why does this molecule take 10 - 12 steps to create (and ~ 12 months to produce)?



I know that shikimic acid:



is the "bottleneck" in the production, according to an article in the NY Times. But a guy I work with retired from Mallinckrodt and said that they sold gallic acid very cheaply (it was an unwanted byproduct in the production of tannic acid):



Why wouldn't it be possible to synthesize shikimic acid by hydrogenating gallic acid and separating the resulting mixture? Currently, Roche gets it shikimic acid from a natural source - star anise.

11/8/2005 2:47:47 PM

study hall?
did you mean to put this in the lounge?

11/8/2005 2:52:57 PM

I dunno but my dads pharmacy sells buttloads of that stuff.

11/8/2005 2:58:26 PM

wait for the stock piling...

11/8/2005 3:02:48 PM

y do u want to know about this so badly that you posted on tww about it?

11/8/2005 3:38:11 PM

Tamiflu lab in his garage.

11/8/2005 4:04:38 PM

Maybe there are chem students that would know?

11/8/2005 11:00:22 PM

sure, ill tell you why.

this bitch is near impossible to hydrogenate. youll observe that some organic textbooks will tell you its IMPOSSIBLE to hydrogenate a benzene ring. however, withe EXTREMELY high temps and EXTREMELY high pressures plus an expensive pt or pd catalyst...you can do it.

im not sure what you would be left with if you tried to saturate that benze ring...maybe something would happen to the alocohol groups....i dont know.

but what i can tell you is that i can sound synthesize tamiflu with a jack knife and a rubber band. thats not the issue. the issue is someones got a patent.

[Edited on November 9, 2005 at 12:04 AM. Reason : =]

11/8/2005 11:45:33 PM

ok....heres a 4 step synthesis.

everyone stock up on those raw materials....

its quite easy to make the stuff. roche just doesnt want to outsouce. they make more money that way.



let me know if you see any problems with it. everything looks square to me.

[Edited on November 9, 2005 at 12:54 AM. Reason : -]

11/9/2005 12:39:52 AM

http://www.nytimes.com/2005/11/05/business/05tamiflu.html

I posted this in study hall because I've seen some good responses to chemistry questions in here. The Times article is about how experts are debating whether or not shikimic acid is the bottleneck in Roche's process.

Roche claims they can make tamiflu faster than any other plant could, so it wouldn't make sense to force them to license production of Tamiflu to generic pharma companies. They claim that it takes 10-12 months to produce, and about a ten-step process.

I work in process development for a generic pharma company. The gallic-acid to shikimic acid route is not the way that Roche manufactures Tamiflu. I was just wondering if someone on here could think of a better way. You can believe whoever finds a faster process (a way to produce (literal) tons of shikimic acid cheaply) is going to be very wealthy. Patent or not, Roche is getting leaned on very heavily to out-source production. Ranbaxy has said they're ready to start making Tamiflu already.

11/9/2005 11:55:54 AM

my way has no bottleneck. and if i can do that in 5 mins, im sure there is a much more elegant synthetic rout out there. i just dont see the need to start with cyclic precursor. it makes sythesis a bitch. making the complex linear stuff is so much easier, and there are so many ring closing mechanisms out there. i just happen to pick diels-alder bc the respective electron donating/withdrawing groups for the dien/ienophile just happen to obey the rules, which is gonna make for a favored cycloaddition and of course becuase the product is a hexene.



^as you guessed, there are of course other ways and it is of course a matter of patent rights and the fact of not wanting to give up corperate profits.


[Edited on November 9, 2005 at 1:01 PM. Reason : -]

11/9/2005 12:48:19 PM

wow, it takes nutball here 5 minutes only 5 minutes you could be rich!!1

dumbass

11/9/2005 12:54:29 PM

One word - stereochemistry

11/9/2005 5:52:14 PM

the stereochemistry of that molecule is NOT complicated, nor a major issue

11/9/2005 8:59:02 PM

::puts two bits on Airlia::

11/9/2005 9:08:21 PM

this thread has nothing to do with patent rights.

Like I said, they aren't going to mean shit for this product. Roche IS BEING FORCED TO ALLOW OTHER COMPANIES TO PRODUCE TAMIFLU

and yes, the stereochemistry is the bitch of it. Your synthesis route would create a mixture of the different enantiomers of shikimic acid. I suspect the *bottleneck* would be in purifying the resulting mixture to get 99% shikimic acid.

But at least we agree that shikimic acid should be synthesized easily enough. Right now it is extracted from star anise, an herb only grown in south china

BTW: 1000s of tons of benzene are hydrogenated every day to refine crude oil. It's basic CHE process

[Edited on November 9, 2005 at 9:13 PM. Reason : benzene crackin]

11/9/2005 9:10:45 PM

^i stand corrected then...but i agree stereochem is an issue, ive just seen worse. 3 stereocenters seems small for your average pharmaceutical.

but if its true that its easy enough to hydrogenate like you say, economically speaking, then i agree thats a promising synthetic route. it just would be more fun to do a cycloaddition imo. i wouldnt really think a purification process would be a bottleneck for the enantiomers. its a bitch, but pharma companies do it all day long.

11/9/2005 9:24:09 PM

star anise is a porn name

11/9/2005 11:44:51 PM

Wait, I thought josh8315 had figured something out that many chemist PhDs have been working on for years couldnt figure out ok

[Edited on November 10, 2005 at 10:11 AM. Reason : xfghsdfg]

11/10/2005 10:11:26 AM

Josh, my process chemist liked your snythesis route.

Only problem: the di-amino enol you used as a starting material (purchase) "doesn't exist" according to him.

And trust me, there weren't "hundreds of PhDs" working on this before like 3 months ago. No one outside of Roche gave two shits about Tamiflu before it got blown up as the only saviour for humanity.

11/10/2005 4:23:34 PM