A just published ASAP paper in Inorganic Chemistry from the Dyson group describes their efforts in preparing potential anti-cancer agents which selectively target tumor cells. The selection criteria that they are attempting to exploit is temperature dependent solubility differences. Cancer cells are known to run hotter than normal cells, while cryogenic and thermotherapy in conjunction with various anti-cancer drugs is known to provide a synergistic effect. According to the paper, however, no chemotherapy agents have been designed with thermoresponsiveness specifically in mind.
The authors chose to make ruthenium complexes containing fluorous tagged phosphines as potential thermomorphic chemotherapy agents. They chose ruthenium complexes, since Ru has been shown to have excellent activity in tumors. Two ruthenium complexes, NAMI-1 and KP1019, are in early clinical trials.
Fluorous ligands were selected due to their well-known steep temperature solubility curves. This behavior has been exploited by various research groups for the design and synthesis of thermomorphic catalysts. Thermomorphic catalysts are completely soluble at elevated reaction temperatures, resulting in homogeneous catalysis. Once the reaction is complete, the mixture is cooled resulting in precipitation of the catalyst which can then be removed as a heterogeneous catalyst.
By making fluorous modified ruthenium compounds the authors hoped that they could produce cytotoxic compounds that are specifically designed to be thermoresponsive. The solubility and uptake of these agents would be temperature dependent. Application of heat to the tumor could then result in more selective therapies.
The researchers produced the compounds shown above and examined their solubility at 37º C and at 42º C. What they found was that compounds 1 and 5 showed the most promising behavior with a 4-fold increase in aqueous solubility at 42º C than at 37º C. When tested in vitro against two cell lines, one cisplatin resistant, a modest temperature dependence in IC50 was observed. The authors attribute this to the increase solubility at higher temperatures resulting in greater cellular uptake, although they caution that other phenomena could be at work.
All in all, an interesting paper which highlights a novel application of an unique physical property of fluorous domains, in this case thermomorphism. While these are probably not viable drug candidates by any means, it does show that the idea of a fluorous thermomorphic therapeutic is feasible.
Earlier this week FTI launched our first newsletter of 2010. The main theme was looking forward and trying to identify applications of fluorous techniques that set to blossom in 2010. We identified three in particular; radiopharmaceuticals, proteomics, and custom carbohydrates. If you would like to read more, then please click here.
The big change for this newsletter though is how its delivered. Previously, we had e-mailed the Newsletter to addresses that we had collected over the years. People move on and standards change, however, so in order to adhere to current best email practices we have switched to a free subscription format. We e-mailed everyone on our list a subscription verification requesting that they subscribe. If you regularly received our Newsletter, but did not see one in your inbox this week, it’s because your subscription wasn’t verified. We will be sending out one last reminder, so keep an eye out for it.
If you have not received the Newsletter in the past and would like to in the future, please click here and fill in the box on the left. You can also view some of our past Newsletters and decide if you would like to subscribe.
The FTI Technical Newsletter, along with F-Blog, is the best way to keep current with everything fluorous.
The synthesis of oligosaccharides is not easy when compared to to other oligomeric biomolecules such as oligonucleotides and peptides. There’s a number of reasons for this, but primarily it comes down to one thing: complexity. For example, DNA is made from four different monomers which are connected by phosphates through the 5′ and 3′ hydroxyls. Peptides are formed from 20 different monomers connected by amide bonds between the C and N terminus. Oligosaccharides, on the other hand, are produced from dozens of different monomers connected by acetal bonds forming a stereocenter which needs to be controlled. On top of that, oligosaccharides need not be linearly attached like peptides and oligonucleotides, but are often branched. With all of this complexity, it’s perhaps not surprising that development of methods for their synthesis and purification have remained an area of intense research for a long time.
Count fluorous methods among those that have been tried and for the most part, very successfully. For a brief summary of some of the fluorous strategies that have been employed, click here.
The Huang group now adds another method; fluorous catch and release. Their strategy is to use a one-pot synthesis to form an oligosaccharide with a terminal reactive functional group linker. A fluorous reagent which reacts with the linker is added to fluorous functionalize the desired oligosaccharide, which is then separated from all the non-fluorous junk using FSPE. The linker is cleaved to release the oligosaccharide from the fluorous tag and a second FSPE used.
The implementation of this strategy requires a linker and fluorous reagent combination that is compatible with oligosaccharide synthesis, selective, react relatively quickly, and can be cleaved from the carbohydrate. The authors first tried to use an azide as the linker and a fluorous phosphine as the reagent, but found that azaylide was not stable to FSPE. They then tried an aldehyde-hydrazine combination, but discovered that the aldehyde was not stable enough. Substituting the aldehyde with a ketone and modifying the hydrazine, however, led to a suitable linker-reagent pair. Once they had found the right combination, they then demonstrated the usefulness of their methodology by preparing several oligosaccharides, including LewisX, a branched trisaccharide which was prepared in 62% yield without silica gel chromatography.
Given all the ways fluorous techniques have been employed in oligosaccharide chemistry, it’ll be interesting to see how the fluorous catch and release strategy is viewed and used within the carbohydrate synthesis community.
A joint 2010 J. Org. Chem. ASAP article from groups in Berlin and India describes the three-component synthesis of pyridines with perfluoroalkyl and perfluoroaryl substitution. In actuality the paper is about 95% trifluoromethyl substituted, but they do include an example of a perfluoroheptyl and a pentafluorophenyl substituted pyridine. The paper is interesting to me for it’s lack of the word “fluorous” or use of fluorous purification techniques.
The chemistry as shown above combines a lithiated alkoxyallene, a nitrile, and an acid to form substituted pyridines. The initial heterocycle forming reaction requires an excess of the allene relative to the nitrile and provides the products in 25-50% yield. The resultant 4-hydroxy group can than be converted to a perfluorobutyl(c4F9) sulfonate, commonly known as a nonaflate analogous to triflate. Like other sp2 sulfonates, the nonaflate is a halogen surrogate which can undergo a number of transformations including Suzuki couplings, Buchwald-Hartwig amidations, Stille couplings, etc. The authors used this strategy to prepare a number of different substituted pyridines.
Given the modest yields and the use of excess allene it’s surprising the authors didn’t try to use a fluorous solid phase extraction based purification rather than the flash chromatography employed. Using perfluorooctylsulfonyl instead of the nonaflate would provide an nice fluorous tag for the FSPE and the subsequent chemistry. The authors were certainly aware of the chemistry considering that two FTI papers using fluorous sulfonates were referenced. If they did try it and it failed, this was not mentioned in the text.
Perhaps, since they used fairly simple allenes and nitriles the separation by flash chromatography was not particularly difficult. If a library were to be made using this chemistry, however, that might not be the case. For example, they already observed that in the case of nitriles with an alpha protein, the major product was not a 4-hydroxy pyridine, but a 4-alkyl pyridine. Upon fluorous sulfonyl tagging, the desired product would be readily separable from the non-fluorous pyridine by FSPE.
About one year ago, a report from the Gouverneur lab in Oxford reported the use of fluorous sulfonates as precursors to 18F imaging agents. Nucleophilic displacement of the fluorous sulfonate by 18F anion provided the desired compounds as shown below. In these types of reactions, the 18F is obviously far more valuable than the precursor, so in general a large excess of the alkylating agent is used. The removal of the excess alkylating agent and their by-products using a simple, fast, and automatable method is desirable to provide the purest imaging agent possible. Hhhhmmm. Simple and fast. Perfect for fluorous methods, which is what they found using a FSPE purification to remove the excess fluorous sulfonate. Their report earned them an Angewandte Chemie cover.
Now comes a report from Prof. Bengt Langstrom’s group at Uppsala where they essentially use a similar strategy for 18F substitution. They report yields, however, that are not quite as good as Prof. Gouverneur’s using the same fluorous sulfonate. They claim better results using a pentafluorophenyl sulfonate rather than a perfluoroalkyl sulfonate. They then utilized pentafluorophenyl modified silica gel for the solid phase extraction. In the test reaction below they report a 77% radiochemical purity after SPE using the pentafluorophenyl sulfonate as opposed to a 50% yield using the perfluoroalkyl sulfonate pictured. For the SPE using the pentafluorophenyl sulfonate the authors used a pentafluorophenyl modified silica gel as the soilid phase.
Langstrom and co-workers report that with the pentafluorophenyl sulfonate they do not observe loss of radioactivity which was observed with the perfluoroalkyl sulfonate. This loss with the perfluoroalkyl sulfonate they ascribe to fluorine scrambling with the 19F atoms of the perfluoroalkyl chain, which the Gouverneur group also observed as a loss of specific activity.
The work provides an potential alternative approach to using a fluorinated sulfonate as a leaving group for 18F substitution. It does, however, remain to be seen how this method compares to the perfluoroalkyl approach in the preparation of actual agents such as FDG, FMISO, and others. The pentafluorophenyl based SPE doesn’t seem as discriminating as the traditional FSPE. The authors collected several fluorophobic fractions with the third being the purest. I would anticipate that the separation could get even more difficult for molecules such as 18F-L-thymidine.
With that caveat, however, the work is another demonstration that a fluorinated or fluorous sulfonate in conjunction with SPE may be an attractive alternative to current methods in nucleophilic 18F substitutions. There are still some issues to be sorted out, but it is an area where fluorous could make a big impact.
Langstrom and co-workers report that with the pentafluorophenyl sulfonate they do not observe loss of radioactivity which was observed with the perfluoroalkyl sulfonate. This loss with the perfluoroalkyl sulfonate they ascribe to fluorine scrambling with the 19F atoms of the perfluoroalkyl chain, which the Gouverneur group also observed as a loss of specific activity.