Anja van de Stolpe, Philips Research, HTC, Eindhoven (firstname.lastname@example.org)
Roughly 10-15 cellular signal transduction pathways (the exact pathway number depends somewhat on definition) regulate basic cellular processes like cell division, differentiation and migration, which are necessary for embryonic development, tissue repair and physiological processes. They also play an important role in the pathophysiology of a large number (if not all) of diseases, like all forms of benign tumors and cancers, immune-mediated diseases (e.g Rheumatoid arthritis, MS), inherited diseases (e.g. epilepsy), inflammatory disease (e.g. atherosclerosis), degenerative diseases (e.g. Alzheimer), and infectious diseases (control of immune response).
The past decades our knowledge on these signal transduction pathways has increased tremendously. They can be categorized as hormonal driven nuclear receptor pathways (androgen receptor, estrogen receptor, progesterone receptor pathways), developmental pathways (Wnt, Hedgehog, TGFβ, and Notch), the highly complex growth factor regulated signalling pathway network, including the RTK-PI3K-AKT-mTOR, JAK-STAT, and MAPK pathways, and the inflammatory NFkB pathway.
Quantitatively measuring functional activity of signal transduction pathways in in vitro disease models or models for healthy organs/tissue, like organ-on-chip models, can help to characterize the (patho-)physiological status of the cell/tissue culture model, for example to compare with in vivo diseased tissue in the patient, or to measure the effect of interventions, like addition of a drug to the culture system.
During the past nine years at Philips Research we have developed novel Pathway Tests (Oncosignal) which can quantitatively measure activity status of these signal transduction pathways in cell and tissue samples, based on computational inference of activity of a signal transduction pathway from measurements of mRNA levels of well-validated direct target genes of the transcription factor associated with the respective signalling pathway (Verhaegh et al. Cancer Research 2014).
Computational model-based Pathway Tests for quantitative measurement of pathway activity of the androgen receptor (AR), progesterone receptor pathway (PR), PI3K, Wnt, Hedgehog (HH), TGFbeta, Notch, NFkB, and STAT1/2/3 pathways have been biologically, and partly clinically, validated on multiple cell/tissue types (cell culture, xenograft mouse models, and patient samples). The first application is cancer diagnostics, reason why the tests are called Oncosignal, however they are applicable to many other diseases.
How do these Oncosignal Pathway Tests work and why does it work?
The left picture shows (simplified) a signal transduction pathway: a signal/ligand (L) from outside the cell, binds specifically to a receptor (R) on/in the cell; the signal is transduced by one or more signalling proteins (S) to activate a transcription factor protein (F) in the nucleus of the cell. The transcription factor regulates production of a number of “target” mRNA molecules (T), which are subsequently translated into proteins, which adapt the function of the cell to the original signal. The right figure shows the most important pathways, in this case for cancer, and a few drugs that interfere with abnormal pathway activity.
With these Oncosignal Pathway Tests, we measure the levels of the direct target mRNA molecules that are produced by the transcription factor; thus, we actually measure as direct as possible the “output” of a transcription factor, and thus of associated signaling pathway that activates the transcription factor. This is done by means of qPCR or Affymetrix microarray, and in the future RNA sequencing. The Pathway computational model subsequently uses these measured levels to calculate a probability that the pathway is active, provided as a pathway activity score. To enable this calculation, the models have been calibrated on a sample set which contains samples in which the pathway is known to be inactive and samples in which the pathway is known to be active. After a one-time calibration on one cell type, the model can be used on multiple, even most, cancer types.
If the measured levels of (a varying number of) the target mRNAs that are present in the cancer cells/tissue sample are higher than normal, the model calculates a high probability for pathway activity. If wanted, the model can be optimized for a specific cell/tissue type or disease by re-calibrating on that tissue/cell type.
Why are these Pathway Tests (Oncosignal) applicable to all kinds of diseases?
The target mRNAs that were selected were chosen on being direct target genes, meaning a minimal involvement of cellular proteins in the production of these target mRNAs, which minimizes tissue specificity, and enables use on a wide variety of cell/tissue types. Also, the target mRNAs where not chosen based on their specific role in cancer, but solely because they are produced by the pathway transcription factor, reason why they can be used for many diseases.
If you would like to have more information (like specific validation results or sample prep) on the tests or on how to use them as functional “readout” for your organ-on-chip culture experiments, you can contact of course always contact me, or send an email to email@example.com or firstname.lastname@example.org.