pH-dependent spontaneous hydrolysis rather than gut bacterial metabolism reduces levels of the ADHD treatment, Methylphenidate

Methylphenidate is absorbed in the small intestine. The drug is known to have low bioavailability and a high interindividual variability in terms of response to the treatment. Gut microbiota has been shown to reduce the bioavailability of a wide variety of orally administered drugs. Here, we tested the ability of small intestinal bacteria to metabolize methylphenidate. In silico analysis identified several small intestinal bacteria to harbor homologues of the human carboxylesterase 1 enzyme responsible for the hydrolysis of methylphenidate in the liver. Despite our initial results hinting towards possible bacterial hydrolysis of the drug, up to 60% of methylphenidate was spontaneously hydrolyzed in the absence of bacteria and this hydrolysis was pH-dependent. Overall, the study shows that pH-dependent spontaneous hydrolysis rather than gut bacterial metabolism reduces levels of methylphenidate and suggest a role of the luminal pH in the bioavailability of the drug.


Introduction 52
Attention-deficit/hyperactivity disorder (ADHD) is one of the most prevalent 53 neurodevelopmental disorders, affecting 6-12% of children and persisting into 54 adulthood in around 60% of the cases [1]. Although a cause-effect relationship has 55 not yet been established for ADHD, altered levels of dopamine and norepinephrine, 56 and their corresponding transporters in the brain, seem to play a key role in the 57 cognitive impairment and dysregulated reward system that characterize ADHD [2,3]. 58 Thus, ADHD is mainly treated with amphetamine-like psychostimulants that improve 59 symptoms by increasing the levels of dopamine and norepinephrine 60 neurotransmitters in the brain.

61
Methylphenidate (MPH), a dopamine reuptake inhibitor, is considered as the golden 62 treatment for ADHD [4,5]. MPH is administered orally and rapidly absorbed into the 63 blood steam through the small intestine (SI), reaching peak concentrations between 64 1-and 3-h post ingestion. Around 70% of MPH is recovered in urine in the form of 65 ritalinic acid (RA). RA is the inactive metabolite of MPH produced in the liver by 66 carboxylesterase 1 (CES1) [6]. Despite its efficacy, MPH has a low bioavailability of 67 around 30%. Moreover, there is a high interindividual variability among patients in 68 terms of their response to the treatment [7,8]. 69 First-pass metabolism could explain the low bioavailability of MPH. Genetic variations 70 in CES1 have been shown to impact enzyme activity towards different substrates in 71 vitro [9], which could account for differences in MPH hydrolysis among patients. 72 Nevertheless, clinical human studies are scarce. Increased concentrations of MPH 73 were found in plasma of individuals carrying a polymorphism in the CES1 gene 74 indicating decreased enzyme activity [10,11]. However, this study was performed on 75 a small number of healthy volunteers administered a single dose of MPH, which 76 cannot be translated to ADHD patients on multiple doses per day of the drug. 77 Recently, absorption of the drug was modelled based on physicochemical properties 78 of the drug, formulation-related information, and differences in gut physiology along 79 the gastrointestinal tract [12]. Importantly, non-specific intestinal loss of MPH had to 80 be introduced in the model in order to obtain plasma profiles of MPH and RA 81 comparable to those found in clinical studies. Thus, the model suggested intestinal 82 loss of MPH prior to absorption and hepatic/systemic metabolism [12] [13]. 83 Nonetheless, the mechanism explaining such non-specific intestinal loss remains to 84 be explained, since the CES1 enzyme is absent in the gastrointestinal tract [14]. 85 The gut microbiota represents a metabolic factory able to synthesize indigenous and 86 exogenous compounds, such as food components and drugs, in the host 87 [15] [16][17] [18]. Bacterial metabolism of MPH could therefore explain the potential 88 intestinal loss of MPH. Indeed, bacterial esterases that are able to hydrolyze carboxyl 89 esters have been previously described [18]. For example, the highly abundant gut 90 bacterium Escherichia coli harbors the esterase yjfP, which is able to hydrolyze the 91 ester 4-nitrophenylacetate [19]. Similarly, Bacillus subtilis pnbA esterase has been 92 shown to hydrolyze 4-nitrophenylacetate [20]. The present study investigates whether 93 gut bacteria can metabolize MPH leading to increased presystemic metabolism and 94 reduced bioavailability of the drug, thereby interfering with the efficacy of MPH 95 medication in ADHD patients.

97
Rat luminal content 98 Luminal small intestinal content of wild-type Groningen (WTG) rats (n = 5) was 99 collected in sterile Eppendorf tubes by gentle pressing along the entire cecum and 100 5 was snap frozen in liquid N2 and stored at -80 o C. 10% (w/v) suspensions of the 101 luminal content were grown in enriched beef broth based on SHIME medium (S1 102 enriched beef broth (S1 Table). For experiments where the effect of pH on MPH 116 hydrolysis was studied (Fig 3), culture media were prepared at different pH values.

117
To do so, buffer solutions of KH2PO4/K2HPO4 were prepared at different 118 concentrations to obtain the desired pH when adding them to the media. Strains 119 which required shaking for proper growth (all E. coli strains, C. ammoniagenes 120 DSM20306, L. salivarius W24 and L. plantarum W1) were grown with continuous 121 agitation at 220 rpm. Bacteria were inoculated from -80 o C glycerol stocks and grown 122 overnight. Before the experiment, cultures were diluted to 1% in fresh enriched beef 123 broth medium and were grown until late exponential phase (S2 Figure). Growth was 124 followed by measuring optical density (OD) at 600 nm in a spectrophotometer. 50 µM 125 6 MPH was added to the cultures and samples were taken at 0 and 24 h for HPLC-126 MS/MS analysis. 128 In order to monitor the levels of MPH and Ritalinic acid (RA) hydrochloride solution (1 129 mg/mL as a free base, Sigma-Aldrich, The Netherlands) in bacterial cultures, 130 samples were collected by adding 100 µL of culture to 400 µL of 100% methanol.

131
The internal standard d10-Ritalinic acid (d10-RA) hydrochloride solution (100 µg/mL 132 as a free base; Sigma-Aldrich, The Netherlands) was added to all samples at a final 133 concentration of 2 ng/µL as an internal standard for accurate quantification. Samples 134 were then centrifuged at 14000 rpm for 15 min at 4 o C. Supernatants were 135 transferred to a clean tube and methanol was evaporated using a Savant speed-136 vacuum dryer (SPD131, Fisher Scientific, Landsmeer, Netherlands). Finally, samples 137 were reconstituted in 500 µL of water.

138
Sample analysis was performed using a Shimazu HPLC system consisting of a SIL-139 20AC autosampler, a CTO-20AC column oven and LC-20AD liquid chromatograph 140 pumps. Chromatography separation was achieved using a Waters CORTECS C18+ 141 column (100x2.1 mm; 2.7 µm). The mobile phase consisted of a mixture of water (A) 142 and acetonitrile (B) both containing 0.1% formic acid. A flow rate of 0.25 mL/min was 143 used with a linear gradient: 5% (B) for 5 min, followed by an increase to 80% (B) in 5 144 min, which was kept for 3 min to wash the column and then returned to initial 145 conditions for 2 min. The HPLC was coupled to an API3000 triple-quadrupole mass       Citrobacter and Faecalibacterium to harbor highly homologous proteins to the human 216 CES1 annotated as esterases or carboxylesterases (Fig 2A) homologous proteins (Fig 2A). 226 Based on the in-silico analysis, a comprehensive screening of gut-associated 227 bacterial strains harboring esterase proteins was performed. Out of all the bacteria 228 found to harbor CES1, yjfP and pnbA homologues, we focused on gut bacteria 229 known to inhabit the small intestine, the major site of MPH absorption [12]. To this The spontaneous hydrolysis of MPH observed in the bacteria growth medium in the 246 absence of bacteria (Fig 2B) led us to investigate the role of pH in MPH hydrolysis.

247
In bacterial cultures where MPH was not metabolized, the pH measured after 24 h 248 ranged from 4.0 -5.5. In contrast, bacterial cultures that showed high levels of MPH 249 11 hydrolysis had a pH between 7.5 -8.0. Moreover, the E. coli BW25113 cultures had a 250 slightly higher average pH of 7.9 compared to E. coli DSM1058 and E. coli 251 DSM12250, where the average pH was 7.5 and this was accompanied by a smaller 252 percentage of MPH hydrolysis, 70% versus 50% respectively (Fig 2B). Indeed,  Table). Enriched beef broth was prepared at different pH values, ranging from 262 5.5 to 8.0 resembling the pH values previously measured in the different bacterial 263 cultures (Fig 2B), and was incubated aerobically with 50 µM MPH for 24 h and 264 analyzed by HPLC-MS/MS. At pH ≤ 6, which resembles the pH measured in bacterial 265 cultures that did not hydrolyze MPH, ³ 80% of MPH remained intact, while 80% of the 266 drug was hydrolyzed to RA at pH 8 (Fig 3A). Pearson r correlation analyses showed 267 a strong positive correlation (r = 0.98, r 2 = 0.96, P value = 0.0005) between the pH 268 value of the medium and the amounts of hydrolyzed MPH.

269
To determine whether the strong correlation found between pH and MPH hydrolysis 270 could explain the differences in MPH metabolism in the bacterial pure cultures, we 271 selected E. coli BW25113 that showed 70% hydrolysis of MPH into RA, and E. When E. coli BW25113 was grown in enriched beef broth at pH ≤ 6.5, a negligible 277 amount of MPH was hydrolyzed to RA after 24 h of incubation with 50 µM MPH and 278 the pH of the 24 h culture dropped to 5.0-5.5. In contrast, when E. coli BW25113 279 was grown in enriched beef broth at pH ≥ 7 (the same pH of the culture plotted in Fig   280   2B), the pH of the culture rose to 7.5-8.5 after 24 h of incubation with MPH and 70-281 90% of MPH was hydrolyzed to RA (Fig 3B). On the other hand, when E. faecium 282 W54 was grown in enriched beef broth at pH ≤ 6.5 the pH of the culture dropped 283 below 5 after 24 h of incubation with MPH and a negligible amount of MPH was 284 hydrolyzed to RA. When E. faecium W54 was grown at pH ≥ 7, the pH of the cultures 285 dropped to values between 6.5-5.5 and this was accompanied by only 20% MPH 286 hydrolysis (Fig 3C). Pearson r correlation analyses showed a positive correlation  Table) 294 changes during the course of bacterial growth and their metabolism, which in turn, 295 could deplete potential hydrolysis catalyzing agents. E. coli BW25113 and E. faecium 296 W54 cultures were grown to late exponential phase (S2 Figure) and the 297 supernatants were collected, filtered, and incubated with 50 µM MPH. pH values of 298 E. faecium W54 supernatants, which were around 5.5, were adjusted to 6.0, 7.0 and 299 13 7.5, respectively, to resemble the pH previously measured in the different bacterial 300 cultures (Fig 2B). Interestingly, incubation of MPH with E. faecium W54 supernatants 301 at pH 6 resulted in 10% hydrolysis of MPH to RA, but levels of hydrolysis increased 302 with increasing pH values; 20-30% at pH 7 and 60% at pH 7.5, respectively (Fig 4A). 303 When E. coli BW25113 supernatants, which had a pH around 7.0 were adjusted to 304 7.5 to resemble the pH after 24h of growth (Fig 2B), the MPH hydrolysis increased 305 from 20-30% at pH 7.0 (Fig 2B) to 60% at pH 7.5 (Fig 4B). Collectively, our results 306 indicate that the majority of the observed hydrolyzed MPH results from pH-dependent 307 spontaneous non-enzymatic conversion rather than from bacterial metabolism.  (Fig 2B), as well as the results from incubation of MPH in growth 313 media in the presence and absence of pure bacterial cultures, uncovered the pH-314 dependent MPH hydrolysis, irrespective of the presence of bacteria (Figs 3 and 4). 315 The complex bacterial community present in the luminal content could have caused ileostomy effluent from an ileostomy patient raised from 5.6 in the morning to 6.8 in 347 the afternoon due to changes in feeding cycles [34] indicating that pH changes can 348 indeed take place in the SI due to bacterial metabolism. Moreover, protein and amino  Collectively, the present study shows that MPH is subject to spontaneous hydrolysis

537
The authors declare no conflicts of interest.