The Tetracycline Regulatory System (Tet System ) was developed as a strategy to regulate the transcription of genes in eukaryotic cells utilizing prokaryotic transcriptional regulatory proteins [29, 30]. There are two variations of this system—one activates transgene expression in the presence of a tetracycline such as doxycycline (Tet-on), while the other system shuts off transgene expression upon doxycycline addition (Tet-off). In both variations, two different transgenes are generated. The first transgene uses a tissue-specific promoter to drive the expression of a tetracycline transactivator (tTA or rtTA). The second transgene contains a tetracycline response element (Tet-O) adjacent to a target gene of interest. The tTA or rtTA protein binds to the Tet-O promoter regulating gene transcription. The presence of doxycycline, prevents binding of the tTA protein to the Tet-O element turning off gene expression (Tet-off) or promotes the binding of the rtTA protein to the Tet-O element turning gene expression on (Tet-on). The Tet system facilitates monitoring of transgene expression at the transcription level in specific tissues within the mouse.

The Tamoxifen System also has been employed to conditionally regulate gene activation posttranscriptionally. MYC fused with the estradiol receptor exhibited conditional oncogene activation [31]. A mutant version of the estradiol receptor, which binds tamoxifen, is utilized to prevent endogenous estradiol from activating gene function [32]. Upon addition of tamoxifen, MYC is active, and withdrawal leads to an inactive product. Subsequently, this strategy was shown to be useful in the generation of conditional transgenic mice [33, 34].

The Tet-system can be combined with the RCAS-TVA system [35, 36]. In this approach a tissue-specific promoter is used to drive the expression of the avian retroviral receptor (TVA) in transgenic mice. The cells of these mice also contain a Tet-O regulated transgene, but lack the rtTA protein. Avian retroviral vectors (RCAS) are used to deliver the rtTA transactivator to cells that express the TVA. The successfully infected cells now contain a transgene whose expression can be regulated by doxycycline.

The Tet system has been used to conditionally overexpress receptors including an oncogenic form of HER2 containing an activating point mutation in its transmembrane domain [52, 53]. Expression was directed to the breast by utilizing the mouse mammary tumor virus (MMTV) promoter to drive the expression of the rtTA protein. Within 4 days of HER2 activation by doxycycline administration, the mice developed hyperplastic abnormalities. Six weeks following oncogene activation, all of the mice developed multiple invasive mammary carcinomas. The tumors were solid invasive carcinomas that often metastasized to the lung. Following 48 h of HER2 inactivation through doxycycline withdrawal, the cancer cells exhibited proliferation arrest and increased apoptosis. The primary carcinomas rapidly and completely regressed in over 90 % of the mice with a mean regression time of 17 days. Within 30 days, the pulmonary metastases had also completely and rapidly regressed. Thus, tumorigenesis appears to be reversible.

However, a majority of the cancers that had a complete regression upon HER2 repression eventually reoccurred. When the primary cancers and metastases were transplanted into syngeneic hosts, they completely regressed only 55–70 % of the time. The relapse tumors all uniformly lacked both endogeneous and transgene protein expression indicating that the cancers had all become HER2 independent [53]. Subsequently, Snail, a transcriptional repressor was found to be activated in relapsed cancers [52]. Therefore, although oncogene inactivation can cause cancer regression, some transgenic tumors are capable of becoming independent of their initiating oncogenic event.

The Tet-On system has been used to generate a conditional model of mutant H-RAS induced melanomas [54, 55]. The tyrosinase gene promoter (Tyr) was used to conditionally overexpress an H-RAS bearing an activating point mutation (V12G) in an Ink4a-deficient background. Approximately 25 % of the mice developed melanomas within 60 days of H-RAS activation. The melanomas were invasive, highly vascular, and notably amelanotic. The tumors exhibited expression of tyrosinase-related-protein-1 (TRP-1), an early melanocyte-specific maker. Within 48 h of H-RAS inactivation through doxycycline withdrawal, the cancers decreased their proliferation and exhibited robust apoptosis. Within 14 days of H-RAS inactivation, the cancers had completely regressed with only microscopically detectable scattered cancer foci. Notably, melanomas transplanted into SCID hosts also regressed upon inactivation of mutant H-RAS. Approximately 30 % of the melanomas resumed growth, even in the absence of H-RAS, but relapsed tumors failed to express TRP-1, suggesting that these tumors were phenotypically different from the primary cancers [55]. Additionally studies have shown that EGFR signaling is required for maintenance of a tumorigenic phenotype in H-RAS induced melanomas. A dominant-negative EGFR reduced the tumorigencity of melanomas, and sustained expression of EGFR can delay cancer regression [56].

The Tet-System has also been used to generate conditional models of mutant K-RAS induced lung adenocarcinoma [57, 58]. The Clara cell secretory protein (CCSP) promoter was used to regulate gene expression in alveolar epithelial cells. Within 7–14 days post induction of K-RAS overexpression, type II pneumocytes exhibited focal hyperplasia, and after 2 months, multiple solid adenomas or adenocarcinomas were present in the lung. The solid adenomas contained a population of macrophages, but lacked invasive growth and stromal elements. The adenocarcinomas had fewer macrophages and cytoplasmic inclusions, but had local invasion of the pleura. Within 3 days of K-RAS inactivation through doxycycline withdrawal, cancers exhibited decreased cellular density and an increased rate of apoptosis. Within 7 days of K-RAS inactivation, only a few patches of hyperplasia were found, and within a month no residual cancer tissue was found in 5/5 mice. The same mice were generated in either a p53 or an Ink4A/Arf deficient background. Cancers grew rapidly in these mice after K-RAS induction, but regressed with the same kinetics. TUNEL assays revealed that regardless of the genetic context cancers were regressing associated with apoptosis [58]. Similarly, CCSP regulated K-RAS (G12C) induced lung adenomas regressed upon oncogene inactivation [57]. Hence, even aggressive lung cancers in a tumor suppressor-deficient background regress upon the inactivation of a single oncogene.

A conditional transgenic model for BCR-ABL leukemias was generated using either the MMTV or SCL (stem cell leukemia) promoter to drive the expression of tTA [59, 60]. Upon induction of BCR-ABL the mice developed B-cell leukemia associated with lymphadenopathy, splenomegaly, and bone marrow infiltration. A third of mice with BCR-ABL under the control of the SCL promoter developed B-cell lymphoblastic disease resembling blast crisis, closely mimicking what is observed in patients with chronic myelogenous leukemia (CML). Inactivation of BCR-ABL induced rapid cancer regression in all mice. BCR-ABL inactivation was associated with the apoptosis of 80 % of the cancer cells within 20 h and complete tumor regression within 5 days. Sustained regression of cancer was observed in tumors arising from three of the four founder lines, as long as the mice had BCR-ABL continuously inactivated. Upon reactivation of BCR-ABL the cancers rapidly recurred. Interestingly, all the mice derived from the fourth founder relapsed within 4 weeks after complete regression. Relapsed cancers lacked continued expression of BCR-ABL protein and mRNA, suggesting that they had become independent of BCR-ABL expression.

The Tet and Tamoxifen systems have been used to demonstrate that MYC inactivation can induce cancer regression in a multitude of different types of cancer (Table 8.1). The Tet-Off System was used to regulate human c-MYC in lymphoid cells under the regulation of the EμSRα promoter [61–63]. When MYC is constitutively activated, 100 % of the mice developed hematopoietic tumors within 5 months. Upon gross examination, the mice exhibited enlargement of the thymus, liver, spleen, and gastrointestinal lymph nodes. Histological examination revealed that cancer cells had invaded all hematopoietic organs as well as liver, kidney, blood, and the lamina propia of the intestines. In one study, tumors were generally immature CD+4/CD+8 T-cell lymphomas and were rarely acute myeloid leukemias [61]. In another study, tumors were either B-cell or T-cell lymphomas [63]. In both studies, the resulting hematopoietic tumors exhibited a high degree of genomic instability reflected by chromosomal gains, losses, or translocations [61, 63]. Despite this genomic complexity, the inactivation of MYC resulted in rapid and sustained cancer regression. Upon MYC inactivation, cancer cells arrested, differentiated, and underwent apoptosis. Over 50 % of tumors exhibit sustained regression for over 30 weeks. Thus, MYC inactivation can induce sustained regression of hematopoietic tumors.

Conditional transgenic mice expressing c-MYC under the control of the Eμ-promoter occasionally developed highly metastatic osteosarcomas [44]. Histological examination of the primary cancer reveals the presence of disorganized bone matrix. MYC inactivation induced rapid cancer regression associated with the differentiation of cancer into mature bone. Continuous video time-lapsed microscopy (CVTL) revealed that upon MYC inactivation, cancer cells ceased to proliferate and differentiated, as confirmed by immunocytochemical analysis. Identically, MYC inactivation in cancers in vivo was associated with the differentiation of malignant cells into mature osteoid. Upon MYC reactivation less than 1 % of the cells were able to regain a proliferative phenotype. Surprisingly, MYC reactivation was also associated with the apoptosis of the now differentiated cancer cells. Moreover, even the transient inactivation of MYC was found to increase the survival of mice with these cancers. Hence, at least in some circumstances even brief oncogene inactivation can induce sustained loss of a neoplastic state.

The tamoxifen system also has been used to evaluate the consequences of MYC inactivation in different types of tumors using MycER^TAM (Table 8.2). MycER^TAM has been expressed in the skin through the involucrin promoter [34, 43]. MYC activation resulted in increased proliferation and blocked differentiation of the suprabasal epidermis. Sustained MYC activation resulted in hyperplasia, dysplasia, angiogenesis, and papillomatosis. MYC inactivation resulted in regression of blood vessels, restoration of cellular differentiation, and the regression of papillomas. A brief inactivation of MYC in keratinocytes caused the cells to differentiate and become unresponsive to MYC reactivation. MYC reactivation could not restore a proliferative phenotype to the differentiated keratinocytes and eventually the cells were sloughed off the skin [43]. Hence, brief inactivation of MYC can induce the sustained loss of neoplastic features in some skin tumors.