Impact of tamoxifen on adipocyte lineage tracing: Inducer of adipogenesis and prolonged nuclear translocation of Cre recombinase

Abstract

Background: The selective estrogen receptor modulator tamoxifen, in combination with the Cre-ER(T2) fusion protein, has been one of the mainstream methods to induce genetic recombination and has found widespread application in lineage tracing studies.

Methods & results: Here, we report that tamoxifen exposure at widely used concentrations remains detectable by mass-spectrometric analysis in adipose tissue after a washout period of 10 days. Surprisingly, its ability to maintain nuclear translocation of the Cre-ER(T2) protein is preserved beyond 2 months of washout. Tamoxifen treatment acutely leads to transient lipoatrophy, followed by de novo adipogenesis that reconstitutes the original fat mass. In addition, we find a "synthetically lethal" phenotype for adipocytes when tamoxifen treatment is combined with adipocyte-specific loss-of-function mutants, such as an adipocyte-specific PPARγ knockout. This is observed to a lesser extent when alternative inducible approaches are employed.

Conclusions: These findings highlight the potential for tamoxifen-induced adipogenesis, and the associated drawbacks of the use of tamoxifen in lineage tracing studies, explaining the discrepancy in lineage tracing results from different systems with temporal control of gene targeting.

Keywords: Adipose tissue; Cre recombinase; Lineage tracing; Tamoxifen.

Figure 1

Tamoxifen treatment induces acute lipoatrophy in mice. Mice were subjected to 5 consecutive daily oral gavages of tamoxifen (100 mg/kg BDW/day) or vehicle oil at a matched volume (5 mL/kg BDW/day). (A) On days 1, 2, 3, 4, 5, 6, and 15, body weight normalized as the percentage of the initial one before the first tamoxifen administration (day 1). (B) Fat pad weight normalized against body weight on day 6 or 15. Both sides of inguinal subcutaneous (sWAT) and epididymal white adipose tissue (eWAT) were measured and averaged. Data are presented as the mean ± SEM. n = 5 (oil) and 9 (tamoxifen) on days 1–5. n = 2 and 4 on day 6. n = 3 and 5 on day 15. *p < 0.05, **p < 0.01. (C and D) Representative H&E stains (C) and perilipin immunofluorescence (D) of sWAT and eWAT on day 6. Lower panels show magnified views of the boxed areas in the corresponding upper panels. Non-specific fluorescence level in the negative control was uniformly subtracted from all the images.

Figure 2

Adipocyte-specific PPARγ knockout aggravates the fat loss induced by tamoxifen. Adn-PPARγ-KO ([Adn-rtTA; TRE-Cre; PPARγ^flox/flox]) and control mice ([Adn-rtTA; PPARγ^flox/flox]) were fed doxycycline diet for 5 consecutive days to eliminate PPARγ in adipocytes. (A) A schematic drawing of the doxycycline-inducible knockout of the PPARγ gene in Adn-PPARγ-KO mice. The Adn-rtTA transgene expresses the reverse tetracycline transcription activator (rtTA) protein specifically in adipose tissue. It binds to the tetracycline response element (TRE) promoter and activates the expression of Cre recombinase, only in the presence of doxycycline. Thus, the Cre-mediated knockout of PPARγ is exclusively in adipocytes, and under temporal control of doxycycline administration. (B) RT-qPCR of PPARγ mRNA in adipose tissue after doxycycline diet. Data are presented as the mean ± SEM. n = 11 (Control) and 20 (Adn-PPARγ-KO). **p < 0.01. (C and D) Concomitantly with doxycycline diet, mice were subjected to 5 daily oral gavages of tamoxifen (100 mg/kg BDW/day) or vehicle oil at a matched volume (5 mL/kg BDW/day). On day 5, the surviving fat pads of sWAT (A) and eWAT (B) were dissected and processed for H&E staining. Lower panels show magnified views of the boxed areas in upper panels.

Figure 3

De novo adipogenesis in fat pads recovered from tamoxifen treatment. AdipoChaser mice ([Adn-rtTA; TRE-Cre; Rosa26-loxP-Stop-loxP-lacZ]) were first fed on doxycycline diet for 1 week to induce β-galactosidase expression in all mature adipocytes, and then switched to regular diet for 1 week to wash out doxycycline-induced Cre expression. Subsequently, mice were subjected to 5 daily oral gavages of tamoxifen (100 mg/kg BDW/day) or vehicle oil at a matched volume (5 mL/kg BDW/day). 10 days after the last tamoxifen administration, the recovered fat pads of sWAT (A) and eWAT (B) were stained for β-galactosidase activity (blue) and counterstained with Nuclear Fast Red (red). Lower panels show magnified views of the boxed areas in upper panels.

Figure 4

Residual tamoxifen extends nuclear translocation of Cre-ER^T2. (A to D) Qualitative mass spectrometry of tamoxifen and 4-hydroxytamoxifen (4-OH tamoxifen) in adipose tissue of mice after 5 daily oral gavages of tamoxifen (100 mg/kg BDW/day) or vehicle oil at a matched volume (5 mL/kg BDW/day). The approximate retention times of tamoxifen and 4-OH tamoxifen are indicated by numbers in the plotting area. (A) Chemically synthesized standards. (B and C) Tissue extract 1 day after the last gavage (day 6) of oil (B) or tamoxifen (C). (D) Tissue extract 10 days after the last gavage (day 15) of tamoxifen. (E and F)Adn-Cre-ER^T2 mice were subjected to 5 daily intraperitoneal injections of tamoxifen (100 mg/kg BDW/day) followed by indicated washout time. Wildtype male or Adn-Cre-ER^T2 female mice without tamoxifen treatment served as negative controls. Paraffin sections of sWAT (E) and eWAT (F) were subjected to immunofluorescence for Cre recombinase. Upper panels: Cre signal. Non-specific fluorescence level in wildtype control was uniformly subtracted from all the images. Lower panels: Cre signal merged with DAPI staining. (G and H) [Adn-rtTA; TRE-Cre; Rosa26-mT/mG] mice were fed on doxycycline diet for 10 days to induce GFP expression in all mature adipocytes, and then switched to regular diet for 3 days to wash out doxycycline-induced Cre expression. Subsequently, mice were subjected to 10 days of cold exposure, concomitantly fed on regular or doxycycline diet. (G) A schematic model of lineage tracing results supporting de novo differentiation of beige adipocytes. The initial doxycycline administration irreversibly switches the expressed reporter in mature adipocytes from tdTomato (red) to GFP (green). After doxycycline washout followed by cold exposure, GFP^-tdTomato⁺ (red) beige adipocytes are from de novo differentiation, while GFP⁺tdTomato⁻ (green) cells are progeny of the pre-existing mature adipocytes. If doxycycline is administrated during cold exposure, all the de novo differentiated beige cells are irreversibly switched to GFP expression. (H) Paraffin sections of sWAT were subjected to co-immunofluorescence of perilipin (red) and GFP (green). The lowest panels show magnification of the boxed areas in the merged images. Arrowheads: GFP-negative multilocular adipocytes.