Neuraminidase Inhibitor Synthesis

With a likely possibility of an influenza epidemic starting from the nearly worldwide H5N1 virus, I have done a large amount of research relating to the synthesis of neuraminidase inhibitor drugs, which prevent the influenza virus from leaving infected cells. Currently, my effort has been focused on oseltamivir phosphate, sold commercially as tamiflu, since it is effective against H5N1 and information was fairly available. It now appears that their are strains resistant to oseltamivir however, so I've started to research other drugs as a contingency plan.

Feel free to discuss other NA inhibitors here, and other related topics. For precursors, try not to dicuss them here if they have another thread dedicated to them.

Here is what I have so far. All of the chemicals in the procedure are actually doable OTC, but I have not yet typed it all up yet. Search rhodium, SM, orgsyn and the hive (helps if you have archives of rhodium and the hive) for details until then.
One thing I didn't see there is making acetonitrile by dehydrating acetamide with thionyl chloride. That would be the easiest otc route, since thionyl chloride is needed in the synth anyway. I think vogel might have a procedure....

Here is the draft of my work so far.
Original procedures that this is based on are found in US patents 6,797,832 and 5,859,284
I put a bunch of refs at the end of the write up. They probably explain all the steps.
For the process based off the first patent, all the solvent volumes are decreased because, compared to the second patent, it seems they went solvent happy being in a well equipped lab and not trying to do anything on any sort of useful scale.

Advantages of my synthesis:
-All starting chemicals are doable OTC for a determined chemist (and since this could be a life or death thing, the chemist is most likely quite determined)
-Unlike the patented procedures, no use of reduction hydrogenation and expensive catalysts for making the drug (you'll need to do a raney nickel hydrogenation for making tetrahydrofuran)
-No obscure chemicals needed for the procedure, anyone here determined enough should be able to pull this off


See the attached pic for a synth diagram, so that all my chemical labels ('A', 'D', etc) make sense.


OTC Oseltamiver Synthesis
By Chris The Great
2006

This document has not yet been tested to ensure that it works as presented.


Step One: Synthesis of A
Methyl acrylate (2.78mol) is added to furan (3.95mol) and cooled to -10*C. AlCl3 in DCM (1.15mol in 1-5M conc.) is added slowly with stirring, keeping the temperature below -10*C. The mixture is then cooled with a cold water bath and stirred for 2 hours. The mixture is added to 1000ml of saturated sodium bicarbonate solution, and the precipitate is filtered out. The aqeous layer is extracted twice with 250ml DCM, and the organic layers are combined and dried over anhydrous sodium sulfate. The solvent is distilled away and the product purified by silica gel column chromatography to give the desired product in 38.6% yield (1.07mol).
[It is likely it can be purified through recrystallization but I have found not details on this. In all likelyhood the unwanted optical would have similar solubility which complicates things.]

Step Two: Synthesis of B
A (1.07mol) is added to NaOH (1.65mol, 3 to 15M concentration) in water with stirring, maintaining a temperature of 5*C. The mixture is allowed to warm up over a period of 30 minutes with stirring, and stirred at room temperature for 2 hours. HCl (310ml of 31% sol.) is added to adjust the pH to below 1, and then the mixture is extracted 4 times with 200ml of DCM. The combined organic layers are dried with Na2SO4, and solvent evaporated and the product recrystallized from ethyl acetate to give the desired product in 87.5% yield (0.936mol).

Step Three: Synthesis of C
B (0.936mol) is dissolved into 850ml water, and NaHCO3 (0.936mol) is added slowly, with stirring to control the foaming. After foaming ceases, bromine (0.936mol) is added dropwise with stirring, and then the mixture is stirred at room temperature for 2 hours. The mixture is extracted with 400ml ethyl acetate, washed with 100ml thiosulfate solution (to remove unreacted bromine) and dried with Na2SO4. The solvent is distilled away and the product purfied by recrystallizing from ethyl acetate, giving the desired product in 89.6% yeild (0.84mol)

Step Four: Synthesis of D
Potassium hydroxide (2.5mol) is dissolved into dimethylacetamide (1000ml) and methanol (150ml), and C (0.84mol) is added slowly as the mixture is heated to a gentle reflux. It is stirred under reflux for 3 hours, then ethyl iodide (1.6mol) is added and the mixture stirred under reflux for another 3 hours. To the mixture is added HCl (300ml 31%) and the solution extracted with ethyl acetate (250ml) eight times. The organic layer is washed with saturated brine and dried with Na2SO4. The solvent is distilled away and the product purified by recrystallization to give the desired product in 95.9% yield (0.804mol).

Step Five: Synthesis of E
A solution of sodium methoxide (0.9mol) in tetrahydrofuran (500ml) is cooled to -30*C, and D (0.804mol) in tetrahydrofuran (500ml) is added dropwise with stirring. After the addition, the mixture is stirred at 30*C for 1hr. To the mixture is added acetic anhydride (1.05mol) and the mixture is stirred at room temperature for 3hr. To the reaction mixture is added saturated ammonium chloride (1000ml), and the mixture is extracted with ethyl acetate (800ml). The organic layer is dried (Na2SO4) and the solvent distilled away. The product is recrystallized (ethyl acetate) to give the desired product in 80.0% yeild (0.644mol).

Step Six: Synthesis of F
E (0.644mol) is dissolved into 3-pentanol (600ml) at reflux, and AlCl3 (0.7mol) is slowly added with stirring. The mixture is stirred at reflux for 1 hour, and then allowed to cool to room temperature. A saturated solution of NaHCO3 is added (1000ml), and the mixture is extracted with 750ml DCM, washed with saturated brine (750ml) and dried with Na2SO4. The solvent is distilled away and the product recrystallized (DCM) to give the desired product in 81.2% yield (0.522mol)

Step Seven: Synthesis of G
F (0.522mol) is dissolved in DCM (750ml) and triethylamine (0.78mol) is added. The mixture is stirred at room temperature for 5 min. and then cooled to 0*C. To the mixture is added tosyl chloride (0.78mol) and the mixture is stirred at room temperature for 1hr. The reaction mixture is washed with saturated sodium bicarbonate solution (1500ml), saturated brine (750ml) and then dried with sodium sulfate. The solvent is distilled off, and the crude product recrystallized from DCM to give G in 97.3% yeild (0.508mol).

Step Eight: Synthesis H
G (0.508mol) is dissolved in ethanol (1000ml) at reflux, and potassium carbonate (0.25mol) is added with stirring. The mixture is stirred at gentle reflux for 1.5 hours, and then cooled to room temperature. Saturated ammonium chloride solution (1000ml) is added. The mixture is extracted with DCM (750ml), washed with saturated brine (1000ml) and dried (Na2SO4). The solvent is distilled away and the crude product recrystallized from DCM to give H in 98.5% yeild (0.5mol).

Step Nine: Synthesis of I
H (0.5mol), sodium azide (0.6mol) and ammonium chloride (0.6mol) are dissolved into water (70ml) and ethanol (250ml), and refluxed for 8 hours. Aqeous NaHCO3 (105ml of 8% solution) is added and the ethanol distilled in vacuum. The aqeous residue is extracted with ethyl acetate (250ml), the extract washed with water (125ml). The wash is back extracted with 125ml of ethyl acetate and the combined organic extracts washed with brine (125ml), dried over Na2SO4, filtered and concentrated in vacuum to give I as a dark brown oil in 102% yield (0.512mol)

Step Ten: Synthesis of J
Step One:
J (0.512mol) and ammonium chloride (1.2mol) are dissolved into ethanol (1500ml), and zinc (0.7mol) is added and the mixture refluxed for 30 minutes. The solvent is distilled off to give Ib, which is used directly in the next step.
Step Two:
Tosysl Chloride (1.13mol) is added portionwise at room temperature to a stirred mixture of Ib (from the step one), K2CO3 (2.5mol) in acetonitrile (1000ml) and stirred for 6hr. Toluene (2500ml) is added, the solid is filtered off and the solvent evaporated to give J in approx 80-90% yeild. It is used directly in the next step without purification.

Step Eleven: Synthesis of K
J (0.494mol), sodium azide (1.2mol) and ammonium chloride (1.2mol) in dimethylacetamide (400ml) is heated to 80-85*C for 5 hours. NaHCO3 (50mmol) in water (250ml) is added, and the mixture extracted with hexanes (6x250ml). The combined hexane extracts are concentrated to 1200ml, and 250ml DCM is added, followed by 1100ml NaHCO3 8% solution and acetic anhydride (0.6mol). The mixture is stirred at room temperature for an hour, and then the aqeous layer is removed. The organic phases are concentrated in vacuum to 430g total weight), dissolved in ethyl acetate (50ml). The mixture is cooled and K crystallizes out and is collected by filtering. The crystals are washed with cold 15% ethyl acetate in hexane (250ml) and dried in a vacuum at room temperature, to give K in 55% yield (0.272mol). [may be slighly higher, ie 60%]

Step Twelve: Synthesis of L (oseltamivir freebase)
K (0.272mol) is dissolved into ethanol (600ml) along with ammonium chloride (0.64mol). Zinc (0.36mol) is added and the mixture refluxed for 30 minutes. The precipitate is filtered out, giving L in 98% yeild (0.266mol). [90-95% more likely]

Step Thirteen: Synthesis of Oseltamivir Phosphate
L (0.266mol) is dissolved in acetone (1000ml) and treated with phosphoric acid (85%, 0.266mol) in absolute ethanol (300ml). The mixture is cooled, and after 12 hours the precipitate is filtered out to give oseltamivir phosphate in 75% yield (0.2mol, 82g).
A second crop of presumably lower purity crystals can be obtained by concentrated the solution and collecting a second crop of product.


References and futher reading:

Direct conversion of ethers to esters by trichloroisocyanuric acid, tetrahedron letters, no 55, pg 5819-5820 (easy prep of ethyl acetate from ether, pdf was found on SM)

See the threads here for stuff on thionyl chloride, especially thionyl chloride from sulfur dichloride and SO3 (and see the thread on that too), acetic anhydride, dimethylformamide (dimethylacetamide is made in pretty much the same procedure but with acetic acid vs. formic acid),

Short and Efficient synthesis of optically active N-tosyl aziridines from 2-amino alchols, Lother W. Bieber & Maria C.F. da Araujo, Molecules 2002, 7, 902-906 (found on the net somewhere)

Reduction of azides to amines or amides with zinv and ammonium chloride as a reducing agent, W. Lin, X. Zhang, Z. He, Y. Jin, L. Gong, Synth. Commun. 32(21), 3279-3284 (2002) (found the essential details from here on a rhodium backup)

Preparation of Methanesulfonyl chloride (see the thread and attached pdf for the synth) in this case applied to making toluenesulfonyl chloride since that is OTC

See US Patent 4,725,311 for details on making easy sodium in decent amounts OTC

Nice work here Pitty I can't get my hands on some of the chemicals mentioned in the synthetic pathway. I will try getting some azides from air-bag systems and buy another couple of chemicals. In my country we are to be given free neuramidase inhibitor (Tamiflu) if one's doctor prescribes it after reporting sickness to the authority. It would still be great to be able to carry out the synthesis and who knows, if all hell breaks loose, it might turn out useful afterall. The supply is unfortunately not that large, where 25% of the population is covered, although I do understand the procedure.

Anyways, let us focus on the neuraminidase inhibitors. There are two other commonly known antiviral drugs, Amatadine and Zanamivir. The former, Amatadine, turns out to be pretty much uneffective, since it is only effective against Influenza A and the influenza is quick to mutate and become resistant to it, besides some other side effects which do not make it feasable. A report (the NICE report if IIRC) has shown that it does not show any correlation with the reduction in the number of hospitalizations and may lead to other complications. Therefore synthesizing it would be a waste of time and resources.

Zamavir on the other hand seems good enough, being able to treat Influenza A and B. It is contraindicated in children smaller than 12 years and in people with asthma or Chronic Pulmonary Disease because it seems to be able to cause bronchospasm. Zamavir is taken as an inhaler, and that is a disadvantage if one intends to produce it OTC.

Thus, Ostelamivir seems to be the best solution. Obviously I would recommend a normal seasonal influenza vaccine. Risking it won't cost anything (or pretty much nothing) and the pandemic influenza might have certain epitopes (resulting from 'fusion' of the avian and seasonal influenza - hybridization) which the bopdy might be primed to recognize by the vaccine and thus one will have a larger and faster immune response to the pandemic.

Oseltamivir prophylaxis has been shown to very effective in reducing the occurence of influenza. Many countries are stockpiling neuramidases for prophylactic treatment of front-line workers (doctors, nurses, surgeons, etc...). Thus in many countries Oseltamivir are being used mainly as a treatment following infection (recommended before 24-48hours max. following infection). Hope this helps. Ask if you need any other info.

Edit: The synthetic pathway of oseltamivir may be found on wikipedia - link. It may be useful for you to check out certain steps in the synthesis.

[Edited on 23-1-2006 by Esplosivo]

Yes, I have seen the pathway detailed on wikipedia before, and have the actual journal with all the details for it. But, for the extra amount of non-easily OTCable chemicals needed for a 1 to 2% increase in total yeild, I decided to stick with the original azide route since the chemicals are cheap and the procedures are easy. Avoiding azides makes sense ofr a 100kg batch but none of us here are going to be doing those. I am thinking 100g batches would be the most you could do easily on a large lab scale if you managed to do everything before on a larger scale.

What are the chemicals in particular you are not able to get. Everything in there is OTC or can be made from OTC starting materials. I am also planning on the airbag route for azides since production on this scale will pose problematic.

Two other neuraminidase inhibitors, both orally available and more active than tamiflu, are "RWJ 207201" and "A 315675". Both have not been released commercially. I have found a patent for A 315675 that might be OTC-able but it looks an order of magnitude more difficult and lower yielding than oseltamivir. The patent in question is #6,455,571... though it took me half an hour to find what reaction scheme produced it (it is structure 50a) and I still haven't figured out what examples detail the reactions.

I have attached structural diagrams of the two drugs.

Another disadvantage with Zanamivir is that administration is difficult when trying to use it against a virus that attacks the lungs. It is more active than oseltamivir, but its poor bioavailability takes away that advantage. The above two compounds have good bioavailability and are also more active than zanamivir. A315675 in particular is a very strong inhibitor, and binds very strongly to nueraminidase. While oseltamivir has a half-life of about 33 minutes, A315675 has a half life of 12 hours! (Note: these values are not representative of in body conditions, and represent an ideal situation for the de-bonding of the drugs to the neuraminidase) It also seems to average out to be roughly 10 times more active.
But, as I said before, I haven't worked out the procedure yet, but it does not look OTC without vast amounts of work, and even then it looks like a poor yielding procedure. I would much rather focus my lab resources towards something I am sure I can make than towards something I might be able to make.
However, if an oseltamivir resitant strian goes pandemic, looks like at least knowing a procedure would be useful since it would them be pretty much your only hope.

[Edited on 24-1-2006 by Chris The Great]

Before you buy a pill-pressing machine for this compound , projects looking smooth on paper pose unpredictible difficulties in the real life situation (lab), esp if the lab is a kitchen. What first seemed to be over in an afternoon can turn out to be very time consuming even in a real lab and sometimes it simply dosen't work even if you follow allready published method exactly. Don't think it's a good idea to plan in detail as all kind of weird things happen on the way...

[Edited on 1-7-2006 by Sandmeyer]