リケラボ論文検索は、全国の大学リポジトリにある学位論文・教授論文を一括検索できる論文検索サービスです。

リケラボ 全国の大学リポジトリにある学位論文・教授論文を一括検索するならリケラボ論文検索大学・研究所にある論文を検索できる

リケラボ 全国の大学リポジトリにある学位論文・教授論文を一括検索するならリケラボ論文検索大学・研究所にある論文を検索できる

大学・研究所にある論文を検索できる 「Network Analysis of Gut Microbiota Including Fusobacterium and Oral Origin Bacteria and Their Distribution on Tumor Surface, Normal Mucosa, and in Feces in Patients with Colorectal Cancer」の論文概要。リケラボ論文検索は、全国の大学リポジトリにある学位論文・教授論文を一括検索できる論文検索サービスです。

コピーが完了しました

URLをコピーしました

論文の公開元へ論文の公開元へ
書き出し

Network Analysis of Gut Microbiota Including Fusobacterium and Oral Origin Bacteria and Their Distribution on Tumor Surface, Normal Mucosa, and in Feces in Patients with Colorectal Cancer

大橋, 彩子 名古屋大学

2023.07.04

概要

主論文の要旨

Network Analysis of Gut Microbiota Including
Fusobacterium and Oral Origin Bacteria and Their
Distribution on Tumor Surface, Normal Mucosa, and
in Feces in Patients with Colorectal Cancer
大腸癌患者における、Fusobacteriumと口腔内細菌を含む
腸内細菌のネットワーク解析と、腫瘍表面・正常粘膜・
便中の腸内細菌の分布について

名古屋大学大学院医学系研究科
病態内科学講座

総合医学専攻

消化器内科学分野

(指導:川嶋 啓揮
大橋 彩子

教授)

【緒言】
近年、大腸癌の発症・進展に複数の腸内細菌の関与が報告されている。Fusobacterium
はその代表で、発癌に関わる分子学的機序も解明が進んでいる。しかし、その他の細
菌については、その役割は十分に解明されていない。
我々は Fusobacterium とその他の腸内細菌が大腸癌の発癌に果たす役割を明らかに
することを目的とし、
「相関関係」と「分布」の二つの観点から、大腸癌患者の大腸内
における腸内細菌を検討した。
【対象及び方法】
2018 年 4 月から 2020 年 6 月に、名古屋大学医学部附属病院で内視鏡的粘膜下層剥
離術もしくは外科手術を受けるため、下部消化管内視鏡検査を施行した 25 例の大腸
癌患者を対象とした。各患者より、腸内細菌を便と粘膜から採取した。便検体は前処
置の前に採取し、粘膜検体は内視鏡検査時に、腫瘍表面・腫瘍近傍の正常粘膜・盲腸
の正常粘膜より、ブラシを用いて採取した。検体から DNA を抽出し、細菌叢を 16S
rRNA 遺伝子の V3–V4 領域をターゲットとした次世代シーケンス(MiSeq)にて解析し
た。細菌叢は QIIME2 で同定し、データベースは SILVA を用いた。
本研究では、既報にて大腸癌との関連が報告される腸内細菌を検討対象とし、採取
部位ごとにそれらの細菌どうしの相関関係をネットワーク解析で評価するとともに、
各細菌の相対存在量の、採取部位間の比較・相関から、各細菌の大腸内における分布
を考察した。ネットワーク解析における二つの細菌間の相対存在量の相関は、スピア
マンの順位相関係数で評価し、|r| > 0.3 かつ p < 0.0005 を有意な相関とした。各細菌
の相対存在量の、採取部位間の相関は、ピアソンの積率相関係数で評価し、p < 0.05
を有意な相関とした。
【結果】
患者背景は、大腸癌患者 25 例中、男性が 11 例、女性が 14 例で、年齢中央値は 73
才(44–80)だった。癌の深達度は、早期癌が 13 例、進行癌が 12 例だった(Table 1)。
細菌叢の構成は、便検体では Bacteroidetes 門が、ブラシ検体では Firmicutes 門が優
位だった。属レベルでは、全部で 260 種類の細菌が同定され、Fusobacterium の割合は
盲腸 0.079 %、腫瘍近傍 0.089 %、腫瘍表面 0.11 %、便 0.022 %だった。
本研究では、Song らと Saus らのレビュー論文にて大腸癌との関連が報告された腸
内細菌のうち、本研究にて実際に検出された細菌を属レベルで検討対象とした(Table 2)。
Figure 1 は、今回検討対象とした細菌の、各採取部位におけるネットワーク解析の
結果である。相対存在量に有意な相関のある細菌が線で結ばれている。本結果より、
すべての採取部位で複数の細菌から成るネットワークが形成されていることが明らか
となり、その相関の仕方は採取部位によって異なっていた。特に腫瘍表面では、
Fusobacterium がネットワーク内で最も相対存在量が大きく重要な細菌であることが
示 唆 さ れ 、 さ ら に 他 の 採 取 部 位 と 比 較 し 、 Fusobacterium と 7 種 類 の 口 腔 内 細 菌

-1-

(Streptococcus, Solobacterium, Gemella, Porphyromonas, Peptostreptococcus, Prevotella,
Parvimonas)が正の相関で密に結びつき、特徴的なネットワークを形成していることが
明らかとなった。
各細菌の各採取部位における相対存在量を比較すると、Streptococcus は便より腫瘍
表面に有意に多かったが、Fusobacterium を含む大部分の細菌は、便・腫瘍表面・腫瘍
近傍・盲腸に偏りなく存在していた。
Figure 2 は、Fusobacterium の相対存在量の、腫瘍表面と便・腫瘍近傍・盲腸との相
関である。腫瘍表面における Fusobacterium の相対存在量は、便(r = 0.862、p < 0.001)、
腫瘍近傍(r = 0.917、p < 0.001)、盲腸(r = 0.851、p < 0.001)のいずれとも有意な正の相
関を示した。先述の 7 種類の口腔内細菌は、いずれも腫瘍表面と腫瘍近傍では相対存
在量に有意な正の相関を示したが、腫瘍表面と便・腫瘍近傍・盲腸のすべてで有意な
相関を認めた細菌はなかった。
【考察】
腸内細菌の分布を考える上で、大腸粘膜の粘液層に注目することが重要である。一
般的に、大腸粘膜の表面には二層構造の粘液層があり、内層は細菌などが侵入できな
い物理的バリアとなっており、外層は常在細菌が定着する場となっている。本研究に
て、便検体は主に管腔内の細菌を、ブラシ検体は主に粘液層の細菌を反映すると考え
られる。
大腸癌ではこの粘液層の働きが破綻して粘液透過性が亢進し、大腸癌表面にバイオ
フィルムが形成されることが明らかとなっている。そして、バイオフィルムが発癌性
を持つことも証明されており、一部の大腸癌では、バイオフィルム中に Fusobacterium
が存在することが確認されている。しかし、Fusobacterium 以外にどのような細菌がバ
イオフィルムを構成しているかは、まだ分かっていない。
本研究のネットワーク解析にて、大腸癌表面における Fusobacterium と口腔内細菌
の相関関係が明らかとなり、大腸癌表面で Fusobacterium と相関する 7 種類の口腔内
細菌が同定された。Fusobacterium も口腔内細菌の一種であり、接着因子によって他の
口腔内細菌とプラークを形成することが知られている。そのため、Fusobacterium と 7
種類の口腔内細菌のネットワークは、大腸癌表面のバイオフィルムを反映している可
能性がある。
また本研究では、Fusobacterium のみが、腫瘍表面と便・腫瘍近傍・盲腸のいずれと
も相対存在量に有意な正の相関を示した。腫瘍表面に Fusobacterium が多い大腸癌患
者は、便や正常粘膜にも Fusobacterium が多く、Fusobacterium が大腸癌表面だけでな
く、大腸全体の粘膜・管腔内に同様に分布していることが示唆された。
【結語】
大腸癌患者の大腸内では、複数の腸内細菌が互いに相関関係を持っており、特に腫
瘍表面では Fusobacterium と 7 種類の口腔内細菌が正の相関のネットワークを形成し

-2-

ていた。これらの細菌の大腸内における分布は異なっており、Fusobacterium のみが、
大腸癌表面と大腸全体の粘膜・管腔内に同様に分布していた。Fusobacterium が広く分
布している大腸では、Fusobacterium が一次的に腫瘍表面に接着して発癌を促し、他の
口腔内細菌は、腫瘍表面で Fusobacterium と二次的に結びつくことで、発癌に適した
局所環境を形成している可能性が示唆された。

-3-

Table 1. Patient characteristics
Patient
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25

Sex
female
male
female
male
male
female
female
male
female
female
male
female
male
male
male
female
female
female
female
male
female
female
female
male
male

Age (years)
79
67
74
76
44
74
73
68
61
80
71
71
64
76
70
79
60
76
66
71
77
71
74
75
79

Location
C
T
R
A
A
C
R
R
T
T
S
T
R
R
D
R
R
A
S
A
R
R
R
S
S

Invasion depth
Tis
Tis
T2
T1b
Tis
T1a
Tis
T2
Tis
T2
T2
T3
T2
T2
T2
T2
Tis
Tis
T1b
Tis
T1b
T1b
T3
T2
T3

Pathology
tub1
tub1
tub1
tub1
tub2
tub2
tub1
tub1
tub1
tub2
tub2
tub1
tub1
tub2
tub2
tub2
tub2
tub1
tub1
pap
tub2
tub2
tub2
tub2
tub2

C: cecum, A: ascending colon, T: transverse colon, D: descending colon, S: sigmoid colon, R: rectum
Invasion depth follows the TNM classification. Tis: mucosa, T1a: submucosa < 1000 μm, T1b: submucosa > 1000 μm,
T2: muscularis propria, T3: subserosa
tub1: well differentiated tubular adenocarcinoma, tub2: moderately differentiated tubular adenocarcinoma, pap:
papillary adenocarcinoma

-4-

-5-

25
19
9
10
1
22
15
16
7
7
0
18
15
10
8
4
0
14
4
2
3
1
0
1

20
19
16
6
2
16
16
17
10
1
19
3
21
20
0

25
19
7
11
0
22
15
20
7
8
1
14
13
11
8
5
2
11
5
1
2
1
1
1

17
20
19
6
7
17
17
14
12
3
23
0
21
19
1

19
19
18
7
6
20
19
17
12
5
22
2
24
20
1

25
19
15
15
1
23
19
18
11
10
0
19
17
16
14
5
4
16
6
5
4
2
1
1
25
22
22
15
13
24
21
19
18
7
23
1
24
21
7

25
18
9
4
1
23
7
18
9
12
1
19
20
2
22
6
1
19
2
6
5
0
4
1
1.17 ± 2.13
3.35 ± 3.44
0.86 ± 0.87
0.06 ± 0.23
0.02 ± 0.05
0.78 ± 1.03
1.04 ± 1.26
0.73 ± 1.18
0.42 ± 0.71
0.01 ± 0.04
3.70 ± 5.25
0±0
3.72 ± 4.17
5.03 ± 8.78
0.43×10-3 ± 0.22×10-2

17.62 ± 11.06
5.72 ± 12.79
0.07 ± 0.17
0.28 ± 0.62
0±0
2.77 ± 2.73
0.40 ± 0.55
3.65 ± 7.35
0.06 ± 0.17
0.24 ± 2.31
0.96×10-4 ± 0.48×10-3
1.10 ± 2.76
0.45 ± 1.64
0.12 ± 0.22
0.06 ± 0.22
0.01 ± 0.21
0.85×10-2 ± 0.96×10-2
0.22 ± 0.36
0.05 ± 0.15
0.16×10-2 ± 0.83×10-2
0.02 ± 0.09
0.92×10-2 ± 0.05
0.22×10-3 ± 0.11×10-2
0.71×10-2 ± 0.04

cecum

** Mean ± standard deviation

Detection number (n) *
near
lesion
feces

* Number of patients in which each bacterium was detected

Enriched in CRC
Bacteroides
Fusobacterium
Peptostreptococcus
Porphyromonas
Mogibacterium
Streptococcus
Gemella
Prevotella
Parvimonas
Akkermansia
Methanobrevibacter
Clostridium
Alistipes
Solobacterium
Bilophila
Desulfovibrio
Staphylococcus
Collinsella
Helicobacter
Slackia
Pseudomonas
Acinetobacter
Lactococcus
Providencia
Depleted in CRC
Bifidobacterium
Faecalibacterium
Blautia
Butyricimonas
Lactobacillus
Eubacterium
Lachnospiraceae
Roseburia
Coprococcus
Enterococcus
Veillonella
Treponema
Ruminococcus
Anaerostipes
Bacillus

cecum

Table 2. Bacteria targeted in our study

0.84 ± 1.13
3.10 ± 2.92
0.77 ± 1.41
0.04 ± 0.08
0.01 ± 0.03
0.93 ± 1.28
1.18 ± 1.20
0.68 ± 1.16
0.32 ± 0.49
0.04 ± 0.12
3.03 ± 3.92
0.02 ± 0.09
3.27 ± 3.24
4.57 ± 7.62
0.38×10-3 ± 0.19×10-2

1.02 ± 1.27
3.55 ± 3.96
0.41 ± 0.49
0.21 ± 0.25
0.06 ± 0.11
1.92 ± 2.16
0.62 ± 0.73
0.59 ± 0.99
0.27 ± 0.36
0.11 ± 0.29
2.27 ± 4.08
0.12×10-2 ± 0.62×10-2
0.89 ± 1.57
0.27 ± 2.04
0.02 ± 0.07

36.75 ± 18.87
4.38 ± 8.99
0.01 ± 0.89
0.21 ± 0.83
0.22×10-3 ± 0.11×10-2
0.48 ± 1.02
0.01 ± 0.22
6.65 ± 12.25
0.02 ± 0.46
2.93 ± 7.44
0.22×10-3 ± 0.11×10-2
0.30 ± 0.60
2.73 ± 2.80
0 ± 0.30
0.34 ± 0.33
0.07 ± 0.09
0.11×10-3 ± 0.02
0.23 ± 0.34
0.02 ± 0.06
0.85×10-2 ± 0.02
0.09 ± 0.44
0±0
0.02 ± 0.11
0.28×10-2 ± 0.01

feces

F, T, S
F, T, S
F, T, S
F, T, S
F, T, S
F, T
F, T
F, T
F, T
F, T
F
F
T
T
T

F, T, S
F, T, S
F, T, S
F, T, S
F, T, S
F, T
F, T
F, T
F, T
F, T
F, T
F
F
F
F
F
F
T
T
T
T
T
T
T

Sample***

*** Type of the sample in the two reviewed articles (F: feces, T: tissue, S: swab)

0.95 ± 1.43
3.15 ± 2.84
0.68 ± 1.21
0.06 ± 0.21
0.41×10-2 ± 0.02
0.74 ± 1.12
1.04 ± 1.14
0.87 ± 1.6
0.29 ± 0.52
0.01 ± 0.04
2.32 ± 3.65
0.01 ± 0.05
3.94 ± 5.74
4.76 ± 9.06
0 ± 0.92×10-4

17.45 ± 8.73
7.65 ± 11.02
1.22 ± 4.03
0.25 ± 0.55
0.94×10-3 ± 0.47×10-2
3.25 ± 4.04
0.34 ± 0.49
4.68 ± 7.36
0.44 ± 1.07
0.19 ± 0.51
0±0
1.20 ± 3.29
0.31 ± 0.43
0.25 ± 0.51
0.07 ± 0.10
0.01 ± 0.03
0.46×10-2 ± 0.49×10-2
0.24 ± 0.33
0.03 ± 0.09
0.01 ± 0.03
0.03 ± 0.10
0.14 ± 0.68
0.36×10-3 ± 0.18×10-2
0.26×10-2 ± 0.01

Relative abundance (%) **
lesion

19.45 ± 9.80
6.49 ± 11.94
0.42 ± 0.65
0.23 ± 0.61
0.12×10-2 ± 0.65×10-2
2.46 ± 2.68
0.31 ± 0.59
5.07 ± 9.31
0.30 ± 1.03
0.18 ± 0.50
0 ± 0.20×10-4
1.52 ± 3.64
0.39 ± 0.55
0.20 ± 0.51
0.05 ± 0.09
0.01 ± 0.02
0 ± 0.18×10-2
0.19 ± 0.24
0.16 ± 0.63
0.30×10-2 ± 0.01
0.02 ± 0.08
0.93×10-2 ± 0.05
0 ± 0.46×10-4
0.31×10-2 ± 0.02

near

Figure 1. Network of the bacteria based on the correlation of their relative abundance
The correlation between the relative abundance of every two bacteria in Table 2 was assessed using the Spearman’s
rank correlation coefficient and significantly correlated bacteria were connected by a line. Significant correlation was
defined at | r | of > 0.3 and p-value of < 0.0005. The blue line shows positive correlation and the red line shows negative
correlation. The line thickness reflects the correlation coefficient. The square surrounding a bacterium is colored by the
phylum and its width reflects the relative abundance of the bacterium. The light blue background color represents
“enriched bacteria” and pink represents “depleted bacteria”. Asterisks indicate oral origin bacteria. The order of the
bacteria is the same in each diagram but the bacteria not correlated with any bacteria are not depicted.
a: Network diagram in cecum (cecum). The number of genera belonging to the round network was the largest of the
four, but the number of correlation lines was small.
b: Network diagram in tumor-adjacent mucosa (near). Fusobacterium and oral origin bacteria were correlated on tumoradjacent mucosa, but the number of their genera was lower than on tumor surface.
c: Network diagram in tumor surface (lesion). Fusobacterium and the seven oral origin bacteria (Streptococcus,
Solobacterium, Gemella, Porphyromonas, Peptostreptococcus, Prevotella, and Parvimonas), which are all “enriched
bacteria”, were densely connected to form a positively correlated network.
d: Network diagram in feces (feces). The number of red lines was higher in fecal samples than in brush samples.

-6-

Figure 2. Correlation of the relative abundance of Fusobacterium between each sampling site
Fusobacterium showed significant positive correlations between fecal and tumor brush samples (a), tumor-adjacent
and tumor brush samples (b), and cecum and tumor brush samples (c). The correlation was assessed using the Pearson
correlation coefficient and significant difference was defined at p-value of < 0.05.

-7-

全国の大学の
卒論・修論・学位論文

一発検索!

この論文の関連論文を見る