© 2006 American Thoracic Society DOI: 10.1165/rcmb.F310
Conditional Control of Gene Expression in the Respiratory EpitheliumA Cautionary NoteCincinnati Children's Hospital Medical Center, Cincinnati, Ohio Whether at the level of subparticle physics, biology, or clinical medicine, the "Uncertainty Principle" articulated by Heisenberg continues to hold sway. The methods used to identify a phenomenon influence the phenomenon. Sisson and coworkers, in "Expression of the reverse tetracycline-transactivator gene causes emphysema-like changes in mice" in this issue of the Journal (pp. 552–560), carefully document the influence of the reverse tetracycline transactivator (rtTA) protein on peripheral airspace size when expressed in a subset of epithelial cells in transgenic mice under control of the Scgb1a1 (CCSP) promoter (1). The work importantly identifies a potentially confounding variable in experiments in which genes are introduced into the genome of cells or animals. While the mechanisms underlying the airspace enlargement in this study remain unclear, the publication should be useful to the field in emphasizing the need for appropriate controls and careful interpretation of data.
Technical advances in gene transfer during the last decade have made the introduction of DNA into cells, in vivo and in vitro, a routine procedure. Gene addition, deletion, and mutation can readily be accomplished in many model systems using microinjection, viral vectors, transfection, and electroporation. Tissue-specific and conditional control of gene expression are now feasible in numerous model systems, including the transgenic mouse. Conditional and tissue-specific transgenic systems are a powerful tool useful for controlling gene expression at precise developmental times, thereby bypassing development toxicities accompanying gene deletion or addition. Tamoxifen-inducible, doxycycline/tetracycline-regulated, and interferon- The Tetracycline Conditional Control System The rtTA/tTA system developed by Gossen and Bujard (2) has been widely used for control of gene expression in vitro and in vivo (2–4). The system consists of the combination of two constructs, one expressing the tetracycline transactivator (a fusion protein consisting of bacterial genes tetR and VP16) (tTA) or rtTA under a tissue-specific promoter and the other expressing the gene of interest under control of an operator, (otet)7CMV. Tet activator is inactivated (or activated) by exposure to tetracycline (doxycycline), thereby expressing the gene of interest at cellular sites expressing the tTA/rtTA. Review of the literature provides little documentation of the potential toxicities of rtTA in this system that has now been used for more than a decade in vivo and in vitro (5). The study by Sisson and colleagues demonstrates progressive airspace enlargement in CCSP-rtTA mice, without evidence of inflammation and in the absence of doxycycline, indicating that toxicity is likely dependent on the rtTA construct. Such airspace enlargement has been previously reported in CCSP-rtTA mouse lines, indicating that abnormalities are not caused by integration site disruption or excessive copy numbers of the transgene (6). Both CCSP-rtTA and SP-C–rtTA (expressing rtTA in subsets of conducting and peripheral airway cells) mice have been widely used for study of gene function. In our experience, airspace remodeling has been observed in a number of CCSP-rtTA lines and has been influenced by strain and inbreeding, the severity varying among distinct mouse lines. To date, SP-C–rtTA lines generated in our laboratory do not develop airspace abnormalities. However, in one line, it has not been possible to generate homozygote SP-C–rtTA mice, indicating the possible toxicity of the SP-C–rtTA transgene. Thus, the design of experiments and appropriate control requires an awareness of the multiple potentially confounding variables that may influence the outcome. These include: (1) insertional mutagenesis (transgenes may insert into critical gene loci), (2) copy number (integrants often insert into the genome as concatamers with multiple copy numbers, (3) doxycycline exposure, (4) potential effect of increased copies of the promoter elements that may lead to promoter squelching, (5) potential toxicity of rtTA/tTA, (6) mouse strain differences that might modify or exacerbate toxicity, (7) male/female differences, (8) age, (9) toxicity of Cre-recombinase if it is controlled by rtTA/tTA to delete or add genes by recombination, and (10) potential immune responses to the proteins expressed by the transgenes. The regulatory elements used in the rtTA/tTA system are usually prokaryotic and thus not likely to alter the function of other host genes, at the doses used, the in mammalian cells (5). Tetracycline has been used for decades in humans and animals, and toxicity has been noted at high concentrations in developing bone. Analogs doxycycline or anhydrotetracycline have the properties of tetracycline but are active at doses 100-fold lower (5). In a systematic study, doxycycline was given to nontransgenic pregnant female mice, either in their drinking water or in their food, at doses cited in the literature for gene induction (7–10). While administration in drinking water at doses up to 5 mg/ml had no adverse effect, Moutier and coworkers observed that 2.5–10 mg/g of doxycycline per wet food during the second half of gestation resulted in a dose-dependent fetoplacental toxicity (11). Doxycycline is a pan-MMP (matrix metalloproteinase) inhibitor. MMPs play a functional role in alveolarization. Two daily injections of doxycycline at 20 mg/kg body weight for 10 d to neonatal rats resulted in alveolar simplification (12), a dose corresponding to 400 mg/kg given orally. We currently use doxycycline at 1 mg/ml in the drinking water or 625 µg/g in dry food pellets. Thus, an adult mouse consumes 1–3 mg of doxycycline per day (100 mg/kg). At these concentrations, we have not observed effects of doxycycline on pregnancy or postnatal alveolarization. Direct toxicity of rtTA or tTA has not been previously reported in vitro or in vivo experiments, but it is not clear that this possibility has been formally tested. The tetR-VP16 fusion protein was not immunogenic in mice (13), and if expressed during mouse development, is not likely to be recognized as non-self. Potential Toxicity of Cre-Recombinase Cre-recombinase is a powerful tool for the manipulation of vertebrate genomes. The enzyme cleaves DNA at a specific target sequences and can religate the newly exposed ends to the cleaved DNA at the second target sequence. Two components are required for the Cre-based recombination: (1) loxP, a 34-bp consensus sequence; and (2) Cre-recombinase, the 38-kD product of the bacteriophage P1 Cre gene (14, 15). Toxicity of Cre-recombinase has been demonstrated in mammalian cells (16–19). Postmeiotic spermatids of transgenic mice (20), mouse embryonic fibroblasts (MEFs) (16, 18, 21), and NIH 3T3 cells (18) were reported to be sensitive to the continuous expression of Cre-recombinase. Cre toxicity has been observed in kidney cell line 293 and osteosarcoma cell line U2OS (18), and in Drosophila cells (22) and also causes phenotypic aberrations in plants (23). Data from Pfeifer and colleagues revealed an accumulation of Cre-expressing cells in the G2/M phase of the cell cycle. The G2/M arrest was found in cells that do not contain engineered loxP sites, indicating that Cre recombinase is toxic (17). The toxicity of Cre depends upon strand cleavage (18) of genomic DNA at sites that share homology with the lox sites of the bacteriophage P1. Such pseudo-lox sites have been discovered in human, mouse, yeast, and Escherichia coli genomes (15, 24–26). "Illegitimate" (20) recombination results in DNA strand breaks and chromosome rearrangements (16, 18). Endogenous pseudo-lox sites share only a limited homology with loxP sites, and Cre-mediated recombination of pseudo-lox sites is less efficient than it is for bona fide loxP sites (26). Thus, limiting the intensity and duration of Cre expression attenuated toxicity in mammalian cells (16, 18). Transient expression of Cre should minimize the chances of recombination at the endogenous pseudo-lox sites or allow repair of DNA damage. Summary In summary, a number of transgenic systems have been developed that enable conditional control of gene expression in vivo and in vitro. These systems have been adapted for use in respiratory epithelial cells. As in all experiments, each variable that is introduced into the system may influence the outcome and interpretation of experiments. The article by Sisson and coworkers demonstrates a potential for toxicity of the rtTA transgene when expressed under control of the CCSP promoter, a finding that may be relevant to other rtTA or tTA constructs. The study emphasizes the need for controls of all genotypes (with and without doxycycline) and for careful interpretation of experiments that are dependent on the introduction of genes in vivo and in vitro. Footnotes Conflict of Interest Statement: Neither author has a financial relationship with a commercial entity that has an interest in the subject of this manuscript. References
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
–inducible constructs are now widely used for the study of gene function. Each system brings with it unique and potentially confounding variables that come with its use. 
