Tumor-associated microbiota in colorectal cancer with vascular tumor-thrombus and neural invasion and association with clinical prognosis

doi:10.21203/rs.3.rs-2337601/v1

Download PDF

Research Article

Tumor-associated microbiota in colorectal cancer with vascular tumor-thrombus and neural invasion and association with clinical prognosis

https://doi.org/10.21203/rs.3.rs-2337601/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Background

Neural invasion (NI) and vascular tumor thrombus (VT) are associated with poor prognosis in patients with colorectal cancer (CRC). But little is known about the changes of gut microbiota in CRC tissues with invasion. In this study, we applied 16S rRNA amplicon sequencing to tumor tissues and adjacent normal tissues in patients with CRC to determine the microbial differences.

Results

A discovery cohort (n = 88), including 30 patients with NI, 23 with VT, and 35 with double-negative CRC tissue, was utilized. We analyzed the relationship between the differential bacterial taxa and survival prognosis in an analysis cohort (n = 82) and explored the association of the special microorganisms with the clinical characteristics in an expanded cohort (n = 286) as well as their association with the survival prognosis in the overall cohort (n = 329). In the discovery cohort, the diversity, composition, and metabolic pathways of the gut microbiome distinctly differed between the tumor and non-tumor tissues in the NI and VT groups. Compared with the species richness of the microbiota communities of the noninvasive tumor tissues of the double-negative group, that of the tumor tissues of the NI and VT groups decreased. At the genus level, the four dominant bacteria in the NI and VT groups were Cupriavidus, Acinetobacter, Sphingobium, and Thermus. A high abundance of Cupriavidus was found to be related to a short survival time of NI CRC in the Kaplan–Meier analysis. Simultaneously, Herbaspirillum was a potential microbial biomarker predicting the prognosis of patients with CRC with NI or VT. Moreover, an abundance of Cupriavidus was associated with the NI status, pN stage, and clinical stage, while that of Herbaspirillum was related to the tumor diameter, NI status, pN stage, distant metastasis, and clinical stage. In the overall cohort, a high abundance of Cupriavidus and Herbaspirillum also indicated a poor prognosis.

Conclusion

This study is the first to comprehensively elaborate the differences in the gut microbiota of patients with CRC with different invasion statuses and to prove the relationship between some gut microbiota and clinical patient characteristics.

colorectal cancer

gut microbiota

16S rRNA

Cupriavidus

Herbaspirillum

Globally, colorectal cancer (CRC) is the third most common malignant cancer and the second most common cause of cancer-related deaths. Based on data from the Global Cancer Observatory, 1,880,725 new cases of CRC and 915,880 CRC-related deaths were recorded in 2020 [1]. Although the treatment strategies for CRC have been improved in the past few years, the overall mortality remains high. In particular, distant invasion has long been a difficult challenge for clinicians because it is the most typical biological characteristic of malignant tumors. Approximately 50% of patients with CRC die of distant metastasis [2]. Therefore, finding specific biomarkers for invasive CRC, optimizing screening strategies and preventing metastasis are the focus of current research.

The vascular system is made up of three types of vessels: arterial, venous, and lymphatic. Invasion of the vascular system is widely regarded as the first stage in the formation of metastases. As an independent prognostic factor in patients with CRC, vascular tumor-thrombus (VT), also known as vascular invasion, includes both lymphatic and blood vascular invasions. CRC spreads as invasive tumor cells that can either enter the lymphatic system and be delivered to regional lymph nodes, or enter the blood vessels and spread throughout the body [3]. Histologically, vascular invasion is defined as the presence of tumor cells in the endothelium-lined luminal space or destruction of the vascular wall by tumor cells [4].

In contrast to direct spread and vascular metastasis, neural invasion (NI) invasively allows tumor cells to penetrate nerve fibers and disseminate along them. Fujita defined NI as tumor invasion of the Auerbach plexus in patients with CRC and identified it as a more important prognostic factor than vascular invasion [5]. Accordingly, NI has been added as a new criterion in the 7th edition of the tumor-node-metastasis classification of malignant tumors [6]. Previous study has shown that the severity of NI has a significant impact on patient prognosis in colon cancer [7]. Moreover, a significant prognostic potential of NI and a higher incidence of local recurrence with escalating severity scores of NI can be confirmed in patients with rectal cancer [8].

A large number of colonic microbiota studies suggest an association between CRC and infection with specific bacterial species, such as Escherichia coli [9]. The gut microbiota can also promote the metastasis of CRC through different mechanisms [10]. Although numerous studies on intestinal bacteria in patients with CRC have been undertaken, previous studies have not specifically addressed the probable changes across individuals with different invasion statuses. Limited data are known about the characteristics of the changes in the gut microbiota in aggressive CRC tissues.

In this study, we hypothesized that compared with double negative (DN) CRC tissues, there may be variations in the abundance and composition of the intestinal microbiome of NI or VT CRC tissues. We verified whether there are differences in the microbiota status between tumor tissues and adjacent normal tissues, discovered the different microbiota of tumor tissues with different invasive states and elucidated the relationship between specific microbiota and clinical characteristics of patients. Accordingly, we used 16SrRNA gene sequencing to evaluate tissue microbiota communities from patients with CRC who had NI or VT, and determine whether the difference in the intestinal microbiome is connected to survival. Clarifying these potential distinctions is crucial for developing new screening and prevention strategies, as well as for promoting personalized treatment for CRC in the future.

Study participants and sample collection

A total of 329 Chinese patients (with 329 tumor tissue samples and 329 adjacent non-tumor tissue samples) from Wuhan Union Hospital of Tongji Medical College, Huazhong University of Science and Technology, Central China, were enrolled in this study. Their data were collected from electronic medical records. Clinical data, including follow-up data and intestinal tissue samples of patients have been seriously evaluated and the inclusion criteria for tissue sample selection were the following: 1) patients with primary CRC who had undergone surgical therapy; 2) non-use of anti-biotics, nonsteroidal anti-inflammatory drugs, immunosuppressants, chemotherapy drugs, and other drugs affecting the gut microbiota 1 month before the collection of samples; and 3) availability of tumor and adjacent normal tissues. The exclusion criteria were as follows: history of alcoholism, smoking, and drug abuse, or unreasonable diet such as a high-fat diet that can disturb the gut microbiota composition [11].

Formalin-fixed, paraffin-embedded samples were used for the matched tumor and normal mucosal tissues. Five serial cuts sized 5 μm per sample were deposited in sterile containers and maintained at 23 ± 2 °C until 16S rRNA MiSeq was performed. All participants provided written informed consent before participation in the study. This study was approved by the Ethics Committee of Tongji Medical College of Huazhong University of Science and Technology (approval no. 2014-041).

DNA extraction, polymerase chain reaction amplification, and MiSeq

Genomic DNA was extracted from tissue samples using the Omega Mag-Bind Soil DNA Kit (Omega Bio-Tek, Norcross, GA, United States). NanoDrop (Thermo Fisher Scientific, Waltham, MA, United States) was utilized for DNA quantification. The size of DNA was checked via 0.8 agarose gel electrophoresis. All extracted DNA samples of the 329 patients were immediately stored at -80°C until analysis.

The extracted DNA was amplified using a series of primers targeting the V3-V4 variable region of the 16S rRNA gene, including the forward primer (5’-ACTCCTACGGGAGGCAGCA-3’) and the reverse primer (5’-GGACTACHVGGGTWTCTAAT-3’). The specific DNA barcode sequences for species identification were attached immediately after the adapter. Polymerase chain reaction (PCR) amplification was performed in a 25 µL reaction system containing 5 µL of 5 × reaction buffer, 5 µL of 5 × GC buffer, 2 µL of dNTP (2.5 mM), 1 µL of forward primer (10 µM), 1 µL of reverse primer (10 µM), 8.75 µL of ddH2O, 0.25 µL of Q5 DNA polymerase (NEB, M0491L), and 2 µL of DNA template. PCR was then performed using a PCR instrument (Applied Biosystems 2720) under the following conditions: 2 min at 98°C; 30 cycles of 15 s at 98°C, 30 s at 55°C, and 30 s at 72°C; and final extension for 5 min at 72°C.

Agarose gel electrophoresis was performed twice to detect the PCR products. Amplicons were purified again and quantified using the Quant-iT PicoGreen dsDNA Assay Kit (Microplate reader, Biotek, FLx800, Winooski, VT, United States). The DNA library was constructed using the TruSeq Nano DNA LT Library Prep Kit, and sequencing was conducted using Illumina MiSeq PE250.

16S rRNA data analysis

The raw reads were quality filtered using QIIME2 (Quantitative Insights Into Microbial Ecology, v2019.4). Error tags, chimera, and low-quality sequences were all removed from the analysis. The clean data were clustered into operational taxonomic units (OTUs) at 97% sequence similarity using USEARCH and the UPARSE-OTU algorithm [12]. To remove sample heterogeneity before analysis, we performed rarefaction on OTU table with QIIME2. Rarefaction allows users to standardize the data obtained from samples with different sequencing efforts, and to compare the OTU richness of the samples using this standardized platform. The ecological approach of rarefaction is to randomly sample the same number of OTUs from each sample so as to predict the observed OTUs and their relative abundance of each sample at the sequencing depth, and use these data to compare the communities at a given level of sampling effort. We selected the sequence with the highest abundance in each OTU as the representative sequence of each OTU. Using QIIME2 and the Greengenes database (release 13.8), the representative sequence and the corresponding abundance in the sample were taxonomically identified, which could be considered as a taxon [13]. The alpha and beta diversities, including rarefaction curve, microbial diversity presented by Chao1, Observed species and Good’s coverage diversity indices, permutational multivariate analysis of variance (PERMANOVA) according to Jaccard distance, principal coordinate analysis (PCoA) conducted by unweighted UniFrac distances, nonmetric multidimensional scaling (NMDS) and profiles of the microbial compositions were calculated with QIIME2 and R according to the relative abundance of OTUs [14]. We also used OTU abundance table to create Venn diagrams, and counted the number of members of each set according to their presence in each group using R software, which represented the number of OTU unique to each group or common among the groups.

The different taxonomies of the groups were identified using the linear discriminant analysis (LDA) effect size (LEfSe), and metabolic pathway prediction was performed by PICRUSt2 (Phylogenetic Investigation of Communities by Reconstruction of Unobserved States, v2.5.0) [15]. The analysis results of LEfSe were shown as a histogram of the LDA value distribution of the significantly different species, which was used to reveal the significantly enriched species and their importance in each group. LEfSe puts more emphasis on searching for stable differential species between groups, that is, only the differential species whose difference in abundance between two groups exceeds the LDA threshold can be considered as a biomarker. In general, the biomarker species with significant differences among groups can be found according to the highest relative abundance, LDA score and p value of each taxon in the comparison. One of its characteristics is that it is not only can analyse the difference of community composition in different sample groups, but also can select the indicator microbial groups that have the obvious performance in different subgroups. PICRUSt2 can generally predict the metabolic pathway in multiple functional databases and we used MetaCyc as the reference database for this time. Based on these data, species with significantly different abundance in the different groups were identiﬁed as biomarkers and the functional potential of bacteria communities could be predicted.

Grouping strategy

Fig. 1 provides the cohort selection flowchart. In this study, a total of 329 CRC matched tumor tissues and adjacent non-tumor tissues comprised the overall cohort collected from the Chinese patients. The above-indicated analysis was performed for all samples, and the annotation of all taxa and their abundance in all tissues, which were marked with the corresponding patient number, were obtained. After a strict pathological diagnosis, 30 paired NI CRC tissue samples (group NI), 23 paired VT CRC tissue samples (group VT), and 35 paired DN CRC tissue samples without NI and VT (group DN) were included as the discovery cohort. Based on the clinical data of the patients and the abundance of various microorganisms detected using QIIME2 in the tumor tissues, we selected all patients with VT or NI (31 and 51 patients, respectively) from the overall cohort to form a prognosis cohort and explored the relationship between high/low abundance of the microbiota and the prognosis of patients with the corresponding invasion status. Subsequently, we selected some microbiota found in the previous analysis and explored the association of their respective abundance in the tumor tissues of 286 patients as an expanded cohort. Finally, we further analyzed the relationship between the abundance of the microbiota and the survival outcomes of the total of 329 patients.

Statistical analysis

One-way analysis of variance and Chi-square test were used to compare demographics and tumor characteristics among patients with CRC with different invasive status in the discovery cohort. The results of 16S rRNA data analysis including the alpha and beta diversities were tested using QIIME2 and R. The post hoc tests were applied to evaluate the differences between tumor / non-tumor tissues among three groups of discovery cohort (Dunnett test for the alpha diversity and Benjamini-Hochberg method for the beta diversity analysis and LEfSe). The enrichment degree of differential microbiota was categorized into “high” and “low” using the median value obtained from the 16S rRNA MiSeq sequencing as cut-off point, while samples with undetectable microbiota expression were categorized as “negative”. Chi-square test, Fisher’s exact test, and t-test were used to compare the relationship between the abundance of selected microbiota and tumor characteristics. The Kaplan-Meier method was used to analyze the relationship between the abundances of microbiota and overall survival (OS) in the patients with CRC using R. All tests performed were two sided, and p or q values of < 0.05 were considered statistically significant.

Study design and patient demographic characteristics

There were a total of 329 CRC matched tumor tissues and adjacent non-tumor tissues collected from Chinese patients. After extracting DNA and detecting the enriched microbiota of all tissue samples by 16S rRNA Illumina MiSeq sequencing, the microbial diversity analysis, LEfSe and metabolic pathway prediction were performed in the discovery cohort which included 30 neural invasion patients with CRC (group NI), 23 vascular tumor-thrombus patients with CRC (group VT) and 35 double negative patients with CRC (group DN). Microbial community analysis was performed among the tumor tissues as well as among the non-tumor tissues so that we could observe the differences in the composition and structure of microorganisms clearly in tumor or non-tumor tissues with different invasive states and identified differential microbiota in the three groups. Then, patients in prognosis cohort were divided into two groups according to the abundance of the most abundant genera respectively, and the relationship between the abundance of several microbiota and the prognosis of patients with invasive status was discussed. Finally, we analyzed the correlation between the abundance of specific microbiota and the clinical characteristics of 286 patients in the clinical characteristics cohort, and illustrated the association of the abundance of specific microbiota in tumor tissues with the overall survival time of patients in the overall cohort (Fig. 1).

The clinical characteristics of the patients in the discovery cohort are shown in Table S1. There were no significant differences in age, gender, tumor diameter, location or differentiation among the three groups (all p > 0.05).

Estimation of sequencing depth

A data set of 203,358 high-quality sequences was obtained from 88 pairs of CRC tissue samples. The median length of each sample was 430 bp. Rarefaction curves based on Faith’s PD showed that the sequencing depth of all samples was close to saturation, indicating that the sequencing results are enough to reflect the diversity contained in the current samples (Fig. 2A). Among them, the genetic diversity of tumor tissues is higher than that of non-tumor tissues. On the one hand, a Venn diagrams illustrating the relationship among the groups showed that compared with DN tumor tissues, 1554 of 9248 OTUs were unique in VT tumor tissues and 1945 OTUs were unique in NI tumor tissues (Fig. 2B). On the other hand, a Venn diagram also showed that only 1725 of 8678 OTUs were unique in VT non-tumor tissues and 1878 OTUs were unique in NI non-tumor tissues (Fig. 2C).

Gut microbial alpha diversity

The alpha diversity indices of Chao1, Observed species and Good’s coverage, which mainly indicated the species richness in the sample, were analyzed by using QIIME2 for estimating the richness and diversity of microbial community in these samples. There were significant differences between the VT tumor tissues and adjacent non-tumor tissues in the Chao1, Observed species and Good’s coverage indices (p = 0.00063, 0.0034, and 0.00021, respectively) (Fig. 2D). The Chao1 and Good’s coverage indices also significantly differed between the NI tumor tissues and adjacent non-tumor tissues (p = 0.032 and 0.022, respectively), but not for the Observed species index (p = 0.7) (Fig. 2E). However, there were no significant differences found in these indices between the DN tumor tissues and adjacent non-tumor tissues (p = 0.73, 0.4, and 0.43, respectively) (Fig. 2F). These results indicate that the non-tumor samples have a greater species richness than tumor samples with VT and NI.

The Chao1 and Observed species indices also significantly differed between the VT and DN tumor tissues (p = 0.025 and 0.019, respectively), except in the non-tumor tissues. (Fig. 3A-B). Meanwhile, there were no significant differences between the NI and DN tissues as well as between the NI and VT tissues (all p > 0.05) (Fig. 3A-B).

Composition of the microbial community

The variations in the microbial compositions between paired samples from three groups were determined using PERMANOVA to measure beta diversity. The neural invasion, vascular tumor-thrombus and double negative tumor tissues and their adjacent non-tumor tissues had a statistically significant difference in beta diversity, according to Jaccard distance (p = 0.001, Table 1). However, the Jaccard distance did not significantly differ between the VT and DN tissues as well as between the NI and DN tissues (all p > 0.05, Table 1), but not between the VT and NI tumor tissues (p = 0.016 and q = 0.048, Table 1).

Table 1

Beta diversity assessed by Jaccard distances.
Metric	Model	F	R2	p-Value	q-Value
Jaccard distances
	Tumor vs. Non-tumor (vascular tumor-thrombus CRC)	2.891	0.062	0.000^***	0.000^***
	Tumor vs. Non-tumor (neural invasion CRC)	2.892	0.047	0.000^***	0.000^***
	Tumor vs. Non-tumor (double negative CRC)	2.617	0.037	0.000^***	0.000^***
	Vascular tumor-thrombus vs. Double negative (tumor)	1.132	0.020	0.193	0.290
	Vascular tumor-thrombus vs. Double negative (non-tumor)	1.121	0.020	0.212	0.636
	Neural invasion vs. Double negative (tumor)	1.017	0.017	0.362	0.362
	Neural invasion vs. Double negative (non-tumor)	0.886	0.014	0.636	0.636
	Vascular tumor-thrombus vs. Neural invasion (tumor)	1.436	0.027	0.016^*	0.048^*
	Vascular tumor-thrombus vs. Neural invasion (non-tumor)	0.958	0.018	0.477	0.636
The q-Value was determined with Benjamini-Hochberg method. CRC, colorectal cancer. ^p < 0.05; ^*p < 0.001; ^q < 0.05; ^***q < 0.001.

The PCoA analysis based on the unweighted UniFrac distance matrix also proved that significant clustering was shaped in the tumor tissues of the three groups, which was distinguished from the non-tumor tissues (Fig. S1A-C). We intuitively discovered the most different microbial species between the tumor and non-tumor tissues using LEfSe (Fig. S1D-F). When the LDA threshold was 4.0, p_Thermi, c_Deinococci, o_Thermales, f_Thermaceae, f_Oxalobacteraceae, g_Thermus and g_Cupriavidus were enriched in tumor tissues. Simultaneously, p_Proteobacteria c_Alphaproteobacteria, o_Sphingomonadales and f_Sphingomonadaceae were enriched in the non-tumor tissues compared with those in all tumor tissues. More interestingly, f_Pseudomonadaceae and g_Pseudomonas were highly enriched in the non-tumor tissues of the VT and DN groups, but not in the NI group.

Beta diversity analysis between the three groups was also evaluated through PCoA based on the unweighted UniFrac distance matrix and NMDS (Fig. S2). Similar to the alpha diversity analysis results, no significant clustering was observed between the NI and DN groups. Although there were differences in the alpha diversity, no significant clustering was shaped between the VT and DN groups.

At the phylum level, the five most fundamental phyla in the 88 CRC tumor tissue samples were Proteobacteria, Thermi, Firmicutes, Bacteroidetes and Actinobacteria, accounting for 98% of the total community (Fig. 4A). According to the relative abundance of the phyla in each group, the general microbial community composition could be clearly seen. At the genus level, the four dominant genera in the tumor tissues were Cupriavidus, Acinetobacter, Sphingobium and Thermus (Fig. 4B). To further evaluate the microbial community differences between the tumor and non-tumor tissues of the three groups, we conducted LEfSe. As expected, we did not find meaningful differences in the non-tumor tissues; the different microbial species in the tumor tissues are shown in Fig. 4C. When the LDA threshold was 2.0, the gut microbiota of the patients with NI was enriched with f_Pseudomonadaceae, g_Pseudomonas, g_Pelomonas, g_Ralstonia, g_Acidovorax and g_Enhydrobacter. In the DN CRC tumor tissues, increased abundances of o_Desulfovibrionales, f_Desulfovibrionaceae, g_Desulfovibrio and g_Subdoligranulum were observed. Only g_Gemmiger was enriched in the patients with VT. The highest relative abundance, LDA score, p value and q value of each taxon among the three groups were also described in Table S2. Furthermore, the different microbiota detected in LEfSe differed between the tumor tissues of the VT and NI groups, as also shown in the previous beta diversity analysis. A total of 18 enriched species were identified (Fig. 4D). g_Cupriavidus, f_Pseudomonadaceae and g_Pseudomonas were enriched in group NI, while g_Meiothermus and g_Gemmiger were enriched in group VT.

Metabolic pathway difference analysis between the tumor and non-tumor tissues

We analyzed the differences in the metabolic pathways between the tumor and adjacent non-tumor tissues using PICRUSt2 and MetaCyc in the discovery cohort. As shown in Table S3, there were fewer differential metabolic pathways in the VT group than in the NI and DN groups, and there was no significantly down-regulated metabolic pathway in the VT tumor tissues. Several metabolic pathways, such as mandelate degradation I, mandelate degradation to acetyl-CoA, and cytidine monophosphate-legionaminate biosynthesis I, were enriched in the tumor tissues of the NI and VT groups. Numerous metabolic pathways, including glyoxylate assimilation, 3-hydroxypropanoate cycle, and superpathway of the 3-hydroxypropanoate cycle, were significantly reduced in the tumor tissues of the DN and NI groups compared with those in the non-tumor tissues. Moreover, seven metabolic pathways, including mevalonate pathway I, superpathway of sulfolactate degradation, peptidoglycan biosynthesis, teichoic acid (poly-glycerol) biosynthesis, geranylgeranyl diphosphate biosynthesis and lactose, superpathway of glycerol degradation to 1, 3-propanediol, and galactose degradation, highly increased in the tumor tissues of the three groups. However, we did not find significantly different metabolic pathways among the tumor tissues of the three groups, which indicates that there is no unique metabolic pathway in tumor tissues with different invasion statuses.

Association of the microbial communities with outcomes and clinical characteristics

Based on the results of the three groups in the discovery cohort, we evaluated the association of high or low abundance microbiota in the tumor tissues and the survival of the patients with CRC with VT or NI in the prognosis cohort (31 and 51 patients, respectively). As shown in Table 2, we used the Kaplan–Meier method and found that among the top 18 highly abundant microbial communities at the genus level, the abundance of Herbaspirillum was significantly associated with the survival time of the patients with CRC with VT or NI (p = 0.026 and 0.014, respectively). A high abundance of Cupriavidus in the NI CRC tissues was associated with a significantly shorter survival time (p = 0.019). Even though Pseudomonas, Pelomonas, Ralstonia and Acidovorax were enriched in tumor tissues of the NI group, their abundance has no significant relationship with the prognosis of patients with neural invasion (all p > 0.05).

Table 2

Association of microbial communities with overall survival of neural invasion and vascular tumor-thrombus CRCs.
Microbial communities	OS in neural invasion patients with CRC (days)			OS in vascular tumor-thrombus patients with CRC (days)
Microbial communities	High	Low or negative	p-Value	High	Low or negative	p-Value
Cupriavidus	457	500	0.019^*	443	582	0.546
Acinetobacter	448	510	0.558	459	574	0.193
Sphingomonas	423	536	0.145	567	459	0.497
Thermus	465	492	0.224	481	550	0.180
Pseudomonas	482	474	0.717	480	552	0.591
Sphingobium	493	463	0.528	586	439	0.486
Brevundimonas	505	450	0.547	491	536	0.413
Massilia	460	497	0.180	454	580	0.707
Ochrobactrum	433	526	0.150	395	642	0.491
Shigella	535	420	0.553	486	545	0.157
Caulobacter	523	432	0.125	464	568	0.683
Lactobacillus	542	413	0.113	496	534	0.544
Arthrobacter	457	501	0.710	552	474	0.871
Herbaspirillum	433	525	0.014^*	485	546	0.026^*
Pelomonas	436	522	0.134	492	539	0.733
Aquabacterium	446	511	0.652	520	509	0.609
Acidovorax	486	470	0.734	427	608	0.151
Ralstonia	442	517	0.691	497	533	0.204
We define the level of bacterial taxa (high and low or negative) based on the median abundance of microbiota in tumor tissues. OS data are shown as the mean. CRC, colorectal cancer; OS, overall survival. ^*p < 0.05.

Considering that the small sample size in our previous analysis may not be widely representative, we verified the association of the status of the different taxa with the clinical characteristics of 286 patients with CRC. We selected four microbes mentioned in the previous experiment, including Cupriavidus, Herbaspirillum, Thermus and Pseudomonas. Interestingly, the status of Cupriavidus was associated with NI (p = 0.042, Table 3), pN stage (p = 0.006, Table 3) and clinical stage (p = 0.001, Table 3). An abundance of Herbaspirillum was also associated with the tumor diameter (p = 0.001, Table 3), NI (p = 0.011, Table 3), pN stage (p = 0.013, Table 3), distant metastasis (p = 0.001, Table 3) and clinical stage (p = 0.001, Table 3). However, there was no significant relationship between the abundance of Herbaspirillum and the incidence of VT (p = 0.214, Table 3). Moreover, the abundance of Thermus was related to the pN stage (p = 0.013, Table S4) and clinical stage (p = 0.043, Table S4). Although Pseudomonas was highly enriched in the tumor tissues of patients with NI, its abundance was not significantly correlated with the clinical characteristics of the patients (all p > 0.05, Table S4).

Table 3

Characteristics according to the relative abundance of Cupriavidus and Herbaspirillum in CRC tissues.
Characteristics	Cupriavidus-high	Cupriavidus-low	p-Value	Herbaspirillum-high	Herbaspirillum–low or negative	p-Value
	n = 143 (%)	n = 143 (%)		n = 143 (%)	n = 143 (%)
Age (years)	57.94 ± 12.93	55.07 ± 13.17	0.064	57.52 ± 12.99	55.49 ± 13.20	0.190
Gender			0.233			0.811
Male	76 (53)	86 (60)		82 (57)	80 (56)
Female	67 (47)	57 (40)		61 (43)	63 (44)
Tumor diameter (cm)	4.51 ± 2.24	4.86 ± 2.24	0.190	4.26 ± 2.08	5.11 ± 2.33	0.000^***
Neural invasion			0.042^*			0.011^*
Absent	90 (63)	106 (74)		88 (62)	108 (76)
Present	53 (37)	37 (26)		55 (38)	35 (24)
Vascular tumor-thrombus			0.890			0.214
Absent	109 (76)	108 (76)		104 (73)	113 (79)
Present	34 (24)	35 (24)		39 (27)	30 (21)
pT stage			0.657			1.000
T1 + T2	10 (7)	12 (8)		11 (8)	11 (8)
T3 + T4	133 (93)	131 (92)		132 (92)	132 (92)
pN stage			0.006^**			0.013^*
N0	63 (44)	86 (60)		64 (45)	85 (59)
N1 + N2	80 (56)	57 (40)		79 (55)	58 (41)
Distant metastasis			0.227			0.000^***
M0	112 (78)	120 (84)		105 (73)	127 (89)
M1	31 (22)	23 (16)		38 (27)	16 (11)
Clinical stage			0.000^***			0.000^***
1 + 2	50 (35)	77 (54)		48 (34)	79 (55)
3 + 4	93 (65)	66 (46)		93 (65)	66 (46)
The p-values for categorical and continuous variables were analyzed by Chi-square test and Fisher’s exact test, and continuous variables were analyzed by two independent-samples t-test. CRC, colorectal cancer. ^p < 0.05; ^p < 0.01; ^**p < 0.001.

Additionally, we tested the association of the different microbiota with the outcomes of the overall cohort based on detailed follow-up records through the Kaplan-Meier analysis. The median survival time of the patients with CRC with highly abundant Cupriavidus was significantly shorter than that of the patients with scarce Cupriavidus (p = 0.025, Fig. 5A). Meanwhile, there was a significant difference between Thermus-high patients and Thermus-low/negative patients (p = 0.019, Fig. 5B). However, in patients with different abundances of Pseudomonas in tumor tissues, there was no significant difference (p = 0.5, Fig. 5C). It can be also observed that the difference in the survival time between Herbaspirillum-high and Herbaspirillum-low/negative patients was statistically significant (p = 0.00031, Fig. 5D).

When compared to the gut microbiota of healthy people, the gut microbiota of people with CRC has a lower abundance of possibly protective taxa and a higher abundance of procarcinogenic taxa [16]. Riquelme found that a lower alpha-diversity in the tumor microbiome was associated with short-term survival in patients with pancreatic adenocarcinoma [17]. Among patients with gastric cancer, those with liver metastasis have a higher alpha diversity and worse prognosis [18]. Our study further illustrates that the total abundance of the gut microbiota in CRC tumor tissues with invasion is lower than that in adjacent non-tumor tissues. Tumors with VT have lower species richness of the gut microbiota than tumors with DN tissue, while the beta diversity does not differ. In contrast, there was a difference in the beta diversity between the tumor tissues of the VT and NI groups, but no difference in the alpha diversity. By using LEfSe, we also discovered that the gut microbiota of patients with NI was enriched with Pseudomonas, Pelomonas, Ralstonia, Acidovorax and Enhydrobacter at the genus level. And increased abundances of Desulfovibrio and Subdoligranulum were observed in DN CRC tumor tissues. Only Gemmiger was enriched in the patients with VT. Taken together, the effects of microbiota diversity on tumor growth and patient survival do not remain the same across different tumor types. However, our results are inconsistent with previous reports that Proteobacteria was enriched in tumor tissues and that tumor microbiomes showed a higher alpha diversity than did normal, patient-matched microbiomes [19]. These incompatible results may be attributed to the fact that patients come from different geographical environments and races and may have different dietary habits, which can affect the nutritional structure of the intestines [20, 21].

Cupriavidus is characterized as a gram-negative, motile, rod-shaped organism with oxidative metabolism and is found in both soil and clinical isolates, particularly in samples from debilitated patients [22]. Previous studies have shown that Cupriavidus is an organophosphorus pesticide-degrading microorganism, which is also highly enriched in the tumor tissues of patients with bladder cancer [23]. In our study, LEfSe showed that Cupriavidus had a large amount of aggregation in the tumor tissues; we proved for the first time that there was a significant correlation between the abundance of Cupriavidus in the tumor tissues and the survival time of patients with CRC, especially those with NI. Although Cupriavidus was not enriched in NI group, a high abundance of this genus was closely associated with the existence of adverse clinical characteristics, including NI and advanced clinical stage; this finding suggests that this genus plays an important role in the development of tumor, which may be an important focus for future prevention and treatment of NI CRCs. We speculate that the enzymes of this genus of bacteria play a crucial role in tumor tissues because this genus has been demonstrated to complete a variety of chemical reactions in vitro, including Mn-oxidizing and producing chemicals from CO₂ [24], however, the specific mechanisms of interaction between this bacterium and CRC need to be confirmed in subsequent studies.

Thermus, a bacterium belonging to the Thermi (also known as Deinococcota or Deinococcus-Thermus) phylum, is a genus of thermophilic bacteria [25]. It was revealed that the genus Thermus is more abundant in tissues from patients with advanced stages of lung cancer or HPV 16 infection [26, 27]. In this study, we found that Thermus was highly enriched in the tumor tissues compared with the non-tumor tissues and was related to the survival time as well as some clinical characteristics of the patients with CRC. Interestingly, in a previous study on chickens, Thermus was significantly abundant in low-body-weight chickens, which showed significantly upregulated inflammatory cytokines in the jejunum [28]. We speculate that the effect of Thermus in the human intestinal tract is also related to inflammatory responses, leading to damage of gut barrier integrity and thus affecting the nutrient absorption of the intestines. However, the role of this genus in the intestinal microbiota, especially in the formation and development of CRC, remains controversial. Although there was no significant relationship found between Thermus and the incidence of NI and VT in CRCs, the impact of this genus on the prognosis of patients requires further investigation.

Concurrently, we found that the abundance of Herbaspirillum, which is a bacterial taxon that was not enriched in the previous analysis, was associated with the survival time of the patients with CRC. As a gram-negative, strictly aerobic, curved or helical bacilli member of the Betaproteobacteria class, Herbaspirillum is a plant bacterium that is motile with polar flagella but not considered a human pathogen previously [29]. With advancements in detection technology, Herbaspirillum has been reported as a new opportunistic pathogen in patients with cancer or those who have undergone hematopoietic stem cell transplant [30]. To our knowledge, this study is the first to elucidate that the massive accumulation of Herbaspirillum in the tumor tissues of the intestines is related to the tumor diameter, NI, distant metastasis, and clinical stage. However, a smaller tumor diameter and a higher incidence of NI were associated with a higher abundance of Herbaspirillum, indicating that this genus of bacteria contributes to the invasion rather than the growth of tumor tissues. A high abundance of Herbaspirillum was also strongly associated with poor prognosis in the patients with VT but had no relationship with the incidence of VT, suggesting that Herbaspirillum infection is a risk factor leading to a short survival time in patients with VT. The virulence of Herbaspirillum may be related to flagella because flagellar movement could enhance the invasion of bacteria to the host and the movement could avoid harmful environments [29]. Faoro also found that the clinical strains of Herbaspirillum acquired new genes and genomic islands, including a cluster of genes related to lipopolysaccharide (LPS) biosynthesis that may be involved in host interactions [31]. Interestingly, the high enrichment of this genus in tumor tissues is consistent with previous findings that LPS is highly present in CRC tumor tissues and is known for its role in intestinal inflammation and progression of CRC [32]. Furthermore, Herbaspirillum infection may be caused by ingestion of undercooked vegetables or contaminated water, as this species is a water- and soil-based organism [33, 34]. Considering that most of our patients resided in places irrigated by rivers, we speculate that there was already a certain amount of Herbaspirillum in their intestines, which may explain why there was no enrichment of this genus in the previous LEfSe. Therefore, we believe that the condition of patients with CRC may be aggravated by severe Herbaspirillum infection, which can cause inflammatory reactions in tumor tissues through biosynthesis of LPS and can also produce host interactions through flagella movement. However, the detailed mechanism of this genus in the microenvironment of CRC tissues is still unknown, and relevant research needs to be continued in the future.

In previous studies, sulfide-producing pathways, cholesterol biosynthesis, and elements of central carbon metabolism were found to be closely related to CRC and gut microbiota [35, 36]. In this study, we also found a few metabolic pathways involved in the above biosynthesis, including the mevalonate pathway, superpathway of sulfolactate degradation and superpathway of glycerol degradation to 1, 3-propanediol, which were enriched in the tumor tissues. Mevalonate pathway enzymes are related to the gut microbiota, and antibiotics can effectively inhibit their concentration, reducing systemic and intestinal inflammations [37]. Therefore, the microbiota within CRC tumor tissues can cause an inflammatory response, which can contribute to the formation of the tumor microenvironment. Moreover, 2, 3-dihydroxypropane-1-sulfonate, one of the most abundant organic sulfur compounds in the biosphere, has been proven to be degraded by Cupriavidus, and its degradation products can impact human health [38, 39]. This may be one reason why the abundance of Cupriavidus in the tumor tissues was higher and correlated with the prognosis of the patients with CRC. Although the relationship between the metabolic pathways enriched in tumor tissues and the high abundance of bacteria found in this study is unknown, the findings provide a new basis for the investigation of CRC via metabolomics.

The strength of this study was that we collected a relatively large number of samples and integrated the basic information of tumor samples and patients. We also used the follow-up survival data to analyze the microbial differences between the tumor tissues and adjacent non-tumor tissues of the patients with CRC with NI or VT for the first time and to explore the relationship between the clinical characteristics (including survival time) of the patients and the abundance of different microbiota.

This study also had several limitations. First, potential DNA contamination, which is the most serious problem, may exist in the samples we analyzed. In particular, a previous study listed Cupriavidus and Herbaspirillum as contaminant genera [40]. To be exact, this article is aimed at the improper operation with the samples containing a low microbial biomass which is less than approximately 10³ to 10⁴ cells. Given our collection methods, careful histological review and samples containing relatively large microbiota, it is unlikely that the tissue samples from patients with CRC were contaminated. Besides, our validation study showed phyla and genera enriched in CRC tumor tissues, which are consistent with the previous study [41]. Second, we did not test additional microorganism specimens, such as saliva and feces, from the patients. Third, the patients were recruited from a single large-scale general hospital, and their information was evaluated retrospectively. Therefore, the geographical environment, dietary structure, and customs of patients may have impacted the results. However, considering the harm of NI and VT in human CRC and the complex composition of the gut microbiota, this retrospective study also provides several new results with important reference value. Prospective investigations are nevertheless needed in the future to verify the heterogeneity of the gut microbiota in patients with CRC with different metastasis status and the specific roles of the above-indicated genera in the tumor tissues. Furthermore, the gut microbiota has also been used to investigate whether it is related to unresponsiveness to cancer treatments, such as immunotherapy, in patients with cancer [42]. These and other studies have shown single microorganisms, consortia of microbes, and microbial metabolites that are being identified to increase the effectiveness of cancer treatments. Aside from CRC prevention, diagnosis, and treatment, potentially feasible opportunities exist with the use of microbiota profiling information in gastrointestinal microorganism-targeted therapies to decrease the toxicities of CRC therapeutics. We believe that the new findings revealed in this study can lay a solid foundation for the implementation of cell or animal experiments to clarify the mechanism of the interaction between the gut microbiota and invasive CRC. Eventually, future work may establish new biomarkers to judge the prognosis of patients and provide new treatments based on the application of probiotic strains.

CRC

colorectal cancer

double negative

DNA

deoxyribonucleic acid

DNn

double negative non-tumor tissues

DNt

double negative tumor tissues

LDA

linear discriminant analysis

LEfSe

linear discriminant analysis effect size

LPS

lipopolysaccharide

neural invasion

Nin

neural invasion non-tumor tissues

Nit

neural invasion tumor tissues

NMDS

nonmetric multidimensional scaling

overall survival

OTUs

operational taxonomic units

PCoA

principal coordinate analysis

PCR

polymerase chain reaction

PERMANOVA

permutational multivariate analysis of variance

PICRUSt

phylogenetic investigation of communities by reconstruction of unobserved states

QIIME

quantitative insights into microbial ecology

vascular tumor-thrombus

VTn

vascular tumor-thrombus non-tumor tissues

VTt

vascular tumor-thrombus tumor tissues

Ethics approval and consent to participate

This experimental protocol was established, according to the ethical guidelines of the Helsinki Declaration, and was approved by the Ethics Committee of Tongji Medical College of Huazhong University of Science and Technology (No. 2014-041). The patients/participants provided their written informed consent to participate in this study.

Consent for publication

Not applicable.

Availability of data and materials

Sequencing data of this study has been uploaded to the SRA database with the accession number from SRR12019460 continuity to SRR12020117 as well as from SRR20281361 continuity to SRR20281373. Additional patients’ clinical information with sample identification was presented in Supplementary Table S5. Further details are available on reasonable request to the corresponding author.

Competing interests

The authors declare that they have no competing interests.

Funding

This work was supported by the Beijing Bethune Public Welfare Foundation (No. BJ GYQZHX2021006); Chen Xiao-Ping foundation for the development of science and technology of Hubei province (No. CXPJJH122006-1003); the National Key R&D Program of China (No. 2018YFC1313300); the National Natural Science Foundation of China (No. 81702392 and No. 81472707); the Chinese South Western Oncology Group (No. CSWOG-CCET005).

Authors' contributions

All the authors took part in design of the study protocol, collection, analysis, and interpretation of the data. HLL and TZ proposed the conception, designed the study and critically reviewed the manuscript, MJL and MJ made substantial contributions to analysis, interpretation of data, and had the main responsibility of preparing the manuscript. LZ, DDY, LLS, and BZ provided assistance with the sample collection. LL provided statistical expertise. KLC provided helpful discussions. XN provided expert skills in histopathology. All the authors have reviewed the final version of the manuscript and approve it for publication.

Acknowledgements

We would like to thank doctors Lu Wen and Linli Shi for collecting samples, and Mingyue Cheng for critically reading the manuscript.

Supplementary material

The Supplementary Material for this article can be found in the following files.

Sung H, Ferlay J, Siegel RL, Laversanne M, Soerjomataram I, Jemal A, et al. Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA: A Cancer. J Clin. 2021;71(3):209–49. doi:10.3322/caac.21660.
Hong S, Li S, Bi M, Yu H, Yan Z, Liu T, et al. lncRNA ILF3-AS1 promotes proliferation and metastasis of colorectal cancer cells by recruiting histone methylase EZH2. Mol Ther Nucleic Acids. 2021;24:1012–23. doi:10.1016/j.omtn.2021.04.007.
van Wyk HC, Roxburgh CS, Horgan PG, Foulis AF, McMillan DC. The detection and role of lymphatic and blood vessel invasion in predicting survival in patients with node negative operable primary colorectal cancer. Crit Rev Oncol Hematol. 2014;90(1):77–90. doi:10.1016/j.critrevonc.2013.11.004.
Zhang K, Ren Y, Xu S, Lu W, Xie S, Qu J, et al. A clinical-radiomics model incorporating T2-weighted and diffusion-weighted magnetic resonance images predicts the existence of lymphovascular invasion / perineural invasion in patients with colorectal cancer. Med Phys. 2021;48(9):4872–82. doi:10.1002/mp.15001.
Fujita S, Nakanisi Y, Taniguchi H, Yamamoto S, Akasu T, Moriya Y, et al. Cancer invasion to Auerbach's plexus is an important prognostic factor in patients with pT3-pT4 colorectal cancer. Dis Colon Rectum. 2007;50(11):1860–6.
Sobin LH. TNM classification of malignant tumours. 7th ed. Oxford: Wiley-Blackwell; 2009.
Liebl F, Demir IE, Rosenberg R, Boldis A, Yildiz E, Kujundzic K, et al. The Severity of Neural Invasion Is Associated with Shortened Survival in Colon Cancer. Clin Cancer Res. 2013;19(1):50–61. doi:10.1158/1078-0432.ccr-12-2392.
Ceyhan GO, Liebl F, Maak M, Schuster T, Becker K, Langer R, et al. The severity of neural invasion is a crucial prognostic factor in rectal cancer independent of neoadjuvant radiochemotherapy. Ann Surg. 2010;252(5):797–804. doi:10.1097/SLA.0b013e3181fcab8d.
Wilson MR, Jiang Y, Villalta PW, Stornetta A, Boudreau PD, Carrá A, et al. The human gut bacterial genotoxin colibactin alkylates DNA. Science. 2019;363(6428); doi:10.1126/science.aar7785.
Chen S, Zhang L, Li M, Zhang Y, Sun M, Wang L, et al. Fusobacterium nucleatum reduces METTL3-mediated mA modification and contributes to colorectal cancer metastasis. Nat Commun. 2022;13(1):1248. doi:10.1038/s41467-022-28913-5.
Bai X, Wei H, Liu W, Coker OO, Gou H, Liu C, et al. Cigarette smoke promotes colorectal cancer through modulation of gut microbiota and related metabolites. Gut. 2022. doi:10.1136/gutjnl-2021-325021.
Edgar RC. Search and clustering orders of magnitude faster than BLAST. Bioinformatics. 2010;26(19):2460–1. doi:10.1093/bioinformatics/btq461.
Bolyen E, Rideout JR, Dillon MR, Bokulich NA, Abnet CC, Al-Ghalith GA, et al. Reproducible, interactive, scalable and extensible microbiome data science using QIIME 2. Nat Biotechnol. 2019;37(8):852–7. doi:10.1038/s41587-019-0209-9.
Jin M, Shang F, Wu J, Fan Q, Chen C, Fan J, et al. Tumor-Associated Microbiota in Proximal and Distal Colorectal Cancer and Their Relationships With Clinical Outcomes. Front Microbiol. 2021;12:727937. doi:10.3389/fmicb.2021.727937.
Langille MGI, Zaneveld J, Caporaso JG, McDonald D, Knights D, Reyes JA, et al. Predictive functional profiling of microbial communities using 16S rRNA marker gene sequences. Nat Biotechnol. 2013;31(9):814–21. doi:10.1038/nbt.2676.
Feng Q, Liang S, Jia H, Stadlmayr A, Tang L, Lan Z, et al. Gut microbiome development along the colorectal adenoma-carcinoma sequence. Nat Commun. 2015;6:6528. doi:10.1038/ncomms7528.
Riquelme E, Zhang Y, Zhang L, Montiel M, Zoltan M, Dong W, et al. Tumor Microbiome Diversity and Composition Influence Pancreatic Cancer Outcomes. Cell. 2019;178(4); doi:10.1016/j.cell.2019.07.008.
Yu D, Yang J, Jin M, Zhou B, Shi L, Zhao L, et al. Fecal Streptococcus Alteration Is Associated with Gastric Cancer Occurrence and Liver Metastasis. mBio. 2021;12(6):e0299421. doi:10.1128/mBio.02994-21.
Burns MB, Lynch J, Starr TK, Knights D, Blekhman R. Virulence genes are a signature of the microbiome in the colorectal tumor microenvironment. Genome Med. 2015;7(1):55. doi:10.1186/s13073-015-0177-8.
Liu N-N, Jiao N, Tan J-C, Wang Z, Wu D, Wang A-J, et al. Multi-kingdom microbiota analyses identify bacterial-fungal interactions and biomarkers of colorectal cancer across cohorts. Nat Microbiol. 2022;7(2):238–50. doi:10.1038/s41564-021-01030-7.
Mentella MC, Scaldaferri F, Ricci C, Gasbarrini A, Miggiano GAD. Cancer and Mediterranean Diet: A Review. Nutrients. 2019;11(9); doi:10.3390/nu11092059.
Vandamme P, Coenye T. Taxonomy of the genus Cupriavidus: a tale of lost and found. Int J Syst Evol Microbiol. 2004;54(Pt 6):2285–9. doi:10.1099/ijs.0.63247-0.
Liu F, Liu A, Lu X, Zhang Z, Xue Y, Xu J, et al. Dysbiosis signatures of the microbial profile in tissue from bladder cancer. Cancer Med. 2019;8(16):6904–14. doi:10.1002/cam4.2419.
Bai Y, Su J, Wen Q, Huang T, Chang Q, Ali A. Characterization and mechanism of Mn(II)-based mixotrophic denitrifying bacterium (Cupriavidus sp. HY129) in remediation of nitrate (NO-N) and manganese (Mn(II)) contaminated groundwater. J Hazard Mater. 2021;408:124414. doi:10.1016/j.jhazmat.2020.124414.
Oren A, Garrity GM. Valid publication of the names of forty-two phyla of prokaryotes. Int J Syst Evol Microbiol. 2021;71(10); doi:10.1099/ijsem.0.005056.
Huang X, Li C, Li F, Zhao J, Wan X, Wang K. Cervicovaginal microbiota composition correlates with the acquisition of high-risk human papillomavirus types. Int J Cancer. 2018;143(3):621–34. doi:10.1002/ijc.31342.
Yu G, Gail MH, Consonni D, Carugno M, Humphrys M, Pesatori AC, et al. Characterizing human lung tissue microbiota and its relationship to epidemiological and clinical features. Genome Biol. 2016;17(1):163. doi:10.1186/s13059-016-1021-1.
Zhang X, Akhtar M, Chen Y, Ma Z, Liang Y, Shi D, et al. Chicken jejunal microbiota improves growth performance by mitigating intestinal inflammation. Microbiome. 2022;10(1):107. doi:10.1186/s40168-022-01299-8.
Li X, Bao X, Qiao G, Wang L, Shi C, Chen S, et al. First Study of Bacteremia Caused by in China: A Brief Research Report and Literature Review. Front Cell Infect Microbiol. 2022;12:882827. doi:10.3389/fcimb.2022.882827.
Bloise I, Guedez-López GV, Tejedor-Rodríguez M, Romero-Gómez MP, García-Rodríguez J, Mingorance J, et al. Bloodstream infection due to Herbaspirillum sp.: case series and review of literature. Eur J Clin Microbiol Infect Dis. 2021;40(4):779–85. doi:10.1007/s10096-020-04075-4.
Faoro H, Oliveira WK, Weiss VA, Tadra-Sfeir MZ, Cardoso RL, Balsanelli E, et al. Genome comparison between clinical and environmental strains of Herbaspirillum seropedicae reveals a potential new emerging bacterium adapted to human hosts. BMC Genomics. 2019;20(1):630. doi:10.1186/s12864-019-5982-9.
Song W, Tiruthani K, Wang Y, Shen L, Hu M, Dorosheva O, et al. Trapping of Lipopolysaccharide to Promote Immunotherapy against Colorectal Cancer and Attenuate Liver Metastasis. Adv Mater. 2018;30(52):e1805007. doi:10.1002/adma.201805007.
Ziga ED, Druley T, Burnham C-AD. Herbaspirillum species bacteremia in a pediatric oncology patient. J Clin Microbiol. 2010;48(11):4320–1. doi:10.1128/JCM.01479-10.
Chen J, Su Z, Liu Y, Sandoghchian S, Zheng D, Wang S, et al. Herbaspirillum species: a potential pathogenic bacteria isolated from acute lymphoblastic leukemia patient. Curr Microbiol. 2011;62(1):331–3. doi:10.1007/s00284-010-9703-5.
Snaebjornsson MT, Janaki-Raman S, Schulze A. Greasing the Wheels of the Cancer Machine: The Role of Lipid Metabolism in Cancer. Cell Metab. 2020;31(1):62–76. doi:10.1016/j.cmet.2019.11.010.
Yachida S, Mizutani S, Shiroma H, Shiba S, Nakajima T, Sakamoto T, et al. Metagenomic and metabolomic analyses reveal distinct stage-specific phenotypes of the gut microbiota in colorectal cancer. Nat Med. 2019;25(6):968–76. doi:10.1038/s41591-019-0458-7.
Bourke CD, Gough EK, Pimundu G, Shonhai A, Berejena C, Terry L, et al. Cotrimoxazole reduces systemic inflammation in HIV infection by altering the gut microbiome and immune activation. Sci Transl Med. 2019;11(486); doi:10.1126/scitranslmed.aav0537.
Chen X, Liu L, Gao X, Dai X, Han Y, Chen Q, et al. Metabolism of chiral sulfonate compound 2,3-dihydroxypropane-1-sulfonate (DHPS) by Roseobacter bacteria in marine environment. Environ Int. 2021;157:106829. doi:10.1016/j.envint.2021.106829.
Denger K, Huhn T, Hollemeyer K, Schleheck D, Cook AM. Sulfoquinovose degraded by pure cultures of bacteria with release of C3-organosulfonates: complete degradation in two-member communities. FEMS Microbiol Lett. 2012;328(1):39–45. doi:10.1111/j.1574-6968.2011.02477.x.
Salter SJ, Cox MJ, Turek EM, Calus ST, Cookson WO, Moffatt MF, et al. Reagent and laboratory contamination can critically impact sequence-based microbiome analyses. BMC Biol. 2014;12:87. doi:10.1186/s12915-014-0087-z.
Fan Q, Shang F, Chen C, Zhou H, Fan J, Yang M, et al. Microbial Characteristics of Locally Advanced Rectal Cancer Patients After Neoadjuvant Chemoradiation Therapy According to Pathologic Response. Cancer Manage Res. 2021;13:2655–67. doi:10.2147/cmar.s294936.
Matson V, Fessler J, Bao R, Chongsuwat T, Zha Y, Alegre M-L, et al. The commensal microbiome is associated with anti-PD-1 efficacy in metastatic melanoma patients. Science. 2018;359(6371):104–8. doi:10.1126/science.aao3290.

No competing interests reported.

Download PDF

Version 1

posted

You are reading this latest preprint version

Tumor-associated microbiota in colorectal cancer with vascular tumor-thrombus and neural invasion and association with clinical prognosis

Status:

Version 1

Abstract

Background

Results

Conclusion

Figures

Background

Methods

Study participants and sample collection

DNA extraction, polymerase chain reaction amplification, and MiSeq

16S rRNA data analysis

Grouping strategy

Statistical analysis

Results

Study design and patient demographic characteristics

Estimation of sequencing depth

Gut microbial alpha diversity

Composition of the microbial community

Metabolic pathway difference analysis between the tumor and non-tumor tissues

Association of the microbial communities with outcomes and clinical characteristics

Discussion

Abbreviations

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1