Figure 4. Efficacy of expression of the 2.0 kbp to 288 bp VPI.EGFP promoter deletion constructs (shown in Fig. 1A ) in the SON.

In these experiments after injection of the rAAVs into the SONs the control (normosmotic) rats were given water to drink for two weeks. In parallel experiments, the salt loaded rats were given water to drink for one week, followed by a second week of access only to water containing 2% NaCl (see Methods). The images in panels A1–6 and B1–6 present endogenous EGFP fluorescence (i.e. without the use of EGFP antibody in immunohistochemistry, −EGFP Ab). This was done in order to better assess quantitative differences between the constructs. In the third column (panels C1–6), IHC using EGFP antibodies together with salt loading was employed in order to maximize the detection of EGFP expression. Photographic image capture (exposure) times are identical for each panel. Under normosmotic conditions, clear expression of EGFP is obtained from injections of the 2.0 kVPI.EGFP (A1) and 1.5 kVPI.EGFP rAAV constructs (A2), and a much higher expression is seen for both under salt loading conditions (B1, B2). For the 950VPI.EGFP construct there was no detectable EGFP fluorescence under normosmotic conditions (A3), but salt loading was shown to significantly increase the EGFP expression in the SON (B3), and a greater increase in fluorescence was observed with both salt loading and IHC (C3). Neither normosmotic nor salt loading conditions produced detectable endogenous EGFP expression from the 543 bp to 288 bp VPI.EGFP constructs (A4–6, B4–6). However, application of both EGFP IHC (+EGFP Ab) and salt loading together clearly show that the EGFP was being expressed in the SON from the 543 bp to 288 bp VPI.EGFP constructs (C4–6). The red arrows denote three putative enhancer domains. Scale bar in panel A6 is 100 µm, and is the same for all the panels. Abbreviation: OC, optic chiasm.