Abstract
Background
Researchers have previously reported that mitochondrial DNA copy number (mtDNA-CN) can play different roles in microsatellite instable/mismatch repair-deficient (MSI/dMMR) and microsatellite stable/mismatch repair-proficient (MSS/pMMR) colorectal cancer (CRC). To support malignancy, dMMR CRC relies on glycolysis, while pMMR CRC favors oxidative phosphorylation. However, it is unclear whether mtDNA-CN changes are related to T cell infiltration in CRC.
Methods
The mtDNA-CN was detected by qRT-PCR in 532 patients, and the expression of CD3 and CD8 in 485 patients was detected by immunohistochemistry. The correlation between mtDNA-CN and the prognosis of CRC patients was further analyzed, and the correlation between mtDNA-CN and T lymphocyte infiltration was also analyzed. Biopsy specimens from the immune checkpoint inhibitors (ICIs) treatment cohort were obtained to verify the correlation between mtDNA-CN and the efficacy of ICIs. The effects of mtDNA-CN and MMR status on gene expression were analyzed by RNA-seq.
Results
Our results show that mtDNA-CN has inverse relationships to CRC prognosis in cases with different MMR statuses, potentially inducing the U-shaped association in CRC. The opposing correlations between mtDNA-CN and T lymphocyte infiltration in cases of dMMR CRC and pMMR CRC further suggest that mtDNA-CN might play an important role in CRC development. More importantly, cases of pMMR CRC with lower mtDNA-CN and of dMMR CRC with higher mtDNA-CN can benefit more dramatically from ICIs. Furthermore, RNA-seq revealed a link between the level of mtDNA-CN and T lymphocyte infiltration in CRC cases with different MMR statuses.
Conclusion
Our study found a potential relationship between mtDNA-CN and CRC development that differs by MMR status, potentially providing a rationale for the use of mtDNA-CN as both a predictive biomarker and a therapeutic target for ICIs.
Subject terms: Cell biology, Oncology
Introduction
At present, the occurrence of colorectal cancer (CRC) has serious effects on human health [1]. The development of immune checkpoint inhibitors (ICIs) has produced unprecedented clinical outcomes for patients with microsatellite instable/mismatch repair-deficient (MSI/dMMR) CRC [2, 3]. CRC with dMMR account for 15% of all CRC [4]. Unfortunately, the vast majority of patients with microsatellite stable/mismatch repair-proficient (MSS/pMMR) CRC do not benefit from immunotherapy due to poor T-cell infiltration [5, 6]. Therefore, it is important to gain a better understanding of the regulatory mechanisms underlying cytotoxic T-cell infiltration.
Mitochondria are critical metabolic organelles involved in metabolism alterations and cellular functions [7, 8]. Increasing evidence has demonstrated that altered metabolism affects TME reprogramming, specifically enhancing immune cell infiltration [9, 10]. Changes in mitochondrial DNA copy number (mtDNA-CN) reflect mitochondrial biogenesis and function, serving as a substitute index of mitochondrial activity [11]. Although the mechanisms regulating mtDNA quantity have not been fully clarified, it has been found that mtDNA-CN affects numerous cellular pathways related to cancer [11]. Previous studies have demonstrated that mtDNA-CN is much lower in MSI CRC than in MSS CRC [12]. The regulation of transcription factor A truncating mutations (TFAM) leads to mtDNA-CN changes and mitochondrial instability, distinguishing most cases of dMMR CRC from pMMR CRC [13]. Therefore, the metabolic patterns of MSI CRC and MSS CRC are distinct during malignant progression [13, 14]. Consequently, the occurrence and development of CRC are likely affected by mtDNA-CN changes.
However, the relationship between mtDNA-CN and T-lymphocyte infiltration in CRC is still unclear. Previous studies have reported that down-regulation of TFAM leads to cytoplasmic leakage of mtDNA and activation of the cGAS/STING pathway, promoting T-lymphocyte infiltration [15]. In this way, mtDNA changes could potentially affect T-lymphocyte infiltration in CRC. It is important to clarify the potential regulatory mechanisms behind cytotoxic T-lymphocyte infiltration in order to develop effective strategies for tumour immunotherapy. Therefore, the purpose of this study was to explore the correlation between mtDNA-CN and T-lymphocyte infiltration in CRC patients with different MMR statuses so as to fill the relevant knowledge gap.
Materials and methods
Patient samples and cell lines
A total of 532 CRC tissue specimens, including 274 pMMR cases and 258 dMMR cases, were obtained by the Sixth Affiliated Hospital of Sun Yat-sen University, China. Among these 532 CRC patients, there are 72 patients (13.5%) with Stage I, 227 patients (42.7%) with Stage II, 173 patients (32.5%) with Stage III and 60 patients (11.3%) with Stage IV. All human tissue samples were collected with written informed consent from donors. In addition, 485 CRC patients (265 pMMR and 220 dMMR patients) who had received surgery at the Sixth Affiliated Hospital of Sun Yat-sen University provided formalin-fixed, paraffin-embedded (FFPE) CRC tissue samples. All paraffin sections for immunohistochemistry experiments were freshly cut, and all sections were obtained from the same batch. Samples were stored for a median 96 months (range: 41–133). Six CRC cell lines, including three with MSI (LoVo, HCT116 and RKO) and three with MSS (SW480, SW620 and HT29), were obtained from the American Type Culture Collection. Cells were cultured according to standard protocol in Dulbecco’s Modified Eagle Medium (DMEM; Gibco, Thermo Fisher Scientific, St. Peters, MO, USA) mixed with 10% foetal bovine serum (FBS; Gibco, Thermo Fisher Scientific, St. Peters, MO, USA) in a 5% CO2 atmosphere. All cell lines were validated by STR DNA fingerprinting. All procedures were conducted with the permission of the Institutional Review Board of the Sixth Affiliated Hospital of Sun Yat-sen University (Project number: E2021156).
Measurement of mitochondrial membrane potential
Mitochondrial membrane potential was measured using the enhanced mitochondrial membrane potential assay kit with JC-1 (Beyotime) according to the manufacturer’s protocol. Briefly, cells were seeded in a confocal microscope dish and cultured for 24 h. An equal volume of JC-1 was added, and cells were incubated for 20 min. Cells were then washed twice with PBS and, after the addition of cell culture medium, observed under a laser confocal microscope. To measure the fluorescence intensity, excitation/emission wavelengths were set to 490/530 nm for JC-1 monomers and to 525/590 nm for JC-1 aggregates.
DNA isolation
Thirty mg sections of tumour tissue were cut for DNA isolation. Genomic DNA was extracted from macro-dissected tumour tissue using a DNA isolation kit (Omega). For DNA isolation of biopsy specimens, a paraffin-embedded tissue DNA extraction kit (Omega) was used. Purity of DNA samples was assessed using Nanodrop technology before mtDNA-CN assessment, and samples without sufficient DNA yield were not included in mtDNA-CN analysis.
Determination of mtDNA-CN
mtDNA-CN was determined using an Applied Biosystems 7500 Fast Real-Time PCR System (Thermo Fisher Scientific). The DNA input was 100 ng, and all samples were measured in triplicate. The mean value was used as the mtDNA-CN score for each individual. The primer sequences used were ND1-F, 5’-CCCTAAAACCCGCCACATCT-3’ and ND1-R, 5’-GAGCGATGGTGAGAGCTAAGGT-3’ for mitochondrial NADH dehydrogenase 1, and HGB-F, 5’-GTGCACCTGACTCCTGAGGAGA-3’ and HGB-R, 5’-CCTTGATACCAACCTGCCCAG-3’, for Haemoglobin (nuclear DNA control), respectively. Quantitative real-time PCR (qPT-PCR) reactions were performed at 95 °C for 30 s followed by 40 cycles of 95 °C for 5 s and 60 °C for 30 s. The copy number was calculated based on threshold cycle values (Ct values) using the 2^ΔCT method.
Immunohistochemistry and IHC scoring
Paraffin-embedded tissues were deparaffinized with dimethylbenzene, followed by antigen retrieval. Tissues were blocked with normal goat serum at 37 °C for 30 min. Next, tissues were incubated overnight at 4 °C with specific primary antibodies. The following primary antibodies were used: anti-CD3 antibody (dilution Cat#ab135372; Abcam) and anti-CD8 antibody (Cat#NBP2-29475; Novus). Finally, tissues were incubated with appropriate secondary antibodies, then incubated with 3, 3′-diaminobenzidine (DAB) complex and counterstained with haematoxylin. Positive CD3 and CD8 densities per patient were determined by counting the number of positive cells per tissue sample using tumour tissue sections, as described previously [16].
Immunofluorescence
For immunofluorescence analysis, cells grown on a confocal microscope dish were incubated with 100 nmol/L MitoTracker Red CMVRos (Invitrogen) for 20 min. Cells were washed, fixed, and blocked in 1% bovine serum albumin-PBS. DAPI staining was performed on cells, which were then washed in PBS. Microscopy was performed using a Zeiss Confocal Laser Scanning Microscope.
Fluorescence in situ hybridisation
Dual-colour fluorescence in situ hybridisation was performed on FFPE tissue samples. After deparaffinization and dehydration, sections of FFPE tissue were treated with protease K, heated in a water bath at 42 °C for 1 h, and washed three times with 0.1% Triton X-100 2×saline-sodium citrate buffer (SSC). The probe was labelled with a human mitochondrial full-length sequence plasmid, and the NICK labelling method was used. Fluorescence microscopy was used to assess labelling efficiency, and agarose gel electrophoresis was used to assess probe length. Samples and probes were incubated in a 90 °C water bath for 10 min, then quenched in ice water. The probe was mixed with hybridisation solution to a final concentration of 0.5 μg/μL. Then, samples were incubated in prehybridization solution for 1 h, after which the prehybridization solution was removed, and 50 μL probe was directly added. Samples were incubated overnight at 42 °C in solution. Next, hybridised slides were washed with 37 °C 50% formamide/2×SSC for 15 min, washed with 37 °C 2×SSC for 15 min, and then washed with 0.1% Triton X-100 1×SSC and 0.1% Triton X-100 0.5×SSC for 15 min. After the addition of 50 μL DAPI (10 ng/mL), samples were stained for 5 min. After being washed with 0.1% Triton X-100 1×SSC and 0.1% Triton X-100 0.5×SSC for 15 min, samples were detected under a fluorescence microscope. After slides were sealed, pictures were collected using a Zeiss Confocal Laser Scanning Microscope.
CRC patient cohort treated with anti-PD-1
Inclusion criteria for patients included pathologic confirmation of CRC, at least 4 weeks of prior treatment with the anti-PD-1 drug pembrolizumab, adequate tumour biopsy samples, complete clinical information, and sufficient archival tissue available for analysis. Patient clinical, radiographic, and treatment data were obtained through the hospital information system. Objective clinical remission was defined according to the response evaluation criteria for solid tumours (RECIST), version 1.1. Ultimately, 30 patients were included in this study, with 11 pMMR cases and 19 dMMR cases.
RNA-Seq and data analysis
In order to explore the transcriptomic landscape changes associated with CRC by mtDNA-CN, six pMMR CRC and six dMMR CRC tissues were collected for RNA-Seq. Tumour samples were divided into four groups based on mtDNA-CN content and MMR status. Differential gene expressions (DEGs) were screened comparing dMMR tumour tissues with high mtDNA-CN to dMMR tumour tissues with low mtDNA-CN, and pMMR tumour tissues with low mtDNA-CN to pMMR tumour tissues with high mtDNA-CN. To study the biological roles of DEGs, Gene Ontology (GO) enrichment analysis was conducted for biological process (BP), cellular component (CC) and molecular function (MF). In addition, DEGs were subjected to Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis.
Statistical analysis
In this study, mtDNA-CN did not seem to be normally distributed (Shapiro–Wilk’s W = 0.8820, P < 0.0001) but skewed towards the right. Therefore, non-parametric statistical tests were used. The association of mtDNA-CN with clinicopathological characteristics (TNM stage, differentiation grade, tumour sublocation and gene mutation, etc.) were investigated using Mann–Whitney’s U and Kruskal–Wallis tests, where applicable. A P value < 0.05 was considered statistically significant for two-sided tests. Classification parameters, expressed as numbers and percentages, were tested by chi-square test. The Kaplan–Meier log-rank test was used to compare the effects of influencing factors on 5-year disease-free survival (DFS) and 5-year overall survival (OS) rates. Data analysis was performed using GraphPad Prism 9 and SPSS version 25.0. RNA-Seq correlation statistical analysis was performed using R version 3.6.3.
Results
CRC cases with different microsatellite and mismatch repair statuses have different mitochondrial contents and membrane potentials
In order to clarify differences in mitochondrial content and membrane potential in CRC cells with different microsatellite statuses, we conducted immunofluorescence and membrane potential experiments on three MSS and three MSI cell lines. Immunofluorescence analysis showed that the mitochondrial content of MSS CRC cells was significantly higher than that of MSI CRC cells (Supplementary Fig. S1A, B). Furthermore, the mitochondrial membrane potential levels of MSS CRC cells were significantly higher than those of MSI CRC cells (Supplementary Fig. S1C–E). As a whole, these results indicate that mitochondria might play different metabolic roles in CRC based on MMR status, such that the content and state of mitochondria are significantly different between dMMR CRC and pMMR CRC.
Correlation of mtDNA-CN with clinicopathological features and prognosis
We next analysed the distributions of mtDNA-CN and mtDNA-CN ratio (cancer tissue versus non-cancerous tissue) (Fig. 1a, b). The mtDNA-CN was lower for cancer tissues than for non-cancerous tissues (P = 0.0001) (Fig. 1c), and the mtDNA-CN of pMMR CRC tissue was significantly higher than that of dMMR CRC tissue (P < 0.0001) (Fig. 1d), but there was no statistical difference for non-cancerous tissues (P = 0.3634) (Fig. 1e). The mtDNA-CN is significantly reduced upon the deletion of any MMR protein, including MLH1 (P < 0.0001), MSH2 (P < 0.0001), MSH6 (P = 0.0274) and PMS2 (P < 0.0001) (Fig. 1f). The increase in mtDNA-CN was particularly marked for Stage T1–T2 (P = 0.0046) (Supplementary Fig. S2C), Stage N positive (P = 0.0283) (Supplementary Fig. S2D) and KRAS mutations (P = 0.0177) (Supplementary Fig. S2L). In addition, mtDNA-CN was higher in patients with rectal cancer (P = 0.0447 and P < 0.0001). No statistically significant differences were observed between mtDNA-CN and other clinicopathological features.
Fig. 1. Distribution characteristics of mtDNA-CN in CRC.
a The distribution of mtDNA-CN in CRC cases. b The mtDNA-CN ratio in tumour tissue versus in non-tumour tissues cases. c Comparison of mtDNA-CN between cancer tissues and corresponding non-cancerous tissues for each case. d Comparison of mtDNA-CN between dMMR cancer tissues and pMMR cancer tissues. e Comparison of mtDNA-CN between dMMR non-cancerous tissues and pMMR non-cancerous tissues. f Comparison of mtDNA-CN between cases with different MMR statuses. MtDNA mitochondrial DNA, DMMR mismatch repair-deficient, PMMR mismatch repair-proficient.
Tumour recurrence occurred in 138 of 485 patients (28.5%) (patients Stage IV were removed), and 132 of 532 patients (24.8%) died of disease during the follow-up period. When inspecting all HRs across quintile categories, the relation between mtDNA-CN and survival appears to be U-shaped. Kaplan–Meier curves show that the DFS rate differs significantly between mtDNA-CN quintiles (P = 0.0087) (Fig. 2c). Patients with first-quintile and fifth-quintile mtDNA-CN values exhibit the worst 5-year DFS rates. Moreover, when we analysed this data according to MMR status, we found that the DFS rates of CRC cases with different MMR statuses exhibit completely opposite relationships with mtDNA-CN. For CRC with pMMR, higher mtDNA-CN is associated with worse DFS (P = 0.0385) (Fig. 2e). In contrast, lower mtDNA-CN is associated with worse DFS in CRC with dMMR (P = 0.0023) (Fig. 2g). When we divided mtDNA-CN values into five equal groups, from low to high, we observed that dMMR CRC and pMMR CRC cases concentrated on opposite sides, with the concentrated side having higher recurrence (Fig. 2i) and death (Fig. 2j) rates. These results likely explain the U-shaped relationship observed between mtDNA-CN and prognosis across all CRC patients.
Fig. 2. Correlation between mtDNA-CN and CRC prognosis.
a Kaplan–Meier analysis of DFS based on mtDNA-CN in the CRC cohort. b Kaplan–Meier analysis of OS based on mtDNA-CN in the CRC cohort. c Kaplan–Meier analysis of DFS based on mtDNA-CN quintile in the CRC cohort. d Kaplan–Meier analysis of OS based on mtDNA-CN quintile in the CRC cohort. e Kaplan–Meier analysis of DFS based on mtDNA-CN in the CRC cohort with pMMR. f Kaplan–Meier analysis of OS based on mtDNA-CN in the CRC cohort with pMMR. g Kaplan–Meier analysis of DFS based on mtDNA-CN in the CRC cohort with dMMR. h Kaplan–Meier analysis of OS based on mtDNA-CN in the CRC cohort with dMMR. i With an increase in mtDNA-CN, the MMR status of CRC and disease recurrence in each quintile. j With an increase of mtDNA-CN, the MMR status of CRC and death in each quintile. MtDNA mitochondrial DNA, DMMR mismatch repair-deficient, PMMR mismatch repair-proficient.
Reverse correlations between mtDNA-CN and T-lymphocyte infiltration in colorectal cancers with different MMR statuses
To verify the potential association between mtDNA-CN and T-lymphocyte infiltration in CRC, we inspected the extent of T-lymphocyte infiltration in 485 CRC samples, including 265 pMMR and 220 dMMR cases. As in previous research, we examined CD3- and CD8-positive cells in two tumour regions: the tumour parenchyma (TP) and the invasive margin (IM) (Fig. 3a). The results showed that the T-lymphocyte infiltration of dMMR CRC is significantly higher than that of pMMR CRC (Fig. 3b), and lower mtDNA-CN values are associated with higher T-lymphocyte infiltration (Fig. 3c). Interestingly, the results show that T-lymphocyte infiltration has different correlations with mtDNA-CN in cases of CRC with different MMR statuses. With the increase in mtDNA-CN, T-lymphocyte infiltration decreased significantly in CRC with pMMR (Fig. 3d), exhibiting a significant negative correlation with T lymphocyte infiltration (Fig. 3f). As for dMMR CRC, T-lymphocyte infiltration increased with the increase in mtDNA-CN (Fig. 3e). We observed a positive correlation in CRC with dMMR (Fig. 3g). The correlation of mtDNA-CN and T-lymphocyte infiltration in pMMR is not as significant as that of CRC with dMMR. Considering that the changes in mtDNA-CN are more significant in dMMR CRC, this stronger linear association suggests that the alteration of mtDNA-CN might be associated with T-lymphocyte infiltration.
Fig. 3. Correlation between mtDNA-CN and T-lymphocyte infiltration in CRC.
a Immunostaining of CD3 and CD8 in tumour samples from representative human CRC cases of different mtDNA-CN with dMMR compared to pMMR. Scale bars: 100 μm (left), 20 μm (right). b Mismatch repair status is correlated with CD3- and CD8-positive densities in TP and IM for the CRC cohort. c MtDNA-CN is correlated with CD3- and CD8- positive densities in TP and IM for the CRC cohort. d MtDNA-CN is correlated with CD3- and CD8-positive densities in TP and IM for the CRC cohort with pMMR. e MtDNA-CN is correlated with CD3- and CD8-positive densities in TP and IM for the CRC cohort with dMMR. f CRC with pMMR shows a negative correlation of CD3 and CD8 with mtDNA-CN. g CRC with dMMR shows a positive correlation of CD3 and CD8 with mtDNA-CN. PMMR mismatch repair-proficient, DMMR mismatch repair-deficient, MtDNA mitochondrial DNA, TP tumour parenchyma, IM invasive margin.
Clinical response to an anti-PD-1 drug is correlated with mtDNA-CN
A total of 30 biopsy tissues from patients (11 cases with pMMR and 19 cases with dMMR) who received ICIs were tested for mtDNA-CN, and the clinicopathological information of patients is shown in Supplementary Fig. S3. We found that, for CRC patients with pMMR, ICI response cases exhibited significantly increased levels of CD3 and CD8 (Fig. 4a). In addition, mtDNA-CN decreased significantly in response cases (Fig. 4c). pMMR CRC patients with lower mtDNA-CN exhibited higher CD3 and CD8 levels (Fig. 4d), with a significant negative correlation (Fig. 4e, f). For CRC patients with dMMR, response cases treated with ICIs had significantly increased CD3 and CD8 levels (Fig. 4b). In addition, mtDNA-CN was significantly increased in response cases (Fig. 4g). dMMR CRC patients with higher mtDNA-CN (Fig. 4h) exhibited higher CD3 and CD8 levels, and this positive correlation was significant (Fig. 4i, j). The results were further confirmed by fluorescence in situ hybridisation (Fig. 4k).
Fig. 4. Clinical verification of the correlation between mtDNA-CN and ICI treatment efficacy.
a Difference in T-lymphocyte infiltration between pMMR CRC responders and non-responders with anti-PD-1 antibody treatment. b Difference in T-lymphocyte infiltration between dMMR CRC responders and non-responders with anti-PD-1 antibody treatment. c Difference in mtDNA-CN between pMMR CRC responders and non-responders with anti-PD-1 antibody treatment. d Correlation between T-lymphocyte infiltration and mtDNA-CN in CRC with pMMR. e Linear-regression correlation between CD3 and mtDNA-CN in CRC with pMMR. f Linear-regression correlation between CD8 and mtDNA-CN in CRC with pMMR. g Difference in mtDNA-CN between dMMR CRC responders and non-responders with anti-PD-1 antibody treatment. h Correlation between T-lymphocyte infiltration and mtDNA-CN in CRC with dMMR. i Linear-regression correlation between CD3 and mtDNA-CN in CRC with dMMR. j Linear-regression correlation between CD8 and mtDNA-CN in CRC with dMMR. k Fluorescence in situ hybridisation analysis of mtDNA (green) and DAPI (blue) in CRC cases with different MMR statuses, immunostaining of CD3 and CD8 in sections of tumour tissues by endoscopic biopsy, and magnetic resonance imaging (MRI) from CRC responders and non-responders with anti-PD-1 antibody treatment. Scale bars: 100 μm (left), 20 μm (right). PMMR mismatch repair-proficient, DMMR mismatch repair-deficient, MtDNA mitochondrial DNA.
Potential research directions
Out of 1029 differential gene expressions (DEGs) (dMMR tumour tissues with high mtDNA-CN vs. dMMR tumour tissues with low mtDNA-CN) across the nuclear genome (Fig. 5a), we identified 545 upregulated genes and 484 downregulated genes (Fig. 5c). The identified DEGs were enriched in several T-lymphocyte infiltration-related GO terms, including T-cell activation, T-cell differentiation, regulation of leukocyte activation and leukocyte migration, etc. KEGG-enriched pathways of identified DEGs included the chemokine signalling pathway, Th17 cell differentiation and Th1 and Th2 cell differentiation, among others. Among a total of 4629 differential gene expressions (DEGs) (pMMR tumour tissues with low mtDNA-CN vs. pMMR tumour tissues with high mtDNA-CN) across the nuclear genome (Fig. 5b), we identified 2899 upregulated genes and 1730 downregulated genes (Fig. 5d). The identified DEGs were enriched in several T-lymphocyte infiltration-related GO terms, including T-cell activation, leukocyte migration and regulation of leukocyte activation, etc. KEGG-enriched pathways of identified DEGs included the chemokine signalling pathway and leukocyte transendothelial migration, among others.
Fig. 5. Identification and functional analysis of DEGs.
a Heatmap of DEGs in dMMR tumour tissues with high mtDNA-CN vs. dMMR tumour tissues with low mtDNA-CN. b Heatmap of DEGs in pMMR tumour tissues with low mtDNA-CN vs. pMMR tumour tissues with high mtDNA-CN. c Volcano plot of differentially expressed genes obtained by DEseq2 analysis in dMMR with high mtDNA-CN vs. dMMR with low mtDNA-CN. d Volcano plot of differentially expressed genes obtained by DEseq2 analysis in pMMR tumour tissues with low mtDNA-CN vs. pMMR tumour tissues with high mtDNA-CN. e, f GO enrichment analysis and KEGG pathway enrichment analysis of DEGs from dMMR tumour tissues with high mtDNA-CN vs. dMMR tumour tissues with low mtDNA-CN. g, h GO enrichment analysis and KEGG pathway enrichment analysis of DEGs from pMMR tumour tissues with low mtDNA-CN vs. pMMR tumour tissues with high mtDNA-CN. DMMR mismatch repair-deficient, PMMR mismatch repair-proficient, MtDNA mitochondrial DNA.
Discussion
This study has two major important findings. First, we identify opposing relationships between mtDNA-CN and prognosis in CRC patients with different MMR statuses, contributing to a U-shaped relationship in CRC. Secondly, we report an inverse association between mtDNA-CN and T-lymphocyte infiltration in CRC patients with different MMR statuses and suggest that mtDNA-CN might be correlated with therapeutic response to ICI. Taken together, these findings offer important directions for the treatment of CRC.
Little is known about how mitochondrial alterations contribute to tumorigenesis. Warburg was the first to reveal that mitochondrial dysfunction triggers aerobic glycolysis in cancer cells [17]. Nevertheless, recent studies show that most tumour mitochondria are functional, and their metabolic processes are changed to support the rapid growth of cancer cells [18]. CRC with different MMR statuses have significantly different characteristics [19]. Several previous studies have demonstrated that CRC cases with dMMR rely on glycolysis, while CRC cases with pMMR favour oxidative phosphorylation to support their malignant status [13, 14]. As a substitute index for mitochondrial activity, mtDNA-CN might play distinct roles in CRC cases with different MMR statuses.
In previous studies, the reported relationships between mtDNA-CN and CRC prognosis have been mixed [20–22]. Interpreting the associations between mtDNA-CN and cancer survival may be complicated. We observed that mtDNA-CN has an inverse relationship with CRC prognosis depending on MMR status. Furthermore, the distribution and poor prognosis of CRC cases with different MMR statuses were concentrated on both extremes of mtDNA-CN values. This evidence might account for the U-shaped relationship between mtDNA and CRC prognosis. In fact, the U-shaped relationship between circulating mtDNA-CN and CRC has been observed in previous studies [23, 24], but its explanation is unclear. Given the correlation between circulating mtDNA-CN and tissue mtDNA-CN [25], our findings might partially explain this association. In addition, the increase in mtDNA-CN was particularly marked in Stages T1–T2, indicating that mtDNA-CN plays an important role during CRC initiation [26]. The progression of colorectal cancer eventually come to depend more on glycolysis and less on mitochondrial ATP synthesis mechanisms [27]. Moreover, stage N positive cases are associated with higher mtDNA-CN, so mtDNA-CN might be involved in the lymph node metastasis of CRC.
The more important question is whether mtDNA-CN is related to T-lymphocyte infiltration. T cells are the major contributors to antitumor immune response. As a unique molecule expressed by T cells, CD3 is able to translate the presence of specific antigens into intracellular signals that trigger anti-tumour immune responses [28, 29]. Moreover, CD8 + T cells recognise antigen CD3 molecules and eliminate tumours, primarily by inducing cell death through perforin granzyme and Fas/Fas ligand pathways [30]. Approximately 85% of patients with CRC have virtually no cytotoxic T-lymphocyte infiltration, and these patients usually have poor survival outcomes and high recurrence risk [31, 32]. However, the mechanisms governing T-lymphocyte infiltration in CRC remain largely unknown. Interestingly, our analysis shows that, in CRC cases with different MMR statuses, the degree of T-lymphocyte infiltration is inversely correlated with mtDNA-CN. It is worth noting that we found a highly significant linear correlation between mtDNA-CN and T-lymphocyte infiltration in CRC patients with dMMR. Compared with non-cancerous tissues, the change of mtDNA-CN in CRC with dMMR is more significant than that in CRC with pMMR. The more significant linear correlation of CRC with dMMR might be evidence that mtDNA-CN changes are related to T-lymphocyte infiltration. In addition, pMMR CRC patients who responded to ICI treatment had lower mtDNA-CN levels in biopsy tissues, while this phenomenon was completely reversed in cases of CRC with dMMR. This special correlation has never been reported in previous studies. Given the current difficulties in converting the CRC microenvironment from “cold” to “hot”, this finding might serve as a new breakthrough for researchers.
In order to better understand the potential relationship between mtDNA-CN and T-lymphocyte infiltration, we sequenced the transcriptomes of pMMR CRC and dMMR CRC tissues with different mtDNA-CN levels. The results of this analysis indicate that DEGs are enriched in cellular functions and pathways related to T-lymphocyte activities in dMMR CRC patients with high mtDNA-CN and in pMMR CRC patients with low mtDNA-CN. However, more work is still needed in the future to confirm specific regulatory genes and mechanisms.
The major findings of our study provide novel insights into the idea of enhancing T-lymphocyte infiltration to reverse immune escape in CRC. At present, only a few patients can benefit from ICI treatment, so there is an urgent need for promising biomarkers to identify treatment-sensitive patients in order to reduce expensive and unnecessary treatment. Our research shows that it might be clinically valuable to use mtDNA-CN as a biomarker to select patients likely to benefit from ICIs. The checkpoint and mechanisms that can trigger reprogramming of the tumour microenvironment have remained mysterious. Our study provides a new direction for the role of mtDNA-CN in TME reprogramming. Our results provide a rationale for the use of mtDNA-CN as both a predictive biomarker and a therapeutic target for ICIs in future clinical trials. It is undeniable that the biopsy specimens for clinical validation are limited. This limitation arises due to the relatively small proportion of dMMR CRC cases and the restricted number of pMMR CRC cases treated with ICIs. Additionally, biopsy specimens are primarily utilised for pathological diagnosis. Therefore, these findings need to be verified in a larger cohort in the future. And further prospective recruitment research samples were evaluated.
Conclusions
The mtDNA-CN plays different roles in CRC cases with different MMR statuses, leading to opposing relationships between mtDNA-CN and prognosis and T-lymphocyte infiltration. In addition, pMMR CRC cases with lower mtDNA-CN and dMMR CRC cases with higher mtDNA-CN are more likely to benefit from ICIs. Overall, mitochondrial regulation is a potentially novel immunotarget for CRC immunotherapy, and our findings offer direction for the treatment of CRC.
Supplementary information
Acknowledgements
This study was supported by National Key Clinical Discipline.
Author contributions
LH and LK contributed to the conception of the study and supervision. MC, HSL and WFL contributed to experiment, data collection, quality assessment and manuscript draft. PZH and FJY contributed to statistical analysis and interpretation of data. YBC, ZXL and LX contributed to the interpretation of data and revision of the manuscript draft. All authors have approved the final draft of the manuscript.
Funding
The programme of Guangdong Provincial Clinical Research Center for Digestive Diseases (2020B1111170004), Science and technology key research and development plan project of Guangzhou (China) (202103000072), Natural Science Foundation of Guangdong Province (China) (2018A030313621), Science and Technology Projects in Guangzhou (202206010062), Sun Yat-sen University Clinical Research 5010 Programme (2016005), Shenzhen “San Ming Projects” Research (Grant No.lc202002).
Data availability
The data from RNA-seq array in this study are available from the corresponding authors upon reasonable request.
Competing interests
The authors declare no competing interests.
Ethics approval and consent to participate
All human tissue samples were obtained by the Sixth Affiliated Hospital of Sun Yat-sen University and collected with written informed consent from donors. In addition, all procedures were conducted with the permission of the Institutional Review Board of the Sixth Affiliated Hospital of Sun Yat-sen University (Project number: E2021156).
Consent for publication
Not applicable.
Footnotes
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
These authors contributed equally: Mian Chen, Huashan Liu, Wenfeng Liang.
Contributor Information
Liang Kang, Email: kangl@mail.sysu.edu.cn.
Liang Huang, Email: huangl75@mail.sysu.edu.cn.
Supplementary information
The online version contains supplementary material available at 10.1038/s41416-023-02568-5.
References
- 1.Xie Y, Shi L, He X, Luo Y. Gastrointestinal cancers in China, the USA, and Europe. Gastroenterol Rep. 2021;9:91–104. doi: 10.1093/gastro/goab010. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Lizardo DY, Kuang C, Hao S, Yu J, Huang Y, Zhang L. Immunotherapy efficacy on mismatch repair-deficient colorectal cancer: from bench to bedside. Biochim Biophys Acta Rev Cancer. 2020;1874:188447. doi: 10.1016/j.bbcan.2020.188447. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Hu H, Kang L, Zhang J, Wu Z, Wang H, Huang M, et al. Neoadjuvant PD-1 blockade with toripalimab, with or without celecoxib, in mismatch repair-deficient or microsatellite instability-high, locally advanced, colorectal cancer (PICC): a single-centre, parallel-group, non-comparative, randomised, phase 2 trial. Lancet Gastroenterol Hepatol. 2022;7:38–48. doi: 10.1016/s2468-1253(21)00348-4. [DOI] [PubMed] [Google Scholar]
- 4.Emiloju OE, Sinicrope FA. Neoadjuvant immune checkpoint inhibitor therapy for localized deficient mismatch repair colorectal cancer: a review. JAMA Oncol. 2023; 10.1001/jamaoncol.2023.3323. [DOI] [PubMed]
- 5.Bruni D, Angell HK, Galon J. The immune contexture and Immunoscore in cancer prognosis and therapeutic efficacy. Nat Rev Cancer. 2020;20:662–80. doi: 10.1038/s41568-020-0285-7. [DOI] [PubMed] [Google Scholar]
- 6.Li S, Na R, Li X, Zhang Y, Zheng T. Targeting interleukin-17 enhances tumor response to immune checkpoint inhibitors in colorectal cancer. Biochim Biophys Acta Rev Cancer. 2022;1877:188758. doi: 10.1016/j.bbcan.2022.188758. [DOI] [PubMed] [Google Scholar]
- 7.Pavlova NN, Thompson CB. The emerging hallmarks of cancer metabolism. Cell Metab. 2016;23:27–47. doi: 10.1016/j.cmet.2015.12.006. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Gentric G, Mieulet V, Mechta-Grigoriou F. Heterogeneity in cancer metabolism: new concepts in an old field. Antioxid Redox Signal. 2017;26:462–85. doi: 10.1089/ars.2016.6750. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Siska PJ, Beckermann KE, Mason FM, Andrejeva G, Greenplate AR, Sendor AB, et al. Mitochondrial dysregulation and glycolytic insufficiency functionally impair CD8 T cells infiltrating human renal cell carcinoma. JCI Insight. 2017;2; 10.1172/jci.insight.93411. [DOI] [PMC free article] [PubMed]
- 10.Li X, Wenes M, Romero P, Huang SC, Fendt SM, Ho PC. Navigating metabolic pathways to enhance antitumour immunity and immunotherapy. Nat Rev Clin Oncol. 2019;16:425–41. doi: 10.1038/s41571-019-0203-7. [DOI] [PubMed] [Google Scholar]
- 11.Cheng SC, Quintin J, Cramer RA, Shepardson KM, Saeed S, Kumar V, et al. mTOR- and HIF-1α-mediated aerobic glycolysis as metabolic basis for trained immunity. Science. 2014;345:1250684. doi: 10.1126/science.1250684. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.van Osch FH, Voets AM, Schouten LJ, Gottschalk RW, Simons CC, van Engeland M, et al. Mitochondrial DNA copy number in colorectal cancer: between tissue comparisons, clinicopathological characteristics and survival. Carcinogenesis. 2015;36:1502–10. doi: 10.1093/carcin/bgv151. [DOI] [PubMed] [Google Scholar]
- 13.Guo J, Zheng L, Liu W, Wang X, Wang Z, Wang Z, et al. Frequent truncating mutation of TFAM induces mitochondrial DNA depletion and apoptotic resistance in microsatellite-unstable colorectal cancer. Cancer Res. 2011;71:2978–87. doi: 10.1158/0008-5472.Can-10-3482. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Sun X, Zhan L, Chen Y, Wang G, He L, Wang Q, et al. Increased mtDNA copy number promotes cancer progression by enhancing mitochondrial oxidative phosphorylation in microsatellite-stable colorectal cancer. Signal Transduct Target Ther. 2018;3:8. doi: 10.1038/s41392-018-0011-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Hu M, Zhou M, Bao X, Pan D, Jiao M, Liu X, et al. ATM inhibition enhances cancer immunotherapy by promoting mtDNA leakage and cGAS/STING activation. J Clin Investig. 2021;131; 10.1172/jci139333. [DOI] [PMC free article] [PubMed]
- 16.Tumeh PC, Harview CL, Yearley JH, Shintaku IP, Taylor EJ, Robert L, et al. PD-1 blockade induces responses by inhibiting adaptive immune resistance. Nature. 2014;515:568–71. doi: 10.1038/nature13954. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Warburg O. On the origin of cancer cells. Science. 1956;123:309–14. doi: 10.1126/science.123.3191.309. [DOI] [PubMed] [Google Scholar]
- 18.Ward PS, Thompson CB. Metabolic reprogramming: a cancer hallmark even Warburg did not anticipate. Cancer Cell. 2012;21:297–308. doi: 10.1016/j.ccr.2012.02.014. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.André T, Cohen R, Salem ME. Immune checkpoint blockade therapy in patients with colorectal cancer harboring microsatellite instability/mismatch repair deficiency in 2022. Am Soc Clin Oncol Educ Book. 2022;42:1–9. doi: 10.1200/edbk_349557. [DOI] [PubMed] [Google Scholar]
- 20.Cui H, Huang P, Wang Z, Zhang Y, Zhang Z, Xu W, et al. Association of decreased mitochondrial DNA content with the progression of colorectal cancer. BMC Cancer. 2013;13:110. doi: 10.1186/1471-2407-13-110. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Wang Y, He S, Zhu X, Qiao W, Zhang J. High copy number of mitochondrial DNA predicts poor prognosis in patients with advanced stage colon cancer. Int J Biol Markers. 2016;31:e382–e8. doi: 10.5301/jbm.5000211. [DOI] [PubMed] [Google Scholar]
- 22.Lee H, Cho JH, Park WJ, Jung SJ, Choi IJ, Lee JH. Loss of the association between telomere length and mitochondrial DNA copy number contribute to colorectal carcinogenesis. Pathol Oncol Res. 2018;24:323–8. doi: 10.1007/s12253-017-0245-z. [DOI] [PubMed] [Google Scholar]
- 23.Thyagarajan B, Wang R, Barcelo H, Koh WP, Yuan JM. Mitochondrial copy number is associated with colorectal cancer risk. Cancer Epidemiol Biomark Prev. 2012;21:1574–81. doi: 10.1158/1055-9965.Epi-12-0138-t. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Guan X, Li M, Bai Y, Feng Y, Li G, Wei W, et al. Associations of mitochondrial DNA copy number with incident risks of gastrointestinal cancers: a prospective case-cohort study. Mol Carcinog. 2023;62:224–35. doi: 10.1002/mc.23478. [DOI] [PubMed] [Google Scholar]
- 25.Qu F, Liu X, Zhou F, Yang H, Bao G, He X, et al. Association between mitochondrial DNA content in leukocytes and colorectal cancer risk: a case-control analysis. Cancer. 2011;117:3148–55. doi: 10.1002/cncr.25906. [DOI] [PubMed] [Google Scholar]
- 26.Feng S, Xiong L, Ji Z, Cheng W, Yang H. Correlation between increased copy number of mitochondrial DNA and clinicopathological stage in colorectal cancer. Oncol Lett. 2011;2:899–903. doi: 10.3892/ol.2011.322. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Ramanathan A, Wang C, Schreiber SL. Perturbational profiling of a cell-line model of tumorigenesis by using metabolic measurements. Proc Natl Acad Sci USA. 2005;102:5992–7. doi: 10.1073/pnas.0502267102. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Straus DB, Weiss A. The CD3 chains of the T cell antigen receptor associate with the ZAP-70 tyrosine kinase and are tyrosine phosphorylated after receptor stimulation. J Exp Med. 1993;178:1523–30. doi: 10.1084/jem.178.5.1523. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Malka D, Lièvre A, André T, Taïeb J, Ducreux M, Bibeau F. Immune scores in colorectal cancer: where are we? Eur J Cancer. 2020;140:105–18. doi: 10.1016/j.ejca.2020.08.024. [DOI] [PubMed] [Google Scholar]
- 30.Golstein P, Griffiths GM. An early history of T cell-mediated cytotoxicity. Nat Rev Immunol. 2018;18:527–35. doi: 10.1038/s41577-018-0009-3. [DOI] [PubMed] [Google Scholar]
- 31.Galon J, Costes A, Sanchez-Cabo F, Kirilovsky A, Mlecnik B, Lagorce-Pagès C, et al. Type, density, and location of immune cells within human colorectal tumors predict clinical outcome. Science. 2006;313:1960–4. doi: 10.1126/science.1129139. [DOI] [PubMed] [Google Scholar]
- 32.Pagès F, Mlecnik B, Marliot F, Bindea G, Ou FS, Bifulco C, et al. International validation of the consensus immunoscore for the classification of colon cancer: a prognostic and accuracy study. Lancet. 2018;391:2128–39. doi: 10.1016/s0140-6736(18)30789-x. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data from RNA-seq array in this study are available from the corresponding authors upon reasonable request.