Generation of an adult smooth muscle cell-targeted Cre recombinase mouse model.

To the Editor: The smooth muscle cell (SMC)–targeted Cre recombination mice are critical tools for in vivo analysis of gene function in the vasculature and for establishing animal models for cardiovascular diseases. Therefore, there is a continuing effort to generate SMCtargeted Cre recombinase mice for in vivo loss-of-gene function studies. Currently, several genetically engineered mice express the Cre-recombinase under the control of SMC-specific promoters such as SM22 (also known as transgelin, a 22-kDa protein that is abundantly and exclusively expressed in SMCs of adult animals) promoters and smooth muscle myosin heavy chain promoters.1–6 However, there are potential limitations in their uses for knockout studies; some of them show relatively low excision efficiency and potential embryonic lethality, which prevent subsequent in vivo analyses in adult SMCs. In our effort to generate an SMC-targeted Cre recombination mouse line that effectively excises loxP-flanked target gene, we obtained a valuable Cre recombinase mouse (SM22 -CreKI) that unexpectedly does not express Cre recombinase in embryonic SMCs and cardiac myocytes but highly expresses the Cre in adult SMCs and cardiac myocytes. This SM22 -CreKI mouse line was generated by knocking in the Cre-recombinase coding sequence into the endogenous SM22 gene locus via homologous recombination of embryonic stem cells (supplemental Figure S1, available online at http://atvb.ahajournals.org). Consistent with previous reports of SM22 knockout mice, our SM22 -CreKI heterozygous and homozygous mice were fertile and appeared phenotypically normal. To determine the temporospatial patterns of Cre-mediated recombination in SM22 -CreKI mice, we crossed the homozygous SM22 CreKI mice with the reporter mice Gt(Rosa)26Sor (The Jackson Laboratory, Bar Harbor, Me; cat. 003309). As shown in the Figure, A, we found that the Cre expression in a heterozygous mouse is sufficient to induce homologous recombination at loxP sites and thus to remove the loxP-flanked STOP signal between the lacZ gene and the Gt(Rosa)26Sor promoter, which leads to -galactosidase ( -gal) activities. We observed -gal–positive staining in almost all adult SMCs that comprise the medial layer of all large and small arteries and veins, including the arterial circle of Willis, aorta, femoral arteries and veins, the pulmonary artery, small arteries in skeletal muscles, and coronary arteries (Figure, A). The Cre-mediated recombination in vascular SMCs occurs in all of the arteries and veins we examined (Figure, A; online supplement S2). Consistent with the expression pattern of endogenous SM22 , -gal–positive staining in the visceral SMCs, including the bladder and gastrointestinal tract in adult mice is observed at high efficiency (Figure, A; online supplement S2). In addition, -gal activity is also found in cardiomyocytes (Figure, A). However, we found no -gal–positive staining in other tissues such as brain, liver, and skeletal muscle cells. Distinct from the endogenous SM22 expression, the SM22 CreKI mice failed to exhibit specific -gal activities in SMCs at all embryonic stages examined, ranging from embryonic day 9 to embryonic day 16.5 (online supplement S3). We detected -gal activities in SMCs and the cardiac myocytes in newborn pups right after birth (day 1). We also bred SM22 -CreKI mice with another reporter mouse model, Gt(Rosa)26Sor–green fluorescent protein (GFP; The Jackson Laboratory, Cat#: 004077) that contains an enhanced GFP (eGFP) gene