To the Editor: The tissue-/cell-specific recombinase system, the Cre/loxP system, has become a very useful tool to conditionally inactivate or activate gene function [1–6]. This system is based on the ability of Cre to catalyse the excision of DNA flanked by loxP site recognition sequences. A number of studies on cell lineage in pancreatic cells have been performed by intercrossing cell-specific Cre-expressing transgenic mice and ROSAβgeo26 (ROSA26) mice, which carry the gene for bacterial beta-galactosidase (lacZ) [7]. In the double transgenic offspring, lacZ is expressed where Cre was expressed at a high enough level to excise the floxed stop codon. Commonly, X-gal staining is used to detect beta-galactosidase activity: beta-galactosidase catalyses the hydrolysis of X-gal to an indolyl alcohol, which then oxidises to form an intense blue indigo stain. While specific induction of lacZ by precise control of Cre, as well as leakage of beta-galactosidase by spontaneous Cre excision in double transgenic mouse, have been well studied, little work has dealt with the contribution of endogenous beta-galactosidase activity to false-positive results. Heart, liver, spleen, kidney, brain and skeletal muscle from mouse, rat and rabbit [8], and cultured human foetal pancreas [9], have been studied for endogenous betagalactosidase activity [8]. In cultured human foetal pancreas, the mRNA expression and enzyme activity of beta-galactosidase are developmentally regulated, peaking at 24 weeks of gestation and declining to low levels in the adult [9]. Endogenous beta-galactosidase activity in these organs was rarely detected following incubation at a pH greater than 7.5. In contrast, bacterial beta-galactosidase activity in the ROSAβgeo26 was strongly detected at pH 8.0–9.0 [8]. In the mouse pancreas, we evaluated endogenous beta-galactosidase expression and studied factors that may contribute to false-positive results. Firstly, to determine expression of the endogenous betagalactosidase mRNA in various tissues, we performed RTPCR on cDNAs from mouse islets, liver, kidney and spleen as well as from pancreatic beta cells from mouse insulin promoter I (MIP)–green fluorescent protein (GFP) mice [10], purified with FACS by their GFP expression (Fig. 1). The sizes of the PCR products for beta-galactosidase and the internal control TATA-binding protein were 330 and 190 bp, respectively. Pancreatic beta cells and islets as well as other organs expressed endogenous beta-galactosidase (Glb1) mRNA (Fig. 1). Since it is important to determine endogenous betagalactosidase enzyme activity, we performed X-gal staining on sections of pancreas and kidney from normal adult C57BL/6 mice; endogenous beta-galactosidase is known to be expressed in the kidney. From mice cardiac-perfused
Read full abstract