The catalytic activity of the ErbB-2 receptor tyrosine kinase is essential for embryonic development

Abstract

Activation of the epidermal growth factor receptor (EGFR) family is thought to play a critical role in both embryogenesis and oncogenesis. The diverse biological activities of the EGFR family are achieved through various ligand-receptor and receptor-receptor interactions. One receptor that has been found to play a central role in this signaling network is ErbB-2/Neu, and it is considered the preferred heterodimerization partner for other members of the EGFR family. To assess the importance of the catalytic activity of ErbB-2 in embryonic development, we have generated mice expressing a kinase-dead erbB-2 cDNA under the transcriptional control of the endogenous promoter. Here, we show that mice homozygous for the kinase-dead erbB-2 allele die at midgestation and display the same spectrum of embryonic defects seen in erbB-2 knockout mutants. These observations suggest that the catalytic activity of ErbB-2 is essential for normal embryonic development.

FIG. 1.

Targeted erbB-2 cDNA knock-in strategy by homologous recombination. For germ line expression, a targeting vector was constructed in which exon 1 was replaced by either (A) a wild-type erbB-2 cDNA or (B) a cDNA encoding the kinase-dead erbB-2 mutation, followed by a PGK-neomycin (Neo) cassette and targeted to the endogenous erbB-2 locus by homologous 5"- and 3"-flanking arms. Digestion of genomic DNA with HindIII (H) and subsequent Southern blot analysis (inset) with an external probe, as indicated, resulted in a 7.5-kb band for the endogenous allele, whereas the knock-in cDNA alleles introduced a HindIII site and resulted in a 4.0-kb band.

FIG. 2.

Embryonic lethality at midgestation in kinase-dead mutants. Mendelian ratios from the progeny of heterozygous matings were determined and compared with the frequency of genotypes observed. (A) No homozygous erbB-2^KD/KD mutant animals were observed at weaning age (3 weeks old). (B) Observations at E11.5 to E13.5 revealed the expected number of mutant embryos; however, all erbB-2^KD/KD (∗∗) embryos were being resorbed, and no heartbeat was detected. (C) At E10.5, all the embryos appeared healthy and were present in proportion with Mendelian frequencies. Solid bars, observed; shaded bars, expected.

FIG. 3.

Defects in heart development in E10.5 mutant embryos. Parasagittal sections of (A) erbB-2^KI/KI, (B) erbB-2^wt/KD, and (C) erbB-2^KD/KD embryos at E10.5 were stained with hematoxylin and eosin. Although heartbeats were detected at the time of dissection, histological examinations of the hearts revealed a lack of trabeculae in the ventricles of erbB-2^KD/KD mutants but were present in heterozygous littermates and in age-matched erbB-2^KI/KI knock-in embryos. t, trabeculae; v, ventricle; ec, endocardial cushion; a, atrium.

FIG. 4.

Lack of sympathetic chain ganglia in kinase-dead mutants. Day E10 to E11 embryos were subjected to whole-mount in situ hybridization analysis using an antisense Phox2a riboprobe. Normal sympathetic chain ganglia development was present in (A and B) erbB-2^wt/wt embryos and (C and D) erbB-2^wt/KD heterozygous embryos, whereas (E and F) erbB-2^KD/KD homozygous mutants lacked proper development or they were delayed in (G and H) erbB-2^KI/KI embryos. The white arrowheads highlight the developing sympathetic chain ganglia.

FIG. 5.

Detection of ErbB-2 expression in E10.5 embryos. Expression of ErbB-2 generated from the cDNA insert in erbB-2^KIKI and erbB-2^KD/KD embryos was determined. For (A) immunoblot analyses, total protein isolated from E10.5 embryos was subjected to SDS-PAGE and incubated with an anti-ErbB-2 antibody. Detection of Grb-2 (lower panel) protein was used to control for equal protein lysate quantification. (B) RNase protection assays with an antisense erbB-2 riboprobe were used to detect erbB-2 transcripts in total RNA isolated from erbB-2^wt/wt (lane 1), erbB-2^KD/KD (lanes 3 and 4), and erbB-2^KI/KI (lanes 6 and 7) embryos. A mouse PGK riboprobe (lower panel) was used as an internal control for equal sample loading in each lane.