As the regular reader of F-Blog knows, fluorous tags can be used for purification, enrichment, and immobilization. Of those three areas fluorous immobilization is the newest and probably the least established at this point. While there has been plenty of work done in the area there are still some knowledge gaps, such as optimal fluorous tag length, linker length and composition, and spotting conditions. A new communication from Prof. Nicola Pohl’s group attempts to fill some of these gaps.
Prof. Pohl was the first to use fluorous glass slides for microarray formation back in 2005. Since that time she has published many papers on the synthesis, purification, and immobilization of oligosaccharides to these surfaces. Her group has primarily used C8F17 tags in conjunction with fluorous solid phase extraction (FSPE) purification and the same tag for immobilization. This latest communication, however, examines the use of bis C6F13 tags instead for purification and immobilization and also provides some valuable tips and insight into fluorous microarray formation.
The synthesis of the bis C6F13 tag is shown above and is relatively straightforward. (Compound 4 was obtained through diallylation of diethyl malonate followed by radical addition of perfluorohexyl iodide to the olefins.) The researchers report that the bis C6F13 tag can be used for FSPE without any modification of the FSPE protocol for the mono-C8F17 tag. Although not mentioned in this report, it is known that the bis-C6F13 tag does have higher fluorous partitioning, and therefore is better retained on fluorous solid phases, than the C8F17 counterpart. How much so is hard to say, but obviously not so much that 100% MeOH cannot be used as the fluorous wash during FSPE.
The authors then tagged three different sugars; glucose, rhambose, and mannose, with a mono-C6F13 tag, a mono-C8F17 tag, and the bis-C6F13 tag to produce nine different fluorous tagged sugars. These were then dissolved in Methanol/DMSO/water (1:3:1) and spotted in equal concentrations to a fluorous slide. The spots were dried, incubated with a solution of FITC-ConA for one hour, washed twice with a 1% BSA in PBS solution and then once with distilled water. The spots were then scanned at 488nm for carbohydrate-ConA binding. ConA is a lectin that is known to selectively bind mannose. The authors found that the morphology of the spots was dependent on temperature, humidity, and drying time. In their hands the best and most uniform spots were formed at 22 degrees C and 70% humidity for 18h followed by 2h at ambient humidity. The results are shown below.
The authors found that the mono-C6F13 tagged mannose provided no signal. The mono-C8F17 tagged mannose-ConA complex was observed, but the signal was clearly less than the bis-C6F13 tagged mannose. The authors also found that the signal intensity of the bis-C6F13 tagged spot did not diminish even after continued washing. They also note that the mono-C8F17 tagged complex only had a slight diminution of signal upon washing. That’s interesting, since the spot clearly looks much less intense. That might indicate then that while some washing off took place with the mono-C8F17 tag that the major reason for the lower signal is less binding due to the difference in the linker. Counting out from the last CF2 atom the mono-C8F17 is nine atoms removed from the anomeric carbon of the sugar. The bis-C6F13 tag on the other hand is 11 atoms removed. This doesn’t even include the presence of the aryl group. Clearly, some further investigation would be desirable here to elucidate what’s important and what isn’t in terms of effect on binding.
One of the stated goals of the research presented here was to find an alternate tag to the C8F17 group for fluorous immobilization. As the authors note the environmental persistence of perfluorooctyl groups is problematic. Shorter fluorous tags are certainly preferred since they are not as persistent and by using multiple shorter fluorous tags the authors have found an alternative. An alternative that provided better signal to boot.