TEMPORAL AND SPATIAL PATTERNS OF TRANSGENE EXPRESSION IN AGING ADULT MICE PROVIDE INSIGHTS ABOUT THE ORIGINS, ORGANIZATION, AND DIFFERENTIATION OF THE INTESTINAL EPITHELIUM

We have used liver fatty acid-binding protein/human growth hormone (L-FABP/hGH) fusion genes to explore the temporal and spatial differentiation of intestinal epithelial cells in 1- to 12-month-old transgenic mice. The intact, endogenous L-FABP gene (Fabpl) was not expressed in the colon at any time. Young adult transgenic mice containing nucleotides -596 to +21 of the rat L-FABP gene linked to the hGH gene (minus its 5' nontranscribed domain) demonstrated inappropriate expression of hGH in enterocytes and many enteroendocrine cells of most proximal and mid-colonic crypts (glands). Rare patches of hGH-negative crypts were present. With increasing age, a wave of "extinction" of L-FABP(-596 to +21)/hGH expression occurred, first in the distal colon and then in successively more proximal regions, leaving by 10 months of age only rare hGH-positive multicrypt patches. At no time during this progressive silencing of transgene expression were crypts observed that contained a mixture of hGH-positive and -negative cells at a particular cell stratum. Young (5-7 weeks) mice containing a L-FABP(-4000 to +21)/hGH transgene also demonstrated inappropriate expression of the transgene in most proximal colonic crypts. However, the additional 3.3 kilobases of upstream sequence resulted in much more rapid extinction of reporter expression, leaving by 5 months of age only scattered single crypts with detectable levels of hGH. This age-related extinction of L-FABP/hGH expression did not involve enterocytes and enteroendocrine cells in the (proximal) small intestine. These results indicate that cis-acting elements outside of nucleotides -4000 to +21 are necessary to fully modulate suppression of colonic L-FABP expression. They also define fundamental changes in colonic epithelial cell populations during adult life. Our data suggest that (i) a single stem cell gives rise to all cells that populate a given colonic crypt, (ii) stem cells represented in several adjacent crypts may be derived from a common progenitor, and (iii) such a progenitor cell may repopulate colonic crypts with stem cells during adult life. Since each colonic crypt contains the amplified descendants of its stem cell, transgenes may be powerful tools for characterizing the spatial and biological features of gut stem cells and their progenitors during life.