A bioluminescence reporter mouse that monitors expression of constitutively active β-catenin

Abstract

This short technical report describes the generation and characterization of a bioluminescence reporter mouse that is engineered to detect and longitudinally monitor the expression of doxycycline-induced constitutively active β-catenin. The new responder transgenic mouse contains the TetO-ΔN89β-CatTMILA transgene, which consists of the tet-operator followed by a bicistronic sequence encoding a stabilized form of active β-catenin (ΔN89β-catenin), an internal ribosome entry site, and the firefly luciferase gene. To confirm that the transgene operates as designed, TetO-ΔN89β-CatTMILA transgenic mouse lines were crossed with an effector mouse that harbors the mouse mammary tumor virus-reverse tetracycline transactivator (MMTV-rtTA) transgene (termed MTB hereon), which primarily targets rtTA expression to the mammary epithelium. Following doxycycline administration, the resultant MTB/CatTMILA bigenic reporter exhibited precocious lobuloalveologenesis, ductal hyperplasia, and mammary adenocarcinomas, which were visualized and monitored by in vivo bioluminescence detection. Therefore, we predict that the TetO-ΔN89β-CatTMILA transgenic responder mouse-when crossed with the appropriate effector transgenic-will have wide-applicability to non-invasively monitor the influence of constitutively active β-catenin expression on cell-fate specification, proliferation, differentiation, and neoplastic transformation in a broad spectrum of target tissues.

Fig 1. Generation of the MTB/Cat^TMILA bigenic mouse.

(A) Design of the TetO- ΔN89β-Cat^TMILA transgene. The ΔN89β-catenin cDNA (2.1kb) was cloned into a single EcoR1 restriction site downstream of the TetO sequence in the TMILA (7.4kb) cloning vector [11]. The ΔN89β-catenin cDNA encodes the truncated Xenopus β-catenin protein with a myc-epitope tag fused in-frame at its N-terminus (black box). The location of the PCR primers for genotyping (black arrowheads) as well as the 13 centrally located Armadillo repeats (Arm repeats) is indicated. The inserted ΔN89β-catenin cDNA is followed by an IRES and a cDNA encoding the firefly luciferase protein. A SV40 polyadenylation signal (PA) serves as a strong transcriptional termination signal. The TetO- ΔN89β-Cat^TMILA transgene was linearized with Not1, isolated from vector sequences, and purified prior to pronuclear microinjection. (B) Schematic depicts the breeding strategy to generate the MTB/Cat^TMILA bigenic mice by crossing the MTB effector transgenic [14] with TetO-ΔN89β-Cat^TMILA responder transgenic. (C) Typical western immunoblot of isolated mammary epithelial cell protein. Lane 1, 2, and 3 denote mammary epithelial protein isolated from wild type ((WT) or non-transgenic) control (without doxycycline), MTB/Cat^TMILA bigenic (without doxycycline), and MTB/Cat^TMILA bigenic mice on food and water supplemented with doxycycline for 1-month respectively. Using antibodies to full-length β-catenin and the myc-epitope tag, the transgene-derived ΔN89β-catenin protein (75kDa) is only detected in the MTB/Cat^TMILA bigenic treated with doxycycline (lane 3); β-actin serves as a loading control. Each lane represents a protein isolate pooled from four individual mice per genotype and treatment.

Fig 2. Doxycycline-induced bioluminescence in the mammary gland of the MTB/Cat^TMILA bigenic.

(A) Overlay of full-body bioluminescence and x-ray images of monogenic control (G6719 (ear tag#)) and MTB/Cat^TMILA bigenic (G6718 (ear tag#)) mice following 0, 24, and 48 hours of doxycycline intake. By 24 hours of doxycycline administration, bioluminescence activity is detected in the #2, #3 (thoracic), and #4 (inguinal) mammary glands of the MTB/Cat^TMILA bigenic (G6718) but not in the monogenic control mouse (G6719). (B) Whole mount of mammary gland from monogenic control mouse (G6719) shows normal ductal morphogenesis (black arrowhead) following 1-week of doxycycline administration. (C) Mammary gland whole mount analysis shows precocious lobuloalveologenesis (white arrowhead) in the similarly treated MTB/Cat^TMILA bigenic (G6718). Scale bar in (B) applies to (C); see S1 Fig for corresponding low magnification images. (D) Myc-epitope tag immunohistochemistry does not detect myc-epitope tagged ΔN89β-catenin expression in the mammary epithelium of the doxycycline treated monogenic (G6719) control (black arrowhead). (E) Myc-tagged ΔN89β-catenin expression is clearly detected in the mammary epithelium of the similarly treated MTB/Cat^TMILA bigenic (white arrowhead); (F) is a higher magnification of (E). (G) Image shows a representative transverse section of an epithelial duct in the mammary gland of the doxycycline-treated monogenic (G6719) control mouse, which scores negative for BrdU incorporation following BrdU immunohistochemistry (black arrowhead). (H) Numerous cells scoring positive for BrdU incorporation are detected in the mammary epithelium of the similarly treated MTB/Cat^TMILA bigenic (white arrowhead); (I) is a higher magnification. Scale bar in (D) and (F) apply to (E, G, and H) and (I) respectively. See S1 Fig for more details and quantitation of BrdU positive cells in the mammary epithelium of both genotypes following doxycycline administration.

Fig 3. Bioluminescence detection of the emergence of doxycycline-induced mammary tumors in the MTB/Cat^TMILA bigenic.

(A) Overlay of whole-body bioluminescence and X-ray images shows mammary tumor enlargement in the MTB/Cat^TMILA bigenic (G6687) following doxycycline administration for the time periods indicated. Note the emergence of a thoracic mammary tumor (white arrowhead) in the bigenic. As expected, luciferase activity is not detected in the similarly treated monogenic control (G6688). Corresponding whole-body X-rays are shown alone in the bottom panels to enable clear visualization of the mammary tumor mass (white arrow head). (B) The MTB/Cat^TMILA bigenic (G6687) is shown following bioluminescence monitoring; note the thoracic mammary tumor detected in (A) above (white arrowhead). (C) and (D) are low and high magnification images respectively of mammary tumor tissue sections immunostained for myc-tagged ΔN89β-catenin expression; note: that most tumor cells score positive for myc-tag immunoreactivity (white arrowhead). (E) and (F) are low and high magnification images of mammary tumor tissue sections stained for BrdU incorporation; many tumor cells are immunopositive for BrdU incorporation (white arrowhead). Scale bar in (C) and (D) applies to (E) and (F) respectively.

Fig 4. Complete penetrance of the mammary tumor phenotype in the MTB/Cat^TMILA bigenic reporter.

(A) Kaplan-Meier tumor free plot for MTB/Cat^TMILA bigenics without doxycycline administration (n = 28 (blue)) and doxycycline-treated MTB/Cat^TMILA bigenics (n = 21 (red)) showing percent tumor free on the Y-axis versus age (in days) on the X-axis. (B) Overlay of whole-body bioluminescence and x-ray images of a MTB/Cat^TMILA (G6274) bigenic revealing two ipsilateral mammary gland tumors (#2 and #3 thoracic mammary glands (white arrowheads)). (C) The MTB/Cat^TMILA (G6274) bigenic reporter exhibiting two mammary tumors shown in (B (white arrowheads)). (D) Hematoxylin and eosin staining reveals that a subset of MTB/Cat^TMILA mammary tumors exhibit histologic characteristics consistent with squamous differentiation as evidence by the presence of pilar-like structures of confluent swirls of laminar keratin [20, 21]; (E) higher magnification of image shown in (D). (F) Typical immunostaining for myc-tagged ΔN89β-catenin expression in these tumors (white arrowhead). (G) Representative staining for BrdU incorporation in these tumors. Note that BrdU positive cells are localized to the region of the tumor that expresses ΔN89β-catenin (compare F with G (white arrowheads)); scale bar in (D) and (E) corresponds to (F) and (G) respectively.

Fig 5. Bioluminescence monitoring of doxycycline-induced ΔN89β-catenin expression in the salivary gland of the MTB/Cat^TMILA bigenic.

(A) Overlay of whole body bioluminescence and X-ray images of MTB/Cat^TMILA bigenic mouse following doxycycline administration for the indicated time periods. Note the localization of the bioluminescence signal to the submandibular salivary gland of the MTB/Cat^TMILA mouse (G8335). (B-D) and (E-G) panels represent salivary gland tissue immunostained for the transgene-derived myc-tagged ΔN89β-catenin protein and BrdU incorporation respectively. (B) Salivary gland tissue from doxycycline-treated monogenic control does not score positive for myc-tagged ΔN89β-catenin expression. (C) Cystic hyperplasia with strong immunostaining for myc-tagged ΔN89β-catenin expression is evident in salivary gland tissue isolated from the MTB/Cat^TMILA bigenic following 336 hours of doxycycline administration (black arrowhead). (D) is a higher magnification image of (C). (E) Salivary gland epithelial cells positive for BrdU are not evident in salivary gland tissue derived from doxycycline-treated monogenic control mice (black arrowhead). (F) Many cells score positive for BrdU incorporation in salivary gland tissue isolated from the similarly treated MTB/Cat^TMILA bigenic (black arrowhead); a higher magnification is shown in (G (black arrowhead)). Note: To date, we have not detected palable salivary tumors in these mice. Scale bar in (B) and (D) apply to (C, E, and F) and (G) respectively.