Skip to Content

Why Not Bromine?

So here’s a question for the medicinal chemists: how come we don’t like bromoaromatics so much? I know I don’t, but I have trouble putting my finger on just why. I know that there’s a ligand efficiency argument to be made against them – all that weight, for one atom – but there are times when a bromine seems to be just the thing. There certainly are such structures in marketed drugs. Some of the bad feelings around them might linger from the sense that it’s sort of unnatural element, as opposed to chlorine, which in the form of chloride is everywhere in living systems.
But bromide? Well, for what it’s worth, there’s a report that bromine may in fact be an essential element after all. That’s not enough to win any arguments about putting it into your molecules – selenium’s essential, too, and you don’t see people cranking out the organoselenides. But here’s a thought experiment: suppose you have two drug candidate structures, one with a chlorine on an aryl ring and the other with a bromine on the same position. If they have basically identical PK, selectivity, preliminary tox, and so on, which one do you choose to go on with? And why?
If you chose the chloro derivative (and I think that most medicinal chemists instinctively would, for just the same hard-to-articulate reasons we’re talking about), then what split in favor of the bromo compound would be enough to make you favor it? How much more activity, PK coverage, etc. do you need to make you willing to take a chance on it instead?

38 comments on “Why Not Bromine?”

  1. Basho says:

    “fruit flies died when bromine was removed from their diet but survived when bromine was restored”. BROMINE BRINGS THE DEAD BACK TO LIFE!

  2. schinderhannes says:

    It’s a N of 1… But I had a case where Chlorine did miracles on an arene compared to F. Br was even better – but Ames positive. CL became dev candidate…

  3. Pete says:

    One reason may be that bromine tends to more of a solubility killer than chlorine (matched molecular pair analysis). If the gain in potency resulting from replacement of chlorine with bromine is smaller than loss of solubility (and deterioration of other ADMET characteristics then the case for making the change is weakened. One situation in which we might anticipate a relatively large increase in potency resulting from replacement of chlorine with bromine is when halogen bonds can form. Halogen bonding tends to be stronger when the halogen is bonded to a relatively electron-withdrawing aromatic ring (e.g. pyrimidine) and the halogen is heavy.

  4. Anonymous says:

    I remember a project biologist being concerned about tox consequenses, i.e. brominism. That should be more associated with bromide salts than bromoarenes though. Does anyone know if bromoarenes can get metabolized in a way that releases bromide ion?

  5. schinderhannes says:

    Side note: How bout BVO? But then again – Mountain Dew ain’t no drug….

  6. Sideline Chemist says:

    It really boils down to your risk tolerance. Heavy halogens like bromine & iodine in drugs have been associated with phototoxicity, photoreactivity, reactive metabolite formation, etc that can led to safety issues. If a Cl atom will give you largely the same properties with a lower safety risk, 99.5% of us will go with the Cl.
    However, there are plenty of examples where Br or I make it into the final drug candidate or even marketed drug. The mother of all examples are the thyroid receptor antagonists and the allosteric MEK inhibitors. It can be done, just a question of you (and your company’s) risk tolerance.

  7. An Old Chemist says:

    Although rare, a Br on an aromatic ring may be displaced by a hot nucleophile like a thiol of a cysteine (iniparib, Sanofi’s PARP inhibitor is an example). Besides, during extensive metabolism, a reactive intermediate may indeed form that could make displacing the Br really easy, e.g.,a quinone methide.

  8. Toad says:

    A quick search of DrugBank suggests there are 12 approved drugs containing a bromine (alkyl and aryl bromides), only one of which was withdrawn from the market. They are for a diverse set of targets and disease states. Hmmm.

  9. Boghog says:

    @ Sideline Chemist:
    When you talk about thyroid hormone antagonists, I assume that you are referring to amiodarone (which contains iodine instead of bromine) which is an effective but a nasty drug associated with a number of side effects which appear to more associated with the extremely long half-life and low water solubility (precipitation in the lungs resulting in pulmonary toxicity). This may be more a reflection of the high log P of bromine/iodine rather than anything intrinsic to these very hydrophobic halogens.

  10. Barry says:

    re: #7 in cases where an aromatic bromide can be displaced, the aromatic chloride or fluoride is displaced faster.
    As to “ligand efficiency”, I think most in this august forum will realize that molecular weight is a proxy for rotatable bonds (which correlate better with ADME) and Ar-Br should be no worse than e.g. Ar-F on that count.
    As to phototox, I know that Ar-I is often photo-labile, but I haven’t seen that with Ar-Br

  11. sulfilimine says:

    Anyone knows how strong those sulfilimine bonds are?

  12. Martin says:

    Possibly irrational but maybe because as practising organic chemists we see what a 5 year old bottle of chlorobenzene looks like vs a 5 yo bottle of bromobenzene. I.e. We’re distrustful of a perceived instability that isn’t really there over the lifetime of the “drug” post-manufacture

  13. anon says:

    I’ve been debating just this very issue. I have several bromoaromatics that are significantly better than the chloroaromatics, with respect to activity in a project (both in vitro and in vivo).
    The biologists have been all excited, and I’ve been trying to find something that works better than the bromine for quite awhile now. CN, CF3, Me, OCF3, F, Cl, SO2Me, etc. all don’t cut it.
    Maybe I should get more excited too?
    Because I don’t really have a good argument, other than “I don’t like the looks of it” and “the MW is too high”
    But cells don’t really have a mass spec to tell them what to take in.

  14. WDR says:

    Interesting that the linked article image has tin as an element essential to life–I’d not seen that claim before.

  15. PorkPieHat says:

    Had a situation very much like #2 @schinderhannes. A brominated compound I was associated with would have been on the market a long time ago, had it not been for a genetox finding.
    Good question to re-examine.

  16. milkshake says:

    For all lovers of heavy atoms: Bromine and Iodine in CF2Br and CF2I groups are perfectly non-electrophilic, and they act as electron pair acceptors – reversibly forming strong complexes with DABCO

  17. Anonymous says:

    The aromatic C-Cl bond is stronger than the C-Br bond by over 10 kcal/mol, so I think chloroaromatics should be more metabolically stable.

  18. Yar says:

    Re #13, if you have a productive halogen bond involving the bromine, the only thing to replace it and bring even better affinity would be an iodioaromatic. (See e.g. J Med Chem review on halogen bonding from sometime in 2013). Somehow I doubt you would go for that. 😉
    Back on topic, I don’t think there is a reason other than lipophilicity for choosing chlorine over bromine. Maybe over time medicinal chemists will become less afraid of it, especially when it leads to improved affinities.

  19. Yar says:

    Re #13, if you have a productive halogen bond involving the bromine, the only thing to replace it and bring even better affinity would be an iodioaromatic. (See e.g. J Med Chem review on halogen bonding from sometime in 2013). Somehow I doubt you would go for that. 😉
    Back on topic, I don’t think there is a reason other than lipophilicity for choosing chlorine over bromine. Maybe over time medicinal chemists will become less afraid of it, especially when it leads to improved affinities.

  20. Yar says:

    Re #13, if you have a productive halogen bond involving the bromine, the only thing to replace it and bring even better affinity would be an iodioaromatic. (See e.g. J Med Chem review on halogen bonding from sometime in 2013). Somehow I doubt you would go for that. 😉
    Back on topic, I don’t think there is a reason other than lipophilicity for choosing chlorine over bromine. Maybe over time medicinal chemists will become less afraid of it, especially when it leads to improved affinities.

  21. Barry says:

    The other (partial) bio-isostere for Ar-Br is ArCCH. If the bromine is accepting electron density from a residue in the target protein, an H-bond from the alkyne can–if the geometry is just right–give you some of that. See the ParkeDavis work on MEK inhibitors

  22. DN says:

    The immune system makes hypobromous acid as a defensive weapon.

  23. DN says:

    The immune system makes hypobromous acid as a defensive weapon.

  24. schinderhannes says:

    Side note: How bout BVO? But then again – Mountain Dew ain’t no drug….

  25. Noel O'Boyle says:

    By coincidence, I’ve just posted on the frequency of different phenyl substituents in the ChEMBL database. You can compare the frequencies, for example, of 4-Br-phenyl and 4-Cl-phenyl. The 4-Br is the 6th most popular 4-derivative, while the 4-Cl is the 2nd most popular.

  26. Anonymous says:

    I would be worried that Br is more likely to be attacked and fall off in vivo, but otherwise there is nothing like an experiment to eliminate such biases. It’s the same irrational issue as racial prejudices in hiring: It’s better just to test anyway.

  27. lcollia says:

    My collegue just gave me an other possible explanation: we are using a lot of coupling methodologies in order to synthesize compounds, and in these reactions Cl-Ar are in general not attacked while it is the case for Br-Ar.
    So maybe also a biais from synthesis.

  28. Anonymous says:

    The simplest reason to use Cl over Br is cost. The bromo compounds usually cost more, weigh more (so more is needed per mmol), and the final product will weigh more, so the pill might be bigger and the amount of API will be higher, (costing more to make, transport, etc, due to higher weight and volume) assuming that the compounds are equipotent.
    But there are cases where the bromine is needed, so the real key is what is the benefit/cost ratio of the change. Often there is a big enough difference between two similar looking molecules that it is clear which is the winner. I just saw a case where a nBu is much better than a nPr, so the extra MW is well worth the extra MW, in the activity, selectivity, PK, etc.

  29. Anonymous says:

    @28: Get serious! API cost is nothing compared to the value/price of an effective drug.

  30. metaphysician says:

    I am so there are awards waiting for the first chemist brave enough to experiment with astatine in drug discovery. 😉

  31. DrSnowboard says:

    A Pfizer mantra that all bromoaromatics are photoreactive? I had a colleague refuse to make one on those grounds, despite it being an interesting SAR comparison. The reasoning that all discovery compounds MUST be potential drugs, constrained in their neat boundaries of less risk and highly ordered retrospective rationality, takes the fun and the creativity out of it for me. Because obviously, if we know what to make, we just make those 3 compound? OR get someone else to make it and claim the intellectual credit.

  32. weirdo says:

    #17: The aromatic C-Cl bond is stronger than the C-Br bond by over 10 kcal/mol, so I think chloroaromatics should be more metabolically stable.
    In what metabolic process is the cleavage of a C-X bond the rate-determining step?
    If it isn’t the RDS, what relevance is the C-X bond strength?
    (No, these are not really rhetorical questions.)

  33. entropyGain says:

    Weight, Derek?
    Is a CF3 all that much worse than a Me because its heavier (ie denser). Silly prejudice.
    d-orbitals are special, and apparently a bit scary to med chemists 🙂

  34. Secondaire says:

    Re #2: I had the reverse, where the fluorine was vastly superior to the chlorine but our cell permeability was less than for the chlorinated compound, but this didn’t translate to poor in vivo bioavailability.
    I second the “bias from synthesis” notion. Something about the massive barrage of Ar-Br functionalization in the past ten years or so tends to wire into our brains that bromides are things that can just pop off under the mildest of conditions, and it’s completely irrational because our enzymes don’t use (PPh3)4Pd as a cofactor…

  35. bromoseltzer says:

    Google “bromoseltzer” and you will see the painkiller that was used in early 20th century.

  36. Morten G says:

    Iodine is terrible in a drug – unless you want a clearance time of several weeks. I think the half-lives of the iodated thyroid hormones is about 7 days. And there aren’t any of the iodated drugs that makes you go “Yes, but this one is fine.”
    The lovely thing about bromine from the viewpoint of a crystallographer is that you have a anomalous signal that is really easy and fast to pick out. To be perfectly honest you don’t always get great ligand density even when the protein is fine…
    And from protein stability viewpoint – NaI is a problem but most proteins are indifferent to NaBr. I don’t know what the Br- toxicity mechanism is but I wouldn’t be worried from a drug metabolism viewpoint. According to trusty Wikipedia “The average concentration of bromide in human blood in Queensland, Australia is 5.3±1.4 mg/L” and while Queenslanders eat a fair amount of fish there are several nationalities that eat much more.

  37. If you are going for finest contents like I do, only pay a visit this
    site every day for the reason that it provides quality contents, thanks

  38. zmatt says:

    It is perhaps interesting to note that in the field of recreational pharmacology there’s “2C-x” series of psychedelic compounds (4-x-2,5-dimethoxyphenethylamine) for various choices of x, of which the bromo appears to be the most common and popular choice. The iodo is also used and both longer acting and more potent (even by weight). In contrast, the chloro is pretty much unheard of.

    It made me wonder if the fluoro derivative also existed, and it turns out 2C-F does in fact have an entry in PiHKAL, but there’s good reason I’d never heard of it. Not only does the synthesis read like typical “fun” with fluorine, even to a non-chemist like me (with the highlight being “This salt was pyrolyzed with the cautious application of a flame, with the needed attention paid to both an explosion risk, and the evolution of the very corrosive boron trifluoride.”), but the fruit of their labour turned out to be completely inactive even in ridiculous dosages.

    Oh well, at least they had fun with the synthesis 😉

Comments are closed.