The purpose of these experiments was to study the in vivo growth characteristics of nonmalignant and malignant epithelial cell cultures maintained in vitro for various lengths of time. All cell cultures tested originated from normal or carcinogen-exposed tracheal epithelium of Fischer 344 rats. The cultures were inoculated into isolated tracheas, which were then grafted to isogenic recipients. At least 10⁴ epithelial cells were required to reestablish a complete epithelial lining in denuded tracheal grafts, whether normal or tumorigenic cell cultures were used. Inoculation of denuded tracheas with 1-week-old primary cultures of normal tracheal epithelium resulted in the reestablishment of a nearly normal mucociliary tracheal lining within 1 to 2 weeks. Injection of primary epithelial cultures older than 1 week failed to reestablish an epithelial lining; instead, the tracheal lumen became obliterated with connective tissue. Inoculation with preneoplastic epithelial cell lines resulted in the establishment of a well-organized, keratinizing squamous epithelium, which remained stable for at least 6 weeks if the recipient was immunosuppressed, but which was rejected at 4 weeks in immunocompetent hosts. Inoculation with a neoplastic cell line resulted in the establishment of a well-organized squamous epithelium for 3 to 4 weeks and the development of disorganization, exophytic growth, and invasion at 4 to 6 weeks. This occurred only in immunosuppressed recipients. In immunocompetent hosts the epithelial lining was completely rejected within 2 weeks. When cells from a highly malignant squamous cell carcinoma line were inoculated into tracheas with an intact epithelial lining, only a few isolated nests of malignant cells were observed at 6 weeks. Repopulation of denuded tracheas with cells from the same tumor line resulted in the establishment of an atypical squamous lining at 1 week. At 2 weeks invasion was widespread, resulting in destruction of the tracheal grafts shortly thereafter.

The epithelial morphologies observed in these studies are highly reminiscent of various preneoplastic and neoplastic “lesions” induced in tracheas of rats by direct action of chemical carcinogens. Our experiments show that this “in vivo culture system” is well suited to the study of growth and differentiation characteristics of carcinogen-altered or preneoplastic epithelial cell populations. This is not easily accomplished otherwise, since nonneoplastic cells injected s.c. or i.m. are often difficult to retrieve. We believe that this new in vivo culture approach will effectively complement the existing in vitro culture systems used in studying the process of epithelial carcinogenesis.