November 15, 2013 (Vol. 33, No. 20)
Miguel Castro, Ph.D. Founder & CEO Bio-Synthesis
Audrey Chang Ph.D. Director of Global Development Services MilliporeSigma
Erin M. Hall Director Genetica DNA Laboratories (a LabCorp Sepcialty Testing Group)
Deborah Lee Dormady Letham, Ph.D. Scientist II Charles River Laboratories Biopharmaceutical Services
Michael Ojo Cell-Line Authentication Specialist LGC Standards
Markus Schmitt, Ph.D. CEO Multiplexion
Doug Storts, Ph.D. Research Head Promega
“The history of cell-line misidentification goes back to more than 50 years ago when the first human cell line, HeLa, was established,” wrote Yvonne Reid and Joe Mintzer in an article entitled “The Current State of Cell Contamination and Authentication—and What It Means for Biobanks” (Biopreservation and Biobanking, Vol. 10, No. 3, 2012, Mary Ann Liebert, Inc., publishers).
They also noted that even though techniques and culture conditions have improved, the problem of misidentified cell lines or cellular cross-contamination remains at a very high level.
GEN recently spoke to a number of cell-line authentication experts to find out just how large a problem this is and to determine what companies are doing to address this issue.
GEN: What types of cells do you work with?
Dr. Castro: We offer genotyping services for authentication of human cell lines and mycoplasm detection. Cell-line authentication verifies whether the genotype of the cell line in question is identical to the genotype of the presumed parental cells.
Cell lines from the following species can be routinely authenticated: Mus musculus (mouse), Rattus norvegicus (rat), Circetulus griseus (Chinese hamster), Chlorocebus pygerythrus (African Green Monkey), Sus scrofa (pig), Oryctolagus cuniculus (domestic rabbit), Canis familiaris (dog), Felis cattus (cat), Equus ferus caballus (horse), Capra aegagrus hircus (domestic goat), Ovis aries (sheep), Bos primigenius (bovine or cattle), and others.
Cell-line authentication is done at our accredited human DNA identity testing facilities, and we are also able to genotype any parental cells not already represented in a database.
Mr. Ojo: A majority of the cells we work with are human cancer cell lines from research institutes and universities looking to either establish or confirm the identity of a particular cell line they are using. We also deal with hospitals, biotech, and gene therapy companies wanting to do a ‘before and after’ comparison on the cell line, to identify whether any genetic changes have occurred as a result of the treatment or manipulation (i.e., addition of a protein or molecule).
We also do work with animal cell lines (mouse, rat, African green monkey, etc.) for speciation purposes. This gives researchers who do not necessarily work with human cell lines the opportunity to authenticate their cells.
Dr. Storts: Promega works with many types of mammalian cells, including cell lines of human origin such as HEK293, HeLa, Jurkat, and HepG2, to name a few. Many of these cell lines have been engineered with luciferase reporter constructs to simplify understanding of complex cell biology with robust cell-based assays.
The American National Standards Institute issued a Standard for Authentication of Human Cell Lines: Standardization of Human Cell Lines (ASN-0002). This standard calls for the use of short tandem repeat (STR) genotyping of human cell lines to confirm identity. STR loci have been used for more than a decade for human forensic and paternity testing because they offer a high degree of discrimination.
The standard recommends a minimum of eight specific STR loci, plus a gender marker. Promega offers a variety of products that meet the standard. These systems can be used to uniquely identify human cell lines derived from different individuals. Because these systems use human-specific primers, they can be used only for human cell lines.
GEN: What tests can be performed for cell-line authentication?
Dr. Castro: Comparative STR analysis can be used if the genotypes for parental cells already have been determined by a recognized repository (ATCC, DSMZ, or JCRB).
Using the repositories, the genotype of the submitted cells can be compared to genotypes in the databases. This approach allows verification of the authenticity of the cell line. The diversity of alleles, repeating segments of DNA two to six base pairs in length that have a diverse number of repeats at a single locus, makes it possible to generate unique genetic profiles by combining analysis results for a number of marker loci. To determine gender, the amelogenin marker found on the X and Y chromosome is routinely used.
Ms. Hall: Historically, cell-line authentication testing had been performed using isoenzyme analysis and karyotyping, among other methods. However, these methods do not allow for the unambiguous identification of the cell line down to the individual human level.
STR DNA profiling has been used for cell-line authentication since the late 1980s and is a widely accepted method for the purpose of human DNA identification. In addition to being the recommended technology by ANSI/ATCC for cell-line authentication, the major advantages to STR profiling include cost efficiency, the availability of universal databases for reference profile comparisons, and the ability to detect low levels of contaminating cells within a single cell culture.
Mr. Ojo: The recommended test for cell-line authentication is STR profiling, and it is the technique we use at LGC Standards. This technique is extremely discriminative as it targets repeated sequences of DNA that are highly variable among individuals, thus allowing us to effectively distinguish one cell line from another.
Another test which can be used is single nucleotide polymorphisms (SNPs), which looks at sequence variation that arises when a single nucleotide in the DNA sequence is altered. However, as it stands, due to the availability of a comprehensive cell-line reference STR database, and its cost-effective nature, STR profiling remains the gold standard.
Dr. Storts: Promega has partnered with the ATCC to offer a human cell-line authentication service to customers in the United States, Canada, and Puerto Rico. The ATCC will perform STR genotyping on the cells using the PowerPlex® 18D system, review the quality of the resultant genotype, and compare the profile against cell lines in their database. Alternatively, laboratories can contact their local core facility. Researchers outside of North America can partner with their local core facility or a variety of genotyping service providers to perform the STR analysis.
GEN: In your opinion, what has been the main source(s) of cell-line misidentification?
Dr. Chang: Although cell-line misidentification may arise from mislabeling, it is often caused by cross-contamination of another cell line into the culture of the originator cell line. Contamination can occur from poor cell culture technique (e.g., nonsegregated biological cabinets and media).
Growth conditions, whether current or further downstream, can favor the contaminant cells, resulting in overgrowth by the contaminant cells. The greatest risks are faced by cell banks with poorly documented histories. These may be derived from uncontrolled sources with minimal or no records as to original sources, the number of subcultures, media, additive traceability, or storage conditions.
This concern over misidentification of cells in the research community prompted an open letter to the Department of Health and Human Services in 2007 by subject matter experts recommending approaches.
Ms. Hall: Cell-line misidentification and cross-contamination were first acknowledged by the biomedical research community nearly 50 years ago. Current research estimates that 18–36% of cell line cultures are affected.
HeLa cells, which are extremely aggressive and fast growing, are associated with mass cell culture contamination, including contamination via aerosol droplets. Another source is human error—a simple flask mislabeling could cause misidentification of the cell line in use.
Not following stringent cell culture rules can also perpetuate the problem; sharing media and other lab supplies among multiple cell lines risks contamination.
Also, sharing cell lines between laboratories could be perpetuating the problem. You could unknowingly receive a misidentified cell line. Failure to authenticate regularly is another source, and problems can be caught by regular testing.
Dr. Letham: Progress is being made in human and nonhuman cell-line characterization with an increased number of journals requiring authentification before publishing results. This is because many researchers do not classify their cells on a regular basis, even when obtained from a biological resource center.
In addition to the classic issues of cross-contamination (HeLa, etc.) and non-subspecies-level testing, the lack of good documentation for cell lines (often passed researcher to researcher) has been a main source of misidentification. Researchers need to be educated on the level of host-cell identification that they need and how to interpret the particular assays for species-level or subspecies-level identification so that they may look for unique identifiers that would establish a profile for each cell line.
GEN: Have companies and/or oversight bodies done enough to reduce or eliminate misidentification?
Dr. Chang: It is important to note that regulatory agencies worldwide mandate that cell lines used in the biopharmaceutical industry undergo mandatory testing including cell-line identification. Validated analytical tests used usually include isoenzyme analysis, karyotyping, and DNA fingerprinting, although new molecular techniques such as STR may also be applicable.
Because of these stringent cGMP practices, studies have shown the incidence of misidentified cell lines in biopharma is very low in direct contrast to the incidence in research. More than 60 years of cell-line-characterization testing performed at BioReliance suggests that the incidence of a misidentified cell line used in biopharma can be classified as an extraordinarily rare event.
Dr. Letham: More work is always needed to reduce misidentification, including supporting efforts to standardize procedures. ANSI’s ASN-0002 guidance for STRs pertains not only to human cell lines, but also new STR profiling protocols being developed by NIST for important African Green Monkey, murine, and CHO lines.
New searchable databases and expanded curated libraries (ATCC.org, NCBI/NIST initiative) should require proper extensive documentation, as multiple entries may cause misleading database search result interpretation. One cell-line notation may represent multiple lines including different recombinant genomes and altered STR profiles due to changes from passage in culture. Researchers and submitters of cell-line-derived products should strive to use multimodal characterization schemes (multilocus and recombinant sequencing, STR analysis, isoenzymes, etc.).
Dr. Schmitt: When asked, the majority of research groups believe that about 10% of cell lines are misidentified (unpublished data). At the same time, less than 35% of research groups have authenticated their cell lines in the past.
The underlying problem appears to be recognized, but it also seems that that certain factors prevent 65% of research groups from authenticating their cell lines. While inexpensive authentication services are now available (prices below €90 or roughly $120 per cell line), the reluctance to authenticate must be seen either in false confidence in the quality of one’s own cell lines or too little pressure from authorities or journals.
While cell-line authentication is increasingly required by scientific journals, it is rarely demanded by granting agencies.
When using misidentified or false cell lines, the best-case scenario is usually seen when the false cell line shares the same origin as the cell line with which the researcher thinks he or she is working. For example, if a breast cancer cell line is confused with another breast cancer cell line, then there is a chance that obtained research results remain valid. Conversely, in vitro testing of novel breast cancer drugs with colon cancer cell lines will have equivocal outcomes.
At Multiplexion, we observe false cell lines in 11.2% of cell lines that were used in ongoing research projects.