Imaging biomolecules with AFM

Yes, these atomic force microscopes give molecular resolution data! With noise floors generally below 100 pm and below 35 pm in the case of the HR-AFM, the performance of the AFM isn’t an issue. The challenges typically come from sample preparation, microscope operation, and interpretation of the data.

The basics of imaging biomolecules

The beauty of AFM is that you see everything down to the nanoscale, the problem with AFM is that you see everything down to the nanoscale! You think that’s funny until you try imaging a DNA fragment that was amplified by PCR that someone hands you with the words “It’s clean and very concentrated!”. You’re not going to see the DNA between the mountains of contaminants. Work VERY clean, purify your material (columns work well), use distilled water for your reagents or at least for your final washes, and use clean, flat substrates to deposit the material on (mica is the default, keep in mind it is negatively charged). Following these steps should eliminate the mountains!

AFM is a mechanical surface characterization technique, and while the forces the tip exerts on the sample are very small, they are still enough to move things around. Getting your biomolecule to stick to the substrate is essential, otherwise you’ll have lots of streaking and your tip will keep picking things up. Imaging dry is usually the first step as it addresses many of these issues and gets you a quick idea if your material is what you think it is. From there work your way up to imaging in liquids and/or force measurements. A good test is to set up the liquid scanning with a dried sample, then add water and see how it goes!

When scanning nanoscale features it is key to reduce the noise on your AFM as much as possible. Pick a good environment, measure the noise floor, calibrate your system on a test sample, and make sure to adjust your scan parameters to what your sample requires. Finally, the interpretation of nanoscale scans can be confusing to new users of AFM. Keep in mind that standard AFM probes have a tip radius of just below 10 nm or more if you are working with a coated tip in liquid. This causes the resulting image to be a convolution of sample shape + tip shape in the XY plane, but the height measurement is accurate, or mostly so. Odd things happen at the nanoscale, dsDNA for example measures less than 2 nm in height.

Imaging DNA

DNA is the most common biomolecule sample in AFM imaging for multiple reasons. It is at the root of many biological questions, relatively easy to obtain in high concentrations, difficult to image by transmission electron microscopy (TEM), and suitable for dynamic imaging by AFM in liquid. There is excellent literature on preparing and imaging DNA, including advanced methods with choices of substrates, probes and modes.

Getting started with DNA imaging mainly requires a high-quality starting material (highly purified), clean handling (DNAses are everywhere!), clean water (distilled water from the store works), and mica. The most straightforward method is the use of divalent cations (magnesium or nickel) in the solution to bind the DNA to the negatively charged mica, followed by washing and drying.

Click for more DNA images and preparation advice!

Other biomolecules

Biomolecules vary dramatically in size and stability and range from large macromolecules to individual proteins. Sample preparation plays a key role in obtaining reliable data and needs to be taken into consideration when interpreting the data. Again, the first test should be done on a dried sample, with the caveat that structures decay to varying degrees during fixation, washing and drying. To demonstrate this we have analyzed some readily available yet surprisingly challenging biomaterials.

Collagen is an example of a large, polymeric and highly relevant biomolecule that is very accessible by AFM. We have found it serendipitously in the eggshell membrane, a great biomaterial for educational outreach programs. With closer inspection we found that the entire surface was still covered by egg white proteins which are readily visualized because of their homogenous appearance (ovalbumin makes up 54% of egg white proteins).

Casein micelles in milk are another readily available though much more fragile and complex macromolecular complex for which AFM imaging is a perfect method. We tested a sample of 2% cow’s milk by filtering the solutes and imaging the filter cake. In this case we expect the drying process to significantly disrupt the structure of the emulsified particles but were able to image what appear to be casein micelles.

Collagen I network found in an eggshell membrane facing the egg white.

Egg white proteins on the eggshell membrane, mostly ovalbumin.

Solutes in 2% cow’s milk enriched by filtration showing casein micelles and fragments.

  • Fotiadis, D.; Scheuring, S.; Müller, S. A.; Engel, A.; Müller, D. J. Imaging and Manipulation of Biological Structures with the AFM. Micron 2002, 33 (4), 385–397.

  • Raspanti, M.; Reguzzoni, M.; Rita Basso, P.; Protasoni, M.; Martini, D. Visualizing the Supramolecular Assembly of Collagen. In The Extracellular Matrix; Vigetti, D., Theocharis, A. D., Eds.; Methods in Molecular Biology; Springer New York: New York, NY, 2019; Vol. 1952, pp 33–44.

  • Kovacs-Nolan, J.; Phillips, M.; Mine, Y. Advances in the Value of Eggs and Egg Components for Human Health. J. Agric. Food Chem. 2005, 53 (22), 8421–8431.

  • Ouanezar, M.; Guyomarc’h, F.; Bouchoux, A. AFM Imaging of Milk Casein Micelles: Evidence for Structural Rearrangement upon Acidification. Langmuir 2012, 28 (11), 4915–4919.