Nonracemic Synthesis of GK−GKRP Disruptor AMG-3969

Hi Folks, it was a busier week than normal and I finally got a chance to work on a post.  There is no witty title here, either.  I couldn’t think of one.  I wanted to find something with some structures to draw because I haven’t posted anything of that nature in a while.  I originally thought I found a process development paper in JOC, but I was in  error.  I decided to post about it anyways, because it hits on some points that process chemists will identify with.

The post can be found in J. Org. Chem. 2014, 79, 3684−3687, doi: 10.1021/jo500336e.  Lead author is Matthew P. Bourbeau and colleagues at Amgen.  The paper is about the rework of an API synthesis.  The original synthesis was used to produce material but when your final product is racemic and being separated using SFC or SMB chromatography at the final step, it will be the bottleneck for anyone’s manufacturing route.  I worked on a project team where we were separating diastereomers in the last step to attain the right product, so I understand.

This is the final synthesis.  You will have to look up the paper to see what the original synthesis was .  This article compares the original to the final synthesis.  Activation of carboxylic acid 2 with HATU to form an amide with 1, followed by deprotection of the Boc group and treatment with ammonia to form the piperidione 3, reduction of the dione to the benzylated piperidine 4.  Palladium mediated cross-coupling with RuPhos gives 5, followed by deprotection of the benzyl group to give 6.  7 undergoes a Sandmeyer reaction to give sulfonyl chloride 8.  8 is coupled to 6 followed by Boc deprotection with TFA.

 

amgen1

 

amgen2
The issues were:  (I just realized I had to create this table in HTML, yuck !!)

Original Synthesis Final Synthesis
7 steps, 14 % yield 5 steps, 26 % yield
Racemic product > 99 % e.e.
Safety concern Reduced safety concern

There was some concern in the original synthesis where 9 (R=Cl) underwent aminolysis at 120 ºC to give the final product (R=NH2).  This step involved a high pressure surge of 110 psi and the DSC thermogram  indicated a large exotherm with an offset temperature at 140 ºC.  This was avoided by forming 7 through 8 and then finally coupling 8 to 6.  To address the issue of the racemic product in the original synthesis, a compound derived from the chiral pool was used, mainly propargyl glycine 2.

End result was a short synthesis, a chiral product and reduced safety concern.  I am curious, though.  Would you need to get a safety profile for each intermediate that had the alkyne in it if it was part of the drug candidate’s structure ?   I dealt with an alkyne in one of my syntheses.  It was used to connect two parts of the molecule.  The safety profile of each side with the alkyne attached helped my team decide which way the molecule should be assembled.  It is the presumably the alkyne that is the high energy functionality that gives the large exotherm.  a had an exotherm at an offset of 114 ºC versus b that had an exotherm at 224 ºC.  Needless to say, we connected our thiophene to b, because at a safety window of 100 ºC, our reaction temperature was still well below 124 ºC.  We reduced the triple bond almost immediately, so it was not in the final structure.

pelitrexol

That’s it for now.  Stay tuned.

Advertisements

3 thoughts on “Nonracemic Synthesis of GK−GKRP Disruptor AMG-3969

  1. Regarding safety profiling – it should really be done anyway, as part of a development package, regardless of whether you think you can see an alerting structure. Our company policy is that all reactions to be run above 100g should be screened for potential thermal runaway or excess pressure generation. We use the ARSST device for this, and have a “proper” calorimeter with which to follow up anything “interesting”. Of course if you do see an alerting structure, DSC is available long before you start to think of scale-up

    Like

  2. It is commendable to use safety data to change your synthesis if you have noticed a highly exothermic event. At 60 gram lots, it doesn’t seem like an issue. What I would have like to know was is the exothermic event due to aminolysis that was performed or because there was an alkyne present. I don’t have any experience with an alkyne that was present during the whole synthesis.

    Like

Any comments ?

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s