Regioisomerism and 5-substitution as promising approaches for molecular design of safer acetaminophen derivatives

Kayo Silva Gustavo Universidade Federal do Pará: Universidade Federal do Para Roberto Barbosa de Morais Universidade Federal do Pará: Universidade Federal do Para Luan Oliveira Ferreira Universidade Federal do Pará: Universidade Federal do Para Valéria Bittencourt Pessoa Mourão Universidade Federal do Pará: Universidade Federal do Para Gerson Maciel Coelho Universidade Federal do Pará: Universidade Federal do Para Tamires Ferreira da Silva Universidade Federal do Pará: Universidade Federal do Para Laine Celestino Pinto Universidade Federal do Pará: Universidade Federal do Para Dulcidéia da Conceição Palheta Universidade Federal Rural da Amazonia Flávia Cristina Matos Oliveira Universidade Federal Rural da Amazonia Natália Lopes Soares Universidade Federal Rural da Amazonia Arnaldo Jorge Martins-Filho Evandro Chagas Institute: Instituto Evandro Chagas Elizabeth Sumi Yamada Universidade Federal Rural da Amazonia Dielly Catrina Favacho Lopes Universidade Federal do Pará: Universidade Federal do Para Rosivaldo dos Santos Borges (  rosborg@ufpa.br ) Universidade Federal do Pará https://orcid.org/0000-0003-4072-7573


Introduction
Acetaminophen (ACP) or paracetamol is a widely used over-the-counter analgesic and antipyretic drug and appears to be safe if used in normal therapeutic doses. However, large doses of ACP produce hepatic and/or renal injury in humans and in experimental animals [1,2]. Its pharmacological effects are generally considered to be based on inhibition of prostaglandin synthesis [3][4][5].
On the other hand, it is metabolized primarily by glucuronidation and sulfation to form major conjugates ACP-glucuronide and ACP-sulfate, respectively. Similarly, a small amount is probably modi ed via a third metabolic pathway, that is, oxidation by the microsomal cytochrome P450-containing mixed-function oxidase system to N-acetyl-p-benzosemiquinone-imine (NAPSQI) and N-acetyl-p-benzoquinone-imine (NAPQI). A glutathione 1,4-Michael adduct of NAPQI and the corresponding cysteine conjugate and mercapturic acid breakdown products were found in human urine after ingestion of ACP [6][7][8].
Its hepatotoxic effect appears to be con ned in compounds that can to form quinoid structures, which are susceptible to direct reaction by irreversible covalent bond and depletion of nonsoluble thiols. The oxidation mechanism of ACP to NAPSQI and NAPQI [15][16][17] was related to bond dissociation energies of phenol (BDE OH ) and acetamide (BDE NH ) moieties, respectively, using calculations at B3LYP level, where the BDE OH is energetically favored over BDE NH and the molecules with high BDE NH values are less hepatotoxic ( Fig. 1) [18,19].
In this work, a molecular strategy was used for drug design of acetaminophen derivatives using calculations of bond dissociation energies in the selection for synthesis and toxicological evaluation. Eight new derivatives were assembled by a rational drug design by using regioisomerism followed by metabolism blocker at 5 position using electron donating (EDGs) or electron withdrawing groups (EWGs) looking for the role played by the structural features of these molecules and in obtaining new compounds endowed with less hepatotoxicity by changes on BDE OH and BDE NH values for the semiquinone and quinone-imine formations (Fig. 2). An acetaminophen regioisomer substituted at the 5-position by chloro, named chloramidaphen (CAP), was selected for toxicity evaluation.

Results And Discussion
Design of acetaminophen regioisomers Our design and development strategy adopted by us in this study is the regioisomerism, changed the phenol moiety of parafor ortho-position. In Fig. 3, the differences on BDE OH and BDE NH between acetaminophen and chloramidaphen are clearly showed. Both compounds have these electronics properties completely different between para (NAPSQI and NAPQI) and ortho of N-acetyl-obenzosemiquinone imine and N-acetyl-o-benzoquinone imine (NAOSQI and NAOQI).
The BDE OH values for these compounds were 85.64 and 81.66 kcal mol -1 and a difference of 3.97 kcal mol -1 was observed between them [18]. Therefore, the regioisomer derivative has a lowest quenching energy than on a rst hydrogen transfer. These results indicate that the regioisomerism increased the quenching capacity, which could be related to better biological activity [20]. In addition, the BDE NH values for these two molecules were 77.30 and 90.72 kcal mol -1 . This second hydrogen abstraction shows a higher difference of 13.41 kcal mol -1 for ortho-regioisomer.
These results are in agreement to nucleophilicity (HOMO), electrophilicity (LUMO), and chemical stability (GAP) [22]. In addition, due to the high chemical stability of halogen on drug metabolism process [23], we choice the chlorinated derivative for synthesis and toxic evaluation. This compound shows a difference of -5.00 and -14.38 kcal/mol for BDE OH and BDE NH values, respectively. Therefore, our methodology shows that new acetaminophen derivative can be a good strategy in drug design for the safer compounds because of the highest values for quinone production as compared to acetaminophen [18]. In fact, ACP toxicity is related to the level of toxic intermediate metabolites such as N-acetyl-p-benzoquinone-imine (NAPQI) as main chemical mechanism [9,14,16]. So, the ability to predict the toxicity pro le of CAP is critical to rational pharmaceutical drug design and development. In addition, a similar electronic mechanism is involved for other phenol derivatives [24].

Synthesis of chloramidaphen
After that, CAP was synthesized from 5-chloro-2-aminophenol using acetic acid, sodium carbonate, and water, as acylating group, alkaline catalytic, and solvent medium, respectively. After recrystallization on methanol, a mp 182. 6

Biochemical changes upon chloramidaphen exposure
Analyzing the results of AST, ALT, creatinine and urea levels of rats treated with 300 mg/kg of CAP (Table  2), our ndings shows that all AST levels measure were increased on days 8 and 15 both treated and control groups and were signi cant different of day 0 measure (p < 0.05) and samples of treated group were different of control group on day 15 (p < 0.05). Besides that, there was no signi cant difference in the ALT, urea and creatinine levels of both treated and control rats throughout the experiment.
The biochemistry parameters for 2000 mg/kg dose were showed on Table 3. The results show that AST levels measure were increased on days 8 and 15 both treated and control groups and were signi cant different of day 0 measure (p < 0.05). On the other hand, ALT level was decreased to control group on day 15 (p < 0.05 vs day 0). In addition, no more signi cant difference in the ALT, urea and creatinine levels were found in both treated and control groups.
The knowledge of normal range or expected values for hematologic and serum chemistry can be a useful tool for interpreting toxicology and safety studies [26]. It is important to notice that several factors can modify these values, like age, diet, handling and environment [27].
Analysis for biochemistry parameters, the toxicological effects were evident to aspartate transaminase (AST, formerly referred to as serum glutamic-oxaloacetic transaminase, SGOT). Here, we showed that high AST levels on both treated groups (300 mg/kg and 2000 mg/kg) and controls groups. These results were different of the values already found in others studies [26][27][28]. We hypothesized that the corn oil and other vegetable oil may be altering liver physiology by alteration of triglycerides synthesis and hepatic lipid peroxidation in some conditions [29][30][31][32][33]. Thus, these ndings showed that the CAP presented neither hepatotoxic effect nor renal toxicity, since that there is no signi cant change in urea and creatinine.
To be one more way to study the hepatotoxic effect of the drug, we also calculated the alterations in the relation AST/ALT levels to additionally suggest liver toxicity, but these levels are not signi cant. Our results for 300 and 2000 mg/kg show low levels of AST and ALT. In both doses, there were few differences among control and treated with chloramidaphen or 300 and 2000 mg/kg. None animal died for 15 days. However, an increase of ALT levels on Swiss-Webster mice treated with 400 and 750 mg/kg acetaminophen was showed of 1552 ± 501 to 4030 ± 829 mU/L. The survivors at 24 hours after acetaminophen treatment was 8 (n = 10) for 400 mg/kg and 4 (n = 10) for 750 mg/kg [34]. In fact, chloramidaphen seems to have a much safer therapeutic index when compared to its relative drug acetaminophen, which have oral LD 50 of 338 mg/kg (oral, mouse) [35].
In accordance to our biological study, some evidence was observed that ortho-regioisomers may be an alternative approach for design and development of low toxic derivative of acetaminophen. The substantial protection afforded by chloramidaphen can be related to decrease of electrophilic mechanism involved for metabolism and toxicity of quinone-imine species due to its high BDE NH values. These mechanisms need to be further investigated.

Conclusion
In this work, a molecular strategy was successful used for understanding of the mechanistic features and drug design of acetaminophen regioisomers using calculations of bond dissociation of phenol (BDE OH ) and acetamide (BDE NH ) moieties. The less toxicity was more related to changes on BDE NH than BDE OH values. Another strategy used was a metabolism blocker at 5 position. The chloramidaphen does not present toxic characteristics in the acute test. The lethal oral toxicity of this compound was estimated to be higher than 2000 mg/kg, classifying as category 5 according to OECD Guide 420, indicating a certain safety margin associated with corn oil. Although the results obtained in this research, it is known that more studies should be performed in order to evaluate the total safety on this compound. Nevertheless, the low toxicity can be related to BDE NH values and NAPQI formation capacity. Furthermore, these results can be used for drug design for safe analgesic compounds of acetaminophen.
Methodology Molecular design of chloramidaphen Geometry optimizations of acetaminophen derivatives have been carried out using density functional theory (DFT) [46]. The calculations were performed with the Gaussian 03 molecular package [47]. All structures were submitted to PM3 [48] geometry conformational search. One hybrid functional of the DFT/B3LYP-31G (d,p) basis set [46,49,50] for the geometry optimizations and hydrogen donations.
The bond dissociation energies of the phenol group or N-acetyl-p-benzosemiquinone imine (NAPSQI) formation were calculated as the energy differences between a neutral molecule and the respective semiquinone plus hydrogen radical as de ned in Equation 1.

BDE OH = (ECH 3 CONHC 6 H 4 O • + EH • ) -ECH 3 CONHC 6 H 4 OH (Equation 1)
The bond dissociation energies of the acylamide group or N-acetyl-p-benzoquinone imine (NAPQI) formation were calculated as the energy differences between a NAPSQI molecule and the respective NAPQI plus hydrogen radical which can be written as Equation 2.

Acute toxicity oral test
Young adults Wistar male rats (n = 12) weighing 220-260g and 8-10 weeks old were used in this study. All animals were housed in a controlled environment (22 ± 2°C; 12/12 h light/dark cycle) with access to food and water ad libitum. All animal experiments complied with the ARRIVE guidelines [52] and were carried out in accordance with the EU Directive 2010/63/EU for animal experiments [53].
The acute toxicity oral test was performed according to the Organization for Economic Cooperation and Development (OECD) guideline 420 [54] and Brazilian Health Regulatory Agency [55], with initial dose at 300 mg/kg. The animals were divided into four groups: control 300 mg/kg (n = 2), treated 300 mg/kg (n = 4), control 2000 mg/kg (n = 2), treated 2000 mg/kg (n = 4). After fasting overnight (8 h), the chloramidaphen was administered in a single dose dissolved in corn oil by oral gavage (day 1). Control animals received corn oil proportional of treated dose.
After dosing, all animals were observed individually for mortality and to monitor physical and behavioral alterations for 14 days, giving special attention to the rst 30 min, then at 1, 2, 4 and 24 h following treatment. After that, the animal observation was performed once per day for the post-dosing period. Symptoms of toxicity such as hypo-activity (lethargy), piloerection, salivation, breathing di culty, diarrhea, tremors, ptosis, hypnosis, muscle tone changes, tail grip response, convulsion and general appearance were evaluated after administration.

Biochemical parameters
The blood samples collected by cardiac puncture underwent evaluation of biochemical and hematological parameters. Regarding the biochemical analysis, tubes without anticoagulant were used and centrifuged, in order to obtain the serum. Serum samples were analyzed to clinical biochemistry parameters included alanine aminotransferase (ALT), aspartate aminotransferase (AST), urea (UREA) and creatinine (CREA).

Data analysis
The results were presented as mean ± standard derivation (SD) and analyzed using one-way analysis of variance (ANOVA). The differences between the means were tested using post hoc Tukey and values of p < 0.05 were considered statistically signi cant. GraphPad Prism for windows version 8.0 was used for the statistical analysis.  Figures Figure 1 Energies to formation of NAPSQI and NAPQI from acetaminophen oxidation Differences on BDEOH and BDENH between acetaminophen (1) and its regioisomer (2) Figure 4 Molecular block of hydroxylation on acetaminophen regioisomers and quinone metabolism. Synthesis reaction of chloramidaphen or 5-chloro-2-hydroxy-acetamide (8)