Determination of the complexity and diversity of the TCR β-chain CDR3 repertoire in bladder cancer using high-throughput sequencing.

The present study aimed to investigate the complexity and diversity of the T lymphocyte immune repertoire in patients with bladder cancer. To do so, the immune state of patients was assessed. The study also aimed to elucidate the aetiology and pathogenesis of bladder cancer to provide a novel theoretical basis for disease prevention, diagnosis, treatment and prognosis monitoring. Cancerous and paracancerous (control) tissue samples were collected from five patients diagnosed with muscle-invasive bladder cancer. Multiplex polymerase chain reaction and Illumina high-throughput sequencing were used to determine the characteristics and clonal diversity of the T-cell receptor (TCR) β-chain complementarity-determining region 3 (CDR3) gene in the cancerous and paracancerous tissues of patients with bladder cancer. The degree of clonal expansion in malignant samples was significantly higher than in adjacent samples. Furthermore, ΤCRβ variable (TRBV), ΤCRβ diversity (TRBD) and ΤCRβ joining (TRBJ) repertoires were significantly different in cancerous samples compared with adjacent samples. In addition, 13 identified V-J pairs were highly expressed in cancerous samples whereas they had low expression in control samples. In conclusion, the degree of T-cell clonal expansion in bladder cancerous tissue was higher than in paracancerous tissue, whereas the immune diversity of the tissues of patients with bladder cancer was significantly lower. The DNA sequence and amino acid sequences, and V-J combination level may be used to comprehensively understand the diversity and characteristics of TCR CDR3 in bladder cancer and paracancerous tissues, and to evaluate the immune status of bladder cancer to develop therapeutic targets and biomarkers for prognosis monitoring.

Introduction

Increasing evidence has revealed that the adaptive immune response to cancer is a prospective prognostic marker and a diagnostic criterion for various types of cancer (1,2). Previously, targeted immune-based therapies have revealed durable responses and may represent novel promising treatments for various types of cancer, including melanoma (3), breast (4), renal (5), non-small cell lung (6) and bladder cancer (7). These emerging immunotherapies, including the immunomodulators Bacillus Calmette-Guérin (BCG) polysaccharide nucleic acid and adoptive cell transfer, highlight the crucial need to better monitor and comprehend tumour-infiltrating lymphocytes (TILs).

T cells in cancer tissues are generated during T-cell differentiation in the thymus (8) and serve a central role in the tumour immune response. T cells recognize antigens displayed on the surface of cancer cells via the membrane protein T-cell receptor (TCR). Each T cell commonly possesses a single TCR. The majority of TCRs contain an α-polypeptide chain and a β-polypeptide chain. Each chain has a variable region (V region), a joining region (J region) and a constant region (C region). The β chain also possesses a diversity region (D region) (8). The V region of the TCR α- and β-chains contains three hypervariable regions, including complementarity determining regions (CDR) 1, 2 and 3, of which CDR3 is the most variable and therefore the primary source of antigen-specific recognition. The diversity of TCR is generated by a random rearrangement of the V, D, J and C regions, extended by non-germline-encoded nucleotides randomly inserted at the junctions of the final segments (9).

Since the diversity of distinct TCRs is ~2.5×1018 in a normal adult, the TCR repertoire is therefore difficult to analyse (10). High-throughput sequencing (HTS) is a novel method that can profile the TCR repertoire and analyse the adaptive immune response; therefore allowing the study of TCR repertoire diversity at a greater depth than previously (11,12). Notably, this technology has been used to investigate and monitor the implication of TCR in tumour immunity through large-scale sequencing of CDR3 (13–15). The present study aimed to evaluate the complexity and diversity of cancer tissues compared with paracancerous tissues by amplifying the CDR3 of TCRs using multiplex polymerase chain reaction (PCR) and HTS. Decrypting the TCRβ chain repertoire in TILs may offer novel insights for the development of future biomarkers and monitoring in bladder cancer.

Materials and methods

Patients

Five patients with muscle-invasive bladder cancer were selected, and cancer and paracancerous specimens were collected at The Shenzhen People's Hospital (Shenzhen, China) between January and April 2016. The diagnosis of bladder cancer was confirmed by pathological diagnosis and clinical evidence. The patient cohort was 44–82 years old, with a mean age of 63.10±19.21 years. These patients had not been treated with chemotherapy prior to the study, and written informed consent was obtained for study participation. This study passed the discussion and approval of the Ethics Committee of Guilin No. 181 Hospital (Guilin, China) and the Shenzhen People's Hospital (Shenzen, China).

Sample preparation

A total of 5–10 cancerous and paracancerous sections were obtained from each patient with bladder cancer, and total DNA was extracted using a standard method. Briefly, dewaxing was performed using xylene, followed by an overnight proteinase K digestion at 56°C. Total DNA in tissue samples was extracted using the QIAamp DNA Mini kit (Qiagen GmbH, Hilden, Germany) according to the manufacturer's protocol. The DNA concentration of each sample was determined using a Qubit fluorometer (Invitrogen; Thermo Fisher Scientific, Inc., Waltham, MA, USA). DNA quality was evaluated via 0.8% agarose gel electrophoresis. The DNA samples extracted from the cancerous tissues of the five patients were mixed together in a 1:1:1:1:1 ratio according to the Qubit value, to obtain a unique cancer sample. Similarly, the DNA samples extracted from the paracancerous tissues of the five patient were mixed together in a 1:1:1:1:1 ratio according to the Qubit value, to obtain a unique control sample.

Multiplex polymerase chain reaction (PCR) amplification of the TCRβ CDR3 region

The human TCRβ sequences were downloaded from the international ImMunoGeneTics information system® for immunoglobulins or antibodies (http://www.imgt.org/) (16). A relatively conserved region in frame region 3, upstream of CDR3, was selected for the putative forward primer region. A cluster of primers (Table I) corresponding to the majority of the V gene sequence family was selected. Reverse primers, corresponding to the J gene family, were similarly designed (17). A total of 30 forward primers and 13 reverse primers were used with the QIAGEN multiplex PCR kit (Qiagen GmbH) to amplify the rearranged TCRβ CDR3 region. The PCR reactions (50 µl) consisted of 2 µl TCRβ variable (TRBV) pool (10 µM), 2 µl TCRβ joining (TRBJ) pool (10 µM), 25 µl 2X QIAGEN Multiplex PCR Master, 5 µl 5X Q-solution, 10 µl template DNA and 6 µl H2O. The PCR conditions were 95°C for 15 min followed by 25 cycles of 94°C for 15 sec and 60°C for 3 min, with a final extension for 10 min at 72°C. The PCR products were purified from the primer sequences by AMPure XP beads (Beckman Coulter, Inc., Brea, CA, USA). A second round of PCR was performed to add a sequencing index to each sample. The PCR reaction (50 µl) consisted of 13.5 µl H2O, 0.5 µl 2X Q5 DNA polymerase, 10 µl 5X Q5 buffer, 1 µl dNTP (10 mM), 1 µl P1 (10 µM), 23 µl DNA, and 1 µl index (10 µM). The PCR condition was set at 98°C for 1 min, followed by 25 cycles of 98°C for 20 sec, 65°C for 30 sec and 72°C for 30 sec, with a final extension for 5 min at 72°C. The library was separated by agarose gel electrophoresis, and the target region was isolated and cleaned using QIAquick Gel Extraction kits according to the manufacturer's protocol (Qiagen GmbH).

Table I.

TRB V primers (5′-3′).

Primer	Sequence
TRBV2	ATTTCACTCTGAAGATCCGGTC
	CAC
TRBV3-1	AAACAGTTCCAAATCGMTTCT
	CAC
TRBV4-1/2/3	CAAGTCGCTTCTCACCTGAATG
TRBV5-1	GCCAGTTCTCTAACTCTCGC
	TCT
TRBV5-4/5/6/8	TCAGGTCGCCAGTTCCCTAAY
	TAT
TRBV6-4.1	CACGTTGGCGTCTGCTGTAC
	CCT
TRBV6-8/5/1.2	CAGGCTGGTGTCGGCTGCTC
	CCT
TRBV6-9/7/1.1/6	CAGGCTGGAGTCAGCTGCTC
	CCT
TRBV6-4.2	AGTCGCTTGCTGTACCCTCT
	CAG
TRRBV6-2/3	GGGGTTGGAGTCGGCTGCTC
	CCT
TRBV7-2/4/6/7/8	GGGATCCGTCTCCACTCTGAM
	GAT
TRBV7-3	GGGATCCGTCTCTACTCTGAA
	GAT
TRBV7-9	GGGATCTTTCTCCACCTTGGA
	GAT
TRBV9	CCTGACTTGCACTCTGAACTAA
	ACCT
TRBV10-1	CCTCACTCTGGAGTCTGCTGCC
TRBV10-2/3	CCTCACTCTGGAGTCMGCT
	ACC
TRBV11-1/2/3	GCAGAGAGGCTCAAAGGAGT
	AGACT
TRBV12-3.2/5.2	GAAGGTGCAGCCTGCAGAAC
	CCAG
TRBV12-3.1/4/5.1	GAAGATCCAGCCCTCAGAACC
	CAG
TRBV13	TCGATTCTCAGCTCAACAGTTC
TRBV14	GGAGGGACGTATTCTACTCTGA
	AGG
TRBV15	TTCTTGACATCCGCTCACCAGG
TRBV16	CTGTAGCCTTGAGATCCAGGC
	TACGA
TRBV18	TAGATGAGTCAGGAATGC
	CAAAG
TRBV19	TCCTTTCCTCTCACTGTGACAT
	CGG
TRBV20-1	AACCATGCAAGCCTGACCTT
TRBV24-1	CTCCCTGTCCCTAGAGTCTGC
	CAT
TRBV25-1	GCCCTCACATACCTCTCAGTA
	CCTC
TRBV27-1	GATCCTGGAGTCGCCCAGC
TRBV28	ATTCTGGAGTCCGCCAGC
TRBV29-1	AACTCTGACTGTGAGCAACATGAG
TRBV30-F5	CAGATCAGCTCTGAGGTGCCCCA

TRB, T-cell receptor β

High-throughput sequencing and data analysis

The library was quantified using an ABI StepOnePlus Real-Time PCR system (Applied Biosystems; Thermo Fisher Scientific, Inc.) and an Agilent 2100 Bioanalyzer instrument (Agilent DNA 1000 Reagents; Agilent Technologies GmbH, Waldbronn, Germany), and sequenced by Illumina MiSeq (Illumina, Inc., San Diego, CA, USA). Briefly, the low-quality reads and the adaptor reads were filtered from the raw data, and the clean data were used in further alignment. Subsequently, the clean data were aligned to the human TRBV database and were analysed using the online IMGT (ImMunoGeneTics)/HighV-QUSET and miTCR (MiLaboratory LLC, Solana Beach, CA, USA) tools. The data included the V, D and J assignments, the CDR3 length distribution, clustering, highly expanded clone and Shannon entropy index (18). Shannon entropy index is a measure of immune diversity in the field of immune library. This index ranges between 0 and 1, where 1 indicates the most diversity, and 0 indicates no diversity at all. The flow chart of biological information is presented in Fig. 1.

Figure 1.

The flow chart of biological information analysis. CDR3, complementarity-determining region 3; IMGT, ImMunoGeneTics; iRAP, integrated RNS-seq analysis pipeline.

Results

T-cell receptor β sequence statistics

A total of 1,789,531 raw reads and 857,022 raw reads were obtained for the cancer sample and the control sample, respectively. The miTCR was used to map the reads to the available databases. The mapping of the immune sequence reads, the total CDR3 sequences, and the unique CDR3 nucleotide sequences per sample are presented in Tables II and III.

Table II.

TRB J primers (5′-3′).

Primer	Sequence
TRBJ1.1	CTTACCTACAACTGTGAGTCTGGTG
TRBJ1.2	CTTACCTACAACGGTTAACCTGGTC
TRBJ1.3	CTTACCTACAACAGTGAGCCAACTT
TRBJ1.4	AAGACAGAGAGCTGGGTTCCACT
TRBJ1.5	CTTACCTAGGATGGAGAGTCGAGTC
TRBJ1.6	CATACCTGTCACAGTGAGCCTG
TRBJ2.1	CCTTCTTACCTAGCACGGTGA
TRBJ2.2	CTTACCCAGTACGGTCAGCCT
TRBJ2.3	CCGCTTACCGAGCACTGTCAG
TRBJ2.4	AGCACTGAGAGCCGGGTCC
TRBJ2.5	CGAGCACCAGGAGCCGCGT
TRBJ2.6	CTCGCCCAGCACGGTCAGCCT
TRBJ2.7	CTTACCTGTGACCGTGAGCCTG

TRB, T-cell receptor β

Table III.

T-cell receptor β sequence statistics.

Data type	Cancer	Control
Total raw reads	1,789,531	857,022
Total clean sequences number	1,081,491	551,870
Immune sequences number	1,059,138	543,038
Unknown sequences number	22,353	8,832
Productive sequences number	744,400	417,942
Nonproductive sequences number	314,738	125,096
In-frame sequences number	763,882	433,569
Out-of-frame sequences number	287,038	105,628
Total CDR3 sequences number	718,791	410,030
Unique CDR3 nt sequences number	13,196	26,051
Unique CDR3 aa sequences number	10,125	21,721

aa, amino acid; CDR3, complementary-determining region 3; nt, nucleotide

Degree of expansion and diversity analysis of T-cell clones

In order to comprehensively and accurately understand the diversity of the TCR clones in the cancer and control samples, the degree of clonal amplification of the DNA sequences was calculated in each sample. The statistical method used to determine the degree of clonal amplification of the sequence in each sample represented the number of times the sequence was present in the test sample. In the present study, a highly expanded clone (HEC) was defined as a TCR clone with a frequency >0.5% of the total reads in each sample. In the cancer sample, 32 DNA sequences were observed and defined as HECs, with a HEC ratio of 0.46. Conversely, only 14 clones were HECs, with a HEC ratio of 0.19, in the control sample. Subsequently, a plot of the frequency distribution of the TCR repertoire was developed in cancer and control samples. Few DNA sequences were present at substantially higher frequencies (>1%), and the majority of the DNA sequences occurred at low frequencies in both cancer and control samples. The percentages of different abundance clones were determined. In the cancer sample, the percentage of clones with an abundance of 0.1–1% was 81.50% of the total T-cell sequences, whereas the percentage of clones with an abundance of 0.01–0.1% was 12.80%. Conversely, in the control sample, the percentage of clones with an abundance of 0.1–1% was 32.60%, whereas the percentage of clones with an abundance of 0.01–0.1% was highest, at 43.86% (Fig. 2). The TCR repertoire in the cancer sample had a greater HEC value (>0.5%), HEC ratio and a higher degree clone percentage (>0.1%) than the control sample. To compare the TCR diversity of the cancer and control samples, the normalized Shannon entropy index was also used. The normalized Shannon entropy index of the cancer sample was lower than that of the control sample (0.40 vs. 0.59) (data not shown). The present study suggested that the entire TCR repertoire in the cancer sample had a substantially more skewed clonotype composition than the control sample.

Figure 2.

Degree of expansion and frequency distribution of the T-cell clones. The x-axis depicts the degree of expansion; the y-axis depicts the percentage of each clone frequency.

Special and shared TCR sequences

To further assess how TCR sequences were shared between the cancer and control samples, deep sequencing was used to characterize the HEC sequences of the respective TCR repertoires. The analysis focused mainly on the DNA and amino acid (aa) sequences of CDR3, which was the most diverse region of the TCR molecule and was relevant to antigen epitope recognition. With regards to the DNA sequences, 12,516 special DNA sequences were obtained in the cancer sample, of which five were HECs. A total of 25,371 special DNA sequences were obtained in the control sample, of which 10 were HECs (Table IV). For the aa sequence, 9,424 special aa sequences were obtained in the cancer sample, of which, seven were HECs, whereas the control sample exhibited 21,020 special aa sequences, of which 10 were HECs (Table V). In addition, 680 shared DNA sequences and 701 shared aa sequences were determined in the cancer and control samples. In a further comparative analysis, all 27 HEC DNA sequences in the cancer sample demonstrated a lower expression (<0.5%) in the control sample. Conversely, four HEC DNA sequences in the control sample had a lower expression (<0.5%) in the cancer sample (Table VI). For the shared aa sequence, all 28 HECs in the cancer sample exhibited a lower expression (<0.5%) in the control sample. In contrast, four HECs in the control sample showed a lower expression (<0.5%) in the cancer sample (Table VII). There were no shared HECs in the DNA and aa sequences between the cancer and control samples. These results suggested that comparisons of the TCR HECs in cancer and control samples may accelerate the screening process for possible novel biomarkers.

Table IV.

A summary of the highly expressed special DNA sequences (>0.5%).

A, Cancer
Clonotype (5′-3′)	Frequency (%)
GCCACCAGTGAGTTCGACAGGGGTAAAGAGACCCAGTAC	0.54
GCCACCAGTGATCAGATTCATGTGGAGACCCAGTAC	0.71
GCCAGCAATACCGGGACAGGGGCATATGGCTACACC	0.72
GCCAGCAGTGATTTGGGGCCAGGCCTGTATGGCTACACC	0.77
GCCAGCTACTCCTATGGCTACACC	0.72
B, Control
Clonotype (5′-3′)	Frequency (%)
GCCACCAGCAAAGGGACGACTAGGGCCGGCGAGCAGTAC	0.90
GCCACCAGTTTTATTAAACAGCCCCAGCAT	1.49
GCCAGCAGCCCGGGACAACCACCCCTCCAC	0.59
GCCAGCAGCTTAGAGTCCGGACCCTACGAGCAGTAC	0.51
GCCAGCAGTGACAACGGGCCAAATCAGCCCCAGCAT	1.41
GCCAGCAGTGACCGGGGTGAGCAGTAC	1.03
GCCAGCAGTTACAGGGGGGGGGAGCTGTTT	1.35
GCCAGTAGCCCCACGGGGGCAGGGAACACTGAAGCTTTC	1.06
GCCTCCACCGACAACAGGGAGACCCAGTAC	1.41
GCCTGGAGTGTTACCGGGGCAGATGAAAAACTGTTT	0.62

Table V.

A summary of the highly expressed special amino acid sequences (>0.5%).

A, Cancer
Clonotype (5′-3′)	Frequency (%)
ASNTGTGAYGYT	0.73
ASSDLGPGLYGYT	0.78
ASSMGRTDTQY	0.53
ASSVEGRTQY	0.50
ASYSYGYT	0.72
ATSDQIHVETQY	0.73
ATSEFDRGKETQY	0.54
B, Control
Clonotype (5′-3′)	Frequency (%)
ASSDNGPNQPQH	1.43
ASSDRGEQY	1.05
ASSLESGPYEQY	0.51
ASSPGQPPLH	0.60
ASSPTGAGNTEAF	1.08
ASSYRGGELF	1.37
ASTDNRETQY	1.42
ATSFIKQPQH	1.51
ATSKGTTRAGEQY	0.92
AWSVTGADEKLF	0.63

Table VI.

A summary of the highly expressed shared DNA sequences (>0.5%).

Clonotype (5′-3′)	Cancer (%)	Control (%)
GCCACCAGCAGAACGGCAGGAGAGACCCAGTAC	0.781	0.001
GCCACCAGCAGAGAAGCTAGCGGGATCCAAGAGACCCAGTAC	0.558	0.001
GCCACCAGCAGAGATACTCTAGGCACAGATACGCAGTAT	0.001	2.191
GCCACCAGCAGAGATGGGGGCGGGGTACAAGAGACCCAGTAC	0.737	0.001
GCCACCAGCAGTCCCTCGGGGATCTATGGCTACACC	1.429	0.014
GCCACCAGCGATGATCAGGGGAGTTTTAATGAAAAACTGTTT	0.748	0.001
GCCACCAGTGATATAGCGGGAGAGACCCAGTAC	1.315	0.001
GCCACCAGTGATCGGGACTACTACAATGAGCAGTTC	1.036	0.002
GCCACCAGTGATTCATCGGATTACGAGCAGTAC	0.501	0.033
GCCACCAGTGATTGGGACAGCCCGACCTACGAGCAGTAC	0.514	0.001
GCCACCAGTGATTTGCGGGGACAGGGCCAAGAGACCCAGTAC	0.896	0.001
GCCACCAGTGATTTGGCCAGGCATTATGGCTACACC	0.650	0.001
GCCACCAGTGATTTTAGGGGCCAACCCCAAGAGACCCAGTAC	1.053	0.008
GCCACCAGTGATTTTGATTCGGGACTTTCGGGGACCCAGCAC	0.656	0.001
GCCACCAGTGATTTTGATTCGGGACTTTCGGGGACCCAGTAC	17.886	0.027
GCCACCAGTGATTTTGGAGGGATCCAAGAGACCCAGTAC	0.595	0.001
GCCACCTCAGGGGACGGCTATGGCTACACC	0.869	0.001
GCCAGCACCCTCTGGGGGAACACTGAAGCTTTC	0.001	1.601
GCCAGCAGCTTTGGGAGCGAGCAGTAC	0.564	0.001
GCCAGCAGGGAATTCGGACAGGGGGCAAAGGCTTTC	1.972	0.001
GCCAGCAGTGAAAGCGGGAGTAGTGCAGATACGCAGTAT	2.047	0.001
GCCAGCAGTGAAGGGGGGCAGGGGCGGCAGACCCAGTAC	0.574	0.001
GCCAGCAGTTCCGCAACGTACCGATACACC	1.303	0.009
GCCAGCATATTCGCGGTAGAGACTGAAGCTTTC	0.001	0.533
GCCAGCCAGGGGTTCGGGGAGCTGTTT	0.811	0.001
GCCAGCCTTCCCGAGACTAGCGGGAGCCACGAGCAGTAC	0.532	0.001
GCCAGTACCCATACGGGTCGAGTGGACGGCTACACC	0.557	0.001
GCCAGTAGTATCAATCTTAACGGCGAGCAGTAC	0.840	0.002
GCCAGTAGTATCGGGACAGGGGGCGCCGGGGGCTACACC	0.839	0.001
GCCAGTGGGGAGGGCCGGACCGGGGAGCTGTTT	2.025	0.001
GCCTCGCGGGAGGGCCCCACCGGGGAGCTGTTT	0.710	0.001
GCCTGGAGTATTTGGACTAACACTGAAGCTTTC	0.001	4.212

Table VII.

A summary of the highly expressed shared amino acid sequences (>0.5%).

Clonotype	Cancer (%)	Control (%)
ASGEGRTGELF	2.038	0.001
ASIFAVETEAF	0.001	0.538
ASLPETSGSHEQY	0.535	0.001
ASQGFGELF	0.815	0.001
ASREFGQGAKAF	1.988	0.001
ASREGPTGELF	0.715	0.001
ASSEGGQGRQTQY	0.579	0.001
ASSEGRKY	0.502	0.001
ASSESGSSADTQY	2.063	0.001
ASSFGSEQY	0.565	0.001
ASSIGTGGAGGYT	0.846	0.001
ASSINLNGEQY	0.846	0.001
ASSSATYRYT	1.311	0.009
ASTHTGRVDGYT	0.560	0.001
ASTLWGNTEAF	0.001	1.624
ATSDDQGSFNEKLF	0.755	0.001
ATSDFDSGLSGTQH	0.663	0.001
ATSDFDSGLSGTQY	18.071	0.027
ATSDFGGIQETQY	0.600	0.001
ATSDFRGQPQETQY	1.062	0.008
ATSDIAGETQY	1.329	0.001
ATSDLARHYGYT	0.656	0.001
ATSDLRGQGQETQY	0.903	0.001
ATSDRDYYNEQF	1.042	0.001
ATSDSSDYEQY	0.492	0.034
ATSDWDSPTYEQY	0.517	0.001
ATSGDGYGYT	0.873	0.001
ATSRDGGGVQETQY	0.753	0.001
ATSRDTLGTDTQY	0.001	2.241
ATSREASGIQETQY	0.582	0.001
ATSRTAGETQY	0.789	0.001
ATSSPSGIYGYT	1.437	0.014
AWSIWTNTEAF	0.001	4.276

Comparison of TRBV, TCRβ diversity (TRBD) and TRBJ repertoires

To determine whether cancer-specific differences exist in the TRBV, TRBD and TRBJ repertoires, the expression levels of the respective TRBV, TRBD and TRBJ repertoires were compared between the cancer and control samples. Results demonstrated that four special TRBV segments (TRBV12-1, TRBV5-7, TRBV6-7 and TRBV6-8) were present in the control sample, whereas none were present in the cancer sample. For the 52 shared TRBV gene segments, the usage frequencies of TRBV2 (14.32 vs. 14.61%) and TRBV25-1 (4.43 vs. 4.04%) were similar (±10%) between the cancer and control samples. For the other distinct TRBV gene segments, the usage frequencies of each TRBV gene exhibited extreme variation. A stacked bar chart was generated to display the frequencies of the 20 top TRBV genes, and differences were revealed between the cancer and control samples (Fig. 3). The usage frequencies of the TRBD gene segment were 38.42 and 61.58% for TRBD1 and TRBD2, respectively in the cancer sample. Conversely, the usage frequencies of the TRBD1 and TRBD2 gene segments were 53.76 and 46.24%, respectively in the control sample (data not shown). With the exception of TRBJ2-2P (0.01 vs. 0.01%), the other 13 TRBJ gene segments usage frequencies displayed extreme variation between the cancer sample and control sample. These data indicated that there might be clonal expansion, in particular for the V and J gene families, in the cancer sample.

Figure 3.

Stacked bar chart for the usage frequencies of the top 20 TRBV gene family segments in the cancer sample and control sample. The x-axis depicts the name of the sample, and the y-axis depicts the frequencies of the TRBV gene subtype, which accounts for the percentage of all the TRBV gene subtypes in the sample. TRBV, T-cell receptor β variable.

To compare the expression difference of the V-J pairs between the cancer and control samples, seven special V-J pairs were obtained in the cancer sample and 132 special V-J pairs in the control sample; there were 483 shared V-J pairs between the cancer and control samples. However, the seven special V-J pairs in the cancer sample and 132 special V-J pairs in the control sample demonstrated lower usage frequency (<1%), whereas for the 483 shared V-J pairs, 23 V-J pairs were found in the cancer sample and 22 V-J pairs in the control sample that exhibited a high usage frequency (>1%) (Table VIII). Among these, 10 V-J pairs were overlapping, and the other V-J pairs revealed extreme variation between the cancer sample and control sample. For example, TRBV24-1-TRBJ2-5 (23.67%) exhibited the highest expression in the cancer sample, with a low expression in the control sample (0.75%). Conversely TRBV30-TRBJ1-1 (4.90%) displayed the highest expression in the control sample, whereas its expression was very low in the cancer sample (0.22%). Taken together, these data indicated that specific V-J dominant pairs were dominant in the cancer sample.

Table VIII.

V-J pair usage percentage (>1%) in the cancer and control samples.

A, Cancer
V-J pairs	Usage (%)
TRBV24-1, TRBJ2-5	23.67
TRBV15, TRBJ2-5	3.79
TRBV19, TRBJ2-2	3.78
TRBV19, TRBJ1-2	3.27
TRBV2, TRBJ2-3	2.65
TRBV2, TRBJ2-7	2.29
TRBV24-1, TRBJ2-7	2.23
TRBV2, TRBJ1-1	2.18
TRBV12-3, TRBJ1-1	2.05
TRBV24-1, TRBJ2-1	1.89
TRBV2, TRBJ2-5	1.82
TRBV2, TRBJ1-5	1.79
TRBV24-1, TRBJ1-2	1.72
TRBV15, TRBJ1-2	1.48
TRBV12-3, TRBJ1-2	1.47
TRBV19, TRBJ2-7	1.41
TRBV10-3, TRBJ2-7	1.38
TRBV15, TRBJ2-7	1.37
TRBV19, TRBJ2-5	1.32
TRBV2, TRBJ1-2	1.26
TRBV15, TRBJ1-4	1.24
TRBV25-1, TRBJ1-2	1.21
TRBV25-1, TRBJ2-7	1.11
B, Control
V-J pairs	Usage (%)
TRBV25-1, TRBJ1-4	1.57
TRBV15, TRBJ2-7	2.14
TRBV19, TRBJ2-5	2.13
TRBV19, TRBJ1-2	1.43
TRBV20-1, TRBJ2-5	1.10
TRBV24-1, TRBJ1-5	1.76
TRBV30, TRBJ1-1	4.90
TRBV2, TRBJ2-1	1.06
TRBV12-3, TRBJ1-2	1.00
TRBV25-1, TRBJ1-1	1.15
TRBV20-1, TRBJ2-7	1.20
TRBV2, TRBJ2-7	3.21
TRBV24-1, TRBJ2-7	1.07
TRBV12-3, TRBJ2-7	1.18
TRBV12-3, TRBJ2-2	1.94
TRBV2, TRBJ1-2	2.09
TRBV12-3, TRBJ1-1	1.12
TRBV19, TRBJ1-1	1.89
TRBV2, TRBJ1-1	2.85
TRBV2, TRBJ2-5	1.63
TRBV15, TRBJ2-3	2.69
TRBV7-2, TRBJ2-7	1.38

Bold indicates shared V-J pairs. TRBV, T-cell receptor β variable

Discussion

Intrinsic and extrinsic factors are important in the development of bladder cancer; however, the exact pathogenesis is still not fully understood. TILs are strictly correlated with patient prognosis and provide the theoretical basis for gene-immunological therapy in bladder cancer (19). Pichler et al revealed that the tumour microenvironment influences the therapeutic response to BCG, which may permit individualized treatment (7). However, most studies on individualized treatment are based on changes and molecular typing of T-cell subsets, and only a few studies focus on the sequence information of TCR binding to tumour antigens. T lymphocytes serve an important role in the tumour immune response; they recognize major histocompatibility complex-bound peptides, which are mediated by TCRs. The CDR3 region of TCRs has a special molecular structure, which represents the different populations of T cells. In the present study, multiplex PCR and HTS were used to investigate the T-cell repertoires in cancerous and paracancerous tissues from patients with bladder cancer. A total of 1,081,491 and 551,870 sequences were obtained from the cancer sample and the control sample, respectively, providing substantial evidence regarding the TCR repertoire in bladder cancer.

HECs are crucial in the immunological repertoire and may be the result of physiological responses to pathogens or antigens. In the present study, the number of HECs and the contribution of these HECs to the total TCR repertoire were analysed. The TCR repertoire in the cancer sample had a greater HEC (>0.5% total TCR analysed) number, HEC ratio, and a higher degree clone percentage (>0.1%) than in the control sample. These results suggested that the degree of amplification in cancer tissue was significantly higher than in paracancerous tissue. Furthermore, to better understand the TCR diversity of the cancer and control samples, the Shannon entropy index (20) was used. The results suggested that the diversity of cancer tissues was lower than that of paracancerous tissues. Nakanishi et al similarly demonstrated that the proportion of CDR3 aa sequence accounting for >0.01% of the total molecular population in the TILs is significantly higher than in peripheral blood lymphocytes in colorectal cancer (15). In addition, Han et al revealed that the HEC ratio of the TCRβ chain CDR3 sequence in carcinoma tissues is significantly higher than in the adjacent tissues from patients with hepatitis B virus-associated hepatocellular carcinoma (21). Velotti et al, via high-resolution PCR, determined the CDR3 size lengths of TCRβ transcripts from patients with bladder cancer, and demonstrated that the number of expanded clones in the cancer tissues is larger than that in peripheral blood (22). This difference may be explained by the prolonged exposure of T lymphocytes to tumour antigens, leading to a targeted rearrangement of the TCR gene, eventually resulting in significant enrichment of the corresponding clones in the cancer tissues. The identification of bladder cancer-specific HECs may therefore accelerate the screening process of potential novel tumour antigens. This screening is critical to better understand the TCR repertoire, and may represent a novel diagnostic tool for patients with bladder cancer.

In the present study, analysis of the expression levels of TRBV, TRBD and TRBJ repertoires was systematically performed and revealed significant differences in the cancer tissues compared with the paracancerous tissues. In addition, 13 identified V-J pairs were highly expressed (>1%) in the cancerous tissue, whereas their expression was very low in the paracancerous tissue. The high expression of one or more TCR BV subfamilies in the cancer tissues from patients with bladder cancer may be associated with the immune pathogenesis of the disease. Therefore, the present study provided a theoretical foundation for further investigations on the immune pathogenesis and gene-immunological therapy of bladder cancer. In the absence of disease, TCR BV is rearranged randomly, and T cells exhibit a positive polyclonal state. However, in the case of disease, prolonged stimulation by specific antigens gives rise to a selective rearrangement and excessive abnormal cloning of one or more TCR BV subfamilies; excessive cloning of the cloned T cell may thus inhibit cloning of other T cells (23,24). Wang et al reported a preferential use of TRBV and TRBJ genes in cancerous tissues compared with adjacent tissues in breast cancer (25). A recent study identified specific V-J pairs that may distinguish the TCR repertoires of patients with liver cancer from healthy subjects. These specific V-J pairs may therefore be used as novel biomarkers in liver cancer (21).

In conclusion, the present study demonstrated that deep analysis of the TCR families from tissue samples may be performed using a HTS platform. This may be a powerful novel approach to evaluate the complexity and diversity of the T-cell immune repertoire in bladder cancer. However, the sample size of this study was small, which did not allow a complete description of the diversity of the TCRβ CDR3 repertoire in various types of bladder cancer. A better understanding of the functional basis of the TCRβ CDR3 clonalities is crucial, and further studies based on larger sample sizes, using tools such as NetMHC and SYFPEITHI, are required to identify certain particular peptides including signal transducer and activator of transcription 3 and glutathione S-transferase alpha 1. This may represent a novel biomarker for the detection, prognosis and development of gene-immunological therapy of bladder cancer.