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

Abstract

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 (K_i < 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.

Graphical abstract:

There are five subtypes of dopamine receptor (D₁R-D₅R) and these are divided based on structural similarity and pharmacological properties into two subgroups – D₁-like which comprises D₁R and D₅R and D₂-like which includes D₂R, D₃R and D₄R. All dopamine receptors are G-protein-receptor coupled and recognize the endogenous neurotransmitter dopamine. D₁-like receptors signal via activation of G_s pathways whereas D₂-like receptors couple to G_i inhibitory proteins.^1–3 Activation of dopamine receptors endogenously is known to play a role in a myriad of physiological effects including movement, cognition and addiction-related behaviors.^{4, 5} There is a high density of D₃R in the mesolimbic region of the brain, an area associated with motivation and drug seeking behaviors.⁶ Thus, the D₃R has received considerable attention, especially over the 2 past decades as a potential target for the treatment of psychostimulant addiction.⁷ In that regard, antagonism or partial agonism by selective D₃R ligands is desirable.⁸ D₃R antagonists have also been shown to increase cognitive performance and reverse cognitive deficits in animal studies and are thus promising as antipsychotic agents.^9–11

Over years of research, there have been significant challenges with obtaining D₃R ligands that are suitable for translational research. Selectivity, especially versus the closely related D₂R and D₄R is one such issue, although there are now available several D₃R subtype selective ligands.¹² However, several D₃R ligands have other drawbacks such as poor oral bioavailability which limits their utility as therapeutics.

Several D₃R ligands have been obtained that conform to a classical D₃R pharmacophore which consists of three main regions: (i) an amine-containing primary pharmacophore region, (ii) a linker region – typically an n-butyl chain and (iii) an arylamide secondary pharmacophore moiety (Fig. 1). Various amine-containing primary pharmacophore groups have been used, among which the phenylpiperazine moiety has been fairly common (e.g. BP 897 and NGB2904, Fig. 1). A number of D₃R ligands have been discovered that contain a 6,7-dimethoxy-1,2,3,4-tetrahydroisoquinoline primary pharmacophore group. However, in addition to binding to D₃R, compounds with this motif have also been reported to possess significant affinity for the σ₂ receptor (σ₂R)¹³. In fact, a number of these 6,7-dimethoxy-1,2,3,4-tetrahydroisoquinoline-containing compounds have been explored as σ₂R selective ligands and as positron emission tomography (PET) cancer imaging agents.^14–18 We recently reported new D₃R ligands that contain a slight variation of this template in containing a 6-methoxy-1,2,3,4,-tetrahydroisoquinolin-7-ol motif.¹⁹ We found that compounds with this motif (e.g. compound 1, Fig. 2) exhibited good selectivity for D₃R over σ₂R (see ref 19 and Table S1 in Supporting Information) and may thus be advantageous as compared to compounds with the 6,7-dimethoxy-1,2,3,4-tetrahydroisoquinoline moiety with regards to retaining D₃R selectivity.

Figure 1.

Structures of the D₃R antagonist pharmacophore and selected D₃R antagonist ligands

Figure 2.

Structures of typical tetrahydroisoquinoline-containing D₃R and σ₂R ligands - RHM-1–86²⁰ and compound 1¹⁹

Previous studies have investigated rigidification of the flexible n-butyl linker region of other D₃R scaffolds (particularly those containing phenyl piperazine-based primary pharmacophore groups) with moieties including cis-alkenyl, trans-alkenyl, alkynyl, xylenyl and cycloalkyl motifs with varying results.^21–24 We were curious to find out the extent to which rigidification of the n-butyl linker unit of a D₃R scaffold containing a 6-methoxy-1,2,3,4,-tetrahydroisoquinolin-7-ol and related tetrahydroisoquinolyl primary pharmacophore groups may lead to improvements in D₃R affinity and potentially increased selectivity versus σ₂R. We considered rigidification with an o-xylenyl motif as this would: i) preserve the 4-atom connectivity between the amine of the primary pharmacophore group and the nitrogen of the arylamide group and ii) preserve the basicity of the nitrogen in the tetrahydroisoquinoline region which is deemed to be necessary towards formation of a critical salt bridge interaction with D₃R.¹⁹ The synthesized compounds were pharmacologically characterized at dopamine D₁R – D₅R and at the σ₂R. Data on these evaluations as well as receptor docking studies of the ligands at D₃R are discussed herein.

In terms of structural diversity of the designed analogues, we targeted the synthesis of compounds with the following structural variations: i) primary pharmacophore group - either a 6-methoxy-1,2,3,4,-tetrahydroisoquinolin-7-ol, 6,7-dihydroxy-1,2,3,4-tetrahydroisoquinoline or 6,7-dimethoxy-1,2,3,4-tetrahydroisoquinoline motif, ii) linker – o-xylene unit and iii) secondary pharmacophore region – a variety of arylamide units, including some found in previously identified D₃R ligands. The compounds were synthesized as outlined in Scheme 1.

Scheme 1.

Synthesis of analogues 5a-t and 6a-c

Reagents and conditions: (a) Appropriate aldehyde, NaBH(OAc)₃, DCM, 12 h, rt, 69–72%; (b) LiAlH₄, THF, 1h, rt, 60–63%; (c) HBTU, appropriate carboxylic acid, TEA, 12 h, rt; (d) conc. HCl, CH₃COOH, 12 h, 40 °C, 55–63% (over two steps); (e) BBr₃, DCM, 0°C, 2h, 87–90%

To commence the synthesis of compounds containing a 6-methoxy-1,2,3,4,-tetrahydroisoquinolin-7-ol motif in tandem with the o-xylenyl linker unit, readily available compound 2a was subjected to reductive amination conditions with 2-formylbenzonitrile yielding 3a. Compound 3a was subsequently reduced with lithium aluminium hydride to afford the amine 4a. Compound 4a in turn underwent acid-amine coupling with various carboxylic acid groups to afford intermediate amides which were debenzylated under acidic conditions (without purification of the intermediate amide) to give the target analogues 5a – r. Analogues 5s and 5t with the 6,7-dimethoxy-1,2,3,4-tetrahydroisoquinoline motif were prepared from 2b in an analogous route to that described for analogues 5a – r (except for the acidic cleavage step). Compounds 5g, 5s and 5t were treated with BBr₃ to effect O-demethylation, thus yielding catechol analogues 6a, 6b and 6c respectively.

The results of radioligand binding evaluations on analogues containing the o-xylenyl linker unit are compiled in Table 1. In general, compounds in this series lacked affinity for D₁R and D₅R. Compound 5a which contains a 4-fluorophenyl arylamide motif was devoid of affinity at all receptors tested. The 2-naphthylamide-containing 5b had only moderate affinity for D₃R (K_i =780 nM); affinity was still moderate but approximately twice as high at D₂R. No affinity was detected for the σ₂R. Compounds with a 2-chloro, 2-bromo and 2-methoxy mono-substituted benzamide motif (i.e. 5c – 5e) displayed moderate affinity for D₃R (K_i = 120 – 540 nM). In all cases these compounds were slightly more selective for D₂R. Affinities generally increased in the order D₄R<D₃R<D₂R. These analogues also had moderate affinity for the σ₂R. Compounds 5f – 5h with the 3-chloro, 3-bromo and 3-methoxy benzamide motifs, displayed a similar binding profile and selectivity trends at D₂R, D₃R, D₄R and σ₂R as their previously described 2-substituted congeners. Compound 5i, with a 3-cyano substituent however displayed strong affinity for D₃R (K_i = 26 nM) with comparable affinity for D₂R and good selectivity versus D₄R (>100-fold). This compound lacked any appreciable affinity for the σ₂R. Compounds 5j – 5m with a 4-halophenyl moiety in the benzamide motif, had similar binding profiles in that they all lacked affinity for D₄R and the σ₂R. Compounds 5j – 5m showed good affinity for D₂R and D₃R; compound 5m had good D₂R affinity (84 nM) but lacked D₃R affinity and was the compound with the best D₂R selectivity identified in this study. Replacement of the 4-halo substituent groups with 4-methoxy or 4-cyano groups (compounds 5n and 5o respectively) resulted in a rebound in (albeit weak: K_i = 1040 – 2800 nM) D₄R affinity. The 2,3-dichloro benzamide analogue 5p displayed a binding profile that was reminiscent of the 2- and 3-substituted analogues described above. Interestingly, the 2,3-dimethoxy benzamide analogue (5q) showed complete selectivity for D₃R (K_i = 57 nM) with no affinity at the other receptors evaluated. This compound has the best D₃R selectivity profile of all compounds evaluated in this study. The sole 3,4-disubstituted benzamide tested (5r) showed strong affinity at D₃R (K_i = 24 nM), comparable to that at D₂R and with low D₁R affinity (K_i = 1970 nM).

Table 1.

Binding affinity of o-xylenyl ring rigidified analogues at dopamine and σ₂ receptors

				Ki ± SEM (nM)a
Cmpd.	R1	R2	R3	D1Rb	D2Rc	D3Rc	D4Rc	D5Rb	σ2d
5a	Me	H		nae	na	na	na	na	na
5b	Me	H		na	310 ± 40	780 ± 100	na	na	na
5c	Me	H		na	190 ± 25	380 ± 49	3900 ± 500	na	440 ± 57
5d	Me	H		na	420 ± 54	540 ± 70	880 ± 110	na	360 ± 46
5e	Me	H		na	78 ± 10	120 ± 15	950 ± 120	na	140 ± 18
5f	Me	H		na	100 ± 13	140 ± 18	950 ± 110	na	240 ± 31
5g	Me	H		na	150 ± 19	170 ± 22	na	na	290 ± 37
5h	Me	H		na	140 ± 18	280 ± 36	1070 ± 1 40	na	610 ± 79
5i	Me	H		na	39 ± 5	26 ± 3.4	3360 ± 430	na	>10000
5j	Me	H		na	52 ± 6.7	140 ± 18	na	na	na
5k	Me	H		na	47 ± 6.1	55 ± 7.1	na	na	na
5l	Me	H		na	76 ± 9.8	160 ± 21	na	na	na
5m	Me	H		na	84 ± 11	na	na	na	na
5n	Me	H		na	74 ± 9.5	180 ± 23	1040 ± 130	na	1400 ± 180
5o	Me	H		na	190 ± 25	310 ± 40	2800 ± 360	na	na
5p	Me	H		na	280 ± 36	460 ± 5.9	1150 ± 150	na	480 ± 62
5q	Me	H		na	na	57 ± 7.4	na	na	na
5r	Me	H		1970 ± 250	27 ± 3.5	24 ± 3.1	na	na	na
5s	Me	Me		510 ± 66	46 ± 5.9	1.2 ± 0.2	57 ± 7.4	300 ± 39	na
5t	Me	Me		58 ± 7.5	50 ± 6.5	3.4 ± 0.4	140 ± 18	340 ± 44	na
6a	H	H		91 ± 12	56 ± 7.2	2.0 ± 0.2	101 ± 13	na	na
6b	H	H		na	900 ± 120	370 ± 48	na	na	na
6c	H	H		na	83 ± 11	28 ± 3.6	-	na	na
(+)-butac1amo1				2.84 ± 0.05
Haloperidol					2.61 ± 0.03				7.21 ± 0.9
Nemonapride						0.86 ± 0.04	0.75 ± 0.06
SKF 83566								1.75 ± 0.1

^aExperiments conducted in triplicate;

^b[3H]-SCH23390 used as radioligand;

^c[3H]N-methylspiperone used as radioligand;

^d[3H]DTG used as radioligand;

^ena- not active- ligands displayed <10% inhibition in a primary assay at a ligand concentration of 10 μM

Two compounds with a 6,7-dimethoxy-1,2,3,4-tetrahydroisoquinoline motif were evaluated – the 3-cyano and 4-cyano benzamide derivatives 5s and 5t respectively. These compounds displayed very high affinity for D₃R (K_i = 1.2 and 3.4 nM; among the most potent D₃R ligands in this study) and exhibited selectivities ranging from 15- to 420-fold versus the other dopamine receptors tested. Interestingly, both compounds lacked σ₂R affinity.

Among the compounds with a 6,7-dihydroxy-1,2,3,4-tetrahydroisoquinoline unit, compound 6a and 6c displayed the highest D₃R affinity. In fact, compound 6a had one of the highest D₃R affinities (2 nM) of all compounds evaluated and had no affinity for D₅R and σ₂R. Selectivity of 6a versus D₁R, D₂R and D₄R was modest (<50-fold). In comparing the catechol-containing analogues 6b and 6c with their dimethoxy counterparts (5s and 5t respectively) and 7-hydroxy-6-methoxy congeners (5i and 5o respectively), it is apparent that D₃R affinity is better tolerated with the dimethoxy functionality than either the catechol or 7-hydroxy-6-methoxy motifs.

On a whole, it appears that 4-substituents on the benzamide moiety are less likely to promote binding to the σ₂R as compared to their 2- or 3-substituted benzamide counterparts. However, compounds with a 2,3-disubstituted benzamide motif do not necessarily follow the same trend as for 2- and 3-monosubstitued benzamides. In that regard, compound 5q stands out in maintaining selectivity versus all other receptors including the σ₂R.

A comparison of the previously evaluated 6-methoxy-1,2,3,4,-tetrahydroisoquinolin-7-ol – containing ligands with a flexible linker group (e.g. 1) with their congeners containing the rigidifying o-xylenyl motif in the present study, also allows some general conclusions to be drawn about the impact of the introduced o-xylenyl sub-structure on D₃R affinity and selectivity. In our previous study the compounds with flexible linker groups displayed strong D₃R affinity (ranging from 2– 28 nM) with moderate or no affinity for other dopamine (See Table S1 in Supporting Information for data on comparator compounds from our previous work).¹⁹ However, in this study the highest D₃R affinity seen for a rigidified compound with the 6-methoxy-1,2,3,4,-tetrahydroisoquinolin-7-ol motif was 26 nM (compound 5i); most such compounds in the present study had affinities >100 nM at the D₃R. Generally then, introduction of the o-xylenyl linker group resulted in a reduction in D₃R affinity receptors (see Table S1). It appears that the inclusion of the o-xylenyl ring improves D₂R affinity as most compounds with flexible linker groups (i.e. from our previous study) had low or no affinity unlike analogous rigidified compounds in the present study. Taken together, the data indicate that the rigidifying o-xylenyl ring motif is detrimental towards D₃R versus D₂R selectivity. One area where the SAR of the flexible and rigid compounds diverged in a positive sense was with regards to D₄R affinity of the 4-halo-substituted benzamide derivatives. With the flexible analogues we found moderate D₄R affinity (95 – 1500 nM) for this sub-group of compounds; for the analogous rigidified compounds there was no D₄R affinity.¹⁹ Therefore, the inclusion of the o-xylenyl ring in this instance is beneficial in promoting D₃R selectivity versus D₄R. There does not seem to be any major impact on D₁R or D₅R affinity as both the flexible and rigidified compounds generally lacked D₁R and D₅R affinity.

A comparison of the 6,7-dimethoxy-1,2,3,4-tetrahydroisoquinoline-containing compound 5t with its flexible counterpart from our previous study (see Table in Supporting Information), indicates that the presence of the rigidifying phenyl unit is beneficial for D₃R affinity (K_i at D₃R = 3.4 and 410 nM for 5t and its more flexible congener respectively). Further investigations are required in order to determine the extent to which this SAR trend holds for other 6,7-dimethoxy-1,2,3,4-tetrahydroisoquinoline-containing compounds.

The compounds with the highest D₃R affinity contained either a 6,7-dimethoxy-1,2,3,4-tetrahydroisoquinoline or 6,7-dihydroxy-1,2,3,4-tetrahydroisoquinoline moiety. This suggests that the presence of an o-xylenyl ring linker motif in tandem with a 6-methoxy-1,2,3,4,-tetrahydroisoquinolin-7-ol moiety is less favorable for strong D₃R affinity.

A molecular docking investigation was conducted to attempt to provide a structural interpretation of the effects of linker rigidification and methylation of substituents on the tetrahydroisoquinoline primary pharmacophore moiety that have been observed in this study. Our previous study showed that, with a flexible linker, compounds with the 6,7-dimethoxy-1,2,3,4-tetrahydroisoquinoline primary pharmacophore moiety displayed less favorable affinity to D₃R relative to 6-methoxy-1,2,3,4,-tetrahydroisoquinolin-7-ol-containing analogues. Based on previous molecular modeling results, this behavior was attributed to the ability of the latter group of compounds to form two hydrogen bond interactions between the primary pharmacophore group and Ser192 of the receptor compared to only one hydrogen bond in the case of the 6,7-dimethoxy-1,2,3,4-tetrahydroisoquinoline-containing compounds.^{19, 25} A similar molecular docking study conducted here indicates that rigidification of the linker induces a mode of binding distinct from the one observed earlier and consistent with the observed SAR. As shown in Figure 3, the three most potent compounds (5s, 5t, and 6a) dock as generally expected with the primary pharmacophore group in the orthosteric pocket and forming the key salt bridge between the alkyl protonated nitrogen and Asp110.

Figure 3.

Docked poses of compounds 5s (a), 5t (b) and 6a (c) at D₃R

However, while the unmethylated compound 6a makes a hydrogen-bond interaction with Ser192 as observed in the previous series, compounds 5s and 5t do not. Interestingly, these compounds offset the lack of hydrogen-bond interactions in the orthosteric pocket with an interaction between the linker and regions further up the D₃R binding cavity. In particular, in 5s and 5t there is an interaction between Cys181 of D₃R and the newly introduced phenyl ring of these analogues (via an aromatic H-bond interaction). A similar interaction with the introduced phenyl ring is also seen in 6a. In compound 5s, the phenyl ring also forms a π-π interaction with Phe106. In addition, the positioning of the phenyl ring observed for 5s and 5t enables additional hydrogen-bonds between the carbonyl of the aryl amide moiety and Thr369 (see Figure 3a and 3b). Such interactions do not occur with compound 6a probably because it is shifted deeper into the orthosteric pocket to establish the hydrogen bond with Ser192 as mentioned above.

Overall, the molecular docking models developed here offer useful structural interpretations of the observed affinity of the compounds with the rigid linkers. The drawback of having the o-xylenyl linker is the decrease in selectivity. The orthosteric binding sites of D₂R and D₃R are highly conserved, but the secondary binding pocket of these receptors differ significantly. Interactions with the D₃R secondary binding pocket are important for achieving D₃R versus D₂R selectivity.^{26, 27} For compounds with both a 6,7-dihydroxy-1,2,3,4-tetrahydroisoquinoline and 6,7-dimethoxy-1,2,3,4-tetrahydroisoquinoline motif, the rigid o-xylenyl linker present in this new series does not allow the arylamide secondary pharmacophore of the molecules to make strong contacts with residues in the secondary binding pocket. This diminished binding of the secondary pharmacophore in the secondary binding pocket may be responsible for the lowered D₃R versus D₂R selectivity in this series.

In conclusion, we examined the effect of rigidification of the linker region of tetrahydroisoquinoline-containing D₃R antagonist chemotypes via the inclusion of an o-xylenyl linker motif. Various oxygenated tetrahydroisoquinoline motifs and benzamide sub-structures were utilized for the primary and secondary pharmacophore regions respectively.

We found that in general, rigidification with an o-xylenyl ring is detrimental to D₃R selectivity versus D₂R. However, selectivity versus the σ₂R seems to be dependent on the benzamide motif employed; 4-subsituted benzamide groups in particular afforded compounds with excellent selectivity versus σ₂R (i.e. low or no affinity for σ₂R). In compounds with a 6-methoxy-1,2,3,4,-tetrahydroisoquinolin-7-ol primary pharmacophore group, D₃R affinity was negatively impacted. However, we identified compounds with 6,7-dihydroxy-1,2,3,4-tetrahydroisoquinoline and 6,7-dimethoxy-1,2,3,4-tetrahydroisoquinoline primary pharmacophore groups with high D₃R affinity. Overall, in this series of o-xylenyl ring rigidified compounds, it appears that the choice of the primary pharmacophore group is critical for D₃R affinity while the benzamide secondary pharmacophore region is important for maintaining D₃R affinity as well as selectivity versus σ₂R. Compounds 5s and 5t were among the most potent D₃R ligands identified in this study. Docking studies indicate that the high D₃R affinity of some compounds may be due to the extra interactions of the phenyl ring of the linker with, especially, Cys181.