A significant part of osteoarthritis research is focused on deepening the understanding of the disease at the molecular level. There is currently no cure for osteoarthritis but through an increased understanding of the biological processes, new potential drugs can be developed. A research article was recently published where we focused on this - investigating osteoarthritis at the molecular level.
In our study, we analyzed synovial fluid from human knees. Synovial fluid is a viscous fluid that is enclosed in the joint and its function is to lubricate, allowing the surfaces to move more smoothly against each other as well as provide nutrients to the cartilage. In osteoarthritis, tissues such as cartilage and meniscus are degraded and components that were previously in these tissues can end up in the synovial fluid.
As remnants of diseased tissue can be present in the synovial fluid, it is of interest to compare synovial fluid samples from healthy knees with samples from knees with osteoarthritis, which is what we have done. More specifically, we have analyzed the proteins present in the fluid using a method that relies on synthetic antibodies which recognize and bind to specific proteins. Through these synthetic antibodies, one can measure how much there is of a certain protein in a sample. This method generates a lot of data, making it suitable for exploratory studies. In our case, an exploratory study means that our data is used to generate hypotheses, rather than us having a clear hypothesis formulated in advance.
In our samples, we found about 6500 proteins. We started by looking at which proteins differed the most between the synovial fluid from healthy and diseased knees. The proteins that differed the most in percentage between the two groups were selected to be included in a network analysis. Before we explain how we used this analysis on our data, we will give a simpler example.
Let’s represent students who take any subjects together in a network. In our example, we will analyze a total of six students. Omar, Marta, Maja, and Elias are all in class 9A. Hugo and Nadia are in class 9B. We let each student’s name be represented in the network and if two students take a subject together, we connect their names with an edge. We start by connecting students who are in the same class as we know they have many common teaching hours during a week.
As most people know, ninth-grade students in Sweden also have an optional extra language. In our case, Omar and Maja are studying German. Three students are studying French: Marta, Elias, and Hugo. Of the six students we have selected, only Nadia is studying Spanish. Let’s now connect the names based on whether the students are taking a language class together. We have now created a network that describes which students are taking any subjects together during a school week.
Now, we go back to our data: we analyzed synovial fluid from healthy knees and knees with osteoarthritis. Of the 6500 proteins we found, we chose to look further at the proteins that differed the most in terms of percentages between the two groups. Specifically, we looked at whether there was any pattern in our data for how the proteins varied together. If our data showed that one protein depended on another, an edge was created between them.
Then, we selected proteins that we identified as particularly interesting. To describe which proteins we selected, let’s go back to our example above. Among the students, we see that Hugo plays a central role in connecting classes 9A and 9B in our network. We looked at the proteins in our data in the same way, as we believe that important proteins are proteins that connect to other parts of the network. We hope that the proteins we identified can contribute to an increased understanding of osteoarthritis. For the curious reader, the published study can be read here. Molecular and cellular proteomics will open in a new tab.
