There was an error in the characteristic information regarding two cell lines published on our website. Both cases were due to errors in the characteristic information received from the depositors. We did report and apologize to users (recipients) of each cell line when we became aware of the error.
(1) HD-Mar2
HD-Mar2 (RCB1981) is a cell line established overseas, but it was deposited and distributed at the Tohoku University Cell Bank. In 2004, HD-Mar2 was deposited from the Tohoku University Cell Bank to the RIKEN Cell Bank, and has been provided as a “Hodgkin’s lymphoma-derived cell line”, which was the characteristic information received from the Tohoku University Cell Bank.
The Swiss Institute of Bioinformatics has been analyzing the characteristic information of various cell lines used in the research community based on information from published papers, etc., and publishing a database called “Cellosaurus” (below).
https://www.cellosaurus.org/
In August 2022, we learned that HD-Mar2 was described as a misclassified cell line in Cellosaurus. After confirming the cell establishment paper (Int J Cancer 25: 583-590 (1980), PubMed ID: 6154663) and other papers using this cell line, it was found that the cell line should have been described as below.
A cell line derived from the pleural effusion of a patient with a recurrent T-lymphoblastic lymphoma after treatment of the originally diagnosed Hodgkin’s lymphoma. The characteristics is T cell lymphoma.
(2) KE-97
KE-97 (RCB1435) is a cell line that was deposited in 1998, and has been provided as a “cell line derived from intraperitoneal (mesenteric) metastasis of human gastric cancer” registered by the depositor who generated KE-97.
In December 2022, we learned in the previously mentioned “Cellosaurus” that KE-97 was described as a misclassified cell line (although it has been used as a human gastric cancer cell line, it was derived from human B cells). Cellosaurus based it on the following paper published in 2018.
Profiling the B/T cell receptor repertoire of lymphocyte derived cell lines. BMC Cancer 18: 940 (2018), PubMed=30285677
The above paper confirms the rearrangement of B cell receptors in KE-97 cells, indicating that KE-97 cells are most likely derived from B cells.
We analyzed the expression of CD19 and CD20, which are B cell markers, and found that KE-97 cells clearly expressed both CD19 and CD20, indicating that the KE-97 cells were most likely derived from B cells.
Consultation with the depositor who generated KE-97 cells:
After communicating all of the above information to the depositor and consulting with them, we have decided to discontinue the distribution of KE-97 cells. (The depositor had long retired from research and did not possess the KE-97 cells.)
Regarding prevention of recurrence and future measures:
Regardless of the cell characteristic information obtained from the depositors, we always conduct the following tests before providing.
Mycoplasma infection inspection:
This test has been conducted since the inception of the cell bank (1987), and although it takes time and effort, the test is performed using the most sensitive DNA staining method used by all major cell banks around the world. We use the PCR test method as well.
Misidentified cell exclusion test for human cells (individual identification):
About 20 years ago, the usefulness of STR polymorphism analysis was published in collaboration with major cell banks around the world, and major cell banks around the world introduced STR polymorphism analysis as a routine test. We continue to use it as a global standard routine test.
Misidentified cell exclusion test for mouse cells (strain identification):
Mouse cells are the second most common cell line after human cells. Many mouse cells are derived from inbred mice, so it is necessary to verify whether the strain name registered by the depositor is correct. We developed STR polymorphism analysis for mouse cells and introduced it as a routine test.
Animal species identification test:
When our cell bank was first established (1987), the isoenzyme test, which was the standard test at that time, was performed. After that, we introduced a mitochondrial gene PCR identification test as a more accurate molecular biological test. Furthermore, a more rigorous identification test based on DNA sequencing (DNA Barcoding method) was established, introduced in 2018, and misidentified 4 cell lines (discrepancies with depositor registration information) were discovered and published. With the introduction of the DNA Barcoding method, identification of the animal species of origin has become almost certain.
If mycoplasma contamination or misidentification is found in the above tests, we will contact the depositor and ask them to resend another cell. However, there are many cell lines that cannot be resent (they are no longer owned by the depositor) or that resending does not resolve the issue, making distribution from the cell bank impossible.
Mycoplasma contamination tests and misidentified cell exclusion tests (STR polymorphism analysis) are the basis of cell quality tests, but they are very basic and can only test a small part of cell characteristics. Of note, STR polymorphism analysis is a test to identify “originating individual” or “originating mouse strain”, and cannot identify errors such as (1) and (2).
By conducting detailed analysis of cell characteristics in the research community, it will become clear after some time after deposit that the cell characteristics (derived cancer type, derived cell type, etc.) are different from the registered information of the depositor. There may be more cases in the future. For this reason, we will actively collect information from the research community, etc., and if cases such as (1) or (2) are confirmed, we will respond appropriately in the same way as this time.
Gene expression analysis (RNA seq) is a candidate method to prevent above cases in cell banks. Comprehensive gene expression analysis, hierarchical clustering, principal component analysis, etc. can verify the derived cell type to some extent. However, statistically significant verification is not possible unless analysis of many cell lines is performed for each cell type (for example, each cancer type and each tissue type). Advances in gene expression analysis technology are remarkable, and thus in the event that significant cost reductions are realized in the future, we plan to conduct gene expression analysis targeting cell types with a sufficient number of cell lines to be analyzed.
We would like to ask for your continued understanding and support for the activities of the RIKEN BioResource Research Center.