|
For drug developers, being able to see where a drug candidate is distributed in a targeted tissue helps them to better assess the potential value of that compound as a pharmaceutical product. To that end, one scientist, Dr. Walter Korfmacher, and researchers from Schering-Plough are now employing mass spectrometry systems from Applied Biosystems/MDS SCIEX to apply a technique known as MS (mass spectrometry) tissue imaging to obtain detailed information about the spatial distribution of drug candidate compounds throughout tissue samples.
Using the Applied Biosystems/MDS SCIEX API QSTAR® Pulsar i hybrid LC/MS/MS system, a QqTOF system equipped with an optional matrix-assisted laser desorption/ionization (MALDI) source, Dr. Korfmacher and a team of researchers have visualized the different regions within rat and mouse tissues where candidate drug compounds were located.
Exploring Applications of MS Tissue Imaging
As director of exploratory drug metabolism at Schering-Plough, Dr. Korfmacher oversees 13 scientists in two labs who work together with drug discovery teams to identify new compounds that the scientists then recommend for development into pharmaceutical products.
“The reason for the existence of our exploratory drug metabolism department is to improve the quality of drugs that go into development. Once you go into development, the cost of moving a drug along the pipeline starts to escalate rapidly,” notes Dr. Korfmacher.
One way that Schering-Plough strives to reduce development costs is to learn as much about a candidate drug compound in the discovery phase, before investing time and money in developing that compound into a marketable drug.
“Typically,” says Dr. Korfmacher, “somewhere between 5 and 10 percent of compounds that go into development actually ever becomes a drug. Even when you reach phase 1 [of clinical trials], only about 20 percent of compounds succeed and become a marketable drug. So, we’re continually trying to improve upon that record. We’d like to have a lower rate of attrition. In other words, if we could raise it from 1 in 5 to 2 in 5 [compounds that succeed], that would be a big improvement.”
According to Dr. Korfmacher, just ten years ago, the main reason that compounds failed in phase 1 was pharmacokinetics, something going awry in the rate of adsorption, distribution, metabolism, or excretion of a candidate drug during testing. Today, in addition to pharmacokinetics complications, he notes that other issues such as toxicology studies frequently reveal problems that can disqualify a candidate compound from further development into a marketable drug.
“So, we’re looking to see if we can come up with tools that can help with these new problems that are emerging,” he says.
One such tool is a technique developed by Dr. Richard Caprioli, Director of the Mass Spectrometry Research Center, Vanderbilt University School of Medicine, and colleagues at Vanderbilt called MS tissue imaging, a sophisticated application of MALDI-TOF MS/MS analysis that provides researchers with reliable localization information for small molecules.
“The goal of tissue imaging is to find out if your compound is present in a tissue, and, if so, to find out how much of it is there, and, if the amount of it varies, depending on the location in the slice of tissue that you’re looking at,” says Dr. Korfmacher.
By applying tissue imaging to drug discovery studies, researchers can track the whereabouts of a particular drug candidate within a target tissue early in the discovery phase before moving the compound to the costlier drug development phase.
“In the past, one never looked at the tissue distribution of a candidate compound before development, but what has changed over the last 10 years is that many things that used to only be done in development are now done in discovery,” explains Dr. Korfmacher.
Besides helping researchers obtain detailed information about where a compound is distributed in a tissue sample, MS tissue imaging can provide answers to questions about the way compounds act in different tissue types.
“One of the reasons that tissue imaging is enticing, and one of the reasons that we’re looking at it, is for central nervous system projects,” notes Dr. Korfmacher.
For example, one question that often confronts drug researchers is: why are some compounds that are intended to treat brain disorders able to get into the brain, but are not active once they get there?
“One of the possibilities,” Dr. Korfmacher predicts, “is that it’s not in the correct part of the brain. Before we had MS tissue imaging, we had no way of answering that question.”
Moving More Compounds from Discovery to Development
At Schering-Plough, Dr. Korfmacher oversees two labs that are equipped with multiple mass spectrometry systems, including four Applied Biosystems/MDS SCIEX mass spectrometry systems: an API 3000™ LC/MS/MS System, two API 4000™ LC/MS/MS Systems, and an API QSTAR® Pulsar i hybrid LC/MS/MS system that the exploratory drug metabolism labs use for the MS tissue imaging technique.
“Our primary function is still quantitative analysis of pharmacokinetics samples that come from laboratory animals,” notes Dr. Korfmacher. “HPLC-MS/MS is the standard technique for doing quantitative analysis of either plasma, or tissue homogenates, and it’s the standard technique that’s used routinely (1).”
While pharmacokinetics applications still represent a significant share of the workload in Dr. Korfmacher’s labs, he sees the great potential of MS tissue imaging for improving the lab’s chances of identifying problems with candidate compounds earlier in the drug discovery process.
“We’re looking for ways to use tissue imaging to help us move [more] compounds from discovery into development,” he says. “If we can obtain a tissue image earlier [in the drug discovery phase], we may be able to answer a question that’s important to the discovery team. We’re trying to improve the chance that when our compound does go into development that there won’t be some problem that we could have seen in discovery that shows up in development and then kills the compound.”
How MS Tissue Imaging Works
Dr. Caprioli, who also works as a consultant for Schering-Plough, first developed the MS tissue imaging technique for locating proteins in tissues samples. The technique was successful at locating proteins and peptides, which have relatively large molecular masses of between 2,000 and 50,000 daltons. Then, about three years ago, with urging from Dr. Korfmacher, Dr. Caprioli adapted the technique to detect small molecules, such as drug candidates, in tissue samples.
Use of the API QSTAR® Pulsar i system from Applied Biosystems/MDS SCIEX helped the researchers to overcome limits of detection, and clearly identify the signals generated by candidate drugs, compounds that generally have molecular masses of around 500 daltons.
In MS tissue imaging, researchers place a tissue sample on a MALDI plate of a QSTAR system, and then view an image of the sample tissue on a computer screen. By using the MALDI System MS Imaging (MMI) software developed by Applied Biosystems/MDS SCIEX, they select the part of the tissue for which they want an image. They then determine how closely they want to space successive laser shots that create the image of the sample. To increase the level of detail of a sampling, the researcher increases the number of laser shots and pixels generated per unit area. For example, 4,000 pixels produce a much more detailed image of the sample than do 200 pixels.
The key to MS tissue imaging is the MMI software. This software allows users to associate a selected pixel location on a tissue image with a mass spectrum of a compound if one is present in the selected area of the tissue.
“The software is critical to making the whole thing work,” notes Dr. Korfmacher.
To evaluate the results of a tissue analysis, the researcher reviews an image on a computer screen filled with colored spots at locations where a particular compound has been detected. The intensity of the color of a spot corresponds to the amount of signal that the laser detects at any one point, or pixel.
“We’ve shown that MALDI MS/MS imaging is semi-quantitative,” says Dr. Korfmacher. “For MS tissue imaging, when the color is more intense in one part of the picture that means that there’s more of the drug in that part of the tissue than in the other part. However, we cannot determine what the actual concentration of the compound is in different parts of the tissue.”
Selecting any pixel that is part of a colored spot on the image will display a mass spectrum, or product-ion spectrum, representative of the compound present in that region of the tissue.
“The product-ion spectrum can then be compared to the authentic standard. This gives you somewhat of a fingerprint match,” explains Dr. Korfmacher.
If the displayed product-ion spectrum matches the pattern of a known compound, the drug researcher knows that that compound is present in the selected region of the tissue.
“We use this feature routinely to get the mass spectrum of an ion of interest,” he says. “This allows you to determine whether it’s indeed the drug that you’re interested in, and not just random noise that happens to show up. You can tell the difference by looking at the mass spectrum generated by the software.”
Drawbacks of Other Methods
Alternative methods for identifying the presence of different candidate drug compounds in tissues include autoradiography, or tissue homogenation followed by electrospray ionization (ESI) or atmospheric pressure chemical ionization (APCI)-MS analysis. However, neither of these approaches can provide the kind of detailed localization information that can be obtained from the MS tissue imaging technique. For instance, as Dr. Korfmacher explains, in rat brain tissue homogenates, it is unlikely that a scientist will be able to determine unambiguously where a drug has been distributed within the brain tissue.
“We know how much [of a candidate drug compound] is in there, but we cannot say anything about where it is, so if indeed it is not distributed evenly throughout the brain, if it is in a certain region, [tissue homogenation] might be misleading,” Dr. Korfmacher says.
In the past, Dr. Korfmacher explains, researchers at Schering-Plough have been able to create tissue homogenates from smaller sections of the brain, and then use ESI or APCI-MS to obtain spatial information about the presence of small molecules in more distinct regions of the tissues, but there are practical complications with this approach.
“Occasionally we’ve done that, and have been able to find an answer. But, as a rat brain is already small, once you cut it up into smaller pieces, you run into practical limitations. If, for example, the compound you are looking for needs to get into the rat cortex, you can separate that from the rest of the rat brain and measure it, but even then you sometimes have to be careful, because if you’re not skilled at doing that sort of separation, you get misleading results,” he says.
In autoradiography, a radioactive isotope or fluorescent tag is attached to compounds so that the whereabouts of a candidate drug can be traced in tissues. However, as Dr. Korfmacher notes, there are drawbacks to this technique as well.
“It takes much more effort to make a radioactive version of compound, than it does to make the compound itself,” explains Dr. Korfmacher.
Also, researchers often must differentiate between the original compound and a metabolite that has retained the radioactive tags. This may result in the researchers monitoring the location of the tag and not the dosed drug. Moreover, radioactive compounds are typically not introduced until late in the drug discovery process, adding to the cost and the length of time needed to carry out this method.
“In drug discovery, you typically do not have radioactive compounds, at least not early in the discovery process,” he adds. “So, we wanted to be able to answer some of these questions with a mass spectrometry technique, without having to use radioactivity.”
In a study undertaken in 2003, Dr. Korfmacher, Dr. Caprioli, and a team of researchers were able to use MS tissue imaging to visualize the distribution of candidate drug compounds in different regions of the rat brain, and different areas of a mouse tumor tissue sample (2).
For this study, the researchers applied the MALDI imaging technique, using an API QSTAR® Pulsar i hybrid LC/MS/MS system. Dr. Caprioli and colleagues then applied the imaging capability of software now available with the QSTAR system to generate 2-D images of mouse tumor tissue samples and rat brain tissue samples from animals previously dosed with candidate drug compounds.
For the mouse tumor tissue samples, the resulting mass spectral images showed higher concentrations of the candidate drug compound in the periphery of the tissue, depicted as darker areas in the outer edges of the resulting image. Likewise, for the rat brain tissue sample, differing intensities of color in different regions of the 2-D image showed how the candidate drug was distributed throughout the rat brain tissue sample. The cortex area of the mass spectral image appeared darker than the striatum. These two experiments showed the effectiveness of the MALDI-QqTOF approach to MS tissue imaging for obtaining detailed information on the spatial distribution of small molecule drugs in tissue samples.
Separating Small Molecules from Matrix Interference
According to Dr. Korfmacher, the main challenge facing Dr. Caprioli and his team of researchers when first applying MS tissue imaging to the localization of small molecules in tissues concerned the way tissue samples are prepared for MALDI-TOF applications.
In MALDI applications, after researchers place samples containing compounds of interest on a plate, they apply a chemical matrix to the sample that facilitates the desorption and ionization of a compound of interest. The mass spectrometry system then generates signals in a mass spectrum that researchers interpret to identify the compound. However, the interpretation of these signals becomes complicated as the matrix itself also generates a signal that can obscure signals generated by compounds of molecular mass less than 1,000 daltons.
“The matrix that we put on the sample in order to make the technique work usually is the biggest signal that you see [in the resulting mass spectrum]. When you look at a mass range of 100-1000 daltons, most of the signal comes from the MALDI matrix,” says Dr. Korfmacher.
Separating this background noise produced by the matrix from the signals generated by a candidate compound was the key to extending tissue imaging applications from experiments capable of locating larger-sized proteins in tissues to ones that can pinpoint the location of drug candidate compounds in tissue samples.
To apply tissue imaging to small molecules, there was a need to find a mass spectrometry system capable of generating signals from compounds that could easily be distinguished from the background interference generated by the matrix.
"Dr. Caprioli had been doing everything on a MALDI-TOF system, but, to do the drug imaging, he had to use a QSTAR system, an MS/MS system that allowed him to do the imaging on a [small molecule] compound of interest," explains Dr. Korfmacher.
What Caprioli’s lab discovered was that, with a MALDI-TOF system, the background ions in the matrix overload the small molecule signals. Therefore, they could only see the drug if there were extremely high levels of it in the sample. In order to detect the analyte—the drug candidate of interest—in a tissue, they would need the MS/MS capabilities of a tandem mass spectrometry system such as the API QSTAR® Pulsar i hybrid LC/MS/MS system (3).
The QSTAR® System Resolves a Matrix Interference Problem
Researchers can obtain more detailed structural information about an analyte or compound by using a tandem mass spectrometry, or MS/MS system. In most MS/MS systems, two analyzers are connected together by a collision cell. The first analyzer generates the protonated “precursor ion” and can be used to select the analyte or compound of interest from the source sample. This precursor ion passes through a collision cell of gas and voltage, which breaks or dissociates the precursor ion into product ions. Analysis of the resulting product ions by the second analyzer generates a mass spectrum of the fragment ions comprising the precursor ion.
By monitoring the transition of a selected precursor ion to its dissociated product ions, tandem mass spectrometry systems give researchers a tool for distinguishing between product ions of similar mass-to-charge ratios, such as matrix ions and small molecule drug candidate compounds.
The QSTAR® system is a tandem analyzer capable of performing MALDI-MS/MS applications. With the QSTAR system, the MALDI produces the precursor ions, and the TOF analyzer provides exceptional mass accuracy and resolution for the product ions.
“[With the QSTAR system],” explains Dr. Korfmacher, “When you look at the product ion mass spectrum you see some [signal] from the matrix, and some from the analyte. Now we had a way to distinguish between the two signals. Using the single time-of flight system, we couldn’t distinguish between them. They showed up as one peak. When you do MS/MS, however, you get a product ion spectrum where the two signals are now far apart in terms of mass, so it’s easy to determine which ions come from the analyte [or candidate drug compound] and which ions came from the matrix. Once you know that, you can zoom in on the analyte product ion and do your imaging based on that transition and it will be very specific for the drug of interest, and the matrix won’t interfere.”
Besides allowing for clear discrimination between matrix and analyte, the higher mass accuracy and resolution of the QSTAR system compared to that of other MS/MS systems give Dr. Korfmacher added confidence when identifying compounds in a tissue image.
“When you’re looking at your data to see if indeed it’s the analyte that you’re trying to get your image of, the fact that you have the higher mass accuracy and resolution gives you increased confidence that [the compound] is indeed what you think it is,” says Dr. Korfmacher.
In addition to providing drug developers with higher mass accuracy and resolution, the QSTAR system uses an orthogonal reflectron time-of-flight analyzer in place of a third scanning quadrupole. Dr. Korfmacher notes how the use of a time-of-flight analyzer in the QSTAR system instead of a third quadrupole provides drug developers additional information in the resulting product ion mass spectrum that helps them to confirm the results of a tissue image analysis.
“With the QSTAR system, you get a spectrum of everything that comes out of a collision cell,” explains Dr. Korfmacher. “It gives you that whole packet of information, whereas a third quadrupole would typically only be used to select the one ion of interest. The reason that that’s useful is that even though to get your image you might just key in on the one transition—the protonated molecule to the ion of interest—you would have that other information there, which is valuable to confirm your results. The value of the time-of –flight analyzer is that you get [the necessary information] without losing sensitivity whereas, on a triple quadrupole, to get that same information you drastically lose sensitivity on your signal.”
A Future Routine Tool of Drug Discovery
While researchers in the exploratory drug metabolism labs at Schering-Plough are currently applying the MS tissue imaging technique to drug discovery projects, Dr. Korfmacher believes that the technique has even greater potential in the future as a tool for both drug discovery and drug development applications.
“At this point,” he notes, “we’re spending a fair amount of our time still refining the technique, trying to understand what makes it work better. In other words, we’d like to know why some compounds ionize better than others, and whether some types of MALDI matrices are better to use than others. It’s still a research tool, but the potential for it is very high, so it will probably become a routine tool somewhere in the next two to five years.”
References
- Korfmacher, W.A., “Bioanalytical Assays in a Drug Discovery Environment,” in “Using Mass Spectrometry for Drug Metabolism Studies,” W. Korfmacher, ed., CRC Press, Boca Raton, Florida, pp. 305-328 (2004).
- Reyzer, M.L., Hsieh, Y., Ng K., Korfmacher, W.A., and Caprioli, R.M., “Direct Analysis of Drug Candidates in Tissue by Matrix-Assisted Laser Desorption/Ionization Mass Spectrometry,” Journal of Mass Spectrometry 38: 1081-1092 (2003).
- Reyzer, M.L. and Caprioli, R.M., “MS Imaging: New Technology Provides New Opportunities,” in “Using Mass Spectrometry for Drug Metabolism Studies,” W. Korfmacher, ed., CRC Press, Boca Raton, Florida, pp. 305-328 (2004).
Trademarks
Applied Biosystems is a registered trademark, and AB (Design) and Applera are trademarks of Applera Corporation or its subsidiaries in the U.S. and certain other countries.
API 3000 and API 4000 are trademarks and QSTAR is a registered trademark of Applied Biosystems/MDS SCIEX, a joint venture between Applera Corporation and MDS Inc.
MDS and SCIEX are trademarks of MDS Inc.
|
