ErbB4 signaling in the mammary gland is required for lobuloalveolar development and Stat5 activation during lactation

Abstract

Signaling by members of the epidermal growth factor receptor family plays an important role in breast development and breast cancer. Earlier work suggested that one of these receptors, ErbB4, is coupled to unique responses in this tissue. To determine the function of ErbB4 signaling in the normal mouse mammary gland, we inactivated ErbB4 signaling by expressing a COOH terminally deleted dominant-negative allele of ErbB4 (ErbB4DeltaIC) as a transgene in the mammary gland. Despite the expression of ErbB4DeltaIC from puberty through later stages of mammary development, an ErbB4DeltaIC-specific phenotype was not observed until mid-lactation. At 12-d postpartum, lobuloalveoli expressing ErbB4DeltaIC protein were condensed and lacked normal lumenal lactation products. In these lobuloalveoli, beta-casein mRNA, detected by in situ hybridization, was normal. However, whey acidic protein mRNA was reduced, and alpha-lactalbumin mRNA was undetectable. Stat5 expression was detected by immunohistochemistry in ErbB4DeltaIC-expressing tissue. However, Stat5 was not phosphorylated at Y694 and was, therefore, probably inactive. When expressed transiently in 293T cells, ErbB4 induced phosphorylation of Stat5. This phosphorylation required an intact Stat5 SH2 domain. In summary, our results demonstrate that ErbB4 signaling is necessary for mammary terminal differentiation and Stat5 activation at mid-lactation.

Figure 1

Expression analysis of ErbB4ΔIC RNA in developmentally staged mammary glands. RNA (20 μg) isolated from the number 4 inguinal mammary gland was hybridized with ³²P-labeled antisense riboprobe corresponding to the COOH-terminal 285-bp of ErbB4ΔIC, including sequences encoding the tandem Flag epitope tags, and subjected to RNase protection analysis (upper panel). Negative controls included 20 μg each of tRNA and RNA from a nontransgenic sibling at 1-d postpartum (control). A probe that hybridizes to β-actin sequences was included to control for RNA integrity and to confirm that equivalent amounts of RNA were added to each reaction (lower panel). Protected fragments corresponding to ErbB4ΔIC and β-actin are indicated.

Figure 2

Immunohistochemical detection of ErbB4ΔIC protein in the mammary gland at 12-d postpartum. Paraffin-embedded section from a 12-d-postpartum nontransgenic sibling control stained with hematoxylin/eosin (A). Sequential sections (B–D) from a 12-d-postpartum mammary gland from an ErbB4ΔIC-expressing mouse were stained with hematoxylin/eosin (B), or stained by immunohistochemistry with rabbit IgG control antibody (C) or a rabbit anti-Flag antibody (D). Expanded secretory lobuloalveoli are indicated by arrows, and condensed atypical lobuloalveoli are indicated by asterisks. Bar in C, 100 μm.

Figure 3

In situ hybridization analysis of milk protein gene expression in ErbB4ΔIC-expressing mammary glands at 12-d postpartum. Sequential sections from a 12-d-postpartum mammary gland from an ErbB4ΔIC-expressing mouse were stained with hematoxylin/eosin (A) or stained by immunohistochemistry with anti-Flag antibody (B). Additional sequential sections were analyzed by in situ hybridization with β-casein sense (C) or antisense (D) riboprobes, WAP sense (E) or antisense (F) riboprobes, or α-lactalbumin sense (G) or antisense (H) riboprobes. Expanded secretory lobuloalveoli are indicated by arrows and condensed atypical lobuloalveoli are indicated by asterisks. Bar in G, 100 μm.

Figure 4

Immunohistochem- ical detection of Stat5 protein in ErbB4ΔIC-expressing mammary glands at 12-d postpartum. High magnification photomicrographs of expanded lobuloalveoli lacking detectable ErbB4ΔIC expression (A–C) or a different region of the same section containing condensed lobuloalveoli expressing high levels of ErbB4ΔIC protein (D–F). Sections were stained with hematoxylin/eosin (A and D), or stained by immunohistochemistry with rabbit serum negative control (B and E) or rabbit anti-Stat5 (C and F). Bar in D, 30 μm.

Figure 6

Coexpression of ErbB4 and Stat5a. 293T cells were transfected with various combinations of ErbB4 and Stat5 expression vectors and lysates of transfected cells were prepared 48 h posttransfection, as described in Materials and Methods. Immunoprecipitations were performed using ErbB4-specific (A–D) or Stat5-specific (E–H) antibodies. Immunoprecipitates were resolved by SDS-PAGE and the 7.5% acrylamide resolving gel was transferred to nitrocellulose. Western blot analysis was performed using antibodies to detect tyrosine phosphorylated proteins (A and E), ErbB4 (B and F), phospho-Stat5 (C and G), or Stat5 (D and H). Stat5 mutants were Stat5/R618V, a point mutation which ablates SH2 function, and Stat5/Y694F, which eliminates the regulatory Y694. Transfections were with empty vectors (lanes 1 and 7), ErbB4 alone (lanes 2 and 8), Stat5a alone (lanes 3 and 9), ErbB4 + Stat5a (lanes 4 and 10), ErbB4 + Stat5/R618V (lanes 5 and 11), and ErbB4 + Stat5/Y694F (lanes 6 and 12). Closed circles in A and E indicate the position of ErbB4 at ∼190 kD; open circles in D, E, and H indicate the position of Stat5a at ∼95 kD.

Figure 5

Immunohistochem- ical detection of Stat5 phosphorylated at Y694 in ErbB4ΔIC-expressing mammary glands at 12-d postpartum. High magnification representations of expanded lobuloalveoli lacking detectable ErbB4ΔIC expression (A–C) or a different region of the same section containing condensed lobuloalveoli expressing high levels of ErbB4ΔIC protein (D–F). Sections were stained with hematoxylin/eosin (A and D), or stained by immunohistochemistry with antiphospho-Stat5 antibody preadsorbed with peptide immunogen and counterstained with methyl green (B and E), or phospho-Stat5 antibody without counterstain (C and F). Bar in D, 30 μm.