SCaBER and RT4 cells were purchased from the American Type Culture Collection. T24 and 5637 cells were purchased from the RIKEN Cell Bank. Cells were cultured in DMEM (Sigma-Aldrich; Merck KGaA) supplemented with 10% fetal bovine serum (FBS; HyClone; Cytiva), 100 U/ml penicillin and 100 µg/ml streptomycin (FUJIFILM Wako Pure Chemical Corporation). Mouse A9 cells containing human chromosome 9q or 4, respectively, tagged with neomycin resistance gene (neo) were maintained in DMEM supplemented with 10% FBS and 800 µg/ml G418 antibiotic (Calbiochem; Merck KGaA) and used as chromosome donors. All cell lines were maintained at 37°C in a humidified incubator with 5% CO₂.

Chromosome transfer via chromosome engineering was performed as previously described (14). A9(neo9q) or A9(neo4) cells were treated with 0.05 µg/ml colcemid at 37°C for 48 h to induce formation of micronuclei, which were then purified by cytochalasin B (10 µg/ml) digestion and centrifugation at 11,900 × g for 60 min at 34°C. The isolated microcells were then resuspended in serum-free DMEM and filtered sequentially through 8, 5 and 3 µm polycarbonate filters (Whatman plc; Cytiva). The purified microcells were collected by centrifugation at 400 × g for 15 min at room temperature (RT) and resuspended in serum-free DMEM containing 50 µg/ml phytohemagglutinin-P (FUJIFILM Wako Pure Chemical Corporation). The microcells were attached to the cell monolayer at 37°C for 15 min, fused with recipient cells in 47% polyethylene glycol solution for 1 min at RT, followed by washing with serum-free DMEM. Cells then maintained in non-selective medium (DMEM) for 24 h at 37°C, trypsinized and divided into six 100 mm dishes containing selection medium (containing 800 µg/ml G418).

The day after microcell fusion, culture medium was changed to selection medium (containing 800 µg/ml G418). One week after fusion, no viable cells were observed. Two weeks after fusion, rapidly proliferating clones were isolated and maintained by serial passaging, as previously described (14).

The presence of the 9p24.2-9q34.3 region on human chromosome 9q contained in A9(neo9) was verified by PCR using 21 specific sequence-tagged site (STS) markers (D9S54, 9p24.2; D9S268, 9p23; D9S285, 9p22.3; D9S165, 9p21.1; D9S200, 9p13.1; SHGC-103793, 9p12; UT801, 9p11.2; SHGC-141463, 9q12; SHGC-146514, 9q13; D9S15, 9q21.12; D9S1122, 9q21.2; D9S153, 9q21.31; D9S777, 9q22.1; D9S318, 9q22.2; D9S287, 9q22.32; D9S277, 9q31.1; D9S177, 9q33.1; D9S290, 9q34.11; D9S66, 9q34.2; and D9S158, 9q34.3). Primer sequences were obtained from the National Center for Biotechnology Information (https://www.ncbi.nlm.nih.gov). PCR was performed with 35 cycles of 30 sec at 94°C, 30 sec at 58-62°C and 30 sec at 72°C.

Total RNA was extracted using RNeasy Mini kit (Qiagen GmbH) from each cell line and treated with DNase I (FUJIFILM Wako Pure Chemical Corporation). First-strand cDNA was synthesized using M-MLV reverse transcriptase (Invitrogen; Thermo Fisher Scientific, Inc.) with random primers (Invitrogen; Thermo Fisher Scientific, Inc.), 5× First Strand Buffer (Invitrogen; Thermo Fisher Scientific, Inc.) and dNTPs (Applied Biosystems; Thermo Fisher Scientific, Inc.). The temperature protocol was 23°C for 12 min for primer annealing, followed by 42°C for 50 min for reverse transcription and then 95°C for 5 min for enzyme inactivation.

The mRNA expression of PPARG (encoding PPARγ), forkhead box A1 (FOXA1) and GATA3 was analyzed using specific primers as follows: PPARG forward, 5′-GACAGGAAAGACAACAGACAAATC −3′ and reverse, 5′-GGGGTGATGTGTTTGAACTTG −3′; FOXA1 forward, 5′-AGGGCTGGATGGTTGTATTG-3′ and reverse, 5′-ACCGGGACGGAGGAGTAG −3′ and GATA3 forward, 5′-GCTTCGGATGCAAGTCCA −3′ and reverse, 5′-GCCCCACAGTTCACACACT-3′. cDNA was amplified using an Applied Biosystems StepOne thermal cycler system and SYBR Green PCR kit (Applied Biosystems; Thermo Fisher Scientific, Inc.). mRNA levels were normalized against GAPDH mRNA (PCR primers: Forward, 5′-AGCCACATCGCTCAGACAC-3′ and reverse, 5′-GCCCAATACGACCAAATCC-3′) using the 2^−ΔΔCq method (15). Thermocycling conditions were 10 min at 95°C for denaturation, followed by 15 sec at 95°C and 60 sec at 60°C for denaturation and annealing/extension for 40 cycles. The experiments were performed in triplicate.

The expression levels of PPARG, FOXA1, and GATA3 of RT4, T24, SCaBER, SCaBER#4 and SCaBER9q were visualized using a freely available web server, Heatmapper (http://www.heatmapper.ca.).

To identify successful transference of chromosome 9q in metaphase spreads, chromosomal FISH was performed using plasmid pSV2neo as a probe. The transferred chromosome 4 was identified using RP11-84C13 BAC DNA as a probe.

The preparation of chromosome slides, probe labeling, hybridization, washing and detection of signals were performed as previously described (16).

Cell proliferation assay was performed to evaluate the proliferation capability of cells. SCaBER, SCaBER#4, and SCaBER#9q cells were seeded at 1.0×10⁵ cells in a 6 cm dish. All cells were cultured in DMEM supplemented with 10% FBS. Measurements were made using a hemocytometer on days 1, 2, 3, 4, and 5 after seeding. Cell counting was performed three times.

Wound healing assay was performed to evaluate the migration capability of cells. SCaBER, SCaBER#4 and SCaBER#9q cells were grown to 100% confluence in a 6-cm dish and a scratch was made with 200-µl pipette tips. The cells were washed with PBS and placed in serum-free medium (DMEM). The wound width at 0 and 24 h was measured and evaluated using a digital camera system (NIS-Elements Documentation, Ver.5.30; Nikon Corporation).

Western blotting was performed as previously described (17). Membranes were blotted with rabbit polyclonal antibodies against human PPARγ (cat. no. #2430; 1:1,000; Cell Signaling Technology, Inc.), PTEN (cat. no. #9188; 1:1,000; Cell Signaling Technology, Inc.) or α-tubulin (cat. no. PM054-7; 1:5,000; Medical and Biological Laboratories Co., Ltd.) and anti-rabbit IgG horseradish peroxidase-linked antibody (cat. no. #7074; 1:2,000; Cell Signaling Technology, Inc.), according to the manufacturer's instructions. Immunoreactive bands were visualized using an enhanced chemiluminescence detection system (cat. no. 32106; Pierce^™ ECL Western Blotting Substrate; Thermo Fisher Scientific, Inc.). Protein bands on western blot films were quantified using ImageJ (ver.1.8.0; National Institutes of Health).

The Cancer Genome Atlas (TCGA) (18) and National Cancer Institute Genomic Data Commons (19) public cancer genome bladder urothelial carcinoma (BLCA) dataset (accessed 20 June, 2021) was visualized and analyzed using UCSC Xena (xena.ucsc.edu). Overall survival curve was obtained using the Kaplan-Meier method with the median expression of each gene as the cutoff (PPARG: <19.31 vs. ≥19.31; FOXA1: <18.80 vs. ≥18.80′ GATA3: <20.52 vs. ≥20.52), and differences in survival were evaluated with the log-rank test.

Data from triplicate experiments are presented as the mean ± standard error of the mean. Data were analyzed using one-way ANOVA with post-hoc Tukey's honestly significant difference test. All statistical analysis was performed using SPSS Statistics software (version 24.0; IBM Corp.) P<0.05 was considered to indicate a statistically significant difference.

The strategy for investigation of tumor suppressor effect of human chromosome 9 in BCa is shown in Fig. 1. We previously generated a library of A9 hybrid cells, each containing one of the human chromosomes (except Y) (20). To confirm the status of human chromosome 9 in A9(neo9) cells, PCR analysis using 21 STS markers located on human chromosome 9 was performed (Fig. 2). PCR analysis showed deletion of the short arm loci of human chromosome 9 (#9delp12-pter) in A9(neo9) cells (Fig. 2). FISH analysis was performed in A9(neo9) cells using neo-plasmid probe. The pSV2neo probe for #9delp12-pter was randomly integrated into two regions, #9p12 and #9q34.3 (Fig. 3A). Additionally, the long arm of human chromosome 9, was independently retained in mouse A9 cells.

Human chromosome 9q was transferred into SCaBER cells using MMCT. Microcell hybrid cells were isolated via three successive chromosome transfer experiments and analyzed to confirm the presence of transferred #9delp12-pter tagged with pSV2neo by FISH. Transferred #9delp12-pter was stably retained in the microcell hybrid clone (SCaBER#9q; Fig. 3B). Microcell hybrids with introduced chromosome 4 (SCaBER#4) were used as a control. The presence of transferred chromosome 4 in SCaBER#4 was confirmed by FISH analysis using a PR11-q4C13 BAC probe containing the 4q22.1 genomic DNA region. The parental SCaBER cells had two copies of chromosome 4, whereas SCaBER#4 microcell hybrid clones had three copies of chromosome 4; this confirmed the presence of the transferred chromosome (Fig. 3C and D).

The morphological features of microcell hybrid clones generated via introduction of a human chromosome were microscopically examined. SCaBER#9q cells were flatter and larger than parental SCaBER and SCaBER#4 cells (Fig. 3E). Thus, human chromosome 9q may have carried genes that regulated this transformed phenotype in SCaBER cells.

The proliferation rate of microcell hybrid clones was examined to determine the effects of chromosome transfer on cell proliferation. SCaBER#9q cells exhibited significantly decreased proliferation compared with control SCaBER#4 and parental cells at days 3-5 (Fig. 4A).

To evaluate the role of chromosome 9q in regulation of malignant potential of basal BCa, wound healing assay was performed using SCaBER microcell hybrid clones. Compared with parental SCaBER and SCaBER#4 cells, SCaBER#9q cells showed decreased mobility (Fig. 4B and C). This indicated that tumor suppressor genes involved in cell proliferation and migration in SCaBER cells were present on 9q.

The luminal subtype of BCa has a lower proliferative and migratory potential than the basal subtype and is less malignant (8). GATA3, FOXA1, and PPARγ are specifically upregulated in the luminal subtype (as previously determined by RNA sequencing analysis) (21). Furthermore, overexpression of GATA3 and FOXA1, which are tumor suppressive, is accompanied by activation of PPARγ in transformation of basal into luminal BCa cells (22–26). Additionally, LOH on the long arm of chromosome 9 is commonly observed in premalignant lesions of bladder carcinogenesis, suggesting that chromosome 9q encodes tumor suppressor genes that serve a key role in development and progression of BCa. The expression levels of luminal markers (FOXA1, GATA3 and PPARG) was examined; compared with SCaBER cells, SCaBER#9q cells exhibited a 2.4-, 1.9- and 5.2-fold increase in the expression of the aforementioned luminal markers (Fig. 5A-C).

A heatmap of differential molecular subtypes was used to evaluate the relative expression of luminal markers (Fig. 5D). The relative expression level of luminal markers was high in RT4 cells (typical luminal subtype), whereas SCaBER (basal) and T24 cells (non-type) exhibited low expression levels (27). Only expression of PPARG was higher in SCaBER#9q cells than in parental, SCaBER#4 and RT4 cells (Fig. 5D). The expression profiles of three luminal marker genes in BCa was further investigated using BLCA dataset. Higher expression of all luminal marker genes was associated with improved survival in patients with BCa (Fig. 6). These results suggested that tumor suppressor genes on 9q may serve an important role in determining the molecular subtype of BCa, which is associated with development of malignancy.

Western blotting was performed to confirm that expression of luminal markers (GATA3, FOXA1. and PPARG) was also increased at the protein level. Only PPARγ was significantly increased in SCaBER#9q cells (3.0-4.4-fold) compared with parental cells (Figs. 7A and B and S1). PPARγ inhibits proliferation, metastasis and invasion of cancer by induction of PTEN expression (28). The expression levels of PTEN were also increased in SCaBER#9q cells (Fig. 7A and B). This suggested that chromosome 9q carried genes that regulate luminal marker of BCa.