Murine SEC24D can substitute functionally for SEC24C during embryonic development

Abstract

The COPII component SEC24 mediates the recruitment of transmembrane cargos or cargo adaptors into newly forming COPII vesicles on the ER membrane. Mammalian genomes encode four Sec24 paralogs (Sec24a-d), with two subfamilies based on sequence homology (SEC24A/B and C/D), though little is known about their comparative functions and cargo-specificities. Complete deficiency for Sec24d results in very early embryonic lethality in mice (before the 8 cell stage), with later embryonic lethality (E7.5) observed in Sec24c null mice. To test the potential overlap in function between SEC24C/D, we employed dual recombinase mediated cassette exchange to generate a Sec24c^c-d allele, in which the C-terminal 90% of SEC24C has been replaced by SEC24D coding sequence. In contrast to the embryonic lethality at E7.5 of SEC24C-deficiency, Sec24c^c-d/c-d pups survive to term, though dying shortly after birth. Sec24c^c-d/c-d pups are smaller in size, but exhibit no other obvious developmental abnormality by pathologic evaluation. These results suggest that tissue-specific and/or stage-specific expression of the Sec24c/d genes rather than differences in cargo export function explain the early embryonic requirements for SEC24C and SEC24D.

Figure 1

Generation of the chimeric Sec24c^c-d allele. (A) Schematic representation of dRMCE to generate the Sec24c^c-d allele. The replacement vector pUC19-Sec24c-d contains the Sec24c intron 2 splice acceptor (yellow), the Sec24d coding sequence beginning with G¹²⁰ (gray), and a stop codon followed by a poly A signal sequence (red). Arrows represent primers used for genotyping, long-range PCR, and RT-PCR (for sequences, see Table S2). (B) The SEC24C-D fusion protein encoded by the dRMCE generated Sec24c^c-d allele, which contains the first 57 amino acids of SEC24C followed by the SEC24D sequence corresponding to the remaining ~ 95% of the SEC24C protein. Val58/41 (Val58 in SEC24C, Val41 in SEC24D) indicates the junction point for this chimeric protein. SEC24C^Met1-Val58 and SEC24D^Met1-Val41 have 34% protein sequence identity. (C) PCR results for dRMCE subclone 12275. Primer combinations are indicated at the top of each lane. Correct targeting was observed for the 5′ recombination (5′ rec) site (lanes 1 and 2) and the 3′ recombination (3′ rec) site (lanes 3 and 4). Additionally, the presence of the Sec24d cDNA (“any ins.”) (lanes 5,6), and the loxP and FRT sites (103 bp and 106 bp products in lanes 7,8, respectively) was confirmed. The signal in lane 9 is due to the Sec24c⁺ allele, confirming that ESC clone 12275 is heterozygous for the Sec24c^c-d allele. Clone 12275 does not carry any random insertions of pCAGGS-iCre (lane 10) or pCAGGS-Flpo (lane 11). (D) A genotyping PCR assay on mouse genomic DNA from tail clip to distinguish between the wild type and Sec24c^c-d allele using primers E, F, and G. (E) Long range PCR confirms correct targeting. Primers GF4 + U were used to amplify the 5′ arm resulting in a 7.6 kb product, and primers F and GR4 were used to amplify the 3′ arm to yield a 5.8 kb product. Primers were located outside the homology arms (GF4 and GR4) and within the Sec24d cDNA (F and U). Neither set of primers yields a band from the Sec24c^wt allele.

Figure 2

Phenotypic analysis of Sec24c^+/c-d intercross progeny at P0. (A) Side views of P0 pups from Sec24c^+/c-d intercross taken shortly after birth. Sec24c^c-d/c-d pups are much paler and smaller than their littermate controls, exhibited little spontaneous movement, and died within minutes of birth. (B) Body weight measurements at P0, 4 weeks, and 6 weeks of age. P0 normalized to average weight of controls (Sec24c^+/+ and Sec24c^+/c-d) within the same litter set at 100%. Four week and 6 week data normalized to the average weight of wild type mice for each sex; data for males and females are pooled; no significant differences in weight were observed between Sec24c^+/+ and Sec24c^+/c-d mice when broken down by sex. (C) Crown to rump length measurements at P0, normalized to average length of controls (Sec24c^+/+ and Sec24c^+/c-d) within the same litter. ** indicates p < 0.0001; all other comparisons not significant, with p > 0.05. Error bars represent the standard deviation.

Figure 3

Histological assessment of Sec24c^+/c-d intercross progeny at P0. (A) H&E stained longitudinal sections of P0 pups from Sec24c^+/c-d intercross taken shortly after birth. A total of 16 animals were analyzed by H&E at this time point. Scale bars = 5 mm. (B) Low and higher magnification views of H&E stained sections through the lung of P0 mice collected shortly after birth. Analysis did not identify gross alterations in the morphology of the respiratory tree, although alveoli from Sec24c^c-d/c-d pups were often uninflated and lined by columnar epithelium compared to the flattened epithelial lining wild type and Sec24c^+/c-d alveoli. Lungs were fixed and sectioned in the context of the whole pup. Scale bars = 500 or 250 microns, as indicated. (C) Longitudinal sections through the vertebral column of Sec24c^c-d/c-d embryos and wild type or Sec24c^+/c-d controls illustrating regions of developing cartilage in forming vertebral bodies by Alcian blue (n = 3 for each genotype, Scale bars = 50 μm) and (D) alizarin red stains (n = 2 for each genotype). (E) Immunohistochemistry for SERT in Sec24c^c-d/c-d and wild type control brain tissues (n = 3 per genotype). Scale bars = 20 μm.

Figure 4

Phenotypic assessment of Sec24c^+/c-d intercross embryos. (A) Side views of embryos from Sec24c^+/c-d intercross at E17.5 (N = 8). Eye pigment variation is normal in inbred mice and did not track with the mouse genotypes. (B) Bouin’s solution fixed E18.5 embryos sectioned at the midline (left) and stained with H&E (center). H&E of 4% PFA fixed lungs from E18.5 embryos are shown at right. A total of 13 embryos were analyzed. (C) Whole mount and H&E images of E15.5 embryos from Sec24c^+/c-d intercross (N = 15 analyzed). (D) Crown to rump length measurements at E17.5–18.5 and E15.5, normalized to average length of controls (Sec24c^+/+ and Sec24c^+/c-d) within the same litter. ** indicates p < 0.0001; * indicates p < 0.002; ns indicates p > 0.05. Error bars represent the standard deviation.

Figure 5

Analysis of Sec24c^c-d allele splicing. A three-primer RT-PCR reaction using qF1, qR1 and qR3 demonstrates the presence of Sec24c^c-d allele splice variants. Primers qF1 and qR3 (see schematic at the right) detected mRNA transcripts from both the wild type allele (exons 2–3-4–5) and the transcript skipping exon 3 (2-4-5) present in mice carrying the Sec24c^c-d allele, resulting from splicing around the dRMCE insertion (lanes 3–6), as well as mice carrying the Sec24c^- allele, in which exon 3 was also excised (lanes 7–8). Primers qF1 and qR1 detect the mRNA transcript of the Sec24d cDNA insert in Sec24c^+/c-d and Sec24c^c-d/c-d mice (lanes 3–6), and this transcript is absent in wild type littermates (lanes 1–2), or Sec24c^+/− mice (lanes 7–8).