Abelian 2024: An analysis of the structure activity relations of 24 2'-Oxo substituted Arycyclohexylamines (Ketamine, MXE, O-PCE, etc. )

 The full text can be found here:

https://ui.adsabs.harvard.edu/abs/2024PhDT........40A/abstract

 

Tl;dr: We now know how different nitrogen substitutions affect the activity of ketamine and MXE analogues. PCE analogues have the highest affinity, and pairing a 3-Methoxy with the 2’-Oxo substitution also offers us high affinity, so it probably makes sense to continue developing compounds along that pattern (though there really isn’t much left to explore at this point). One really interesting thing that stood out is that an N-cyclobutyl substitution has decently high affinity! This is definitely something that should be explored more.

 

Another fantastic and comprehensive paper has come from a friend and former colleague, the wonderful Dr. Anush Abelian! In her PhD dissertation she sets out to synthesize and analyze a series known as the β-Ketoarylcyclohexylamines- This means arylcyclohexylamines that have the 2’-Oxo group, also labeled as a β-Ketone. This substitution lends unique subjective properties to the dissociative experience- compounds with the β-Ketone often feel heavier, more hole-y and sedating compared to the manic and stimulating phenyl-substituted base arylcyclohexylamines. They tend to be less potent, which makes doses more forgiving for casual use. They tend to have shorter durations. Perhaps the most famous dissociative of all contains the β-Ketone structure: Ketamine, which could also be labeled as 2-Cl-2’-Oxo-PCM.

The observation of the unique effects of ketamine inspired further development along the lines of compounds containing both the 2-Oxo substitution paired with various aryl substitutions. In research for example, Sleigh et al. 2015 and Jose et al. 2013 delve into various long reverse ester chains on the secondary amine for very short acting ketamine analogues. Tiletamine, which contains a thiophene instead of a phenyl ring (+ an ethylamine, could be written as 2’-Oxo-ThCE, or ThXE using drug nomenclature), was developed as a veterinary anesthetic.

Primarily though, development of novel compounds and data on the in-vivo activity of these drugs has been driven by the grey area research chemical market and community. Fastanbulbous on Bluelight developed MXE based on hypotheses about SAR and started a revolution with arylcyclohexylamine development. Many compounds followed suit after MXE’s extinction, using the same _X_ naming convention to capitalize on the Hype- like MXM, MXPr, MXiPr, 2-FXE. (Even unrelated diarylethylamines got caught up in the hype: MXP). Others were developed as base β-Ketones with no phenyl substitutions, like DCK and O-PCE, both of which proved to be worthy and interesting compounds. Various analogues directly inspired by ketamine appeared on the market too, like 2-F-DCK, 2-Br-DCK. Needless to say, this is a really exciting area of drug development and there is so much potential for so many unique and interesting compounds!

Dr. Abelian has for the first time done a comparative analysis of an entire series of these compounds, detailing their synthesis, comparing their binding affinities and projected pKa’s and laying out a map for the structure-activity relations of the entire family of compounds along with potential future compounds!

Her findings are multitudinous and the inclusion of more experimental compounds helps expand the potential for future developments.

The following paragraph is all chem synth skip it if that’s not of interest.

The synthesis of each compound is laid out clearly! For the chem folks, it’s a classic modified Calvin Stevens ketamine synthesis with a ring rearrangement starting with a substituted phenylcyclopentyl ketone (Stevens et al 1966). This can be formed from reacting a cyclopentane Grignard with the corresponding benzonitrile. The starting unsubstituted and 2-Cl substituted phenylcyclopentyl ketone were inexpensive and readily available for laboratory purposes (I’ll never forget that odd sickening strong celery smell that seemed impossible to wash off of anything…) so the Grignard step was usually unnecessary. The ketone is then brominated at the α position with aluminum chloride and bromine, which can then be converted into the α-hydroxyketone via reacting it with a strong base. With that, a light-sensitive imine is formed by reacting with the corresponding alkylamine, which then undergoes a rearrangement under microwave radiation, yielding the final β-ketoarylcyclohexylamine. All reactions were done under argon and all products were purified by column, then formed into the HCl salt by the addition of a molar equivalent of HCl. Further purification by recrystallization yielded excellent product! It should be noted that with the 2-Chloro substituted compounds, microwave radiation yielded a highly fluorescent byproduct. In those cases, the bromine intermediate was reacted with 6 equivalences of base to yield the α-2-chlorophenyl-hydroxycyclohexylketone. This was aminated by reacting it with Methanesulfonyl chloride, followed by the corresponding alkylamine. Each compound is extensively analyzed and characterized, which can be used as reference for drug testing programs.

For the pharmacology and structure activity relations nerds, we look at table 3.1, which lays out the binding affinities for the entire series of compounds:

 

 

A lot of really interesting observations we can make from this beautiful data! A lot of patterns we can draw! Let’s look at the different  categories of compounds. Here are the affinities displayed visually with the molecules in order from highest affinity to lowest affinity. It’s important to remember that a lower number signifies higher affinity.

 

But……….
First we gotta talk about nomenclature! We need to lay this out and communicate this clearly!  How we name these things- it's confusing! There’s the scientific way of naming things, and there’s the market way of naming things. As I discuss these compounds I will follow the naming conventions listed below.

 

There is the substituted xCx naming convention for arylcylohexylamines. You list the substitutions, and then the base structure. The first x is the aromatic ring, almost always a phenyl ring, P,  the C is the cyclohexyl ring, the last x denotes the amine, always changing, which is the object of interest here. It yields us things like 3-MeO-PCiPr etc.

 By this scheme ketamine would be listed as 2-Cl-2’-Oxo-PCM.

Then MXE came out, everything wanted to be like MXE! Our naming conventions got fucked. Now we use the MXE denotation of MXPr, MXiPr, FXE, DMXE etc. Gotta ride the hype! The way this nomenclature is structured is: 2’-Oxo compounds with a substitution of any kind on the phenyl ring get the xXx label, The first character being the corresponding phenyl substitution, (F for Fluorine in FXE for example, M for 3-Methoxy, DM for methyl (“desoxy”), H for hydroxy (as in HXE); a 3- position substitution is assumed and if that is not the case, the number substitution is affixed to the front, as in 2-FXE), the last character corresponding to the amine. By previous convention, MXE is 3-MeO-2’-Oxo-PCE. The capital X in the middle represents the 2’-Oxo group in this nomenclatural structure.

Lastly, there are the compounds with no aromatic substitutions- the most popular ones are DCK and O-PCE, which seem to follow no rhyme or reason. I hate this shit! We should have a system and a standard. O-PCx. I think this is fine nomenclature. DCK should technically be called O-PCM. But I will use this convention for these compounds, just for ease of writing.

So we end up with these naming schemes. Things have been called names like n-ethylnorketamine which make things more confusing! We just name things in relation to other things. That aforementioned compound is 2-Cl-2’-Oxo-PCE! You could call it ClXE (if a chlorine is involved it seems everyone just calls it ketamine!). I think the easiest name for that compound is probably just ethylketamine though. Having 3 different naming schemes depending on the substitutions the drug has is not a good way to do things, it breeds confusion and misunderstanding and obscures the patterns of relations between these compounds. But that is where we are and we’re just kinda stuck with it.

 

Anyways,

 

 

Let’s look at the patterns and structure activity relations!

 

Start with the unsubstituted 2’-Oxo-compounds: These are the ones that can fall under the nomenclature of O-PCx. They have beta ketone on the cyclohexane as ketamine and every compound in this study have. These ones have no substitution on the aromatic ring. These are familiar compounds: DCK, O-PCE. The ones listed in this study are:

O-PCA (2’-Oxo-PCA)

O-PCM (2’-Oxo-PCM, Deschloroketamine, DCK)

O-PCE (2’-Oxo-PCE)

O-PCPr (2’-Oxo-PCPr)

O-PCiPr (2’-Oxo-PCiPr)

O-PCEtOH (2’-Oxo-PCEtOH)

O-PCAl (2’-Oxo-PCAl)

O-PCcP (2’-Oxo-PCcP)

This affords us a really interesting view of the effect of different amines in 2’-Oxo substituted ACH receptor affinity. They are in order of affinity- O-PCE> O-PCiPr>O-PCM> O-PCPr>O-PCAl>O-PCA>O-PCcP>O-PCEtOH- this seems to map pretty predictably and reliably on in-vivo potency and also mostly follow the pattern seen in non 2’-Oxo subbed ACHs: O-PCE is quite potent, O-PCM (DCK) a little less so, and O-PCPr less so than that. One could predict that the potency of O-PCiPr would fall between that of DCK and O-PCPr. The more exotic amine substitutions didn’t show a higher affinity than ketamine, so would expectedly be less potent than it, though this says nothing about the actual qualitative experience. O-PCcP would be pretty impotent though and would require the ingestion of a large amount of powder. O-PCEtOH has such a low affinity one would presume it bordesr on potentially being inactive. However, two active compounds with this substitution have been observed: First with a ketamine analogue (That would be 2-Cl-2’-Oxo-PCEtOH) where this compound was actually the most potent of a series of ketamine esters and ethers (Jose et al. 2013). An unsubstituted version of this compound was also allegedly sold on the market and was supposedly active in-vivo though there is no detailed information on the effects (Morris Wallach 2014). It is entirely possible for an arylcyclohexylamine to have very low affinity but still be appreciably active in vivo- this is seen in 3F-PCP, where affinity would indicate it is wholly inactive, when this clearly isn’t the case (Wallach 2014). It appears that there is a sweet spot with the ethylamine for highest affinity, and that affinity is conserved by keeping to roughly that size and shape (hence why isopropyl shows higher affinity than propyl).

 

From there we can compare the next category of compounds- non halogenated phenyl substitutions-The MX style compounds

MXM (3- MeO-2’-Oxo-PCMe)

MXE (3-MeO-2’Oxo-PCE)

MXPr (3-MeO-2’-Oxo-PCPr)

MXiPr (3-MeO-2’-Oxo-PCiPr)

DMXE (3-Me-2’-Oxo-PCE)

DMXM (3-Me-2’-Oxo-PCM)

In order of affinity: MXiPr>MXPr>MXE>MXM>DMXE>DMXM. This is an interesting case in affinity not correlating to in vivo potency- any seasoned dissonaut can tell you that MXPr and MXiPr were not at all more potent than MXE. I cannot begin to imagine what the reason for this discrepancy is, but it’s an important lesson in not really knowing what a drug does until you try it. I am curious if this pattern of affinity just increasing with larger carbon chains on the amine holds true for this particular substitution pattern- does it then hold true for 3-HO substitutions? What about a 3,4-MD substitution? It all remains to be seen. Most of these compounds have been made and tested in people so this data is more just curious for how it doesn’t align with in-vivo activity than predictive or anything.

 

Then we look at the compounds directly adjacent to ketamine. I don’t really know what to call these.

We have:

Norketamine (2-Cl-2’-Oxo-PCA)

Ketamine (2-Cl-2’-Oxo-PCM)

Ethylketamine (2-Cl-2’-Oxo-PCE, N-Ethylnorketamine)

Propylketamine (2-Cl-2’-Oxo-PCPr, N-Propylnorketamine)

Cyclopropylketamine (2-Cl-2’-Oxo-PCcP, N-Cyclopropylnorketamine)

Allylketamine (2-Cl-2’-Oxo-PCAl, N-Allylnorketamine)

Cyclobutylketmaine (2-Cl-2’-Oxo-PCcBu, N-Cyclobutylnorketamine)

For these compounds, we have in order of affinity: Ethylketamine> Cyclobutylketamine >Ketamine>Propylketamine> >Norketamine>Allylketamine>Cyclopropylketamine. These compounds seem to follow the pattern laid out by the normal 2’-Oxo substituted compounds. Interestingly it seems at the very least the 2-halogen position doesn’t run into the odd pattern seen with the 3-alkane or alkoxy substituted compounds. I don’t know if this is a function of the 2 position or a function of the substitution being a halogen. Other structures would have to be tested to elucidate any pattern, like a series of 3F or 3Cl substituted compounds, or a series of 2-MeO substituted compounds. Interestingly, cyclobutylketamine has a higher affinity than normal ketamine! I wonder if this property extends to other substitutions with the cyclobutylamine- or why this seems to go against the trend of bulkier chains losing affinity- would be super curious to compare to straight chain butyl or other butylamine isomers, particularly sec-butyl. This shows a lot of promise in developing new compounds with this amine. In correspondence with Dr. Abelian, she suggested this discrepancy is because the binding pose of ketamine at the PCP site within the NMDA channel is unstable, shifting between 2 different conformations that bind to 2 different parts of the binding site (Zhang et al. 2021). It is not yet known what may modulate one conformation over the other or if it’s seemingly totally random, but either way this could potentially result in SAR discrepancies

 

Lastly, there are 3 tertiary amines with 2 alkane chains:

Dimethylketamine (2-Cl-2’-Oxo-PCDMe)

Methylethylketamine (2-Cl-2’-Oxo-PCMeE)

Methylpropylketamine (2-Cl-2’-Oxo-PCMePr)

These in rank order of affinity are Methylethyl>Methylpropyl>Dimethyl. These have not been explored really, it is theorized that aryclcyclohexylamines with tertiary dialkene substitutions are metabolized into 2 sets of the corresponding secondary amines (Cho et al. 1993). It is not known if this is also seen when there are 2’-Oxo substitutions- it is also likely that these compounds are active in their own right when they are not metabolized, though it is possible that the fate of them in vivo is to always be metabolized into those 2 secondary amines.

Is there anything else to look at?

One last pattern to analyze is the relations across the different ring substitutions? We can in 2 cases compare the 2’-Oxo cyclohexane substituted ACH’s across 4 aromatic substitution patterns: Unsubstituted, 3-Methyl, 3-Methoxy, and 2-Chloro. We have this dataset for PCM and PCE. For both compounds, affinity is greatest along the following pattern: Unsubstituted>3-Methoxy>2-Chloro>3-Methyl. I would be curious to see where 3-halogenated substitutions would fall on this pattern- interestingly, 3-Methyl substitutions are higher affinity for non 2’-Oxo substituted ACH’s- they seem to drop off here though. Just goes to show that MXE was a really brilliant design!

 

Conclusion, tl;dr, going forward:
So now we have a really nice map of the structure-activity relations of 2’-Oxo substituted compounds. The 2’-Oxo substitution grants arylcyclohexylamines a “heavier” feel and is responsible for the compounds we tend to assign the feeling of “holing”. In general, compounds with 2’-Oxo substitutions seem to follow many of the same patterns seen in the non 2’-Oxo ACHs, in terms of how the amine and the length of the carbon chain on the amine affects activity.

If we want to rank the highest affinity (and likely most potent) compounds in this study, we have MXiPr>MXPr>O-PCE>Ethylmethylketamine>MXE>O-PCiPr>DCK>O-PCPr>N-Ethylnorketamine>Cyclobutylketamine>MXM>DMXE. All of these compounds are higher affinity than ketamine, and Allylketamine is close in affinity to ketamine, being only slightly less. Of these, the ones that haven’t been made yet are Ethylmethylketamine, O-PCiPr, Cyclobutylketamine, and Allylketamine. These all represent potential future developments for arylcyclohexylamines. Notably, the cyclobutyl and allyl N-substitutions conserve a good amount of activity, and are probably worth exploring as regular non-2’-Oxo phenyl substituted arylcyclohexylamines (eg, 3-MeO-PCcBu, 3,4-MD-PCcBu, 3-MeO-PCAl etc.).

 The most interesting new finding is perhaps the high affinity of the cyclobutylamine, which has not really been investigated. It would be really interesting to see what this is actually like qualitatively and how that structure affects the activity of other substituted ACH’s!

 

References:

Abelian A (2024). Design, Syntheses, and Pharmacological Evaluations of β-ketoarylcyclohexylamines. Doctoral Dissertation, Saint Joseph’s University. Astrophysics Data System.

 

Cho AK, Hiramatsu M, Schmitz DA, Vargas HM, Landaw EM (1993) A behavioral and pharmacokinetic study of the actions of phenylcyclohexyldiethylamine and its active metabolite phenylcyclohexylethylamine. The Journal of Pharmacology and Experimental Therapeutics 264(3):1401-5

 

Jose J, Gamage SA, Harvey MG, Voss LJ, Sleigh JW, Denny WA (2013) Structure-activity relationships for ketamine esters as short-acting anaesthetics. Bioorg Med Chem. 21(17):5098-106.

 

Morris H, Wallach J. From PCP to MXE: a comprehensive review of the non-medical use of dissociative drugs. (2014) Drug Test Anal. 6(7-8):614-632.

 

Stevens C, Thuillier A, Taylor KG, Daniher FA, Dickerson JP, Hanson HT, Nielsen NA, Tikotkar NA, Weier RM (1966). Amino Ketone Rearrangements. VII.1 Synthesis of 2-Methylamino-2-Substituted Phenylcyclohexanones. The Journal of Organic Chemistry 31(8), 2601-7

 

Wallach J (2014). Structure Activity Relationship (SAR) Studies of Arylcycloalkylamines as N-Methyl-D-Aspartate Receptor Antagonists. Doctoral Dissertation, University of the Sciences in Philadelphia. Proquest Dissertations and Theses database.

 

Zhang Y, Ye F, Zhang T, Lv S, Zhou L, Du D, Lin H, Guo F, Luo C, Zhu S (2021). Structural basis of ketamine action on human NMDA receptors. Nature 596:301-5