Histochemistry, the study of chemical components within the context of cell and tissue structure, using stains and/or antibodies, is a mainstay of most labs. With evolutionary roots that stretch back to ancient times when scientists first set out to discover what role “humors” or chemicals played in the function of the human body, the field marries microscopy, chemistry, immunology, and molecular biology.
While the core template for performing histochemical analysis of tissues hasn’t changed much over the decades, scientists are discovering new ways to circumvent common and costly problems associated with the technique.
Antibodies—the costly double-edged sword of the research world
When histochemistry utilizes antibodies to detect the presence or localization of specific proteins in tissue, it is referred to as immunohistochemistry (IHC). Cell or tissue morphology is first preserved using fixative. Samples are washed and blocked to prevent nonspecific protein-protein interactions, and then a primary antibody (or antibodies) is applied. Microscopic visualization of the protein within the cells or architecture of the tissue is achieved through the use of enzymatic or fluorescent reporter systems.
Formalin fixed paraffin embedded human tonsil stained with a high molecular weight cytokeratin antibody labeled and HIGHDEF® blue IHC chromogen (HRP) (Prod. no. ADI-950-151) produces a brilliant blue color. Image courtesy of Enzo Life Sciences.
Although straight-forward, unfortunately, there are myriad opportunities for the process to go awry. Usually these issues can be attributed to poor technique, insufficient controls, or problematic reagents.
Antibodies with poor specificity, for example, often lead to false-positive staining, making it impossible to draw relevant conditions from research, according to Simon Renshaw, principal imaging scientist at Abcam. He says that these poorly characterized antibodies cause scientists across the globe to waste nearly $800 million, annually.
It is a generally accepted plight, in the histochemical realm, that antibodies vary wildly, requiring researchers to be fastidious in the validation of antibodies (even one that has been used in the lab previously). Less than 5% of polyclonal antibodies bind the intended target and functionality varies between batches. Monoclonal versions suffer from other frustrating afflictions: hybridoma cell lines may die off or lose antibody genes. Monoclonals may also bind to more than one target and so are fraught with frustrations similar to those encountered with polyclonal antibodies.
Recombinant antibodies avoid some of the quirkiness inherent to antibody production. Manufactured by cloning the immune-specific heavy and light antibody chains into a high-yield mammalian expression vector, which is then expressed in bacteria, yeast, or mammalian cell hosts, Renshaw points out that these antibodies “are highly reproducible as they eliminate batch-to-batch variation since they are generated from a known DNA sequence, making each antibody that is produced identical.”
While invaluable when wielded correctly, antibodies are a tool that will always require a healthy dose of skepticism as accompaniment—“Controls aren’t glamorous, but they are essential to accurately interpreting results,” warns Renshaw. These include simple steps like using a “no primary antibody control” to show that staining only occurs where the antibody is present. Scientists should also be cognizant of trying to minimize variability by standardizing protocols and tissue preparation.
CRISPR technology is helping to make validation less painful, since there is now increased availability of knockout cells for use as a negative control, says Renshaw. “At Abcam, we’re using knockout cell lines generated via CRISPR-Cas9 to validate the specificity of our primary antibodies.” The company currently offers 1,000 knockout validated antibodies.
Biocompare’s IHC Search Tool
Find, compare and review IHC
tools from different suppliers Search
Jonathan Weinreich, product manager at ENZO, explains that false-positives can also be avoided by using non-biotin systems, since endogenous biotin may be present in tissue. “Recently, the gold standard for IHC moved to polymeric detection, where horseradish peroxidase [is] conjugated to a bound polymer, which in turn is bound to a secondary antibody ... Since the nanopolymer has a high surface area, it will be able to bind many horseradish peroxidase molecules, increasing the number of enzymes capable of reacting with substrate to produce color per slide.”
Heavy metal IHC—pushing the limits of multiple protein detection
“There has been a growing demand for simultaneous detection of multiple protein targets on a single tissue sample without significant loss in spatial resolution, especially in a rapidly growing research area like immuno-oncology,” says Renshaw.
Imaging mass cytometry, which combines IHC with mass spectroscopy, utilizes rare earth metals conjugated to antibodies as reporters. According to Renshaw, after antibodies stain the tissue section, the sample is nebulized and metals are ionized and measured through time-of-flight (TOF) mass spectrometry. The data for each metal isotope corresponds to its original tagged antibody and the amount of heavy metals detected is proportional to the number of proteins bound by the antibodies.
Renshaw say that this opens up the possibility of simultaneously detecting a hundred or more targets.
Fluidigm’s Hyperion system and IONpath are two imaging platforms currently on the market that are capable of performing imaging mass cytometry on tissue samples.
These types of advancements allow for the use of antibodies against housekeeping proteins for reference and normalization, providing more reproducible and accurate IHC results, according to Renshaw. “This opens the door to IHC as an even more trustworthy clinical and diagnostic application.”
Moving beyond antibodies with MALDI imaging
Matrix-assisted laser desorption/ionization (MALDI) imaging or tissuetyping is another mass spectroscopy based technique that can determine the distribution of potentially hundreds to thousands of proteins and small molecules from a single tissue section.
MALDI IMS of a lung adenocarcinoma tissue section displaying different areas defined from distinct molecular contents in the tissue.
a) Hematoxylin and eosin staining of the section after MALDI acquisition.
b) Combined image of two peptide masses discriminating tumor (red) from non-tumor (light blue) areas.
The peptide in red was identified as Cytokeratin 7, a well known marker expressed in adenocarcinoma. MALDI data are acquired with Bruker’s rapiflex MALDI Tissuetyper. Image courtesy of Bruker.
“[Bruker’s] MALDI Tissuetyper enables the assessment of spatial molecular arrangements in tissue sections; it actually goes far beyond microscopy in providing hundreds of different molecular images from a single scan without the need of target-specific reagents,” says Juergen Tressel, business development manager & team leader MALDI Tissuetyper business unit at Bruker.
Rather than using antibodies or probes to seek out specific targets, MALDI imaging uses direct label-free measurement of proteins, peptides, lipids, drugs, and metabolites from tissue by analyzing the mass to charge ratio (m/z).
According to Tressel, MALDI imaging is a non-destructive technique that allows for subsequent histological staining of the same section. The molecular data can be overlaid with microscopic data.The result is a high-throughput system that can determine if say, a section of a biopsy is malignant or benign. “Tissuetyping for classification of the tumor for typing and grading and for screening could be a fast, sensitive, and cost-effective workflow and reduces the bias of a subjective evaluation of a pathologist,” notes Tressel. He also points out that as less invasive diagnostic procedures yield smaller biopsy samples, there is simply less material to properly analyze with traditional histological and IHC methods.
When a tissue section is analyzed with tissuetyping, a spectral output that plots intensity against the mass to charge ratio is generated. The predominant peaks of this molecular signature can then be compared against classifiers (previously validated spectra). In the case of a specific type of cancer, for example, the classifier spectrum might contain particular molecules that have already been implicated in the disease allowing for relatively straight-forward comparison to biopsies.
“I like to think of the observed peak pattern as a ‘molecular phenotype’, and the classifiers are trained based on those molecular phenotypes of known tissues,” says Soeren Deininger, scientific affairs manager for the MALDI Tissuetyper at Bruker. “Later, unknown tissue can be classified based on those classifiers.”
While MALDI imaging does not directly identify proteins (as targeted approaches such as IHC do), that can be achieved by follow-up using a top-down approach, as is possible with Bruker’s rapifleX TOF/TOF upgrade, or with liquid chromatography-mass spectrometry.
“[The] MALDI Tissuetyper is the next biggest thing within pathology,” says Tressel. “It has the power and potential to uncover new markers for diagnostic purposes or markers that correlate with disease severity as well as prognosis and therapeutic response.”
References
1. Mark, R, Wick, MD. Histochemistry as a tool in morphological analysis: a historical review. Ann Diagn Pathol. 16:71-8. 2012. [PMID: 22261397]
2. Bradbury, A, Plückthun, A. Reproducibility: Standardize antibodies used in research. Nature. 518:27-9. 2015. [PMID: 25652980]
3. Rimm, DL. Next-gen immunohistochemistry. Nat Methods. 11:381-3. 2014. [PMID: 24681723]
Image: Immunohistochemical analysis of paraffin-embedded human mantle cell lymphoma tissue, labeling Cyclin D1 with unpurified ab134175 at 1/100 dilution. Image courtesy of Abcam.