Recombinant adeno-associated viral vector 2/8 model of il6 mediated chronic inflammation

Abstract Background Interleukin 6 (IL6) is a pleiotropic cytokine mostly known as a proinflammatory cytokine. There is a need for a reliable in vivo model to study IL6 mediated chronic inflammation. Purpose The aim of our study was to create a mouse model mimicking the chronic inflammatory state as observed in cancer and inflammatory diseases. We opted for a recombinant adeno-associated viral vector (rAAV) approach to mimic the acute initiation of chronic inflammation later in life. Methods Our ultimate aim was to identify a rAAV subtype that gave a stable low-grade expression of IL6 in liver and serum, with minimal expression in other organs, particularly the heart. We selected the rAAV2/8 subtype based on the differential organ expression pattern of three different capsids, Cap7, Cap8 and Cap9. Instead of the potent viral CMV promoter, we opted for the human EF1a short (EFS) promoter, a short intron-less version derived from human EF1α for stable expression of IL6 and firefly luciferase led by an internal ribosome entry site (IL6-IRES-fLuc) DNA cassette. rAAV2/8_EFS-IL6-IRES-fLuc and the control vector rAAV2/8_EFS-MCS-IRES-fLuc were generated and different dilutions were tested to obtain a reliable long-term low grade IL6 overexpression. Results All mice treated with the highest dose of rAAV2/8_EFS-IL6-IRES-fLuc vector (8.8 1010 genome copies (GC)/animal) showed severe weight loss, starting between week 4-6 post injection. IL6 levels in plasma increased over time, fLuc activity started rising at 2 weeks with a peak at week 5 (Figure 1 panel A). In addition, mice suffered from severe hepatosplenomegaly. Histological examination of the liver showed features of chronic hepatitis as seen in viral or auto-immune hepatitis (Figure 1 panel B). In the spleen, the histological structure of red and white pulp was disturbed. In the heart we see infiltrating cells. The ejection fraction and fractional shortening at the end of the experiment remained within normal range. Assuming that the previous results were due to an overexpression of IL6 due to too high amounts of rAAV2/8_EFS-IL6-IRES-fLuc, we subsequently challenged mice with a rAAV2/8_EFS-IL6-IRES-fLuc dilution series to test which vector dose could be used for reliable long-term low grade IL6 overexpression. Weight loss was only observed following the injection of 4.4 1010 rAAV2/8_EFS-IL6-IRES-fLuc GC (Figure 2 panel A). With lower doses, there were less infiltrating cells in the liver, yet the disturbed anatomy of the spleen was still present (Figure 2 panel B). After extensive testing, the ideal dose for 12 weeks of IL6 overexpression model in 20g mice was considered between 2.2 and 2.93 1010 rAAV2/8_EFS-IL6-IRES-fLuc GC. Conclusion and discussion We successfully generated and characterized a mouse model of IL6 overexpression to mimic chronic IL6 mediated inflammation. To the best of our knowledge this is the first model employing IL6 overexpression using a rAAV viral vector.High dose rAAV2/8-IL6 vectorTitration of rAAV2/8-IL6 vector