What are PAINS?

Author: Mackenzie Fuller

Learning Objectives

  1. Identify what PAINS are.
  2. Discuss the mechanisms that PAINS use to produce their effect.
  3. Evaluate why it is important to identify PAINS in a screen and how to do so.

 

Graphical Abstract

What are Pan-assay Interference Compounds (PAINS)?
What are Pan-assay Interference Compounds (PAINS)? An overview of what PAINS are and why it is necessary to identify them in a screen.

Legend. (4,5,6) PAINS are compounds that produce a false positive in a high–throughput screen. For a true hit to be produced, the compound has to fit perfectly into the binding site on the protein and then the desired effect is produced. PAINS on the other hand, can produce this effect without fitting perfectly into the binding site. There are many classes of compounds that produce PAINS and it is necessary to identify the compounds in your screen and eliminate them. PAINS cannot be developed into a drug that is successful at treating the disease, but this may not be discovered until many years and a lot of money later. (Picture 1 is a protein binding site, Picture 2 is a robot used for high-throughput screens, Picture 3 is toxoflavin, which is a known PAIN) Image created using Canva Pro.

 

Summary

When researchers are looking for new drug treatments for diseases, they first identify and isolate a protein that contributes to the disease. They then run compound screens on the protein looking for compounds that produce a hit. These hits correspond to compounds that produce the desired result of the assay. Usually this desired effect is either the inhibition or the activation of the protein. However, many of these hits can be false positives and do not actually have a direct interaction with the protein to produce the desired effect. The compounds responsible for
producing these false positives in high-throughput screens are called pan-assay interference compounds, commonly referred to as PAINS (1). These compounds come from several different classes of compounds and can be identified by specific substructural features (3).

A compound or drug that produces a true hit inhibits or activates a protein by fitting directly into the binding site of the protein. Instead of binding directly to the protein’s binding site, many PAINS cause reactions that produce a false signal that mimics the drug-like interaction. There
are many mechanisms by which these false positives are produced. Some produce fluorescence or a color change that mimics what a true hit produces. Others can trap metals that are used to produce molecules in the library or the reagents used in the assays. These metals can then produce a positive signal that does not reflect on how the compound interacts with the protein. Even further, PAINS can alter the structure and function of proteins through interactions that do not require them fitting into the binding site. For example, one class of PAINS is redox cyclers. These compounds react to produce hydrogen peroxide, which can inhibit a protein without the compound ever having to bind specifically to it. Thus when the results are analyzed, it will show that this compound produced a hit because the protein has been inhibited. However, there was no specific interaction of the compound with the protein and is therefore not considered a hit (1). In addition to fluorescence and redox reactions, PAINS produce false positives through covalent modifications, chelation, and degradation reactions (3).

Due to the fact that PAINS operate using many different mechanisms and cover a wide range of compound classes, they can be difficult to identify in a compound screen, which usually consist of thousands of compounds. If PAINS are not identified from the hits produced in a compound screen, researchers can spend many years and thousands of dollars trying to further their research and in the end not have a drug that produces the desired effect. That is why many researchers have been developing filters that attempt to identify and eliminate them. These filters identify substructures of the compounds that correspond to specific compound classes that often interfere in assays (2). To this day, more than 450 compound classes have been identified and are used to filter out PAINS from compound screens (3). However, these filters are not comprehensive and the compounds can be mislabeled as either a hit or a PAIN. This is because the compounds are labeled purely based on structural characteristics from a specific library. Some compounds that are labeled as a hit may actually be a PAIN because other functional groups within the compound cause false positives, but they are not listed as a criteria for PAINS in the filter (2). Additionally, researchers also run the risk of labeling a true hit as a PAIN due to a compound’s similarity to other PAINS. However, in some cases, the substructures that would trigger labeling the compound as a PAIN are crucial for the compound
to produce the true hit (3). Using filters to help identify PAINS in compound screens is a necessary step, but researchers must be careful and use their own judgement to account for any mislabeling that may occur.

In conclusion, PAINS, or pan-assay interference compounds, are compounds that produce false positives in high-throughput screens. Unlike true hits, where the compound or drug binds to the binding site of the protein, PAINS produce a reaction that the result mimics the drug interaction. This can occur through many different mechanisms, including fluorescence, covalent modification, chelation, redox reactions, and degradation. If these compounds are not identified as PAINS, researchers run the risk of wasting years and thousands of dollars on research that
in the end does not result in a drug that is able to treat the disease of interest.

 

Audio Recording

References

  1. Baell, J., & Walters, M. A. (2014). Chemical Con Artists Foil Drug Discovery. Nature, 513, 481–483. Retrieved from
    https://www.nature.com/news/polopoly_fs/1.15991!/menu/main/topColumns/topLeftColumn/pdf/513481a.pdf
  2. Baell, J. B., & Nissink, J. (2018). Seven Year Itch: Pan-Assay Interference Compounds (PAINS) in 2017-Utility and Limitations. ACS chemical biology , 13 (1), 36–44. https://doi.org/10.1021/acschembio.7b00903
  3. Jasial, S., Hu, Y., & Bajorath, J. (2017). How Frequently Are Pan-Assay Interference Compounds Active? Large-Scale Analysis of Screening Data Reveals Diverse Activity Profiles, Low Global Hit Frequency, and Many Consistently Inactive Compounds. Journal of Medicinal Chemistry , 60 (9), 3879–3886. doi: 10.1021/acs.jmedchem.7b00154
  4. Bartlett, M. (n.d.). Chemical Genomics Robot . Photograph.
  5. Toxoflavin . (n.d.). Photograph.
  6. Shafee, T. (n.d.). Hexokinase induced fit . photograph.

 

Questions

  1. What are PAINS? PAINS, or pan-assay interference compounds, are compounds that produce false positives in high-throughput compound screens.
  2. How do PAINS work? PAINS produce a false positive reading by mimicking the drug response. There are many mechanisms that PAINS can work through to produce the false positive. These include fluorescence, chelation, redox reactions, and degradation. Through these mechanisms, the protein is either activated or inhibited, depending on the protein’s role in the disease.
  3. Why is identifying PAINS in a compound screen important? If PAINS are not identified, researchers can spend years and a lot of money trying to further develop the compounds into a treatment but in the end come up with no successful treatment.
  4. What are the consequences of a true hit being mislabeled as a PAIN? If a true hit is mislabeled as a PAIN, the compound is “discarded” and no further research would be done on it. This could result in researchers overlooking/missing out on a potential treatment for a disease.