An Aminoalkylindole Pharmacophore at the CB1 Receptor
Marie Castro, Traci Hunter, Julie Mericle,Judy Norris, Minal Patel, Sharmistha Basu Dutt,
Dow Hurst and Patricia Reggio

Chemistry Department
Kennesaw State University
Kennesaw, GA 30144 USA

Hypothesis
Methods
Results
Conclusions
Acknowledgments
References

Click on images to enlarge them.

I   Hypothesis

The important molecular features of aminoalkylindoles (AAIs) for their interaction with the CB1 receptor are

(1)      a shape that avoids bulk above the plane of the
           indole system near the bottom portion of this system;

(2)      the relative positions of the indole and aroyl moieties; and,

(3)      the carbonyl oxygen of the aroyl moiety.

Futhermore, the primary interaction of AAIs with the cannabinoid CB1 receptor is hypothesized to be aromatic stacking.

II   Methods

To probe this hypothesis, sets of high affinity/high efficacy and low affinity/low efficacy AAIs were used.

A. Conformational Analyses

For each set of compounds, conformational analyses were performed using the semi-empirical AM1 method. For selected analogs, the results of the AM1 conformational analysis were compared to those using an HF-SCF calculation at the STO-3G level.

B. Receptor Essential Volume (REV) Calculations

Where appropriate, the Active Analog Approach was used to calculate regions of steric interference (termed receptor essential volumes, REVs) at the binding site for AAIs at CB1. To this end,

(1)      accessible conformers identified above were superimposed
          at their indole rings using the Chem -X modeling program;

(2)      the Van der Waal’s (VdW) volume of each conformer was calculated;

(3)      the union of the VdW volumes of the high affinity conformers
           and the VdW volumes of low affinity conformers were each calculated
           using the SET MAP/COMBINE facility in Chem-X; finally,

(4)      using this same facility, a logical NOT operation was performed
           to calculate the volume occupied by atoms of the low affinity
           conformers that is not occupied by atoms of the high affinity conformers, 
           yielding the REV map.

C. Receptor Docking

Ligands were docked into a computer model of the CB1 transmembrane helix bundle using interactive graphics.

III   Results

A. Differential Affinities of AAI Stereoisomers Lead to REV 1.

Compounds 1-3 were used to calculate a Receptor Essential Volume map, REV 1

                                                
           R-(+)-                          S-(-)- 
      K_I=106±11nM¹          K_I=17,500±6.19nM¹                              K_I=1.4%@1mM¹ 
        [³H]-WIN                      [³H]-WIN                                               [³H]-WIN

REV1 was used to screen all other conformers to be used in further AAI pharmacophore development.

B. Relative Positions of Aroyl and Indole Moieties

1.      The Influence of the C2 substituent

        The AAIs exist in two major conformational families, the S-cis and S-trans forms.

To study the influence of the C-2 substitutent on conformation, two compounds below were studied:

Table 1

       R                IC₅₀ (nM)¹                                %cis                  %trans
                   [³H]-WIN        MVD

        H           249±17         44.5±9.8                  42.38                  57.62
        Me        152±17         123±13                    95.07                  4.93

We found that when C2 = H, nearly equal amounts of S-cis and S-trans conformers are found. However, the S-trans form predominates. When C2 = Me, the S-cis conformer overwhelmingly predominates. ( SeeTable 1 )

2.              Influence of Aroyl Moiety

Table 2

 R                                           IC₅₀(nM)²                 AM1                     STO-3G
                                      [³H]-Win      MVD      %cis     %trans       %cis       %trans

p-methoxyphenyl       3155±54      319±63           90.1           9.9         84.1         15.9

1-napthyl                    19± 2           15± 2            88.1         11.9         87.6         12.4 

Changing the aroyl moiety from p-methoxyphenyl to 1-napthyl did not change the relative percentages of S-cis and S-trans conformers appreciably. The increased affinity and efficacy of the 1- napthyl derivative cannot be explained based on conformational preferences. (See Receptor Interaction Section below).

C. C-2 SAR: The Influence of Greater Bulk at C-2

The increase in bulk at C-2 from C-2 = Me to C-2 = Et results in significant loss in CB1 affinity. This loss in affinity could not be explained by conformational preferences.

                                          Table 3

   R                IC₅₀ (nM)¹          %cis                  %trans
                     [³H]-WIN

   H                 249±17               42.38                   57.62
   Me              152± 17               95.07                     4.9
   Et                27% at 1µM        95.48                     4.52

These results lead us to hypothesize that there is another sterically occluded region near the C-2 position (i.e., enlargement of the C-2 substitutent from Me to Et apparently prevents the Et analog from binding due to steric problems). Using the compounds in Table 3, an REV for this region,termed REV 2 was calculated.

D. The Importance of the Carbonyl Oxygen

We have designed and synthesized rigid indene analogs of the AAIs. These compounds are locked in either an S-cis or S-trans conformation but lack the carbonyl oxygen. Our results show that both the Z (cis) and E (trans) analogs bind to CB1, with the E (trans) compounds exhibiting higher CB1 affinity. We concluded from this that the carbonyl oxygen is not a primary interaction site for AAIs at CB1.

E. C-4 SAR

Table 4

 IC₅₀(nM)¹
[³H]-WIN                    MVD

42% @ 3000             >10,000

Our conformational analyses revealed that the C-4 analog cannot adopt either an S-cis or an S-trans conformation. This may be one reason why its affinity is so low. (See Receptor Interaction below)

F. CB1 Receptor Interactions

        We have hypothesized that the primary interaction of AAIs at the CB1 receptor is aromatic stacking.

(a)       All binding sites identified in docking studies are consistent with our
            REV 1 and REV 2 results. The change from a p-methoxyphenyl to a
            1-napthyl analog results in additional aromatic stacking interactions.
            This may explain the increased CB1 affinity of 1-napthyl derivatives.

(b)       Both S-cis and S-trans conformers can fit at the CB1 binding site.

(c)       The conformation adopted by the C4 = Me analog in order to dock at the receptor
           requires an energy expense of 15.0 kcal/mol. In addition, the only significant
           interaction of the 4' analog at CB1 is a single Hydrogen bond.

IV.   Conclusions

A.     For an AAI to bind to the CB1 Receptor it must:  

         (1)      Be shaped to avoid REV 1 and REV 2

         (2)      Exist in S-cis/S-trans conformations

B.      The primary interaction of AAIs with CB1 appears to be aromatic stacking.

C.      The carbonyl oxygen does not appear essential for high CB1 affinity.

V.   Acknowledgments

This work was supported by NIDAGrant DA03934.

VII.   References

1. T.E. D’Ambra et al J. Med. Chem. 35,124-135, 1992.

2. M.A. Eissenstat et al. J. Med. Chem.38, 3094-3105, 1995.