Preclinical in vitro evaluation of immune suppression induced by GYM329, Fc-engineered sweeping antibody.

Fc-engineering is commonly used to improve the therapeutic potency of antibody (Ab) treatments. Because FcγRIIb is the only inhibitory FcγR that contains an immunoreceptor tyrosine-based inhibition motif (ITIM), Fc-engineered Abs with enhanced binding affinity to FcγRIIb might provide immune suppression in clinical contexts. GYM329 is an anti-latent myostatin Fc-engineered Ab with increased affinity to FcγRIIb which is expected to improve muscle strength in patients with muscular disorders. Cross-linking of FcγRIIb by immune complex (IC) results in phosphorylation of ITIM to inhibit immune activation and apoptosis in B cells. We examined whether the IC of Fc-engineered Abs with enhanced binding affinity to FcγRIIb causes phosphorylation of ITIM or B cell apoptosis using GYM329 and its Fc variant Abs in human and cynomolgus-monkey (cyno) immune cells in vitro. IC of GYM329 with enhanced binding affinity to human FcγRIIb (×5) induced neither ITIM phosphorylation nor B cell apoptosis. As for GYM329, FcγRIIb should work as an endocytic receptor of small IC to sweep latent myostatin, so it is preferable that GYM329 induces neither ITIM phosphorylation nor B cell apoptosis to prevent immune suppression. In contrast, IC of myo-HuCy2b, the Ab with enhanced binding affinity to human FcγRIIb (×4), induced ITIM phosphorylation and B cell apoptosis. The result of the present study demonstrated that Fc-engineered Abs with similar binding affinity to FcγRIIb had different effects. Thus, it is important to also investigate FcγR-mediated immune functions other than binding to fully understand the biological effects of Fc-engineered Abs.

Priming treatment with T-cell redirecting bispecific antibody ERY974 reduced cytokine induction without losing cytotoxic activity in vitro by changing the chromatin state in T cells.

CD3 bispecific constructs are anticipated to become an important form of cancer immunotherapy, but they frequently cause cytokine release syndrome (CRS) that is difficult to manage in clinical contexts. A combination of intra-patient dose escalation and immunosuppressive treatment is widely used to mitigate CRS. Studies suggest that CRS after subsequent doses of CD3 bispecific constructs is less severe than after the priming dose, and that step-up dosing reduces cytokine levels in animals and humans. However, the mechanism underlying the reduced cytokine induction after priming treatment with CD3 bispecific constructs is unclear. To understand human T-cell activation and chromatin states after priming treatment with CD3 bispecific construct targeting CD3ɛ and glypican 3 (ERY974), we examined cytokine levels, cytokine mRNA expression, CD3ɛ expression, CD3-mediated signal transduction, T cell activation markers, cytotoxicity against target cells, and chromatin states in T cells after ERY974 priming treatment or negative control. The second ERY974 treatment decreased cytokines on Day 8, and ERY974 priming treatment changed the chromatin state in T cells. CD3ɛ expression, CD3-mediated signal transduction, T cell activation markers, and cytotoxicity were similar between the priming treatment with ERY974 and negative control. The present study suggests that chromatin state changes in T cells after the priming treatment was a pivotal factor in the mitigation of cytokine release after the second ERY974 treatment.

In vitro toxicological support to establish specification limit for anti-CD3 monospecific impurity in a bispecific T cell engager drug, ERY974.

An emerging structure for anti-tumor antibody drugs utilizes a bispecific antibody (BiAb) that recognizes a tumor surface antigen and CD3 on T cells. An impurity that commonly contaminates these BiAb products is an anti-CD3 monoclonal antibody (mAb). The most plausible cause of toxic activity by an anti-CD3 mAb is the induction of cytokines via T cell activation. In this in vitro study, we compared cytokine induction and T cell activation after treatment with an anti-glypican-3/CD3 BiAb (ERY974), anti-CD3 mAb impurity (aCD3), or ERY974 spiked with 5% aCD3. We found that contamination with up to 5% aCD3 did not affect cytokine release by ERY974. Cytokine levels induced by ERY974 in the presence of target cells were significantly higher than those induced by aCD3, but were very similar to those by the spiked treatment. The results supported the specification of a 5% limit for aCD3. OKT-3 had much higher activity to induce cytokines from peripheral blood mononuclear cells in an in vitro assay than aCD3. This suggests that specification limit should be decided for each type of anti-CD3 impurity that affects T cell-activating BiAb drug products. In vitro cytokine assays can provide useful information for determining these specification limits.

Daily ascending dosing in cynomolgus monkeys to mitigate cytokine release syndrome induced by ERY22, surrogate for T-cell redirecting bispecific antibody ERY974 for cancer immunotherapy.

CD3 bispecific constructs show promising therapeutic potential as anti-tumor antibodies, but it has concurrently been difficult to manage cytokine release syndrome (CRS) in clinical use. Currently, the most effective measure for reducing CRS is considered a combination of intra-patient/animal dose escalation and corticosteroid premedication. To examine how effectively an intra-animal ascending dose regimen without premedication would mitigate CRS, we compared plasma cytokine levels in two groups of cynomolgus monkeys; one group was given a single dose, and the other a three-fold daily ascending dose of a CD3 bispecific construct that targets and cross-reacts with both glypican 3 and CD3 (ERY22). Ascending doses up to 1000 µg/kg of ERY22 dramatically reduced the peak cytokine levels of IL-6, TNF-α, and IFN-γ, IL-2 as well the clinical severity of CRS compared with a single dose of 1000 µg/kg. Peak cytokine levels following the single and ascending doses were 60,095 pg/mL and 1221 pg/mL for IL-6; 353 pg/mL and 14 pg/mL for TNF-α; 123 pg/mL and 16 pg/mL for IFN-γ; and 2219 pg/mL and 42 pg/mL for IL-2. The tolerance acquired with daily ascending doses up to 1000 µg/kg remained in effect for the following weekly doses of 1000 µg/kg.

High-content imaging analyses of γH2AX-foci and micronuclei in TK6 cells elucidated genotoxicity of chemicals and their clastogenic/aneugenic mode of action.

BACKGROUND: The in vitro micronucleus (MN) test is an important component of a genotoxicity test battery that evaluates chemicals. Although the standard method of manually scoring micronucleated (MNed) cells by microscope is a reliable and standard method, it is laborious and time-consuming. A high-throughput assay system for detecting MN cells automatically has long been desired in the fields of pharmaceutical development or environmental risk monitoring. Although the MN test per se cannot clarify whether the mode of MN induction is aneugenic or clastogenic, this clarification may well be made possible by combining the MN test with an evaluation of γH2AX, a sensitive marker of DNA double strand breaks (DSB). In the present study, we aimed to establish a high-content (HC) imaging assay that automatically detects micronuclei (MNi) and simultaneously measures γH2AX foci in human lymphoblastoid TK6 cells. RESULTS: TK6 cells were fixed on the bottom of each well in 96-well plates hypotonically, which spreads the cells thinly to detach MNi from the primary nuclei. Then, the number of MNi and immunocytochemically-stained γH2AX foci were measured using an imaging analyzer. The system correctly judged 4 non-genotoxins and 13 genotoxins, which included 9 clastogens and 4 aneugens representing various genotoxic mechanisms, such as DNA alkylation, cross-linking, topoisomerase inhibition, and microtubule disruption. Furthermore, all the clastogens induced both γH2AX foci and MNi, while the aneugens induced only MNi, not γH2AX foci; therefore, the HC imaging assay clearly discriminated the aneugens from the clastogens. Additionally, the test system could feasibly analyze cell cycle, to add information about a chemical's mode of action. CONCLUSIONS: A HC imaging assay to detect γH2AX foci and MNi in TK6 cells was established, and the assay provided information on the aneugenic/clastogenic mode of action.

Advantages of evaluating γH2AX induction in non-clinical drug development.

γH2AX, the phosphorylated form of a histone variant H2AX at Ser 139, is already widely used as a biomarker to research the fundamental biology of DNA damage and repair and to assess the risk of environmental chemicals, pollutants, radiation, and so on. It is also beginning to be used in the early non-clinical stage of pharmaceutical drug development as an in vitro tool for screening and for mechanistic studies on genotoxicity. Here, we review the available information on γH2AX-based test systems that can be used to develop drugs and present our own experience of practically applying these systems during the non-clinical phase of drug development. Furthermore, the potential application of γH2AX as a tool for in vivo non-clinical safety studies is also discussed.

Different players generate positive responses in two in vitro cytokine assay formats with aqueous and immobilized TGN1412 analog.

To detect potential risk of severe cytokine release syndrome, in vitro assay formats with human cells have been developed. The two major testing platforms are a combination of whole blood with aqueous-phase test articles (whole blood cytokine assay, WBCA) and peripheral blood mononuclear cells with solid-phase articles (PBMC assay). Significant induction of cytokines was seen in both assays after treatment with a widely used control agent, TGN1412 or its analog CD28SA, but the WBCA cytokine profile differed from what was expected from clinical experience. In the WBCA, potential risk of CD28SA was detected by elevation of IL-8 whereas IL-2, a key cytokine after stimulation of CD28, was not induced in approximately 40% of donor samples. Therefore, further mechanistic understanding of the different responses in the in vitro assay was needed. In this study of donor samples treated with CD28SA, we compared the induction of cytokines and identified the cytokine-producing cells in the two assays. IL-2 was markedly elevated in all the donors in the PBMC assay but only in 1 of 3 donors in the WBCA. IL-8, the most sensitive biomarker in the WBCA, was produced by monocytes and granulocytes. T cells, the most relevant player in the PBMC assay with CD28SA, did not contribute to the positive response seen in two donors in the WBCA, which suggests that different players caused the positive cytokine responses to CD28SA in the two assays.

An anti-glypican 3/CD3 bispecific T cell-redirecting antibody for treatment of solid tumors.

Cancer care is being revolutionized by immunotherapies such as immune checkpoint inhibitors, engineered T cell transfer, and cell vaccines. The bispecific T cell-redirecting antibody (TRAB) is one such promising immunotherapy, which can redirect T cells to tumor cells by engaging CD3 on a T cell and an antigen on a tumor cell. Because T cells can be redirected to tumor cells regardless of the specificity of T cell receptors, TRAB is considered efficacious for less immunogenic tumors lacking enough neoantigens. Its clinical efficacy has been exemplified by blinatumomab, a bispecific T cell engager targeting CD19 and CD3, which has shown marked clinical responses against hematological malignancies. However, the success of TRAB in solid tumors has been hampered by the lack of a target molecule with sufficient tumor selectivity to avoid "on-target off-tumor" toxicity. Glypican 3 (GPC3) is a highly tumor-specific antigen that is expressed during fetal development but is strictly suppressed in normal adult tissues. We developed ERY974, a whole humanized immunoglobulin G-structured TRAB harboring a common light chain, which bispecifically binds to GPC3 and CD3. Using a mouse model with reconstituted human immune cells, we revealed that ERY974 is highly effective in killing various types of tumors that have GPC3 expression comparable to that in clinical tumors. ERY974 also induced a robust antitumor efficacy even against tumors with nonimmunogenic features, which are difficult to treat by inhibiting immune checkpoints such as PD-1 (programmed cell death protein-1) and CTLA-4 (cytotoxic T lymphocyte-associated protein-4). Immune monitoring revealed that ERY974 converted the poorly inflamed tumor microenvironment to a highly inflamed microenvironment. Toxicology studies in cynomolgus monkeys showed transient cytokine elevation, but this was manageable and reversible. No organ toxicity was evident. These data provide a rationale for clinical testing of ERY974 for the treatment of patients with GPC3-positive solid tumors.

Giemsa-stained pseudo-micronuclei in rat skin treated with vitamin D3 analog, pefcalcitol.

BACKGROUND: Pefcalcitol, an analog of vitamin D3 (VD3), is an anti-psoriatic drug candidate that is designed to achieve much higher pharmacological effects, such as keratinocyte differentiation, than those of VD3, with fewer side effects. Genotoxicity of the compound was evaluated in a rat skin micronucleus (MN) test. RESULTS: In the rat skin MN test, pefcalcitol showed positive when specimens were stained with Giemsa, whereas neither an in vitro chromosome aberration test in CHL cells nor an in vivo bone marrow MN test in rats indicated clastogenicity. To elucidate the causes of the discrepancy, the MN specimens were re-stained with acridine orange (AO), a fluorescent dye specific to nucleic acid, and the in vivo clastogenicity of the compound in rat skin was re-evaluated. The MN-like granules that had been stained by Giemsa were not stained by AO, and AO-stained specimens indicated that pefcalcitol did not increase the frequency of micronucleated (MNed) cells. Histopathological evaluation suggested that the MN-like granules in the epidermis were keratohyalin granules contained in keratinocytes, which had highly proliferated after treatment with pefcalcitol. CONCLUSIONS: Pefcalcitol was concluded to be negative in the rat skin MN test. The present study demonstrated that Giemsa staining gave a misleading positive result in the skin MN test, because Giemsa stained keratohyalin granules.

Is an in vitro whole blood cytokine assay useful to detect the potential risk of severe infusion reaction of monoclonal antibody pharmaceuticals?

After the life-threatening cytokine release syndrome (CRS) occurred in the clinical study of the anti-CD28 monoclonal antibody (mAb) TGN1412, in vitro cytokine release assays using human blood cells have been proposed for non-clinical evaluation of the potential risk of CRS. Two basic assay formats are frequently used: human peripheral blood mononuclear cells (PBMC) with immobilized mAbs, and whole blood with aqueous mAbs. However, the suitability of the whole blood cytokine assay (WBCA) has been questioned, because an unrealistically large sample size would be required to detect the potential risk of CRS induced by TGN1412, which has low sensitivity. We performed a WBCA using peripheral blood obtained from 68 healthy volunteers to compare two high risk mAbs, the TGN1412 analogue anti-CD28 superagonistic mAb (CD28SA) and the FcγR-mediated alemutuzumab, with a low risk mAb, panitumumab. Based on the cytokine measurements in this study, the sample size required to detect a statistically significant increase in cytokines with 90% power and 5% significance was determined to be n = 9 for CD28SA and n = 5 for alemtuzumab. The most sensitive marker was IL-8. The results suggest that WBCA is a practical test design that can warn of the potential risk of FcγR-mediated alemtuzumab and T-cell activating CD28SA but, because there was apparently a lower response to CD28SA, it cannot be used as a risk-ranking tool. WBCA is suggested to be a helpful tool for identifying potential FcγR-mediated hazards, but further mechanistic understanding of the response to CD28SA is necessary before applying it to T cell-stimulating mAbs.

The predominant role of apoptosis in γH2AX formation induced by aneugens is useful for distinguishing aneugens from clastogens.

The phosphorylated form of the histone protein H2AX, called γH2AX, is recognized as a useful biomarker not only for DNA double-strand breaks but also for a wide range of other DNA damage. An increasing number of publications propose γH2AX to be measured when determining genotoxicity, phototoxicity, and the effectiveness of cancer therapy. Because γH2AX is also generated by apoptosis, a γH2AX-assay might assess genotoxic risk incorrectly. The aim of this study was to elucidate the influence of apoptosis on measurements of γH2AX by flow cytometry, with the clastogens mitomycin C (MMC) and etoposide (ETP), and the aneugens vinblastine (VB) and paclitaxel (PT), which do not react directly with DNA. TK6 human lymphoblastoid cells were treated with the clastogens and the aneugens, stained for the apoptotic biomarker caspase-3 and for γH2AX, and then analyzed by flow cytometry. All the test compounds caused a dose-dependent increase of γH2AX-positive (γH2AX+) cells. The γH2AX+ cell population included both caspase-3-positive (γH2AX+/caspase-3+) and caspase-3-negative (γH2AX+/caspase-3-) cells. The increase in γH2AX+ cells after treatment with the aneugens corresponded to the increase in caspase-3+ cells. The increase in γH2AX+/caspase-3- cells after treatment with the clastogens was significant, but there was only a slight increase after treatment with the aneugens. This reflects the fact that the apoptotic pathway of a clastogen starts from DNA damage, whereas that of an aneugen starts from cell-cycle arrest in the M-phase. Therefore, the two pathways contribute differently to apoptosis. Double staining for γH2AX and caspase-3 provided helpful information for the different mechanistic effects of aneugens and clastogens that induce γH2AX.

HSP90 inhibitor CH5164840 induces micronuclei in TK6 cells via an aneugenic mechanism.

Heat-shock protein 90 (HSP90) is a promising druggable target for therapy of conditions including cancer, renal disease, and chronic neurodegenerative diseases. Despite the possible beneficial effects of HSP90 inhibitors, some of these agents present a genotoxicity liability. We have examined the mode of action of micronucleus formation in TK6 cells by a novel and highly specific HSP90 inhibitor, CH5164840, by means of an in vitro micronucleus test with fluorescence in situ hybridization (FISH), γH2AX staining to detect DNA damage, and microscopic observation of chromosomal alignment in mitotic cells. The percentage of centromere-positive micronuclei induced by CH5164840 (FISH analysis) was significant, but the percentage of centromere-negative ones was not, suggesting that induction of micronuclei was due to a mechanism of aneugenicity rather than DNA reactivity. This conclusion was further supported by the result of co-staining γH2AX and the apoptosis marker caspase-3; the predominant elevation of apoptotic γH2AX rather than non-apoptotic γH2AX indicated little involvement of DNA-reactivity mechanisms. Microscopic observation revealed asymmetric spindle microtubules and chromosomal misalignment of metaphase cells. These data indicated that CH5164840 causes spindle dysfunction that induces micronuclei. The risk/benefit ratio must be considered in the development of HSP90 inhibitors.

New DNA probes to detect aneugenicity in rat bone marrow micronucleated cells by a pan-centromeric FISH analysis.

When characterizing the genotoxicity of chemicals that induce micronuclei, it is practical to be able to classify the chemicals as aneugens or clastogens. This classification gives information on the mechanistic properties of chemicals and is indispensable for setting the threshold safety margins for genotoxicity in pharmaceutical development. A widely used method for detecting aneugens is fluorescence in situ hybridization (FISH) but, even though the rat is an experimental animal generally used in preclinical studies in drug development, DNA probes that hybridize to all the centromeres of rat chromosomes have not yet been established. In the present study, in addition to the previously known satellite I sequence, we identified two novel satellite sequences, satellite II and satellite III, from the rat genome database. DNA probes with a mixture of these satellite DNA sequences were used to establish a FISH method for pan-centromeric staining of rat chromosomes. To confirm the feasibility of the method, vinblastine (VBS) and mitomycin C (MMC) were administered to rats as a typical aneugen and clastogen, respectively. Micronucleated polychromatic erythrocytes (MNPCE) from bone marrow were enriched by sorting in flow cytometry and subjected to the FISH method. As a result, the ratio of centromere-positive MNPCE increased in VBS-treated rats but not in MMC-treated ones. Since the FISH method using the novel DNA probes clearly discriminates the aneugens from the clastogens, we suggest this method as a useful tool for providing mechanistic information for micronucleus induction in vivo.

Fluorescent dye-based simple staining for in vivo micronucleus test with flow cytometer.

Flow cytometry (FCM) has become known as a useful tool for examining numerous cells in a micronucleus test in a short time. To successfully count micronuclei, immature erythrocytes and micronuclei need to be specifically stained and CD71-based FCM, with anti-CD71 antibody for immature erythrocytes and propidium iodide (PI) for micronuclei is a widely accepted tool. Because staining with fluorescent dyes may be much simpler compared to immunostaining, attempts are being made to develop a fluorescent dye-based FCM (FD-FCM). The aim of this study was to provide a practical FD-FCM method. Peripheral blood (PB) erythrocytes and bone marrow (BM) erythrocytes were obtained from rats treated with cyclophosphamide at a dose of 20mg/kg for two days. Nucleic cells of BM samples were eliminated using a cellulose column. Then erythrocytes were fixed, stained with Hoechst 33258 and PI and examined with FCM. Mean FD-FCM values of micronucleated immature erythrocytes in PB and BM were respectively 110% and 77% of the values obtained by microscopy. Percentages of mean immature erythrocyte values by FCM to those by microscopy were 74% and 94%. These data suggest that the simple method, composed of column purification of erythrocytes, methanol fixation, fluorescent dye staining and FCM, was useful for automated scoring in micronucleus testing of rat BM and PB.

Whole cell-ELISA to measure the gammaH2AX response of six aneugens and eight DNA-damaging chemicals.

The phosphorylated form of the histone protein H2AX (gammaH2AX) plays a central role in sensing and repairing DNA damage and is a sensitive marker for DNA double-strand breaks (DSB). Although a wide range of genotoxic agents that do not initiate DSB induce gammaH2AX, the range of chemicals that cause H2AX phosphorylation is not clear. We designed a novel, whole cell enzyme-linked immunosorbent assay (cell-ELISA) that can accurately quantify gammaH2AX levels and identify chemical compounds that induce gammaH2AX formation; our novel assay is more convenient than microscopic examination of gammaH2AX foci or flow cytometry. We measured gammaH2AX levels in CHL, CHO and V79 cells exposed to DNA-damaging, non-genotoxic and aneugenic chemicals using the cell-ELISA assay. The cell-ELISA results for the DNA-damaging compounds (methyl methanesulfonate, N-ethyl-N'-nitro-N-nitrosoguanidine, mitomycin C, cisplatin, irinotecan, etoposide, methotrexate and 5-fluorouracil) assayed showed that there was a concentration-dependent increase in gammaH2AX, which was 1.5-fold greater than the negative control; the only exception was a negative response of CHO cells to 5-fluorouracil. None of the 10 non-genotoxic compounds assayed showed similar increases in gammaH2AX and all exhibited concentration-dependent growth inhibition of the cells. The highest levels of gammaH2AX found from treatment with aneugens (vincristine, colcemid, paclitaxel, griseofulvin, 17-allylaminogeldanamycin and CH3310395), which are compounds that cause spindle dysfunction and have no genotoxic activity in the Ames test, were 1.5-fold lower than the negative control. In contrast, mitomycin C and etoposide, which both have aneugenic and DNA-damaging activities, induced a positive response. None of the aneugens caused an increase in gammaH2AX at concentrations that induce micronuclei. The chemical classes that show positive results in the cell-ELISA are different from those that are positive in the Ames or in vitro micronucleus test. By using the cell-ELISA for the level of gammaH2AX, we were able to distinguish DNA-damaging agents from non-genotoxic compounds or aneugens.

Two structurally distinct inhibitors of glycogen synthase kinase 3 induced centromere positive micronuclei in human lymphoblastoid TK6 cells.

Glycogen synthase kinase 3 (GSK3) is an attractive novel pharmacological target. Inhibition of GSK3 is recently regarded as one of the viable approaches to therapy for Alzheimer's disease, cancer, diabetes mellitus, osteoporosis, and bipolar mood disorder. Here, we have investigated the aneugenic potential of two potent and highly specific inhibitors of GSK3 by using an in vitro micronucleus test with human lymphoblastoid TK6 cells. One inhibitor was a newly synthesized maleimide derivative and the other was a previously known aminopyrimidine derivative. Both compounds elicited statistically significant and concentration-dependent increases in micronucleated cells. One hundred micronuclei (MN) of each were analyzed using centromeric DNA staining with fluorescence in situ hybridization. Both the two structurally distinct compounds induced centromere-positive micronuclei (CMN). Calculated from the frequency of MN cells and the percentage of CMN, CMN cell incidence after treatment with the maleimide compound at 1.2 microM, 2.4 microM, and 4.8 microM was 11.6, 27.7, and 56.3 per 1000 cells, respectively; the negative control was 4.5. CMN cell incidence after the treatment with the aminopyrimidine compound at 1.8 microM, 3.6 microM, and 5.4 microM was 6.7, 9.8 and 17.2 per 1000 cells, respectively. Both compounds exhibited concentration-dependent increase in the number of mitotic cells. The frequency of CMN cells correlated well with mitotic cell incidence after treatment with either compound. Furthermore, both inhibitors induced abnormal mitotic cells with asymmetric mitotic spindles and lagging anaphase chromosomes. These results lend further support to the hypothesis that the inhibition of GSK3 activity affects microtubule function and exhibits an aneugenic mode of action.

A newly established GDL1 cell line from gpt delta mice well reflects the in vivo mutation spectra induced by mitomycin C.

In order to create a novel in vitro test system for detection of large deletions and point mutations, we developed an immortalized cell line. A SV40 large T antigen expression unit was introduced into fibroblasts derived from gpt delta mouse lung tissue and a selected clone was established as the gpt delta L1 (GDL1) cell line. The novel GDL1 cells were examined for mutant frequencies (MFs) and for molecular characterization of mutations induced by mitomycin C (MMC). The GDL1 cells were treated with MMC at doses of 0.025, 0.05, and 0.1 microg/mL for 24h and mutations were detected by Spi- and 6-thioguanine (6-TG) selections. The MFs of the MMC-treated cells increased up to 3.4-fold with Spi- selection and 3.5-fold with 6-TG selection compared to MFs of untreated cells. In the Spi- mutants, the number of large (up to 76 kilo base pair (kbp)) deletion mutations increased. A majority of the large deletion mutations had 1-4 base pairs (bp) of microhomology in the deletion junctions. A number of the rearranged deletion mutations were accompanied with deletions and insertions of up to 1.1 kbp. In the gpt mutants obtained from 6-TG selection, single base substitutions of G:C to T:A, tandem base substitutions occurring at the 5'-GG-3' or 5'-CG-3' sequence, and deletion mutations larger than 2 bp were increased. We compared the spectrum of MMC-induced mutations observed in vitro to that of in vivo using gpt delta mice, which we reported previously. Although a slight difference was observed in MMC-induced mutation spectra between in vitro and in vivo, the mutations detected in vitro included all of the types of mutations observed in vivo. The present study demonstrates that the newly established GDL1 cell line is a useful tool to detect and analyze various mutations including large deletions in mammalian cells.