Since we started cell repository work in 1987, we had examined the animal species origin of deposited cell lines by the conventional isozyme analysis of two different isozymes, lactate dehydrogenase and nucleotide phosphorylase, similarly to other cell repositories such as ATCC in USA. This isozyme analysis depends on the biochemical features of the enzymes that are commonly present in several animal species but show different pattern in electrophoresis. However, in 2015 we found that this isozyme analysis was not sufficient to identify the animal species origin of deposited cell lines (please refer to URL below).
Therefore, we introduced a robust molecular method which is the PCR analysis to detect mitochondrial genes (mitochondria PCR analysis), and we finished this analysis by 2016. This analysis is based on the difference of mitochondrial genes (12S ribosomal RNA, 16S ribosomal RNA or cytochrome b) and to identify animal species origin by PCR products using species-specific primers (Ref. 1, Figure 1). However, this analysis was developed for 14 animal species from which relatively large number of cell lines in cell repositories were derived or of which demands were high (Ref. 1), and the origins of 61 cell lines in our cell bank derived from 26 other animal species remained unconfirmed.
In order to identify the origin of more animal species, a method based on the DNA sequence of approximately 600bp PCR product derived from mitochondrial cytochrome c oxidase subunit 1 (COI, indicated with green rectangle in Figure 1) was developed (COI sequence analysis). This analysis is based on the database of COI sequences which are publicly disclosed for example by National Center for Biotechnology Information (NCBI) (Ref. 2). This analysis became possible by the accumulation of COI sequences of many animal species in publicly available databases following the development of DNA sequence technology with relatively low cost. In 2017, we introduced COI sequence analysis for the cell lines of which we could not confirm the animal species with mitochondria PCR analysis. However, for 8 cell lines derived from 3 animal species the accumulation of COI sequences in NCBI etc. was not sufficient enough to identify the origin. We searched databases for these cases and found that DNA sequence of mitochondrial cytochrome b gene was useful for the identification of origin.
As the result, with respect to 57 cell lines out of 61 cell lines the results were identical to those registered by depositors (Table 1). However, the origins of 4 cell lines were different from those registered by depositors as below.
① Hepa-T1 derived from “Tilapia” (Figure 2)
Hepa-T1 (RCB1156) was deposited in 1995 by a scientist of foreign country as a cell line derived from “Tilapia”.
As shown in Figure 2, the COI sequence of Hepa-T1② is quite different from that of tilapia registered in NCBI① (concordance % is 79.8), but 99.8% identical to that of eel registered in NCBI⑤ and also to that of Hepa-E1 (RCB1155)④ deposited as eel origin in our center.
We also tested the cells cryopreserved immediately after the deposition③, and the result was same (Figure 2) indicating that this misidentification occurred prior to the deposition to our center.
Of note, both Hepa-T1 and Hepa-E1 cell lines were deposited simultaneously by the same scientist. At the deposition, Hepa-E1 was described as a cell line derived from Eel. Therefore, it is highly likely that Hepa-T1 was replaced with Hepa-E1 in the depositor’s lab without being noticed before deposition. Many similar cases of misidentification have occurred in other cell lines as well (Ref. 3)
② FRI-SpIm-1229 derived from “Lemyra imparilis” (Figure 3)
FRI-SpIm-1229 (RCB0281) was deposited in 2000 by a scientist in Japan as a cell line derived from “Lemyra imparilis” (a kind of moth).
COI sequence of Lemyra imparilis is not available from publicly available databases at the moment. However, as show in Figure 3 the COI sequence of FRI-SpIm-1229① is 99.5% identical to that of “Mamestra brassicae”③ (another kind of moth).
We also tested the cells cryopreserved immediately after the deposition②, and the result was almost same (Figure 3) indicating that this misidentification occurred prior to the deposition to our center.
③ LAH1 and LAH2 cell lines derived from frog (Figure 4, Figure 5, Figure 6)
LAH1 (RCB1733) and LAH2 (RCB1735) were deposited in 2001 by a scientist in Japan as cell lines derived from frog. In addition, LAH3 (RCB1734) was also deposited simultaneously as a cell line derived from frog by the same scientist.
With regard to frog species, registration of COI sequences is not sufficient to distinguish them precisely. However, registration of sequences of mitochondrial cytochrome b gene is sufficient to distinguish them (Ref. 4). Therefore, we performed sequence analyses of cytochrome b genes of LAH1, LAH2 and LAH3 cell lines.
LAH1 (②③ in Figure 5)
Information registered by the depositor: Wild-type of “Pelophylax nigromaculatus”
The sequence of cytochrome b gene of LAH1② is not identical to that of “Pelophylax nigromaculatus” (① in Figure 5) (concordance % is 74.5), but 99.5% identical to that of “Bufo japonicas formosus” (④ in Figure 5).
LAH2 (②③ in Figure 6)
Information registered by the depositor: Albino of “Pelophylax nigromaculatus”
The sequence of cytochrome b gene of LAH2 is not identical to that of “Pelophylax nigromaculatus” (① in Figure 6) (concordance % is 90.5), but 100% identical to that of “Pelophylax porosus brevipodus” (⑥ in Figure 6).
As described above, the origins of LAH1 and LAH2 were different from those registered by the depositor
We also tested the cells cryopreserved immediately after the deposition, and the results were same (Figure 5, Figure 6) indicating that LAH1 was derived from “Bufo japonicas formosus” and LAH2 was derived from “Pelophylax porosus brevipodus” when they were deposited to our center.
On the other hand, the origin of LAH3 was that registered by the depositor as below.
LAH3 (④⑤ in Figure 6)
Information registered by the depositor: Albino of “Pelophylax porosus brevipodus”
The sequence of cytochrome b gene of LAH3④ is 100% identical to that of “Pelophylax porosus brevipodus” (⑥ in Figure 6).
The results of LAH2 and LAH3 were identical. Since all three cell lines (LAH1, LAH2, LAH3) were deposited by the same scientist as mentioned already, we cannot deny a possibility that LAH2 was replaced with LAH3 in the depositor’s lab without being noticed before deposition.
We understand that this misidentification might cause serious scientific damage. But, it is our hope that the damage is the minimum. Please accept our sincere apologies and we ask your kind understanding.
A newly developed technology sometimes reveals mistake, mishandling, or misidentification that were made in the past and results in uncomfortable outcomes. Although the outcomes may be as painful as could be, we believe it is a necessary process for proper development of sciences. Therefore, we will continuously pursue a higher quality control of our cell lines adopting the newly developed technologies to serve better to the scientific community. We appreciate you for your support of our activities.
1. Ono, K. et al. Species identification of animal cells by nested PCR targeted to mitochondrial DNA. In Vitro Cell Dev Biol Anim 43: 168-175 (2007)
2. Almeida, J.L., Cole, K.D., and Plant, A.L. Standards for Cell Line Authentication and Beyond. PLoS Biol 14: e1002476 (2016).
3. Yoshino, K. et al. Essential role for gene profiling analysis in the authentication of human cell lines. Hum Cell 19: 43-48 (2006)
4. Huang, Z., Yang, C., and Ke, D. DNA barcoding and molecular phylogeny in Ranidae. Mitochondrial DNA A DNA Mapp Seq Anal 27: 4003-4007 (2016)
Yukio NAKAMURA, M.D., Ph.D.
Cell Engineering Division (Cell Bank)
RIKEN BioResource Research Center (RIKEN BRC)