The BBB maintains brain homeostasis and selectively excludes most blood-borne and xenobiotic substances from entering the brain, protecting it from systemic and exogenous influences [72, 73]. Deterioration in BBB functions have been clearly shown to play a major role in the pathogenesis and progression of a steadily increasing number of neurological disorders such as multiple sclerosis, Alzheimer's disease, neoplasia, hypertension, dementia, epilepsy, infection, and trauma . While the harmful effects of smoking on public health have been well demonstrated and detailed in many organs, the BBB has received much less attention despite the strong evidence for an association between tobacco smoke and vascular impairment. Recently published data  have shown that TS negatively affects vascular endothelial physiology and directly promotes the pro-inflammatory activation of BBB endothelium.
Many of the over 4000 components contained in TS are thought to facilitate the development of a proatherosclerotic environment [25, 28, 75–77] by triggering a complex pro-inflammatory response. This is mediated by the recruitment of leukocytes through cytokine signalling (e.g., IL-2, IL-4, IL-6, IL-8, INF-γ and GM-CSF), matrix metalloproteinase upregulation (e.g., MMP-1, MMP-8, and MMP-9), and by promoting the adherence of monocytes to the endothelial wall [25, 78].
TS's status as a powerful pro-inflammatory agent is suggested by the observation of elevated levels of white blood cells, primarily neutrophils and monocytes, in smokers . Elevated C-reactive protein (CRP) levels caused by TS promotes endothelial dysfunction by lowering the production of NO and diminishing its bioactivity , thereby limiting the smoker's ability to vasodilate via NO release.
One well accepted yet poorly understood mechanism by which smoking can directly hamper BBB viability and function relays upon oxidative stress caused by the many highly reactive oxygen species and free radicals of which TS is highly enriched. The gas-phase of tobacco smoke alone is a complex mixture of high concentrations of ROS and free radicals (approximately one quadrillion radicals per puff ). This represents an exorbitant load of oxidant to body tissues which can lead to BBB breakdown by oxidative damage, tight junction (TJ) modification, and matrix metalloproteinases (MMP) activation . The results presented herein suggest that antioxidant vitamins may provide an effective shielding against oxidative stress thus, protecting the BBB during TS exposure.
Our data have shown that the administration of physiological concentration of vitamin E and/or C reduced the BBB endothelial pro-inflammatory activation solicited by TS exposure (see Figure 1). While the precise mechanism/s by which TS triggers an immune vascular response at the BBB cannot be determined from these experiments, our data strongly suggest that: 1) the imbalance between tissue/plasma concentrations of antioxidants and reactive oxygen species caused by TS played a major role in the process; 2) this can be partially prevented by administration of antioxidant vitamins.
The BBB endothelium is not the only tissue affected by TS. Our data also showed that TS directly activated monocytes which resulted in the release of IL-1β and TNF-α and MMPs (see Figure 2). We also observed a time dependent increase in the activity of MMP-2 and MMP-9 which initiated as early as 15 minutes of TS exposure and increased substantially after 60 minutes. The increased activity of matrix metalloproteinases reflected the continuous release of these enzymes during TS exposure suggesting an ongoing pro-inflammatory stimulation of the cells. Similar to what observed in BBB ECs cultures, pre-treatment with antioxidant vitamins reduced monocytes pro-inflammatory activation by TS.
In this study we used a final concentration of vitamin C (80 μM) equivalent to that found independently of gender in healthy non-smoker subject (35.3-107.1 μM; 6.2-18.8 mg/L ) according to literature data . Not surprisingly the anti-inflammatory activity of vitamin C was limited to the initial 15 minutes of TS exposure. Vitamin C is an hydrophilic reducing agent making it the first line of defense against ROS and the first to be consumed in the process. Therefore, it very likely that higher concentrations and/or a sustained supplementation of vitamin C are necessary to maintain a protective effect for a longer period of time. This is agreement with the decision of Food and Nutrition Board of the National Academy of Sciences to raise the RDA of vitamin C for smokers to over 200 mg/day (≅ 160 μM/l cconsidering an estimated 70% absorption which decreases as intake increases) versus the recommended 90 mg/day for non-smokers. This help counteracting vitamin C depletion caused by TS-derived ROS over time.
On the other hand, when vitamin C was co-administered with vitamin E (a powerful fat soluble reducing agent and the primary antioxidant present in the lipid-rich cell membranes; see also Figure 3) the protective effect was significantly enhanced both in efficacy and duration. The interaction between vitamins E and C has been recognized and extensively studied showing an ascorbate-mediated regeneration of oxidized membrane alpha-tocopherol back to its reduced form. A process that extends the protective effect of vitamin E against oxidative stress .
Whether ROS can directly hamper BBB viability was also clearly demonstrated. Our results showed a time and dose dependent loss of BBB viability when in the absence of physiological antioxidants protection the DIV-BBB modules were exposed to H2O2 levels comparable to those a chronic smoker is routinely exposed (see Figure 5). Our findings are in agreement with studies by others showing that plasma antioxidant vitamins (especially ascorbic acid) levels in chronic smokers are significant decreased due to accelerated turnover in response to TS-evoked oxidative stress . This is also in agreement with the fact that administration of antioxidant vitamins C and E (particularly when combined) prevented the loss of BBB integrity.
Not surprisingly, direct BBB exposure to TS showed similar results. We observed a similar time and dose dependent pattern of BBB impairment (see Figure 4A), which was paralleled by the release of adenylate kinase (a well-established marker of cell damage; see Figure 4B). This suggests that the increased paracellular permeability is not exclusively dependent (at least in the short term) on the exposure to nicotine  contained in TS but also to cell damage caused by ROS. This hypothesis is supported by the fact that pre-treatment with antioxidant vitamins significantly reduced the loss of BBB viability during TS exposure (see Figure 4C). This was accompanied by a similar decreased in the release of AK (see Figure 4D) and that of pro-inflammatory cytokines (see Figure 4E). The latter is also in agreement with previous studies showing that vitamin C plays an important (yet not fully understood) role in the modulation of cytokines release as well as other immune surveillance activities .
Another mechanism through which tobacco smoke can impair BBB viability is through the alteration of nitric oxide bioavailability. NO is a molecular mediator involved in a variety of physiologic and pathological processes. Three isoforms of NOS have been described: neuronal NOS (nNOS), endothelial (eNOS) and inducible or immunological NOS (iNOS). Vascular endothelial cells generate small bursts of NO from L-arginine via increase in intracellular Ca2+ through a calcium-dependent constitutive NO synthase (eNOS). In contrast, iNOS induced in pathological/inflammatory state is Ca-independent manner and releases NO continuously in large amounts. Our data showed a sudden burst in NO (see Figure 6A) immediately after the initial exposure to tobacco smoke. Consequent to the sudden raise of NO we observed a steady decline on NO levels which suggests a rapid reaction of NO with other TS-released ROS and its conversion in peroxynitrite (ONOO-) . This hypothesis is in agreement with the fact that NO decline was prevented by pre-administration of antioxidant vitamins, which neutralized the ROS contained in TS. This being the case, it is possible that antioxidant treatments may indeed prolong the initial NO-dependent vasodilatation experience by smokers during TS exposure.
By contrast, during the recovery time (in the absence of TS) NO dropped below the baseline level. The magnitude of this decline appeared to be proportional to the amount of TS to which the corresponding BBB modules were previously exposed. While the precise mechanisms by which TS affected NO bioavailability cannot be entirely determined from these experiments, some inferences can be still derived. Nicotine contained in TS has been shown to inhibit eNOS activity, thus decreasing the constitutive level of NO release by the vascular endothelium [86, 87]. As shown in Figure 6B pre-treatment with antioxidant vitamins did not affect the decline of NO during the recovery phase. This supports the hypothesis that perhaps nicotine rather than ROS are responsible for the observed decline of NO during this phase leading to vasoconstriction and aggravating the vascular impairment of the brain [88, 89].
Tobacco smoking has also been associated with impairments of glucose metabolism and glycemic regulation . Our data have clearly shown that TS exposure determined a decrease in glucose consumption (Figure 7B left panel), paralleled by an increase in lactate production. This is indicative of a potential change in the BBB bioenergetic behaviour toward a more predominant anaerobic metabolism (see Figure 7B right panel). The mechanisms through which TS affect glucose metabolism may be related to free radicals and other ROS contained in tobacco smoke . In vivo, this effect can be further exacerbated by the carbon monoxide (CO) also contained in tobacco smoke. CO through binding to hemoglobin, myoglobin and mitochondrial cytochrome oxidase can cause a relatively prolonged impairment of glucose oxidative metabolism and a decrease in glucose utilization [92, 93]
One way through which BBB endothelial cells can try to counteract the increased levels of ROS is through the generation of reductant equivalents such as glutathione peroxidase (of the principal antioxidant defense enzymes). Generation of glutathione peroxidase requires oxidized glutathione to be reduced by glutathione reductase using NADPH generated in the pentose phosphate pathway (also called the phosphogluconate pathway). This process generates NADPH (oxidative phase) and pentoses (5-carbon sugars; non-oxidative synthesis) . A shift toward the pentose phosphate pathway to compensate for the excess of ROS would explain the reduced glucose consumption we observed during TS exposure.
Furthermore, BBB endothelial cells could also use the Krebs cycle reducing agents (normally utilized to sustain the energy-producing pathway based on oxidative phosphorylation) to counteract TS-derived ROS. This would leave the cellular mitochondria underequipped to sustain the oxidative glucose metabolism and would explain the increased lactate production also observed during TS exposure. This hypothesis is supported by our data which showed that administration of antioxidant vitamins C and E gradually restore the normal bioenergetic metabolism of the BBB (see Figure 7B) [11, 94]. However, a more specific and detailed study will be necessary to confirm this hypothesis and unravel the underlining mechanism/s.