High-Dose Acetaminophen Alters the Integrity of the Blood–Brain Barrier and Leads to Increased CNS Uptake of Codeine in Rats

The consumption of acetaminophen (APAP) can induce neurological changes in human subjects; however, effects of APAP on blood–brain barrier (BBB) integrity are unknown. BBB changes by APAP can have profound consequences for brain delivery of co-administered drugs. To study APAP effects, female Sprague–Dawley rats (12–16 weeks old) were administered vehicle (i.e., 100% dimethyl sulfoxide (DMSO), intraperitoneally (i.p.)) or APAP (80 mg/kg or 500 mg/kg in DMSO, i.p.; equivalent to a 900 mg or 5600 mg daily dose for a 70 kg human subject). BBB permeability was measured via in situ brain perfusion using [14C]sucrose and [3H]codeine, an opioid analgesic drug that is co-administered with APAP (i.e., Tylenol #3). Localization and protein expression of tight junction proteins (i.e., claudin-5, occludin, ZO-1) were studied in rat brain microvessels using Western blot analysis and confocal microscopy, respectively. Paracellular [14C]sucrose “leak” and brain [3H]codeine accumulation were significantly enhanced in rats treated with 500 mg/kg APAP only. Additionally, claudin-5 localization and protein expression were altered in brain microvessels isolated from rats administered 500 mg/kg APAP. Our novel and translational data show that BBB integrity is altered following a single high APAP dose, results that are relevant to patients abusing or misusing APAP and/or APAP/opioid combination products.


Introduction
The blood-brain barrier (BBB) is essential for maintenance and regulation of brain homeostasis. The BBB selectively and dynamically regulates solute exchange between the central nervous system (CNS) and systemic circulation and simultaneously restricts entry of harmful substances into the brain. The BBB is susceptible to disruption or modulation by stress factors including disease states (e.g., ischemic stroke [1][2][3], inflammatory pain [4][5][6], and Alzheimer's disease [7][8][9][10]), as well as the presence of circulating intrinsic regulators (e.g., hormones [11,12]) and xenobiotics including environmental toxins [13][14][15] and drugs [16,17]). For example, morphine is an inducer of ATP-binding cassette transporters at the BBB including P-glycoprotein and breast cancer resistance protein [18][19][20], which reduces blood-to-brain permeability of many drugs and is suggested to be an underlying mechanism for development of opioid tolerance following chronic use. Besides opioid analgesics, other common psychoactive substances including cocaine [21,22], nicotine [23,24], alcohol [25,26], methamphetamine [27][28][29][30] and methylenedioxymethamphetamine (MDMA, tight junction barrier proteins at the brain microvascular endothelium, thereby exacerbating CNS opioid exposure through the "leak" of codeine into the brain. To test this hypothesis, we studied, in vivo, paracellular solute "leak" and localization/expression of critical tight junction proteins at the BBB (i.e., claudin-5, occludin and ZO-1) following APAP administration. Our work demonstrated, for the first time, that an acute high dosage (500 mg/kg) of APAP increases paracellular leak of the BBB to both sucrose (i.e., a vascular marker) and codeine, a commonly prescribed opioid that is concomitantly administered with APAP. Of particular significance, low-dose APAP (80 mg/kg) did not have significant effects on BBB permeability to sucrose or codeine. In addition to BBB permeability changes in response to high-dose APAP, we showed that tight junction protein complex disruption by acute APAP exposure is characterized by altered expression of claudin-5 in rat brain microvessels.

Animals and Treatments
Animal protocols were approved by the University of Arizona Institutional Animal Care and Use Committee (Protocol #18-377; Approval Date: 25 February 2021) and were conducted in compliance with both National Institutes of Health and Animal Research: Reporting In Vivo Experiments (ARRIVE) guidelines. Female Sprague-Dawley rats were purchased from Envigo (Madison, WI, USA). At the time of experimentation, rats were 3-4 months old with body weights of 200-250 g. We purposely focused our study on female experimental animals to enable robust comparison with our previous study on APAP effects at the BBB [52]. Animals were housed under controlled conditions (22.2-22.4 • C; 50% relative humidity; 12 h light/dark cycle) with free access to food and water for a minimum of seven days. For low-and high-dose drug treatments groups, APAP (Millipore-Sigma, St. Louis, MO, USA) was dissolved in vehicle (100% dimethyl sulfoxide; DMSO) to achieve doses of 80 mg/mL and 500 mg/mL, respectively. Animals received a single intraperitoneal (i.p.) injection of either APAP or vehicle (1 mL/kg). Three hours after treatment and prior to further experimentation, animals were anesthetized with 100 mg/mL ketamine with 10 mg/mL xylazine (i.p.).

In Situ Brain Perfusion
In situ brain perfusion was performed as described previously by our laboratory [53,54]. Briefly, after anesthesia, experimental animals (n = 6) were given an i.p. dose of heparin (10,000 U/kg) to prevent coagulation. Common carotid arteries were exposed and bilaterally canulated to connect to the perfusion circuit. Both jugular veins were severed to provide drainage. The perfusion buffer (117 mM NaCl, 4.7 mM KCl, 0.8 mM MgSO 4 , 1.2 mM KH 2 PO 4 , 2.5 mM CaCl 2 , 10 mM d-glucose, 3.9% dextran (75,000 g/mol), and 1.0 g/L bovine serum albumin; pH 7.4) was warmed to 37 • C and oxygenated with 95% O 2 and 5% CO 2 . Evan's blue dye was added to the perfusion buffer as a visual indicator for tight junction intactness. Rat brains were perfused for 10 min at a total flow rate of 3.6 mL per min under a perfusion pressure of 95 to 105 mmHg. In control experiments, we monitored EKG and respiratory waveforms in animals subjected to in situ brain perfusion for up to 30 min. In all animals, these physiological parameters remained within normal limits, which implies that our in situ brain perfusion method allows for evaluation of BBB integrity and permeability in a stable, well-controlled environment for the entire duration of the perfusion. For measurement of the BBB paracellular leak, [ 14 C]sucrose (Specific Activity = 0.5000 mCi/mL; PerkinElmer Life and Analytical Sciences, Boston, MA, USA) was used as a vascular marker and infused into the perfusate at 0.5 mL per min using a slow-drive syringe pump (Harvard Apparatus, Cambridge, MA, USA). For the measurement of cerebral exposure to opioids, [ 3 H]codeine (Specific Activity = 2.0000 mCi/mL; Research Triangle Institute, Research Triangle, NC, USA) was infused into the perfusion circuit with identical settings to the sucrose experiments. Immediately after perfusion, rat brains were extracted and processed by removing cerebellum, meninges and choroid plexus. The processed brains were divided into three parts and solubilized for two days using 1 mL TS2 tissue solubilizer. At this time, a 2 mL Optiphase SuperMix liquid scintillation cocktail (PerkinElmer Life and Analytical Sciences) was added to each tube to enable the measurement of radioactivity and 100 µL 30% (v/v) glacial acetic acid was added to quench background counts. Radioactivity was measured with a 1450 Liquid Scintillation and Luminescence Counter (PerkinElmer Life and Analytical Sciences) and reported as brainto-perfusate radioactivity ratios (Rbr %; pmol/mg brain tissue) by dividing the measured amount of radioisotope in brain per brain weight by the known amount of radioisotope in the perfusate: RBr (%; pmol/mg brain tissue) = C Brain /C Perfusate × 100% (1) The brain vascular volume in rats has been previously shown to range between 6 and 9 µL/g of brain tissue in perfusion studies utilizing a saline-based bicarbonate buffer [55]. Since brain tissue was processed immediately after perfusion with radiolabeled substrate, all uptake values obtained for [ 14 C]sucrose or [ 3 H]codeine required correction for brain vascular volume (i.e., 8.0 µL/g brain tissue as calculated from data reported by Takasato and colleagues [55]).

Microvessel Isolation
Rat brain microvessels were isolated according to a protocol developed and published by our laboratory [56]. Briefly, animals (n = 9) were euthanized with ketamine/xylazine and decapitated. Brains were removed, processed (i.e., removal of cerebellum, meninges and choroid plexus), and homogenized in ice-cold brain microvessel buffer (pH 7.4) with 0.1% protease inhibitor cocktail (Millipore-Sigma, St. Louis, MO, USA). After thoroughly mixing the brain homogenate with 26% dextran (75,000 g/mol), samples were centrifuged at 6500× g at 4 • C for 30 min. Following centrifugation, the supernatant was aspirated, and pellets were resuspended in the same buffer, mixed with 26% dextran, and were once again centrifuged for 30 min under the same conditions. The microvessel-enriched pellets from the second centrifugation were resuspended in Pierce™ IP Lysis Buffer (Thermo Fisher Scientific, Waltham, MA, USA) with cOmplete™ Mini Protease Inhibitor Cocktail (Roche, Basel, Switzerland), and Phosphatase Inhibitor Cocktail II and III (Research Products International) and stored at −80 • C until further analysis.

Confocal Microscopy
Confocal microscopy was performed on a Leica SP8 confocal microscope (Leica Biosystems, Wetzlar, Germany) with 488 and 552 nm excitation lasers. Emitted light was detected with a Leica hybrid detector (HyD). Images were acquired from randomly chosen areas of the somatosensory cortex between 2 mm rostral to 1 mm caudal relative to Bregma. Images from control and treated animals were acquired using matching laser power and gain settings with emission windows set to prevent bleed-through between fluorophores. Adjustments for brightness and contrast levels were performed with ImageJ software (NIH) in an identical manner for both control and treated images.

Statistical Analysis
In situ brain perfusion brain-to-perfusate radioactivity ratios (RBr %) and Western blot densitometric analysis data (normalized to loading control) were reported as mean ± S.D. Statistically significant differences between control and treatment groups were determined with one-way analysis of variance (ANOVA) followed by post hoc two-tailed unpaired homoscedastic Student's t-test to examine differences between groups. A value of p < 0.05 was considered to be statistically significant.

Increased BBB Paracellular "Leak" in Response to High-Dose APAP Treatment
To investigate whether there is a change in BBB paracellular permeability (i.e., "leak") in the presence of APAP, radiolabeled sucrose (i.e., [ 14 C]sucrose) was used. We administered 80 mg/kg of APAP, a low dose emulating approximately 900 mg for an average-weight human adult (i.e., 70 kg), and 500 mg/kg of APAP, an acute dosage equivalent to approximately 5600 mg for an average-weight human adult (i.e., 70 kg), for the low-and high-dose treatments, respectively, 3 h prior to in situ brain perfusion with [ 14 C]sucrose. As shown in Figure 1A, only the high-dose APAP treatment resulted in significantly elevated codeine and sucrose radioactivity in perfused rat brains, which indicates disruption of BBB functional integrity leading to paracellular leak. tered 80 mg/kg of APAP, a low dose emulating approximately 900 mg for an averageweight human adult (i.e., 70 kg), and 500 mg/kg of APAP, an acute dosage equivalent to approximately 5600 mg for an average-weight human adult (i.e., 70 kg), for the low-and high-dose treatments, respectively, 3 h prior to in situ brain perfusion with [ 14 C]sucrose. As shown in Figure 1A, only the high-dose APAP treatment resulted in significantly elevated codeine and sucrose radioactivity in perfused rat brains, which indicates disruption of BBB functional integrity leading to paracellular leak. In situ brain perfusion with [ 14 C]sucrose as a vascular paracellular permeability marker shows significantly elevated radioactivity level represented in brain-to-perfusate radioactivity ratios (RBr %) in brains of high-dose APAP-treated rats (n = 6) in comparison to vehicle or low-dose APAP-treated rats. No significant change radioactivity level was measured in the low-dose APAP (80 mg/kg) treated rats. (B): In situ perfusion with [ 3 H]codeine following APAP injection shows significantly higher radioactivity represented in brain-toperfusate radioactivity ratios (RBr %) in brains of APAP (500 mg/kg)-treated rats (n = 6) in comparison to vehicle. Data are expressed as mean ± S.D. (** p < 0.01; *** p < 0.001).

Elevated CNS Exposure to Codeine after APAP Treatment
To investigate whether the increase in BBB paracellular permeability due to APAP stress contributes to elevated CNS exposure to drugs, we utilized in situ brain perfusion to measure [ 3 H]codeine uptake into brain tissue 3 h following APAP injection (80 mg/kg or 500 mg/kg; i.p.). Codeine was purposely selected for these experiments since it is commonly co-administered with APAP in combination products including, but not limited to, Tylenol #3 ® . It is also an opioid analgesic drug that accesses the CNS primarily by passive diffusion and not by facilitated transport processes [54,57], which renders it an ideal drug to use in the evaluation of altered BBB functional integrity by APAP. Codeine radioactivity levels measured in brain tissue of high-dose APAP (500 mg/kg)-treated rats were significantly increased as compared to those of low-dose APAP (80 mg/kg) or vehicle controls ( Figure 1B). This indicates that concomitantly administered medications such as high-dose APAP and codeine can interact at the BBB, thereby causing a serious risk of adverse effects associated with the codeine due to the elevated brain penetration of this opioid analgesic drug in the presence of a high dose of APAP.

Increased Claudin-5 Expression in Brain Microvessel after High-Dose APAP Treatment
Since claudin-5 is considered to be the primary determinant of BBB tight junction integrity [17], we measured changes in claudin-5 content in rat brain microvessels 3 h after In situ brain perfusion with [ 14 C]sucrose as a vascular paracellular permeability marker shows significantly elevated radioactivity level represented in brain-to-perfusate radioactivity ratios (RBr %) in brains of high-dose APAP-treated rats (n = 6) in comparison to vehicle or low-dose APAP-treated rats. No significant change radioactivity level was measured in the low-dose APAP (80 mg/kg) treated rats. (B): In situ perfusion with [ 3 H]codeine following APAP injection shows significantly higher radioactivity represented in brain-to-perfusate radioactivity ratios (RBr %) in brains of APAP (500 mg/kg)-treated rats (n = 6) in comparison to vehicle. Data are expressed as mean ± S.D. (** p < 0.01; *** p < 0.001).

Elevated CNS Exposure to Codeine after APAP Treatment
To investigate whether the increase in BBB paracellular permeability due to APAP stress contributes to elevated CNS exposure to drugs, we utilized in situ brain perfusion to measure [ 3 H]codeine uptake into brain tissue 3 h following APAP injection (80 mg/kg or 500 mg/kg; i.p.). Codeine was purposely selected for these experiments since it is commonly co-administered with APAP in combination products including, but not limited to, Tylenol #3 ® . It is also an opioid analgesic drug that accesses the CNS primarily by passive diffusion and not by facilitated transport processes [54,57], which renders it an ideal drug to use in the evaluation of altered BBB functional integrity by APAP. Codeine radioactivity levels measured in brain tissue of high-dose APAP (500 mg/kg)-treated rats were significantly increased as compared to those of low-dose APAP (80 mg/kg) or vehicle controls ( Figure 1B). This indicates that concomitantly administered medications such as high-dose APAP and codeine can interact at the BBB, thereby causing a serious risk of adverse effects associated with the codeine due to the elevated brain penetration of this opioid analgesic drug in the presence of a high dose of APAP.

Increased Claudin-5 Expression in Brain Microvessel after High-Dose APAP Treatment
Since claudin-5 is considered to be the primary determinant of BBB tight junction integrity [17], we measured changes in claudin-5 content in rat brain microvessels 3 h after an acute low dose (80 mg/kg; i.p.) or high dose (500 mg/kg; i.p.) of APAP in accordance with our in situ brain perfusion experiment. Western blot analysis of microvessel samples indicated a significant elevation in claudin-5 expression in response to high-dose APAP treatment as compared to low-dose APAP (p < 0.0001) or vehicle-treated rats (p < 0.0001) (Figure 2A). Both a cross-section ( Figure 2B) and longitudinal section ( Figure 2C) of a rat brain microvessel show localization of claudin-5 at endothelial cell margins, evidence for its role as a critical constituent of tight junction protein complexes. Confocal imaging of microvessels in rat brain cryosections corroborated with our Western blot data, demonstrating an overall increase in claudin-5 staining throughout cortices of rats three hours after high-dose APAP treatment as compared to vehicle ( Figure 2D). treatment as compared to low-dose APAP (p < 0.0001) or vehicle-treated rats (p < 0.0001) (Figure 2A). Both a cross-section ( Figure 2B) and longitudinal section ( Figure 2C) of a rat brain microvessel show localization of claudin-5 at endothelial cell margins, evidence for its role as a critical constituent of tight junction protein complexes. Confocal imaging of microvessels in rat brain cryosections corroborated with our Western blot data, demonstrating an overall increase in claudin-5 staining throughout cortices of rats three hours after high-dose APAP treatment as compared to vehicle ( Figure 2D).

Figure 2. Claudin-5 level increases in BBB microvessel endothelial cells three hours after high-dose APAP treatment. (A) Western blot (top) and densitometric analysis (bottom) of claudin-5 levels in
isolated brain microvessel of rats (n = 9) treated with low-dose APAP (80 mg/kg) and high-dose APAP (500 mg/kg) shows statistical significance between the three groups (p = 0.000012), as well as between vehicle and the high-dose groups (p = 0.00018; Student's t-test) and between low-dose and high-dose groups (p = 0.000074). No statistical significance was detected between the vehicle and the low-dose APAP group (p = 0.84). Densitometric data are expressed as mean ± S.D. (** p < 0.0001). and densitometric analysis (bottom) of claudin-5 levels in isolated brain microvessel of rats (n = 9) treated with low-dose APAP (80 mg/kg) and high-dose APAP (500 mg/kg) shows statistical significance between the three groups (p = 0.000012), as well as between vehicle and the high-dose groups (p = 0.00018; Student's t-test) and between low-dose and high-dose groups (p = 0.000074). No statistical significance was detected between the vehicle and the low-dose APAP group (p = 0.84). Densitometric data are expressed as mean ± S.D. (** p < 0.0001). High magnification confocal microscopy images of a cross-section (B) and a longitudinal section (C) of a cortical microvessel showing claudin-5 localization (green) within the microvessel labeled with the vascular marker tomato lectin (red). (D) confocal microscopy images indicate higher overall claudin-5 expression in cortical microvessel of rats treated with high-dose APAP (500 mg/kg) than vehicle, but not in those treated with low-dose APAP (80 mg/kg). Arrows show sites of claudin-5 upregulation. Microvessels were stained with lectin (red) and are shown in the inset to demonstrate vascular localization of claudin-5. Scale bar = 20 µm.

High magnification confocal microscopy images of a cross-section (B) and a longitudinal section (C) of a cortical microvessel showing claudin-5 localization (green) within the microvessel labeled with the vascular marker tomato lectin (red). (D) confocal microscopy images indicate higher overall
We also evaluated APAP effects on other tight junction proteins including the transmembrane protein occludin and the intracellular accessory protein zonula occludens-1 (ZO-1). In contrast to our results with claudin-5, Western blot analysis and confocal microscopy of occludin ( Figure 3A,C) and ZO-1 ( Figure 3B,D) in cortical microvessels showed no detectable change in either APAP-treated or untreated rats at the three-hour timepoint. Taken together, these observations suggest that the changes in claudin-5 is likely involved in the process leading up to the observed BBB paracellular "leak" in response to high-dose APAP treatment, while occludin and ZO-1 are likely not responsible for the disruption of BBB tight junctions in the acute phase of the high-dose APAP stress response.
upregulation. Microvessels were stained with lectin (red) and are shown in the inset to demonstrate vascular localization of claudin-5. Scale bar = 20 μm.
We also evaluated APAP effects on other tight junction proteins including the transmembrane protein occludin and the intracellular accessory protein zonula occludens-1 (ZO-1). In contrast to our results with claudin-5, Western blot analysis and confocal microscopy of occludin ( Figure 3A,C) and ZO-1 ( Figure 3B,D) in cortical microvessels showed no detectable change in either APAP-treated or untreated rats at the three-hour timepoint. Taken together, these observations suggest that the changes in claudin-5 is likely involved in the process leading up to the observed BBB paracellular "leak" in response to high-dose APAP treatment, while occludin and ZO-1 are likely not responsible for the disruption of BBB tight junctions in the acute phase of the high-dose APAP stress response. Western blot (top) and densitometric analysis (bottom) shows comparable occludin levels in isolated brain microvessel of APAP (500 mg/mL)-treated and control rats (n = 9). Densitometric data are expressed as mean ± S.D. (B) Western blot (top) and densitometric analysis (bottom) shows comparable ZO-1 levels in isolated brain microvessel of APAP (500 mg/mL)-treated and control rats (n = 9). Densitometric data are expressed as mean ± S.D. (C) Confocal microscopy shows similar overall occludin expression in cortical microvessel of rats treated with APAP (80 or 500 mg/kg) as

Discussion
The BBB plays an essential role in regulation of brain homeostasis and preservation of the optimal microenvironment for proper neuronal function. Dysregulation of BBB permeability is associated with many pathologies [17] and can lead to increased vulnerability (i.e., leak) of the brain to harmful substances in the systemic circulation [52,54,[57][58][59][60][61][62]. For example, peripheral inflammatory pain can cause BBB dysfunction that is manifested by increased paracellular "leak" to circulating small molecule solutes including drugs [54]. Such enhancement in paracellular permeability was shown to result from modulation of tight junction protein complexes at the brain microvascular endothelium. Claudins, as well as occludin, are critical transmembrane tight junction proteins at the BBB that form protein complexes responsible for physically sealing paracellular gaps between adjacent endothelial cells [17,63]. Changes to protein expression levels of claudins and occludin are frequently observed in modulation of stress responses by tight junctions at the BBB [17]. ZO-1 is an important intracellular accessory protein at the tight junction that links transmembrane proteins to the actin cytoskeleton, thus forming complex networks essential for the dynamic regulation of the BBB [17]. In the present study, we are the first to demonstrate that acute treatment of APAP disrupts tight junction protein complexes in the cerebral microvasculature, as evidenced by an increase in paracellular permeability at the BBB and an upregulation of claudin-5. We believe these changes are due specifically to APAP administration rather than the DMSO solvent. Our previous work has shown that BBB permeability to sucrose or small molecule drugs in control (i.e., untreated animals) is not altered in vivo following acute (i.e., up to 3 h) administration of DMSO vehicle [53,64]. Preliminary time-point experiments showed that this increase in claudin-5 levels was sustained until 6 h after APAP treatment and returned to baseline by 12 h (data not shown). Additionally, as increased levels of claudin-5 in the systemic circulation have been implicated in BBB breakdown [65,66], we conducted an ELISA analysis of rat plasma that detected a slight but statistically significant (p = 0.03) increase in plasma claudin-5 protein levels at 3 h following APAP administration (data not shown). In contrast, occludin and ZO-1 were not affected by APAP treatment, indicating that claudin-5 is likely the primary tight junction protein that controls the response to high-dose APAP-induced stress and, by extension, enhanced paracellular "leak" associated with APAP administration.
Although a simultaneous increase in both claudin-5 expression and BBB permeability may seem paradoxical, it is essential to note that many publications in the scientific literature have reported that claudin-5 upregulation is associated with an increase in barrier permeability or neuronal injury due to pathophysiological stressors. For example, studies in the kainic acid model of temporal lobe epilepsy demonstrated a significant increase in claudin-5 levels 24-72 h after kainic acid injection in correspondence with neuronal injury [67]. Our laboratory's studies in a rat model of peripheral inflammatory pain showed a 6-fold increase in microvessel claudin-5 levels with BBB paracellular "leak" but no detectable change in claudin-5 localization [6]. Further investigation into the underlying mechanism revealed that transforming growth factor (TGF)-β-signaling mediated this disruption of the BBB that was manifested by sucrose extravasation and claudin-5 upregulation at the brain microvascular endothelium [53]. Claudin-5 is also elevated in epithelial tight junctions in lung tissue isolated from alcoholic rats with diminished barrier functions, likely a result of disrupted protein-protein interactions between claudins [68]. Interestingly, overexpression of claudins 1 and 3 in stably transfected NIH/3T3 or IB3.1 cells leads to a decrease in permeability to dextrans but overexpression of claudin-5 in these same cells results in an increase in permeability to dextrans [69]. Additionally, it is essential to consider that changes in monomeric tight junction protein expression following exposure to a pathological or pharmacological stressor may reflect altered tight junction integrity via collapse of oligomeric structures. Indeed, claudin-5 has been shown to form oligomers in HEK293 cells transfected with native claudin-5 [70] or chimeric claudin-5 proteins [71]. The ability of claudin-5 to assemble into higher order structures was proposed to be required for the formation of tight junction strands and, subsequently, the sealing function of tight junction protein complexes in endothelial cells [70][71][72]. These findings highlight the fact that elevated claudin-5 protein expression in brain microvascular endothelial cells is not necessarily indicative of improved neurovascular integrity. Rather, any change in claudin-5 expression, regardless of directionality, can be reflective of tight junction dysregulation and paracellular leak, depending on the specific pathological or mechanistic processes involved and the resulted state of functional intact nature of tight junction assembly. Future work by our group will elucidate the mechanism of action pertaining to the upregulation of claudin-5 in response to APAP treatment.
APAP may also affect BBB permeability through interactions with phospholipids in the plasma membrane. APAP directly binds to phospholipids and significantly increases membrane fluidity in a dose-dependent manner [73]. Alterations in physicochemical properties of the plasma membrane are likely to alter the function of the tight junction protein complex by dynamically affecting the organization, oligomerization, and structure of the proteins comprising the tight junction at the BBB [74][75][76]. These changes may affect paracellular permeability at the BBB independent of changes in expression levels of individual proteins comprising the TJ complex.
In pharmacological studies, BBB disruption is frequently studied as a clinical goal for optimization of drug delivery into the CNS, while its role as an adverse effect of pharmaceutical agents, polypharmacy, and risk factor for brain microenvironment dysregulation is frequently overlooked. APAP is an extraordinarily effective and useful substance used in hundreds of combination medications and consumed by millions of individuals each day in the United States alone [38]. Contrary to APAP popularity, there is limited knowledge on effects of APAP and its potential for drug-drug interactions at the BBB. Previous studies indicate that, while APAP has been shown to preserve brain endothelial cell survival and reduce the neurovascular inflammatory response under oxidative stress in vitro and at low doses in vivo [77,78], APAP at a higher dose (>200 mg/kg) induces cortical oxidative stress and produces reactive astrocytosis without achieving acute liver injury [50,78]. As another example, one prominent site for potential drug interactions at the BBB is P-glycoprotein, a critical efflux transporter at the BBB responsible for restricting entrance of a wide range of substances into the CNS. Our laboratory has previously shown that APAP at 500 mg/kg significantly increases the expression level of P-glycoprotein in brain microvessels and thereby reduces morphine uptake and antinociception [52]. Of particular significance, we demonstrated that APAP activated endothelial signaling pathways such as the constitutive androstane receptor, a nuclear receptor known to be involved in the induction of drug metabolizing enzymes and transporters [52]. In the present study, we further demonstrate that a high dose of APAP elicits drug-drug interactions at the BBB by increasing paracellular permeability and, as a result, can also increase CNS exposure to concomitantly administered drugs that are not transport substrates for P-glycoprotein (i.e., codeine). Importantly, we found that APAP administered at a dose considered safe and effective for analgesia did not alter BBB integrity. Taken together with our previous work, these data imply that an acute, high dose of APAP can have profound effects on cerebral microvascular homeostasis and contribute to unexpected changes in CNS drug disposition.
In addition to drug exposures, the BBB is susceptible to modulation by various physiological stress stimuli. As APAP is the most prevalent and most used analgesic on the market, it is important to note that inflammatory pain by itself could damage tight junction integrity at the BBB. We have previously shown that the induction of inflammatory pain in the hind paw with several different agents (i.e., λ-carrageenan, complete Freund's adjuvant, formalin) also increases paracellular "leak" of the BBB to sucrose and codeine. Associated with these changes in the λ-carrageenan model are alterations to the tight junction proteins claudin-5, occludin, and ZO-1 as well as the cytoskeletal scaffolding protein, actin [4][5][6]. Studies with the λ-carrageenan model further illustrated prolonged BBB tight junction injury, most prominently at the 3 and 48 h timepoints, which correlated with increased CNS codeine uptake and enhanced codeine analgesia [57]. Given that APAP is commonly used in the presence of pain, it is of our interest to further investigate the effect of APAP in inflammatory pain models and understand their concurrent effects on the function of the BBB.
In conclusion, our novel data show that high-dose APAP increases paracellular permeability of the BBB, an effect that is correlated with increased protein expression of claudin-5 in brain microvessels, an effect that indicates dysregulation of tight junction assembly. The APAP-induced paracellular "leak" contributes to higher CNS exposure to codeine. As noted, many human subjects regularly consume excessive amounts of APAP, including doses greater than or equal to those used in this study on rats, on a daily basis in order to manage acute or chronic pain [43]. Our data suggest that further investigation into effects of APAP on BBB integrity at both the molecular and functional level is warranted. Such studies are likely to yield paradigm-shifting findings that will lead to safer prescribing of APAP and improved formulation of APAP-containing combination products to lower occurrence of accidental overdose of concomitantly administered drugs such as opioids.
Supplementary Materials: The following supporting information can be downloaded at: https:// www.mdpi.com/article/10.3390/pharmaceutics14050949/s1, Figure S1: Western blot and confocal microscopy demonstrate antibody specificity.  Data Availability Statement: If reasonably requested or needed, data and samples/models will be made available for sharing to qualified parties provided that such a request does not compromise intellectual property interests, interfere with publication, or betray confidentiality. Data that are shared will include standards and notations required to accurately interpret the data, following commonly accepted practices in the field. Data and samples/materials will be available for access and sharing as soon as reasonably possible and no longer than two years after acquisition of the data.

Conflicts of Interest:
The authors have no competing interests to declare.