At A Glance
Name: Pankaj Desai
Position: Associate Professor, University of Cincinnati College of Pharmacy
Background: PhD, University of South Carolina — 1990
Pankaj Desai, an associate professor in the University of Cincinnati’s College of Pharmacy, has worked in pharmacology for more than 15 years. His current research focuses on enzymatic induction, a mechanism responsible for many unwanted drug-drug interactions during cancer treatment. Desai took a few moments to talk with Inside Bioassays about this mechanism and the recent paper he co-authored on the subject (Cancer Chemotherapy and Pharmacology, 2004 Jun 3 [Epub ahead of print])
The goal of your study was to examine the effects of paclitaxel and docetaxel on the expression of a particular protein in hepatocytes. Can you elaborate on the importance of studying the expression of this protein?
The protein that we are most interested in is cytochrome P450 — CYP3A4. Basically, there’s a superfamily of enzymes, and there’s a lot of members of this family, and with regards to transformation of drugs in our system, these enzymes are abundantly present in the liver and other parts of our body. They are involved in elimination of the drug. They participate in mono-oxidization of the drugs. It essentially can [change] the drug, in most cases, from an active to an inactive species. So changes in the level of this protein essentially changes the persistence of the active drug in the body. There are also examples where the compound you administer to the subject is not really the active one, but it requires an enzyme like this to work upon it and change it to the active form. So either way, changes in the level of CYP 3A4 can accelerate the elimination of other co-administered drugs. This is a very common mechanism that underlies many serious drug-drug interactions.
So this is a problem because a co-administered drug will be inappropriately eliminated from the body?
Right. A co-administered drug can be eliminated at a rate that’s much faster than in the absence of this enzyme-inducing compound. So that’s one possibility. The other is that sometimes this particular protein actually metabolizes the compound that is inducing itself. This is a phenomenon called auto-induction.
Besides using Western and Northern blot analyses, you also used cell-based assays in your research. Can you describe what types of assays you used and how you performed them?
Before even talking about the cell-based assays, Western and Northern blots are important, of course. They will tell you whether transcriptional activation is taking place or not, or for that matter, do you actually see increased protein levels. But ultimately, from a clinical pharmacology perspective, what matters to people like me is, ‘Does that translate into changing the activity of the enzyme?’ And to measure the activity, one can employ two or three different kinds of compounds, or you can actually use all of the compounds, but in my lab, we just go after one compound as a marker, and that is testosterone. The testosterone is converted to a hydroxylated product that is predominately metabolized by CYP3A4. So once we do our induction experiments, we try to compare the level of this hydroxylation in drug-treated cells versus controls. From a clinical perspective, that is probably the most important thing.
How do you actually perform the cellular assays?
What happens is we are collaborating with a couple of transplantation centers. And essentially, when they receive donor tissue that is surgically unsuitable, they typically used to throw away those kinds of tissues. But now you can actually bring those to your lab and work with a pathologist who will provide us with freshly isolated cells which are shipped to us as primary cultures, so we see those cells within two days of isolations. And when we see those, we immediately start the experiments and treat our cells with test compounds — in this case with paclitaxel and docetaxel. But we also include compounds that are very well known inducers, like rifampicin and phenobarbitol. These are compounds where we know they cause a bunch of problems in the clinic because they induce enzymes. So those are our typical prototype inducers that serve as positive controls, and all of these get compared to a set of cells that do not add the drug — all we do is basically add the solvent that we employed for dissolving our compounds. So that serves as our control. At the end of three days of experimentation — we change the media every 24 hours so [the cells] are incubated with fresh drug-containing media — we pull out the media and incubate the cells in drug-free media, so whatever drug was taken up by the cells will be given a chance to be effluxed out. That time frame is about two hours or so, and at that time we add testosterone — and these are still intact cells that are attached to the plate — for about 30 minutes. We then pull out a little aliquot from the supernatant bathing these cells and inject it into a high-performance liquid chromatography system so we can measure the formation of the metabolite. After this, we can then work with the cells and fractionate them and collect RNA and proteins and do the Western and Northern blots.
Are there any specific challenges to working with hepatocytes as opposed to other types of mammalian cells?
Sure. The biggest challenge, of course, is the short supply of hepatocytes. You usually have to wait for a while, because there’s a demand for these: A number of people are trying to use these and pharmaceutical companies are using them. This is probably the system that’s closest to doing the experiments in humans. Interspecies differences are really well known, but in particular, for induction, it is remarkable how humans differ from mouse and rat, so for that we have to depend on human hepatocytes.
[Second], sometimes we don’t know if the cells are going to make it or not. In other words, cells may not be healthy and may undergo apoptosis after the experiment has started. And depending on the age, gender, and smoking or drug-use history [of the donor], there is some variability that one needs to address in the experiments, which is tricky.
So what were your major findings of the drug induction experiments?
One of the critical things that we looked at is whether paclitaxel — an anti-cancer compound that was discovered and developed in the eighties and introduced as a drug in the US in the early nineties — has the potential to induce CYP3A4 and therefore influence the metabolism of other compounds. Like any other scientists we were first working with rat tissue and rat hepatocytes, and we saw there was a lot of induction, so we said: “Let’s go and see what happens in humans.” And we see that there is some pretty potent induction with paclitaxel. When docetaxel — a compound that has a very similar chemical structure to paclitaxel — was introduced, we did not know what its induction potential would be. When we worked with rat tissue we saw induction, but when we started doing the work with human hepatocytes, we did not see any induction in human tissue. This was a remarkable thing in that two compounds that are so structurally and mechanistically similar are showing a striking difference is that one induces in rats and humans, and the other only in rats. That was a major finding. The first time my graduate student showed me the data, I didn’t believe it. But he pretty much worked with six batches of hepatocytes to ensure that what we were seeing was real.
So what are the implications of that for cancer treatment?
For cancer compounds, often a cocktail of medications is used, and painkillers and other supportive medications are used. Typically a cancer patient is taking many drugs. Chances are, when considering paclitaxel versus docetaxel, if you were to use docetaxel you would have a reduced propensity for drug-drug interactions that involve enzyme induction — whereas with paclitaxel that may be an issue.
So what could possibly explain this difference? In the last four or five years, tremendous progress has been made — primarily by molecular biologists and pharmacologists — to understand how CYP3A4 is regulated. It turns out that one of the key receptors that regulates its expression is called PXR, or pregnane X receptor. In humans we call it hPXR. And it turns out there is a significant difference between rodent PXR and human PXR. A lot of these PXR molecules are similar across species — the amino acid sequence with regards to the DNA-binding site. But there is a significant divergence in the ligand-binding domain sites. There are examples of compounds that induce rat CYP3A4 enzyme, but not human CYP3A4. Vice-versa, there are compounds that induce in humans and not rats — rifampicin is one such compound. That may be explained by the fact that rifampicin may not be a substrate for rat PXR. So going back to the paclitaxel versus docetaxel example, I think the striking difference that we see could at least partially be explained that docetaxel is not a very strong activator of human PXR, whereas paclitaxel is. And that is where we used the cell-based assays.