Previously, biochemical analyses of hippocampal tissue from tau transgenic mice (rTg4510; [16
]) injected with MB showed reduced tau levels after 24 hours [15
]. Hippocampi of wildtype mice were then injected with either 1 mM or 0.1 mM MB (based on a recent study in Zebrafish [17
]) to determine its effective therapeutic window for reducing tau levels. Only 1 mM MB reduced tau levels (Figure ). Based on these results, 1 mM MB (1 mM; n = 6) or saline (n = 7) was administered to the right hippocampi [18
] of 7-month old rTg4510 mice for 1 month using mini-osmotic pump implantation (Figure ). The accumulation of soluble mutant human tau in the forebrain of this model causes spatial memory deficits as early as 3 months of age, which precede neuronal loss [16
]. Morris water maze (MWM) during the final week of treatment revealed that MB-treated mice showed significant improvements in learning the location of the escape platform compared to those receiving saline (Figure ). Probe trial analysis using target quadrant discrimination, number of platform crossings and search strategy imaging showed significantly improved cognitive recall in MB-treated mice compared to those treated with saline (Figure ). Biochemical analysis of hippocampal lysates showed that both phosphorylated (S202/T205) and total tau levels were significantly reduced (Figure ). Thus, despite the focal distribution of osmotic pump administration and the age of the mice, MB was still able to improve cognitive function; an effect that was concomitant with reductions in hippocampal tau levels. Contrastingly, pathology was unaffected (Figure ). Furthermore, nissl staining was done in order to determine if any damage was done to the hippocampus due to pump implantation. No such damage was found (Figure ).
Figure 1 Direct hippocampal infusion of methylene blue by mini-osmotic pump reverses spatial navigation related learning and memory deficits by reducing tau levels in rTg4510 mice. (A) Four groups of wildtype mice (n = 3) were injected with 100 μM MB, (more ...)
Figure 2 MB treatment reduces tau levels, but does not affect pathology in mice with pre-existing tangle formation and neurodegeneration. (A) Western blot analysis of tau protein levels from hippocampal lysates. (B) Quantitation of the optical density as a percentage (more ...)
Based on this evidence, a new trial was initiated in rTg4510 mice to test the effects of long-term MB administration not only on behavior and tau biochemistry, but also on neuronal survival and tau pathology. Practical limitations with osmotic pump applications required that this study be done using non-invasive peripheral administration. Dose selection for this study was based on two factors: 1) FDA conversion tables show that a 10 mg/kg dose in a mouse is equivalent to ~1 mg/kg in humans, which is within the range of current MB clinical applications [19
], and 2) pharmacokinetic analyses showed that MB could concentrate in the brain 500-fold, making the effective concentration. (> 100 μM) possible(Figure )
Figure 3 Pharmacokinetic analysis of MB following peripheral administration. (A) A single intravenous (I.V.) administration of 20 mg/kg MB was given to wildtype mice at different time points (n = 3 per time point), and cerebellar and plasma concentrations were (more ...)
Thus, two groups (n = 10) of 3 month-old rTg4510 mice and two groups of 10 wildtype littermates received either ~10 mg/kg(165 μM; 5× maximum recommended dose) of MB via drinking water supplemented with 2 mM saccharine or saccharine water alone (Figure ). Following 12 weeks of treatment, behavioral assessment showed no overt alterations in motor coordination or task acquisition (See additional file 1
: Figure ). MWM was then used to assess cognitive function. Probe trial analysis and search strategy imaging showed that MB, but not saccharine, prevented the significant progressive impairment in target quadrant discrimination that is a hallmark of the rTg4510 phenotype (Figure ). Furthermore, wildtype littermates showed normal spatial memory recall irrespective of MB treatment. Biochemical analyses of half-brain homogenates (excluding cerebellum) showed a reduction in soluble tau levels in some mice, but not others (Figure ). Histochemical analyses revealed no change in tau pathology in any mice (Figure ).
Figure 4 Chronic treatment with methylene blue has moderate effect in behavior. (A) Experimental design, n = 10. (B) Percent of time spent in each quadrant during the probe trial of MWM (F(15,144) = 8.781, p < 0.0001). MB rTg4510 recognize target versus (more ...)
Figure 5 Chronic dosing of methylene blue leads to reductions in soluble tau but pathology is unaffected. (A) Half-brain lysates (no cerebellum) were analyzed for tau protein levels by Western blot. (B) Optical density of tau levels is shown as a percentage of (more ...)
Given the variability in behavioral performance as well as reductions in tau levels, we hypothesized that the MB concentration in the brain may have also varied due to the ad libitum
administration strategy. To test this idea, brain concentrations of MB were assessed using LC-MS analysis of the cerebellar tissue from these mice. Indeed, MB concentration was positively correlated with MWM performance (p < 0.05) and was inversely correlated with soluble tau levels (p < 0.05) (Figure ). The differences in brain MB concentration between mice could not be attributed to body weight or gender (See additional file 1
: Figure ). Moreover, mice with >470 μM MB brain concentration accounted for a preponderance of the effects on memory function and tau reductions, consistent with our previous results showing that very high concentrations of MB were required for anti-tau efficacy (Figure ). Surprisingly, stereological assessment of five different brain regions from these mice showed that MB treatment significantly delayed neurodegeneration by ~30% in all forebrain regions of all rTg4510 mice (Figure ), but neuronal number failed to correlate with memory performance or soluble tau levels (Figure and data not shown).
Figure 6 Chronic dosing of methylene blue enhances neuronal survival. (A) Parenchymal drug concentrations correlated significantly with memory retention in the probe trial of MWM p = 0.016, Pearson r = 0.766, r2 = 0.587. (B) Tau levels inversely correlate significantly (more ...)
Post-hoc analysis of the behavioral performances of rTg4510 mice with brain MB concentrations above or below ~470 μM (High [MB] or Low [MB], respectively), elucidated that the High [MB] cohort performed equivalent to wildtype mice, while the Low [MB] cohort was significantly impaired (Figure ). We again used the camera-tracking software to map the areas of the pool most traversed by each cohort (Figure ). High [MB] rTg4510 mice were predominantly found in the target quadrant, while the Low [MB] cohort displayed an unguided search strategy. Furthermore, we wanted to see if MB treatment in the High [MB] mice had an effect on motor learning. The mice were subjected to the rotorod task for two days and the latency to fall onto a spring-cushioned lever was measured. We found that motor learning from day 1 to day 2 of MB treated rTg4510 was significantly improved, dissimilar to the saccharine treated rTg4510 analogs (Figure ).
Figure 7 High concentrations of MB repair spatial and motor memory deficits. (A) (A)Post-hoc analyses reveal that High [MB] rTg4510 mice (n = 5) perform similar to wild-type MB treated mice, while Low [MB] rTg4510 mice (n = 4) did not display memory retention. (more ...)