SEARCH

SEARCH BY CITATION

Keywords:

  • Crim1;
  • gene-trap;
  • organogenesis;
  • microphthalmia;
  • syndactyly;
  • skin blebbing

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

Crim1 is a transmembrane protein, containing six vWF-C type cysteine-rich repeats, that tethers growth factors to the cell surface. A mouse line, KST264, generated in a LacZ insertion mutagenesis gene-trap screen, was examined to elucidate Crim1 function in development. We showed that Crim1KST264/KST264 mice were not null for Crim1 due to the production of a shortened protein isoform. These mice are likely to represent an effective hypomorph or a dominant-negative for Crim1. Transgene expression recapitulated known Crim1 expression in lens, brain, and limb, but also revealed expression in the smooth muscle cells of the developing heart and renal vasculature, developing cartilage, mature ovary and detrusor of the bladder. Transgene expression was also observed in glomerular epithelial cells, podocytes, mesangial cells, and urothelium in the kidney. Crim1KST264/KST264 mice displayed perinatal lethality, syndactyly, eye, and kidney abnormalities. The severe and complex phenotype observed in Crim1KST264/KST264 mice highlights the importance of Crim1 in numerous aspects of organogenesis. Developmental Dynamics 236:502–511, 2007. © 2006 Wiley-Liss, Inc.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

There are a growing number of proteins described that contain vWF-C type cysteine-rich repeats (CRR), including proteins important in development, such as chordin, kielin, and amnionless, as well as the extracellular matrix component procollagen IIa (Garcia Abreu et al.,2002). One such protein, Crim1, was identified as a protein expressed in a spatially and temporally restricted manner during organogenesis of the limbs, kidney, lens, pinna, erupting teeth, and testis (Georgas et al.,2000; Kolle et al.,2000; Lovicu et al.,2000). The involvement of Crim1 in blood vessel biology was suggested by the observation that knockdown of Crim1 expression in human umbilical vein endothelial cells disrupted formation of vascular tubes in culture (Glienke et al.,2002). In addition, morpholino-mediated knockdown of Crim1 in the zebrafish embryo resulted in aberrant formation of the intersegmental vessels and dorsal longitudinal anastomotic vessel, further implicating Crim1 in vascular development (Kinna et al.,2006).

The presence of CRR motifs within Crim1 suggested that Crim1 may act as a modulator of the action of members of the transforming growth factor-beta (TGF-β) superfamily, as does chordin (Larrain et al.,2000). We have demonstrated that Crim1 can bind to bone morphogenetic protein (BMP) -2, -4 and -7, but only within the context of the cell and not in solution, suggesting a cell-autonomous mode of action (Wilkinson et al.,2003). Crim1 forms a complex with the preprotein forms of such ligands, tethering the preproteins to the cell surface, thus retarding their secretion as mature active dimers (Wilkinson et al.,2003). While possibly antagonistic, Crim1-ligand binding may act to prolong ligand activity by allowing slow release of the mature protein or by restricting the distance over which ligand can act.

In this study, we have investigated the role of Crim1 in mouse development by characterizing the KST264 mouse line. This line was generated as part of a previously reported gene-trap screen designed to identify novel proteins containing signal sequences (secreted and transmembrane proteins; Skarnes et al.,1995; Leighton et al.,2001). KST264 mice have a β-galactosidase/neomycinres (β-Geo) cassette inserted into intron 1 of the Crim1 gene, which results in a fusion of the coding region of exon 1 of the Crim1 gene and β-Geo. We refer to homozygotes for this insertion as Crim1KST264/KST264 mice. This insertion results in perinatal lethality in homozygous animals with defects in a variety of organ systems, including the limbs, eyes, and kidneys. We show that the Crim1KST264/KST264 mouse line is not a “null” mutant but retains expression of a minor, alternatively spliced isoform of Crim1. Analysis of transgene expression in Crim1+/KST264 mice has revealed additional sites of Crim1 expression, particularly in the smooth muscle cells of specific developing arterial vascular beds.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

Crim1KST264/KST264 Mice Display a Disruption of Crim1 Expression, but Are Not Null

The present study used a mouse line mutant for Crim1, KST264, generated in a secretory gene-trap screen (Leighton et al.,2001; Fig. 1A). We have used this mouse line to further our understanding of Crim1 biology. We began by confirming that the mouse line represented a bona fide Crim1 mutation and analyzing Crim1 transcript levels. Reverse transcriptase-polymerase chain reaction (RT-PCR) was performed on kidney mRNA from wild-type, heterozygous Crim1+/KST264, and homozygous Crim1KST264/KST264 15.5 days post coitum (dpc) embryos (Fig. 1B). PCR using a Crim1 exon 1 primer and a β-Geo primer on heterozygous Crim1+/KST264 15.5 dpc kidney cDNA yielded an amplicon of 1,046 bp, but none was apparent from wild-type cDNA. Subcloning and sequencing of the RT-PCR amplicon confirmed that the transcript was chimeric for exon 1 of Crim1 and the transgene (data not shown), suggesting that the gene-trap cassette was inserted into intron 1.

thumbnail image

Figure 1. The KST264 line is a Crim1 mutant. A: Ideogram of the Crim1 locus and a schematic representation of the transgene, insertion site of the gene-trap and Crim1 mRNA splice variants (discussed below) (SA, splice acceptor; TM, transmembrane domain; PLAP, placental alkaline phosphatase; β-Geo, a fusion between β-galactosidase and neomycin phosphotransferase; IRES, internal ribosome entry site; pA, polyadenylation site). B: Reverse transcriptase-polymerase chain reaction (RT-PCR) of wild-type (+/+) and heterozygous Crim1+/KST264 (+/KST264) 15.5 days post coitum (dpc) kidney mRNA, showing amplification of a 1,046-bp amplicon using a Crim1 exon 1 primer and a β-Geo primer in the +/KST264 lane.C: RT-PCR analysis of wild-type (+/+) and homozygous Crim1KST264/KST264 (KST264/KST264) 15.5 dpc kidney mRNA. An mRNA splice variant lacking exon 2 (261 bp) is weakly expressed in wild-type samples but appears more prominently in Crim1KST264/KST264 homozygous kidneys (KST264/KST264) (X1-X3). Transcripts containing exon 2 were not detected in Crim1KST264/KST264 kidneys (X2). RT-PCR using primers to the 3′ portion (exons 12 and 16) of Crim1 also indicate a down-regulation of Crim1 in Crim1KST264/KST264 kidneys (X12-X16). Primers to glyceraldehyde-3-phosphate dehydrogenase (GAPDH) were used as a control. D: Quantitative RT-PCR using real-time PCR was conducted to determine the change in Crim1 transcript levels between wild-type, Crim1+/KST264 and Crim1KST264/KST264 15.5 dpc kidneys. Primer pairs specific for exon 1, exon 2, exon 3, exons 1–3, exon 11, and exon 17 were used and transcript levels expressed as a proportion of wild-type levels. Solid bars, wild-type; striped bars, Crim1+/KST264; open bars, Crim1KST264/KST264. Error bars represent the standard deviation of the mean (SDM). E: RT-PCR reveals the existence of an exon 2-minus splice variant of Crim1 in wild-type 8.5 dpc embryos and wild-type 15.5 dpc lung, eye, rib, and heart. This transcript is also present in 8.5 dpc and 10.5 dpc Crim1KST264/KST264 embryos, 13.5 dpc Crim1KST264/KST264 hindlimb, and 15.5 dpc Crim1KST264/KST264 lung, brain, eye, rib, and heart. F: Crim1 protein in the membrane fraction of 17.5 dpc wild-type, Crim1+/KST264, and Crim1KST264/KST264 whole kidney extracts. Note that the predominant band of almost 150 kDa in wild-type sample diminishes in Crim1+/KST264 sample. Conversely, an immunoreactive protein of just over 25 kDa in size is abundant in the Crim1KST264/KST264 sample, detectable in the Crim1+/KST264 sample, and undetectable in the wild-type sample.

Download figure to PowerPoint

RT-PCR on 15.5 dpc embryonic kidney mRNA with primers designed to exon 1 and exon 3 produced an amplicon (435 bp) of the expected size for the reported transcript in wild-type samples. No such amplicon was detected in Crim1KST264/KST264 samples (Fig. 1C). However, a shorter amplicon (261 bp) was apparent in both wild-type and Crim1KST264/KST264 samples. Subcloning and sequencing of this RT-PCR amplicon from Crim1KST264/KST264 homozygotes demonstrated that an in-frame Crim1 transcript lacking exon 2 was produced (data not shown). Furthermore, RT-PCR with exon 2-specific primers failed to detect any evidence of a transcript containing exon 2 in Crim1KST264/KST264 homozygotes. Analysis of the 3′ portion of the Crim1 transcript with RT-PCR using primers to exons 12 and 16 produced an amplicon of the expected size (797 bp) in wild-type and Crim1KST264/KST264 samples, in addition to a shorter, minor PCR product (Fig. 1C). The level of transcript in Crim1KST264/KST264 homozygotes appeared reduced relative to the wild-type. This finding was confirmed by quantitative real-time PCR on whole kidney cDNA from 15.5 dpc wild-type, Crim1+/KST264, and Crim1KST264/KST264 embryos (Fig. 1D). Relative to wild-type, we found similar mRNA levels for Crim1+/KST264 transcripts with the exception of exon 2-containing transcripts, that were somewhat reduced. Transcript levels for Crim1KST264/KST264, however, were greatly diminished, with exon 2-containing transcripts remaining undetectable.

We also tested for the presence of an exon 2-minus Crim1 transcript in other tissues. We found that exon 2-minus splice variants were present in wild-type 8.5 dpc embryos and 15.5 dpc lung, eye, rib, and heart (Fig. 1E). Moreover, RT-PCR revealed exon 2-minus Crim1 transcripts were detectable in 8.5 dpc and 10.5 dpc Crim1KST264/KST264 embryos, 13.5 dpc Crim1KST264/KST264 hindlimb and 15.5 dpc Crim1KST264/KST264 lung, brain, eye, ribs, and heart (Fig. 1E). Taken together, the data demonstrate that there is at least one minor splice variant of Crim1, which splices around the gene-trap cassette that lacks exon 2, and continues to be expressed in the Crim1KST264/KST264 homozygotes. This variant would be expected to have a deletion of 57 amino acids in a cysteine-rich region N-terminal to the first CRR repeat.

To determine whether there was a change in Crim1 protein levels in the mutant mice, we performed immunoblotting on the membrane fraction of 17.5 dpc wild-type, Crim1+/KST264, and Crim1KST264/KST264 whole kidney extracts. Using an antibody that detects the C-terminal, cytoplasmic portion of Crim1 (Wilkinson et al.,2003), we detected a predominant band of almost 150 kDa in wild-type samples (Fig. 1F), comparable to that expected for full-length protein (Wilkinson et al.,2003). This immunoreactive protein was slightly diminished in Crim1+/KST264 samples and faintly detected in Crim1KST264/KST264 samples. Conversely, an immunoreactive protein of just over 25 kDa was abundant in the Crim1KST264/KST264 sample, detectable in the Crim1+/KST264 sample, and undetectable in the wild-type sample. Crim1KST264/KST264 homozygous mice retain isoforms of Crim1 mRNA and residual protein and, therefore, cannot be regarded as null. Therefore, we use the terminology Crim1+/KST264 and Crim1KST264/KST264 for mice heterozygous and homozygous for this mutation, respectively.

Expression of the Transgene Reveals Additional Sites of Crim1 Expression

In Crim1+/KST264 adult kidneys, Crim1-β-Geo was expressed strongly in the entire arterial plexus, including the afferent arterioles, but was not detected in the venous plexus (Fig. 2A). Section data of adult Crim1+/KST264 kidneys revealed X-Gal staining in the GECs, podocytes, mesangial cells, proximal tubules (Fig. 2B), and distal tubules (Fig. 2D). To verify that no artifacts were introduced due to the method chosen for X-Gal staining, Crim1+/KST264 adult kidneys were cryosectioned then incubated with X-Gal directly on the tissue section. The staining pattern observed for these tissues (Fig. 2C,E) was consistent with those stained in whole-mount and sectioned after paraffin embedding.

thumbnail image

Figure 2. Expression of β-galactosidase in Crim1+/KST264 heterozygotes reveal novel sites of Crim1 expression. A: An X-Gal stained Crim1+/KST264 adult kidney shown in hemi-mount reveals Crim1-β-gal expression in an afferent artery and arterioles (arrow) and glomeruli (arrowheads). B: A cross-section of an adult Crim1+/KST264 kidney reveals Crim1-β-gal expression in proximal tubules (arrow), mesangial cells (double arrowhead), glomerular epithelial cells (open arrowhead), and podocytes (arrowhead). C: A section of an adult Crim1+/KST264 kidney comparable to that in B after cryosectioning and X-Gal staining was performed directly on the section. Note that a pattern of expression was observed similar to the adult Crim1+/KST264 kidney in B, with Crim1-β-gal expression in vascular smooth muscle (open arrowhead) and structures of the glomerulus (arrow). D: A cross-section of an adult Crim1+/KST264 kidney reveals Crim1-β-gal expression in distal tubules (arrow). E: A section of an adult Crim1+/KST264 kidney comparable to that in D after cryosectioning and X-Gal staining was performed directly on the section. Again, a pattern of expression was observed similar to the adult Crim1+/KST264 kidney in D, with Crim1-β-gal expression in distal tubules (arrow). F: Posterior view of an 8.5 days post coitum (dpc) Crim1+/KST264 embryo showing Crim1-β-gal expression in the yolk sac (arrow). Note the localized expression in the blood islands of the yolk sac (open arrowhead). G: A 10.5 dpc Crim1+/KST264 embryo showing Crim1-β-gal expression in the heart (double arrowhead), eye, branchial arches (arrowhead), and restricted sites of the brain (open arrowhead) and spinal chord (arrow). H: A 12.5 dpc Crim1+/KST264 embryo showing Crim1-β-gal expression in the eye, the primordium of the pinna (arrow), and spinal cord (open arrowhead). I: A 15.5 dpc Crim1+/KST264 embryo showing Crim1-β-gal expression in the eye, cartilage within the pinna (arrow), and vibrissae (open arrowhead). J: A cross-section of the embryo in G showing Crim1-β-gal expression in the endothelium of the dorsal aorta (open arrowhead) and the motor neurons (arrow). K: A cross-section of the embryo in G showing Crim1-β-gal expression in the pericardium (arrow) and epicardial cells adjacent to the myocardium (open arrowheads). L: A forelimb of a 12.5 dpc Crim1+/KST264 embryo showing Crim1-β-gal expression associated with the developing digits. M: Crim1-β-gal expression persists in the forelimb of a 15.5 dpc Crim1+/KST264 embryo. N: A cross-section of a 16.5 dpc Crim1+/KST264 forelimb showing Crim1-β-gal expression in the basal layer of the epidermis (arrow) and cartilage condensates (open arrowhead). O: The heart of a 17.5 dpc Crim1+/KST264 embryo after X-gal staining. Note the expression of Crim1-β-gal in the large coronary artery (arrow) and adventitia of the outflow vessels (open arrowhead), as well as the epicardial covering of the ventricles and atria. P–R: Cross-sections of the heart in (O) detailing Crim1-β-gal expression in epicardial cells (arrows, P). Q: Crim1-β-gal is expressed in the smooth muscle cells (arrow) and the endothelium (open arrowhead) of a coronary artery (lumen denoted by an asterisk). R: Crim1-β-gal expression in endocardial endothelial cells of a trabeculated portion of the left ventricle (open arrowheads). S,T: The 15.5 dpc Crim1+/KST264 embryonic kidney sections reveal Crim1-β-gal expression in the vascular smooth muscle (arrow, S) and the urothelium (open arrowhead, S), in addition to the developing tubules (arrows, T), mesangial cells (double arrowhead, T), glomerular epithelial cells (open arrowhead, T), and podocytes (arrowhead, T). U: Expression of Crim1-β-gal in the suprarenal ganglion of the adrenal gland (arrow) and cells of the adrenal gland proper (open arrowheads) of a 15.5 dpc Crim1+/KST264 embryo. V: Expression of Crim1-β-gal in trophoblasts of the placenta of a 17.5 dpc Crim1+/KST264 embryo. W: Expression of Crim1-β-gal in the epithelium of the 17.5 dpc Crim1+/KST264 lung. X: Expression in the basal layer of the epidermis (arrow) and epithelium of a developing hair follicle (open arrow) in the skin of a 17.5 dpc Crim1+/KST264 embryo. Y: Expression in the transitional epithelium (arrowhead), lamina propria (open arrowhead), and muscular layer (arrows) of the adult Crim1+/KST264 bladder. Z: Expression in the vascular smooth muscle of an artery in the adult Crim1+/KST264 bladder. AA: Expression in Sertoli cells (arrow), smooth muscle cells (arrowhead), and Leydig cells (open arrowhead) of an adult Crim1+/KST264 mouse testis. BB: Expression of Crim1-β-gal in the granulosa cells of an ovarian follicle of a 17.5 dpc Crim1+/KST264 embryo. CC: Expression of Crim1-β-gal and in the atresing follicle of an adult Crim1+/KST264 ovary. Expression is also seen in the ovarian surface epithelium (BB, CC). da, dorsal aorta; ep, epicardium; lp, lamina propria; mu, muscularis; my, myocardium; nt, neural tube; pc, pericardium; te, transitional epithelium; tr, trabeculated myocardium.

Download figure to PowerPoint

Whole-mount X-Gal staining revealed an onset of Crim1-β-Geo expression around 8.5 dpc. Staining was evident in the yolk sac and around the ectoplacental cone (Fig. 2F). High levels of staining were concentrated in regions of the yolk sac adjacent to the posterior part of the embryo proper and in the nascent blood islands. By 10.5 dpc, Crim1-β-Geo expression was detected in the embryo proper, with X-Gal staining in the eye, forebrain, and branchial arches (Fig. 2G), consistent with previously reported sites of Crim1 expression (Georgas et al.,2000; Kolle et al.,2000; Lovicu et al.,2000). At 12.5 dpc, strong staining was observed in the eye, ear, spinal cord, limbs, and peridermal cells (Fig. 2H). At 15.5 dpc, Crim1-β-Geo expression remained strong in the eye, ear, limbs, and the chordal regions of the spinal cord, and was also detected in the nascent vibrissae follicles (Fig. 2I). Section data at 10.5 dpc revealed Crim1-β-Geo expression in the dorsal aorta, motor neurons of the spinal cord (Fig. 2J), the pericardium, and developing epicardium (Fig. 2K). Crim1-β-Geo expression was present in the developing limbs (Fig. 2L,M) and cartilage of the developing digits (Fig. 2N). Crim1-β-Geo expression remained prominent during heart development, with X-Gal staining in the epicardium, the coronary arteries, and some endocardial cells at 17.5 dpc (Fig. 2O–R).

In contrast to other organs examined, the X-Gal staining observed in the kidney did not completely replicate previously described Crim1 expression. Crim1 mRNA expression was reported in renal vesicles, comma- and S-shaped bodies of the developing nephrons of the kidney (Georgas et al.,2000). Whereas Crim1 mRNA expression was reported to be detectable from 12.5 dpc (Georgas et al.,2000), X-Gal staining was not observed in the kidneys until approximately 14 dpc. At 15.5 dpc, X-Gal staining was evident in the urothelium and weakly in the surrounding muscle layers of the ureter and in smooth muscle of the renal arteries (Fig. 2S). X-Gal staining was also present in the glomerular epithelial cells (GECs) of the developing nephrons with some weak staining in the podocytes (Fig. 2T). Section data also revealed Crim1-β-Geo expression in cells of the adrenal gland and the suprarenal ganglion of the adrenal gland (Fig. 2U), giant trophoblast cells of the placenta (Fig. 2V), and epithelial cells of the developing lung (Fig. 2W). At 17.5 dpc, expression was detected in the basal layer of the epidermis and epithelium of a developing hair follicle in the skin (Fig. 2X).

In the Crim1+/KST264 adult, Crim1-β-Geo expression was detected in the basal layer of the transitional epithelium, the lamina propria, and strongly in the detrusor muscle of the bladder (Fig. 2Y). Strong staining was also observed in the smooth muscle of the arteries of the bladder (Fig. 2Z). In the adult testis, Crim1-β-Geo expression was observed in Sertoli cells, smooth muscle cells, and the interstitial Leydig cells (Fig. 2 AA). During early gonad development, we previously reported sexual dimorphism of Crim1, with expression restricted to the Sertoli cells of the developing testis (Georgas et al.,2000). However, analysis of the ovaries of postnatal Crim1+/KST264 female mice also revealed strong X-Gal staining in the ovarian surface epithelium, maturing follicles, and corpus lutea (Fig. 2 BB,CC).

Crim1KST264/KST264 Mice Are Perinatal Lethal and the Mutation Results in Cerebral Hematoma, Eye Defects, Syndactyly, and Skin Blebbing

The effect of the Crim1KST264/KST264 mutation on embryo survival was gauged throughout development. There was no significant deviation from expected Mendelian ratios at any stages up until the end of gestation (approximately 19 days in the C57Bl/6 inbred strain). However, of all litters allowed to go to term from heterozygote intercrosses (>50), no homozygous neonates were viable, although on some occasions stillborn, homozygous mice were retrieved. Hence, mice homozygous for the Crim1KST264/KST264 mutation appear to die perinatally. Crim1KST264/KST264 homozygous embryos displayed anomalies in several organ systems. At 15.5 dpc and 16.5 dpc, embryos often displayed prominent hematoma formation above the nose (Table 1; Fig. 3A). By 16.5 dpc, there was an obvious abnormality of the eye in Crim1KST264/KST264 mice (Fig. 3B). Mid-sagittal sections through the center of the optic cups revealed a reduction in the size of the lens in homozygous animals (Fig. 3B). In addition, there was an accumulation of cells of endothelial appearance in the posterior eye chamber. This aberrant cell mass that accumulates in eyes of homozygous embryos may be precursor cells to the fine capillaries that normally comprise the tunica vasculosa lentis of the hyaloid vasculature, which regresses later in development. In homozygous embryos, the smaller lens appears to promote the encroachment of the optic cup anteriorly, resulting in a more restricted aperture of the pigmented anterior optic cup, which is evident when viewed from the exterior (Fig. 3B). At 12.5 dpc, all Crim1KST264/KST264 embryos displayed blebbing of the skin on the midline of the head and the dorsum (Table 1; Fig. 3C). The blebbing appeared to resolve in Crim1KST264/KST264 embryos, as it was only rarely seen at 15.5 dpc and never observed after that stage (Table 1). By 13.5 dpc, homozygous embryos were able to be reliably phenotyped based upon the presence of syndactyly and the anomalies of the eyes (Table 1). Homozygous embryos genotyped after 12.5 dpc showed closer spacing of the hind limb digits 3 and 4 (Fig. 3D). By 15.5 dpc, all homozygous embryos displayed syndactyly of these digits (Fig. 3D). Other digits in the forelimb and hind limb were affected to a variable degree (data not shown), but this did not involve digits 1 and 5, and the hindlimbs were always affected. To examine limb defects, we crossed the Crim1+/KST264 mice with outbred CD1 mice for one generation and observed a small number of viable homozygotes at postnatal day 0 (P0). Together with soft tissue syndactyly, the skeleton preparations suggested hard tissue syndactyly of the most proximal phalanges (Fig. 3D). Syndactyly of the metacarpals or metatarsals was never observed. By 15.5 dpc, the kidneys of Crim1KST264/KST264 embryos were smaller than those of their wild-type littermates (Fig. 3E–H).

Table 1. Frequency of Phenotypes Among Crim1KST264/KST264 Embryosa
Stage (dpc)Number of Crim1KST264/KST264 embryosPhenotype
Eye and lens sizeLimb syndactylyKidney sizeFacial hematomaSkin blebbing
  • a

    DPC, days post coitum; n.d.; no data.

10.550 (0%)0 (0%)n.d.0 (0%)0 (0%)
11.570 (0%)0 (0%)n.d.0 (0%)2 (29%)
12.584 (50%)4 (50%)n.d.0 (0%)8 (100%)
13.533 (100%)3 (100%)0 (0%)0 (0%)3 (100%)
14.544 (100%)4 (100%)0 (0%)0 (0%)0 (0%)
15.52121 (100%)21 (100%)21 (100%)4 (19%)1 (5%)
16.577 (100%)7 (100%)7 (100%)2 (29%)0 (0%)
17.51212 (100%)12 (100%)12 (100%)1 (8%)0 (0%)
thumbnail image

Figure 3. Homozygosity for a gene-trap insertion into Crim1 results in cerebral edema; eye, rib, kidney, and skull defects; and syndactyly. A: Frontal views of the heads of 14.5 days post coitum (dpc) wild-type and Crim1KST264/KST264 embryos reveal hematoma on the forehead of the homozygote (arrow, right). B: Whole-mount (left panels) and representative hematoxylin and eosin-stained, mid-sagittal sections through the center of the optic cups (right panels) of eyes from 16.5 dpc Crim1+/KST264 (top panels) and Crim1KST264/KST264 (lower panels) show a restriction of the aperture of the anterior eye chamber (bars), a reduction in lens size and an accumulation of cells in the posterior eye chamber (closed arrowhead) in the Crim1KST264/KST264 mice. C: An X-Gal stained 12.5 dpc Crim1KST264/KST264 embryo showing blebbing of the skin (arrows). D: Analysis of limb development at 12.5 dpc, 15.5 dpc, and postnatal (P) 0 in Crim1+/KST264 (top panels) and Crim1KST264/KST264 (lower panels) animals reveals syndactyly in homozygotes. At 12.5 dpc, there is an indication of fusion of digits 3 and 4 in homozygous limbs. At 15.5 dpc, there is clear fusion of digits 3 and 4 in homozygous limbs. Analysis of the perinatal forelimbs and hindlimbs using Alcian blue and Alizarin red staining reveals evidence for soft and hard tissue syndactyly in homozygous animals. E–G: 15.5 dpc embryonic kidneys shown in whole-mount from wild-type (E), Crim1+/KST264 (F), and Crim1KST264/KST264 (G) littermates. H: The relative length of fresh, unfixed 15.5 dpc embryonic kidneys collected from wild-type, Crim1+/KST264, and Crim1KST264/KST264 embryos are graphically represented. Error bars represent the standard deviation of the mean (SDM). Kidneys from homozygotes were smaller than wild-type littermates (P < 0.05).

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

The net outcome of secreted growth factor expression results from more than the presence of receptor and the transduction of signal in response to binding. For many growth factor superfamilies, there are interacting modulatory proteins that affect posttranslational modifications, interactions with the extracellular matrix, rate of degradation or secretion from the cell, receptor binding, and receptor function. One such protein is Crim1. We have previously shown that this transmembrane protein, which also contains an insulin-like growth factor binding protein (IGFBP) motif and six vWF-C type cysteine-rich repeats, can modulate the activity of BMPs in a cell-autonomous manner in vitro (Wilkinson et al.,2003). In this study, we have characterized the phenotype of the Crim1KST264 gene-trap mouse strain and detailed new sites of Crim1 expression based on a β-Geo reporter gene. Another noteworthy aspect of the β-Geo reporter, is that it did not detect some previously described sites of Crim1 mRNA expression, namely the comma- and S-shaped bodies of the developing kidney (Georgas et al.,2000). A possible explanation is that, in such sites, the β-Geo reporter gene is expressed but is not active, as has been reported in some instances (Couffinhal et al.,1997). In addition, the β-Geo reporter gene may not detect some isoforms of Crim1 transcripts due to alternative splicing. Nevertheless, although the reporter gene used in this study may in some instances underestimate the sites of Crim1 expression, it has revealed new sites of Crim1 expression.

The insertion of the gene-trap cassette into intron 1 of Crim1 did not create a null. Our RT-PCR analysis reveals a minor splice variant lacking exon 2. This alternative splicing event results in an in-frame transcript that would be translated into a Crim1 isoform lacking a cysteine-rich region between the putative IGFBP motif and the first CRR. Immunoblotting analysis of Crim1+/KST264 and Crim1KST264/KST264 tissue suggests that the resultant protein is only 25–30 kDa and is, therefore, likely to have spliced out additional exons, although immunoreactivity indicates that the C-terminus is translated. Importantly, it has been shown that the growth factor-binding ability of Crim1 largely resides in the CRR domains. While a deletion mutant lacking the region encoded by exon 2 but retaining the IGFBP and CRR domains was capable of binding BMPs (Wilkinson et al.,2003), we would predict that the protein present in Crim1KST264/KST264 mice is too short to contain a significant complement of CRR domains. As such, it may behave as an effective null. An alternative possibility is that the exon 2-minus isoform acts in a dominant-negative manner. Quantitative RT-PCR suggests that it is only present at a low level in Crim1KST264/KST264, consistent with a hypomorphic mode of action, but this is not the case at the protein level. Importantly, the Crim1+/KST264 mice do not display the phenotypes observed in Crim1KST264/KST264 mice, arguing against a dominant-negative action for the shorter Crim1 isoform. The use of alternative splicing at the Crim1 gene locus may well vary between tissues and cell types. We have also detected the presence of another Crim1 alternative splice form using the primers from exons 12 to 16. However, the precise functions of the various Crim1 isoforms, including the exon 2-minus form, remain to be determined. What is clear from our analyses is that Crim1KST264/KST264 may not reflect what is likely to be seen in a Crim1 null animal. Our understanding of Crim1 biology would, therefore, be advanced using an allelic series of Crim1 mutations in a tissue-specific manner.

We have previously proposed that Crim1 may act as a modulator of the action of members of the TGF-β superfamily by means of the presence of CRR motifs (Georgas et al.,2000; Kolle et al.,2000; Wilkinson et al.,2003). Indeed, Crim1 can bind to BMP-2, -4, and -7 within the context of the cell, but not in solution (Wilkinson et al.,2003). While complicated by not being a null mutation, the complex phenotype observed in the Crim1KST264/KST264 mice is not easily explained by disruption to a single known BMP or TGF-β, although some aspects are similar to those observed in disruption to members of this superfamily. The reduced lens and microphthalmia observed in Crim1KST264/KST264 mice are perhaps most similar to those in BMP7 null mice (Dudley et al.,1995; Luo et al.,1995; Karsenty et al.,1996; Jena et al.,1997) and a conditional knockout of BMPR-1A (Beebe et al.,2004). Similarly, BMP7 null mice display dysplastic kidneys (Dudley et al.,1995; Luo et al.,1995; Karsenty et al.,1996; Jena et al.,1997) as do Crim1KST264/KST264, albeit less severely. Yet the digit syndactyly in Crim1KST264/KST264 mice could be regarded as the opposite of the polydactyly observed in BMP7 null mice (Dudley et al.,1995; Luo et al.,1995; Jena et al.,1997). Despite the difference in phenotypes between Crim1 and BMP7 mutants, Crim1 could still be a physiological partner for this ligand in some tissues and the phenotypic differences may reflect a mode of regulation that is not simply antagonism of TGF-β superfamily members. Alternatively, Crim1 may play a role in the tethering and delivery of other growth factor families.

Aspects of the Crim1KST264/KST264 phenotype are reminiscent of those observed for molecules directly involved in adhesion and migration events during development. These include the skin blistering in α3 (Kreidberg et al.,1996; DiPersio et al.,1997), α6 (Georges-Labouesse et al.,1996), β1 (Raghavan et al.,2000), and β4 integrins (Gil et al.,1994; Vidal et al.,1995). Mutation of both α3 and α6 integrins in mice resulted in lens defects and syndactyly of the digits similar to that in Crim1KST264/KST264 mice (De Arcangelis et al.,1999). Furthermore, in addition to other defects, targeted disruption of the laminin α5 chain also results in syndactyly similar to that in Crim1KST264/KST264 mice (Miner et al.,1998). Other genes that, when mutated in the mouse, also result in defects that include syndactyly and eye and kidney malformations include the “blebs” genes in mice (reviewed by Smyth and Scambler,2005). These encode the proteins, Fras1 (McGregor et al.,2003; Vrontou et al.,2003), Frem1 (Smyth et al.,2004), Frem2 (Jadeja et al.,2005), and Grip1 (Takamiya et al.,2004). Fras1, Frem1, and Frem2 are large, multidomain proteins and possess discrete motifs that are commonly implicated in interactions with extracellular matrix components as well as TGF-β superfamily biology (Smyth and Scambler,2005). It is noteworthy that, like Crim1, Fras1 contains six vWF-C type cysteine-rich repeats (McGregor et al.,2003; Vrontou et al.,2003).

A loss-of-function study on Crim1 has been conducted in zebrafish using antisense morpholinos (Kinna et al.,2006), and, as with the Crim1KST264/KST264 gene-trap mouse, a pleiotrophic phenotype was observed. The Crim1 morphant fish exhibited abnormal somite patterning, smaller eyes and heads, an expansion of ventral mesoderm-derived intermediate cell mass, as well as a range of vascular defects. The presence of eye and vascular defects in both species suggests a concordance of function. The abnormal somite patterning and expansion of the intermediate cell mass in the fish is evidence of Crim1 involvement in early patterning and hemangiogenesis in that species. No evidence is provided by the Crim1KST264/KST264 mutation for a role in early patterning in the mouse embryo. Such a difference may reflect the evolution of various CRR-containing proteins and their physiological ligands and the resultant combinatorial bindings or the fact that the Crim1KST264/KST264 mouse is not a true null.

In summary, the analysis of the Crim1KST264 gene-trap line has revealed some unexpected roles for Crim1 in development, as well as highlighting the complications that can arise with large, potentially alternately spliced, proteins. We conclude that normal Crim1 isoform ratios are critical for the proper limb, eye, and kidney development. The underlying biology of this molecule is likely to be quite complex, possibly involving interactions with members of the TGF-β superfamily of growth factors and interactions with the extracellular milieu. The Crim1KST264 mouse line has provided a useful resource in beginning to elucidate the in vivo role of Crim1.

EXPERIMENTAL PROCEDURES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

Maintenance of the Gene-Trap Mouse Line

Use of animals in this study was done in accordance with the AEC, University of Queensland (CMCB/535/04/NHMRC). Live Crim1+/KST264 mice were obtained from Marc Tessier-Levigne (University of California, San Francisco, CA). This line was produced in 129/Ola embryonic stem cells and bred on a C57Bl/6 genetic background (Leighton et al.,2001). Mice were housed and bred in the Physiology and Pharmacology animal facility and the Institute for Molecular Bioscience animal facility, University of Queensland. Heterozygous males were crossed with wild-type C57Bl/6 females, and the offspring were genotyped by PCR with neomycinres-specific primers (forward, 5′-GCAGCAGTTTTTCCAGTTCC-3′; reverse, 5′-GGTTTTCCGCCAGACGCCAC-3′) to detect the transgene. Heterozygous F1-generation male and female mice were interbred and plug checked, and embryos were analyzed at the desired developmental stage. The genotyping of embryonic samples was done by PCR for the presence of the transgene in genomic DNA (neomycinres-specific primers) and by RT-PCR for detection of the mutant transcript (neomycinres-specific primers) and for detection of the wild-type transcript (exon1- and exon3-specific primer pair; see below).

RT-PCR Analyses

Tissue destined for RNA extraction was dissected in ice-cold phosphate buffered saline (PBS) and homogenized in TRIzol Reagent (Invitrogen, San Diego, CA) and extracted according to the manufacturer's specifications. One microgram of RNA was used in a reverse transcriptase reaction using standard procedures described elsewhere (Pennisi et al.,2000). PCR was performed on the resultant cDNA using the following sets of primers: Crim1 exon 1–β-Geo PCR; forward, 5′-TCTCGCTGCTGGGGCTGCTG-3′; reverse, 5′-GGGGGATGTGCTGCAAGGCG-3′; Crim1 exon 1–exon 3 PCR; forward, 5′-TCTCGCTGCTGGGGCTGCTG-3′; reverse, 5′-CACGCAGCGGCTGGGTAAAG-3′; Crim1 exon 2 PCR; forward, 5′-CCGGAATTCAGATGAGGACTGGGATGATG-3′; reverse, 5′-CCGGAATTCCTTCTTCGATCCTCTTTAATG-3′; Crim1 exon 12–exon 16; forward, 5′-GCTCAGCACCCCTTCTATTTG-3′; reverse, 5′-GTGATGAGTCTTCGCCTGGATG-3′; GAPDH PCR; forward, 5′-TCGGTGTGAACGGATTTG-3′; reverse, 5′-ATTCTCGGCCTTGACTGT-3′. To quantify the levels of Crim1 transcript, quantitative real-time PCR was conducted on whole kidney cDNA from Crim1+/+, Crim1+/KST264 and Crim1KST264/KST264 15.5 dpc embryos using standard techniques with the following primer pairs; exon 1 forward, 5′-CCCTGTGACGAGTCCAAGTG-3′; exon 1 reverse, 5′-ATGTAGCAGCAGCCGCAGAC-3′; exon 2 forward, 5′-AATGGGAAATGCGAATGTGG-3′; exon 2 reverse, 5′-GGAAACTCAAAGGGATTGTTGC-3′; exon 3 forward, 5′-GAAGTGCGGTTCTCTCCACG-3′; exon 3 reverse, 5′-AGCGGCTGGGTAAAGGACAG-3′; exons 1–3 forward, 5′-CCCTGTGACGAGTCCAAGTG-3′; exons 1–3 reverse, 5′-AGCGGCTGGGTAAAGGACAG-3′; exon 11 forward, 5′-ACGGAAAGGAAATGTGCG-3′; exon 11 reverse, 5′-AACGAATGGTGGGGTTGCC-3′; exon 17 forward, 5′-CACCTCTTGACCTGACCACTG-3′; exon 17 reverse, 5′-GTAGCCCCTGGAAATGCCTG-3′; TFIID forward, 5′-ACGGACAACTGCGTTGATTTT-3′; TFIID reverse, 5′-ACTTAGCTGGGAAGCCCAAC-3′. Expression levels were normalized to TFIID levels and expressed as a proportion of wild-type levels.

Immunoblotting

Kidneys from wild-type, Crim1+/KST264, and Crim1KST264/KST264 embryos were lysed in ice-cold RIPA buffer with protease inhibitors (Complete without EDTA, Roche), precleared by centrifugation and membrane fraction isolated. Lysate was run on a sodium dodecyl sulfate-polyacrylamide gel electrophoresis under reducing conditions. Proteins were transferred to nitrocellulose membranes that were subsequently blocked in 5% nonfat milk protein. Membranes were then incubated overnight at 4°C with a rabbit polyclonal antibody to the C-terminus (cytoplasmic portion) of Crim1 described previously (Wilkinson et al.,2003) and subsequently probed with a horseradish peroxidase-conjugated secondary antibody (Promega). Immunoreactive proteins were then detected by an enhanced chemiluminescence system (Pierce).

Sample Preparation and X-Gal Staining

Samples were dissected in PBS, fixed in 4% paraformaldehyde (PFA) in PBS at 4°C for 4–6 hr and washed in PBS at room temperature. For X-Gal staining, tissues were incubated in X-Gal staining buffer (5 mM potassium ferricyanide, 5 mM potassium ferrocyanide, 2 mM magnesium chloride, 0.01% sodium deoxycholate, 0.02% NP-40, 1 mg/ml X-Gal) at 37°C for 4–16 hr. Samples were then washed extensively in PBS before photography and further processing. Samples for histochemical analyses were dehydrated in a graded ethanol series, cleared, and infused with paraffin. Serial, 4-μm sections were cut and mounted on SuperFrost microscope slides (Menzel-Glaser). Samples were then processed for routine hematoxylin and eosin staining or nuclear fast red counterstaining. All X-Gal staining described is specific for mice carrying the Crim1KST264 allele. In no case was staining detected in wild-type tissue. Some adult Crim1+/KST264 kidneys were dissected in PBS, fixed in 4% PFA in PBS at 4°C for 4–6 hr, washed in PBS at room temperature, and embedded in OCT for cutting 20-μm sections on a cryostat. Sections were dried, washed in PBS, and incubated with X-Gal staining buffer as above. Sections were then washed in PBS and mounted with an aqueous mountant.

Skeletal Preparations

Alcian blue staining of cartilage and bone was performed by fixing embryos overnight in 95% ethanol and then staining in 0.05% Alcian blue, 5% acetic acid overnight. Digestion in 2% potassium hydroxide was then performed for 24 hr followed by clearing in 1% potassium hydroxide, 20% glycerol. The skeletons were then stained in Alizarin red S in 1% potassium hydroxide overnight before further clearing in 1% potassium hydroxide and storage in 20% ethanol containing 20% glycerol.

Statistics

Data are presented as mean ± standard deviation of the mean. Statistical significance was determined using an unpaired Student's t-test (for kidney size) or χ2 (for any deviation from expected Mendelian ratios).

Acknowledgements

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

We thank Daying Wen and Marc Tessier-Levigne for scientific advice and technical assistance. We acknowledge the support of staff in the School of Biomedical Sciences Animal Facility, the Transgenic Animal Service of Queensland, and the Queensland Histology Service, Queensland Institute of Medical Research. T.A.S.Q. is supported by the Australian Research Council SRC Centre for Functional and Applied Genomics. G.K. and M.J.P. held Australian Postgraduate Awards. M.L.S. is a Foundation for Advanced Cancer Studies and Merck Fellow of the Life Sciences Research Foundation. M.H.L. is an NHMRC Principal Research Fellow. We remember our colleague, Toshiya Yamada.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES
  • Beebe D, Garcia C, Wang X, Rajagopal R, Feldmeier M, Kim JY, Chytil A, Moses H, Ashery-Padan R, Rauchman M. 2004. Contributions by members of the TGFbeta superfamily to lens development. Int J Dev Biol 48: 845856.
  • Couffinhal T, Kearney M, Sullivan A, Silver M, Tsurumi Y, Isner JM. 1997. Histochemical staining following LacZ gene transfer underestimates transfection efficiency. Hum Gene Ther 8: 929934.
  • De Arcangelis A, Mark M, Kreidberg J, Sorokin L, Georges-Labouesse E. 1999. Synergistic activities of alpha3 and alpha6 integrins are required during apical ectodermal ridge formation and organogenesis in the mouse. Development 126: 39573968.
  • DiPersio CM, Hodivala-Dilke KM, Jaenisch R, Kreidberg JA, Hynes RO. 1997. alpha3beta1 Integrin is required for normal development of the epidermal basement membrane. J Cell Biol 137: 729742.
  • Dudley AT, Lyons KM, Robertson EJ. 1995. A requirement for bone morphogenetic protein-7 during development of the mammalian kidney and eye. Genes Dev 9: 279527807.
  • Garcia Abreu J, Coffinier C, Larrain J, Oelgeschlager M, De Robertis EM. 2002. Chordin-like CR domains and the regulation of evolutionarily conserved extracellular signaling systems. Gene 287: 3947.
  • Georgas K, Bowles J, Yamada T, Koopman P, Little MH. 2000. Characterisation of Crim1 expression in the developing mouse urogenital tract reveals a sexually dimorphic gonadal expression pattern. Dev Dyn 219: 582587.
  • Georges-Labouesse E, Messaddeq N, Yehia G, Cadalbert L, Dierich A, Le Meur M. 1996. Absence of integrin alpha 6 leads to epidermolysis bullosa and neonatal death in mice. Nat Genet 13: 370373.
  • Gil SG, Brown TA, Ryan MC, Carter WG. 1994. Junctional epidermolysis bullosis: defects in expression of epiligrin/nicein/kalinin and integrin beta 4 that inhibit hemidesmosome formation. J Invest Dermatol 103: 31S38S.
  • Glienke J, Sturz A, Menrad A, Thierauch KH. 2002. CRIM1 is involved in endothelial cell capillary formation in vitro and is expressed in blood vessels in vivo. Mech Dev 119: 165175.
  • Jadeja S, Smyth I, Pitera JE, Taylor MS, van Haelst M, Bentley E, McGregor L, Hopkins J, Chalepakis G, Philip N, Perez Aytes A, Watt FM, Darling SM, Jackson I, Woolf AS, Scambler PJ. 2005. Identification of a new gene mutated in Fraser syndrome and mouse myelencephalic blebs. Nat Genet 37: 520525.
  • Jena N, Martin-Seisdedos C, McCue P, Croce CM. 1997. BMP7 null mutation in mice: developmental defects in skeleton, kidney, and eye. Exp Cell Res 230: 2837.
  • Karsenty G, Luo G, Hofmann C, Bradley A. 1996. BMP 7 is required for nephrogenesis, eye development, and skeletal patterning. Ann N Y Acad Sci 785: 98107.
  • Kinna G, Kolle G, Carter A, Key B, Lieschke GJ, Perkins A, Little MH. 2006. Knockdown of zebrafish crim1 results in a bent tail phenotype with defects in somite and vascular development. Mech Dev 123: 277287.
  • Kolle G, Georgas K, Holmes GP, Little MH, Yamada T. 2000. CRIM1, a novel gene encoding a cysteine-rich repeat protein, is developmentally regulated and implicated in vertebrate CNS development and organogenesis. Mech Dev 90: 181193.
  • Kreidberg JA, Donovan MJ, Goldstein SL, Rennke H, Shepherd K, Jones RC, Jaenisch R. 1996. Alpha 3 beta 1 integrin has a crucial role in kidney and lung organogenesis. Development 122: 35373547.
  • Larrain J, Bachiller D, Lu B, Agius E, Piccolo S, De Robertis EM. 2000. BMP-binding modules in chordin: a model for signalling regulation in the extracellular space. Development 127: 821830.
  • Leighton PA, Mitchell KJ, Goodrich LV, Lu X, Pinson K, Scherz P, Skarnes WC, Tessier-Lavigne M. 2001. Defining brain wiring patterns and mechanisms through gene trapping in mice. Nature 410: 174179.
  • Lovicu FJ, Kolle G, Yamada T, Little MH, McAvoy JW. 2000. Expression of Crim1 during murine ocular development. Mech Dev 94: 261265.
  • Luo G, Hofmann C, Bronckers AL, Sohocki M, Bradley A, Karsenty G. 1995. BMP-7 is an inducer of nephrogenesis, and is also required for eye development and skeletal patterning. Genes Dev 9: 28082820.
  • McGregor L, Makela V, Darling SM, Vrontou S, Chalepakis G, Roberts C, Smart N, Rutland P, Prescott N, Hopkins J, Bentley E, Shaw A, Roberts E, Mueller R, Jadeja S, Philip N, Nelson J, Francannet C, Perez-Aytes A, Megarbane A, Kerr B, Wainwright B, Woolf AS, Winter RM, Scambler PJ. 2003. Fraser syndrome and mouse blebbed phenotype caused by mutations in FRAS1/Fras1 encoding a putative extracellular matrix protein. Nat Genet 34: 203208.
  • Miner JH, Cunningham J, Sanes JR. 1998. Roles for laminin in embryogenesis: exencephaly, syndactyly, and placentopathy in mice lacking the laminin alpha5 chain. J Cell Biol 143: 17131723.
  • Pennisi D, Bowles J, Nagy A, Muscat G, Koopman P. 2000. Mice null for Sox18 are viable and display a mild coat defect. Mol Cell Biol 24: 93319336.
  • Raghavan S, Bauer C, Mundschau G, Li Q, Fuchs E. 2000. Conditional ablation of beta1 integrin in skin. Severe defects in epidermal proliferation, basement membrane formation, and hair follicle invagination. J Cell Biol 150: 11491160.
  • Skarnes WC, Moss JE, Hurtley SM, Beddington RS. 1995. Capturing genes encoding membrane and secreted proteins important for mouse development. Proc Natl Acad Sci U S A 92: 65926656.
  • Smyth I, Scambler P. 2005. The genetics of Fraser syndrome and the blebs mouse mutants. Hum Mol Genet 14(Spec No. 2): R269R274.
  • Smyth I, Du X, Taylor MS, Justice MJ, Beutler B, Jackson IJ. 2004. The extracellular matrix gene Frem1 is essential for the normal adhesion of the embryonic epidermis. Proc Natl Acad Sci U S A 101: 1356013565.
  • Takamiya K, Kostourou V, Adams S, Jadeja S, Chalepakis G, Scambler PJ, Huganir RL, Adams RH. 2004. A direct functional link between the multi-PDZ domain protein GRIP1 and the Fraser syndrome protein Fras1. Nat Genet 36: 172177.
  • Vidal F, Aberdam D, Miquel C, Christiano AM, Pulkkinen L, Uitto J, Ortonne JP, Meneguzzi G. 1995. Integrin beta 4 mutations associated with junctional epidermolysis bullosa with pyloric atresia. Nat Genet 10: 229234.
  • Vrontou S, Petrou P, Meyer BI, Galanopoulos VK, Imai K, Yanagi M, Chowdhury K, Scambler PJ, Chalepakis G. 2003. Fras1 deficiency results in cryptophthalmos, renal agenesis and blebbed phenotype in mice. Nat Genet 34: 209214.
  • Wilkinson L, Kolle G, Wen D, Piper M, Scott J, Little M. 2003. CRIM1 regulates the rate of processing and delivery of bone morphogenetic proteins to the cell surface. J Biol Chem 278: 3418134188.