## Understanding Accelerator Mass Spectrometry

Accelerator mass spectrometry (AMS) is an analytical method used to detect the amount of radioactive carbon in a biological sample. It is an extremely sensitive methodology that can be used in early clinical research when conventional radiometric detection methods such as liquid scintillation counting are not possible. AMS is a powerful technique; however, it is different from other mass spectrometry methods and requires specialized techniques for optimal use. In this post, I will review some of the basic ideas and principles. At the end of this post, I hope that you will understand how AMS works, and how you can use it in an upcoming clinical study.

AMS is a very sensitive technique of measuring carbon atoms. AMS can differentiate between carbon-12, carbon-13, and carbon-14, by detecting individual atoms. AMS does not detect molecules, and AMS can only be tuned for one atom at a time (eg, carbon vs. oxygen). Thus AMS detects atoms, and traditional MS detects molecules. This is the primary, and most important, difference between the two methodologies.

AMS analysis consists of the following steps:

1. Biological sample is converted to graphite (pure carbon)
2. Graphite sample is loaded onto AMS instrument
3. AMS analysis is performed
4. Ratio of 14C to 13C and of 13C to 12C are measured
5. Ratio of 14C to 12C is calculated, and converted to ng-equivalents/mL

### Biological sample is converted to graphite (pure carbon)

The conversion of a biological sample to graphite requires several steps. Great care is taken during sample preparation to avoid contaminating the sample with carbon material from the lab or the technician. Any organic material will contaminate the sample and skew the results. As shown in the image, the sample (normally 20 μL of blood, urine, feces, serum, tissue, etc.) is placed in a quartz tube. The quartz tube is then placed inside of a handling tube. The technician uses the handling tube to avoid contact with the quartz sample tube while performing the necessary sample preparation steps. The sample is dried in a vacuum centrifuge without heat. The quartz tube that contains the dried sample is transferred to a clean tube free from carbon and the new tube is sealed with a vacuum (0 atm inside the tube). The sealed sample is then heated at 900°C overnight. Any organic material turns into CO2 and the pressure inside of the sealed sample rises to 10 atmospheres. This step is called combustion. It takes approximately 24 hours to process a single set of samples from the biological state to completing combustion.

Following combustion, the pressurized sample contains carbon dioxide (CO2). Using a vacuum system, the CO2 is reduced to graphite (or pure carbon) using zinc and iron under freezing conditions (liquid N2) to catalyze the reaction. The reaction is completed by heating the sample for 2 hours at 500°C. At the conclusion of this step all organic carbon material has been converted to graphite or pure carbon. The graphite will contain all forms of carbon in the sample (14C, 13C, and 12C).

### Graphite sample is loaded onto AMS instrument

The graphite samples are then loaded onto a sample wheel that is used to inject the sample into the AMS instrument. To load the samples, graphite is loaded into a pellet using the AMS pellet press. These pellets are a uniform diameter and thickness and fit into the AMS sample wheel. A single sample is loaded onto each spot on the wheel, then the wheel is loaded onto the AMS instrument. The conversion of a combustion sample to graphite and loading the AMS sample wheel takes approximately 24 hours. An example of the sample press and sample wheel are shown below:

### AMS analysis is performed

The AMS instrument starts by firing a Cesium (CS) ion source at a single graphite sample on the AMS sample wheel. This CS ion beam creates negatively charged carbon atoms that are attracted to the positively charged 500 kV ion accelerator. The negatively charged carbon atoms accelerate toward the ion source, and in the middle of the ion source the negatively charged atoms pass by an electron stripper which removes one electron from each carbon atom. After the electron stripper, the uncharged carbon atoms are repelled from the positive ion accelerator achieving an effective 1 million volt acceleration (500 kV approaching the ion source, and 500 kV leaving the ion source). These accelerated carbon atoms move toward a detector for 13C and 12C. These atoms have differ in mass by 1 atomic mass unit (AMU) thus they can be separated due to the 1 million volt acceleration. These detectors actually create a continuous chain of carbon atoms between the source sample and the detector, thus a microamperage (μA) is measured for 13C and 12C separately. This amperage is proportional to the amount of each atom present in the sample. The 14C atoms continue to move through the detectors and arrive at a particle detector. These particle detectors are very sensitive and can detect cosmic rays. Mathematical algorithms are used to differentiate random atomic “hits” from real 14C “hits”. The measurement of 14C is based on the number of particles that contact the detector.

The distribution of natural carbon is as follows:

• 12C represents 99% of natural carbon
• 13C represents 1% of natural carbon
• 14C represents <0.0000000001% of natural carbon

### Ratio of 14C to 13C and of 13C to 12C are measured

Using experimental data, two key ratios are calculated:

• $\frac{^{14}C}{^{13}C}$
• $\frac{^{13}C}{^{12}C}$

These ratios are calculated because direct calculation of 14C relative to 12C would result in too much variability and error. In the above equations, 12C and 13C are compared and differ by ~102. Similarly, the quantity of 13C is closer to the quantity of 14C than is the more abundant 12C.

### Ratio of 14C to 12C is calculated, and converted to ng-equivalents/mL

From these two ratios, the amount of 14C relative to the amount of 12C can be determined by multiplying the two ratios listed above:

• $\frac{^{14}C}{^{12}C}={\frac{^{14}C}{^{13}C}}*{\frac{^{13}C}{^{12}C}}$

The resulting value has units of disintegration per minute (dpm) per mg of carbon. This dpm/mg of carbon can be converted to dpm/mL plasma/blood/urine based on the total carbon content of the sample. The carbon content of plasma or blood samples is relatively constant at 1 mg of carbon or each 20 μL of plasma or blood. The carbon content of other tissues and fluids (e.g., urine) varies by sample and it must be determine for each sample.

After converting to dpm/mL, the specific activity of the drug measured as dpm/ng can be used to convert from dpm/mL to ng-equivalents/mL in the sample. The limit of quantitation for 14C using AMS is 0.5 dpm/1 mg of carbon. Given a 20 %mu;L sample, the lower limit of quantitation is 0.01 dpm/mL.

I appreciate that this explanation was long and somewhat complicated; however, an understanding of the methodology will likely help in planning your next study with AMS. This analytical technique is much more sensitive than liquid scintillation counting (lower limit of detection is ~15 dpm/mL) which allows for detection of very small amounts of radioactivity in samples. This is very useful when the amount of radioactivity absorbed is very small due to the route of administration (e.g., topical ocular) or low bioavailability and large volume of distribution. There is great value in understanding that AMS detects atomic carbon and utilizes a destructive analytical technique that results in total loss of molecular structure.

I hope you have more information and can now speak with confidence about accelerated mass spectrometry or AMS.

Here are some pictures from the AMS lab at Accium Biosciences in Seattle, WA.

Today, drug development is carried out in human subjects and animals. However, as computing power and the number of sophisticated technology platforms grow exponentially, and our knowledge of human health and disease increases, the virtualization of clinical research and development will grow steadily. Read this article to learn more.