G-CSF is the major regulator of neutrophil development and controls the proliferation, differentiation and survival of myeloid progenitors in both steady state and “emergency” granulopoiesis. More recently however, novel activities of the G-CSF/G-CSF receptor (CSF3R) axis have been suggested, especially after severe forms of tissue damage. For instance, upon myocardial infarction, G-CSF was shown to protect cardiomyocytes from apoptosis, reducing the extent of myocardial damage. Likewise, in ischemic stroke models, G-CSF treatment reduced infarct volumes by promoting neuronal survival and differentiation of neural stem cells in the brain. Finally, G-CSF has been suggested to promote hepatic regeneration after severe liver damage by directly stimulating the proliferation of oval cells. To study whether and to what extent CSF3R expression in damaged tissues reflects reactivation of a normal development program, we generated a csf3r-Cre knock-in model. To this end, the csf3r locus was targeted with a construct containing the cre gene and a puromycin selection cassette flanked by a 2.7 kb fragment of the csf3r 5′UTR and a 4 kb fragment directly downstream of the initiation codon of csf3r. Correct insertion of the cre gene was confirmed by Southern blotting and sequence analysis and subsequent crossing with FLP deleter mice resulted in successful germ line transmission and deletion of the puromycin cassette. These mice were crossed with the R26R LacZ reporter mouse and β-galactosidase (β-gal) activity was examined by FACS and histochemical analysis. At E9.5, no β-gal activity was detected, indicating that csf3r is not expressed at that stage of development. In contrast, in E12.5 embryos, β-gal staining was observed in fetal liver, heart, kidney, brain and intestine. Analysis of adult tissues showed that cardiomyocytes were β-gal positive in a mosaic pattern. A similar mosaicism was observed in liver, kidney and the intestines. In liver both hepatocytes and vascular endothelial cells stained positive, whereas in the kidney β-gal staining was detected in vascular endothelium, glomeruli and afferent and efferent vessels. In ileum and colon, β-gal activity was mainly seen in the stem and progenitor cell-containing crypts and the developing progeny migrating upwards along the villus. In bone marrow, high levels (>40%) of β-gal activity were detected in Gr1+ and CD11b+ myeloid cells, CD19+ (B) and CD3ε+ (T) lymphoid cells, NK cells and CD11c+ dendritic cells, indicating that G-CSFR was expressed in a common progenitor of these cell types. In agreement with this, β-gal activity was also detected in the primitive LSK population. In conclusion, our data indicate that CSF3R is expressed in both hematopoietic and multiple non-hematopoietic organs from embryonic development stage E12.5 onward. This may have important ramifications for a more wide-spread application of G-CSF in tissue regeneration.

Author notes

Disclosure: No relevant conflicts of interest to declare.