Deletion of Ultraconserved Elements Yields Viable Mice

Ultraconserved elements have been suggested to retain extended perfect sequence identity between the human, mouse, and rat genomes due to essential functional properties. To investigate the necessities of these elements in vivo, we removed four noncoding ultraconserved elements (ranging in length from 222 to 731 base pairs) from the mouse genome. To maximize the likelihood of observing a phenotype, we chose to delete elements that function as enhancers in a mouse transgenic assay and that are near genes that exhibit marked phenotypes both when completely inactivated in the mouse and when their expression is altered due to other genomic modifications. Remarkably, all four resulting lines of mice lacking these ultraconserved elements were viable and fertile, and failed to reveal any critical abnormalities when assayed for a variety of phenotypes including growth, longevity, pathology, and metabolism. In addition, more targeted screens, informed by the abnormalities observed in mice in which genes in proximity to the investigated elements had been altered, also failed to reveal notable abnormalities. These results, while not inclusive of all the possible phenotypic impact of the deleted sequences, indicate that extreme sequence constraint does not necessarily reflect crucial functions required for viability.


INTRODUCTION
The use of evolutionary conservation has become a powerful means for identifying functionally important genomic sequences [1,2]. Ultraconserved elements have been defined as a group of extremely conserved sequences that show 100% identity over 200bp or greater between the human, mouse, and rat genomes [3]. This category of extreme evolutionary sequence conservation is represented by 481 sequences in the human genome of which over half show no evidence of transcription. Further analysis of the distribution of these non-coding ultraconserved elements demonstrate that they tend to cluster in regions that are enriched for transcription factors and developmental genes [3], and a limited number of functional studies suggest a role for some of these noncoding elements in gene regulation [4][5][6].
Several hypotheses have been proposed to explain the extreme sequence constraint of ultraconserved elements including strong negative selective pressure and/or reduced mutation rates [3]. The negative selection hypothesis postulates that crucial functions such as vital gene regulatory information is embedded within these sequences, while the reduced mutation rate hypothesis suggests that these sequences exist in a hyperrepaired or hypomutable state [3]. Recent analysis of human variation in these non-coding ultraconserved elements provides compelling evidence supporting negative selection as contributing to their extreme evolutionary conservation [7]. Furthermore, non-coding ultraconserved elements have also been shown to be significantly depleted in human segmental duplications and copy number variants, suggesting that disruption of their normal copy number may lead to reduced fitness [8]. In this study, we removed four carefully chosen non-coding ultraconserved elements in the mouse genome to directly explore a functional role for these elements in vivo.

Generation and general characterization of ultraconserved knockout mice
To increase the probability of observing an associated phenotype in the ultraconserved null mice, we employed a variety of criteria in selecting the non-coding ultraconserved elements for deletion. We chose elements that showed tissue-specific in vivo enhancer activity in a mouse transgenic reporter assay that tended to recapitulate aspects of the expression pattern found in genes that were in their proximity ( Figure 1) [6]. Other factors that were taken into account in prioritizing elements for deletion included their proximity to genes whose inactivation and alterations in their expression results in specific phenotypes that we could screen for in the ultraconserved element deletion mice (Table   1). Elements meeting most of these criteria were chosen for removal and included: uc248, uc329, uc467, and uc482 ( Figure 1) [3]; representing 222, 307, 731, and 295 base pairs respectively of 100% identity between human, mouse, and rat.
All four non-coding ultraconserved elements were deleted from the mouse genome using standard mouse genetic engineering techniques, and removal was confirmed by PCR and Southern blot hybridization (Supplementary Methods). We first examined each line for the viability of homozygous/hemizygous knockout mice in mixed crosses, and observed that all four lines showed no reduction in the expected number of homozygous/ hemizygous mice that were generated (Table 2). Homozygous matings within the four lines revealed no significant differences in viability and litter size compared to the wildtype littermates (Table 2). We next examined body weight (up to ten weeks of age;  Table 1), with one exception. The exception was one uc329 homozygous male having unilateral renal agenesis. Additional analysis of 102 uc329 homozygous null mice revealed a total of 2 mice (including the initial propositus) with one kidney compared to none within the 30 uc329 wild-type littermates that were screened. Unilateral renal agenesis is estimated to occur in 1 to 1000 live births in humans [9] and is asymptomatic and unassociated with a reduction in survival rate [10]. Possible explanations for unilateral renal agenesis in ~2% of uc329 homozygous null mice in this study include a spontaneous event unassociated with the deleted element or a low penetrance phenotype caused by the absence of this element.

Screens for phenotypes of adjacent genes
In addition to the above general screens, we screened each of these mouse lines for phenotypes specifically associated with the inactivation or dosage abnormality of the genes in proximity to the deleted ultraconserved elements. Ultraconserved element 248 is bracketed by the genes doublesex and mab3-related transcription factor 1 (DMRT1; Entrez Gene ID 1761) and doublesex and mab3-related transcription factor 3 (DMRT3; Entrez Gene ID 58524) ( Figure 1A). In humans, haploinsufficiency due to chromosomal aberrations within this region leads to XY sex reversal [11]. In mice, Dmrt1 homozygous knockouts exhibit defects in testicular development [12], while DMRT3 function is unknown. In order to identify the phenotype associated with Dmrt3 deficiency for these studies we deleted Dmrt3 from the mouse genome. All Dmrt3 null homozygous mice died from starvation at two months of age due to dental malocclusions, and in addition some of the males exhibited male sexual development abnormalities (N. Ahituv unpublished results). Based on these results, we extensively phenotyped uc248 homozygous null mice for sexual and dental abnormalities. Pathological analysis of both male and female sexual organs and teeth in 6 week old uc248 null mice revealed no obvious defects (Supplementary Table 1). In addition, heterozygous and homozygous crosses exhibited no reduction in expected homozygous offspring ( Table 2).
Ultraconserved element 467, the longest solitary non-coding ultraconserved element in the human genome (731bp), lies inside the last intron of polymerase alpha (POLA; Entrez Gene ID 5422) adjacent to the aristaless-related homeobox (ARX; Entrez Gene ID 170302) gene ( Figure 1C). Mutations in ARX in humans leads to a wide range of neurological and sexual development disorders [13,14], while hemizygous Arx null male mice die shortly after birth and have small brains and male sexual development abnormalities [15]. In addition, a duplication of this region in mice, caused by insertional mutagenesis, leads to embryonic lethality due to exencephaly accompanied by anophthalmia [16]. Detailed pathological examination of the reproductive organs and neuroanatomical examination of the brains of uc467 null mice revealed no apparent abnormalities (Supplementary Table 1). In addition, the mice showed no obvious differences in the offspring expected from the hemizygous x heterozygous and hemizygous x homozygous crosses ( Table 2). Pax6 (OMIM #607108) have a variety of phenotypes, the most notable being kidney and eye abnormalities respectively. Detailed pathological analysis of the kidneys and eyes of the uc329 null mice revealed no significant differences compared to the wild-type littermates (Supplementary Table 1), other than the ~2% unilateral renal agenesis discussed above. Clinical chemistry tests revealed slightly higher urea N levels compared to the wild-type littermates (33. 16 (Table 2). Growth hormone abnormalities would be expected to lead to body weight irregularities, none of which were detected (Figure 2). Calcium levels were also normal (Supplementary Figure 1) supporting a lack of marked abnormalities in parathyroid gland function.

DISCUSSION
Based on the compelling evidence that ultraconserved elements are conserved due to functional constraint, it has been proposed that their removal in vivo would lead to a significant phenotypic impact [7,8]. Accordingly our results were unexpected. It is likely that our assays were not able to detect dramatic phenotypes that under a different setting, for instance outside the controlled laboratory setting, would become evident. Moreover, possible phenotypes might become evident only on a longer timescale, such as longer generation time. It is also possible that subtler genetic manipulations of the ultraconserved elements might lead to an evident phenotype due to a gain of function type mechanism. All four elements examined in this study demonstrated in vivo enhancer activity when tested in a transgenic mouse assay (Figure 1) [6], which would suggest regulatory element redundancy as another possible explanation for the lack of a significant impact following the removal of these specific elements. Just as gene redundancy has been shown to be responsible for the lack of phenotypes associated with many seemingly vital genes knockouts, regulatory sequence redundancy [22] can similarly provide a possible explanation for the lack of a marked phenotype. While our studies have not defined a specific need for the extreme sequence constraints of noncoding ultraconserved elements, they have ruled out the hypothesis that these constraints reflect crucial functions required for viability.

Generation of ultraconserved element null mice. The basic technology used for gene
targeting and screening has been described previously [23]. Briefly, the four selected ultraconserved elements were removed in W4/129S6 mouse embryonic stem cells   . a, uc248 region. b, uc329 region. c, uc467 region. d, uc482 region. A black oval represents each ultraconserved element while the above embryos represent observed positive enhancer activities captured through transgenic mouse testing at e11.5 for that element [6]. Stained embryos in boxes represent whole-mount in situ hybridizations of the specific gene at e11.5 (genes without stained embryos were negative for this assay at this time point). Southern blot hybridization (gel results below for each element), and checked for G418 sensitivity. Positive colonies were injected into C57BL/6J blastocyst stage embryos. Chimeric mice were subsequently crossed to C57BL/6J mice, generating agouti offspring that were heterozygous/hemizygous for the ultraconserved element deletion and were intercrossed to generate homozygous ultraconserved null mice.

Genotyping
Genomic DNA was extracted from a 1-cm section of tail that was incubated overnight in lysis buffer (containing 50mM Tris-HCl pH 8.0, 1mM EDTA, 1% SDS, 20mM NaCl and 1mg/ml Proteinase K) at 55 degrees Celsius. Genotyping was performed by PCR, with primers outside the deleted segments and within the targeting vector (see primer tables below for each element).