Figure 2.

Intrathecal injection of pCpG+, pCpG−, LPS or vehicle. (a–c) Rats were injected intrathecal with equimolar amounts of pCpG+ (120 µg, 7.5 µg/µl, 770 CpG) or pCpG− (100 µg, 7.6 µg/µl, 0 CpG) containing undetectable (<4 pg of LPS/100 µg DNA) amounts of LPS, 15 pg of LPS (10 pg/µl) or vehicle. Lumbosacral CSF was collected (spontaneous needle outflow) just prior to injection (naive group) and at 3, 16 and 49 h post-injection [n = 4–5 per group except IL-6 at 16 and 49 h (n = 2–4), IL-1 (n = 2–4) and naive (n = 7–8)] and assayed for TNF-α, IL-6 or IL-1 via ELISA. (a) At the 3-h time point 15 pg of LPS (n = 5) elicited 1 ± 0.1 pM TNF-α, whereas vehicle (n = 4) elicited 0.4 ± 0.05 pM TNF-α. (d–f) Rats were injected with 100 µg of the additionally purified pCpG+ preparation used in (a–c) (n = 5, again undetectable LPS) or pCpG− (n = 4)with added LPS (measured at 13 pg of LPS/100 µg plasmid). Lumbosacral CSF was collected and assayed as above (naive n = 4). Molar values were calculated assuming 51-kDa TNF-α, 22-kDa IL-6 and 18-kDa IL-1β standards. Data are the mean ± SEM