Bogdan Budnik, PhD, director of proteomics at the Harvard University Mass Spectrometry and Proteomics Resource Laboratory, talks to contributing editor Tanuja Koppal, PhD, about the types of projects they are involved with using mass spectrometry (MS). He shares some useful insights when it comes to choosing instruments and software tools for MS analysis and how his team works with their users/clients to provide the best possible information to address their biological questions.
Q: Can you tell me a few things about your center and what you do?
A: We are a small resource lab in the Center for Systems Biology at Harvard University. Our main users come from this center, but we also have scientists from the departments of molecular and cell biology, chemistry, and chemical biology, who come to us to process and analyze their samples. Sometimes people from the Harvard School of Engineering also bring us their samples for analysis. So we serve a large, diverse pool of scientists who want us to do sample preparation and analysis for them, and this brings diversity into the work that we do. Our main interest is in proteomics, and we analyze a lot of different proteins using mass spectrometry (MS). There is another mass spectrometry group here that focuses on analysis of small molecules, such as drug metabolites. We also have a cell-sorting and a RNA-sequencing facility located here, so scientists can make use of all these resources for their work.
Q: What are the main types of projects that you are involved with?
A: Our group is involved with proteomics, and we cover a broad range of projects. Most of the samples that we process involve some sort of pull-down proteomics, where we isolate materials from biological samples and then try to identify what’s in there. Another big application is the analysis of posttranslational modifications (PTM) of proteins. People like to know about phosphorylation, acetylation, methylation, ubiquitination, or other types of modifications that are present in proteins once they are translated. The largest group of samples that we process is for quantitative proteomics. This is when people are looking at different states of the same sample or looking at the changes to the proteome in the sample after different treatments. Users bring us several samples, and we analyze the proteins using a technique called tandem mass tag (TMT). We analyze about six to 10 samples in one single run using MS, after they have all been digested and labeled at the peptide level. We can then see differences in the protein abundance in the samples, and how it changes with time or following different treatments and environmental conditions.
Q: What types of mass spectrometers do you use, and why?
A: For proteomics, we have three main MS instruments, and they are all different types of orbitrap analyzers; and for our purpose, they work the best. We have samples that range from millions of cells to a couple thousand cells, and we need instruments with different dynamic ranges and sensitivity to analyze them. Which instrument you choose largely depends on your own personal experiences. Once you get familiar with a certain instrument, the software tools, and how to do the sequence searches, which is very important for proteomics work, that usually drives the instrument you pick. You also have to look carefully at the types of samples you routinely work with and figure out what is the best instrument for analyzing those samples. You can determine this based on recent publications, or I suggest you prepare a sample [that] is a representative of the majority of the samples that you are going to work with, and ask different vendor companies to analyze it for you in their demo labs. Then you can clearly see and compare the results and decide on the quality of the instrument, depth of coverage, sensitivity, or other factors that may be important for your future work.
There are many different choices in terms of the software that you can use for data handling and for performing protein sequence searches. I was one of the beta testers for the Proteome Discoverer software by Thermo Fisher Scientific, so I know it well. There are several open source software [programs] as well, but I prefer the one designed by people at the Max Planck Institute, called MaxQuant. These two tools are [among] the best for normal proteome analysis. We have used other software, such as Byonic, which has worked well for peptides with PTM.
Q: Have the applications for MS changed in recent years?
A: Clearly, proteomics is becoming more quantitative. People don’t just want to know what is in the sample but [also] how much is there. People are doing RNA sequencing in a single cell and doing hundreds of them in parallel, and proteomics is heading in the same direction. It’s not there yet in terms of sensitivity, but in a few years, it will be. Previously, when we did proteome analysis we would need millions of cells. Later that number dropped to hundreds of cells, and today we can get away with just 10 to 50 cells for analysis. We are now working toward single-cell proteomics. The data that we can get at the protein and RNA levels from a single cell will be very important to biologists in terms of finding out whether differences in translation exist, the types of differences that exist, and whether those differences are important or not. MS can certainly play a big role in that type of analysis.
Q: What recommendations do you have for your users and our readers looking to analyze samples using MS?
A: We work with several different types of samples that are sent to us from labs across the world, and that brings diversity to our work. However, we do not accept samples from anyone without having a conversation with them first. We learned a long time ago that people don’t necessarily know how to prepare a biological sample for MS analysis. So, we decided to do everything for them from start to finish. The first and the main questions that we ask the client are what is the biology you are looking for, and what kind of answers do you expect to get from the MS data? Based on their answers, we go ahead and decide how the samples need to be prepared, how the protein is to be digested, and what needs to be done later. If cells need to be analyzed, we ask the user to just bring us the cells washed in phosphate-buffered saline (PBS). We then take it from there all the way to the MS analysis so we can be confident of the results that we get. Proteomic analysis is not as straightforward as it seems, and probably that is why we have core labs with certain expertise set up in academic centers and within companies.
The instruments that we work with these days are also very complex. So, I personally would not recommend saving on the maintenance contracts and trying to fix things yourself. Even though we know the instruments well and have the necessary engineering help, the instrument time is worth more than the cost of having a contract in place to get instruments fixed by the vendors. For our business model, having a fast response time to get the instruments fixed is very important.
Bogdan Budnik, PhD, is the director of proteomics at the Harvard University Mass Spectrometry and Proteomics Resource Laboratory. He obtained an undergraduate degree in general physics and a master’s degree in nuclear and plasma physics from Kharkov State University in Ukraine. He has worked at the Harvard University Mass Spectrometry and Proteomics Resource Laboratory for 10 years, first as scientific manager, and he has been the director of proteomics since 2014. The lab serves a wide range of biological projects, primarily from the local Harvard University community, as well as from other academic groups and commercial organizations.