Chemical Forums
Chemistry Forums for Students => Organic Chemistry Forum => Organic Chemistry Forum for Graduate Students and Professionals => Topic started by: Babcock_Hall on November 23, 2021, 05:15:06 PM
-
DOI: 10.1002/anie.200802036 Zhu and Schmidt Angewandte Chemie 2009
I am planning a synthesis of an O-glycoside bearing other functional groups. I envision making an O-glycosidic bond, then oxidizing a sulfide to a sulfone with Oxone or MCPBA, then performing a Horner reaction, then deprotecting. In my reading about protection of glycosides, I was surprised to learn that some, but not all, protecting groups affect the reactivity of the activated glycoside. The review article above refers to the protecting groups that alter reactivity as "arming" or disarming." How important is this issue in choosing the best protection/deprotection strategy? Does anyone have another good review article on the synthesis of O-glycosides at their fingertips?
EDT
When I did a literature search, I obtained both good and bad news. The good news was that oxidation of a sulfide to a sulfone by MCPBA in the presence of a benzoyl-protected glycoside is well precedented. I also saw one example of acetyl protection. The bad news is that when there is a two-carbon spacer between the sulfone and the oxygen, the sulfone and the two carbons are lost under basic conditions. LDA in THF will bring about this reaction, and so will sodium methoxide in methanol at 0 °C. This turns the sulfur-containing portion of the molecule from a base-stable protecting group into a base-labile one. The details of what happened were not given in this 1993 JCS Chem Comm paper (pp. 825-826): https://pubs.rsc.org/en/content/articlepdf/1993/c3/c39930000825. What is not clear is what would happen with a longer number of carbons, other than the cost of the synthesis going up.
-
Is it not E1cb? The alphaprotons on the sulfone are acidic.
-
I was originally thinking about something akin to a Conant-Swan fragmentation, but now that you say E1cb, I think that you are probably right. Our typical Horner conditions are 2 hours at room temperature, and one could try a weaker nitrogen base than DBU, like DIPEA. But there is also the deprotection of the glycoside itself to consider.
If the spacing between O and S were 3-carbons, then I don't see how an elimination would happen.
-
Or use a Wittig reagent instead, they are much less basic.
-
On the sulfur of the sulfone is a benzyl group. It was pointed out to me that the S-CH2-Ph hydrogen is more acidic than the hydrogen on the other side of the sulfone. The person who pointed it out to me was wondering about the actual mechanism of fragmentation. It may be not worth worrying about, but if I could understand better what was going on, I feel as though I have a better chance at preventing it.
I just now found a review article on protecting groups in oligosaccharide synthesis: https://doi.org/10.1002/asia.201901621. The authors are Ghosh and Kulkarni, and the review covers 2009-2019.
-
I do not understand why you are not using thioaryl glycosides to conduct the chemistry which is more precedented. The Kahne-Crich glycosylation is the most well known to work with stubborn alpha/beta ratios which was a significant problem with mannose.
However, I would try radical conditions. This pretty much sounds like the julia olefination https://www.organic-chemistry.org/namedreactions/julia-olefination.shtm
I remember danishefsky used a dissolving sodium reduction to deprotect benzyl groups because palladium chemistry failed to remove all ~20 of them in the final step of a large oligosaccharide; i.e. sodium metal in ammonia at -78. 10.1021/ja307628w and related references
-
I may not have been clear, but the sulfone will be well away from the glycosidic portion of the molecule. I am only just now looking at the various methods of making glycosidic bonds and protecting or deprotecting the alcoholic oxygen atoms. I will look into the Kahne-Crich method. Presently I was thinking in terms of making the glycosidic bond well in advance of making the vinyl sulfone.
-
I think the problem with E1cb will persist even if you have the glycosidic bond early in the synthesis?
-
At this point I am planning on a 3-methylene spacer between the oxygen of the glycoside and the sulfur of the sulfone. The third carbon adds considerably to the cost of the starting material, but it is worth it if it does not eliminate. I have yet to pick out a preferred method of making the glycoside, but I found the Handbook of Chemical Glycosylation edited by Demchenko, which might help.
https://onlinelibrary.wiley.com/doi/pdf/10.1002/9783527621644
-
OK. Is it important to control the stereochemistry in the glycosylation? I think if you use acetyl as PG you will get neighbouring group partiticipation and this will give you mainly one stereoisomere this may also be true uf you use benzoyl. Its not a difficult reaction, you can use a protected bromo-sugar and mild base and mix with your alcohol, its similar to protect a alcohol with a MEM-group. If you need you can make the bromo-sugar from the corresponding -OMe glycoside and TMSBr.
-
It depends on the specificity of the putative enzyme that might hydrolyze. I was thinking that I wanted a beta-linkage, but I need to continue my reading in this area.
-
Maybe it would be good to make both alpha and beta? They can have very different biological properties. There are good ways to controll the stereochemistry in the glycosylation. I think this step will be trivial, its standard chemistry.
-
I think this step will be trivial, its standard chemistry.
Always easy to say from behind a keyboard when someone else is going to have to run the reactions! :)
-
I have been so focused on ironing out actual or imagined problems in the early steps, that I have only just started to answer the question of how to perform the glycosylation reaction. Based on my preliminary assessment, I will see if I can obtain both alpha and beta linkages synthetically.
-
I think this step will be trivial, its standard chemistry.
Always easy to say from behind a keyboard when someone else is going to have to run the reactions! :)
Very true, I just compare with the other steps and glycosylation is more standard chemistry. But a lot can happen when you have multiple functional groups so…
-
Based on conversations with various authors and the literature searching that I did, I think that the 3-carbon spacer between the glycosidic oxygen and the sulfonyl sulfur will not have the elimination issue seen with the 2-carbon spacer. I also think that the presence of the glycoside will protect the oxygen against oxidation of the sulfide to the sulfone. I need to research the deprotection, to make sure that the functional groups will survive it.
-
https://doi.org/10.1016/j.tet.2005.01.015
Scheme 3 in this paper indicates that the 3-carbon spacer between the oxygen on C-1 of the glycoside and the sulfur of the sulfone is inert to base. The 2-carbon spacer gives elimination, leading to loss of the glucose portion of the molecule. In this paper there was a phenyl group on sulfur; therefore, there was only one acidic proton alpha to the sulfur. The authors did not cite the paper that I did earlier in this thread (T-H Chan and C-H Fei, 1993 J Chem Soc Chem Comm, 825), suggesting that they were unaware of it.
-
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4331121/
The β-trimethylsilylethoxymethyl (SEM) group, which can be added with 2-(Trimethylsilyl)ethoxymethyl chloride, came up in a different thread on glycosides. "The trimethylsilylethoxymethyl group easily survives under bromination, basic hydrolysis, oxidation and other harsh conditions.[3]... In general, however, SEM deprotection from alcohols seems facile[9–14]..." This quote comes from the link above.
If I were to deprotect this group in the presence of a vinyl sulfone, does anyone foresee a problem in deprotection with, for example, magnesium bromide? I did a quick SciFinder search to see if anyone had made a tetra-SEM glycoside, but I have not found one yet. I did find two examples where SEM protected the O-1 oxygen and one example where the oxygen atoms of a disaccharide carried various orthogonal PGs, including one SEM.
-
Yes, for E1cb mechanism needs a labile group in the 2-position, if it is in the 3-position you get no elimination.
Magnesium bromide sounds mild, but why not use some fluoride-reagent? Neutral or mild basic conditions sounds good if you have a glycoside, magnesium bromide is a weak Lewis acid, no?
-
Fluoride sounds great, because we have previously used it in the presence of a vinyl sulfone without difficulty. However, I am concerned that I have not found any examples of a tetra-SEM glycoside in the literature so far. Possibly there is some difficulty of which I am unaware.
-
In one of my proposed routes a primary alcohol bearing a sulfide linkage forms a glycosidic bond to an electrophilic, protected derivative of glucose. I wonder whether or not there will be a competition between the oxygen and the sulfur. The sulfur would bear a positive charge if it attacked. One way around this putative problem would be to oxidize the sulfur.
-
No, there is actually a way of doing glycosylation where you have a methylthio glycoside and uses methyl triflate to activate this by forming a charged thio glycoside, the charged dimethyl sulfide act as a good leaving group. I dont think a sulfide will compete with the alcohol. The activated glycoside is a very hard electrophile and is attacked by the hard alcohol nucleophile, the sulphide is a very soft nucleophile and does not go so well together with the activated glycoside.
-
Fluoride sounds great, because we have previously used it in the presence of a vinyl sulfone without difficulty. However, I am concerned that I have not found any examples of a tetra-SEM glycoside in the literature so far. Possibly there is some difficulty of which I am unaware.
https://doi.org/10.1002/chem.201101163
I have now found one paper in which four SEM groups were used to protect the oxygen atoms at carbons 2-4 and 6 of a glycoside conjugated with a steroid. I did not see a deprotection step in their supplemental, but I would assume that one would use a standard recipe. Given that I have some concerns about the deprotection step if we used acetyl protection, protection with SEM looks potentially useful.
-
Great!
-
Our first glycosylation attempts will use the Schmidt glycosyl donor, a trichloroacetimidate. The choice of base strongly influences the configuration of the trichloroacetimidate. One thing that I am not yet sure about is what factors determine the configuration of the final product, the glycoside. I have seen some schemes in which retention occurs and others in which inversion occurs, and my tentative view is that it depends on whether or not there is neighboring group participation, but it might also depend on the acidity of the catalyst. As I learn more, I will post it here.
EDT
In the Handbook I mentioned in a previous comment, Schmidt wrote, "The anomeric stereochemistry is derived from the anomeric configuration of glycosyl trichloroacetimidates (inversion or retention), anchimeric assistance, the influence of solvents or thermodynamic or kinetic effects....In addition, the nature of the counteranion in catalysts is very influential in controlling the stereoselectivity of the Schmidt glycosidation, as reflected by comparing entries 1 and 2, or 3 and 4 in Table 3.1 [394], but how the anions work has not yet been understood." Sounds like I will be burning the midnight oil.
-
I think the NGP-effect is strong so if you have acetyl-PG you get mostly beta.
-
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6259426/
doi: 10.3390/molecules15107235
Hi Rolnor,
This is a 2010 review that is more focused on carbohydrate protecting groups. After reading portions of it and looking at Scheme 1, I think I understand your comment a bit more clearly. Thanks.
-
Good!