Studies on engineered autocrine systems: requirements for ligand release from cells producing an artificial growth...

Transplantation of artificial ligand-secreting cells is envisioned as a promising technology in tissue engineering. To achieve practical control of these systems, sufficiently high levels of ligand must be produced to provide adequate receptor binding and signal generation, but ligand spread into adjacent tissue regions must also be controlled. Mathematical models predict that the relative amount of ligand found either in the extracellular medium or bound to cell receptors is governed by ligand and receptor synthesis rates, receptor turnover rate, cell density, and exogenous blocking antibody concentration. To experimentally elucidate the relative contribution of these parameters, we constructed an artificial autocrine system in which all major parameters were under direct experimental control. A synthetic gene for a secretory form of human epidermal growth factor (EGF) was constructed consisting of the mature protein product fused to a foreign signal seqeunce. This was inserted into a vector behind an inducible promoter. Cells lacking endogenous receptors for EGF (EGFR) were first transfected with the gene for the human EGFR and then with the gene for EGF. Cells were selected that expressed high levels of both receptor and ligand. A tetracycline-sensitive promoter system for the artificial EGF gene allowed us to experimentally vary EGF secretion rates 20- to 200-fold. Using this system, we found that no significant amount of EGF was found in the culture medium unless the production rate of the ligand exceeded that of the receptor. Alternately, the use of EGFR-blocking antibodies allowed ligand escape into the medium. Even in the presence of high concentration of antagonistic antireceptor antibodies, however, cells were still able to consume EGF produced in an autocrine fashion. Induction of high levels of EGF production resulted in an almost 90% reduction in total receptor mass through down-regulation, but cells continued to rapidly bind, internalize, and degrade EGF. Our data suggest that cells have an unexpectedly high capacity to both bind and utilize growth factors produced in an autocrine fashion. In addition, interrupting autocrine loops may be more difficult than originally envisioned. Our artificial autocrine system should prove useful in understanding both how these systems are normally regulated and how they can be manipulated for purposes of tissue engineering.