End-to-end sequence-structure-function meta-learning predicts genome-wide chemical-protein interactions for dark proteins.

Systematically discovering protein-ligand interactions across the entire human and pathogen genomes is critical in chemical genomics, protein function prediction, drug discovery, and many other areas. However, more than 90% of gene families remain "dark"-i.e., their small-molecule ligands are undiscovered due to experimental limitations or human/historical biases. Existing computational approaches typically fail when the dark protein differs from those with known ligands. To address this challenge, we have developed a deep learning framework, called PortalCG, which consists of four novel components: (i) a 3-dimensional ligand binding site enhanced sequence pre-training strategy to encode the evolutionary links between ligand-binding sites across gene families; (ii) an end-to-end pretraining-fine-tuning strategy to reduce the impact of inaccuracy of predicted structures on function predictions by recognizing the sequence-structure-function paradigm; (iii) a new out-of-cluster meta-learning algorithm that extracts and accumulates information learned from predicting ligands of distinct gene families (meta-data) and applies the meta-data to a dark gene family; and (iv) a stress model selection step, using different gene families in the test data from those in the training and development data sets to facilitate model deployment in a real-world scenario. In extensive and rigorous benchmark experiments, PortalCG considerably outperformed state-of-the-art techniques of machine learning and protein-ligand docking when applied to dark gene families, and demonstrated its generalization power for target identifications and compound screenings under out-of-distribution (OOD) scenarios. Furthermore, in an external validation for the multi-target compound screening, the performance of PortalCG surpassed the rational design from medicinal chemists. Our results also suggest that a differentiable sequence-structure-function deep learning framework, where protein structural information serves as an intermediate layer, could be superior to conventional methodology where predicted protein structures were used for the compound screening. We applied PortalCG to two case studies to exemplify its potential in drug discovery: designing selective dual-antagonists of dopamine receptors for the treatment of opioid use disorder (OUD), and illuminating the understudied human genome for target diseases that do not yet have effective and safe therapeutics. Our results suggested that PortalCG is a viable solution to the OOD problem in exploring understudied regions of protein functional space.

Fluoroalkoxylated C-3 and C-9 (S)-12-bromostepholidine analogues with D1R antagonist activity.

To illuminate the tolerance of fluoroalkoxylated groups at the C-3 and C-9 positions of tetrahydroprotoberberines (THPBs) on D1R activity, C-3 and C-9 fluoroalkoxylated analogues of (S)-12-bromostepholidine were prepared and evaluated. All compounds examined were D1R antagonists as measured by a cAMP assay. Our structure-activity studies herein indicate that the C-3 position tolerates a 1,1-difluoroethoxy substituent for D1R antagonist activity. Compound 13a was the most potent cAMP-based D1R antagonist identified and was also found to antagonize ß-arrestin translocation in a TANGO assay. Affinity assessments at other dopamine receptors revealed that 13a is selective for D1R and unlike other naturally-occurring THPBs such as (S)-stepholidine, lacks D2R affinity. In preliminary biopharmaceutical assays, excellent BBB permeation was observed for 13a. Further pharmacological studies are warranted on (S)-stepholidine congeners to harvest their potential as a source of novel, druggable D1R-targeted agents.

Structure-Activity Relationship Studies on 6-Chloro-1-phenylbenzazepines Leads to the Identification of a New Dopamine D1 Receptor Antagonist.

The 1-phenylbenzazepine template has yielded a number of D1R-like ligands, which, though useful as pharmacological tools, have significant drawbacks in terms of selectivity versus D5R as well as pharmacokinetic behavior. A number of 1-phenylbenzazepines contain a 6-chloro functional group, but extensive SAR studies around the 6-chloro-1-phenylbenzazepine framework have not been reported in the literature. To further understand the tolerance of the 6-chloro-1-phenylbenzazepine template for various substituent groups towards affinity and selectivity at D1R, we synthesized two series of analogs with structural variations at the C-7, C-8, N-3, C-3' and C-4' positions. The series 2 analogs differed from series 1 analogs in possessing a nitrogenated functionality at C-8 and lacked a C-4' substituent, but were otherwise similar. Analogs were assessed for affinity at D1R, D2R and D5R. For both series, we found that the analogs lacked affinity for D2R and showed modest D1R versus D5R selectivity. For series 1 analogs, an N-3 methyl substituent group was better tolerated than N-H or an N-3 allyl substituent. The C-8 position appears to be tolerant of amino and methanesulfonamide substituents for high D1R affinity, but C-8 amides displayed low to moderate D1R affinities. A C-3' methyl substituent appeared to be critical for the D1R affinity of some analogs, but the C-4' substituents tried (hydroxy and methoxy; series 1) did not result in any significant boost in D1R affinity. Compound 15a was the most potent and selective D1R ligand identified from these studies (Ki at D1R = 30 nM; 6-fold selectivity versus D5R). Further functional activity assessments indicate that 15a functions as a D1R antagonist towards cAMP-mediated signaling. The predicted drug-like properties of 15a are encouraging for further pharmacological assessments on the compound.

Semisynthetic Transformations on (+)-Boldine Reveal a 5-HT2A/2CR Antagonist.

Aporphine alkaloids have shown affinity for serotonin receptors (5-HTRs), and there has been a recent upsurge of interest in aporphines as 5-HT2CR ligands. 1,2,9,10-Tetraoxygenated aporphine alkaloids in particular have demonstrated good affinity for 5-HTRs. In continued efforts to understand the impacts of structural modification of the 1,2,9,10-tetraoxygenated aporphine template on affinity, selectivity, and activity at 5-HT2R subtypes, we used (+)-boldine (8) as a semisynthetic feedstock in the preparation of C-2-alkoxylated (+)-predicentrine analogues. Compound 10n, which contains a benzyloxy group at C-2, has been identified as a novel 5-HT2CR ligand with strong affinity (4 nM) and moderate selectivity versus 5-HT2BR and 5-HT2AR (12-fold and 6-fold, respectively). Compound 10n functions as an antagonist at 5-HT2A and 5-HT2C receptors. Computational experiments indicate that several hydrophobic interactions as well as strong H-bond and salt bridge interactions between the protonated amine moiety in 10n and Asp134 are responsible for the potent 5-HT2CR affinity of this compound. Furthermore, compound 10n displays favorable predicted drug-like characteristics, which is encouraging toward future optimization.

Further studies on C2'-substituted 1-phenylbenzazepines as dopamine D1 receptor ligands.

The 1-phenylbenzazepine scaffold has yielded several D1R targeting ligands, but some gaps remain in our understanding of the structure-activity relationships in this scaffold. In particular, there is a paucity of studies that have investigated the effects of substituents at the C2' position of 1-phenylbenzazepines on their affinity and selectivity towards D1R. In this study, a set of methyl- and fluoro- C2'-substituted 1-phenylbenzazepines, with ring A catechol or 8-hydroxy-7-methoxy moieties in tandem with N-methyl or N-allyl substituent groups, was synthesized and evaluated for affinity at a subset of dopamine receptors - D1R, D2R and D5R. These studies indicate that an N-methyl group is generally preferred over N-unsubstituted or N-allyl groups for strong D1R affinity. In addition, it was revealed that compounds with a ring A 8-hydroxy-7-methoxy motif displayed stronger D1R affinity than analogous compounds with a ring A catechol moiety. Furthermore, the presence of a C2' substituent does not significantly impact D1R selectivity over D5R. However, for all analogs assessed, D1R selectivity over D2R was maintained. D1R vs D5R selectivity was generally poor or modest (less than 10-fold) among members of the series. A new high affinity selective D1R ligand - 10b (Ki = 5.7 nM), was identified in this study; further pharmacological characterization indicates that 10b is an antagonist at D1R (IC50 = 10.7 nM). Docking studies on 10b indicate that a number of interactions with hydrophobic residues (Trp321, Val317, Phe313, Phe289, Phe288, Phe285, Phe203, Tyr194, Leu190, Ser188, His 164, Ile104, Val100 and Trp99) in addition to the typical N-Asp103 salt bridge are important for its D1R affinity.

New tetrahydroisoquinoline-based D3R ligands with an o-xylenyl linker motif.

The effect of rigidification of the n-butyl linker region of tetrahydroisoquinoline-containing D3R ligands via inclusion of an o-xylenyl motif was examined in this study. Generally, rigidification with an o-xylenyl linker group reduces D3R affinity and negatively impacts selectivity versus D2R for compounds possessing a 6-methoxy-1,2,3,4,-tetrahydroisoquinolin-7-ol primary pharmacophore group. However, D3R affinity appears to be regulated by the primary pharmacophore group and high affinity D3R ligands with 6,7-dihydroxy-1,2,3,4-tetrahydroisoquinoline and 6,7-dimethoxy-1,2,3,4-tetrahydroisoquinoline primary pharmacophore groups were identified. The results of this study also indicate that D3R selectivity versus the σ2R is dictated by the benzamide secondary pharmacophore group, this being facilitated with 4-substituted benzamides. Compounds 5s and 5t were identified as high affinity (Ki < 4 nM) D3R ligands. Docking studies revealed that the added phenyl ring moiety interacts with the Cys181 in D3R which partially accounts for the strong D3R affinity of the ligands.

Identification of C10 nitrogen-containing aporphines with dopamine D1 versus D5 receptor selectivity.

New aporphines containing C10 nitrogen substituents (viz. nitro, aniline or amide moieties), were synthesized and evaluated for affinity at human serotonin 5-HT1A and 5-HT2A receptors and at human dopamine D1, D2 and D5 receptors. Two series of analogs were investigated: series A which contain a sole C10 nitrogen substituent on the tetracyclic aporphine core and series B which are 1,2,10-trisubstituted aporphines. Remarkably, compounds from both series lacked affinity for the D5 receptor, thus attaining D1 versus D5 selectivity. Compound 20c was the most potent D1 ligand identified. Docking studies at D1 and D5 receptors indicate that the binding mode of 20c at the D1 receptor allows for stronger hydrophobic contacts, (primarily with Phe residues) as compared to the D5 receptor, accounting for its D1 versus D5 selectivity. Considering the lack of affinity for the D5 receptor (and low affinity at other receptors tested), compound 20c represents an interesting starting point for further structural diversification of aporphines as sub-type selective D1 receptor tools.

Synthesis and dopamine receptor pharmacological evaluations on ring C ortho halogenated 1-phenylbenzazepines.

A series of 1-phenylbenzazepines containing bromine or chlorine substituents at the ortho position of the appended phenyl ring (2'-monosubstituted or 2',6'- disubstituted patterns) were synthesized and evaluated for affinity towards dopamine D1R, D2R and D5R. As is typical of the 1-phenylbenzazepine scaffold, the compounds displayed selectivity towards D1R and D5R; analogs generally lacked affinity for D2R. Interestingly, 2',6'-dichloro substituted analogs showed modest D5R versus D1R selectivity whereas this selectivity was reversed in compounds with a 2'-halo substitution pattern. Compound 10a was identified as a D1R antagonist (Ki = 14 nM; IC50 = 9.4 nM).

Structural manipulation of aporphines via C10 nitrogenation leads to the identification of new 5-HT7AR ligands.

Aporphine alkaloids containing a C10 nitrogen motif were synthesized and evaluated for affinity at 5-HT1AR, 5-HT2AR, 5-HT6R and 5-HT7AR. Three series of racemic aporphines were investigated: 1,2,10-trisubstituted, C10 N-monosubstituted and compounds containing a C10 benzofused aminothiazole moiety. The 1,2,10-trisubstituted series of compounds as a group displayed modest selectivity for 5-HT7AR and also had moderate 5-HT7AR affinity. Compounds from the C10 N-monosubstituted series generally lacked affinity for 5-HT2AR and 5-HT6R and showed strong affinity for 5-HT1A or 5-HT7AR. Compounds in this series that contained an N6-methyl group were up to 27-fold selective for 5-HT7AR over 5-HT1AR, whereas compounds with an N6-propyl substituent showed a reversal in this selectivity. The C10 benzofused aminothiazole analogues showed a similar binding profile as the C10 N-monosubstituted series i.e. strong affinity for 5-HT1AR or 5-HT7AR, with selectivity between the two receptors being similarly influenced by N6-methyl or N6-propyl substituents. Compounds 29 and 34a exhibit high 5-HT7AR affinity, excellent selectivity versus dopamine receptors and function as antagonists in 5-HT7AR cAMP-based assays. Compounds 29 and 34a have been identified as new lead molecules for further tool and pharmaceutical optimization.

New halogenated tris-(phenylalkyl)amines as h5-HT2B receptor ligands.

A series of compounds in which various halogen substituents were incorporated into a phenyl ring of a tris-(phenylalkyl)amine scaffold, was synthesized and evaluated for affinity to h5-HT2 receptors. In general, all compounds were found to have good affinity for the 5-HT2B receptor and were selective over 5-HT2A and 5-HT2C receptors. Compound 9i was the most selective compound in this study and is the highest affinity 5-HT2B receptor ligand bearing a tris-(phenylalkyl)amine scaffold to date.

Tetrahydroprotoberberine alkaloids with dopamine and σ receptor affinity.

Two series of analogues of the tetrahydroprotoberberine (THPB) alkaloid (±)-stepholidine that (a) contain various alkoxy substituents at the C10 position and, (b) were de-rigidified with respect to (±)-stepholidine, were synthesized and evaluated for affinity at dopamine and σ receptors in order to evaluate effects on D3 and σ2 receptor affinity and selectivity. Small n-alkoxy groups are best tolerated by D3 and σ2 receptors. Among all compounds tested, C10 methoxy and ethoxy analogues (10 and 11 respectively) displayed the highest affinity for σ2 receptors as well as σ2 versus σ1 selectivity and also showed the highest D3 receptor affinity. De-rigidification of stepholidine resulted in decreased affinity at all receptors evaluated; thus the tetracyclic THPB framework is advantageous for affinity at dopamine and σ receptors. Docking of the C10 analogues at the D3 receptor, suggest that an ionic interaction between the protonated nitrogen atom and Asp110, a H-bond interaction between the C2 phenol and Ser192, a H-bond interaction between the C10 phenol and Cys181 as well as hydrophobic interactions of the aryl rings to Phe106 and Phe345, are critical for high affinity of the compounds.

An alternative synthesis and x-ray crystallographic confirmation of (-)-stepholidine.

A formal enantioselective synthesis of (-)-stepholidine that provides an alternative preparation of key lactone intermediate 2 is described. The stereostructure of (-)-stepholidine prepared via this method was confirmed by x-ray diffraction.

C4 phenyl aporphines with selective h5-HT(2B) receptor affinity.

A group of aporphine alkaloids related to (±)-nantenine (1) and bearing a C4 phenyl and various C1 or N-substituents, was synthesized and evaluated for affinity to h5-HT receptors. In general, unlike nantenine, the analogs lack affinity for the h5-HT(2A) receptor and other 5-HT receptors but bind selectively to the h5-HT(2B) receptor. With regards to 5-HT(2B) affinity, there appears to be a low tolerance for bulky C1 or N-substituents when the C4 phenyl moiety is present. Compound 5a had the highest 5-HT(2B) affinity of the compounds tested, was found to be an antagonist and is selective vs other CNS receptors.

Synthesis and evaluation of aporphine analogs containing C1 allyl isosteres at the h5-HT(2A) receptor.

A series of C1 aporphine analogs related to compound 5 and that contain substituted allylic, alkynyl, nitrile, ester and benzyl groups was synthesized and evaluated for affinity at h5HT2A and α1A receptors in functional activity assays that measure calcium release. The presence of branched allylic substituent groups diminished affinity for the h5HT2A receptor. Likewise, the alkynyl, nitrile and ester derivatives evaluated displayed lower 5-HT2A receptor affinity as compared to 5. Hydrophobic, steric and electronic effects impact the affinity of p-substituted benzyl derivatives 8i-8k but in different ways. High hydrophobicity and size favor 5-HT2A affinity whereas, high electronegativity disfavors 5-HT2A affinity. p-Bromobenzyl analog 8k was identified as a 5-HT2A receptor selective ligand, with the highest 5-HT2A receptor affinity of any aporphine known to date. Most of the other analogs were selective for the 5-HT2A versus the α1A receptor. ChemScore binding energies from docking studies correlated qualitatively with the observed trends in affinity for 8i-8k, although the binding energies were not well differentiated quantitatively.

Semisynthetic Studies on and Biological Evaluation of N-Methyllaurotetanine Analogues as Ligands for 5-HT Receptors.

N-Methyllaurotetanine (1) has been reported to display good affinity for the 5-HT1A receptor, but no structure-affinity studies have been performed to date. The commercially available alkaloid boldine (2) was used as the starting material for synthesis of various C-9 alkoxy analogues of N-methyllaurotetanine in order to gauge the effect of C-9 alkylation on affinity and selectivity at 5-HT1A, 5-HT2A, and 5-HT7 receptors. Mitsunobu reactions were implemented in the alkylation steps leading to the analogues. Modest improvement in 5-HT1A affinity was observed upon alkylation for most analogues. Thus, the C-9 hydroxy group of 1 is not critical for affinity to the 5-HT1A receptor. Some analogues displayed high affinity for the 5-HT7 receptor, comparable to N-methyllaurotetanine, with moderate selectivity vs 5-HT1A and 5-HT2A receptors.

A divergent route to 9, 10-oxygenated tetrahydroprotoberberine and 8-oxoprotoberberine alkaloids: synthesis of (±)-isocorypalmine and oxypalmatine.

A new route which is germane to the synthesis of 9,10-oxygenated tetrahydroprotoberberines and 8-oxoprotoberberines is described. The route features the use of a diester (14) generated from reaction of dimethylmalonate with an aryl halide in the presence of n-butyllithium. The amide 17 prepared in subsequent steps is a versatile precursor for the synthesis of tetrahydroprotoberberine and 8-oxoprotoberberine scaffolds using standard high-yielding reactions. In this manner, (±)-isocorypalmine and oxypalmatine have been synthesized in 23% and 22 % yields respectively.

Evaluation of structural effects on 5-HT(2A) receptor antagonism by aporphines: identification of a new aporphine with 5-HT(2A) antagonist activity.

A set of aporphine analogs related to nantenine was evaluated for antagonist activity at 5-HT2A and α1A adrenergic receptors. With regards to 5-HT2A receptor antagonism, a C2 allyl group is detrimental to activity. The chiral center of nantenine is not important for 5-HT2A antagonist activity, however the N6 nitrogen atom is a critical feature for 5-HT2A antagonism. Compound 12b was the most potent 5-HT2A aporphine antagonist identified in this study and has similar potency to previously identified aporphine antagonists 2 and 3. The ring A and N6 modifications examined were detrimental to α1A antagonism. A slight eutomeric preference for the R enantiomer of nantenine was observed in relation to α1A antagonism.

Natural Product-Inspired Dopamine Receptor Ligands.

Due to their evolutionary bias as ligands for biologically relevant drug targets, natural products offer a unique opportunity as lead compounds in drug discovery. Given the involvement of dopamine receptors in various physiological and behavioral functions, they are linked to numerous diseases and disorders such as Parkinson's disease, schizophrenia, and substance use disorders. Consequently, ligands targeting dopamine receptors hold considerable therapeutic and investigative promise. As this perspective will highlight, dopamine receptor targeting natural products play a pivotal role as scaffolds with unique and beneficial pharmacological properties, allowing for natural product-inspired drug design and lead optimization. As such, dopamine receptor targeting natural products still have untapped potential to aid in the treatment of disorders and diseases related to central nervous system (CNS) and peripheral nervous system (PNS) dysfunction.

A New Route to Azafluoranthene Natural Products via Direct Arylation.

Microwave-assisted direct arylation was successfully employed in the synthesis of azafluoranthene alkaloids for the first time. Direct arylation reactions on a diverse set of phenyltetrahydroisoquinolines produces the indeno[1,2,3-ij]isoquinoline nucleus en route to a high yielding azafluoranthene synthesis. The method was used as a key step in the efficient preparation of the natural products rufescine and triclisine. As demonstrated herein, this synthetic approach should be generally applicable to the preparation of natural and un-natural azafluoranthene alkaloids as well as "azafluoranthene-like" isoquinoline alkaloids.

Discovery of Selective Dopamine Receptor Ligands Derived from (-)-Stepholidine via C-3 Alkoxylation and C-3/C-9 Dialkoxylation.

We evaluated C-3 alkoxylated and C-3/C-9 dialkoxylated (-)-stepholidine analogues to probe the tolerance at the C-3 and C-9 positions of the tetrahydroprotoberberine (THPB) template toward affinity for dopamine receptors. A C-9 ethoxyl substituent appears optimal for D1R affinity since high D1R affinities were observed for compounds that contain an ethyl group at C-9, with larger C-9 substituents tending to decrease D1R affinity. A number of novel ligands were identified, such as compounds 12a and 12b, with nanomolar affinities for D1R and no affinity for either D2R or D3R, with compound 12a being identified as a D1R antagonist for both G-protein- and ß-arrestin-based signaling. Compound 23b was identified as the most potent and selective D3R ligand containing a THPB template to date and functions as an antagonist for both G-protein- and ß-arrestin-based signaling. Molecular docking and molecular dynamics studies validated the D1R and D3R affinity and selectivity of 12a, 12b, and 23b.