Construction of a doxycycline inducible adipogenic lentiviral expression system

Abstract

To provide a tool for research on regulating adipocyte differentiation, tetracycline inducible (Tet on) lentiviral expression vectors under the control of an adipose-specific promoter were constructed. The lowest basal expression in the absence of doxycycline and most efficient dose-dependent, doxycycline-induced transient overexpression was observed using vectors constructed with a combination of Tetracycline Responsive Element (TRE) and reverse tetracycline-controlled TransActivator advanced (rtTAadv), transfected in white (3T3-L1) and brown (HIB-1B) preadipocytes cell lines. The results demonstrate that doxycycline adipogenic inducible expression can be achieved using a pLenti TRE / rtTA adv under the control of the truncated aP2 promoter in HIB-1B preadipocytes.

Fig. 1

Schematic working process of Tet On lentiviral expression vectors. The TRE is located upstream of the CMV promoter (P_CMV) driving the target gene LucGFP which is silent in the absence of activation by the reverse tetracycline-controlled transactivator (rtTA). The rtTA is transcribed and translated and binds to the TRE but only activates transcription of the target gene LucGFP in the presence of Dox.

Fig. 2

Schematic of the lentiviral LucGFP expression vectors. (A) The constitutive pLenti LucGFP expression vector was created by 3-fragment MultiSite Gateway® cloning system (LR reaction) using pLenti6/V5 and entry clones of LucGFP, P′_CMV and rtTA. (B) The TRE (1, 3) or TRE tight (2, 4) element was inserted upstream of the CMV promoter (P_CMV) in the pLenti6/V5 vector using BamHI and XhoI. The LucGFP, P′_CMV and rtTA or rtTA adv entry clones were cloned into the modified backbones by LR reaction. (C) The full length (P_aP2) or truncated (P′_aP2) aP2 promoter was inserted upstream of the transactivator rtTA adv element in either the TRE or TRE tight P_CMV / LucGFP pLenti backbone by the 3-fragment MultiSite Gateway® cloning system.

Fig. 3

Comparison of different Tet on LucGFP vectors (CMV driven) by luciferase assay in 3T3-L1 (A) and HIB-1B (B) preadipocytes. HIB-1B and 3T3-L1 preadipocytes were transiently transfected with different Tet On CMV-LucGFP vectors, the empty vector pLenti TRE (negative control) or the constitutive vector pLenti-LucGFP (CMV, positive control). Cells were given 1 μg/ml doxycycline (Dox) 24 h post-transfection and cultured for 24 h. Luciferase activity was expressed as fold change relative to the value from the cells transfected with pLenti TRE tight-LucGFP-rtTA adv vector without doxycycline induction. Results represent mean ± S.E.M from 2 independent replicate experiments performed in triplicate wells. Student’s t-test was used to analyze the data. ^∗P < 0.05 with respect to basal TRE rtTA; #P < 0.05 with respect to the same vector without Dox treatment.

Fig. 4

The Tet on LucGFP vectors responded to doxycycline in a dose-dependent manner. 3T3-L1 preadipocytes were transfected with either TRE LucGFP rtTA adv or TRE tight LucGFP rtTA adv vector. Doxycycline was added into the cells 24 h post-transfection at the concentration indicated in the graph. Firefly luciferase activity was measured 24 h after the addition of doxycycline. The error bars represent S.E.M from 2 independent replicate experiments performed in triplicate wells.

Fig. 5

Adipogenic stimulation of aP2 expression by acute treatment of 3T3-L1 and HIB-1B cells. 10 μM rosiglitazone (Rosi) were added to confluent HIB-1B and 3T3-L1 (A) cells and 10 μM forskolin (Fsk) was added to the culture 12 h later. RNA was extracted from the cells 12 h after adding forskolin and aP2 mRNA expression levels in treated 3T3-L1 and HIB-1B cells was determined by quantitative real time PCR and normalized against 36B4 housekeeping gene expression. (B) 3T3-L1 preadipocytes were transfected with C/EBPβ and PPARγ by FugeneHD®. 10 μM rosiglitazone (Rosi) were added into the cells 24 h post-transfection and RNA was extracted from the cells 24 h after adding rosiglitazone. (C) The differentiated 3T3-L1 cells (dif 3T3-L1) were derived from a standard differentiated protocol outlined in the Methods section. The level of aP2 mRNA in treated cells was determined by quantitative real time PCR and normalized against 18s housekeeping gene expression, relative to the value from the 14 h pcDNA transfected cells. Each bar represents the mean ± S.E.M from 2 independent replicate experiments performed in duplicate wells. (A) ^∗P < 0.05 by Student’s t-test with respect to controls. (B) (C) Student’s t-test: ^∗P < 0.05 caused by time of Rosi treatment or differentiation with respect to the control at 14 h; #P < 0.05 caused by overexpression with respect to the control at the same time point.

Fig. 6

Activity of adipogenic and doxycycline inducible lentiviral LucGFP expression vectors in 3T3-L1. The inducible LucGFP lentiviral vectors with full length aP2 promoter (Tet On P_aP2 vectors, A) or with truncated aP2 promoter (Tet on P′_aP2 vectors, B) were co-overexpressed with C/EBPβ and PPARγ into 3T3-L1 preadipocytes. 10 μM rosiglitazone (Rosi) and 1 μg/ml doxycycline (Dox) were added to the cells 24 h post-transfection as indicated and treated for 24 h before luciferase assay. Firefly luciferase activity was interpreted relative to the control value (no co-overexpression, no drug treatments) of each group. Each bar represents the mean ± S.E.M from 2 independent replicate experiments performed in triplicate wells. Data was analyzed by two-way ANOVA for each lentiviral vector to examine the effects of Dox treatment and application of adipogenic conditions (C/EBPβ + PPARγ co-expression + Rosi treatment). (A) TRE P_aP2 (left): Dox (P < 0.001) and adipogenic condition (P < 0.001) both significantly increased P_aP2 activity and there was no interaction (P = 0.18) between the two treatments; TRE tight P_aP2 (right): Dox (P < 0.001) significantly increased P_aP2 activity while adipogenic conditions had no effect (P = 0.31) and there was no interaction (P = 0.091). (B) Neither Dox nor adipogenic condition treatment had significant effects on P′_aP2 activity.

Fig. 7

Activity of adipogenic and doxycycline inducible lentiviral LucGFP expression vectors in 3T3-L1 and HIB-1B preadipocytes. The two fat specific expression vectors, pLenti TRE-LucGFP P_aP2 rtTA adv and pLenti TRE-LucGFP P′_aP2 rtTA adv were transiently expressed into HIB-1B (A) or 3T3-L1 (B) preadipocytes. 10 μM Rosiglitazone (Rosi) and 10 μM forskolin (Fsk), as indicated, were added to the cells 24 and 36 h post-transfection respectively. 1 μg/ml doxycycline (Dox) was added to the cells 24 h post-transfection to induce the expression of the Tet on vectors. Luciferase assay was carried out 48 h after transfection and firefly luciferase activity was interpreted relative to the control value (no Rosi, no Fsk, and no Dox) of each group. Each bar represents the mean ± S.E.M from 2 independent replicate experiments performed in triplicate wells. Data was analyzed by two-way ANOVA with each lentiviral vector. (A) left: no interaction between Dox and Fsk + Rosi treatments (P = 0.111) but Dox (P < 0.001) and Fsk + Rosi (P < 0.001) both independently increased P_aP2 activity; right: no interaction between Dox and Fsk + Rosi treatments (P = 0.459) while Dox (P < 0.001) and Fsk + Rosi (P = 0.008) both independently increased P′_aP2 activity. (B) No interaction between Dox and Fsk + Rosi treatments (P > 0.05) and Dox and fsk + Rosi independently increased both P_aP2 and P′_aP2 activity (P < 0.001).