Taken by mouth for type 2 diabetes. It blocks the enzyme that breaks down GLP-1, so the body's own signal to release insulin after a meal lasts longer. Sold as the phosphate salt.
DPP-4 inhibitor
FDA 2006
C16H15F6N5O · 407.32 g/mol
CAS 486460-32-6
Chemical resolution route
5 steps · 11% overall · 96% ee
The whole point of this route is that it installs the chiral amine without a noble metal. The enamine is reduced with cheap NaBH4 to give the racemate, and the two enantiomers are then separated with (−)-di-p-toluoyl-L-tartaric acid. That trades yield for cost: 11% overall against Merck's 52% first-generation process, but no rhodium or ruthenium, no high-pressure hydrogenation, and no need to strip precious metal out of the product stream. The separated filtrate still holds the S enantiomer at 90% ee, so neither half is thrown away.
Scale: laboratory scale, 5 to 10 g per step
no public CAS
+
no public CAS
Step 187% isolated
CDI, dry THF 50 °C, 5 h
Meldrum's acid acts as the acyl anion equivalent · CDI replaces pivaloyl chloride here
no public CAS
no public CAS
+
no public CAS
Step 267% isolated
i-Pr2NEt, isopropyl acetate 75–80 °C, 5 h
amide formation onto the triazolopyrazine, giving the β-ketoamide
no public CAS
no public CAS
Step 392% isolated
NH4OAc, aqueous NH3, MeOH 58 °C, then cooled
the enamine crystallises straight out of the mixture · 99 wt% by simple filtration
no public CAS
no public CAS
Step 460% isolated
NaBH4 with MsOH, THF / i-PrOH −15 °C, 4.5 h
gives the racemate · NaBH4 alone does nothing (0% conversion); the acid additive is what makes it go, and MsOH was chosen over BF3·OEt2 on toxicity grounds
no public CAS
no public CAS
Step 533% isolated
(−)-Di-p-toluoyl-L-tartaric acid, MeOH / i-PrOH, 65 °C then aq. NH3, then 85 wt% H3PO4, 78 °C
the tartrate of the R enantiomer crystallises out, is freed with ammonia, and is taken to the phosphate salt · 96% ee, 99 wt%
The same β-ketoamide is hydrogenated to the alcohol, the alcohol is turned into its mesylate, and ammonia then displaces the mesylate to give the amine with inversion. Compounds 14 and 15 were made here for the first time. The authors report it as an alternative rather than a better route: the optical purity collapses to 7% ee by the end, so the stereocentre set in the first step is largely lost in the displacement.
Scale: laboratory scale, 3 to 10 g per step
no public CAS
Step 198% assay
[Ru(cymene)Cl]2 with (S)-Ts-DPEN, MeOH H2 at 435 psi, 65 °C, 12 h
asymmetric reduction of the ketone · formic acid / Et3N as the hydrogen source did not work at all, even at 80 °C
no public CAS
no public CAS
Step 242% isolated
MsCl, Et3N, ethyl acetate −10 °C, 2 h
the alcohol becomes a leaving group · 98 wt% after chromatography, but already only 7% ee
no public CAS
no public CAS
Step 333% isolated
NH3 in MeOH (13 wt%) 65 °C, 8 h at 145 psi, then H3PO4
ammonia displaces the mesylate, inverting the centre to R · 99 wt% but 7% ee