Expression of transgenes targeted to the Gt(ROSA)26Sor locus is orientation dependent

Abstract

Background: Targeting transgenes to a chosen location in the genome has a number of advantages. A single copy of the DNA construct can be inserted by targeting into regions of chromatin that allow the desired developmental and tissue-specific expression of the transgene.

Methodology: In order to develop a reliable system for reproducibly expressing transgenes it was decided to insert constructs at the Gt(ROSA)26Sor locus. A cytomegalovirus (CMV) promoter was used to drive expression of the Tetracycline (tet) transcriptional activator, rtTA2(s)-M2, and test the effectiveness of using the ROSA26 locus to allow transgene expression. The tet operator construct was inserted into one allele of ROSA26 and a tet responder construct controlling expression of EGFP was inserted into the other allele.

Conclusions: Expression of the targeted transgenes was shown to be affected by both the presence of selectable marker cassettes and by the orientation of the transgenes with respect to the endogenous ROSA26 promoter. These results suggest that transcriptional interference from the endogenous gene promoter or from promoters in the selectable marker cassettes may be affecting transgene expression at the locus. Additionally we have been able to determine the optimal orientation for transgene expression at the ROSA26 locus.

Figure 1. Targeting strategy for introducing to ROSA26locus.

(A). Diagram of targeting constructs to introduce Dox responsive transgenes into locus. A1 and A2 introduce the same activator transgene in opposite orientations. B1 and B2 introduce the same responder transgene in opposite orientations. (B) Four different cell lines were produced with the activator and responder transgenes in different orientations (C) Southern blots probed with 5′ and 3′ flanking probes produce the correct band sizes to demonstrate appropriate targeting of constructs.

Figure 2. Dox regulated gene expression works appropriately at the ROSA26 locus.

(A) EGFP expression induced in A1B1 cells by addition of dox to a final concentration of 1 µgml⁻¹. (i) and (iii) Phase contrast. (ii) and (iv) EGFP fluorescence (B) EGFP expression in single targeted and double targeted cell lines in reponse to 1 µgml⁻¹ dox. Induction of EGFP is only seen in the A1B1 cells (F(2,30) = 1167.61, p<0.0001). Each point is an average of three measurements each from two independently targeted cell lines. Error bars denote standard deviations. Asterisks indicate statistically significant differences. (C) Dose response curve EGFP expression in A1B1 cells in response to dox.

Figure 3. Expression level is dependent on orientation.

(A) Diagram of orientation of constructs in cell lines. (B) A1B1 and A2B2 cells induced with doxycycline (i) A1B1 phase contrast (ii) A1B1 EGFP expression (iii) A2B2 phase contrast (iv) A2B2 EGFP expression (C) Graph of expression levels in induced and uninduced cell lines quantitated by fluorimetry. A2 cell lines have higher EGFP expression compared with the A1 cell lines (F(1,40) = 346.09, p<0.0001). Expression levels between the B1 and B2 cell lines were not significantly different (F(1,40) = 0.05, p = 0.8211). Each point is an average of three measurements each from two independently targeted cell lines. Error bars denote standard deviations. Asterisks indicate statistically significant differences. (D) Expression level of rtTA measured by RT-PCR.

Figure 4. Expression level can be increased when selectable marker is removed.

(A) Diagram of orientation of constructs in cell lines. (B) A2B2 and A2B2Δ cells induced with doxycycline (i) A2B2 phase contrast (ii) A2B2 EGFP expression (iii) A2B2Δ phase contrast (iv) A2B2Δ EGFP expression (C) Graph of expression levels in induced and uninduced cell lines quantitated by fluorimetry. Expression of EGFP is higher when these cell lines have lost the selectable marker cassette (F(1,40) = 77.25, p<0.0001). The orientation of the TRE EGFP makes no difference to expression level (F(1,40) = 1.14, p = 0.2910). Each point is an average of three measurements each from two independently targeted cell lines. Error bars denote standard deviations. Asterisks indicate statistically significant differences.

Figure 5. Expression level is decreased when selectable marker is removed and promoter is in same orientation as ROSA26 promoter.

(A) Diagram of orientation of constructs in cell lines. (B) Α1Β1 ανδ Α1Β1Δ cells induced with doxycycline (i) A1B1 phase contrast (ii) A1B1 EGFP expression (iii) A1B1Δ phase contrast (iv) A1B1Δ EGFP expression (C) Graph of expression levels in induced and uninduced cell lines quantitated by fluorimetry. Expression of EGFP is lower when the selectable marker is removed from the A1 cell lines (F(1,40) = 51.77, p<0.001). Again the orientation of the TRE-EGFP makes no significant difference to the expression levels observed (F(1,40) = 0.05, p = 0.8295). Each point is an average of three measurements each from two independently targeted cell lines. Error bars denote standard deviations. Asterisks indicate statistically significant differences.