Analysis of in vivo functions of Memo in embryonic and mammary gland development

Studies from our lab recently led to the discovery of Memo (mediator of ErbB2-driven cell motility), a novel 297 amino acid protein shown to be required for ErbB2- and other receptor tyrosine kinase-driven cell motility in breast tumor cells. Inhibition of Memo expression had consequences on the microtubule network which could not grow towards the periphery of the cells upon heregulin (a ligand activating ErbB2/ErbB3 and ErbB2/ErB4 heterodimers) stimulation. It also had consequences on the actin cytoskeleton, since more actin stress fibers were seen.
To explore the biological function of Memo, and in order to check if Memo also plays a role in in vivo cell migration events, we generated animal models deficient for Memo. We found that Memo is expressed ubiquitously in adult organs as well as in organs of the developing embryo. Unexpectedly, we did not see any defect in migration in vivo, despite the presence of a lot of migrating events during development like gastrulation or migration of the neural crest derivatives or of the somitomeres. Instead, we found that Memo seems to play a role in vascular integrity, as demonstrated by the presence of hemorrhages and the dilated small vessels in the Memo deficient embryos. This leads to the death of Memo deficient embryos after 13 days of embryonic development.
To study the in vivo role of Memo in the lactating mammary gland, we generated mice deficient for Memo in luminal alveolar epithelial cells (the cells that produce and secrete milk during lactation). We measured a decrease in the weight of pups from Memo deficient mothers, indicating that they were unable to correctly nurse them. The weight of the mammary gland itself was smaller in the Memo deficient females compared to control females. By histological analyis we saw the abnormal presence of shed cells in the lumen of Memo deficient glands in the first days of lactation. We saw a progressive loss of alveoli (formed by epithelium) which were replaced by adipocytes. Increased apoptosis (controlled
cell death) was measured in the Memo deficient glands. Consistent with this apoptosis seen at the histological level, we could see an increase in the levels of pro-apoptotic P-Stat3 and Bax at protein level. We also could see improper localization of the adherens junction proteins E-cadherin and ß-catenin in the Memo deficient mammary glands. We therefore propose that in the mammary gland Memo plays a role in epithelial cell-cell adhesion, and that if this role is not properly achieved, the cells undergo apoptosis and are shed in the lumen of alveoli which progressively disappear. This leads to improper feeding of the pups.