Generation and characterization of a conditional deletion allele for Lmna in mice

https://doi.org/10.1016/j.bbrc.2013.08.082Get rights and content

Highlights

  • A new Lmna conditional knockout allele (KO) deletes exon 2 upon Cre expression.

  • Mice deleted of exon 2 by germline Cre expression phenocopy conventional Lmna KO mice.

  • The new Lmna conditional KO mouse is a useful tool to examine the tissue specific role of A-type lamins.

Abstract

Extensive efforts have been devoted to study A-type lamins because mutations in their gene, LMNA in humans, are associated with a number of diseases. The mouse germline mutations in the A-type lamins (encoded by Lmna) exhibit postnatal lethality at either 4–8 postnatal (P) weeks or P16–18 days, depending on the deletion alleles. These mice exhibit defects in several tissues including hearts and skeletal muscles. Despite numerous studies, how the germline mutation of Lmna, which is expressed in many postnatal tissues, affects only selected tissues remains poorly understood. Addressing the tissue specific functions of Lmna requires the generation and careful characterization of conditional Lmna null alleles. Here we report the creation of a conditional Lmna knockout allele in mice by introducing loxP sites flanking the second exon of Lmna. The Lmnaflox/flox mice are phenotypically normal and fertile. We show that Lmna homozygous mutants (LmnaΔ/Δ) generated by germline Cre expression display postnatal lethality at P16–18 days with defects similar to a recently reported germline Lmna knockout mouse that exhibits the earliest lethality compared to other germline knockout alleles. This conditional knockout mouse strain should serve as an important genetic tool to study the tissue specific roles of Lmna, which would contribute toward the understanding of various human diseases associated with A-type lamins.

Introduction

The nuclear lamina (NL) is a meshwork of proteins that lines the inner nuclear membrane and serves as a nuclear scaffold. Lamins, which belong to the type V intermediate filaments, are the major components of the NL. Lamin proteins were initially subdivided based on immunological and biochemical criteria [1], [2]. A-type lamins have nearly neutral isoelectric points (pI) and are almost completely solubilized during mitosis, whereas B-type lamins have acidic pI and remain associated with nuclear envelope derived membranes during mitosis. Since the first cDNA cloning of A-type lamins in humans [3], lamins have been identified in many vertebrates. Sequence comparisons of lamin proteins have revealed that despite of overall similarity, A- and B-type lamins exhibit subclass-specific sequence variations [4], [5], [6]. It is also known that both A and B-type lamins are initially carboxymethylated and farnesylated at the C-terminal CaaX motif, but these modifications are removed from lamin-A by proteolysis (reviewed in [7]).

In the mouse, Lmnb1 and Lmnb2 encode lamin-B1 and lamin-B2, respectively. Lmnb2 also expresses lamin-B3 through alternatively splicing in the testes. On the other hand, Lmna expresses several forms of A-type lamins, including lamin-A, -C, -AΔ10, and -C2, via alternative splicing. While B-type lamins are expressed in most cell types throughout development, the expression of A-type lamins is either lacking or very low during early development with increasing levels of expression in some cell types as development progresses [8], [9], [10]. For example, during mouse embryogenesis, the expression of A-type lamins is limited to a few tissues such as the trunk and epidermis, while in many other tissues, lamin-A/C is not detected until well after birth. In adult mice, lamin-A/C are expressed in many tissues at high levels [9].

Since the discovery that mutations in LMNA cause autosomal dominant Emery–Dreifuss muscular dystrophy (EDMD) in humans [11], many additional LMNA mutations have been shown to cause a large group of human diseases ranging from muscular dystrophy to premature aging disease. Several germline Lmna mutant mouse models have been generated with the aim of understanding the mechanisms by which A-type lamins function in health and diseases [12], [13]. Sullivan and colleagues showed that homozygous mice deleted of exons 8–11 of Lmna (referred to here as LmnaΔ811/Δ811) in the germline develop to term with no overt abnormalities [14]. However, their postnatal growths are severely retarded, and these mice present with a subset of pathologies similar to those caused by LMNA mutations, including muscular dystrophy, dilated cardiomyopathy, and Charcot–Marie–Tooth syndrome in humans [15]. These mice die at around 4–8 weeks after birth. Despite the widespread use of this mouse model, recent studies show that these mice still express a truncated lamin-A protein of 54 kDa, which corresponds to the 468 amino acids encoded by exons 1–7 and 12, through alternative splicing [16]. The expression of this large lamin-A fragment might function dominant negatively because heterozygous Lmna+/Δ811 mice develop dilated cardiomyopathies 4 weeks after birth [17], which complicates phenotypic interpretation.

Recently, another Lmna germline knockout mouse line has been created using gene trap (GT) technology [13]. In the LmnaGT−/− mouse, a gene trap cassette, consisting of a splicing acceptor and a reporter gene, is introduced into the upstream sequence of exon 2 of Lmna, and only the N-terminal 118 amino acids of lamin-A is expressed as a fusion with the reporter protein. LmnaGT−/− mice are overtly normal during embryogenesis. However, they succumb to death at 16–18 days after birth, which is much earlier than LmnaΔ811/Δ811 mice. Therefore, LmnaGT−/− mice are the most severe loss-of-function mutants for A-type lamins to date.

Although mutations of LMNA are associated with diverse human diseases, a careful look at the diseases and animal models reveals that the defects caused by different mutations are manifested in limited tissue types including striated muscles, adipose tissues, and peripheral nerves [18]. How mutations in the broadly expressed lamin-A cause tissue-restricted defects is unclear. Several nonexclusive hypotheses have been proposed to account for the molecular basis of this selectivity [19]. For example, the mechanical stress model posits that cell types such as muscle cells that are under constant mechanical stress might be more susceptible to the alteration of the nuclear lamina [20]. The gene expression model, on the other hand, proposes that lamins influence gene expression profiles in a cell type specific manner [21], [22]. Although studies using the germline Lmna mutant mouse models created thus far have provided support to aspects of the above hypotheses, understanding the function of Lmna in a given tissue requires the creation of a conditional Lmna knockout allele. Recently, Solovei et al. reported the use of an Lmna conditional knockout allele, which leads to the deletion of the last three exons (exons 10–12) of Lmna upon Cre expression [23]. The predicted lack of polyA signal in Lmna transcripts from exons 1–9 may cause their degradation. However, these transcripts could utilize a cryptic polyA signal in the absence of exons 10–12, which would lead to the generation of truncated lamin-A/C proteins. Unfortunately, the authors provided no characterization of this Lmna conditional knockout allele. Thus it remains unclear whether truncated lamin-A/C proteins are present upon Cre expression in these mice. To facilitate the study of the tissue autonomous function of A-type lamins, we report the generation and characterization of a different conditional Lmna knockout mouse model that deletes the majority of lamin-A/C proteins upon Cre-mediated recombination.

Section snippets

Construction of the targeting vector

The targeting vector to generate conditional knockout alleles of Lmna was constructed using the recombineering protocol established by the Capecchi laboratory [24]. Briefly, an 8 kb genomic region from 2 kb upstream of exon 2 through the middle of exon 11 of Lmna was retrieved from a mouse BAC clone RP24-265E18, and was cloned into the pStart-K vector using Red-recombinase mediated cloning method. The construct was modified to add the 5′ and 3′ loxP sites at 100 bp upstream and downstream,

Generation of Lmnafflox/+, Lmnaflox/+, and Lmnaflox/flox mice

Wild-type Lmna allele consists of 12 exons, which encode 665 amino acids, and an alternative first exon, 1c, that contributes to 7 amino acids at the N-terminus of lamin-C2 (Fig. 1A). Exon 1 and 1c of Lmna are separated by an intronic region of over 9 kb, which could make the deletion of the two exons inefficient. This would be undesirable for a conditional allele. Therefore, to remove all isoforms of lamin-A including lamin-C2, we set out to create a conditional exon 2 deletion allele. We

Acknowledgments

We thank O. Martin for technical support, and the members of the Zheng lab for critical comments. Supported by HHMI, R01 GM56312, and Ellison Medical Foundation.

References (26)

  • F.D. McKeon et al.

    Homologies in both primary and secondary structure between nuclear envelope and intermediate filament proteins

    Nature

    (1986)
  • G. Krohne et al.

    Nuclear lamin LI of Xenopus laevis: cDNA cloning, amino acid sequence and binding specificity of a member of the lamin B subfamily

    EMBO J.

    (1987)
  • B. Burke et al.

    The nuclear lamins: flexibility in function

    Nat. Rev. Mol. Cell Biol.

    (2013)
  • Cited by (39)

    • Genetic Ablation of the DNA Damage Response Pathway Attenuates Lamin-Associated Dilated Cardiomyopathy in Mice

      2022, JACC: Basic to Translational Science
      Citation Excerpt :

      Generation and phenotypic characterization of the Myh6-Cre:LmnaF/F mice have been published.17 In brief, to delete the Lmna gene in cardiac myocytes, the mice with the floxed exon 2 of the Lmna gene (LmnaF/F) were crossed to the Myh6-Cre transgenic mice, as published.17-19 Mice were housed in a 12-hour light/dark cycle facility with food and water available ad libitum.

    • Integrating mechanical signals into cellular identity

      2022, Trends in Cell Biology
      Citation Excerpt :

      Consistent with a mechanically attenuated state, loss of function for most, if not all, genes implicated in both indirect and direct mechanotransduction have minimal deleterious effects on pluripotent cells. For example, although mouse embryos null for SRF [47], MKL1/2 [48,49], lamins [50–52], emerin [53], Nesprins [54,55], and the broadly expressed SUN proteins (Sun1 and Sun2 [56,57]), or for YAP/TAZ [58,59], have various defects, the pluripotent stem cells themselves invariably appear normal. We note that cortical actin tension has recently been reported to be required for maintenance of pluripotency [60,61].

    • Matrix remodeling controls a nuclear lamin A/C-emerin network that directs Wnt-regulated stem cell fate

      2022, Developmental Cell
      Citation Excerpt :

      Mmp14+/- mice (Swiss Black background), were obtained from the NIH (Holmbeck et al., 1999) and mated to generate Mmp14-/- and Mmp14+/+ mice. Lmnaf/f mice were provided by Yixian Zheng (Carnegie Institution for Science) and crossed into C57BL/6 mice for more than six generation (Kim and Zheng, 2013). Lmna+/- and Dermo1-Cre knock-in mice were obtained from the Jackson Laboratory (Maine, USA).

    View all citing articles on Scopus
    View full text