The tetracycline COL-3 reduces the inflammatory response of
microglial cells challenged with either LPS or αSa
Figure 2A confirms that amyloid fibrils were present in αS shaken
samples for 96h (lower image), whereas non-shaken samples did not
produce fibrilar αSa species.
To evaluate the anti-inflammatory potential of the non-antibiotic
tetracycline COL-3, we used microglial cells activated with the
bacterial cell wall component LPS (10 ng/ml), a prototypical agonist for
Toll-like receptor-4 (TLR4) [36]. After 24 hours of treatment, LPS
strongly stimulated the production/release of the proinflammatory
cytokine TNF-α. When the cultures were exposed to either 10 or 20 μM
COL-3, the production of TNF-α induced by LPS was significantly reduced
by more than 30 and 75%, respectively (Figure 2B). No significant
effect was observed, however, at 1 µM, indicating that the suppressive
action of the tetracycline derivative was concentration-dependent and
only significant after 10 µM. COL-3 at 20 µM appeared as effective as
the reference tetracycline DOX, at 50µM. The reference anti-inflammatory
drug DEX (2.5 µM) wasmore effective than either 20 µM COL-3 or 50 µM DOX
to reduce TNF-α release.
Cellular expression of the phenotypic activation marker Iba-1 was also
estimated under the same experimental conditions. We established that
COL-3 suppressed the LPS-induction of microglial Iba-1 expression only
at 10 or 20 µM (Figure 2B). At 50 µM, DOX also caused a robust decrease
of the Iba-1 immunosignal. Note that 2.5µM DEX reduced Iba-1 expression
with an efficacy that was similar to that of COL-3 and DOX, at 20 and
50µM, respectively. Photomicrographs illustrated LPS-treated microglial
cell cultures, exposed or not to COL-3 (20 µM) or DOX (50 µM) and then
processed sequentially for CD11b and Iba-1 fluorescence immunodetection,
binding proteins that are constitutively expressed in microglia (Figure
2C).
To model inflammatory reactions as they may occur in PD we used αSa to
trigger an inflammogenic response in microglial cell cultures. As
expected, a 24-hour exposure to αSa caused a robust increase of TNF-α
release whereas the monomeric form of αS (αSm) does not induce a
significant effect compared to control levels (Figure 2D). At 10 or 20
µM, COL-3 significant reduced TNF-α release induced by αSa. DOX exerted
a comparable repressive effect at 50µM. DEX at 2.5 µM was as effective
as COL-3 and DOX at 20 and 50 µM, respectively.
Coherent with previous observations, we also demonstrated that the
induction of Iba-1 expression induced by αSa was efficiently curtailed
by COL-3 (10 or 20 µM), DOX (50 µM) and DEX (2.5 µM) (Figure 2E).
Micrographs illustrate microglial cultures treated with αSa in the
presence or not of COL-3 (20µM) or DOX (50 µM) and then processed
sequentially for CD11b and Iba-1 fluorescence immunodetection (Figure
2F). It is interesting to point out that αSm does not induce any effect.
The tetracycline COL-3 reduces glucose accumulation in
microglial cells challenged with either LPS or αSa
Recent studies have demonstrated that glucose is an essential fuel for
microglial inflammatory processes [32,37]. Accordingly, we found
that a 24-h treatment with 10 ng/ml LPS leads to an almost 2-fold
increase in [3H]-2-DG uptake in microglial
cultures (Figure 3A). [3H]-2-DG uptake was
significantly reduced when LPS-treated cultures were exposed to either
10 or 20 µM COL-3, whereas a lower COL-3 concentration (1 µM) was not
effective (Figure 3A). A similar effect was obtained with DOX used at a
concentration of 50 µM. Surprisingly, DEX (2.5 µM) failed to alter
[3H]-2-DG uptake, indicating that the
anti-inflammatory effects of the glucocorticoid and the non-antibiotic
tetracycline derivative may occur through distinct mechanisms. The
uptake of [3H]-2-DG was also strongly reduced by
inhibiting NADPH oxidase activity with 300µM APO or by exposing the
cultures to 500 µM unlabeled 2-DG, a concentration of the glucose analog
susceptible to restrain glucose utilization by these cells without
affecting their survival (dos-Santos Pereira et al, 2020). As expected,
glucose uptake was also efficiently prevented when 50 mM glucose was
added acutely to the cultures to compete with
[3H]-2-DG during uptake measurement (Figure 3B).
Interestingly, αSa also robustly increased
[3H]-2-DG in microglial cell cultures (Figure 3C),
suggesting that the greater demand of glucose by microglial cells was
not circumscribed to LPS-induced inflammatory events. In contrast, αSm
did not induce statistical effects. Noticeably, COL-3 (10 or 20 µM)
reduced substantially glucose uptake when applied concomitantly to αSa.
This effect was mimicked by DOX at 50 µM. A co-treatment of αSa-treated
cultures with 300 µM APO or 500 µM 2-DG also led to a significant
reduction in [3H]-2-DG uptake. As expected
[3H]-2-DG uptake was also prevented when glucose
was added in excess to compete with the radioligand when performing
uptake measurements (Figure 3D).
COL-3 prevents the accumulation of NADPH in microglial cells
challenged with either LPS or αSa
Glucose may have a key impact on microglial inflammatory processes by
stimulating the pentose phosphate pathway and consequently the synthesis
of NADPH, the requisite substrate for the superoxide producing enzyme
NADPH oxidase [32,37]. Coherent with this view, we found that NADPH
levels were increased substantially in microglial cells activated by
either 10 ng/ml LPS or 70 µg/ml αSa but not with αSm (Figure 4A,B). Of
interest, COL-3 (20 µM) caused a significant reduction of NADPH levels
in both inflammation paradigms (Figure 4A,B). As expected, the
suppressive effect of COL-3 was mimicked by DOX at a concentration of 50
µM.
Both the inactive glucose analog 2DG (500 µM) and the NADPH oxidase
inhibitor APO (300 µM) mimicked the inhibitory effect of COL-3 on NADPH
production in microglia activated by LPS or αSa, indicating that the
non-antibiotic tetracycline may interfere with a glucose-dependent
mechanism that promotes NADPH synthesis and as a consequence ROS
production via the NADPH oxidase enzyme.