta}800 mice had drastically reduced cell numbers in the outer nuclear layer along with a reduction in the photoreceptor inner (PIS) and outer segments (POS), all suggestive of retinal degeneration (Fig. 6D versus A). Anti-glialfibrillary acidic protein (GFAP), a non-specific marker of retinal injury, was also upregulated in the Lmo7{Delta}800 mutants (Fig. 6E versus B). A closer histologic evaluation of retinal photoreceptors indicated that, in most mutants analyzed, the photoreceptor layer was only one cell layer thick and membranous debris filled the subretinal space (data not shown).



View larger version (107K):
[in this window]
[in a new window]
  		Figure 6. Effects of Lmo7{Delta}800 and Uchl3{Delta}3-7 on retina. Muscle sections were stained with H&E (first column), antisera to GFAP (second column) and antisera to ubiquitin (third column). (A–C) Lmo7{Delta}800 heterozygotes, (D–F) Lmo7{Delta}800 homozygotes, (G–I) Uchl3{Delta}3-7 heterozygotes, (J–L) Uchl3{Delta}3-7 homozygotes. The Lmo7{Delta}800 immunolabeling was performed on frozen sections, and Uchl3{Delta}3-7 immunolabeling was performed on 4% PFA fixed sections. RPE, retinal pigment epithelium; POS, photoreceptor outer segments; PIS, photoreceptor inner segments; ONL, outer nuclear layer; INL, inner nuclear layer; GCL, ganglion cell layer.

 
Uchl3{Delta}3-7 homozygotes exhibited a retinal phenotype that was indistinguishable from Lmo7{Delta}800 mutants with a highly reduced outer nuclear layer thickness and dramatic shortening of the distal photoreceptor region (Fig. 6J versus G). Immunolabeling with antibodies to ubiqutin revealed no differences in the staining intensity between homozygous mutants and heterozygous controls in both Lmo7{Delta}800 (Fig. 6F versus C) and Uchl3{Delta}3-7 animals (Fig. 6L versus I).

To determine whether the retinal phenotype was progressive or a congenital anomaly, we analyzed Lmo7{Delta}800 mice at 30 dpp. At this age, the retinas of both heterozygous and homozygous mutants possessed an outer nuclear layer of ~12 cells in thickness. In addition, photoreceptors contained elongated and properly structured inner and outer segments (data not shown), in marked contrast with the structure of the retina at 3 months post-partum. We conclude that the retinal defects are degenerative and are directly related to the loss of Uchl3 gene function. Uchl3{Delta}3-7 did not exhibit vision impairments based on their normal performance in the visible version of the Morris water maze.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
MATERIALS AND METHODS
 REFERENCES
 
The Lmo7{Delta}800 deletion does not recapitulate the Ednrbs-1Acrg embryonic lethality
The goal of this work was to continue our functional dissection of the 1.5–1.7 Mb region of the genome, termed the Acrg minimal region, that harbors one or more genes essential for embryonic survival. This region, which is thought to contain only four genes, is part of a larger gene desert on mouse chromosome 14 (22). We had previously shown that the removal of one of the genes, Uchl3, could not account for the embryonic lethality of the Acrg deletion (23). The Lmo7{Delta}800 deletion removed Uchl3 and Lmo7 as well as an additional 500 kb downstream of the 3' end of Lmo7 where no additional genes are thought to reside. We found that mice homozygous for this deletion are viable, thus ruling out Lmo7 as a single gene candidate for the Ednrbs-1Acrg embryonic lethality. Furthermore these results eliminate the possibility of a synthetic embryonic lethality caused by the loss of both Uchl3 and Lmo7 genes. If the embryonic lethality can be attributed to the removal of a single gene, then the obvious remaining candidates are Ak000009 and Kiaa0603, the other two genes within the Acrg minimal region.

Alternatively, the lethality could be a synthetic phenotype resulting from the loss of function of some other combination of the four genes within the minimal region, or possibly by misregulation of genes just outside of the Acrg minimal region. For example, the comparative sequence analysis of the Acrg region revealed the presence of many small non-exonic conserved segments whose function remains to be determined (22). However, based on the prediction that no genes lie 1.3 Mb proximal or 1 Mb distal to the cluster of four genes that we identified, there is no reason to suspect that additional genes were affected by the Lmo7{Delta}800 deletion (22). Finally, it is possible that the embryonic lethality requires the loss of one or more genes within the Ednrbs-1Acrg minimal region as well as additional genes within the extent of the entire Acrg deletion. The endpoints of the Acrg minimal region were originally defined by complementation analysis of two radiation-induced deletions, Ednrbs-1Acrg and Ednrbs-36Pub (Fig. 1A). Mice carrying Ednrbs-36Pub die at birth with defects in the skeleton and central nervous system while Ednrbs-1Acrg mice die during early embryogenesis (19). Compound heterozygotes between the two mutants survive embryogenesis and die at birth, with phenotypes similar to Ednrbs-36Pub. Therefore it is possible that the early embryonic lethality associated with the Acrg minimal region requires the removal of genes both within and outside the complementation-defined Acrg minimal region. To test whether heterozygosity for genes outside the minimal region would exacerbate the Lmo7{Delta}800 phenotype, we generated a compound heterozygote between Lmo7{Delta}800 and Ednrbs-1Acrg. Histological and growth analyses of Lmo7{Delta}800/Ednrbs-1Acrg mice were indistinguishable from Lmo7{Delta}800 homozygotes, indicating that embryonic lethality is fully recessive (data not shown).

The Lmo7{Delta}800 deletion exhibits postnatal lethality and growth retardation
While the Lmo7{Delta}800 deletion did not recapitulate the Acrg embryonic lethality, it uncovered postnatal phenotypes that allowed us to assign new functions to both Uchl3 and Lmo7. With the ability to compare the phenotypes of the Uchl3{Delta}3-7 and Lmo7{Delta}800 mutants, we were able to assess the individual contribution of Uchl3 and Lmo7 genes toward the defects observed.

We had reported previously that Uchl3{Delta}3-7 mutants are proportionately smaller than their control littermates from birth, and reach about 80% normal weight in adulthood. Body and organ size in mammals is to a large extent a function of cell number that is dependent on cell cycle regulation. Because the levels of a number of key cell cycle regulators are controlled by ubiquitination, it is possible that the deletion of Uchl3 can lead to a reduced rate of cell division, and as a result, to decreased overall body size (38,39).

The Lmo7{Delta}800mutation leads to a more dramatic growth abnormality than Uchl3{Delta}3-7, producing a severely affected class of mice (class 2) that constitutes ~40% of all mutants. The majority of mice in that class die, while surviving mice are severely runted. Most class 2 Lmo7{Delta}800mutants die either at birth or at weaning, during periods when the levels of corticosterone in mice are sharply elevated, indicating a response to stress (32,33). This suggests that Lmo7 gene function may be required during periods of elevated stress, consistent with the slower stress response of Lmo7{Delta}800mutants in the tail suspension test.

The slower growth of wild-type mice during weaning is exacerbated in the more severely affected Lmo7{Delta}800 mutants, who actually lose weight, and either die or slow down their growth considerably. Even the less severely affected Lmo7{Delta}800 class 1 mutants progress more slowly during this period, indicating that this may be a Lmo7-specific phenotype. During the third week of life complex behavioral and physiological changes occur that accompany the switch from mother's milk to solid food, including the development of feeding controls, gastrointestinal and liver metabolic adaptations, alterations in brain metabolism and hormonal changes (40). The transient cessation of growth at weaning in Lmo7{Delta}800 mutants could result from defects in any of these processes. It is unlikely, however, that the growth defect results from inadequate digestion or absorption of food, due to the absence of diarrhea in the mutants.

The Lmo7{Delta}800 deletion leads to muscle degeneration
Both the Lmo7{Delta}800 and Uchl3{Delta}3-7 mutants displayed morphological features symptomatic of muscular degeneration. These included increases in the number and size of nuclei as well as the age-related appearance of tubular sarcoplasmic reticulum inclusions within muscle fibers. However the Lmo7{Delta}800 mice displayed an earlier onset and higher penetrance of the muscular degeneration phenotype than Uchl3{Delta}3-7mutants. Based on in vitro substrate specificity studies, UCHL3 is thought to be involved in ubiquitin recycling and maintaining free pools of monomeric ubiquitin for proteolysis (41,42). Thus the inclusions probably reflect the accumulation of denatured proteins, similar to the protein aggregates that are characteristic of neurodegenerative disorders such as Parkinson's and Alzheimer's diseases (43). Indeed, as shown previously, Uchl3{Delta}3-7 mutants exhibited dorsal root ganglia cell body degeneration (31). Furthermore, when the mutation in Uchl3 was combined with a mutation in its closely related family member, Uchl1 (Uchl1gad), the double homozygotes exhibited an enhanced number of protein inclusions in the axons of the gracile tract of the medulla and spinal cord (31).

Loss of function of Uchl3 leads to retinal degeneration
The Lmo7{Delta}800 and Uchl3{Delta}3-7 mutants displayed indistinguishable morphological features symptomatic of retinal degeneration, suggesting that this is a Uchl3-related phenotype. The primary features of the degeneration included a highly reduced thickness of both the photoreceptor inner and outer segments and the outer nuclear layer, the retinal layer in which photoreceptor nuclei are located. The retina was morphologically normal at 30 dpp, suggesting that the mutations do not interefere with retinal eye development, but rather cause dramatic and selective degeneration of the photoreceptor cell population with age.

The health and structural maintenance of photoreceptors is highly dependent upon the ability of the retinal pigment epithelia (RPE) to degrade the membranous outer segment packets (44,45). The retinal phenotype of homozygous Uchl3{Delta}3-7 mutants is plausibly explained by a reduction in ubiquitin recycling, leading to a decrease in ubiquitin-dependent phagocytosis and degradation of outer segment membranes, consistent with the appearance of membranous debris in the subretinal space. This phenotype is very similar to what has been described in the Royal College of Surgeons (RCS) rat, in which a mutation in the receptor protein tyrosine kinase-encoding Mertk gene inhibits the RPE from phagocytosing shed photoreceptor outer segments, leading to a progressive loss of rod and cone photoreceptors (46).

Unlike the retinal phenotype, which could be ascribed exclusively to Uchl3 loss-of-function, the growth retardation and muscle degeneration were more severe in Lmo7{Delta}800 mice than in Uchl3{Delta}3-7 mutants. While we cannot rule out the possibility that the increased severity was due to differences in genetic background between the two mutants, we think that this is unlikely. Both mutations were generated on a 129 Sv substrain background, but maintained after outcrossing the chimeras to C57Bl/6 by F1 brother–sister matings. No differences were observed within either strain of mice among the progeny of different F1 intercrosses, and no phenotypic differences were detected after multiple generations. We also carried out a pilot experiment in which we crossed Lmo7{Delta}800 homozygotes to Uchl3{Delta}3-7/Lmo7{Delta}800 compound heterozygotes for two generations. We found one class of Uchl3{Delta}3-7 mutants and two classes of Lmo7{Delta}800 mutants with neonatal lethality observed only among the Lmo7{Delta}800/Lmo7{Delta}800 mice (data not shown). Thus we do not believe that the increased severity of the Lmo7{Delta}800 is due to genetic background effects. Instead it is worth considering whether the increased severity reflects an interaction between Lmo7 and Uchl3 in the ubiquitin pathway. One of the variants of LMO7 contains an F-box domain, originally identified as a subunit of the E3 ubiquitin ligase complex called SCF (47). The LMO7 F-box-containing splice variant is the predominant transcript in brain, skeletal muscle and heart (data not shown). Interestingly, MAFbx, an F-box containing protein, and Atrogin-1, a striated muscle-specific F-box and PDZ domain-containing protein, were shown to be dramatically up-regulated in multiple models of muscle atrophy and were suggested to be critical components in the enhanced muscle proteolysis leading to muscle atrophy (48,49). Thus the lack of Lmo7 in the Lmo7{Delta}800 mutant animals may specifically contribute to the Uchl3-mediated ubiquitin pathway deregulation in heart and skeletal muscle and lead to more severe muscle degeneration.

Likewise, the additional growth defects in Lmo7{Delta}800 could reflect a mild form of dysphagia, an inability to swallow food. In a previous study we showed that mice with homozygous mutations in both Uchl3{Delta}3-7 and a closely related ubiquitin hydrolase Uchl1 display progressive axonal degeneration of the nucleus tractus solitarius, a region essential for central nervous system control of swallowing (31). The double mutants died of starvation due to a failure to feed, while the single mutants were each ~20% smaller than normal. It may be that in combination with loss of Uchl3, the loss of Lmo7 has an impact on protein degradation, just as the loss of Uchl3 increases the severity of the Uchl1 mutant phenotype.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
REFERENCES
 
Library screening, targeting of ES cells and Cre recombination
The 5' Hprt genomic library was screened using a probe located between exon 2 and 3 of Uchl3. The probe was generated from cloned DNA with the following primers: 5'-GCAGAGTTCGACTAAGCTCTT and 5'-GCTACGGATAGGCAGTGAAGC. The 3' Hprt library was screened with a probe located downstream of the distal breakpoint of the Acrg minimal region, using the primers 5'-GAGGGCTGTCCCTCTCTTC and 5'-ATCCTTCAGGCAATTCAGGGG and the mouse AC080021 BAC. Two positives were obtained from the 5' Hprt library and one positive from the 3' Hprt library. Positive clones were purified and used to infect the Cre-expressing bacterial strain BNN132 to excise the clones as plasmids. The plasmids were digested with AscI to flip the inserts to obtain the proper orientation of the inserts relative to the loxP sites. Following restriction mapping, gaps were generated in each insert to facilitate insertional targeting of the plasmid into ES cells. A 1.1 kb KpnI/SmaI gap was generated in the 5' Hprt plasmid leaving 2.9 and 4.3 kb arms. A similar gap was generated in the 3' Hprt clone using HpaI, leaving 1.3 and 7.5 kb arms. The resulting vectors were used for ES cell targeting.

Hprt-deficient AB2.2 129/SvEv ES cells were obtained from Lexicon and grown on neomycin resistant PMEF feeder cells and then switched to neomycin and puromycin resistant SNL feeder cells, a gift from Dr Alan Bradley's laboratory. The 5' Hprt gapped vector was electroporated into the ES cells, and neomycin resistant clones were obtained. Ten positive clones were electroporated with the 3' Hprt gapped vector and selected for puromycin resistance. Six double-targeted ES clones were electroporated with supercoiled plasmid pOG231, which encodes Cre recombinase driven by the CMV promoter. After 48 h cells were transferred to HAT (hypoxanthine, aminopterin, thymidine)-containing medium for 1 week followed by HT (hypoxanthine, thymidine)-containing medium for 3 days before picking colonies. HAT-resistant clones were obtained with 0.3% efficiency. PCR with the primers 5'-CCTCATGGACTAATTATGGAC and 5'-CCAGTTTCACTAATGACACA, which are within exons 2 and 9 of the human Hprt minigene, respectively, was used to diagnose cells that had undergone recombination within the Hprt gene. This was verified by Southern analysis.

Generation of chimeric animals
ES cells carrying the deletion were injected into C57BL/6 blastocysts, which were implanted into pseudopregnant mice. Chimeras were bred to C57BL/6 mice and tail DNA of all progeny was genotyped by PCR using the Hprt primers described above. In F1 intercrosses primers contained within the Lmo7 gene (5'-GCCCTAGATCCCGACTTAG and 5'-CTCTCAACCATCAGCCGC) were used to detect the presence of the wild-type chromosome.

DNA FISH analysis
Metaphase chromosome spreads from ES cells cultures were prepared from double-targeted cells before and after Cre excision. For FISH analysis, slides were treated with RNAse, washed quickly in 2xSSC, and dehydrated at 25°C in 70, 90 and 100% ethanol for 2 min each. DNA denaturation was achieved by treatment with 70% formamide, 2xSSC at 70°C for 2 min. The slides were transferred to ice cold 70% ethanol for 2 min and dehydrated at room temperature (RT) in 70, 90, and 100% ethanol for 2 min each before hybridization to chromosome-specific probes. BAC AC074357 was biotin-labeled and BAC AC079638 was digoxigenin-labeled. After hybridization at 37°C overnight, slides were washed, blocked and treated with rhodamine-labeled anti-digoxigenin antibody (Roche) followed by treatment with FITC-conjugated avidin. Before viewing, DNA was stained with DAPI.

Southern and northern analysis
Southern blots were hybridized in Church Buffer (50) at 65°C and washed in 0.1xSSC/0.1% SDS at 23 and 65°C. Northern blots were hybridized in ExpressHyb (Clontech, Catalog no. 8015-2) according to the manufacturer's instructions. Radiolabeled probes were prepared from the corresponding mouse cDNA clones: Kiaa0603 (AW988132), Ak000009 (5' fragment of AA915052), Uchl3 (exons 1–2, 8–10), Lmo7 (exons 23–27).

Growth analysis
To analyze the effect of the mutations on growth, mice derived from seven Lmo7{Delta}800/+xLmo7{Delta}800/Lmo7{Delta}800 families and four Uchl3{Delta}3-7/+xUchl3{Delta}3-7/Uchl3{Delta}3-7 families were weighed in the afternoon every 2 days for the first 30 days after birth (day 0 dpp) followed by every 5 days until they reached 60 dpp and every 10 days thereafter. The data were subjected to cluster analysis based on principal components analysis that identified groups (clusters) with common characteristics that are maintained for the whole period of observation. Initial data classification was carried out according to standard procedure ISODATA followed by the Wilcoxon rank sum test analysis.

Histological analysis
Brain, thymus, heart, lung, liver, pancreas, stomach, spleen, small bowel, colon, kidney, adrena, lymph node, skeletal muscle, uterus and testes were analyzed histologically. All tissues were fixed in 4% paraformaldehyde, sectioned, stained with hematoxylin and eosin (H&E) and analyzed microscopically. Muscle tissue was fixed in 4% paraformaldehyde or frozen in isopentane in liquid nitrogen. Frozen muscle tissue was postfixed in 1% osmium tetroxide and stained with 2% aqueous uranyl acetate. Ultrathin 60–90 nm sections were cut using a diamond knife on a RMC MT6000 ultramicrotome and stained with 2% aqueous uranyl acetate and lead citrate and viewed with a Philips EM400 transmission electron microscope.

Retinal analysis
Mice were lightly anesthetized with an intraperitoneal injection of Avertin (1.25% 2,2,2-tribromoethanol and 0.8% tert-pentyl alcohol in water, 0.3 ml). To facilitate examination of the fundus, pupils were dilated with 1% Cyclomydril ophthalmic drops (Alcon Pharmaceuticals, Fort Worth, TX, USA). The fundus was evaluated by indirect ophthalmoscopy and photographs were taken with a Kowa Genesis small animal fundus camera (Torrance, CA, USA) along with the aid of a 90 diopter condensing lens (Volk, Mentor, OH, USA) as described by Hawes et al. (51). Eyes that were used for gross ocular histology and immunohistochemistry were either fresh frozen on dry ice and embedded in TBS tissue freezing medium (IMEB Inc., Chicago, IL, USA) followed by cryosectioning or were fixed in 4% paraformaldehyde followed by paraffin sectioning using standard protocols. Representative tissue sections were stained with H&E to facilitate visualization of ocular structures. To visualize specific antigen localization, anti-GFAP (six drops per 1000 µl; Immunon, Pittsburgh, PA, USA) and anti-ubiquitin (1 : 200; Dako, Carpinteria, CA, USA) were utilized in conjunction with the VECTASTAIN ABC kit (Burlingame, CA, USA) according to manufacturer's protocols. To allow for a more thorough evaluation of photoreceptor structure, additional eyes were fixed in 2.5% glutaraldehyde, post-fixed in 1% osmium tetroxide and embedded in Araldite/Embed 812 (Electron Microscopy Sciences, Fort Washington, PA, USA). One micron-thick sections were sectioned through the posterior pole of the eye and were stained with Toluidine blue O. All tissue sections were viewed on a Nikon Eclipse E800 compound microscope equipped with a CoolSnap Color Camera (Photometrics, Tucson, AZ, USA), and the images were collected with MetaMorph Imaging System software (Universal Imaging Corporation, West Chester, PA, USA).


   ACKNOWLEDGEMENTS
 
We would like to thank Dr Allan Bradley for providing us with the 5' and 3' Hprt libraries and SNL feeder cells. We wish to thank Marcelo Wood, Ted Abel and members of the Tennessee Mouse Genome Consortium, particularly Drs Dabney Johnson and Karen Goss for the initial behavioral analysis of the mutant mice. We also thank the University of California at Davis Mutant Mouse Histopathology Laboratories, particularly Dr Alexander Borowsky, for the histological analysis of muscle, as well as the Integrated Microscopy Center at the University of Memphis for assistance with the histological processing of the eyes. We are also grateful to Robert Ingram, Drs Laurie Jo Kurihara and Aaron Bowman as well as Dr Susan Henning, Baylor College of Medicine, for many helpful discussions. This work was supported by a predoctoral fellowship to E.S. from the New Jersey Commission on Cancer Research, by the Howard Hughes Medical Institute and an unrestricted grant from Research to Prevent Blindness Inc., New York, to the Department of Ophthalmology at the University of Tennessee Health Science Center. S.M.T. was an Investigator of the Howard Hughes Medical Institute.


   FOOTNOTES
 
* To whom correspondence should be addressed at: One Nassau Hall, Princeton University, Princeton, NJ 08544, USA. Tel: +1 6092586100; Fax: +1 6092581615; Email: smt{at}princeton.edu Back


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 

  1. Waterston, R.H., Lindblad-Toh, K., Birney, E., Rogers, J., Abril, J.F., Agarwal, P., Agarwala, R., Ainscough, R., Alexandersson, M., An, P. et al. (2002) Initial sequencing and comparative analysis of the mouse genome. Nature, 420, 520–562.[CrossRef][Medline]

  2. Venter, J.C., Adams, M.D., Myers, E.W., Li, P.W., Mural, R.J., Sutton, G.G., Smith, H.O., Yandell, M., Evans, C.A., Holt, R.A. et al. (2001) The sequence of the human genome. Science, 291, 1304–1351.[Abstract/Free Full Text]

  3. Lander, E.S., Linton, L.M., Birren, B., Nusbaum, C., Zody, M.C., Baldwin, J., Devon, K., Dewar, K., Doyle, M., FitzHugh, W. et al. (2001) Initial sequencing and analysis of the human genome. Nature, 409, 860–921.[CrossRef][Medline]

  4. Davis, A.P. and Justice, M.J. (1998) An Oak Ridge legacy: the specific locus test and its role in mouse mutagenesis. Genetics, 148, 7–12.[Free Full Text]

  5. Thomas, J.W., LaMantia, C. and Magnuson, T. (1998) X-ray-induced mutations in mouse embryonic stem cells. Proc. Natl Acad. Sci. USA, 95, 1114–1119.[Abstract/Free Full Text]

  6. You, Y., Browning, V.L. and Schimenti, J.C. (1997) Generation of radiation-induced deletion complexes in the mouse genome using embryonic stem cells. Methods, 13, 409–421.[CrossRef][ISI][Medline]

  7. Ramirez-Solis, R., Liu, P. and Bradley, A. (1995) Chromosome engineering in mice. Nature, 378, 720–724.[CrossRef][Medline]

  8. Smith, A.J., De Sousa, M.A., Kwabi-Addo, B., Heppell-Parton, A., Impey, H. and Rabbitts, P. (1995) A site-directed chromosomal translocation induced in embryonic stem cells by Cre-loxP recombination. Nat. Genet., 9, 376–385.[CrossRef][ISI][Medline]

  9. Cleary, M.A., van Raamsdonk, C.D., Levorse, J., Zheng, B., Bradley, A. and Tilghman, S.M. (2001) Disruption of an imprinted gene cluster by a targeted chromosomal translocation in mice. Nat. Genet., 29, 78–82.[CrossRef][ISI][Medline]

  10. Buchholz, F., Refaeli, Y., Trumpp, A. and Bishop, J.M. (2000) Inducible chromosomal translocation of AML1 and ETO genes through Cre/loxP-mediated recombination in the mouse. EMBO Rep., 1, 133–139.[CrossRef][ISI][Medline]

  11. Collins, E.C., Pannell, R., Simpson, E.M., Forster, A. and Rabbitts, T.H. (2000) Inter-chromosomal recombination of Mll and Af9 genes mediated by cre-loxP in mouse development. EMBO Rep., 1, 127–132.[CrossRef][ISI][Medline]

  12. Lindsay, E.A., Botta, A., Jurecic, V., Carattini-Rivera, S., Cheah, Y.C., Rosenblatt, H.M., Bradley, A. and Baldini, A. (1999) Congenital heart disease in mice deficient for the DiGeorge syndrome region. Nature, 401, 379–383.[CrossRef][Medline]

  13. Tsai, T.F., Jiang, Y.H., Bressler, J., Armstrong, D. and Beaudet, A.L. (1999) Paternal deletion from Snrpn to Ube3a in the mouse causes hypotonia, growth retardation and partial lethality and provides evidence for a gene contributing to Prader–Willi syndrome. Hum. Mol. Genet., 8, 1357–1364.[Abstract/Free Full Text]

  14. Yu, Y. and Bradley, A. (2001) Engineering chromosomal rearrangements in mice. Nat. Rev. Genet., 2, 780–790.[CrossRef][ISI][Medline]

  15. Schumacher, A., Faust, C. and Magnuson, T. (1996) Positional cloning of a global regulator of anterior-posterior patterning in mice. Nature, 384, 648.[Medline]

  16. Bell, J.A., Rinchik, E.M., Raymond, S., Suffolk, R. and Jackson, I.J. (1995) A high-resolution map of the brown (b, Tyrp1) deletion complex of mouse chromosome 4. Mamm. Genome, 6, 389–395.[CrossRef][ISI][Medline]

  17. Hosoda, K., Hammer, R.E., Richardson, J.A., Baynash, A.G., Cheung, J.C., Giaid, A. and Yanagisawa, M. (1994) Targeted and natural (piebald-lethal) mutations of endothelin-B receptor gene produce megacolon associated with spotted coat color in mice. Cell, 79, 1267–1276.[CrossRef][ISI][Medline]

  18. Shin, M.K., Levorse, J.M., Ingram, R.S. and Tilghman, S.M. (1999) The temporal requirement for endothelin receptor-B signalling during neural crest development. Nature, 402, 496–501.[CrossRef][Medline]

  19. O'Brien, T.P., Metallinos, D.L., Chen, H., Shin, M.K. and Tilghman, S.M. (1996) Complementation mapping of skeletal and central nervous system abnormalities in mice of the piebald deletion complex. Genetics, 143, 447–461.[Abstract]

  20. Peterson, K.A., King, B.L., Hagge-Greenberg, A., Roix, J.J., Bult, C.J. and O'Brien, T.P. (2002) Functional and comparative genomic analysis of the piebald deletion region of mouse chromosome 14. Genomics, 80, 172–184.[CrossRef][ISI][Medline]

  21. Welsh, I.C. and O'Brien, T.P. (2000) Loss of late primitive streak mesoderm and interruption of left-right morphogenesis in the Ednrb(s-1Acrg) mutant mouse. Dev. Biol., 225, 151–168.[CrossRef][ISI][Medline]

  22. Kurihara, L.J., Semenova, E., Miller, W., Ingram, R.S., Guan, X.J. and Tilghman, S.M. (2002) Candidate genes required for embryonic development: a comparative analysis of distal mouse chromosome 14 and human chromosome 13q22. Genomics, 79, 154–161.[CrossRef][ISI][Medline]

  23. Kurihara, L.J., Semenova, E., Levorse, J.M. and Tilghman, S.M. (2000) Expression and functional analysis of Uch-L3 during mouse development. Mol. Cell. Biol., 20, 2498–2504.[Abstract/Free Full Text]

  24. Harris, B.Z. and Lim, W.A. (2001) Mechanism and role of PDZ domains in signaling complex assembly. J. Cell. Sci., 114, 3219–3231.[ISI][Medline]

  25. Hung, A.Y. and Sheng, M. (2002) PDZ domains: structural modules for protein complex assembly. J. Biol. Chem., 277, 5699–5702.[Free Full Text]

  26. Retaux, S. and Bachy, I. (2002) A short history of LIM domains (1993–2002): from protein interaction to degradation. Mol. Neurobiol., 26, 269–281.[CrossRef][ISI][Medline]

  27. Bach, I. (2000) The LIM domain: regulation by association. Mech. Dev., 91, 5–17.[CrossRef][ISI][Medline]

  28. Ikeda, W., Nakanishi, H., Miyoshi, J., Mandai, K., Ishizaki, H., Tanaka, M., Togawa, A., Takahashi, K., Nishioka, H., Yoshida, H. et al. (1999) Afadin: A key molecule essential for structural organization of cell–cell junctions of polarized epithelia during embryogenesis. J. Cell. Biol., 146, 1117–1132.[Abstract/Free Full Text]

  29. Zheng, B., Mills, A.A. and Bradley, A. (1999) A system for rapid generation of coat color-tagged knockouts and defined chromosomal rearrangements in mice. Nucl. Acids Res., 27, 2354–2360.[Abstract/Free Full Text]

  30. Zheng, B., Sage, M., Sheppeard, E.A., Jurecic, V. and Bradley, A. (2000) Engineering mouse chromosomes with Cre-loxP: range, efficiency, and somatic applications. Mol. Cell. Biol., 20, 648–655.[Abstract/Free Full Text]

  31. Kurihara, L.J., Kikuchi, T., Wada, K. and Tilghman, S.M. (2001) Loss of Uch-L1 and Uch-L3 leads to neurodegeneration, posterior paralysis and dysphagia. Hum. Mol. Genet., 10, 1963–1970.[Abstract/Free Full Text]

  32. Martin, R.J. and Gahagan, J.H. (1977) The influence of age and fasting on serum hormones in the lean and obese-Zucker rat. Proc. Soc. Exp. Biol. Med., 154, 610–614.[Medline]

  33. Diez, J.A., Sze, P.Y. and Ginsburg, B.E. (1976) Postnatal development of mouse plasma and brain corticosterone levels: new findings contingent upon the use of a competitive protein-binding assay. Endocrinology, 98, 1434–1442.[Abstract]

  34. Zhou, Q., Chu, P.H., Huang, C., Cheng, C.F., Martone, M.E., Knoll, G., Shelton, G.D., Evans, S. and Chen, J. (2001) Ablation of Cypher, a PDZ-LIM domain Z-line protein, causes a severe form of congenital myopathy. J. Cell. Biol., 155, 605–612.[Abstract/Free Full Text]

  35. Nakagawa, N., Hoshijima, M., Oyasu, M., Saito, N., Tanizawa, K. and Kuroda, S. (2000) ENH, containing PDZ and LIM domains, heart/skeletal muscle-specific protein, associates with cytoskeletal proteins through the PDZ domain. Biochem. Biophys. Res. Commun., 272, 505–512.[CrossRef][ISI][Medline]

  36. Guy, P.M., Kenny, D.A. and Gill, G.N. (1999) The PDZ domain of the LIM protein enigma binds to beta-tropomyosin. Mol. Biol. Cell., 10, 1973–1984.[Abstract/Free Full Text]

  37. Pashmforoush, M., Pomies, P., Peterson, K.L., Kubalak, S., Ross, J. Jr, Hefti, A., Aebi, U., Beckerle, M.C. and Chien, K.R. (2001) Adult mice deficient in actinin-associated LIM-domain protein reveal a developmental pathway for right ventricular cardiomyopathy. Nat. Med., 7, 591–597.[CrossRef][ISI][Medline]

  38. Trumpp, A., Refaeli, Y., Oskarsson, T., Gasser, S., Murphy, M., Martin, G.R. and Bishop, J.M. (2001) c-Myc regulates mammalian body size by controlling cell number but not cell size. Nature, 414, 768–773.[CrossRef][Medline]

  39. Tworkowski, K.A., Salghetti, S.E. and Tansey, W.P. (2002) Stable and unstable pools of Myc protein exist in human cells. Oncogene, 21, 8515–8520.[CrossRef][ISI][Medline]

  40. Henning, S.J. (1981) Postnatal development: coordination of feeding, digestion, and metabolism. Am. J. Physiol., 241, G199–214.[Medline]

  41. Larsen, C.N., Krantz, B.A. and Wilkinson, K.D. (1998) Substrate specificity of deubiquitinating enzymes: ubiquitin C-terminal hydrolases. Biochemistry, 37, 3358–3368.[CrossRef][Medline]