Derek Lowe, an Arkansan by birth, got his BA from Hendrix College and his PhD in organic chemistry from Duke before spending time in Germany on a Humboldt Fellowship on his post-doc. He's worked for several major pharmaceutical companies since 1989 on drug discovery projects against schizophrenia, Alzheimer's, diabetes, osteoporosis and other diseases.
To contact Derek email him directly: derekb.lowe@gmail.com
Twitter: Dereklowe
A couple of commenters took exception to my words yesterday about thiophene not being a "real" heterocycle. And I have to say, on reflection, that they're right. When I think about it, I have seen an example myself, in a project some years ago, where thiophene-for-phenyl was not a silent switch. If I recall correctly, the thiophene was surprisingly more potent, and that seems to be the direction that other people have seen as well. Anyone know of an example where a thiophene kills the activity compared to a phenyl?
That said, the great majority of the times I've seen matched pairs of compounds with this change, there's been no real difference in activity. I haven't seen as many PK comparisons, but the ones I can think of have been pretty close. That's not always the case, though: Plavix (clopidogrel) is the canonical example of a thiophene that gets metabolically unzipped (scroll down on that page to "Pharmacokinetics and metabolism" to see the scheme). You're not going to see a phenyl ring do that, of course - it'll get oxidized to the phenol, likely as not, but that'll get glucuronidated or something and sluiced out the kidneys, taking everything else with it. But note also that depending on things like CYP2C19 to produce your active drug for you is not without risks: people vary in their enzyme profiles, and you might find that your blood levels in a real patient population are rather jumpier than you'd hoped for.
So I'll take back my comments: thiophene really is (or at least can be) a heterocycle all its own, and not just a phenyl with eye makeup. But one of the conclusions of that GSK paper was that it's not such a great heterocycle for drug development, in the end.
Drawing programs like ChemDraw render 5-membered rings as regular pentagons by default, with sides and angles of equal measure (108 degrees for the latter). This is a bit of a distortion (literally) when applied to thiophene:
C-C bond length (distal to S) = 1.41 A
C-C bond lengths (proximal to S) = 1.34 A
C-S bond lengths = 1.70 A
C-S-C angle = 93 degrees (!)
C-C-S angles = 109 degrees
C-C-C angles = 114 degrees
Crazily enough, the C-S-C bond angle in thiazole is 89 degrees – it's an acute angle!
Stick by your guns Derek. I agree with your original assessment. Although thiophene is technically a heterocycle by the strictest definition, I don't consider it one from a med chem perspective. Just because it can show differences in activity over phenyl, that doesn't, in my book, change its designation. For me, heterocycles are nitrogenous and impact the properties of the compound, which in general thiophene does not. Thiophene, along with furan and N-phenyl pyrrole are what I call "academic heterocycles". That is, they are used in transition metal catalyzed methodology papers to claim the method works with "heterocycles".
I think the thiophene was a carryover from the older antiplatelet agent ticlodipine. That was identified ca 1970 so quite possibly found in an in vivo screen
What about the phenyl-for-thiophene analog of amphetamine being offered from various online vendors as a "research chemical, not for human consumption". Haven't looked into it myself, but would be curious to know how its activity differs from amphetamine.
"Now I say heterocycle but any _reasonable_ person would know that I was in fact talking about an aromatic compound with one or more integral nitrogen atoms since they are the only ones _I_ find useful."
Miranda Priestly has got nothing on you. Not you Derek, '4. Rock'.
15. David Borhani on July 20, 2012 12:51 PM writes...
How about tiotropium (Spiriva)? Minimal metabolism (with *TWO* thiophene rings). From the prescribing info:
Metabolism
The extent of metabolism appears to be small. This is evident from a urinary excretion of 74% of unchanged substance after an intravenous dose to young healthy
volunteers. Tiotropium, an ester, is nonenzymatically cleaved to the alcohol N-methylscopine and dithienylglycolic acid, neither of which binds to muscarinic receptors.
In vitro experiments with human liver microsomes and human hepatocytes suggest that a fraction of the administered dose (74% of an intravenous dose is excreted
unchanged in the urine, leaving 25% for metabolism) is metabolized by cytochrome P450-dependent oxidation and subsequent glutathione conjugation to a variety of
Phase II metabolites. This enzymatic pathway can be inhibited by CYP450 2D6 and 3A4 inhibitors, such as quinidine, ketoconazole, and gestodene. Thus, CYP450
2D6 and 3A4 are involved in the metabolic pathway that is responsible for the elimination of a small part of the administered dose. In vitro studies using human liver
microsomes showed that tiotropium in supra-therapeutic concentrations did not inhibit CYP450 1A1, 1A2, 2B6, 2C9, 2C19, 2D6, 2E1, or 3A4.
Elimination
The terminal elimination half-life of tiotropium was between 5 and 6 days following inhalation. Total clearance was 880 mL/min after an intravenous dose in young
healthy volunteers with an inter-individual variability of 22%. Intravenously administered tiotropium was mainly excreted unchanged in urine (74%). After dry powder
inhalation, urinary excretion was 14% of the dose, the remainder being mainly non-absorbed drug in the gut which was eliminated via the feces. The renal clearance of
tiotropium exceeds the creatinine clearance, indicating active secretion into the urine. After chronic once-daily inhalation by COPD patients, pharmacokinetic steady
state was reached after 2 to 3 weeks with no accumulation thereafter.
1. alig on July 19, 2012 1:02 PM writes...
I have seen it go both ways. Phenyl being >1000X more potent than thiophene and vice versa.
Permalink to Comment2. Giagan on July 19, 2012 2:44 PM writes...
Drawing programs like ChemDraw render 5-membered rings as regular pentagons by default, with sides and angles of equal measure (108 degrees for the latter). This is a bit of a distortion (literally) when applied to thiophene:
C-C bond length (distal to S) = 1.41 A
C-C bond lengths (proximal to S) = 1.34 A
C-S bond lengths = 1.70 A
C-S-C angle = 93 degrees (!)
C-C-S angles = 109 degrees
C-C-C angles = 114 degrees
Crazily enough, the C-S-C bond angle in thiazole is 89 degrees – it's an acute angle!
Permalink to Comment3. Toad on July 19, 2012 4:26 PM writes...
Benzene and thiophene look quite different when you look at the field and the associated point charges in something like GAMESS.
Permalink to Comment4. Rock on July 19, 2012 9:42 PM writes...
Stick by your guns Derek. I agree with your original assessment. Although thiophene is technically a heterocycle by the strictest definition, I don't consider it one from a med chem perspective. Just because it can show differences in activity over phenyl, that doesn't, in my book, change its designation. For me, heterocycles are nitrogenous and impact the properties of the compound, which in general thiophene does not. Thiophene, along with furan and N-phenyl pyrrole are what I call "academic heterocycles". That is, they are used in transition metal catalyzed methodology papers to claim the method works with "heterocycles".
Permalink to Comment5. Chris on July 20, 2012 1:41 AM writes...
Out of curiosity does anyone know how Plavix was discovered, it seems an unlikely deliberate choice as a prodrug? Was it an in vivo screen?
Permalink to Comment6. petros on July 20, 2012 2:12 AM writes...
I think the thiophene was a carryover from the older antiplatelet agent ticlodipine. That was identified ca 1970 so quite possibly found in an in vivo screen
Permalink to Comment7. gippgig on July 20, 2012 2:24 AM writes...
Has anyone looked at the thiophene analog of DOPA? I've wondered about that one for a long time...
Permalink to Comment8. chris on July 20, 2012 2:41 AM writes...
Thanks petros, a search reveals Ticlodipine was discovered in an in vivo screen looking for anti-inflammatory compounds.
Permalink to Comment9. BD on July 20, 2012 3:28 AM writes...
What about the phenyl-for-thiophene analog of amphetamine being offered from various online vendors as a "research chemical, not for human consumption". Haven't looked into it myself, but would be curious to know how its activity differs from amphetamine.
Permalink to Comment10. iupac-fan on July 20, 2012 5:00 AM writes...
this saga is for me incomprehensible. is it cyclic? does it contain a heteroatom?
how about cellulose? how many thousands of heterocycles therein? thousands, or none?
if you mean aromatic nitrogen heterocycle, they you'd better say so as it's rather more specific....
Permalink to Comment11. Morten G on July 20, 2012 6:10 AM writes...
"Now I say heterocycle but any _reasonable_ person would know that I was in fact talking about an aromatic compound with one or more integral nitrogen atoms since they are the only ones _I_ find useful."
Miranda Priestly has got nothing on you. Not you Derek, '4. Rock'.
Permalink to Comment12. Myma on July 20, 2012 7:48 AM writes...
Thiophene smells better than benzene.
Permalink to Comment13. drug_hunter on July 20, 2012 9:45 AM writes...
1000X difference between benzene and thiophene!?
Everyone, references please for large (>10X) differences -- would be welcome and instructive.
Permalink to Comment14. Med Chem on July 20, 2012 11:03 AM writes...
here is a good example of thiophene to phenyl switch:
http://pubs.acs.org/doi/abs/10.1021/jm1011726
provided a switch between Plk and Nek enzymes
Permalink to Comment15. David Borhani on July 20, 2012 12:51 PM writes...
How about tiotropium (Spiriva)? Minimal metabolism (with *TWO* thiophene rings). From the prescribing info:
Metabolism
Permalink to CommentThe extent of metabolism appears to be small. This is evident from a urinary excretion of 74% of unchanged substance after an intravenous dose to young healthy
volunteers. Tiotropium, an ester, is nonenzymatically cleaved to the alcohol N-methylscopine and dithienylglycolic acid, neither of which binds to muscarinic receptors.
In vitro experiments with human liver microsomes and human hepatocytes suggest that a fraction of the administered dose (74% of an intravenous dose is excreted
unchanged in the urine, leaving 25% for metabolism) is metabolized by cytochrome P450-dependent oxidation and subsequent glutathione conjugation to a variety of
Phase II metabolites. This enzymatic pathway can be inhibited by CYP450 2D6 and 3A4 inhibitors, such as quinidine, ketoconazole, and gestodene. Thus, CYP450
2D6 and 3A4 are involved in the metabolic pathway that is responsible for the elimination of a small part of the administered dose. In vitro studies using human liver
microsomes showed that tiotropium in supra-therapeutic concentrations did not inhibit CYP450 1A1, 1A2, 2B6, 2C9, 2C19, 2D6, 2E1, or 3A4.
Elimination
The terminal elimination half-life of tiotropium was between 5 and 6 days following inhalation. Total clearance was 880 mL/min after an intravenous dose in young
healthy volunteers with an inter-individual variability of 22%. Intravenously administered tiotropium was mainly excreted unchanged in urine (74%). After dry powder
inhalation, urinary excretion was 14% of the dose, the remainder being mainly non-absorbed drug in the gut which was eliminated via the feces. The renal clearance of
tiotropium exceeds the creatinine clearance, indicating active secretion into the urine. After chronic once-daily inhalation by COPD patients, pharmacokinetic steady
state was reached after 2 to 3 weeks with no accumulation thereafter.
16. sean on July 21, 2012 5:30 PM writes...
In reference to Giagan's comment - ChemOffice leaves a lot to desire.
Permalink to Comment