Index

• Faculty Mentor: John Hanson (hanson@ups.edu)
• Project TA: Rollie Williams (jrwilliams@pugetsound.edu)
• See the notes from 2009, 2006, 2005
• Students working on this project
• Table of Reagents and Amounts Available
• Step 1. Formation of 2-Acetamido-3,4,6-tri-O-acetyl-2-deoxy-a-D-glucosyl Chloride
• Step 2. Reaction of 2-Nitrophenol with the Glycosyl Chloride
• Step 3. Removal of Acetate Esters
• Step 4. Formation of the 4,6-Isopropylidene Group

T4 lysozyme is one of the most thoroughly studied enzymes from a structural point of view. Unfortunately, there is currently no convenient small-molecule assay for the enzyme. As part of an effort to synthesize a substrate for this enzyme I would like to prepare the o-nitrophenyl muramic acid derivative 6. I used a similar route to prepare the corresponding p-nitrophenyl derivative. N-Acetylgluosamine (1) is treated with acetyl chloride to produce the tetraacetyl chloride 2. Base-catalyzed substitution of o-nitrophenol should afford the glycoside 3. Removal of the acetate esters with methoxide, followed by selective protection of the 4 and 6 hydroxyls as an acetonide, affords the 3-hydroxy derivative 5. Alkylation with ethyl 2-bromopropionate or 2-bromopropionic acid should afford the desired muramic acid derivative.

The table below lists the chemicals that we will have available for this project. If you need something that is not on this list, consult with John Hanson. Also note the "Amount/group" column. This is the total amount of material available for each group to use on the project.

Step 1: Formation of 2-Acetamido-3,4,6-tri-O-acetyl-2-deoxy-a-D-glucosyl chloride

Literature References

1. Horton, D. Organic Syntheses Coll. Vol. 5, p.1 or Organic Syntheses 1966, 46, 1.
2. Heidlas, J.E.; Lees, W.J.; Pale, P.; Whitesides, G.M. J. Org. Chem. 1992, 57, 146-151.

Notes and Suggestions

1. See the observations from 2005.
2. Acetyl chloride should be handled with care. It reacts vigorously with water to generate HCl. Dispense in the hood.
3. I found that I had to heat the reaction gently to get it to react. It can then stir at room temperature for several days until you decide to work it up.
4. The pentaacetyl glucosamine is a lower running side product. It won't interfere with our next reaction.
5. 50:50 EtOAc:Hexane is a good TLC solvent. The desired product has an Rf of about 0.5 while the pentacetate is around 0.2.
6. Try doing this reaction starting with 10 g of N-Acetylglucosamine.
7. Store the product in a dessicator containing NaOH or KOH. (Remind me to put out some of the small dessicators for this purpose.

Notes, Observations, and Suggestions

1. (3/25/10) -- Derek and Morgan set up their reaction around noon today. I had them make their drying tube using a type of silica based dessicant rather than calcium chloride. I put the bottle of dessicant and drying tubes in the lysozyme hood. After adding the N-acetylglucosamine, not much seemed to happen, the solid N-acetylglucosamine was just stirring in the acetyl chloride solution. After 2 hours there didn't seem to be much change, but when Christine and Brian came in to set up their reaction around 3:30 pm they saw Derek and Morgan's reaction start to boil/reflux and the solid went into solution. In other news....Caitlin and Hallie decided to set up their reaction on Monday.
2. (3/26/10) Christine and Brian's reaction also become homogenous after stirring overnight, but it is so viscous that it is not really stirring well. I have seen this behavior before and the reaction still seemed to work well. When I left at around 6 pm, Alli and Rachel's reaction was refluxing and becoming clear. Robert and Susan's reaction, which had been set up earlier in the day, was already clear by this point. It appears that this reaction has an induction period of several hours before it starts to show appreciable signs of reacting. But then, over a relatively short time (15 min?), it begins to reflux and eventually everything goes into solution.
3. After your reaction has finished this initial reaction phase, you may remove the condenser and drying tube and securely cap the reaction flask with a yellow cap until you are ready to work it up. After you are done with the drying tube, place it in the large jar with dessicant so it will stay dry.
4. (3/31/10) Derek and Morgan obtained about 8 g of product. Their NMR looks very good, only a small amount (5%?) of the pentaacetate. I haven't heard from anyone else yet on their yield. But the Tues. Eve. and Wed. Aft. labs look like they are getting lots of crystals, so it looks like this first step is working well.
5. (4/2/10) Christine and Brian got 9.5 g of product and the NMR looked good.

Step 2: Reaction of 2-Nitrophenol with the Glycosyl Chloride

Literature Reference

1. Adapted from: Kur'yanov, V.O.; Chupakhina, T.A.; Zemlyakov, A.E.; Chirva, V.Y.; Ischenko, V.V.; Khilya, V.P. Chemistry of Natural Compounds 2001, 37, 39.

Notes and Suggestions

1. The procedure from the reference listed above used 15-crown-5, but I had better luck using 18-crown-6.
2. I used 1 mmol glycosyl chloride (1), 1 mmol nitrophenol, 1 mmol anhydrous K₂CO₃, 4 mL anhydrous CH₃CN, 0.1 mmol 18-crown-6. (I subsequently did it successfully on a 10-fold larger scale.) Save 0.5 g of your product from step 1 in case you need more characterization. Use the rest for this step.
3. Christina Donnelly (a former research student) tried an alternative approach that worked well; instead of using potassium carbonate and 18-crown-6 she used cesium carbonate. This worked very well for the 4-nitrophenol reaction, and I think it would work equally well with the 2-nitrophenol. She used 5.5 mmol of the glycosyl chloride (1), 5.5 mmol of the nitrophenol, 5.5 mmol of Cs₂CO₃, in 20 mL of acetonitrile. I recommend trying this.
4. I was careful to keep everything dry, including flame-drying the flask and running the reaction under nitrogen atmosphere.
5. I followed the reaction by TLC using 70% ether/30% ethyl acetate as eluant. It seemed to be done after 24 hrs.
6. My workup procedure involved adding chloroform and water to the reaction mixture (approximately 2.5 times the amount of acetonitrile used) then separating the chloroform layer and reextracting the remaining aqueous layer with half-again as much chloroform. The combined chloroform layers were washed 2 x 1 M KOH, and 1 x sat. NaCl, dried (Na₂SO₄), filtered, and evaporated to yield a yellow solid. Let's substitute dichloromethane for chloroform since it is less toxic.
7. Several solvents have been used successfully for the recrystallization including isopropanol/ether and ethyl acetate/hexanes. It is a bit too soluble in straight isopropanol for this to be an optimal recrystallization solvent. We might want to try methanol.

Notes, Observations, and Suggestions

1. (4/6/10) Potassium Carbonate/18-crown-6 Method. Derek and Morgan did the reaction with potassium carbonate/18-crown-6 over the weekend and the reaction mixture solidified. They worked it up and got about 7 g of a yellow solid. The crude NMR looked great! They recrystallized from about 150 mL of methanol to give white needles. They are letting it sit in the refrigerator overnight before collecting it. But this looks like it worked great!
2. (4/6/10) Rachel and Ali are going to try the cesium carbonate method. Brian and Christine are also trying the cesium carbonate method. (4/7/10) Both of these have solidified after stirring overnight at room temperature. Similar to what Derek and Morgan's looked like.
3. (4/15/10) It looks like recrystallization from methanol is working well. All groups, regardless of method, seem to be getting good yields (~60%) of product. The NMR's are looking spectacular. Shown below is the spectrum from Ali and Rachel
   
   

Step 3: Removal of Acetate Esters

Literature Reference

1. Adapted from: Kur'yanov, V.O.; Chupakhina, T.A.; Zemlyakov, A.E.; Chirva, V.Y.; Ischenko, V.V.; Khilya, V.P. Chemistry of Natural Compounds 2001, 37, 39.

Notes and Suggestions

1. The p-nitrophenyl glycoside was not completely soluble in methanol, so I ended up using a mixture of methanol/dichloromethane as the solvent. When I ran the reaction using 3.3 g of glycoside I ended up using a mixture of 100 mL methanol and 85 mL of dichloromethane. Start by adding an appropriate amount of methanol and then add dichloromethane, with gentle warming if necessary to get the glycoside to dissolve.
2. For every mole of of glycoside I used about 0.5 mol of sodium methoxide. The reaction was complete within an hour. Follow the reaction by TLC. A white precipitate of the product started forming in the reaction mixture after about 15 minutes.
3. After an hour I filtered the reaction mixture through a sintered glass funnel (Medium porosity) and washed with methanol. You might be able to use a Buchner funnel instead of the sintered glass funnel. Keep both the solid and the filtrate! The filtrate was treated with Dowex 50W-X8 (H+ form, approximately 1 gram/10 mg of the sodium methoxide) for 15 minutes. The Dowex was removed by gravity filtration and the solvent was removed in vacuo to give a white solid. This white solid was combined with the white solid that was collected above.
4. The product of this reaction doesn't seem to be very soluble in most organic solvents, although it seems to go into DMSO OK. We were also able to take an NMR by using deteuromethanol, a little CDCl₃, and then heating.

Notes, Observations, and Suggestions

1. (4/12/10) Derek and Morgan had about 4.5 g of the starting material. They were able to dissolve it in a combination of about 130 mL of methanol and 40 mL of dichloromethane. About 10 minutes after adding the methoxide (0.2 g) white solid started to form. After about an hour they did a TLC in 70% ether/hexanes and the starting material was gone and their was only a spot on the baseline. The Rf of the starting material was about 0.4. Next time try using pure ether as the eluant to get it to run a little higher. They collected the solid product by Buchner filtration and treated the filtrate with Dowex that had been washed with methanol. They had some white solid forming in the flask, so I told them to add a little more methanol/dichloromethane to try to solubilize it before filtering off the Dowex.
2. (4/14/10) Christine and Brian ran the reaction on a slightly larger scale (7 g?), but were able to get it all to dissolve in a similar amount of solvent as Derek and Morgan. They used ~250 mg of methoxide and over the course of an hour lots of white solid formed in the flask, essentially stopping the stir bar. They collected the product by filtration, and then neutralized the filtrate with Dowex, and removed the solvent by rotary evaporation to get about 0.4 g of solid. We don't have a yield yet on the solid that they filtered, but it looked like it was a fair amount.
3. (4/15/10) Christine took an NMR of their product in DMSO. (See below.) It looks good.
   
   
4. (4/15/10) We had three groups do this reaction tonight. They had between 4-7 g of starting material. I had them all dissolve the starting material in 100 mL of methanol and about 50 mL dichloromethane (with gentle heating if necessary). They then added about 0.2 g of sodium methoxide and stirred at room temperature. The solutions started getting cloudy within a few minutes, and after one hour the flask was full of white solid that usually stopped the stirrer. TLC's in pure there showed only baseline material. The solid was collected by Buchner filtration, and the filtrate was treated with 20 g of Dowex-50W-H+, filtered, and rotovapped to give a white solid. They then left them open to the air to dry.

Step 4: Formation of Acetonide

Notes and Suggestions

1. I ran this reaction with 2.2 g of the p-nitrophenyl glycoside, 50 mL of 1,4-dioxane, 8 mL of 2,2-dimethoxypropane and 75 mg of TsOH.
2. The reaction mixture was heated at 50 degrees for about an hour. There was always white solid present, it never all dissolved.
3. I turned the heat off after 1 hour and let it stir for two more. It looked like all the starting material was gone. I added 75 microliters of pyridine to neutralize the TsOH and then rotovapped. You should substitute imidazole for the pyridine.
4. Follow the reaction by TLC. The product should run higher than the starting material. I looked in my notebook and I used 10% ethanol/90% ethyl acetate as my TLC eluent. Under these conditions the product had an Rf of about 0.4. The starting material had an Rf of ~0.1. The nitrophenol probably runs near the top of the plate.
5. I purified the reaction by flash chromatography, but it would be more convenient if we could recrystallize. But given the limited amount of time left you might not have time to purify the product. But characterize it by NMR.
6. This reaction did not go very well last year, so I want you to do the reaction on one-fifth the scale I did (i.e., start with 0.44 g of the nitrophenyl glycoside). If the TLC doesn't look like much is happening after an hour, add more TsOH.

Notes, Observations, and Suggestions

1. (4/20/10) Derek and Morgan tried this reaction and after heating for 1 hour, there was still lots of undissolved solid, so they added a little more TsOH and let it stir overnight at RT. Still milky looking. I had them take a TLC at this point and we saw a fair amount of product. Robert and Susan also saw similar results.
2. (4/20/10) Christine and Brian did the reaction but used about 20 mL of dioxane instead of the 10 mL that the other groups used. After stirring at 50 degrees for about 30 minutes most of the solid had dissolved! TLC showed what looks like mostly product and a little starting material. They let the reaction go for a total of about 1.5 h and then worked it up by adding imidazole to neutralize the tosic acid. They then rotovapped to give an oil, took this up in dichloromethane (a very small amount of solid was present), washed with sat. bicarbonate followed by sat. NaCl, dried, filtered and rotovapped to give a yellow oil (approximately 0.7 g, >100% yield!).
3. (4/20/10) TLC of the reaction mixture in 10%EtOH/90%EtOAc showed the following:
   • Starting triol: Rf=0.3 (often streaking, probably due to limited solubility)
   • Product: Rf=0.6
   • Transient intermediate (faint): Rf=0.4
4. (4/20/10) Brian and Christine's NMR of their product after an aqueous workup (see below) looks great. It still contains some dioxane, and there is a little aromatic impurity, but otherwise it looks good.