Cell Line Authentication Using Isoenzyme Analysis

Confirmation of purity and identity of cell cultures is a necessary step in the production of biotherapeutics. Manipulation of multiple cell lines in the same facility introduces the possibility that cross-contamination may occur. Different cell lines may proliferate with varying growth rates such that a single cell from a rapidly growing line, introduced into a culture of slower growing cells, can overtake the original culture in the course of a few passages. In addition, cultures may be mislabeled during manipulation, again resulting in a misidentification of the culture.

Raymond W. Nims, Ph.D

The need to verify the purity of cell lines on an ongoing basis is critical.¹ In some cases it may be possible to visually identify a contaminating colony of cells in an otherwise pure culture. In many cases, however, an unintentional co-culture cannot be determined by visual inspection. It is therefore necessary to repeatedly verify the identity and purity of cell cultures when used as substrates for the manufacture of biotechnology products. This is commonly done at the master and working cell bank levels. Pre-bank and end-of-production cultures may also be evaluated.

Isoenzyme analysis is commonly performed as part of cell line authentication and characterization. It often is part of an overall testing battery for authentication of seed stocks submitted to cell repositories.^2,3 In addition, authentication is typically performed on master and working cell banks as part of US and EU's mandatory overall characterization testing.^4,5 Because isoenzyme analysis is technically simple, robust, and rapid, it is used in conjunction with, or often in place of, karyotyping or DNA fingerprinting. With commercially available kits, it only takes a few hours for an investigator to confirm a cell line's purity and species of origin.

Karyotyping and DNA fingerprinting are useful for these purposes but can be expensive, time consuming, and may require comparison with a karyotype or fingerprint taken from the original culture. These techniques may be required to establish identity at the individual cell line level. For routine speciation and assessment of purity, however, the relatively rapid and inexpensive isoenzyme analysis method is commonly employed as a useful alternate.

METHOD OVERVIEW

The isoenzyme analysis method utilizes electrophoretic banding patterns to examine slight differences from species to species in the structure and mobility of individual isoforms for a number of intracellular enzymes. We perform assays with a commercially available kit, the AuthentiKit system, manufactured by Innovative Chemistry, Inc.

⁶

The intracellular enzymes typically evaluated in speciation are nucleoside phosphorylase (NP), malate dehydrogenase (MD), glucose-6-phosphate dehydrogenase (G6PD), lactate dehydrogenase (LD), peptidase B (PepB), aspartate amino transferase (AST), and mannose 6-phosphate isomerase (MPI). Reagents for detection of these enzymes are provided as part of the kit. Also provided is a standard reagent, murine L929 cell extract, to be loaded onto each gel, which allows the migration data of the test sample to be corrected for day-to-day test variability.

The kit manufacturer offers a chart of standardized electrophoretic migration distances for various intracellular enzymes. This listing contains values for 25 animal species. Using this chart and a systematic process of elimination, species assignments for test samples can be made on the basis of the corrected electrophoretic migration distances.

The kit also includes a control reagent (human HeLa extract), which is loaded onto each gel and is used to determine system suitability. If the corrected migration distance for the control band in a particular gel is different by more than 2 mm from the expected migration distance provided by the AuthentiKit, the gel fails system suitability.

TESTING STRATEGIES FOR SPECIATION

Accurate determination of species of origin for test cell extracts is dependent on the operator's ability to clearly resolve and determine migration distances for the bands associated with the individual isoforms of the assay's standard and control reagents and the test cell extracts. The color development of each gel must be monitored and the development reaction stopped at the appropriate time, in a manner analogous to image development in black and white photography. Figure 1 demonstrates what happens when gels with enzyme reagents are left to incubate too long. It becomes difficult to resolve individual bands in gels for the enzymes where multiple isoforms are expected, such as LD and PepB.

Figure 1. The Effect of Excessive Color Development on Band Resolution in the Electrophoresis of Innovative Chemistry's Standard and Control Reagents (Extracts of Mouse L929 and Human HeLa cells, Respectively)

The test cell extract migration distances should be corrected for day-to-day variability. This is done by calculating a correction factor for each gel on the basis of the observed and expected migration distances for the standard reagent. Test-article band-migration distances are then multiplied by the same correction factor. Performance of the isoenzyme analysis assay in this manner should optimize the accuracy of the method for determining cell line species of origin.

In the course of evaluating nearly 900 cell submissions, we have detected only two misidentifications and two interspecies co-cultures. This surprisingly low frequency of unexpected findings is a reflection, in part, of the type of testing done at our testing facility. Typically, our submissions consist of GMP working and master cell banks, end-of-production cells, and only a few pre-bank and research samples. It would appear that GMP practices implemented in the biotechnology industry have helped to ensure that misidentifications are rare.

DETECTING CROSS-CONTAMINATION

Interspecies cell line cross-contamination, leading to relatively stable mixed populations of the original and contaminating cell lines, is a concern for laboratories performing tissue culture using two cell types that possess similar growth characteristics (i.e., doubling time and degree of contact inhibition). It is possible to detect such cases by means of the isoenzyme analysis method,

^3,7

provided the appropriate enzymes are evaluated and sufficient proportions of the two cell types exist in the stable mixed cultures.

Figure 2. Species of Origin for Test Cell Extracts by Frequency of Submission Over a Seven-year Period

By far, the most commonly employed cell lines in the biotechnology industry originate from three animal species: Chinese hamster, mouse, and human. Figure 2 displays the frequency, by species of origin, of cell lines evaluated at our testing facility over the past seven years. Cross-contaminations involving any of these three cell lines likely represent the greatest risk. As shown in Table 1, specific enzymes may be required, or may be optimal, for detecting cell line cross-contamination with specific mixtures of animal species. Consider this information when performing the analyses and interpreting the results.

Table 1. Enzyme Selection for Detection of Cell Line Cross-Contamination with Isoenzyme Analysis

A striking example involves the mixture consisting of Chinese hamster and mouse cells (Figure 3).⁸ The enzyme PepB works best to definitively differentiate the cells of the two species. It is apparent from the figure that only PepB will detect a stable cross-contamination of the two species. The enzyme LD is not generally useful in this case because a single band or more than one band may be displayed for each species, depending on the specific cell line being analyzed.

Figure 3. Electrophoresis of Extracts of the Indicated Proportional Mixtures of Mouse (L929) and Chinese Hamster (CHO-K1) Cells.

Isoenzyme analysis also has the potential for detecting the early stages of interspecies cell line cross-contamination (i.e., the period of time between the initial contamination event and the eventual replacement of the original culture with the contaminating cell line). This can occur in cases where differences in growth parameters favor one of the cell types over the other. In 1992 it was estimated that the extent of the total cell population represented by contaminating cells must approach 25% before the presence of two species within the cell culture would be detected by isoenzyme analysis.⁹ Newer studies have demonstrated that when the contaminating cells represent about 10% of the total population, the cross-contamination can be detected in one or more of the gels evaluated by isoenzyme analysis (Figures 3, 4A).⁸

Figure 4. Electrophoresis

Evaluations of mixtures involving a fast growing Chinese hamster cell line (CHO-K1) and a slower growing human cell line (MRC-5) revealed that extracts prepared from mixtures of the two cell types resulted in visualization of distinct LD bands for each species, when each cell type represented at least 11% of the total number of cells extracted (Figure 4A). In a scenario more directly analogous to a cross-contamination event, a small number (~100) of CHO-K1 cells were deliberately inoculated into a confluent 75-cm2 flask containing human MRC-5 cells. The presence of the Chinese hamster cells was detected in the human cell culture evaluating the LD enzyme within two passages of the resulting culture (Figure 4B).

Figure 5. Electrophoresis of Extracts from Cercopithecus Monkey (Vero) cells; Mixture (a Suspected Co-culture of the Two Cell Lines); and Human (MRC-5) Cells

ISOENZYME ANALYSIS IN ACTION

We shall present two examples of actual cell line cross-contamination detection using isoenzyme analysis. In the first case, the contamination of a slow growing human diploid cell line (MRC-5) with a fast growing cercopithecus monkey cell (suspected to be Vero cells) was apparent from microscopic examination of the culture. An analysis of the cells harvested from this mixed culture (Figure 5) revealed the expected bands for the human cells. Upon inspection of the AST gel, an additional band was revealed with a migration distance consistent with that of cercopithecus monkey. The MD gel for the culture also revealed a doublet for the mitochondrial band, i.e., that migrating close to the gel origin. One of these bands was in a region typical for human, while the other was characteristic for cercopithecus monkey (based on historical data). These results demonstrate that for detection of cross-contaminations involving human and cercopithecus monkey cells, the enzyme AST is especially useful. In the absence of AST data, a doublet for the mitochondrial band of MD enzyme should be regarded as an indication that a contamination may be present and that performance of an AST gel is required (Table 1).

Figure 6. Morphologic Appearance of an Apparent Co-culture of Bovine (BT) and Cercopithecus Monkey Cells

In a second case, the microscopic appearance of a bovine turbinate (BT) cell culture suggested the presence of a second cell type (Figure 6). On the basis of the morphology of the contaminating cells and knowledge of the other cell lines being used in the laboratory, it was suspected that the contaminant was the cercopithecus monkey cell line Vero. This was borne out by the results of the isoenzyme analysis (Figure 7). In this case, the presence of the contaminant was evident in the gels for each enzyme evaluated. Measurement of the migration distances for the bands not attributed to the bovine cell enzymes confirmed the presence of cercopithecus monkey cells.

The animal species of origin of the cell line being characterized, as well as the potential species representing the greatest risk of contamination, should be considered when selecting the enzymes to be evaluated. Interspecies cell line cross-contamination may be readily detected in certain cases (e.g., bovine cells contaminated with cercopithecus monkey cells), with the extra (contaminant) bands appearing in gels for most of the enzymes evaluated. In other cases, such as a mixture of Chinese hamster and mouse cells, only specific enzymes will differentiate the two animal species sufficiently to enable detection of a co-cultivation or total replacement of one cell with another. Optimization of the testing strategy with regard to the factors discussed earlier (color development time, correction of migration distances for each gel on the basis of the migration of the standard reagent, and selection of the appropriate enzymes) will provide the greatest assurance that this method will accurately confirm the species of origin and purity of the cell culture being characterized.

Figure 7. Electrophoresis of Extracts from Mouse (L929), Human (HeLa), Chinese Hamster (CHO-K1), and a Suspected Mixture of Bovine (BT) and Cercopithecus Monkey (Vero) Cells

Raymond W. Nims, Ph.D. is a lab director, Biologics Testing Division, BioReliance, Invitrogen bioservices, 14920 Broschart Road, Rockville, MD 20850, 301. 610.2778, fax: 301. 738.1036, ray.nims@invitrogen.com.

Christopher J. Herbstritt is a lab director, Biologics Testing Division, BioReliance, Invitrogen bioservices, 301.610.2256, fax: 301.738.1036, chris.herbstritt@invitrogen.com.

REFERENCES

1. Markovic O, Markovic N. Cell cross-contamination in cell cultures: The silent and neglected danger.

In Vitro Cell. Dev. Biol

. 1998; 34:1-8.

2. Hay RJ. The seed stock concept and quality control for cell lines. Anal.Biochem. 1988; 171:225–237.

3. Steube KG, et al. Isoenzyme analysis as a rapid method for the examination of the species identity of cell cultures. In Vitro Cell. Dev. Biol. 1995; 31:115–119.

4. CBER. Points to Consider in the Characterization of Cell Lines Used to Produce Biologicals. Bethesda, MD: Office of Biologics Research and Review, FDA; 1993.

5. ICH. Guideline on Quality of Biotechnological/Biological Products: Derivation and Characterization of Cell Substrates Used in the Production of Biotechnological/Biological Products. Q5D. Geneva, Switzerland: International Conference on Harmonization; 1997.

6. Innovative Chemistry, Inc. The AuthentiKit System. Handbook for Cell Authentication and Identification. 2nd ed., Marshfield, MA; 1988.

7. Halton DM, et al. Cell culture quality control by rapid isoenzymatic characterization. In Vitro 1983; 19:16–24.

8. Nims RW, et al. Sensitivity of isoenzyme analysis for the detection of interspecies cell line cross-contamination. In Vitro Cell. Dev. Biol. 1998: 34:35–39.

9. Hay RJ, et al. Reply to "Cells, lines, and DNA fingerprinting" In Vitro Cell. Dev. Biol. 1992; 28A: 593–594.