Abstract

Girls with MLS syndrome have microphthalmia with linear skin defects of face and neck, sclerocornea, corpus callosum agenesis and other brain anomalies. This X-linked dominant, male-lethal condition is associated with heterozygous deletions of a critical region in Xp22.31, from the 5′ untranslated region of MID1 at the telomeric boundary to the ARHGAP6 gene at the centromeric boundary. HCCS, encoding human holocytochrome c-type synthetase, is the only gene located entirely inside the critical region. Because single gene analysis is not feasible in MLS patients (all have deletions), we generated a deletion of the equivalent region in the mouse to study the molecular basis of this syndrome. This deletion inactivates mouse Hccs, whose homologs in lower organisms (cytochrome c or c1 heme lyases) are essential for function of cytochrome c or c1 in the mitochondrial respiratory chain. Ubiquitous deletions generated in vivo lead to lethality of hemizygous, homozygous and heterozygous embryos early in development. This lethality is rescued by expression of the human HCCS gene from a transgenic BAC, resulting in viable homozygous, heterozygous and hemizygous deleted mice with no apparent phenotype. In the presence of the HCCS transgene, the deletion is easily transmitted to subsequent generations. We did obtain a single heterozygous deleted female that does not express human HCCS, which is analogous to the low prevalence of the heterozygous MLS deletion in humans. Through the study of these genetically engineered mice we demonstrate that loss of HCCS causes the male lethality of MLS syndrome.

INTRODUCTION

Microphthalmia with linear skin defects (MLS), or MIDAS (microphthalmia, dermal aplasia and sclerocornea) syndrome is a rare X-linked dominant neurodevelopmental disorder. It is characterized by the presence at birth of linear, erythematous skin defects on the face and neck that heal to form hyperpigmented areas in nearly 100% of patients; eye abnormalities (microphthalmia, corneal opacities, sclerocornea and orbital cysts) in over 75% of cases; brain abnormalities (agenesis of the corpus callosum, ventriculomegaly, microcephaly) in about 40%; mild to severe mental retardation in 24%; and cardiac abnormalities (arrhythmias, septum defects, cardiomyopathy) in 18% (14). Most patients are female, but rare males with a 47,XXY or a 46,XX karyotype with Xp;Y translocations have been described (49). Deletions of the short arm of one X chromosome involving Xp22.3 to Xpter have been identified in all girls with MLS syndrome, except for one atypical patient who has only the skin defects and in whom a chromosomal defect has not been found (3).

Cytogenetic analysis combined with breakpoint mapping on somatic cell hybrids from patients with various phenotypes and terminal deletions involving Xp assigned the critical deletion interval for MLS to Xp22.31, centromeric to the Kallmann syndrome gene (KAL) and telomeric to the gene for amelogenin (AMELX) (1012) between the telomeric breakpoint of patient BA333, the largest deletion seen in a patient who has no features of MLS and the centromeric breakpoint of patient BA325, the smallest deletion seen in a patient with typical MLS syndrome (12).

Three genes with functional characteristics that make them good candidates to contribute to the phenotypic features of MLS syndrome are associated with the human critical region (1315). The centromeric boundary-breakpoint (BA325) is within ARHGAP6, which encodes a rho GTPase activating protein (GAP). This gene is transcribed in a centromere to telomere direction and spans 500 kb. A newly described patient with typical features of MLS has a breakpoint centromeric to exon 6 of ARHGAP6, which is the first exon that encodes the GAP domain of this protein (16). Rho GTPases belong to the superfamily of ras-related GTPases that participate in signal transduction to regulate the actin cytoskeleton and are implicated in diverse processes such as neurogenesis, cell motility, apoptosis and neural crest cell migration (1719). The ARHGAP6 protein has specific GAP activity for rhoA (20).

The second gene in the MLS region is HCCS, which is located 15 kb telomeric to ARHGAP6 and encodes the human holocytochrome c-type synthetase. This highly conserved gene spans 11 kb of genomic DNA and is ubiquitously expressed, with highest levels in heart and skeletal muscle (13,21). Homologs of HCCS (cytochrome c heme lyases; CCHL) (2224) localize to the mitochondrial intermembrane space where they trap apocytochrome c when it translocates through the outer mitochondrial membrane (25). CCHL also directly promotes covalent attachment of heme to apocytochrome c, resulting in formation of holocytochrome c (24,2628). This essential function suggested to us that deletion of HCCS could result in the male lethality and possibly other features of MLS syndrome.

The coding region of the third gene, MID1, is strictly outside the MLS critical deletion interval, at its telomeric boundary, but it was identified because 5′ untranslated exons of this gene in humans (15) reside within the critical region, and their deletion may thus contribute to part of the phenotype (15,29). MID1 had already been identified by Quaderi et al., who found it to be mutated in X-linked Opitz G/BBB syndrome (30). Opitz G/BBB syndrome is not male lethal and loss-of-function mutations of MID1 appear to be compatible with male viability (2931). Mid1 is expressed in the first and second branchial arches at E10.5 (30) and the MID1/midin protein participates in a signaling pathway that promotes microtubule stabilization (3234).

The clinical findings of MLS syndrome overlap significantly with two other X-linked dominant conditions, Aicardi syndrome and Goltz syndrome or focal dermal hypoplasia (FDH), that are exclusively (Aicardi syndrome; 100%) or predominantly (FDH; 95%) observed in females (3,35) and in which genomic rearrangements have never been observed. The classic triad of Aicardi syndrome consists of chorioretinal lacunae, agenesis of the corpus callosum and infantile spasms (36), but skin defects are not present. Two MLS patients with an Xp22.3 deletion had pigmentary retinal lesions that were similar to the punched-out lacunae seen in Aicardi syndrome (35,37,38). Goltz patients commonly have microphthalmia, linear skin lesions, skeletal anomalies of the limbs and less commonly have agenesis of the corpus callosum (39,40). Based on the overlapping phenotype, both Aicardi syndrome and FDH were assigned to Xp22.3 (35,38) and we and others have hypothesized that they are allelic to MLS and caused by mutations in the gene or genes that are deleted or disrupted in MLS syndrome (4,1315,41). We performed mutation screening of the coding regions of these three genes (ARHGAP6, HCCS and MID1) on genomic DNA from Aicardi and Goltz patients, but no mutations have been found to date (1315). This probably indicates that these conditions are not allelic with MLS syndrome, although alternatively the pathologic mutations may not be in the coding sequence of these genes.

We therefore chose an alternative approach to understand the contribution to the MLS phenotype of the genes in the critical region by engineering mouse models with overlapping deletions similar to the human MLS deletion. In this report we describe the generation and initial characterization of these mutant mice. We found that inactivation of the Hccs gene results in early embryonic lethality of the mutant mice. This phenotype can be rescued by expression of a human HCCS transgene. Such ‘rescued’ male and female animals have no obvious phenotype, and we conclude that loss of HCCS causes the male lethality of human MLS syndrome.

RESULTS

A null mutation of Hccs is incompatible with survival of mouse embryonic stem cells

We replaced exons 2, 3 and 4 of the mouse Hccs gene with a targeting cassette containing a neomycin resistance gene and exons 1–2 of an HPRT minigene. Since ES cells used in gene targeting are male and Hccs is on the X chromosome, we expected that Southern analysis of correctly targeted clones with the 5′ and 3′ external probes would reveal a single 10.5 kb fragment after digestion with ScaI and a 6.9 kb fragment after digestion with SpeI (Fig. 1A). After electroporation of the ES cells with the targeting construct, we observed poor growth of the colonies under negative selection. Southern analysis of 100 surviving ES cell clones showed only four that appeared targeted, but they also retained the wild-type band (Fig. 1B). To understand the origin of this wild-type band, we performed FISH analysis with a probe within the MLS region on metaphase spreads derived from such ES cells and found that they were all polyploid and contained wild-type X chromosomes (data not shown). Occasional polyploid cells can exist in ES cell cultures and we hypothesized that the absence of Hccs function in correctly targeted cells led to preferential survival of these colonies. The Hccs genomic locus is only 11 kb and contains small introns; attempts to generate a conditional allele of this gene yielded similar results, suggesting that the insertion of the targeting cassette caused enough disruption of the gene's expression in these cultured ES cells (data not shown).

Detailed characterization of a previously generated null mutation of the functional domain (rhoGAP domain) of the Arhgap6 gene did not reveal a phenotype in hemizygous, homozygous and heterozygous animals (20). However, since these mutant mice still expressed a truncated mRNA, it could not be excluded that the exons encoding the remaining 470 amino acid protein still contribute to the MLS phenotype.

To definitively understand the relative contribution of ARHGAP6, HCCS and MID1 to the MLS phenotype, we designed a deletion strategy.

Inactivation of 61b3r, the mouse homolog of a 5′ untranslated exon of MID1, creates an anchor point for the murine MLS deletion

An STS marker determines the telomeric end of the human MLS critical region (15). This sequence, named 61B3R, was found to be a 5′ untranslated exon of the MID1 gene that is only expressed in human cerebellum (15). Since we did not detect expression of the conserved mouse exon in a wide range of examined adult and embryonic tissues (not shown), we reasoned that its inactivation would not result in a significant phenotype. We therefore chose it as the telomeric anchor point to generate a deletion between the rhoGAP domain of Arhgap6 and 61b3r, similar to the human MLS critical region. A 2.0 kb genomic fragment containing the 61b3r sequence was replaced with a targeting cassette containing exons 3–9 of an HPRT minigene, a puromycin resistance gene and a loxP site in the correct orientation for later generation of a deletion when double targeted with the Arhgap6 construct (Fig. 2). The length of the targeting vector's flanking arms for homologous recombination was 4.5 kb. We first tested the 61b3r replacement allele as a single targeting event. As expected, homozygous and hemizygous mice were fertile and had no obvious phenotype. Targeted alleles were transmitted in the expected Mendelian ratios and Mid1 mRNA expression was normal, as was histological screening of adult tissues and baseline neurobehavioral analysis (data not shown). We concluded that deletion of this sequence has no overt functional consequences in the mouse.

Generation of mice targeted to create an MLS deletion

Prior to generating deletions of the orthologous region in the mouse, we also demonstrated that the order and orientation of the three genes (MID1, HCCS and ARHGAP6) associated with the human MLS critical region is conserved in mice (15) (Fig. 2A). The 61b3r targeting construct was then electroporated into 129/SvEv ES cell clone 72D3, which was previously targeted with the Arhgap6 construct and had given germline transmission of this mutation (20). This double-targeted allele is hereafter referred to as MLS2loxP. Targeting at 61b3r was identified by an 8.4 kb and a 6 kb fragment; targeting of Arhgap6 was identified by a 10.5 kb ApaI and a 11.5 kb BamHI fragment on Southern analysis as previously described (20) (Fig. 2B and D). Three independent electroporations resulted in 69 of 245 (28%) correctly double-targeted ES cell clones. Because the loci are on the X chromosome and ES cells are derived from a male embryo, all double-targeted ES cell clones have the similarly oriented loxP sites in cis. This allows for the generation of a ∼400 kb deletion that can be detected by Southern analysis with NdeI. The deletion replaces the targeted fragments of 6.5 and 9.2 kb at the Arhgap6 and 61b3r loci, respectively, by a single novel 7.7 kb fragment (Fig. 2C). This deleted allele is hereafter referred to as MLSΔ. We attempted to induce this deletion at the ES cell stage by electroporation of a Cre recombinase-expressing plasmid. This resulted in poor growth of the cultures and yielded no clones that contained only a deleted allele. These results were consistent with our observations during generation of the Hccs null allele and again supported that loss of Hccs function, in this instance by deletion of the entire genomic locus, leads to lethality of the male ES cells. Two different ES-cell clones were then injected into C57BL6/J blastocysts to generate two lines of mice carrying the MLS2loxP double-targeted allele, as a source to subsequently generate the deletions in vivo. As expected from previous observations on single targeted 61b3r and Arhgap6 mice, homozygous and hemizygous MLS2loxP mice appeared healthy and fertile, and both lines were used for further experiments.

Mice with the MLSΔ deletion only survive when the deletion is mosaic

To circumvent the ES cell-lethality, we chose to generate deletions by mating MLS2loxP double-targeted mice to a line of mice that express a Cre-recombinase transgene driven by an EIIa viral promoter. This promoter drives expression in the oocyte and earliest embryonic cell divisions, and has been successfully used to generate ubiquitous deletions in vivo (42,43). As predicted, when we intercrossed mice carrying the Cre recombinase transgene to MLS2loxP mice, we never observed any hemizygous (MLSΔ/Y) or homozygous (MLSΔ/MLSΔ) deleted animals. However, we also did not observe heterozygous (MLSΔ/X) deleted females, but reproducibly found a variable number of offspring that still retained their wild-type and/or targeted alleles in addition to a deletion-specific band on Southern analysis of tail genomic DNA with NdeI. This band was consistently fainter than the wild-type and/or targeted alleles in the same lanes (Fig. 2D, middle panel), suggesting that the deletions were present in a mosaic pattern, but were never ubiquitously generated in all cells. Table 1A also shows that more mosaic alleles were obtained when the ratio of double-floxed alleles compared with wild-type X chromosomes in the parents is higher (compare row 1 with rows 2 and 3). To confirm this observation, we mated hemizygous MLS2loxP/Y double-targeted males to wild-type (X/X) females. Three separate matings resulted in a total of seven out of 25 (28%) female offspring with both a double-targeted and a deleted allele (Table 1A, row 3). Since the double-targeted X chromosome is contributed by the male, such offspring must be mosaic for the deletion. We then analysed various tissues of these mice for levels of mosaicism and found tissue-specific differences (Fig. 3). Both male and female mice with mosaic deletions were apparently healthy and fertile. However, we also observed a trend toward smaller litter sizes (LS: 4.5) and a higher percentage of mosaic deleted offspring when more double-targeted or mosaic deleted alleles were present in the parental genotypes. As seen in Table 1, there are no offspring with deletions when a single mosaic deleted allele is present in the parents (Table 1B; top row); in comparison, 60% of all females or 20% of all offspring are mosaic when both parents carry a mosaic or floxed allele (Table 1B; bottom row). Together, these findings indicate that ubiquitous deletions of the Hccs gene are lethal in the mouse. The fact that we did not observe any heterozygous deleted females in these initial matings may explain why MLS syndrome is a rare disorder in humans. Most heterozygous affected females may not survive to birth unless they have skewed X inactivation in favor of the non-deleted X chromosome. The observed mosaicism in the surviving male deleted mice is conceptually comparable to the skewed X-inactivation patterns seen in female patients with MLS syndrome.

A human HCCS transgene rescues the lethality of the MLSΔ deletion

The preceding data indicated that loss of function of Hccs is responsible for the male lethality of MLS syndrome. However, Hccs is not the only gene that may be affected by the deletion. Exons 6 to 14 of Arhgap6 and the 5′ region of Mid1 are also involved and their disruption could contribute to the MLS phenotype. To address this issue, we generated transgenic mice expressing a human BAC clone (GS602M16; 65716 bp) containing the 11 kb HCCS gene with >40 kb of upstream regulatory sequence and exons 12–14 of ARHGAP6 (Fig. 4A). The transgene is detected by a human-specific 7.5 kb band by Southern analysis of HindIII-digested tail DNA from the founders and their offspring (Fig. 4B). Two founder animals carried the transgene in approximately equal copy numbers, but only one line with approximately five copies of the transgene showed expression by Northern and RT–PCR analysis. Detailed analysis revealed that the BAC was rearranged in the non-expressing line (not shown). The expressing line of mice was then used for all further experiments. Homozygous and heterozygous HCCS-transgenic mice are healthy and fertile and have no obvious phenotype. Gross pathological and microscopic analysis of brain, heart, muscle, kidney, liver and spleen also did not reveal any anomalies.

We then crossed double-targeted MLS2loxP;EIIa-Cre recombinase--positive animals (hereafter referred to as MLS2loxP; Cre+) with mice that carry the HCCS BAC-transgene (hereafter referred to as HCCS+). In contrast to our findings without the transgene (Table 1A), such matings not only resulted in mosaic deleted mice, but also resulted in viable hemizygous (MLSΔ/Y;Cre+;HCCS+), homozygous (MLSΔ/ MLSΔ;Cre+; HCCS+) and heterozygous (MLSΔ/X;Cre+;HCCS+) offspring with deletions in all examined tissues (Table 2A, Fig. 5). MLSΔ/Y;Cre+;HCCS+ and MLSΔ/MLSΔ;Cre+;HCCS+ mice are indistinguishable from non-deleted littermates and produce offspring with deletions (Table 2B), confirming that germline transmission of the MLSΔ allele is possible on an HCCS-transgenic background. All such offspring with deletions carry the HCCS BAC transgene, except for one heterozygous deleted MLSΔ/X;Cre+ female (discussed below). We examined the expression of human HCCS, mouse Hccs, Mid1 and Arhgap6 by RT–PCR analysis on total RNA extracted from brain, heart, liver and muscle of a MLSΔ/Y;Cre+;HCCS+ male. As expected, the human HCCS transgene was expressed, while the deletion abolished the expression of the endogenous mouse Hccs gene, but did not affect Mid1, and resulted in the expression of a truncated Arhgap6 mRNA (Fig. 5B). We previously showed that a mutation generating a truncated Arhgap6 protein is compatible with normal survival and has no obvious phenotype in female homozygous and male hemizygous mice (20). These combined data provide further support for the inactivation of HCCS causing the male lethality of MLS syndrome.

To date, 25 of 26 MLSΔ/X females from various matings between deleted parents carry the HCCS transgene. We would have expected only 17/26 to carry the transgene if rescue of the lethality by HCCS were not needed for survival of mice with the MLSΔ/X genotype (this number was derived from the different parental genotypes in these matings, not shown). The difference between 1/26 observed and 9/26 expected MLSΔ/X;Cre+ animals is statistically significant (P=0.004; Fisher's exact test). The single MLSΔ/X;Cre+ animal appears smaller than her littermates with smaller eyes (Fig. 5C) and at 9 months of age has only produced seven liveborn offspring, none of whom carry the deleted allele. The rarity of the MLSΔ/X genotype in mice is consistent with human MLS syndrome and is further evidence that the generated deletion is a faithful model for this disorder. It is possible that the phenotype is more severe in the mouse than in humans, but an alternative, more likely explanation is that the majority of heterozygous human females affected with this deletion also do not survive until birth. This is consistent with the rarity of MLS syndrome in humans and may be explained by patterns of X chromosome inactivation in surviving individuals that favor the non-deleted X chromosome. Our clinical observation that a patient with an MLS deletion and a mild phenotype, because of favorably skewed X-chromosome inactivation, miscarried an anencephalic female fetus with the same deletion corroborates this interpretation (4).

Variable embryonic phenotype of deletions

The reduced litter sizes and absence of deleted viable offspring observed in matings aimed at generating deletions indicated that the deletions are embryonic lethal. To investigate the timing and cause of this lethality, we set up timed matings. First, we examined embryos at embryonic day E10.5 and E9.5 from matings between MLS2loxP/MLS2loxP;Cre+ females and MLS2loxP/Y;Cre+ males without the human HCCS transgene. We found that 15 out of 21 embryos appeared abnormal, with the severity of the defects ranging from complete resorption to delayed development and abnormalities suggesting gastrulation defects. PCR-based genotyping of DNA extracted from yolk sacs revealed that 11 of 17 embryos that had sufficient tissue to be genotyped contained both targeted alleles and deleted alleles. However, the PCR-based genotyping would not allow us to determine unequivocally whether these embryos were mosaic or ubiquitously deleted in all tissues. We then performed timed matings between (MLSΔ/Y;Cre+;HCCS+) males and (MLSΔMLSΔ;Cre+;HCCS+), (MLSΔ/X;Cre+;HCCS+) or wild-type (X/X) females. We evaluated pregnancies at E8.5, E9.5 and E10.5. The numbers of embryos recovered at each stage were similar to expected values based on average litter size, and 42% appeared abnormal (Fig. 6; Table 3). We also found variability in the severity of the embryonic phenotype of the deletions from these matings (Fig. 6), as well as normal-appearing deleted, HCCS-negative embryos present until E9.5, but not at E10.5 (Table 3). These data are consistent with lethality of the deletions prior to E10.5. The number of deleted embryos that do not express human HCCS is much smaller than those that do express this transgene. Interestingly, the percentage of abnormal embryos in the group with MLSΔ;HCCS+ genotypes (22/47; 47%) was higher than in the group with wild-type genotypes (12/38; 32%). This suggests that other variables may affect the completeness of the rescue by HCCS. All experiments were performed on a mixed genetic background with various contributions of 129/SvEv, C57BL6/J and FVB genotypes. Hence, it is possible that the phenotype is influenced by modifier alleles such as the Xce locus (44,45). This, together with possible subtle variations in expression of the human HCCS transgene compared with endogenous Hccs, could explain the embryonic abnormalities. Nevertheless, we obtained adult mice that were completely rescued from the lethal effects of the deletion by HCCS expression, which allows us to conclude that HCCS is a critical gene for lethality of MLS syndrome.

DISCUSSION

Understanding the pathogenesis of MLS syndrome is uniquely challenging: the condition is rare, X-linked dominant male lethal and, in all patients known to date, caused by deletions of the short arm of the X chromosome involving the region Xp22.31 (4). Three genes are affected by this deletion (1315), but since no point mutations are known, the contribution of each gene to the MLS syndrome pathogenesis can best be determined via the generation of mouse models. In addition to the generation of single-gene loss-of-function mutations, the availability of chromosome engineering techniques using Cre-loxP mediated recombination allowed us to create a model of the complete MLS syndrome deletion in the mouse. We demonstrated that an Hccs null mutation is not compatible with ES cell survival. In vivo generated deletions resulting in complete loss of Hccs lead to embryonic lethality that can be rescued by overexpressing the human HCCS gene from a BAC that contains all 5′ regulatory elements of this gene. These data provided the proof that HCCS results in the male lethality of MLS syndrome.

HCCS is a highly conserved gene (13) and even though detailed studies have not been performed in mammals, information about its homologs in lower organisms such as C. elegans and N. crassa indicate that it has an essential role in the proper incorporation and function of holocytochrome c in the mitochondrial respiratory chain (2228). Holocytochrome c plays an essential role in oxidative phosphorylation by shuttling electrons from complex III to complex IV of the respiratory chain. In addition, release of cytochrome c from mitochondria and its subsequent binding to procaspase 3 in the cytosol leading to caspase 3 activation has recently been shown to play a major role in the initiation of apoptosis through the ‘mitochondrial’ or ‘stress-induced’ pathway (46,47). Apoptosis plays an important role in neuronal migration and patterning of the brain; a null mutation of mouse cytochrome c leads to early lethality of developmentally delayed embryos (47). There are precedents for mutations in genes affecting mitochondrial oxidative phosphorylation as a cause for neurological abnormalities. Deficiency of cytochrome c oxidases and mutations in nuclear genes SURF1 and SCO2 that encode proteins with a role in cytochrome c oxidase assembly result in neurodevelopmental disorders (Leigh syndrome) (48,49). Mutations in another component that is important for mitochondrial respiration, the X-linked E1α subunit of the pyruvate dehydrogenase (PDH) complex, are found in patients with Leigh syndrome (50) and agenesis of the corpus callosum (51).

There are several possible explanations for the variability in the phenotype of the embryos. Variable patterns of X chromosome inactivation may be present in heterozygous MLSΔ/X embryos. Since Hccs is the most important contributor to the lethality, there may be residual mitochondrial oxidative phosphorylation in the early embryo because of the presence of maternally inherited mitochondria. Also, all experiments were performed on a mixed genetic background (129/SvEv, C57BL6/J and FVB/N) and modifier genes may influence the phenotype. Specifically, 129 and C57 mice carry different alleles of the X-controlling element (Xce), a regulatory locus that influences which X chromosome is selected for inactivation. Female offspring of intercrosses between C57 and 129 may carry two Xce alleles, causing skewed X inactivation patterns, whether or not a deleterious mutation is present (44,45). To our knowledge, the Xce allele of FVB/N has not been characterized. Developmental defects such as agenesis of the corpus callosum, a major phenotypic feature of human MLS syndrome, are certainly strain-dependent. Further studies on animals that have been backcrossed to pure genetic backgrounds will address these issues.

Having established that human HCCS rescues the lethality and all obvious features of MLS, it is intriguing that we observed a higher than expected rate of embryonic loss even in the presence of this transgene. This could be due to genetic background effects or subtle differences in levels or patterns of expression of the human transgene compared with the endogenous mouse gene.

Our data indicate that, even though rescue of embryonic lethality is not 100%, the surviving deleted mice that express the human HCCS appear healthy. The single heterozygous deleted (MLSΔ/X) female that does not express the HCCS transgene most likely survived because of patterns of X chromosome inactivation favoring the wild-type allele. Currently, we are still breeding this animal to investigate whether she will produce offspring with the same genotype and further pathological studies to characterize her phenotype in more detail are planned in the future. These findings suggest that HCCS is not only responsible for male lethality of MLS syndrome, but may also play a significant role in the other clinical features. Current investigations underway in our laboratory will further address this by comparing the phenotype of this line of mice to that of two other partially overlapping deletions that we have generated. Both also result in loss of Hccs, but in addition one affects additional 5′ untranslated exons of Mid1, while the other removes the entire Arhgap6 gene. We are also generating tissue-specific deletions by crossing double-targeted mice to mice expressing Cre-recombinase in forebrain, eye and skin. These experiments will be critical to pinpoint the exact role of HCCS in the development of the corpus callosum, eyes and skin, which are the structures most commonly affected in human MLS syndrome.

MATERIALS AND METHODS

Generation of Hccs, 61b3r and Arhgap6 targeted mice

The Hccs null mutation, the 61b3r replacement and the Arhgap6 rhoGAP-domain deletion were all generated by replacing critical regions of their respective genomic loci by replacement vectors PL13 and PG12 (52) (gift from Allan Bradley, Baylor College of Medicine, currently at the Sanger Centre) that after homologous recombination introduce a properly oriented loxP site and exons 1–2 of a human hypoxanthine–guanine phosphoribosyl-transferase (HPRT) minigene and a neomycin-resistance cassette (Hccs and Arhgap6 replacement constructs), or introduce exons 3–9 of the HPRT minigene and a puromycin-resistance cassette (61b3r construct). The Arhgap6 replacement construct assembled in this vector was previously described in detail (20). To generate the Hccs and 61b3r replacement constructs we screened a 129/SvEv genomic library (Allan Bradley) with an Hccs cDNA probe (13) and with a 61B3R genomic probe (15). After linearization, the assembled targeting vectors were electroporated into Hprt-deficient 129S7/SvEvBrd-Hprtb-m2 ES cells; individual colonies were screened by Southern hybridization with 5′ and 3′ external probes represented in Figures 1 and 2; positive clones were injected in C57BL6/J blastocysts and offspring of chimeric males were screened by Southern analysis of tail DNA as described (20).

Generation of HCCS BAC-transgenic mice

Human BAC clone GS602M16 (Genbank number AC003657) was completely sequenced by the BCM Human Genome Sequencing Center and contains the HCCS gene, exons 12–14 of Arhgap6 and 40 kb of the region 5′ to Hccs. BAC DNA was linearized by restriction digestion, purified from a pulsed-field gel, diluted to a concentration of 1 pg/µl and microinjected into a total of 73 FVB oocytes. Southern analysis with HindIII on tail DNA from transgenic mice detects a 7.5 kb human-specific genomic fragment.

Generation of mice with the MLSΔ deletion

The 61b3r targeting construct was electroporated into Arhgap6 rhoGAP-domain targeted ES cell clones that had been successfully transmitted to the germline (72D3) (20). Electroporated clones were plated onto embryonic fibroblast feeder layer cells and grown without selection for 24 h, then double selection for neomycin and puromycin resistance was introduced for 10 days. DNA from individual colonies was screened for correct integration of both targeting constructs using the enzyme combinations shown in Figure 2A and B. To detect deletions we performed Southern analysis with NdeI as shown in Figure 2C.

To generate the deletion in ES cells, doubly targeted ES-cell clones were electroporated with a Cre-expressing plasmid. To generate deletions in vivo, doubly targeted ES cell clones were injected into C57BL6/J blastocysts; chimeras and offspring were screened using the same Southern analysis protocols (Fig. 2A and B). Doubly targeted animals were mated to EIIA-Cre recombinase-expressing transgenic mice (from Heiner Westphal, NICHD, Bethesda, MD, USA) and screened for the presence of the Cre-recombinase transgene by PCR amplification with primers CreA1 (5′-CCGGGCTGCCACGACCAA-3′) and CreA2 (5′-GGCGCGGCAACACCATTTTT-3′) at annealing temperature (Ta) of 55°C, and for the deletion by Southern analysis with NdeI (Fig. 2C).

Pathological analysis

Tissues were fixed in 4% formaldehyde, dehydrated and embedded in paraffin. Sections were cut at 10 µm thickness, stained with hematoxylin and eosin and examined at 200–630× magnification with an Axiophot microscope (Carl Zeiss, Thornwood, NY, USA)

Northern analysis and RT–PCR analysis

To examine HCCS expression from the human BAC transgene, total RNA was prepared from tissues using Trizol (Life Technologies-GIBCOTM, Carlsbad, CA, USA) and northern blots prepared as previously described (20). A 700 bp fragment from the human HCCS cDNA was used as a probe. For various RT–PCR analyses, total RNA was reverse-transcribed using the Superscript II reverse transcriptase kit (Life Technologies-GIBCOTM, Carlsbad, CA, USA) with a standard protocol; 0.5–2 µl of the reverse-transcribed cDNA was amplified with following primer pairs, using standard PCR conditions—for mouse Hccs: 10–14F (5′-TGGGTATTTTAGATTGGGAGTGA-3′) and 10–14R (5′-CTTTATGCATCGGGCATCCT-3′) (Ta 55°C); for human HCCS: 14cbX (5′-CTTATTCTCAGCCGCAGTG-3′), 14cbE (5′-TGGGGACCACGGGCTGGAG-3′) (Ta 61°C); for Arhgap6 3′-to the rhoGAP domain: 13756 (5′-TCAGGTAATTCGGAGGACT-3′) and 19411 (5′-CGGGCTT ATACTAGGGTTTC-3′) (Ta 55°C); for exon 1 of Arhgap6: 19522 (5′-CAGAGCCTGCTGCACAG-3′) and 13749 (5′-GGCCATGCTCTTCTTGAGC-3′) (Ta 56°C); for exons 2–4 of Arhgap6: GAP2F (5′-TGCCCTTATCCCAAGTCATTGC-3′) and GAP4R (5′-GGTTCTGGAGTATTCGGTGATG-3′) (Ta 58°C); for the Mid1 coding region: 1416 (5′-TCCATCAACGCCTCCCAGT-3′) and 1417 (5′-CCTTCTCATCCTCGTGCTC-3′) (Ta 58°C).

Embryo dissection and genotyping

Timed matings were established and the morning on which the plug was found designated as day E0.5. After female mice were anesthetized and sacrificed (at days E8.5, E9.5 and E10.5), the uteri were dissected and immediately transferred to sterile cold PBS; individual conception sites were dissected in PBS under a Zeiss SV11 stereoscope in PBS and embryos were fixed for 15 min to 2 h (depending on size) in 4% freshly prepared paraformaldehyde in PBS and stepwise dehydrated and transferred to 70% ethanol for storage. DNA for PCR analysis was prepared according to an adapted established protocol: yolk sacs or tissues were put in 20–50 µl lysis buffer (20 mm Tris pH 8.4; 50 mm KCl; 0.045% NP40; 0.045% Tween-20; 50 µg/ml proteinase K) overnight at 55°C, heated to 95°C to inactivate the proteinase K and then centrifuged for 10 min. In standard PCR reactions 0.5–2 µl of the supernatant were used with primer for each genotype. To determine the sex of the embryos, primers XY-F (5′-TGAAGCTTTTGGCTTTGAG-3′) and XY-R (5′-CCGCTGCCAAATTCTTTGG-3′) (Ta 55°C) amplified a 300 bp fragment from females and a 300 and 320 bp fragment from males. To detect the presence of the HCCS transgene, we used primers 27581 (5′-TAATCTACTTCTCTCATCCTCTGG-3′) and 27582 (5′-TTCTTTTTGTCAAGTGGGACT-3′) (Ta 53°C) specific for the human BAC containing the HCCS gene. The wild-type 61b3r sequence was amplified with primers 61b3rforw (5′-CTTTTTGTCTTTGTTTTTGATG-3′) and 61b3rrev (5′-ATCACTAACCAACTTCCTTCCT-3′) (Ta 56°C), while the wild-type Arhgap6 locus sequence was amplified with primers rhoGAPforw (5′-AGAATTTGACCGTGGGGTTG-3′) and rhoGAPrev (5′-GAGTGTTGATAAATGCAGTG-3′) (Ta. 55°C); both disappear upon insertion of the targeting constructs. To detect targeted alleles, we used sets of primers that amplify the neomycin and puromycin resistance markers in the targeting vectors as follows: Neo1 (5′-AGAGGCTATTCGGCTATGACTG-3′) with Neo2 (5′-TTCGTCCAGATCATCCTGATC-3′) (Ta 54°C) and Puro1 (5′-GCGCAGCAACAGATGGAAGGC-3′) with Puro2 (5′-CCGCTCGTAGAAGGGGAGGTTG-3′) (Ta 60°C). To detect deletions we used primers HPRT1 (5′-GTTATGACCTTGATTTATTTTGC-3′) and HPRT2 (5′-GCTTATATCCAACACTTCGTG-3′) (Ta 54°C) that amplify between the 5′ and 3′ HPRT exons, which become in proximity of each other when the intervening sequence is deleted. All of these primers were first validated on DNA extracted from mouse tails that was previously genotyped by standard Southern methods. Because this PCR reaction amplifies a 2 kb fragment, it is not 100% successful from the embryonic DNA samples, therefore the presence of deletions in embryos was also inferred where this was possible from comparison of parental genotypes with results on other PCRs, such as sex and presence of wild-type allele.

ACKNOWLEDGEMENTS

The authors wish to thank Dr Paul Overbeek and Gabriele Shuster from the Mental Retardation Research Center transgenic core facility for assistance with the human HCCS transgene injections and Yang Liu for technical assistance. This work was supported by NIH-grant K08-HD01171 (I.V.), the March of Dimes Birth Defects Foundation (FY01-541) (I.V.), the Aicardi syndrome foundation (I.V.), the Baylor College of Medicine Mental Retardation Research Center (NIH grant HD2764064) (I.V., D.A., H.Z.) and the Howard Hughes Medical Institute (H.Z.).

*

To whom correspondence should be addressed at: Department of Obstetrics and Gynecology, Baylor College of Medicine, 6550 Fannin, Suite 901, Houston TX 77030, USA. Tel: +1 7137984914; Fax: +1 7137985060; Email: iveyver@bcm.tmc.edu

The authors wish it to be known that, in their opinion, the two first authors should be regarded as joint First Authors.

Figure 1. The Hccs null mutation is incompatible with ES-cell survival. (A) Targeting construct of the murine Hccs locus: exons 2, 3 and 4 were replaced by neomycin for negative selection, a loxP site and the 5′ half of an HPRT minigene. These features were introduced because we planned to use this allele as an anchor point to generate larger genomic deletions. The untargeted locus yields a 13.5 kb fragment when digested with ScaI and hybridized with the 5′ external probe and a 12.5 kb fragment when digested with SpeI and hybridized with the 3′ external probe. The targeted alleles yield a 10.5 and 6.9 kb fragment, respectively. The upward and downward triangles represent the start and stop codons of the Hccs gene. (B) Southern analysis of genomic DNA from expanded ES cell clones reveals that (male) ES cell clones F4 and G4 contain both a targeted and wild-type X chromosome and thus must be 41,XXY or polyploid (H5 only showed two alleles with the 5′ probe and may represent a rearranged clone).

Figure 1. The Hccs null mutation is incompatible with ES-cell survival. (A) Targeting construct of the murine Hccs locus: exons 2, 3 and 4 were replaced by neomycin for negative selection, a loxP site and the 5′ half of an HPRT minigene. These features were introduced because we planned to use this allele as an anchor point to generate larger genomic deletions. The untargeted locus yields a 13.5 kb fragment when digested with ScaI and hybridized with the 5′ external probe and a 12.5 kb fragment when digested with SpeI and hybridized with the 3′ external probe. The targeted alleles yield a 10.5 and 6.9 kb fragment, respectively. The upward and downward triangles represent the start and stop codons of the Hccs gene. (B) Southern analysis of genomic DNA from expanded ES cell clones reveals that (male) ES cell clones F4 and G4 contain both a targeted and wild-type X chromosome and thus must be 41,XXY or polyploid (H5 only showed two alleles with the 5′ probe and may represent a rearranged clone).

Figure 2. Targeting of the 61b3r and Arhgap6 to generate MLS deletion in the mouse. (A) Wild-type loci. The genes (Mid1, Hccs, Arhgap6) and conserved marker, 61b3r, are boxed. On the left is the 5′ region of the Mid1 genomic locus with the restriction enzyme sites used to detect correct targeting of 61b3r (BamHI) and generation of the deletion (NdeI). On the right is the region surrounding exons 5–9 of the Arhgap6 genomic locus with restriction enzyme sites used to detect correct targeting of exons 6–8 (rhoGAP domain; BamHI and ApaI) and generation of the deletion (NdeI). (B) Replacement of 61b3r with a puromycin resistance gene (P), loxP site (black triangle) and the 3′-half of the HPRT minigene (3-H) replaces a wild-type 17 kb fragment with a 6 and 8.4 kb fragment, detected with the 5′ and 3′ external probes on Southern analysis with BamHI. Replacement of exons 6–8 of Arhgap6 with a neomycin resistance gene (N), a loxP site (black triangle) and the 5′-half of HPRT (5-H) replaces a 14 kb BamHI wild-type fragment with an 11.5 kb fragment, detected with the 5′ probe, and a 22 kb ApaI wild-type fragment with a 10.5 kb fragment, detected with the 3′ probe (probes are indicated by open boxes under the detected fragments). (C) After recombination between the two loxP sites, NdeI restriction digestion detects a novel 7.7 kb fragment with both deletion probes (shown as hatched boxes), that replaces a 9.2 kb fragment at the 61b3r and a 6.5 kb fragment at the Arhgap6 targeted alleles. Not shown are the wild-type allele sizes detected in this Southern analysis, which are <1 kb, and not detected on Southern analysis at the Arhgap6 locus and 5.5 kb at the 61b3r locus. (D) Southern analysis results of targeting described in C. Left panel: targeting at 61b3r; middle panel: detection of the 7.7 kb deleted fragment (note the fainter bands compared to wild-type and targeted alleles); right panel: targeting at Arhgap6.

Figure 2. Targeting of the 61b3r and Arhgap6 to generate MLS deletion in the mouse. (A) Wild-type loci. The genes (Mid1, Hccs, Arhgap6) and conserved marker, 61b3r, are boxed. On the left is the 5′ region of the Mid1 genomic locus with the restriction enzyme sites used to detect correct targeting of 61b3r (BamHI) and generation of the deletion (NdeI). On the right is the region surrounding exons 5–9 of the Arhgap6 genomic locus with restriction enzyme sites used to detect correct targeting of exons 6–8 (rhoGAP domain; BamHI and ApaI) and generation of the deletion (NdeI). (B) Replacement of 61b3r with a puromycin resistance gene (P), loxP site (black triangle) and the 3′-half of the HPRT minigene (3-H) replaces a wild-type 17 kb fragment with a 6 and 8.4 kb fragment, detected with the 5′ and 3′ external probes on Southern analysis with BamHI. Replacement of exons 6–8 of Arhgap6 with a neomycin resistance gene (N), a loxP site (black triangle) and the 5′-half of HPRT (5-H) replaces a 14 kb BamHI wild-type fragment with an 11.5 kb fragment, detected with the 5′ probe, and a 22 kb ApaI wild-type fragment with a 10.5 kb fragment, detected with the 3′ probe (probes are indicated by open boxes under the detected fragments). (C) After recombination between the two loxP sites, NdeI restriction digestion detects a novel 7.7 kb fragment with both deletion probes (shown as hatched boxes), that replaces a 9.2 kb fragment at the 61b3r and a 6.5 kb fragment at the Arhgap6 targeted alleles. Not shown are the wild-type allele sizes detected in this Southern analysis, which are <1 kb, and not detected on Southern analysis at the Arhgap6 locus and 5.5 kb at the 61b3r locus. (D) Southern analysis results of targeting described in C. Left panel: targeting at 61b3r; middle panel: detection of the 7.7 kb deleted fragment (note the fainter bands compared to wild-type and targeted alleles); right panel: targeting at Arhgap6.

Figure 3. Mosaicism of deletions. Southern analysis with NdeI of genomic DNA extracted from brain (B), lung (L), spleen (S), heart (H), kidney (K) and muscle (M) from one heterozygous targeted (F94) and two homozygous targeted (F211, F255) females. The asterisk indicates nearly absent deleted band in spleen in F94 compared with other tissues; the black circle indicates that brain is the only tissue where a deletion was found in F255. The lower panel shows the ethidium bromide-stained gel for comparison of loading.

Figure 3. Mosaicism of deletions. Southern analysis with NdeI of genomic DNA extracted from brain (B), lung (L), spleen (S), heart (H), kidney (K) and muscle (M) from one heterozygous targeted (F94) and two homozygous targeted (F211, F255) females. The asterisk indicates nearly absent deleted band in spleen in F94 compared with other tissues; the black circle indicates that brain is the only tissue where a deletion was found in F255. The lower panel shows the ethidium bromide-stained gel for comparison of loading.

Figure 4. HCCS BAC transgene. (A) Genomic structure of the human BAC transgene is shown with the coding exons of HCCS, and exons 12–14 of ARHGAP6 are shown. Southern hybridization with a probe in the 3′-region of HCCS (open rectangle) after HindIII (H) digestion detects a human-specific fragment of 7.5 kb. (B) Southern analysis of founder animals and offspring shows the human-specific band in positive lanes, as well as a fainter 5.5 kb mouse-specific, cross-hybridizing band in all lanes.

Figure 4. HCCS BAC transgene. (A) Genomic structure of the human BAC transgene is shown with the coding exons of HCCS, and exons 12–14 of ARHGAP6 are shown. Southern hybridization with a probe in the 3′-region of HCCS (open rectangle) after HindIII (H) digestion detects a human-specific fragment of 7.5 kb. (B) Southern analysis of founder animals and offspring shows the human-specific band in positive lanes, as well as a fainter 5.5 kb mouse-specific, cross-hybridizing band in all lanes.

Figure 5. Analysis of deleted mice. (A) Genotyping results of surviving mice with homozygous (F247), heterozygous (F248) and hemizygous (M247) deletions. (B) RT–PCR of total RNA extracted from brain (B), heart (H), liver (L), muscle (M) and kidney (K) of a MLSΔ/Y;HCCS+ animal, compared with RNA extracted from a wild-type (WT) mouse (a positive control for murine Hccs expression and negative control for human HCCS expression) and from an HCCS-transgenic mouse, which expresses both genes. RT:+ or − indicates reaction with or without reverse transcriptase. Panels from top to bottom are RT–PCR results from reactions with oligos complementary to mRNA of mouse Hccs (10–14F/10–14R), human HCCS (14cbX/14cbE), Arhgap6 exon 1 (19522/13749) and exons 2–4 (GAP2F/GAP4R), both 5′ to the rhoGAP domain, Arhgap6 3′ to the rhoGAP domain (13756/19411) and Mid1 (1416/1417). (C) Female heterozygous deleted mouse (MLSΔ/X) not expressing the human HCCS BAC-transgene. The smaller eyes can be seen in the MLSΔ/X female (top and left lower panel), compared with wild-type female littermate (right lower panel).

Figure 5. Analysis of deleted mice. (A) Genotyping results of surviving mice with homozygous (F247), heterozygous (F248) and hemizygous (M247) deletions. (B) RT–PCR of total RNA extracted from brain (B), heart (H), liver (L), muscle (M) and kidney (K) of a MLSΔ/Y;HCCS+ animal, compared with RNA extracted from a wild-type (WT) mouse (a positive control for murine Hccs expression and negative control for human HCCS expression) and from an HCCS-transgenic mouse, which expresses both genes. RT:+ or − indicates reaction with or without reverse transcriptase. Panels from top to bottom are RT–PCR results from reactions with oligos complementary to mRNA of mouse Hccs (10–14F/10–14R), human HCCS (14cbX/14cbE), Arhgap6 exon 1 (19522/13749) and exons 2–4 (GAP2F/GAP4R), both 5′ to the rhoGAP domain, Arhgap6 3′ to the rhoGAP domain (13756/19411) and Mid1 (1416/1417). (C) Female heterozygous deleted mouse (MLSΔ/X) not expressing the human HCCS BAC-transgene. The smaller eyes can be seen in the MLSΔ/X female (top and left lower panel), compared with wild-type female littermate (right lower panel).

Figure 6. Embryonic phenotype of MLSΔ deletions. Developmental stages E8.5, E9.5 and E10.5 for each genotype are indicated on the left. The genotype of the embryos is indicated at the top. X/Y;HCCS+ or − are wild type male embryos with or without the HCCS transgene for comparison; MLSΔ/X;HCCS+ are heterozygous deleted females carrying the HCCS transgene; MLSΔ/X;HCCS− are heterozygous deleted females not carrying the HCCS transgene. Magnifications are indicated on the first frame of each row, except for the E10.5 embryos. The asterisks indicate abnormal or developmentally delayed appearing embryos. Phenotypes include partial resorption, abnormal rotation, delayed development with disorganization of the somites and brain hypoplasia with distortion and thinness of the neural folds.

Figure 6. Embryonic phenotype of MLSΔ deletions. Developmental stages E8.5, E9.5 and E10.5 for each genotype are indicated on the left. The genotype of the embryos is indicated at the top. X/Y;HCCS+ or − are wild type male embryos with or without the HCCS transgene for comparison; MLSΔ/X;HCCS+ are heterozygous deleted females carrying the HCCS transgene; MLSΔ/X;HCCS− are heterozygous deleted females not carrying the HCCS transgene. Magnifications are indicated on the first frame of each row, except for the E10.5 embryos. The asterisks indicate abnormal or developmentally delayed appearing embryos. Phenotypes include partial resorption, abnormal rotation, delayed development with disorganization of the somites and brain hypoplasia with distortion and thinness of the neural folds.

Table 1.

Results from matings to induce (A) and transmit (B) mosaic deletions generated by EIIa Cre recombinase transgene expression

 Genotypes of parents LS M/F MLS2LoxP/ΔF/all F MLS2LoxP/ΔF/all F+M 
MLS2LoxP/X;Cre+×MLS2LoxP/Y;Cre+ 7.3 34/36 1/33 (3%) 1/70 (1.4%) 
 MLS2LoxP/MLS2LoxP;Cre+×MLS2LoxP/Y;Cre+ 6.3 27/41 7/41 (17%) 7/68 (10%) 
 X/X;Cre+×MLS2LoxP/Y;Cre+ 8.9 10/25 7/25 (28%) 7/35 (20%) 
MLS2LoxP/Δ;Cre+a×X/Y;Cre+ 4.5 15/15 0/15 (0%) 0/30 (0%) 
 MLS2LoxP/Δ;Cre+a×MLS2LoxP/Y;Cre+ 4.5 10/5 3/5 (60%) 3/15 (20%) 
 Genotypes of parents LS M/F MLS2LoxP/ΔF/all F MLS2LoxP/ΔF/all F+M 
MLS2LoxP/X;Cre+×MLS2LoxP/Y;Cre+ 7.3 34/36 1/33 (3%) 1/70 (1.4%) 
 MLS2LoxP/MLS2LoxP;Cre+×MLS2LoxP/Y;Cre+ 6.3 27/41 7/41 (17%) 7/68 (10%) 
 X/X;Cre+×MLS2LoxP/Y;Cre+ 8.9 10/25 7/25 (28%) 7/35 (20%) 
MLS2LoxP/Δ;Cre+a×X/Y;Cre+ 4.5 15/15 0/15 (0%) 0/30 (0%) 
 MLS2LoxP/Δ;Cre+a×MLS2LoxP/Y;Cre+ 4.5 10/5 3/5 (60%) 3/15 (20%) 

LS, average littersize; M/F, male to female ratio of offspring; MLS2LoxP/Δ, mosaic deleted femaleaThese include females that are homozygous (MLS2LoxP/Δ/MLS2LoxP/Δ) and heterozygous (MLS2LoxP/Δ/X) for the mosaic allele. The numbers for offspring of these females were combined.

Table 2.

Results from matings with MLS2LoxP;Cre+;HCCS+ transgenic mice

 Genotypes of parents LS MLS2LoxP/ΔF/all F MLSΔ F/all F MLSΔM/all M 
MLS2LoxP/MLS2LoxP;Cre+;HCCS+×MLS2LoxP/Y;Cre+;HCCS6.4 25% 4% 20% 
X/X×MLSΔ/Y;Cre+;HCCS+ (n=2) n/a 6/6 0/6 
 MLSΔ/X;Cre+;HCCS+×MLSΔ/Y;Cre+;HCCS+ (n=6) 5.6 n/a 9a/15 10/15 
 MLSΔ/X;Cre+;HCCS×X/Y (n=6) 7.4 n/a 16/32 3/20 
 Genotypes of parents LS MLS2LoxP/ΔF/all F MLSΔ F/all F MLSΔM/all M 
MLS2LoxP/MLS2LoxP;Cre+;HCCS+×MLS2LoxP/Y;Cre+;HCCS6.4 25% 4% 20% 
X/X×MLSΔ/Y;Cre+;HCCS+ (n=2) n/a 6/6 0/6 
 MLSΔ/X;Cre+;HCCS+×MLSΔ/Y;Cre+;HCCS+ (n=6) 5.6 n/a 9a/15 10/15 
 MLSΔ/X;Cre+;HCCS×X/Y (n=6) 7.4 n/a 16/32 3/20 

Columns are labeled as follows: LS, average littersize; MLS2LoxP/Δ F are mosaic deleted females; MLSΔF are HCCS+ females with heterozygous or homozygous deletions and MLSΔ M are HCCS+ males with hemizygous deletions.

aOne of the female offspring with a deletion does not express HCCS (see text).

Table 3.

Embryonic genotype and phenotype from offspring of MLSΔ/Y;HCCS+;Cre+ mated with X/X animals or with MLSΔ/X;HCCS+;Cre+ animals

Developmental stage E8.5 E9.5 E10.5 
Genotype Phenotype 
MLSΔ/X;HCCS+, MLSΔ/MLSΔ;HCCS+, MLSΔ/Y;HCCSNormal 14 
Abnormal 12 
 Observed/expecteda 26/20 16/15 5/3 
MLSΔ/X;HCCS−, MLSΔ/MLSΔ;HCCS−, MLSΔ/Y;HCCS− Normal 
Abnormal 
 Observed/expecteda 2/9 6/11 0/2 
X/Y (HCCS+ or −) Normal 10 12 
Abnormal 
 Observed/expecteda 14/13 20/16 4/4 
Total  42 42 
Developmental stage E8.5 E9.5 E10.5 
Genotype Phenotype 
MLSΔ/X;HCCS+, MLSΔ/MLSΔ;HCCS+, MLSΔ/Y;HCCSNormal 14 
Abnormal 12 
 Observed/expecteda 26/20 16/15 5/3 
MLSΔ/X;HCCS−, MLSΔ/MLSΔ;HCCS−, MLSΔ/Y;HCCS− Normal 
Abnormal 
 Observed/expecteda 2/9 6/11 0/2 
X/Y (HCCS+ or −) Normal 10 12 
Abnormal 
 Observed/expecteda 14/13 20/16 4/4 
Total  42 42 

This table shows results of embryo genotyping from 13 timed matings: six at E8.5, five at E9.5, and two at E10.5. The average number of embryos found per mating was 7.7 at E8.5, 8.6 at E9.5, and 5.0 at E10.5.

a‘Expected’ refers to the total number of embryos of each genotype that would be expected based on Mendelian segregation of the parental genotypes.

References

1
Temple, I.K., Hurst, J.A., Hing, S., Butler, L. and Baraitser, M. (
1990
) De novo deletion of Xp22.2-pter in a female with linear skin lesions of the face and neck, microphthalmia, and anterior chamber eye anomalies.
J. Med. Genet.
 ,
27
,
56
–58.
2
al-Gazali, L.I., Mueller, R.F., Caine, A., Antoniou, A., McCartney, A., Fitchett, M. and Dennis, N.R. (
1990
) Two 46,XX,t(X;Y) females with linear skin defects and congenital microphthalmia: a new syndrome at Xp22.3.
J. Med. Genet.
 ,
27
,
59
–63.
3
Zvulunov, A., Kachko, L., Manor, E., Shinwell, E. and Carmi, R. (
1998
) Reticulolinear aplasia cutis congenita of the face and neck: a distinctive cutaneous manifestation in several syndromes linked to Xp22.
Br. J. Dermatol.
 ,
138
,
1046
–1052.
4
Lindsay, E.A., Grillo, A., Ferrero, G.B., Roth, E.J., Magenis, E., Grompe, M., Hulten, M., Gould, C., Baldini, A. and Zoghbi, H.Y. (
1994
) Microphthalmia with linear skin defects (MLS) syndrome: clinical, cytogenetic, and molecular characterization.
Am. J. Med. Genet.
 ,
49
,
229
–234.
5
Paulger, B.R., Kraus, E.W., Pulitzer, D.R. and Moore, C.M. (
1997
) Xp microdeletion syndrome characterized by pathognomonic linear skin defects on the head and neck.
Pediatr. Dermatol.
 ,
14
,
26
–30.
6
Ogata, T., Wakui, K., Muroya, K., Ohashi, H., Matsuo, N., Brown, D.M., Ishii, T. and Fukushima, Y. (
1998
) Microphthalmia with linear skin defects syndrome in a mosaic female infant with monosomy for the Xp22 region: molecular analysis of the Xp22 breakpoint and the X-inactivation pattern.
Hum. Genet.
 ,
103
,
51
–56.
7
Stratton, R.F., Walter, C.A., Paulgar, B.R., Price, M.E. and Moore, C.M. (
1998
) Second 46,XX male with MLS syndrome.
Am. J. Med. Genet.
 ,
76
,
37
–41.
8
Kobayashi, M., Kiyosawa, M., Toyoura, T. and Tokoro, T. (
1998
) An XX male with microphthalmos and sclerocornea.
J. Pediatr. Ophthal. Strabismus
 ,
35
,
122
–124.
9
Kono, T., Migita, T., Koyama, S. and Seki, I. (
1999
) Another observation of microphthalmia in an XX male: microphthalmia with linear skin defects syndrome without linear skin lesions.
J. Hum. Genet.
 ,
44
,
63
–68.
10
Schaefer, L., Ferrero, G.B., Grillo, A., Bassi, M.T., Roth, E.J., Wapenaar, M.C., van Ommen, G.J., Mohandas, T.K., Rocchi, M. and Zoghbi, H.Y. (
1993
) A high resolution deletion map of human chromosome Xp22.
Nat. Genet.
 ,
4
,
272
–279.
11
Wapenaar, M.C., Bassi, M.T., Schaefer, L., Grillo, A., Ferrero, G.B., Chinault, A.C., Ballabio, A. and Zoghbi, H.Y. (
1993
) The genes for X-linked ocular albinism (OA1) and microphthalmia with linear skin defects (MLS): cloning and characterization of the critical regions.
Hum. Mol. Genet.
 ,
2
,
947
–952.
12
Wapenaar, M.C., Schiaffino, M.V., Bassi, M.T., Schaefer, L., Chinault, A.C., Zoghbi, H.Y. and Ballabio, A. (
1994
) A YAC-based binning strategy facilitating the rapid assembly of cosmid contigs: 1.6 Mb of overlapping cosmids in Xp22.
Hum. Mol. Genet.
 ,
3
,
1155
–1161.
13
Schaefer, L., Ballabio, A. and Zoghbi, H.Y. (
1996
) Cloning and characterization of a putative human holocytochrome c-type synthetase gene (HCCS) isolated from the critical region for microphthalmia with linear skin defects (MLS).
Genomics
 ,
34
,
166
–172.
14
Schaefer, L., Prakash, S. and Zoghbi, H.Y. (
1997
) Cloning and characterization of a novel rho-type GTPase-activating protein gene (ARHGAP6) from the critical region for microphthalmia with linear skin defects.
Genomics
 ,
46
,
268
–277.
15
Van den Veyver, I.B., Cormier, T.A., Jurecic, V., Baldini, A. and Zoghbi, H.Y. (
1998
) Characterization and physical mapping in human and mouse of a novel RING finger gene in Xp22.
Genomics
 ,
51
,
251
–261.
16
Kayserili, H., Cox, T.C., Cox, L.L., Basaran, S., Kilic, G., Ballabio, A. and Yuksel-Apak, M. (
2001
) Molecular characterisation of a new case of microphthalmia with linear skin defects (MLS).
J. Med. Genet.
 ,
38
,
411
–417.
17
Liu, J.P. and Jessell, T.M. (
1998
) A role for rhoB in the delamination of neural crest cells from the dorsal neural tube.
Development
 ,
125
,
5055
–5067.
18
Wunnenberg-Stapleton, K., Blitz, I.L., Hashimoto, C. and Cho, K.W. (
1999
) Involvement of the small GTPases XRhoA and XRnd1 in cell adhesion and head formation in early Xenopus development.
Development
 ,
126
,
5339
–5351.
19
Lucas, J.M., Nikolic, I. and Hens, M.D. (
2002
) cDNA cloning, sequence comparison, and developmental expression of Xenopus rac1.
Mech. Devl.
 ,
115
,
113
–116.
20
Prakash, S.K., Paylor, R., Jenna, S., Lamarche-Vane, N., Armstrong, D.L., Xu, B., Mancini, M.A. and Zoghbi, H.Y. (
2000
) Functional analysis of ARHGAP6, a novel GTPase-activating protein for RhoA.
Hum. Mol. Genet.
 ,
9
,
477
–488.
21
Van den Veyver, I.B., Subramanian, S. and Zoghbi, H.Y. (
1998
) Genomic structure of a human holocytochrome c-type synthetase gene in Xp22.3 and mutation analysis in patients with Rett syndrome.
Am. J. Med. Genet.
 ,
78
,
179
–181.
22
Drygas, M.E., Lambowitz, A.M. and Nargang, F.E. (
1989
) Cloning and analysis of the Neurospora crassa gene for cytochrome c heme lyase.
J. Biol. Chem.
 ,
264
,
17897
–17906
23
Dumont, M.E., Ernst, J.F. and Sherman, F. (
1988
) Coupling of heme attachment to import of cytochrome c into yeast mitochondria. Studies with heme lyase-deficient mitochondria and altered apocytochromes c.
J. Biol. Chem.
 ,
263
,
15928
–15937.
24
Dumont, M.E., Cardillo, T.S., Hayes, M.K. and Sherman, F. (
1991
) Role of cytochrome c heme lyase in mitochondrial import and accumulation of cytochrome c in Saccharomyces cerevisiae.
Mol. Cell. Biol.
 ,
11
,
5487
–5496.
25
Diekert, K., Kispal, G., Guiard, B. and Lill, R. (
1999
) An internal targeting signal directing proteins into the mitochondrial intermembrane space.
Proc. Natl Acad. Sci. USA
 ,
96
,
11752
–11757.
26
Mayer, A., Neupert, W. and Lill, R. (
1995
) Translocation of apocytochrome c across the outer membrane of mitochondria.
J. Biol. Chem.
 ,
270
,
12390
–12397.
27
Steiner, H., Kispal, G., Zollner, A., Haid, A., Neupert, W. and Lill, R. (
1996
) Heme binding to a conserved Cys-Pro-Val motif is crucial for the catalytic function of mitochondrial heme lyases.
J. Biol. Chem.
 ,
271
,
32605
–32611.
28
Kranz, R., Lill, R., Goldman, B., Bonnard, G. and Merchant, S. (
1998
) Molecular mechanisms of cytochrome c biogenesis: three distinct systems.
Mol. Microbiol.
 ,
29
,
383
–396.
29
Cox, T.C., Allen, L.R., Cox, L.L., Hopwood, B., Goodwin, B., Haan, E. and Suthers, G.K. (
2000
) New mutations in MID1 provide support for loss of function as the cause of X-linked opitz syndrome.
Hum. Mol. Genet.
 ,
9
,
2553
–2562.
30
Quaderi, N.A., Schweiger, S., Gaudenz, K., Franco, B., Rugarli, E.I., Berger, W., Feldman, G.J., Volta, M., Andolfi, G., Gilgenkrantz, S. et al. (
1997
) Opitz G/BBB syndrome, a defect of midline development, is due to mutations in a new RING finger gene on Xp22.
Nat. Genet.
 ,
17
,
285
–291.
31
Gaudenz, K., Roessler, E., Quaderi, N., Franco, B., Feldman, G., Gasser, D.L., Wittwer, B., Horst, J., Montini, E., Opitz, J.M. et al. (
1998
) Opitz G/BBB syndrome in Xp22: mutations in the MID1 gene cluster in the carboxy-terminal domain.
Am. J. Hum. Genet.
 ,
63
,
703
–710.
32
Cainarca, S., Messali, S., Ballabio, A. and Meroni, G. (
1999
) Functional characterization of the Opitz syndrome gene product (midin): evidence for homodimerization and association with microtubules throughout the cell cycle.
Hum. Mol. Genet.
 ,
8
,
1387
–1396.
33
Liu, J., Prickett, T.D., Elliott, E., Meroni, G. and Brautigan, D.L. (
2001
) Phosphorylation and microtubule association of the Opitz syndrome protein mid-1 is regulated by protein phosphatase 2A via binding to the regulatory subunit alpha 4.
Proc. Natl Acad. Sci. USA
 ,
98
,
6650
–6655.
34
Trockenbacher, A., Suckow, V., Foerster, J., Winter, J., Krauss, S., Ropers, H.H., Schneider, R. and Schweiger, S. (
2001
) MID1, mutated in Opitz syndrome, encodes an ubiquitin ligase that targets phosphatase 2A for degradation.
Nat. Genet.
 ,
29
,
287
–294.
35
Naritomi, K., Izumikawa, Y., Nagataki, S., Fukushima, Y., Wakui, K., Niikawa, N. and Hirayama, K. (
1992
) Combined Goltz and Aicardi syndromes in a terminal Xp deletion: are they a contiguous gene syndrome?
Am. J. Med. Genet.
 ,
43
,
839
–843.
36
Donnenfeld, A.E., Packer, R.J., Zackai, E.H., Chee, C.M., Sellinger, B. and Emanuel, B.S. (
1989
) Clinical, cytogenetic, and pedigree findings in 18 cases of Aicardi syndrome.
Am. J. Med. Genet.
 ,
32
,
461
–467.
37
Ropers, H.H., Zuffardi, O., Bianchi, E. and Tiepolo, L. (
1982
) Agenesis of corpus callosum, ocular, and skeletal anomalies (X-linked dominant Aicardi's syndrome) in a girl with balanced X/3 translocation.
Hum. Genet.
 ,
61
,
364
–368.
38
Donnenfeld, A.E., Graham, J.M., Jr., Packer, R.J., Aquino, R., Berg, S.Z. and Emanuel, B.S. (
1990
) Microphthalmia and chorioretinal lesions in a girl with an Xp22.2-pter deletion and partial 3p trisomy: clinical observations relevant to Aicardi syndrome gene localization.
Am. J. Med. Genet.
 ,
37
,
182
–186.
39
Goltz, R.W., Peterson, W.C., Gorlin, R.J. and Ravits, A.G. (
1962
) Focal dermal hypoplasia.
Arch. Dermatol.
 ,
708
–717.
40
Goltz, R.W., Henderson, R.R., Hitch, J.M. and Ott, J.E. (
1970
) Focal dermal hypoplasia syndrome. A review of the literature and report of two cases.
Arch Dermatol.
 ,
101
,
1
–11.
41
Prakash, S.K., Van den Veyver, I.B., Franco, B., Volta, M., Ballabio, A. and Zoghbi, H.Y. (
1999
) Characterization of a novel chromo domain gene in xp22.3 with homology to Drosophila msl-3.
Genomics
 ,
59
,
77
–84.
42
Lakso, M., Pichel, J.G., Gorman, J.R., Sauer, B., Okamoto, Y., Lee, E., Alt, F.W. and Westphal, H. (
1996
) Efficient in vivo manipulation of mouse genomic sequences at the zygote stage.
Proc. Natl Acad. Sci. USA
 ,
93
,
5860
–5865.
43
Hirotsune, S., Fleck, M.W., Gambello, M.J., Bix, G.J., Chen, A., Clark, G.D., Ledbetter, D.H., McBain, C.J. and Wynshaw-Boris, A. (
1998
) Graded reduction of Pafah1b1 (Lis1) activity results in neuronal migration defects and early embryonic lethality.
Nat. Genet.
 ,
19
,
333
–339.
44
Avner, P., Prissette, M., Arnaud, D., Courtier, B., Cecchi, C. and Heard, E. (
1998
) Molecular correlates of the murine Xce locus.
Genet. Res.
 ,
72
,
217
–224.
45
Percec, I., Plenge, R.M., Nadeau, J.H., Bartolomei, M.S. and Willard, H.F. (
2002
) Autosomal dominant mutations affecting X inactivation choice in the mouse.
Science
 ,
296
,
1136
–1139.
46
Liu, X., Kim, C.N., Yang, J., Jemmerson, R. and Wang, X. (
1996
) Induction of apoptotic program in cell-free extracts: requirement for dATP and cytochrome c.
Cell
 ,
86
,
147
–157.
47
Li, K., Li, Y., Shelton, J.M., Richardson, J.A., Spencer, E., Chen, Z.J., Wang, X. and Williams, R.S. (
2000
) Cytochrome c deficiency causes embryonic lethality and attenuates stress-induced apoptosis.
Cell
 ,
101
,
389
–399.
48
Zhu, Z., Yao, J., Johns, T., Fu, K., De Bie, I., Macmillan, C., Cuthbert, A.P., Newbold, R.F., Wang, J., Chevrette, M. et al. (
1998
) SURF1, encoding a factor involved in the biogenesis of cytochrome c oxidase, is mutated in Leigh syndrome.
Nat. Genet.
 ,
20
,
337
–343.
49
Papadopoulou, L.C., Sue, C.M., Davidson, M.M., Tanji, K., Nishino, I., Sadlock, J.E., Krishna, S., Walker, W., Selby, J., Glerum, D.M. et al. (
1999
) Fatal infantile cardioencephalomyopathy with COX deficiency and mutations in SCO2, a COX assembly gene.
Nat. Genet.
 ,
23
,
333
–337.
50
DiMauro, S. and De Vivo, D.C. (
1996
) Genetic heterogeneity in Leigh syndrome.
Ann. Neurol.
 ,
40
,
5
–7.
51
Shevell, M.I., Matthews, P.M., Scriver, C.R., Brown, R.M., Otero, L.J., Legris, M., Brown, G.K. and Arnold, D.L. (
1994
) Cerebral dysgenesis and lactic acidemia: an MRI/MRS phenotype associated with pyruvate dehydrogenase deficiency.
Pediatr. Neurol.
 ,
11
,
224
–229.
52
Ramirez-Solis, R., Liu, P. and Bradley, A. (
1995
) Chromosome engineering in mice.
Nature
 ,
378
,
720
–724.