The Magic Methyl and Its Tricks in Drug Discovery and Development

One of the key scientific aspects of small-molecule drug discovery and development is the analysis of the relationship between its chemical structure and biological activity. Understanding the effects that lead to significant changes in biological activity is of paramount importance for the rational design and optimization of bioactive molecules. The “methylation effect”, or the “magic methyl” effect, is a factor that stands out due to the number of examples that demonstrate profound changes in either pharmacodynamic or pharmacokinetic properties. In many cases, this has been carried out rationally, but in others it has been the product of serendipitous observations. This paper summarizes recent examples that provide an overview of the current state of the art and contribute to a better understanding of the methylation effect in bioactive small-molecule drug candidates.


Introduction
The small, monovalent, and lipophilic methyl group (-CH 3 ) is versatile and of great importance in the design or optimization of bioactive compounds, whether in terms of pharmacodynamic or pharmacokinetic properties [1]. Its role in drug design and hitto-lead optimization processes is broad, including the displacement of water molecules during molecular recognition, i.e., the realization of hydrophobic interactions [2,3]; the participation in van der Waals interactions; the modulation of physicochemical properties, such as LogP and aqueous solubility [1]; and the control of the conformational properties of a given scaffold [1]. The control of the number of conformations in a given system by methylation correlates with the strategy of conformational restriction [4,5]. Other drug design strategies, such as bioisosterism [6][7][8] and homologation [9], can also benefit from methyl group insertion. During the drug discovery process, controlling conformational behavior can not only favor the adoption of a bioactive conformation, generating a potency gain for pharmacological target modulation, but can also help break to planarity and symmetry, resulting in increased aqueous solubility while increasing lipophilicity [10,11].
Other uses of the methyl group include modulating metabolic reactions by preventing their occurrence through stereoelectronic effects, by serving as a metabolic point to prevent the formation of toxic metabolites, or by modulating the metabolic profile, making molecules softer for metabolic reactions [1].
This plethora of effects mediated by the methyl group is commonly referred to as the "methyl effect", the "methylation effect", or even the "magic methyl" effect. It is important to mention that there are previous works that have already reviewed this topic and are EZH2 Y641F mutant. The authors suggest that the cyclization abolished the "magic methyl" effect previously reported to be key to the FDA-approved drug 8. Therefore, the authors selected the open-ring analog (14) for further structure-activity relationship (SAR) exploration, resulting in compound 16, a derivative with a second methyl group at the pyridine ring that showed selectivity over 22 other methyl transferases [19]. The methylation effect in the discovery of EZH2-selective inhibitors [19].

PI3K/mTOR Inhibitors
A series of 2-methyl-1H-imidazo [4,5-c]quinolines were reported [20] based on ring bioisosterism with the 1,3-dihydro-2H-imidazo [4,5-c]quinolin-2-one system present in the phosphoinositide 3-kinases (PI3Ks) and mammalian target of rapamycin (mTOR) inhibitor and clinical candidate BEZ235 (17) [21] (Figure 3). Targeting the PI3K/AKT/mTOR pathway is a validated strategy for cancer treatment because it is aberrantly activated in several human cancers and plays an essential role in cell growth, proliferation, differentiation, and apoptosis [22,23]. The rationale for the modification was to utilize the methyl group to preserve cell permeability and cell absorption capacity while reducing the number of polar heteroatoms (i.e., the oxygen atom of the carbonyl group of candidate 17). Hence, a series of compounds were synthesized in order to explore the potential of the 2-methyl-1H-imidazo [4,5-c]quinoline scaffold and to improve its drug-like profile. This study resulted in compound 18 having the best profile of kinase selectivity, cellular antiproliferative activity, western blot and immunohistochemical analyses, antitumor efficacy in vivo, and pharmacokinetic properties [20].

PI3K/mTOR Inhibitors
A series of 2-methyl-1H-imidazo [4,5-c]quinolines were reported [20] based on ring bioisosterism with the 1,3-dihydro-2H-imidazo [4,5-c]quinolin-2-one system present in the phosphoinositide 3-kinases (PI3Ks) and mammalian target of rapamycin (mTOR) inhibitor and clinical candidate BEZ235 (17) [21] (Figure 3). Targeting the PI3K/AKT/mTOR pathway is a validated strategy for cancer treatment because it is aberrantly activated in several human cancers and plays an essential role in cell growth, proliferation, differentiation, and apoptosis [22,23]. The rationale for the modification was to utilize the methyl group to preserve cell permeability and cell absorption capacity while reducing the number of polar heteroatoms (i.e., the oxygen atom of the carbonyl group of candidate 17). Hence, a series of compounds were synthesized in order to explore the potential of the 2-methyl-1Himidazo [4,5-c]quinoline scaffold and to improve its drug-like profile. This study resulted in compound 18 having the best profile of kinase selectivity, cellular antiproliferative activity, western blot and immunohistochemical analyses, antitumor efficacy in vivo, and pharmacokinetic properties [20].

Selective κ-Opioid Receptor Antagonists
Tetrahydroisoquinoline derivatives have been described as selective κ-opioid receptor antagonists and as compounds of interest for the treatment of several CNS disorders, such as substance abuse, depression, and anxiety [24]. Compound 19 was first discovered as a potent antagonist of this receptor [25], and subsequent SAR evaluations were performed that focused, among other modifications, on the study of the methylation pattern of the piperidine ring. The results revealed that the 4-methylated analog (20) had an 18-fold increase in the affinity for κ-opioid receptors compared to 19 [26] (Figure 4).

Cannabinoid Receptor Modulators
Modulation of the endocannabinoid system by targeting G-protein-coupled cannabinoid receptors has broad therapeutic applications ranging from pain to cancer treatment [27,28]. A series of oxazolo [5,4-d]pyrimidines (22) were designed via the bioisosterism strategy as new cannabinoid receptor 1 (CB1R) and cannabinoid receptor 2 (CB2R) modulators ( Figure 5) [29]. A classical bioisosteric heterocycle replacement strategy was applied to compound 21, a CB2R agonist developed by Eli Lilly [30]. SAR studies revealed the importance of methylation at position 5 of this core when 23 was compared with the unmethylated derivative of the series (24). Compound 23 was characterized as a selective CB2R antagonist with high binding affinity in the low nanomolar range [29].

Selective κ-Opioid Receptor Antagonists
Tetrahydroisoquinoline derivatives have been described as selective κ-opioid receptor antagonists and as compounds of interest for the treatment of several CNS disorders, such as substance abuse, depression, and anxiety [24]. Compound 19 was first discovered as a potent antagonist of this receptor [25], and subsequent SAR evaluations were performed that focused, among other modifications, on the study of the methylation pattern of the piperidine ring. The results revealed that the 4-methylated analog (20) had an 18-fold increase in the affinity for κ-opioid receptors compared to 19 [26] (Figure 4).

Selective κ-Opioid Receptor Antagonists
Tetrahydroisoquinoline derivatives have been described as selective κ-opioid receptor antagonists and as compounds of interest for the treatment of several CNS disorders, such as substance abuse, depression, and anxiety [24]. Compound 19 was first discovered as a potent antagonist of this receptor [25], and subsequent SAR evaluations were performed that focused, among other modifications, on the study of the methylation pattern of the piperidine ring. The results revealed that the 4-methylated analog (20) had an 18-fold increase in the affinity for κ-opioid receptors compared to 19 [26] (Figure 4).

Cannabinoid Receptor Modulators
Modulation of the endocannabinoid system by targeting G-protein-coupled cannabinoid receptors has broad therapeutic applications ranging from pain to cancer treatment [27,28]. A series of oxazolo [5,4-d]pyrimidines (22) were designed via the bioisosterism strategy as new cannabinoid receptor 1 (CB1R) and cannabinoid receptor 2 (CB2R) modulators ( Figure 5) [29]. A classical bioisosteric heterocycle replacement strategy was applied to compound 21, a CB2R agonist developed by Eli Lilly [30]. SAR studies revealed the importance of methylation at position 5 of this core when 23 was compared with the unmethylated derivative of the series (24). Compound 23 was characterized as a selective CB2R antagonist with high binding affinity in the low nanomolar range [29].

Cannabinoid Receptor Modulators
Modulation of the endocannabinoid system by targeting G-protein-coupled cannabinoid receptors has broad therapeutic applications ranging from pain to cancer treatment [27,28]. A series of oxazolo [5,4-d]pyrimidines (22) were designed via the bioisosterism strategy as new cannabinoid receptor 1 (CB 1 R) and cannabinoid receptor 2 (CB 2 R) modulators ( Figure 5) [29]. A classical bioisosteric heterocycle replacement strategy was applied to compound 21, a CB 2 R agonist developed by Eli Lilly [30]. SAR studies revealed the importance of methylation at position 5 of this core when 23 was compared with the unmethylated derivative of the series (24). Compound 23 was characterized as a selective CB 2 R antagonist with high binding affinity in the low nanomolar range [29]. Mugnaini et al. [31] reported that 2-(1-adamantanylcarboxamido)thiophene derivatives (25)(26)(27)(28) are selective CB2R agonists ( Figure 6). The chemical starting point, compound 25, had weak activity against CB2R, and the simple addition of the methyl group (26) resulted in a 50-fold increase in the affinity. The authors remarked on the crucial role that methyl groups play in biologically active small molecules and emphasized that the effect was likely due to 26 s ability to successfully insert its methyl into the receptor binding site to establish effective hydrophobic contacts. This theory was supported by the fact that the n-propyl analog (27) obtained only a threefold increase in the affinity for CB2 receptors. These results are in stark contrast to studies suggesting that adding a methyl group to a lead molecule can result in a 10-fold increase in activity in only 8% of cases, while a 100-fold increase in potency is much less likely, occurring in 0.4% of cases [14,31]. Garai and colleagues [32] employed the magic methyl effect to increase the potency and efficacy of GAT211 (29) [33], a cannabinoid type-1 receptor (CB1R) agonist-positive allosteric modulator (ago-PAM) (Figure 7). The strategic placement of a methyl group at the alpha position of the nitro functional group was hypothesized to be advantageous in terms of activity and functional selectivity, as it generated two diastereomers and an additional chiral center. Results from studies with the two diastereomers highlighted the Mugnaini et al. [31] reported that 2-(1-adamantanylcarboxamido)thiophene derivatives (25)(26)(27)(28) are selective CB 2 R agonists ( Figure 6). The chemical starting point, compound 25, had weak activity against CB 2 R, and the simple addition of the methyl group (26) resulted in a 50-fold increase in the affinity. The authors remarked on the crucial role that methyl groups play in biologically active small molecules and emphasized that the effect was likely due to 26's ability to successfully insert its methyl into the receptor binding site to establish effective hydrophobic contacts. This theory was supported by the fact that the n-propyl analog (27) obtained only a threefold increase in the affinity for CB 2 receptors. These results are in stark contrast to studies suggesting that adding a methyl group to a lead molecule can result in a 10-fold increase in activity in only 8% of cases, while a 100-fold increase in potency is much less likely, occurring in 0.4% of cases [14,31]. Mugnaini et al. [31] reported that 2-(1-adamantanylcarboxamido)thiophene derivatives (25)(26)(27)(28) are selective CB2R agonists ( Figure 6). The chemical starting point, compound 25, had weak activity against CB2R, and the simple addition of the methyl group (26) resulted in a 50-fold increase in the affinity. The authors remarked on the crucial role that methyl groups play in biologically active small molecules and emphasized that the effect was likely due to 26 s ability to successfully insert its methyl into the receptor binding site to establish effective hydrophobic contacts. This theory was supported by the fact that the n-propyl analog (27) obtained only a threefold increase in the affinity for CB2 receptors. These results are in stark contrast to studies suggesting that adding a methyl group to a lead molecule can result in a 10-fold increase in activity in only 8% of cases, while a 100-fold increase in potency is much less likely, occurring in 0.4% of cases [14,31]. Garai and colleagues [32] employed the magic methyl effect to increase the potency and efficacy of GAT211 (29) [33], a cannabinoid type-1 receptor (CB1R) agonist-positive allosteric modulator (ago-PAM) (Figure 7). The strategic placement of a methyl group at the alpha position of the nitro functional group was hypothesized to be advantageous in terms of activity and functional selectivity, as it generated two diastereomers and an additional chiral center. Results from studies with the two diastereomers highlighted the Garai and colleagues [32] employed the magic methyl effect to increase the potency and efficacy of GAT211 (29) [33], a cannabinoid type-1 receptor (CB 1 R) agonist-positive allosteric modulator (ago-PAM) (Figure 7). The strategic placement of a methyl group at the alpha position of the nitro functional group was hypothesized to be advantageous in terms of activity and functional selectivity, as it generated two diastereomers and an additional chiral center. Results from studies with the two diastereomers highlighted the increased potency and efficacy of erythro, (±)-30 compared to threo, (±)-31. The analysis of This result represents the first example of a diastereoselective CB 1 R allosteric modulator interaction [32].

Histamine 1 Receptor Antagonists
To identify new fragment-like [34,35] histamine 1 receptor (H1R) antagonists, a virtual screening campaign was performed, which led to the identification of compound 32 ( Figure 8) [36]. Next, 32 was used for SAR exploration and to investigate the role of the well-defined receptor binding pockets, i.e., (1) the amine binding region, (2) the upper and lower aromatic binding regions, and (3) the effect of binding site (de)solvation on H1R antagonist binding. Molecular modeling analysis combined with SAR exploration indicated the amine binding region as the primary binding hotspot, preferentially binding small tertiary amines, which is related to hydrophobic interactions. The methylation effect is clear when comparing 32 and 33, since the N-methylation strongly increased the binding affinity for H1 receptors [37].

Histamine 1 Receptor Antagonists
To identify new fragment-like [34,35] histamine 1 receptor (H 1 R) antagonists, a virtual screening campaign was performed, which led to the identification of compound 32 ( Figure 8) [36]. Next, 32 was used for SAR exploration and to investigate the role of the well-defined receptor binding pockets, i.e., (1) the amine binding region, (2) the upper and lower aromatic binding regions, and (3) the effect of binding site (de)solvation on H 1 R antagonist binding. Molecular modeling analysis combined with SAR exploration indicated the amine binding region as the primary binding hotspot, preferentially binding small tertiary amines, which is related to hydrophobic interactions. The methylation effect is clear when comparing 32 and 33, since the N-methylation strongly increased the binding affinity for H 1 receptors [37].

Histamine 1 Receptor Antagonists
To identify new fragment-like [34,35] histamine 1 receptor (H1R) antagonists, a virtual screening campaign was performed, which led to the identification of compound 32 ( Figure 8) [36]. Next, 32 was used for SAR exploration and to investigate the role of the well-defined receptor binding pockets, i.e., (1) the amine binding region, (2) the upper and lower aromatic binding regions, and (3) the effect of binding site (de)solvation on H1R antagonist binding. Molecular modeling analysis combined with SAR exploration indicated the amine binding region as the primary binding hotspot, preferentially binding small tertiary amines, which is related to hydrophobic interactions. The methylation effect is clear when comparing 32 and 33, since the N-methylation strongly increased the binding affinity for H1 receptors [37].

Inhibitors of Phosphopantetheine Adenylyltransferase
In a study conducted by Novartis [38], a fragment-based screening approach [34,35] was used to identify inhibitors of phosphopantetheine adenylyltransferase (PPAT) for the discovery of new antibiotics for the treatment of infections caused by multidrug-resistant and pan-drug-resistant Gram-negative bacteria [38]. Fragment 34 was one of the identified hits, and hit-to-lead optimization based on C5 methylation of the imidazo [4,5-b]pyridine core resulted in 35, which had a 15-fold increase in potency (Figure 9), related to additional interactions with a hydrophobic pocket (V135, M105, and L131) of the target [38].

Inhibitors of Phosphopantetheine Adenylyltransferase
In a study conducted by Novartis [38], a fragment-based screening approach [34,35] was used to identify inhibitors of phosphopantetheine adenylyltransferase (PPAT) for the discovery of new antibiotics for the treatment of infections caused by multidrug-resistant and pan-drug-resistant Gram-negative bacteria [38]. Fragment 34 was one of the identified hits, and hit-to-lead optimization based on C5 methylation of the imidazo[4,5-b]pyridine core resulted in 35, which had a 15-fold increase in potency (Figure 9), related to additional interactions with a hydrophobic pocket (V135, M105, and L131) of the target [38]. Moreover, another fragment hit (36) was optimized, and the methylation pattern of this hit profoundly altered its potency ( Figure 10). An X-ray cocrystal of the structurally related hit 37 revealed that this triazolopyrimidinone is bound in a similar manner to 34. The bioisosteric replacement of the bromine atom of 37 by a chlorine atom (38) did not change potency significantly. Surprisingly, substitution of the benzylic position of the benzylamine with a methyl group resulted in a 30-fold activity boost, as observed for the (R)-methyl analog 39, which is probably related to hydrophobic interactions [38].

Genetic Depletion of the Mitotic Aurora Kinase B (AURKB)
AURKB is a gene encoding mitotic Aurora Kinase B that is overexpressed in some tumor cells, making it an interesting therapeutic target [39]. Huang and colleagues [40] employed a methyl group scanning strategy to enable hit-to-lead optimization ( Figure 11) of compounds identified by mechanism-informed phenotypic screening [41], evaluating the genetic depletion of Aurora Kinase B (AURKB) [40]. The authors modified the benzene ring of hit 40 and synthesized ortho-, meta-, and para-methyl-substituted analogs. The parasubstituted compound (41) demonstrated the best polyploidy-inducing activity, with a minimum effective concentration for polyploidy (MECP) of 0.625 µM. The authors further optimized lead 41, resulting in compound 42 (MECP = 0.019 µM). This compound displayed substantial cytotoxic activity in several cancer cell lines and promoted the loss of function in Aurora Kinase B (AURKB) phenotypes [40]. Moreover, another fragment hit (36) was optimized, and the methylation pattern of this hit profoundly altered its potency ( Figure 10). An X-ray cocrystal of the structurally related hit 37 revealed that this triazolopyrimidinone is bound in a similar manner to 34. The bioisosteric replacement of the bromine atom of 37 by a chlorine atom (38) did not change potency significantly. Surprisingly, substitution of the benzylic position of the benzylamine with a methyl group resulted in a 30-fold activity boost, as observed for the (R)-methyl analog 39, which is probably related to hydrophobic interactions [38].

Inhibitors of Phosphopantetheine Adenylyltransferase
In a study conducted by Novartis [38], a fragment-based screening approach [34,35] was used to identify inhibitors of phosphopantetheine adenylyltransferase (PPAT) for the discovery of new antibiotics for the treatment of infections caused by multidrug-resistant and pan-drug-resistant Gram-negative bacteria [38]. Fragment 34 was one of the identified hits, and hit-to-lead optimization based on C5 methylation of the imidazo[4,5-b]pyridine core resulted in 35, which had a 15-fold increase in potency (Figure 9), related to additional interactions with a hydrophobic pocket (V135, M105, and L131) of the target [38]. Moreover, another fragment hit (36) was optimized, and the methylation pattern of this hit profoundly altered its potency ( Figure 10). An X-ray cocrystal of the structurally related hit 37 revealed that this triazolopyrimidinone is bound in a similar manner to 34. The bioisosteric replacement of the bromine atom of 37 by a chlorine atom (38) did not change potency significantly. Surprisingly, substitution of the benzylic position of the benzylamine with a methyl group resulted in a 30-fold activity boost, as observed for the (R)-methyl analog 39, which is probably related to hydrophobic interactions [38].

Genetic Depletion of the Mitotic Aurora Kinase B (AURKB)
AURKB is a gene encoding mitotic Aurora Kinase B that is overexpressed in some tumor cells, making it an interesting therapeutic target [39]. Huang and colleagues [40] employed a methyl group scanning strategy to enable hit-to-lead optimization ( Figure 11) of compounds identified by mechanism-informed phenotypic screening [41], evaluating the genetic depletion of Aurora Kinase B (AURKB) [40]. The authors modified the benzene ring of hit 40 and synthesized ortho-, meta-, and para-methyl-substituted analogs. The parasubstituted compound (41) demonstrated the best polyploidy-inducing activity, with a minimum effective concentration for polyploidy (MECP) of 0.625 µM. The authors further optimized lead 41, resulting in compound 42 (MECP = 0.019 µM). This compound displayed substantial cytotoxic activity in several cancer cell lines and promoted the loss of function in Aurora Kinase B (AURKB) phenotypes [40].

Genetic Depletion of the Mitotic Aurora Kinase B (AURKB)
AURKB is a gene encoding mitotic Aurora Kinase B that is overexpressed in some tumor cells, making it an interesting therapeutic target [39]. Huang and colleagues [40] employed a methyl group scanning strategy to enable hit-to-lead optimization ( Figure 11) of compounds identified by mechanism-informed phenotypic screening [41], evaluating the genetic depletion of Aurora Kinase B (AURKB) [40]. The authors modified the benzene ring of hit 40 and synthesized ortho-, meta-, and para-methyl-substituted analogs. The para-substituted compound (41) demonstrated the best polyploidy-inducing activity, with a minimum effective concentration for polyploidy (MECP) of 0.625 µM. The authors further optimized lead 41, resulting in compound 42 (MECP = 0.019 µM). This compound displayed substantial cytotoxic activity in several cancer cell lines and promoted the loss of function in Aurora Kinase B (AURKB) phenotypes [40].

Neurokinin-3 Receptor Antagonists
The discovery of new neurokinin-3 receptor (NK3R) antagonists for the treatment of sex hormone disorders was described [42]. Starting with an HTS campaign, hit 43 was identified as an interesting starting point for optimization, but several issues such as poor solubility, microsomal stability, and off-target safety profile led to the selection of the parent compound 44 as the starting point ( Figure 12). Despite being significantly less active, 44 presented improved off-target and PK profiles, making it more suitable for optimization. From the SAR analysis, it was possible to perceive the methylation effect by installing a methyl group at the 8-position of the tetrahydro-[1,2,4]triazolo[4,3-a]pyrazine core, where (R)-46 presented increased potency, reaching the nanomolar scale. A second methylation at C6 of the terminal pyridine ring further increased potency, resulting in low nanomolar activity (48 and 49) [42].

Cereblon Ligands for Targeted Protein Degradation
With the goal of obtaining new cereblon ligands for targeted protein degradation [43,44], Xie and coworkers [45] explored the ortho-effect produced by a methyl group ( Figure 13). The authors modified phenyl dihydrouracil (PDHU) (50) (cereblon Kd = 3.05 µM) and observed that the ortho-substituted methyl analog (51) had improved binding potency (cereblon Kd = 1.24 µM). Given this result, the authors selected this compound for further modification and explored the vector at the meta-position for attachment of the

Neurokinin-3 Receptor Antagonists
The discovery of new neurokinin-3 receptor (NK 3 R) antagonists for the treatment of sex hormone disorders was described [42]. Starting with an HTS campaign, hit 43 was identified as an interesting starting point for optimization, but several issues such as poor solubility, microsomal stability, and off-target safety profile led to the selection of the parent compound 44 as the starting point ( Figure 12). Despite being significantly less active, 44 presented improved off-target and PK profiles, making it more suitable for optimization. From the SAR analysis, it was possible to perceive the methylation effect by installing a methyl group at the 8-position of the tetrahydro-[1,2,4]triazolo[4,3-a]pyrazine core, where (R)-46 presented increased potency, reaching the nanomolar scale. A second methylation at C6 of the terminal pyridine ring further increased potency, resulting in low nanomolar activity (48 and 49) [42].

Neurokinin-3 Receptor Antagonists
The discovery of new neurokinin-3 receptor (NK3R) antagonists for the treatment of sex hormone disorders was described [42]. Starting with an HTS campaign, hit 43 was identified as an interesting starting point for optimization, but several issues such as poor solubility, microsomal stability, and off-target safety profile led to the selection of the parent compound 44 as the starting point ( Figure 12). Despite being significantly less active, 44 presented improved off-target and PK profiles, making it more suitable for optimization. From the SAR analysis, it was possible to perceive the methylation effect by installing a methyl group at the 8-position of the tetrahydro-[1,2,4]triazolo[4,3-a]pyrazine core, where (R)-46 presented increased potency, reaching the nanomolar scale. A second methylation at C6 of the terminal pyridine ring further increased potency, resulting in low nanomolar activity (48 and 49) [42].

Cereblon Ligands for Targeted Protein Degradation
With the goal of obtaining new cereblon ligands for targeted protein degradation [43,44], Xie and coworkers [45] explored the ortho-effect produced by a methyl group ( Figure 13). The authors modified phenyl dihydrouracil (PDHU) (50) (cereblon Kd = 3.05 µM) and observed that the ortho-substituted methyl analog (51) had improved binding potency (cereblon Kd = 1.24 µM). Given this result, the authors selected this compound for further modification and explored the vector at the meta-position for attachment of the

Cereblon Ligands for Targeted Protein Degradation
With the goal of obtaining new cereblon ligands for targeted protein degradation [43,44], Xie and coworkers [45] explored the ortho-effect produced by a methyl group ( Figure 13). The authors modified phenyl dihydrouracil (PDHU) (50) (cereblon K d = 3.05 µM) and observed that the ortho-substituted methyl analog (51) had improved binding potency (cereblon K d = 1.24 µM). Given this result, the authors selected this compound for further modification and explored the vector at the meta-position for attachment of the linker and of the "protein of interest" ligand subunit. The authors identified compound 52 with the best affinity (K d = 0.21 µM) for cereblon. The ortho-methylated cereblon ligands were explored and allowed the identification of potent bromodomain-containing protein 4 (BRD4) degraders [45]. linker and of the "protein of interest" ligand subunit. The authors identified compound 52 with the best affinity (Kd = 0.21 µM) for cereblon. The ortho-methylated cereblon ligands were explored and allowed the identification of potent bromodomain-containing protein 4 (BRD4) degraders [45].

Phosphodiesterase Inhibitors
The methylation effect was also shown to have an impact on multitarget small molecule discovery. The compounds (59 and 60) were designed through molecular hybridization [47] and bioisosteric replacement [7] strategies using 57 and 58 as starting points ( Figure 15). As previously reported, 57 is an adenosine A2A receptor agonist [48], and 58 is a phosphodiesterase 4 (PDE4) inhibitor [49]. Insertion of the methyl group at the

Phosphodiesterase Inhibitors
The methylation effect was also shown to have an impact on multitarget small molecule discovery. The compounds (59 and 60) were designed through molecular hybridization [47] and bioisosteric replacement [7] strategies using 57 and 58 as starting points ( Figure 15). As previously reported, 57 is an adenosine A2A receptor agonist [48], and 58 is a phosphodiesterase 4 (PDE4) inhibitor [49]. Insertion of the methyl group at the

Phosphodiesterase Inhibitors
The methylation effect was also shown to have an impact on multitarget small molecule discovery. The compounds (59 and 60) were designed through molecular hybridization [47] and bioisosteric replacement [7] strategies using 57 and 58 as starting points ( Figure 15). As previously reported, 57 is an adenosine A 2A receptor agonist [48], and 58 is a phosphodiesterase 4 (PDE4) inhibitor [49]. Insertion of the methyl group at the amide nitrogen of the N-acylhydrazone (60) moiety significantly increased the percent inhibition of PDE4A1A compared to the non-methylated analog (59). Further evaluation showed that 60 had an IC 50 of 1.08 µM for PDE4A1A inhibition and a moderate affinity for the adenosine A 2A receptor (K i = 1.5 µM), making this compound interesting for the treatment of pulmonary arterial hypertension. Regardless of the presence of methyl, there is a σ-hole intramolecular interaction between the sulfur atom of the thiophene ring and the nitrogen atom of the imine, which establishes the bioactive conformation for the system, as already described for N-acylhydrazone derivatives [50,51]. However, with the N-methylation of the amide, there is a greater stabilization of this conformation and, consequently, an improvement in biological activity (Figure 15) [50]. amide nitrogen of the N-acylhydrazone (60) moiety significantly increased the percent inhibition of PDE4A1A compared to the non-methylated analog (59). Further evaluation showed that 60 had an IC50 of 1.08 µM for PDE4A1A inhibition and a moderate affinity for the adenosine A2A receptor (Ki = 1.5 µM), making this compound interesting for the treatment of pulmonary arterial hypertension. Regardless of the presence of methyl, there is a σ-hole intramolecular interaction between the sulfur atom of the thiophene ring and the nitrogen atom of the imine, which establishes the bioactive conformation for the system, as already described for N-acylhydrazone derivatives [50,51]. However, with the N-methylation of the amide, there is a greater stabilization of this conformation and, consequently, an improvement in biological activity ( Figure 15) [50]. Brullo and colleagues [52] designed and synthesized methylated PDE4 inhibitors as possible candidates for Alzheimer s disease treatment due to their role in pro-cognitive effects. The authors observed that the methylated open-chain linkers were superior to both de-methylated and cyclic conformationally constrained analogs. For example, the methylated compound 62 showed an IC50 of 0.47 µM (PDE4D3), and the de-methylated 61 compound had an IC50 of 11 µM (PDE4D3) (Figure 16). In addition, crystallographic studies showed that the methyl group was able to interact with the binding site and Brullo and colleagues [52] designed and synthesized methylated PDE4 inhibitors as possible candidates for Alzheimer's disease treatment due to their role in pro-cognitive effects. The authors observed that the methylated open-chain linkers were superior to both de-methylated and cyclic conformationally constrained analogs. For example, the methylated compound 62 showed an IC 50 of 0.47 µM (PDE4D3), and the de-methylated 61 compound had an IC 50 of 11 µM (PDE4D3) (Figure 16). In addition, crystallographic studies showed that the methyl group was able to interact with the binding site and improve improve potency while maintaining the linker flexibility necessary for inhibitors to interact with PDE4 [52]. Nunes et al. [53] reported the optimization of the sulfonamide prototype LASSBio-448 (63) [54], a PDE4 inhibitor (PDE4A IC50 = 0.7 µM; PDE4D IC50 = 4.7 µM), for the treatment of pulmonary inflammatory diseases such as asthma (Figure 17. In this work, the authors investigated the methyl effect by designing and synthesizing methylated homologs on the Nsp3 of a series of sulfonamides and sulfonylhydrazones. While the nonmethylated sulfonylhydrazone LASSBio-1624 (64) was inactive against PDE4, the Nmethylated sulfonylhydrazone derivative, LASSBio-1632 (65), was active, showing antiasthmatic activity associated with the inhibition of PDE4A (IC50 = 0.5 µM) and PDE4D (IC50 = 0.7 µM). The authors also reported that the lead compound was able to block airway hyperreactivity and TNF-α production in lung tissue [53].

Human Ghrelin Receptor Agonist
Other interesting examples of the use of methyl groups in optimization processes are macrocycles and peptides, which normally lack adequate physicochemical and pharmacokinetic properties [63,64]. In these cases, the methyl effect can be exploited to optimize these properties through conformational restriction.
A key example is the discovery of ulimorelin (74), a compound that has reached Phase 3 clinical trials. Ulimorelin (74) acts as an agonist of the human ghrelin receptor (also known as the growth hormone secretagogue receptor-GHSR) and has gastroprokinetic properties. The development of 74 was initiated by an HTS campaign that led to the identification of 73 (Figure 20). Despite its high potency, 73 did not show adequate pharmacokinetic properties, and modification of the macrocycle methylation pattern helped to stabilize the bioactive conformation of this compound, resulting in the discovery of ulimorelin (74), which was four-to fivefold more potent for receptor activation and showed minimally adequate pharmacokinetic properties to enter the clinical phase [65]. It is important to note that during the SAR investigation, the side chain modification of isoleucine to cyclopropyl and the introduction of para-fluor on the phenyl ring of 74 did not significantly affect the affinity of the compound. The authors reported that cyclopropyl is more metabolically stable than the side chain of isoleucine and that para-fluor resulted in an optimization of the ligand lipophilicity efficiency (LLE) [65].

Human Ghrelin Receptor Agonist
Other interesting examples of the use of methyl groups in optimization proce macrocycles and peptides, which normally lack adequate physicochemi pharmacokinetic properties [63,64]. In these cases, the methyl effect can be exp optimize these properties through conformational restriction.
A key example is the discovery of ulimorelin (74), a compound that has Phase 3 clinical trials. Ulimorelin (74) acts as an agonist of the human ghrelin (also known as the growth hormone secretagogue receptor-GHSR) a gastroprokinetic properties. The development of 74 was initiated by an HTS ca that led to the identification of 73 (Figure 20). Despite its high potency, 73 did n adequate pharmacokinetic properties, and modification of the macrocycle met pattern helped to stabilize the bioactive conformation of this compound, resultin discovery of ulimorelin (74), which was four-to fivefold more potent for activation and showed minimally adequate pharmacokinetic properties to e clinical phase [65]. It is important to note that during the SAR investigation, the si modification of isoleucine to cyclopropyl and the introduction of para-fluor on th ring of 74 did not significantly affect the affinity of the compound. The authors r that cyclopropyl is more metabolically stable than the side chain of isoleucine para-fluor resulted in an optimization of the ligand lipophilicity efficiency (LLE)

Pan-Genotype NS3/4A Protease Inhibitors
The results reported by Sun and colleagues [66] highlight the effect of the group in improving bioavailability following oral administration to rats of pan-g NS3/4A protease inhibitors for the treatment of hepatitis C virus infection [67]. F authors incorporated two methyl groups on 75 (IC50 = 51 nM) to produce 76, a com with improved activity (IC50 = 8 nM) against the genotype 3a (GT-3a) NS3/4A ( Figure 21). Based on this compound, a series of macrocycles were designed to better in vivo profile. The authors addressed the metabolic liability of 76 by explo deuteration strategy and highlighted that an optimal profile was obtai incorporating a CF3 into the Boc group and an additional methyl next to th acylsulfonamide moiety (77, IC50 = 4.8 nM). These modifications led to improvem both in vivo distribution and metabolic stability [66].

Pan-Genotype NS3/4A Protease Inhibitors
The results reported by Sun and colleagues [66] highlight the effect of the methyl group in improving bioavailability following oral administration to rats of pan-genotype NS3/4A protease inhibitors for the treatment of hepatitis C virus infection [67]. First, the authors incorporated two methyl groups on 75 (IC 50 = 51 nM) to produce 76, a compound with improved activity (IC 50 = 8 nM) against the genotype 3a (GT-3a) NS3/4A protease ( Figure 21). Based on this compound, a series of macrocycles were designed to obtain a better in vivo profile. The authors addressed the metabolic liability of 76 by exploring the deuteration strategy and highlighted that an optimal profile was obtained by incorporating a CF 3 into the Boc group and an additional methyl next to the polar acylsulfonamide moiety (77, IC 50 = 4.8 nM). These modifications led to improvements in both in vivo distribution and metabolic stability [66].

Class I Histone Deacetylase (HDAC) Inhibitors
The authors of [68] investigated the impact of the presence of the methyl group in the design of selective Class I HDAC inhibitors as interesting candidates for cancer treatment. For example, in the macrocyclic prototype 79 (HDAC1-3 range of IC50 = 3.1-8.9 nM) [69], removal of the methyl group from the propenyl group resulted in compound 78 ( Figure  22). Removal of the methyl group (78) was detrimental, resulting in IC50 activities in the range of 69-110 nM (HDAC1-3). When a second methyl group was added (80), a small decrease in the inhibitory activity was observed (HDAC1-3 range of IC50 = 11-21 nM). Theoretical modeling studies suggested that the binding pocket better fits the dehydrobutyrine moiety of 79, which contains only one methyl group in the olefin subunit and seems to be important for the inhibition of HDACs from Class I [68].

Trypanocidal Analogs of Benznidazole
To design new analogs of benznidazole (81), Alcantara and coworkers [70] made changes to the imidazole ring, moving the nitro group to position 4 and incorporating the methyl group in position 2 ( Figure 23). The authors added the methyl group based on

Class I Histone Deacetylase (HDAC) Inhibitors
The authors of [68] investigated the impact of the presence of the methyl group in the design of selective Class I HDAC inhibitors as interesting candidates for cancer treatment. For example, in the macrocyclic prototype 79 (HDAC1-3 range of IC 50 = 3.1-8.9 nM) [69], removal of the methyl group from the propenyl group resulted in compound 78 ( Figure 22). Removal of the methyl group (78) was detrimental, resulting in IC 50 activities in the range of 69-110 nM (HDAC1-3). When a second methyl group was added (80), a small decrease in the inhibitory activity was observed (HDAC1-3 range of IC 50 = 11-21 nM). Theoretical modeling studies suggested that the binding pocket better fits the dehydrobutyrine moiety of 79, which contains only one methyl group in the olefin subunit and seems to be important for the inhibition of HDACs from Class I [68].

Class I Histone Deacetylase (HDAC) Inhibitors
The authors of [68] investigated the impact of the presence of the methyl group in the design of selective Class I HDAC inhibitors as interesting candidates for cancer treatment. For example, in the macrocyclic prototype 79 (HDAC1-3 range of IC50 = 3.1-8.9 nM) [69], removal of the methyl group from the propenyl group resulted in compound 78 ( Figure  22). Removal of the methyl group (78) was detrimental, resulting in IC50 activities in the range of 69-110 nM (HDAC1-3). When a second methyl group was added (80), a small decrease in the inhibitory activity was observed (HDAC1-3 range of IC50 = 11-21 nM). Theoretical modeling studies suggested that the binding pocket better fits the dehydrobutyrine moiety of 79, which contains only one methyl group in the olefin subunit and seems to be important for the inhibition of HDACs from Class I [68].

Trypanocidal Analogs of Benznidazole
To design new analogs of benznidazole (81), Alcantara and coworkers [70] made changes to the imidazole ring, moving the nitro group to position 4 and incorporating the methyl group in position 2 ( Figure 23). The authors added the methyl group based on

Trypanocidal Analogs of Benznidazole
To design new analogs of benznidazole (81), Alcantara and coworkers [70] made changes to the imidazole ring, moving the nitro group to position 4 and incorporating the methyl group in position 2 ( Figure 23). The authors added the methyl group based on studies showing that potency and solubility could be improved, and they moved the nitro group to position 4 based on results suggesting that such derivatives are non-toxic. In addition, the authors performed molecular hybridization based on the N-acylhydrazone cruzain inhibitor 82 (IC 50 = 0.6 µM). The imidazole-N-arylhydrazone hybrids were tested against trypomastigote forms, and the results showed that the 4-chlorophenyl derivative (83) had the best trypanocidal activity with an IC 50 of 206.98 µM [70]. studies showing that potency and solubility could be improved, and they moved the nitro group to position 4 based on results suggesting that such derivatives are non-toxic. In addition, the authors performed molecular hybridization based on the N-acylhydrazone cruzain inhibitor 82 (IC50 = 0.6 µM). The imidazole-N-arylhydrazone hybrids were tested against trypomastigote forms, and the results showed that the 4-chlorophenyl derivative (83) had the best trypanocidal activity with an IC50 of 206.98 µM [70].

Antibacterial Agents
Based on previously described β-ketoacyl acyl carrier protein synthase (FabH) inhibitors, compounds 84-86 [71], a series of furoxan-sulfonylhydrazone derivatives (87) were designed as new antibacterial agents ( Figure 24) [72]. From the SAR studies, compound 88 was identified as the most potent of the series, in which the methyl group proved to be an important structural feature compared to other substituents [72].

Phosphonate Derivatives as Anticancer Agents
A series of bis-(3-indolyl)methane phosphonate derivatives were synthesized as anticancer agents (89-92). Overall, compounds methylated at position 5 of the bis-indole core (89 and 91) showed increased potency for inhibiting cell proliferation of ovarian and lung cancer cell lines compared to unmethylated analogs (90 and 92) (Figure 25) [73].

Antibacterial Agents
Based on previously described β-ketoacyl acyl carrier protein synthase (FabH) inhibitors, compounds 84-86 [71], a series of furoxan-sulfonylhydrazone derivatives (87) were designed as new antibacterial agents ( Figure 24) [72]. From the SAR studies, compound 88 was identified as the most potent of the series, in which the methyl group proved to be an important structural feature compared to other substituents [72]. studies showing that potency and solubility could be improved, and they moved the nitro group to position 4 based on results suggesting that such derivatives are non-toxic. In addition, the authors performed molecular hybridization based on the N-acylhydrazone cruzain inhibitor 82 (IC50 = 0.6 µM). The imidazole-N-arylhydrazone hybrids were tested against trypomastigote forms, and the results showed that the 4-chlorophenyl derivative (83) had the best trypanocidal activity with an IC50 of 206.98 µM [70].

Antibacterial Agents
Based on previously described β-ketoacyl acyl carrier protein synthase (FabH) inhibitors, compounds 84-86 [71], a series of furoxan-sulfonylhydrazone derivatives (87) were designed as new antibacterial agents ( Figure 24) [72]. From the SAR studies, compound 88 was identified as the most potent of the series, in which the methyl group proved to be an important structural feature compared to other substituents [72].

Methylation Effect on Aqueous Solubility
A series of N-acylhydrazone derivatives were designed as HDAC6/8-selective inhibitors for cancer treatment [74]. The series was designed from the natural product trichostatin A (93) using bioisosteric replacement [6,7] and conformational restriction [5] strategies. The most potent compounds in the series were 94 and 95, which differed structurally by a single methyl group ( Figure 26). In this case, the magic methyl did not significantly change the activity, but the aqueous solubility was significantly increased by its presence [74], which is probably a consequence of the strong conformational effect caused by the N-methylation of N-acylhydrazone derivatives [75].

Methylation Effect on Plasma Stability
In the next case study, morpholin-2-one derivatives (96-99) were identified as fungicidal agents against Candida and Aspergillus species (Figure 27) [76]. However, the development of this series was hampered by low plasmatic stability, probably related to lactone hydrolysis. The introduction of methyl groups at the 6-position of the morpholin-2-one scaffold (96-99) led to a significant improvement in plasmatic stability while maintaining in vitro antifungal activity. The gemdimethyl derivative 99 was the most stable derivative as a consequence of the higher steric hindrance of lactone hydrolysis [76].

Methylation Effect on Aqueous Solubility
A series of N-acylhydrazone derivatives were designed as HDAC6/8-selective inhibitors for cancer treatment [74]. The series was designed from the natural product trichostatin A (93) using bioisosteric replacement [6,7] and conformational restriction [5] strategies. The most potent compounds in the series were 94 and 95, which differed structurally by a single methyl group ( Figure 26). In this case, the magic methyl did not significantly change the activity, but the aqueous solubility was significantly increased by its presence [74], which is probably a consequence of the strong conformational effect caused by the N-methylation of N-acylhydrazone derivatives [75].

Methylation Effect on Aqueous Solubility
A series of N-acylhydrazone derivatives were designed as HDAC6/8-selective inhibitors for cancer treatment [74]. The series was designed from the natural product trichostatin A (93) using bioisosteric replacement [6,7] and conformational restriction [5] strategies. The most potent compounds in the series were 94 and 95, which differed structurally by a single methyl group ( Figure 26). In this case, the magic methyl did not significantly change the activity, but the aqueous solubility was significantly increased by its presence [74], which is probably a consequence of the strong conformational effect caused by the N-methylation of N-acylhydrazone derivatives [75].

Methylation Effect on Plasma Stability
In the next case study, morpholin-2-one derivatives (96-99) were identified as fungicidal agents against Candida and Aspergillus species (Figure 27) [76]. However, the development of this series was hampered by low plasmatic stability, probably related to lactone hydrolysis. The introduction of methyl groups at the 6-position of the morpholin-2-one scaffold (96-99) led to a significant improvement in plasmatic stability while maintaining in vitro antifungal activity. The gemdimethyl derivative 99 was the most stable derivative as a consequence of the higher steric hindrance of lactone hydrolysis [76].

Methylation Effect on Plasma Stability
In the next case study, morpholin-2-one derivatives (96-99) were identified as fungicidal agents against Candida and Aspergillus species (Figure 27) [76]. However, the development of this series was hampered by low plasmatic stability, probably related to lactone hydrolysis. The introduction of methyl groups at the 6-position of the morpholin-2-one scaffold (96-99) led to a significant improvement in plasmatic stability while maintaining in vitro antifungal activity. The gemdimethyl derivative 99 was the most stable derivative as a consequence of the higher steric hindrance of lactone hydrolysis [76].

Methylation Effect on hERG Potassium Channel Inhibition
Jin and coworkers [77] reported the introduction of methyl groups into the aminopropylamine chain of compound 100 (checkpoint kinase 1-CHK1 IC50 = 20.9 nM) to provide a series of CHK1 inhibitors. These compounds showed excellent inhibitory activity, and compound 101 was the most potent (CHK1 IC50 = 16.1 nM). Additionally, 101 showed reduced inhibition of the human ether-à-go-go-related (hERG) potassium channel (35.5% at 10 µM) compared to 100 (43.4% at 10 µM) ( Figure 28). Furthermore, the authors suggested that the introduction of the gem-dimethyl group improved in vivo metabolic stability compared to linear amines [77]. In another study, Ma and colleagues [78] designed analogs of the mu opioid receptor (MOR) ligand NAN (102) [79], a 6α-N-7′-indolyl-substituted naltrexamine derivative, which showed promising pharmacological effects but had significant hERG potassium channel liability ( Figure 29). According to in vivo morphine-induced antinociception assays, compound 103 was the most potent antagonist. This compound (103) bears a methyl group at the 2′ position of the indole ring and had a sevenfold lower potency for hERG potassium channel inhibition compared to NAN (101) [78].

Methylation Effect on hERG Potassium Channel Inhibition
Jin and coworkers [77] reported the introduction of methyl groups into the aminopropylamine chain of compound 100 (checkpoint kinase 1-CHK1 IC 50 = 20.9 nM) to provide a series of CHK1 inhibitors. These compounds showed excellent inhibitory activity, and compound 101 was the most potent (CHK1 IC 50 = 16.1 nM). Additionally, 101 showed reduced inhibition of the human ether-à-go-go-related (hERG) potassium channel (35.5% at 10 µM) compared to 100 (43.4% at 10 µM) ( Figure 28). Furthermore, the authors suggested that the introduction of the gem-dimethyl group improved in vivo metabolic stability compared to linear amines [77].

Methylation Effect on hERG Potassium Channel Inhibition
Jin and coworkers [77] reported the introduction of methyl groups into the aminopropylamine chain of compound 100 (checkpoint kinase 1-CHK1 IC50 = 20.9 nM) to provide a series of CHK1 inhibitors. These compounds showed excellent inhibitory activity, and compound 101 was the most potent (CHK1 IC50 = 16.1 nM). Additionally, 101 showed reduced inhibition of the human ether-à-go-go-related (hERG) potassium channel (35.5% at 10 µM) compared to 100 (43.4% at 10 µM) ( Figure 28). Furthermore, the authors suggested that the introduction of the gem-dimethyl group improved in vivo metabolic stability compared to linear amines [77]. In another study, Ma and colleagues [78] designed analogs of the mu opioid receptor (MOR) ligand NAN (102) [79], a 6α-N-7′-indolyl-substituted naltrexamine derivative, which showed promising pharmacological effects but had significant hERG potassium channel liability ( Figure 29). According to in vivo morphine-induced antinociception assays, compound 103 was the most potent antagonist. This compound (103) bears a methyl group at the 2′ position of the indole ring and had a sevenfold lower potency for hERG potassium channel inhibition compared to NAN (101) [78].  In another study, Ma and colleagues [78] designed analogs of the mu opioid receptor (MOR) ligand NAN (102) [79], a 6α-N-7 -indolyl-substituted naltrexamine derivative, which showed promising pharmacological effects but had significant hERG potassium channel liability ( Figure 29). According to in vivo morphine-induced antinociception assays, compound 103 was the most potent antagonist. This compound (103) bears a methyl group at the 2 position of the indole ring and had a sevenfold lower potency for hERG potassium channel inhibition compared to NAN (101) [78].

Methylation Effect on hERG Potassium Channel Inhibition
Jin and coworkers [77] reported the introduction of methyl groups into the aminopropylamine chain of compound 100 (checkpoint kinase 1-CHK1 IC50 = 20.9 nM) to provide a series of CHK1 inhibitors. These compounds showed excellent inhibitory activity, and compound 101 was the most potent (CHK1 IC50 = 16.1 nM). Additionally, 101 showed reduced inhibition of the human ether-à-go-go-related (hERG) potassium channel (35.5% at 10 µM) compared to 100 (43.4% at 10 µM) ( Figure 28). Furthermore, the authors suggested that the introduction of the gem-dimethyl group improved in vivo metabolic stability compared to linear amines [77]. In another study, Ma and colleagues [78] designed analogs of the mu opioid receptor (MOR) ligand NAN (102) [79], a 6α-N-7′-indolyl-substituted naltrexamine derivative, which showed promising pharmacological effects but had significant hERG potassium channel liability ( Figure 29). According to in vivo morphine-induced antinociception assays, compound 103 was the most potent antagonist. This compound (103) bears a methyl group at the 2′ position of the indole ring and had a sevenfold lower potency for hERG potassium channel inhibition compared to NAN (101) [78].  Figure 29. The methylation effect in hERG inhibition profile optimization of mu opioid ligands [78].

Methylation Effect on Metabolism
Liu and coworkers [80] reported the modulation of linkers of phosphoinositide 3kinase delta inhibitors and found that by introducing the "magic methyl" group they had the best balance between oxidative metabolism, stability, and potency. The quinazolinone derivative 104 showed significant inhibitory potency on PI3Kδ with an IC 50 value of 0.008 µM (Figure 30). However, compound 104 showed a high clearance with a Cl int value of 21.80 µL/mg/min in human liver microsomes (HLMs). Additional metabolite identification studies of compound 104 revealed that oxidation of the five-membered pyrrolidine linker was the main soft spot for metabolic reactions. This led to the design of new analogs of 104, resulting in compound 105, which demonstrated favorable bioavailability in Sprague-Dawley rats following intravenous and oral treatment. In addition, compound 105 had a PI3Kδ IC 50 of 0.014 µM and activated basophils and B cells and was effective in a collagen-induced arthritis model [80]. Liu and coworkers [80] reported the modulation of linkers of phosphoinositide 3kinase delta inhibitors and found that by introducing the "magic methyl" group they had the best balance between oxidative metabolism, stability, and potency. The quinazolinone derivative 104 showed significant inhibitory potency on PI3Kδ with an IC50 value of 0.008 µM ( Figure 30). However, compound 104 showed a high clearance with a Clint value of 21.80 µL/mg/min in human liver microsomes (HLMs). Additional metabolite identification studies of compound 104 revealed that oxidation of the five-membered pyrrolidine linker was the main soft spot for metabolic reactions. This led to the design of new analogs of 104, resulting in compound 105, which demonstrated favorable bioavailability in Sprague-Dawley rats following intravenous and oral treatment. In addition, compound 105 had a PI3Kδ IC50 of 0.014 µM and activated basophils and B cells and was effective in a collagen-induced arthritis model [80].

Perspectives
Analyzing the state of the art in the use of the methyl effect in medicinal chemistry, it is evident that its applicability to the discovery and optimization of new small-molecule drug candidates is indisputable. In this review, the importance of this group for improving pharmacodynamic properties has been discussed, highlighted by the discovery of the recently approved anticancer drug tazemetostat (8), where the authors found that four methyl groups inserted at different positions resulted in a stunning >100,000-fold improvement in activity. Indeed, there are many examples focusing on the effect of methylation on the pharmacodynamic properties of bioactive molecules. However, in this review some examples of the influence of the methyl group on the pharmacokinetic and physicochemical profile of drug candidates have been presented, covering its use to block metabolic soft spots, reduce hERG liability, improve aqueous solubility, and increase plasma stability. From its participation in the molecular recognition process of pharmacological targets to the modulation of ADMET properties, the "magic methyl" never ceases to surprise us. We hope that the key examples discussed here will help the scientific community to further understand either the relationship between the structure and biological activity of new chemical entities or the rational application of methylation and what can be expected from this process.

Perspectives
Analyzing the state of the art in the use of the methyl effect in medicinal chemistry, it is evident that its applicability to the discovery and optimization of new small-molecule drug candidates is indisputable. In this review, the importance of this group for improving pharmacodynamic properties has been discussed, highlighted by the discovery of the recently approved anticancer drug tazemetostat (8), where the authors found that four methyl groups inserted at different positions resulted in a stunning >100,000-fold improvement in activity. Indeed, there are many examples focusing on the effect of methylation on the pharmacodynamic properties of bioactive molecules. However, in this review some examples of the influence of the methyl group on the pharmacokinetic and physicochemical profile of drug candidates have been presented, covering its use to block metabolic soft spots, reduce hERG liability, improve aqueous solubility, and increase plasma stability. From its participation in the molecular recognition process of pharmacological targets to the modulation of ADMET properties, the "magic methyl" never ceases to surprise us. We hope that the key examples discussed here will help the scientific community to further understand either the relationship between the structure and biological activity of new chemical entities or the rational application of methylation and what can be expected from this process.