Abstract
Repurposing doxycycline for the treatment of amyloidosis has recently been put forward because of the antiaggregating and anti-inflammatory properties of the drug. Most of the investigations of the therapeutic potential of doxycycline for neurodegenerative amyloidosis, e.g., prion and Alzheimer disease (AD), have been carried out in mouse models, but surprisingly no data are available regarding the concentrations reached in the brain after systemic administration. We filled this gap by analyzing the pharmacokinetic profile of doxycycline in plasma and brain after single and repeated intraperitoneal injections of 10 and 100 mg/kg, in wild-type mice and the APP23 mouse model of AD. The main outcomes of our study are: 1) Peak plasma concentrations ranged from 2 to10 μg/ml, superimposable to those in humans; 2) brain-to-plasma ratio was ∼0.2, comparable to the cerebrospinal fluid/serum ratios in humans; 3) brain Cmax 4–6 hours after a single dose was ∼0.5 (10 mg/kg) and ∼5 μM (100 mg/kg). Notably, these concentrations are lower than those required for the drug’s antiaggregating properties as observed in cell-free studies, suggesting that other features underlie the positive cognitive effects in AD mice; 4) elimination half-life was shorter than in humans (3–6 vs. 15–30 hours), therefore no significant accumulation was observed in mouse brain following repeated treatments; and 5) there were no differences between doxycycline concentrations in brain areas of age-matched wild-type and APP23 mice. These data are useful for planning preclinical studies with translational validity, and to identify more reliably the mechanism(s) of action underlying the central in vivo effects of doxycycline.
Introduction
Tetracyclines are a well known class of broad-spectrum antibiotics discovered in the late 1940s (Chopra and Roberts, 2001). More recent, extensive evidence suggests other interesting properties of these drugs, including antiapoptotic, anti-inflammatory, and antiaggregating activities (Griffin et al., 2010; Stoilova et al., 2013), which has prompted their repurposing to treat amyloidosis, a group of diseases caused by misfolding and abnormal aggregation of specific proteins that eventually accumulate as insoluble amyloid deposits (Chiti and Dobson, 2006). Doxycycline is a tetracycline derivative widely studied in these conditions.
Doxycycline disrupts transthyretin amyloid in vitro (Cardoso et al., 2003) and in vivo in animal models (Cardoso and Saraiva, 2006); in a phase 2 clinical trial, doxycycline plus tauroursodeoxycholic acid stabilized the disease in the majority of patients with hereditary transthyretin amyloidosis (Obici et al., 2012). In vitro and in vivo preclinical data also showed positive effects of doxycycline on amyloid light chain (AL) amyloidosis (Ward et al., 2011); this was followed by a report that the supplementation with doxycycline of standard chemotherapy in AL patients reduced early cardiac mortality (Wechalekar and Whelan, 2017). Doxycycline also inhibits β2-microglobulin (β2M) fibrillogenesis (Giorgetti et al., 2011), prompting the proposal of “off-label” treatment of a severe form of dialysis-related β2M amyloidosis in three patients who displayed a significant reduction in articular pain (Montagna et al., 2013). The use of doxycycline as an orphan drug for the treatment of hereditary amyloid polyneuropathy caused by β2M was recently approved by the European Committee for Orphan Medicinal Products (http://www.emea.europa.eu/docs/en_GB/document_library/Orphan_designation/2012/05/WC500127736.pdf). Doxycycline might also have therapeutic potential for neurodegenerative amyloidosis, such as in prion diseases and Alzheimer disease (AD), according to in vitro and animal data (Forloni et al., 2001, 2002; De Luigi et al., 2008; Diomede et al., 2010). A recent phase 2 study in Creutzfeldt-Jakob disease (CJD) patients at an early stage demonstrated the superiority of doxycycline over control (Varges et al., 2017), contrasting with the negative results of an earlier study (Haik et al., 2014) carried out however in CJD patients at a very late stage. Contrasting results were also obtained in clinical trials with doxycycline in AD patients (Loeb et al., 2004; Molloy et al., 2013). A preventive clinical trial in patients with fatal familial insomnia, a genetic prion disease, is currently ongoing (Forloni et al., 2015).
Different variables affect the in vivo activity of doxycycline for treating amyloidosis, including the stage of the disease when treatment starts, and whether the treatment regimen can achieve adequate drug concentrations at the active site (e.g., in the brain for neurodegenerative amyloidosis).
Many of the main pharmacokinetic (PK) features of doxycycline in humans have already been investigated in detail (Cunha et al., 1982; Saivin and Houin, 1988): At the usual oral doses of 100–200 mg/day, it is absorbed rapidly and almost completely, with peak serum concentrations from 1.7 to 5.9 μg/ml 2–3 hours after dosing, and elimination half-life of 15–30 hours (Saivin and Houin, 1988; Binh et al., 2009; Montagna et al., 2013). After long-term administration of 100 mg/day, concentrations at steady-state reach 0.7–1.5 μg/ml (Binh et al., 2009; Montagna et al., 2013). Plasma protein binding of doxycycline is 80%–90% (Cunha et al., 1982; Saivin and Houin, 1988) and there is no significant metabolism (Yim et al., 1985; Saivin and Houin, 1988). Because of its lipophilicity—much higher than that of tetracycline—doxycycline easily penetrates and distributes within body tissues. However, very few data are available on its ability to cross the blood-brain barrier (BBB): In patients with neurosyphilis (Yim et al., 1985) or suspected tick-borne neuroborreliosis (Dotevall and Hagberg, 1989), cerebrospinal fluid concentrations were 15%–25% of serum levels. Brain concentrations of doxycycline were also measured in autopsy samples from late-stage CJD patients given 100 mg daily (Haik et al., 2014): The drug crossed the BBB and persisted in the brain for days after the end of treatment, possibly because of its ability to bind the prion protein aggregates present in the brain of CJD patients.
A more detailed analysis of BBB passage can be obtained in animal models, particularly rodents, which also offer the possibility of correlating a pharmacological effect (e.g., on amyloid load or neuroprotection) with the actual brain concentrations. When planning treatment schedules in animals, one must also take into account that the drug’s PK profile varies as a function of body weight (Boxenbaum, 1982).
Very few reports describe the PK of doxycycline in rodents. The half-time of elimination from serum is 3–4 hours (Schach won Wittenau et al., 1972; Bocker et al., 1981), i.e., significantly faster than in humans; the brain-to-plasma concentration ratio was 0.31 at a single time point (4 hours) after intravenous injection of 25 mg/kg in rats (Colovic and Caccia, 2003). To our knowledge, no data are available on BBB passage in mice.
Given the paucity of published data and the importance of this information for planning preclinical studies with translational validity, we carried out ad hoc PK studies in mice, measuring plasma and brain levels of doxycycline after single and repeated intraperitoneal injections. We also did a single-dose study to measure plasma and brain levels of doxycycline in APP23 mice, a transgenic model of AD. The analytical method, employing high-performance liquid chromatography (HPLC)–tandem mass spectrometry (MS/MS), was developed and validated according to accepted guidelines.
Materials and Methods
Mice
Three groups of mice were used: 1) 7-week-old, male C57Bl/6 CRI mice (Charles River Laboratories Italia, Calco, Italy), 2) 20-month-old, female and male APP23 transgenic (Tg) mice, and 3) sex- and age-matched wild-type (WT) female and male littermates. Mice were housed three to four per cage at constant room temperature (21 ± 1°C) and relative humidity (60%) with a 12-hour light cycle (lights on 7:00 AM to 7:00 PM) with food and water ad libitum (Global Diet 2018S; Envigo, Somerset, NJ). Procedures involving animals were conducted at the Instituto di Ricerche Farmacologiche “Mario Negri” (IRCCS), which adheres to the principles set out in the following laws, regulations, and policies governing the care and use of laboratory animals: Italian Governing Law (D.lgs 26/2014; Authorization n.19/2008-A issued March 6, 2008 by Ministry of Health); Mario Negri Institutional Regulations and Policies providing internal authorization for persons conducting animal experiments which includes ad hoc members for ethical issues (authorization code 1/05-D) (Quality Management System Certificate—UNI EN ISO 9001:2008—Reg. No 6121). The NIH Guide for the Care and Use of Laboratory Animals (2011 edition), and EU directives and guidelines (EEC Council Directive 2010/63/UE). The Statement of Compliance (Assurance) with the Public Health Service (PHS) Policy on Human Care and Use of Laboratory Animals has been recently reviewed (9/9/2014) and expired on September 30, 2019 (Animal Welfare Assurance no. A5023-01).
Drug Treatments
Doxycycline hyclate HCl (MilliporeSigma, St. Louis, MO) was dissolved in sterile 0.9% saline at 10 and 100 mg/ml (free base) and injected intraperitoneally in volumes of 10 ml/kg to have final doses of 10 and 100 mg/kg.
Single Treatment in 7-Week-Old Male C57Bl/6 Mice.
Mice weighing 24.4 ± 1.5 g were treated with 10 and 100 mg/kg (36 mice for each dose) and killed by decapitation at 0.5, 1, 2 4, 6, 8, 10, 20, and 24 hours after the injection (4/time point/dose).
Repeated Treatment in 7-Week-Old Male C57Bl/6 Mice.
Mice weighing 25.3 ± 1.3 g were given 10 and 100 mg/kg doxycycline (24 mice/dose) once a day (9:00 AM) for four injections (12) or twice a day (17:00 and 9:00) for eight injections (12). Mice were killed just before, or 2 and 6 hours after, the last injection (4 mice/time point).
Single Treatment in 20-Month-Old Male and Female WT Mice.
Mice weighing 39.8 ± 10.8 g were treated with 100 mg/kg and killed by decapitation at 2, 6, 8, 10, 16, and 24 hours after the injection (n = 3–6/time point).
Repeated Treatments in 20-Month-Old Male and Female APP23 and WT Mice.
WT and APP23 mice weighing 33.0 ± 9.8 g were given 100 mg/kg (three for group) once a day for four consecutive days, and killed by decapitation 24 hours after the last injection.
Blood samples were transferred to heparinized tubes and kept in ice until centrifugation at 2000g for 15 minutes (4°C), to obtain plasma that was stored at −80°C until HPLC-MS/MS analysis. Immediately after death brain tissues were removed, dissected into two halves, and kept on dry ice before storage at −80°C. The day of analysis, the brain tissues and dissected cerebral areas were homogenized in phosphate buffer 0.01 M pH 7.4 (1 g in 6 ml) with an Ultra Turrax T10 basic (IKA-Werke GmbH & Co., Staufen, DE) and processed as described below.
Doxycycline Concentrations in Plasma and Brain
The analytical method and procedures for its validation are described fully in the Supplemental Material. Briefly, after addition of internal standard (IS, demeclocycline), 50 μl of plasma or 200 μl brain homogenate were purified by protein precipitation with five volumes of cold acetonitrile (with 0.1% of formic acid), and centrifuged. The supernatants were evaporated under a nitrogen flow, and the residues resuspended in 0.1% HCOOH in water/acetonitrile (98/2, v/v) and injected into the HPLC-MS/MS. Separation was done on a Kinetex EVO C18 column with HCOOH 0.1% in water (mobile phase A, MP-A) and acetonitrile (mobile phase B, MP-B); elution started with 98% of MP-A held for 2 minutes followed by a 10-minute nonlinear gradient (curve 8) to 98% of MP-B. Doxycycline and the IS were acquired in positive multiple reaction mode monitoring the quantitative ion transitions mass-to-charge ratio (m/z) 445.1 → m/z 428.1 (collision energy 20 eV) and m/z 465.1 → m/z 448.1 (collision energy 15 eV), respectively.
Plasma and brain samples of treated mice were analyzed in parallel with quality control samples (two replicates at three concentrations) and with freshly prepared calibration curves linear in the range 0.1–10 μg/ml for plasma and 60–6000 ng/g for brain.
The GraphPad Prism program (GraphPad Software, Inc. La Jolla, CA), was used for plotting the calibration curves and for quantifying the unknown concentrations of doxycycline in plasma and brain. The same software was used for statistical analysis and graphics. Plasma and brain PK profiles were analyzed using a noncompartmental model for extravascular administration to obtain the main PK parameters. The peak concentration (Cmax) and the time taken to reach it (Tmax) were taken directly from the data; elimination half-life (t1/2), area under the curve from 0 to the last time point (AUC0–t) and from 0 to infinity (AUC0–inf), apparent volume of distribution (V/F), and apparent clearance (Cl/F) were obtained with PKSolver, a freely available menu-driven add-in program for Microsoft Excel (Zhang et al., 2010).
Identification and Analysis of Metabolites in Plasma and Brain
In vivo formation of doxycycline metabolites was investigated using a high-resolution mass spectrometer (Q Exactive Orbitrap; Thermo Fisher Scientific, Waltham, MA) coupled with a liquid chromatographic system (1200 series; Agilent Technologies, Santa Clara, CA). Extraction procedure from biologic matrices and HPLC-MS/MS analysis is described in the Supplemental Material.
Results
HPLC-MS/MS Method Validation
We developed an HPLC-MS/MS method and validated it following EMA guidelines (http://www.ema.europa.eu/docs/en_GB/document_library/Scientific_guideline/2011/08/WC500109686.pdf).
Analysis of plasma and brain homogenates spiked with different amounts of doxycycline (respectively, 0.1–10 μg/ml and 60–6000 ng/g) allowed evaluation of the linearity of response and limits of quantifications (0.1 μg/ml and 60 ng/g), the selectivity (i.e., the ability of the method to distinguish doxycycline and the IS from other components of the biologic matrices), the recovery and matrix effects, accuracy and precision, carryover, and analyte stability in the different experimental conditions. All these validation data are detailed in the Supplemental Material.
Pharmacokinetic Studies
Single Treatment in 7-Week-Old C57Bl/6 Mice.
Figure 1 shows the pharmacokinetic profile of doxycycline in plasma and brain of male C57Bl/6 mice, at different times after a single i.p. dose of 10 or 100 mg/kg. After the lowest dose (Fig. 1A), plasma concentration (Cmax) of 2.26 ± 0.34 μg/ml (mean ± S.D., four mice) peaked at 2 hours. The terminal phase had a half-life of 3.2 hours; concentrations were not detectable at 24 hours. AUC0–20h and AUC0–inf were similar. The analysis of brain tissues (Fig. 1A) indicated a rapid distribution of doxycycline, with quantifiable levels 30 minutes after the dose; the Cmax (0.22 ± 0.04 μg/g) was reached at 6 hours after the treatment. Brain doxycycline declined with a t1/2 of 3.9 hours. AUC0–24 hours and AUC0–inf were also closed. The ratio of AUC0–inf brain to AUC0–inf plasma was 0.17.
After the highest dose plasma Cmax was reached at 1 hour (Fig. 1B), the terminal half-life calculated on the last three points was 6.5 hours. In the brain, Cmax was reached after 4 hours and concentrations declined then with a t1/2 of 6.5 hours. At this dose, the ratio of AUC0–inf brain to AUC0–inf plasma was 0.22. The ratio of AUC0–inf (100 mg/kg) to AUC0–inf (10 mg/kg) was 8.5 in plasma and 10.6 in brain tissue, indicating dose-proportionality.
Repeated Treatment in 7-Week-Old C57Bl/6 Mice.
Data after a single intraperitoneal injection were used to simulate the expected PK profiles after repeated doses of 10 and 100 mg/kg doxycycline, once or twice a day. These simulations are shown in Fig. 2 (10 mg/kg) and Fig. 3 (100 mg/kg) for brain and in Supplemental Figures 4 and 5 for plasma, with their variability (as S.D.). We considered these simulations useful 1) to visually show that the fast elimination of doxycycline does not allow significant accumulation of the drug during repeated treatment, with the partial exception of the higher dose twice a day, and 2) to verify experimentally that the concentrations after repeated administrations are as expected, thus excluding possible changes in the PK profile during chronic treatment (e.g., in absorption or elimination mechanisms).
Figures 2 and 3 show that brain doxycycline concentrations measured just before, or 2 and 6 hours after, the last of four once-a-day injections or eight twice-a-day injections, are mostly within the expected values, and not significantly different from the concentrations after single injections, although we noticed a slight accumulation after eight twice-a-day injections with both doses. Plasma doxycycline concentrations after repeated treatment were mostly in line with expected values (Supplemental Figures 4 and 5).
Single Treatment in 20-Month-Old C57Bl/6 Mice
Figure 4 shows the PK profile of doxycycline in plasma and brain of 20-month-old C57Bl/6 mice at different times after a single i.p. dose of 100 mg/kg. The aim was to gain PK information for the best design of studies in age-matched APP23 mice; thus, this analysis included both male (n = 12) and female (n = 10) mice, comparably distributed at the different time points. No clear-cut differences were noted (data not shown), so the values for the two genders were combined.
The time courses of plasma and brain levels in these 20-month-old mice were superimposable on those of younger mice, except at the longest time points (Fig. 4). Analysis of the last four points gave t1/2 of 11.1 (plasma) and 15.5 hours (brain), about double those in younger mice. The absolute Cl/F were comparable, whereas the absolute apparent volume of distribution during terminal phase after nonintravenous injection (Vz/F) was higher in older mice (332 ml vs. 180 ml in younger mice). The brain-to-plasma ratio (AUC0–inf brain/AUC0–inf plasma) was about 20% for both ages.
Repeated Treatment in 20-Month-Old APP23 Mice.
On the basis of the data in 20-month-old WT mice, doxycycline levels were measured 24 hours after the fourth i.p. injection (1/day) of 100 mg/kg in three APP23 mice (two females and one male) and three sex- and age-matched WT mice, in several brains areas: cortex, hippocampus, cerebellum, striatum, and the “rest of the brain.” Doxycycline levels were measured at this time point on the basis of the PK profile observed in age-matched mice treated once (Fig. 4); in fact, distribution equilibrium is more probably achieved at the end of the dosing interval and, therefore, a trough value allows a better estimation of the degree of accumulation than Cmax (Rowland and Tozer, 2011). Moreover, mice were treated with the highest doxycycline dose (100 mg/kg) to have clearly measurable levels after 24 hours. Since preliminary studies in old WT mice showed some toxicity at this dose twice a day, we decided to treat APP23 mice once a day.
Brain levels of doxycycline were similar in WT and APP23 mice in all the regions considered (Table 1).
Identification and Analysis of In Vivo Doxycycline Metabolite.
Since previous data suggested minimal metabolism of doxycycline (Bocker, 1983), initial analyses were carried out in a pool of plasma and brain samples obtained from the young WT mice chronically treated with the highest dose of doxycycline. N-demethylated doxycycline was the only metabolite identified in plasma (Supplemental Figure 6), whereas no metabolites were found in the brain samples. Subsequent semiquantitative analysis in all the plasma samples of the mice treated with 100 mg/kg doxycycline (all time points after single or repeated doses, in both young and old WT mice) showed that N-demethylated metabolite never exceed 5% of the parent drug (Supplemental Figure 7) (more details can be found in the Supplemental Material).
Discussion
Preclinical studies are often limited by inadequate evaluation of whether the drug dose actually achieves plasma and/or tissue (e.g., brain) concentrations with translational validity (meaning, that result in a similar exposure), and are compatible with the desired pharmacological effect. We reassessed the PK parameters of doxycycline in mice with the main aim of providing the information needed to fine-tune preclinical studies in mouse models of neurodegenerative amyloidosis.
The peak plasma concentrations of doxycycline in the 7-week-old male C57BL76 mice, observed 1–2 hours after single intraperitoneal injections of 10 or 100 mg/kg, were 2–10 μg/ml, superimposable with those in humans taking the usual oral doses of 100–200 mg [1.7–5.9 μg/ml; (Saivin and Houin, 1988; Binh et al., 2009; Montagna et al., 2013)]. This is consistent with empirical allometric extrapolations (McCann and Ricaurte, 2001), following the principle of interspecies drug dose scaling, which suggests that a dose of 100 mg in humans should correspond approximately to a dose of 15 mg/kg in mice. A nonsignificant increase in drug concentrations appeared 6 hours after the treatment with both 10 and 100 mg/kg, in comparison with the concentrations measured after 4 hours. The presence of a second peak may be consistent with the enterohepatic recycling previously described for doxycycline in mice (Bocker at al., 1981) and tetracycline in rat (Adir, 1975); a second peak had also been reported in the plasma of patients treated with doxycycline (Fabre et al., 1966; Pedersen and Miller, 1980; Malmborg, 1984). We also found that the AUC for doxycycline in mice was proportional to the dose administered, at least within the 10–100 mg/kg range.
A slower t1/2 was observed after treatment with 100 mg/kg than after 10 mg/kg (6.5 vs. 3.2 hours, respectively). Since this is accompanied by a less-than-proportional Cmax (10 vs. 2 μg/ml with 100 and 10 mg/kg), we suggest that at the highest dose absorption became a limiting factor, a condition in which the terminal half-life also reflects rate and extent of absorption and not just the elimination process (Toutain and Bousquet-Mélou, 2004). However, plasma clearance was similar for the two doses.
There was a remarkable difference, however, in the half-time of elimination from plasma, which is 15–30 hours in humans (Saivin and Houin, 1988; Binh et al., 2009; Montagna et al., 2013) but significantly shorter in young mice [3–6 hours, in line with previous data (Schach won Wittenau et al., 1972; Bocker et al., 1981)]. Our old mice had an intermediate value of 11 hours (this is discussed below). These results have implications for treatment schedules, since a single daily dose may be justified in humans but not for mice. Doxycycline metabolism appeared negligible, confirming previous data (Bocker, 1983). In fact, plasma levels of N-demethyl-doxycycline, the only detectable metabolite, never exceed 5% of the levels of the parent drug; no metabolites were detected in brain samples of doxycycline-treated mice.
Mean maximum brain concentrations of 0.24 and 1.79 μg doxycycline/g of brain were reached within 4–6 hours after 10 and 100 mg/kg doxycycline. Interestingly, the values in mice are comparable to those in autopsy samples from CJD patients chronically treated with 100 mg/day: In individuals who received the last dose within 24 hours of death, levels were 0.6–3.0 μg/g (Haik et al., 2014).
The present study describes for the first time the brain-to-plasma ratio of doxycycline in mice, which was about 0.2 (calculated on the AUC values), after either 10 or 100 mg/kg, in both young and old mice. Assuming that cerebral blood accounts for 2% of total blood (Edvinsson et al., 1973; Modak et al., 1978; Chugh et al., 2009), it follows that brain levels cannot be accounted for by the residual cerebral blood. This value is very similar to the cerebrospinal fluid/serum ratios in patients with neurosyphilis (Yim et al., 1985) or suspected tick-borne neuroborreliosis (Dotevall and Hagberg, 1989). Assuming a mean plasma protein binding of 80%–95% (Saivin and Houin, 1988; Riond and Riviere, 1989, 1990; Riond et al., 1990; Davis et al., 2006), the concentration gradient of doxycycline—i.e., the ratio of brain to free plasma concentration—is 1 to 2 with both doses, in agreement with that measured in rats (Colovic and Caccia, 2003), dogs (Barza et al., 1975), and humans (Yim et al., 1985; Dotevall and Hagberg, 1989).
The mean elimination half-times from the brain of young mice (3.9 and 6.5 hours, after 10 and 100 mg/kg) were identical to those in plasma. This fast elimination suggests that no significant drug accumulation can be expected after repeated doses, even twice a day, as can be visually appreciated in the simulations shown in Figs. 2 and 3. Accordingly, the brain concentrations after 4-day treatment with 10 or 100 mg/kg were not significantly different from those after a single dose. These data show that maximal brain concentrations (Cmax) approaching 7 μM on average can only be reached after 100 mg/kg doxycycline twice a day, whereas one dose a day gives up to 5 μM. After repeated doses of 10 mg/kg, either once or twice a day, Cmax remains at ∼0.5 μM.
Elimination was slower from both plasma and brain of 20-month-old mice (11.1 and 15.4 hours, respectively); these data led us to estimate that brain concentrations up to 10 μM might be reached after repeated doses (data not shown). Analysis of PK data also indicated greater exposure to the drug in older mice owing to the higher apparent volume of distribution as a consequence of the different weight (i.e., more adipose tissue), and this is a common observation for lipophilic drugs like doxycycline (Hanley et al., 2010).
Finally, doxycycline levels were similar in brain areas of 20-month-old WT and APP23 mice 24 hours after the fourth intraperitoneal injection, indicating no significant changes in BBB passage of the drug in AD mice.
The new details on doxycycline pharmacokinetics in the present study must be taken into account for fine tuning in vivo animal studies. In addition, the drug concentrations reached in the brain under specific treatment conditions may help identify the mechanism(s) of action of the central in vivo effects of doxycycline in mice. For example, it was recently shown (Balducci et al., 2018) that memory deficits in APP/PS1Tg mice are significantly rescued by single or repeated intraperitoneal doses of 10 mg/kg doxycycline, which results in brain concentrations always lower than 1 μM, i.e., lower than the concentrations required for antiaggregating effects in vitro (≥10 μM) (Forloni et al., 2001, 2002; Cardoso et al., 2003; Giorgetti et al., 2011; Ward et al., 2011; De Luigi et al., 2015; Gonzalez-Lizarraga et al., 2017). This suggests that other properties of doxycycline might be at play, in accordance with the drug’s pleiotropic activities (Stoilova et al., 2013). In fact, the positive cognitive effects of doxycycline in AD mice were not associated with any reduction of the Aβ plaque load, but there was a significant normalization effect on glial cells, whose activation in the AD mouse brain contributes to the memory impairment (Balducci et al., 2017), thus highlighting an important contribution of the drug’s anti-inflammatory effects (Balducci et al., 2018). It has also been shown that the inhibitory effect of doxycycline on poly (ADP-ribose) polymerase-1 (PARP-1), involved in microglial activation, inflammation, and cell death (Kauppinen and Swanson, 2005), also induced by Aβ (Kauppinen et al., 2011), occurs at submicromolar concentrations (Alano et al., 2006), consistently with the concentrations actually measured in the brain of our mice.
Acknowledgments
We thank Dr. Diego Albani and Dr. Federica Fusco for providing APP23 and WT mice. We thank Dr. Pietro La Vitola for his help with treatment of the mice.
Authorship Contributions
Participated in research design: Lucchetti, Fracasso, Balducci, Forloni, Salmona, Gobbi.
Conducted experiments: Lucchetti, Fracasso, Passoni, Balducci.
Performed data analysis: Lucchetti, Fracasso, Gobbi.
Wrote or contributed to the writing of the manuscript: Lucchetti, Gobbi.
Footnotes
- Received July 20, 2018.
- Accepted October 26, 2018.
↵This article has supplemental material available at dmd.aspetjournals.org.
Abbreviations
- AUC
- area under the curve
- BBB
- blood-brain-barrier
- CJD
- Creutzfeldt-Jakob disease
- Cmax
- peak concentration
- HCOOH
- formic acid
- HPLC
- high-performance liquid chromatography
- IS
- internal standard
- MS/MS
- tandem mass spectrometry
- PK
- pharmacokinetic(s)
- t1/2
- elimination half-life
- WT
- wild type
- Copyright © 2018 by The American Society for Pharmacology and Experimental Therapeutics