The use of Transmission Electron Microscopy (TEM) in preclinical studies has significantly waned in the last 20 or so years. There are multiple reasons for this. TEM is expensive and labor intensive. Compared to standard histopathology, TEM requires significantly more time and money to analyze samples. The cost of maintaining an electron microscope is practical only for large organizations. Interpreting TEM images is challenging, as training programs and opportunities to practice the discipline are limited and thus expertise is difficult to develop. Confounding this are newer molecular techniques and additional imaging technologies such as confocal microscopy that can answer some of the questions previously reserved only for TEM in a much quicker and cost-effective manner. Despite all these challenges, TEM still has an important role in preclinical safety assessment. It is critical, however, to understand the strengths and weaknesses of TEM so that time and effort are not wasted in producing TEM data that does not add to the scientific assessment of a preclinical study.
The first and most important aspect of using TEM in preclinical studies is to formulate a limited, and well-defined question that TEM is well suited to answer. In almost all cases this means that some other aspect of the study has produced a finding which TEM can help further characterize. Examples of this include vacuolation of hepatocytes observed as a light microscopy finding. Here, TEM is well suited for characterizing the nature of the vacuoles and their location within the cell; both of which may have important implications for risk assessment. If urine protein levels are elevated, TEM may be helpful in assessing if structural changes to the glomerulus are the cause of proteinuria. A third application of TEM in preclinical studies could be the assessment of animal models. If an animal model for glomerular damage/loss is being used, TEM would be helpful to ensure that the animal model has the desired glomerular changes. TEM should never be used to “screen” for changes within tissues when there isn’t some other data to support that change. Furthermore, TEM should not be used to set NOEL or NOAEL because it is not practical to evaluate incidence and severity of TEM findings across dose groups.
The second most important aspect of using TEM in preclinical studies is to understand the importance of sample preparation. Because most preclinical laboratories do not routinely process tissue samples for TEM, any study that plans to use TEM should consult with the TEM lab that will process the samples well before necropsy and tissue harvest. There are a number of references that are helpful in preparing TEM samples for shipment, such as Dykstra et al. 2002. The critical aspects of proper tissue preparation involve trimming the tissue to the proper size (usually no thicker than 1 mm in at least one dimension), selecting the correct fixative for TEM, and, depending on the tissue type, trimming with proper orientation. Orientation in tissues with more uniform consistency, such as kidney and liver, is less important, where as in eye and peripheral nerve it is critical for proper analysis. There is no post-collection step in TEM processing that can correct for poor or inadequate sample preparation. If it is not done correctly, the TEM images will be suboptimal, as will be the information gathered from them.
Although the use of TEM has undoubtedly declined in preclinical safety studies over the past 20 years, it still remains a useful tool to answer questions in preclinical studies that no other modality can. EPL has conducted TEM analysis of tissue from preclinical safety studies for over 25 years. We currently have a contract with the Laboratory of Neurotoxicity Studies at Virginia Tech University. Samples are sent to Virginia Tech where images are produced in consultation with pathologists at EPL. An EPL pathologist will then interpret the images and write a report. These studies can be done using GLP or non-GLP processes, as needed. Other uses of TEM, such as in investigative studies, are also possible and we have worked closely with clients to produce images and data that fits their needs. Veterinary pathologists experienced in TEM processing and analysis are available at both EPL and Virginia Tech. In addition, the TEM technicians at Virginia Tech have over 30 years of experience in processing tissue samples and producing high quality TEM images.