Org Prep Daily

December 19, 2007

Frustrating: Meet the Bricks

Filed under: procedures — milkshake @ 6:11 am

bricks.jpg  Credit: Adolf Born

Orally active drugs need to dissolve in water or in lipids. When the compounds are crashing out even from straight ethyl acetate one should urgently look for better-behaved analogues before going too far down the SAR potency/selectivity road. One should not hope to fix the solubility later – there is only so much that the solubilizing side-chains can do. And when a poorly-soluble polycyclic molecule is decorated with amines a mirage of improved solubility and potency in cell-based assay may appear but very little of it usually goes towards a sound SAR.  (Also the PK of such series is most likely going to be awful). 

Compounds generally appear more soluble when prepared in form of a lyophilisate. Scaling-up a poorly-soluble clinical candidate that was never before prepared in a pure crystalline form is bound to produce unpleasant surprises…

Developing poorly soluble compounds is a frequent source of frustration in kinase projects; the kinase ligand-binding sites naturally like polyheterocyclic compounds. Nowadays many medchem programs start with the high-throughput screening of chemical  libraries – the lead compounds obtained from combichem/parallel-synthesis are notorious for the high number of NH bonds. Also, over-optimistic crystallography-driven drug design can lead to terrible molecules: X-ray co-crystal picture offers the illusion of a great insight (“we should try to pick a H-bond here and reach a pocket back there”)  and it encourages decorating the molecule with too large greasy or excessively-polar substituents – the fast way into potent undruggable series. 

When trying to fix a poor compound solubility:

1. Abandon the series and work on something else – the most reliable option. One should consider this seriously when the compounds are so terribly insoluble that they form gels.
2. Get rid of some NH bonds – alkylate the nitrogens, replace them with  C or O, etc. Tertiary amides are less troublesome than primary, secondary amides or sulfonamides. Stay away from N,N’-diaryl ureas. Amine NHs are fine.
3. If you must have the NH bonds in your molecule, put some groups right next to the NH, to shield it  – in ortho or alpha positions. An ortho methyl substituent can make all the difference.
4. Get rid of some hydrogen bond acceptors  – remove sulfonyls, carbonyls etc.
5. Avoid a combination of el deficient with el rich aromatic rings within the same molecule – the charge transfer between electronically unequal partners produces exceptionally strong pi-stacking. (Picric acid forms a stable co-crystal with naphtalene)
6. Mess up the symmetry: Avoid big symmetrical substituents like p-Br-phenyl, try to add some ortho substituents to keep rings out of plane
7. Add solubilizing side-chains containing tertiary (or secondary) amines or try to pro-drug
8. Disconnect rings or replace aromatic rings with non-aromatic ones. Polycyclic aromatic systems with more than 2 rings are Very Bad
9. Reduce the size of the molecule

Purifying insoluble compounds is hard, especially when some structurally-related impurity carries over into the solids. You cannot do a silica column and the reverse-phase prep HPLC purification is no picnic either (you inject your material in DMF solution – then pray it wont crash out in the injection loop, autoinjector needle, connecting tubing or on the top of the very expensive column). If you must make poorly-soluble compounds, it is best to rigorously purify the intermediates right up to the moment when the material becomes very insoluble (such as the deprotection or coupling step) and spend enough time optimising the reaction conditions and workup in the following steps so that the impurities are kept at low levels from then on.


  1. Polycyclic aromatic systems are Very Bad

    And probably carcinogenic! If you want to make PAH’s soluble though, add TIPS or TMS groups, or slap on some t-butyl groups to disturb the pi-stacking.

    Comment by excimer — December 19, 2007 @ 10:27 am

  2. Better yet, TIPS or TES ethynyl groups–they’re soluble AND they still pi-stack, though I don’t know if they’d fare so well in drugs.
    It’s a little ironic that some of what’s on your list is desirable if you’re looking to tune crystal packing.

    Comment by Ψ*Ψ — December 19, 2007 @ 2:03 pm

  3. Poorly Soluble Marketed Drugs Display Solvation Limited Solubility

    Christel A. S. Bergstro¨m,* Carola M. Wassvik, Kajsa Johansson, and Ina Hubatsch

    J. Med. Chem, DOI:10.1021/jm0706416

    Comment by Wavefunction — December 19, 2007 @ 3:18 pm

  4. Nicely put. Management like to count numbers and reach indignificant milestones for bonuses however. But I say damn the naysayers and roll-the-dice. This is why they call it reseach. I have had the same experience ou have seen. It has historically been easier to handle in oncology because the bar was so low, whereas for chronic treatment the drugs had to have better biopharma properties. Now the oncology programs are required to have better starting and endpoints, so “everyone pay head to what milkshake says,” and stop the flat-brick, pi-stacking, amide, urea, biphenyl, sulfonamide crap and get to work. Maybe we should learn how to make those natural product (or at least NP-like) compounds. This is really what we went to school to do.

    Comment by watchin the wheels — December 19, 2007 @ 4:22 pm

  5. hey milkshake, ever done a mcmurry coupling? any advice on how to make them suck less?

    Comment by excimer — December 20, 2007 @ 2:34 am

  6. Oh, I have never done it myself but I remember there was a very powerful version of McMurry coupling, I think from Furstner (Umlaut u) that was catalytic in titanium and supposedly worked on things that classic McMurry couldn’t touch. I think he published it in 90s. Also I can ask my former advisor, he spent almost 3 years as a postdoc with McMurry so chances are he did it himself.

    Why does it suck: workup? Catalytic Ti should help with that

    Comment by milkshake — December 20, 2007 @ 4:50 am

  7. yeah, it’s the workup. You get a ton of blue gunk which is undoubtedly titanium. I’ll take a look at that furstner pub. Thanks

    Comment by excimer — December 20, 2007 @ 10:14 am

  8. Furstner- JACS, 1995, (117) 4468-4475; Tet, 1995 (51), 8875-8888.
    Never run any of this myself, but have always have excellent results with all of Furstner’s experimentals.

    Comment by Jose — December 20, 2007 @ 12:10 pm

  9. Hey.. don’t mean to highjack this thread…

    Anyway.. I am doing an aromatic iodination and am using mercuric oxide… the work up calls for na2s2o5 which I am assuming is a reducing agent to help remove the mercuric shit…

    Anyway I don’t have this stuff… any alternatives you guys can think of?

    Comment by tom — December 21, 2007 @ 5:26 pm

  10. I would try N-iodosuccinimide in neat triflic acid or in BF3.H2O as a solvent

    Comment by milkshake — December 21, 2007 @ 6:36 pm

  11. Yeah that looks like a much nicer prep… but I am going with a common lit prep for the intermediate I am making.. makes more sense for me to use up less time trying to work with a procedure i have never used before.

    Mercuric oxide is nasty… but it seems like a pretty easy reaction.. just dump and stir. I am doing it on like 10+ g so I won’t ever have to do it again as long as I get like 2-3 g of material.

    Comment by Tom — December 22, 2007 @ 2:29 am

  12. ah how I love med chem. I only got a 20% yield on the final product…. but it doesn’t really matter as I have plenty of material…

    Quite nice. Optimizing yields is such a pain in the ass.

    Comment by tom — December 24, 2007 @ 3:44 am

  13. you have to leave something for the process chemists

    Comment by milkshake — December 24, 2007 @ 6:49 am

  14. .. I am doing an aromatic iodination and am using mercuric oxide………

    What about old fashioned Iodine monochloride? I have used it a lot (reflux in DCE or CHCl3). Works esp well in e- rich systems, of course. Even gives polyiodination products, which was quite useful when you’re making imaging agents. However, controlling stoichiometry should give selective iodination. It’s cheap, somewhat nasty to handle, but compared with HgO should be less of a pain to work with……..

    Comment by moody blue — December 27, 2007 @ 1:33 pm

RSS feed for comments on this post. TrackBack URI

Leave a Reply

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

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

Google photo

You are commenting using your Google 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 )

Connecting to %s

Blog at

%d bloggers like this: