Accepted: 20 October 2021
The landscape of prognostic and immunological role of myosin light chain 9 (MYL9) in human tumors
Minghe Lv1,2 Lumeng Luo1, 1,2 Xue Chen1,2 |
1Department of Radiation Oncology, Fudan University Shanghai Cancer Center, Fudan University, Shanghai, China
2Department of Oncology, Shanghai Medical College, Fudan University, Shanghai, China
Correspondence
Xue Chen, Department of Oncology, Shanghai Medical College, Fudan University, Shanghai 200032, China. Email: xuechen17@fudan.edu.cn
Abstract
Introduction: Recent studies have shown that myosin light chain 9 (MYL9) plays a vital role in immune infiltration, tumor invasion, and metastasis; however, the prognostic and immunological role of MYL9 has not been reported. The purpose of this study was to explore the potential prognostic and immunological roles of MYL9 in human cancers by public datasets mainly including the cancer genome atlas (TCGA) and Gene expression omnibus.
Methods: The expression pattern and prognostic value of MYL9 were ana- lyzed across multiple public datasets in different cancer. The correlations between MYL9 expression and immune infiltration among multiple cancers were analyzed by using the TIMER2.0. The MYL9-related gene enrichment analysis was implemented by mainly using KEGG and GO datasets.
Results: MYL9 was lowly expressed in most cancers, such as breast cancer, lung adenocarcinoma and squamous cell carcinoma, and stomach adenocarcinoma; but it was highly expressed in several cancers, such as cholangiocarcinoma, head and neck squamous cell carcinoma, and liver he- patocellular carcinoma. Furthermore, MYL9 expression was distinctively as- sociated with prognosis in adrenocortical carcinoma, colon adenocarcinoma, brain glioma, lung cancer, ovarian cancer, gastric cancer, breast cancer, blood cancer, and prostate cancer patients. The expressions of MYL9 were significantly associated with the infiltration of cancer-associated fibroblasts, B cell, CD8+T cell, CD4+T cell, macrophage, neutrophil, dendritic cell in different tumors as well as immune markers. In addition, we found that the functional mechanisms of MYL9 involved muscle contraction and focal adhesion.
Conclusion: MYL9 can serve as a prognostic signature in pan-cancer and is associated with immune infiltration. This pan-cancer study is the first to show a relatively comprehensive understanding of the prognostic and im- munological roles of MYL9 across different cancers.
KEYWORDS
cancer, immune infiltration, MYL9, prognosis
This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.
@ 2021 The Authors. Immunity, Inflammation and Disease published by John Wiley & Sons Ltd.
1 INTRODUCTION |
Due to the complexity of tumor genesis and develop- ment, it is very important to analyze the expression of any valuable genes and evaluate their correlation with clinical survival prognosis and possible molecular me- chanisms. The public databases, such as TCGA1,2 and Gene expression omnibus (GEO), contain functional genomics datasets for different cancers,3,4 and that set the stage for us to do a generalized cancer analysis.
Myosin is an actin-dependent molecular motor that uses the energy of adenosine triphosphatase hydrolysis to move along actin filaments and generate force. Myosin has many functions, including cell signaling, cell con- tractility, vesicle trafficking, endocytosis, and protein/ RNA localization.5,6 Myosin consists of two heavy chains and four light chains. Myosin light chain 9 (MYL9), a regulatory subunit of the forcing-producing ATPase nonmyosin II (NMII),7 may regulate muscle contraction by regulating ATPase activity in the myosin head. It binds to actin filaments to control cytoskeletal dynamics and is subsequently involved in cell shape establishment, migration, polarity, adhesion, and signal-mechanical transduction.8,9 Recently, many studies showed that MYL9 played the role of a promoter in tumor invasion and metastasis.10 For example, the MAL/SRF complex was involved in platelet formation and megakaryocyte migration by regulating MYL9 (MLC2) and MMP9,11 suggesting that it has clinical significance in human tis- sues of different tumor types. In addition, phosphoryla- tion of MYL9 is known to be key to cell migration on solid substrates.12 Therefore, MYL9 may play a crucial role in the genesis and development of tumors. However, based on large clinical data, there is currently no evi- dence of a generalized relationship between MYL9 and multiple tumor types.
In this study, we attempted to mainly use the TCGA and GEO databases to explore the prognostic and im- munological role of MYL9 across different tumors, and to investigate the potential molecular mechanism of MYL9 in the pathogenesis or clinical prognosis of different tu- mors via gene expression, survival prognosis, immune infiltration, and enrichment analysis.
2 METHODS |
2.1 The analysis of MYL9 gene expression |
We first used the Oncomine dataset to obtain the ex- pression of MYL9 between cancers and corresponding normal tissues (https://www.oncomine.org/). Then, the
TIMER2.0 (tumor immune estimation resource; version 2) web (http://timer.cistrome.org/) and GEPIA2 (Gene Ex- pression Profiling Interactive Analysis; version 2) web server (http://gepia2.cancer-pku.cn/#analysis) were used to further analyze the expression difference between hu- man tumors and normal tissues. In GEPIA2, we set p-value cutoff = 0.01, log2FC (fold change) cutoff = 1, and “Match TCGA normal and GTEx data.” Furthermore, we also employed the GEPIA2 to observe the correlation be- tween MYL9 expression and the pathological stages (Stage I, Stage II, Stage III, and Stage IV) of cancers, and the results were shown by the box or violin plots.
2.2 | The survival prognosis analysis of MYL9
The prognostic role of MYL9 was firstly analyzed by the “Survival Map” module of GEPIA2. Then, we used the data from Prognoscan (http://www.abren.net/ PrognoScan/) and Kaplan-Meier plotter (https:// kmplot.com/analysis/) databases to further analyze the effects of MYL9 expression on prognosis in different tu- mors. Kaplan-Meier plotter tool was used to analyze the effects of clinicopathological factors and MYL9 on the prognosis of patients with gastric cancer and ovarian cancer.
2.3 Immune infiltrating analysis and prognosis analysis |
The association between MYL9 expression and cancer- associated fibroblasts (CAFs) across all TCGA tumors was obtained by using the TIMER2 tool. The EPIC, MCPCOUNTER, XCELL, and TIDE algorithms were used to estimate immune penetration. The p and partial correlation (COR) values were obtained by the Spearman rank correlation test with purity adjustment. The data were visualized as a heatmap and a scatter plot. Fur- thermore, we employed the TIMER2 tool to set a Cox Proportional Hazard Model to evaluate CAF, age, stage, gender, race, purity, and MYL9, via the EPIC, MCPCOUNTER, XCELL, and TIDE algorithms. Ad- ditionally, the relationship between MYL9 expression and immune infiltrating cells, including B cell, CD8+T cell, CD4+T cell, macrophage, neutrophil, dendritic cell (DC) were determined by using the TIMER (http:// cistrome.org/TIMER/) databases. We further obtained the Kaplan-Meier curve for cumulative survival (CS), under the settings of 50% split expression percentage of patients, 50% split infiltration percentage of patients, and survival time between 0 and 200 months.
2.4 The enrichment analysis of MYL9-related gene |
We first used the STRING website (https://string-db.org/) to obtain the Top 50 of MYL9 binding proteins. Then, using the GEPIA2 website, we obtained the top 100 of MYL9-related targeted genes based on the data set of all TCGA cancers. We further used the GEPIA2 tool to conduct a pairwise gene Pearson correlation analysis of MYL9 and selected genes. Moreover, we used the TIMER2 tool to offer the heatmap data of these genes, which contains the partial correlation (cor) and p value in the purity-adjusted Spearman’s rank correlation test. We use bioinformatics and evolutionary genomics web- site (http://bioinformatics.psb.ugent.be/webtools/Venn/) interaction analysis to compare MYL9 combination and interaction of genes. Moreover, KEGG (Kyoto En- cyclopedia of Genes and Genomes) path analysis and GO (Gene Ontology) analysis were performed based on the two sets of data. The gene lists were uploaded to DAVID (database for annotation, visualization, and integrated discovery) for the data of the functional annotation chart. The bubble plots of enriched pathways were finally vi- sualized with the bioinformatics web (http://www. bioinformatics.com.cn/).
2.5 Statistical analysis |
Data from the Oncomine database was presented as p values determined by t test, fold changes, and Gene Rank. We used the PrognoScan, Kaplan-Meier plotter, TIMER and TIMER2, and GEPIA2 websites to conduct survival figures in respective analyses, with data includ- ing either hazard ratio (HR) and p values or p values derived from a log-rank test. Spearman’s and Pearson correlation analyses were used to gauge the degree of correlation between particular variables. MYL9-related gene enrichment analysis was analyzed by using KEGG and GO databases. p <. 05 was considered statistically significant, if not specially noted.
3 RESULTS |
3.1 The analysis of MYL9 gene expression in different cancers |
In this study, we first assessed the expression of MTL9 in multiple tumors and normal tissue types using the On- comine database, and results revealed that the expression of this gene was reduced compared with normal tissues for bladder, breast, kidney, lung, ovarian, and prostate
cancers, but was elevated compared with normal tissues for esophageal cancer, leukemia, liver cancer, lymphoma, and pancreatic cancer (Figure 1A). We additionally used the TIMER2.0 tool to analyze the expressions of the MTL9 gene in TCGA data set for all types of cancers, and found that the expressions of MYL9 gene in BLCA (bladder urothelial carcinoma), BRCA (breast invasive carcinoma), CESC (cervical squamous cell carcinoma and endocervical adenocarcinoma), COAD (colon adenocarcinoma), KICH (kidney chromophobe), KIRP (kidney renal papillary cell carcinoma), LUAD (lung adenocarcinoma), LUSC (lung squamous cell car- cinoma), PCPG (pheochromocytoma and para- ganglioma), PRAD (prostate adenocarcinoma), READ (rectum adenocarcinoma), STAD (stomach adenocarci- noma), UCEC (uterine corpus endometrial carcinoma) were significantly lower than the corresponding control tissues; however, the expressions of MYL9 gene in CHOL (cholangiocarcinoma), GBM (glioblastoma multiforme), HNSC (head and neck squamous cell carcinoma), LIHC (liver hepatocellular carcinoma) were obviously elevated (Figure 1B). Due to the lack of the normal tissue of ACC (adrenocortical carcinoma), DLBC (lymphoid neoplasm diffuse large B-cell lymphoma), LAMC (acute myeloid leukemia), LGG (brain lower grade glioma), OV (ovarian serous cystadenocarcinoma), SARC (sarcoma), SKCM (skin cutaneous melanoma), TGCT (testicular germ cell tumors), THYM (thymoma), and UCS (uterine carcino- sarcoma) in the TCGA database, we further evaluated the difference in MYL9 expression between normal and tu- mor tissues among ACC, DLBC, LAMC, LGG, OV, SARC, SKCM, TGCT, THYM, and USC, by using normal tissues from the GTEx data set as controls, and results showed that the expressions of MTL9 gene in DLBC and THYM were higher than corresponding normal tissues, but the expressions of MTL9 gene in LAMC, SKCM, TGCT, and UCS were lower than normal control tissues (Figure 1C). We also used the “pathological staging map” module of GEPIA2 to observe the correlation between MYL9 expression and tumor pathological staging, in- cluding COAD, KIRC (kidney renal clear cell carcino- ma), THCA (thyroid carcinoma), BLCA, TGCT, OV, STAD (Figure 1D). The data of this part indicated that the MYL9 gene played a different role in different tu- mors, which deserved further investigation.
3.2 Survival analysis data of MYL9 gene in different cancers |
To further explore the effects of the MTL9 gene on dif- ferent tumors, we divided tumor cases into high- expression group and low-expression group according to
(A)
(B)
Cancer
Cancer vs. Cancer
15
Normal
Multi-cancer
MYL9 Expression Level (log2 TPM)
**
**
*
*
**
Analysis Type by Cancer
Cancer
Histology
”
D
-
…
V
Bladder Cancer
4
.
1
1
1
Brain and CNS Cancer
3
.
1
Breast Cancer
6
S
Cervical Cancer
Colorectal Cancer
2
S
2
2
1
1
Esophageal Cancer
1
Gastric Cancer
7
1
1
4
Head and Neck Cancer
-
Kidney Cancer
3
2
1
1
-
Leukemia
2
9
Un
Liver Cancer
2
1
Lung Cancer
11
Lymphoma
8
1
3
3
Melanoma
1
%
Myeloma
2
Other Cancer
1
1
Ovarian Cancer
1
1
0
Pancreatic Cancer
2
ACC.Tumor (n=79)
BLCA. Tumor (n=408)
BLCA.Normal BRCA.Tumor (n=1093)
BRCA.Normal (n=112)
BRCA-Basal. Tumor way
BRCA-Her2.Tumor (n=82)
BRCA-LumA. Tumor (n=564)
BRCA-LumB. Tumor (n=217)
CESC.Tumor (n=304)
CESC.Normal (n=3)
CHOL. Tumor (n=36)
CHOL.Normal (n=9)
COAD.Tumor (n=457)
COAD.Normal (n=41)
DLBC.Tumor (n=48) ESCA.Tumor (n=184)
GBM.Normal (n=5)
HNSC.Tumor (n=520)
HNSC.NOIma MEN
HNSC-HPV-,Tumor (n=421)
KIRC.Normal (n=72)
LGG. Tumor (n=516)
LIHC. Tumor (n=371)
LUAD.Normal (n=59)
LUSC.Norma MESO.Tumor (n=87)
PCPG.Normal (n=3)
READ. Tumor (n=166)
THCA. Tumor (n=501)
THCA.Normal (n=59)
UCEC.Normal (n=35)
UCS.Tumor (n=57)
Prostate Cancer
S
3
ESCA.Normal (n=11)
GBM. Tumor (n=153)
HNSC-HPV+.Tumor (n=97)
KICH. Tumor (n=66)
KICH.Normal (n=25)
KIRC. Tumor (n=533)
KIRP. Tumor (n=290)
KIRP.Normal (n=32)
LAML. Tumor (n=173)
LIHC.Normal (n=50)
LOAD. Tumor (n=512)
LUSC. Tumor (n=501)
OV.Tumor (n=303)
PAAR monate
PAAD.Normal (n=4)
PCPG.Tumor (n=179)
PRAD. Tumor (n=497)
PRAD.Normal (n=52)
READ.Normal (n=10)
SARC.Tumor (n=259)
SKCM.Tumor (n=103)
SKCM.Metastasis (n=368)
STAD.Tumor (n=415)
STAD.Normal (n=35)
IOCh lumor (n= 150)
HAT Tumor ‘n=120)
UCEC.Tumor (n=545)
Sarcoma
2
4
8
6
1
Significant Unique Analyses Total Unique Analyses
19
38
19
15
8
13
428
714
257
1
5
10
10
5
1
Gene rank percentile(%)
%
(C)
2
2
A
2
Expression -log2(TPM+1)
W
8
2
SA
Z
៛
2
2
2
0
ACC
(num(T)=77; num(N)=128)
DLBC (num(T)=47; num(N)=337)
LAML (num(T)=173; num(N)=70)
LGG (num(T)=518; num(N)=207)
OV (num(T)=426; num(N)=88)
2
₾
Expression-log2(TPM+1)
2
AR
8
A
R
A
6
.
+
2
(D)
SARC
(num(T)=262; num(N)=2)
SKCM
(num(T)=461; num(N)=558)
TGCT (num(T)=137; num(N)=165)
THYM (num(T)=118; num(N)=339)
UCS (num(T)=57; num(N)=78)
MYL9 expression (log2(TPM+1))
$
COAD
F value = 3.12 Pr(>F) = 0.0268
2
KIRC
F value = 4.9 Pr[>F) = 0.0023
二
THCA
F value = 4.38 Pr[>F) = 0.00468
ㅎ
2
10
®
00
-
→
-
.
-
.
.
4
4
Stage I
Stage II
Stage III
Stage IV
Stage 1
Stage II
Stage III
Stage IV
Stage I
Stage II
Stage II
Stage IV
MYL9 expression (log2(TPM+1))
BLCA
F value = 20.1 Pr[>F) = 4.75€-09
=
TGCT
F value = 3.13 Pr[>F) = 0.0469
2
OV
F value = 5.23
Pr(>F) = 0.00572
=
STAD
F value = 4.16 Pr[>F) = 0.00643
2
2
:
:
:
00
0
유
8
8
.
៛
®
6
2
-
=
៛
0
Stage II
Stage III
Stage IV
Stage I
Stage II
Stage III
Stage II
Stage III
Stage IV
Stage I
Stage II
Stage
Stage IV
UVM.Tumor (n=80)
the expression level of MYL9 and used the GEPIA2 tool to investigate the effects of the MYL9 gene on overall survival (OS) and disease-free survival (DFS). The heat- map and Kaplan-Meier plot of OS in different tumors were displayed, and results showed that the low MYL9 group had a longer OS than the high MYL9 group in COAD, LGG, and MESO (mesothelioma) patients; how- ever, the results were adverse in ACC patients. We fur- ther found that lowly expressed MYL9 was linked to favorable prognosis of DFS for COAD patients, but see- mingly suggested a poor prognosis of DFS for THYM patients (Figure S1). To further investigate and verify the effects of MYL9 on prognosis in different tumors, we obtained the survival analysis data of MYL9 in Prog- noScan (Figure S2-6) and Kaplan-Meier datasets. We found that in the Prognoscan data set, multiple cancers showed a marked correlation between the prognosis of patients and the expression level of MYL9 including lung, ovarian, blood, prostate, brain, breast, and color- ectal cancer (Figure 2A-L). We also used the Kaplan-Meier plotter dataset to assess how the expres- sion of MYL9 related to prognosis in a series of tumor types, revealing that its reduction was significantly linked with a greater OS in ovarian cancer and gastric cancer, a greater first progression (FP) in lung cancer and gastric cancer, and a better post progression survival (PPS) in breast cancer, ovarian cancer and gastric cancer (Figure 2M-T). In addition, we used the data of Kaplan-Meier plotter RNA-seq dataset to further study the effects of the mRNA expression levels of MYL9 on prognosis in different tumors, results showing that its reduction was obviously linked with a better OS and relapse-free survival (RFS) in the majority of cancers except for sarcoma (Figure S7). Therefore, our pan- cancer analysis showed that MYL9 is a potential prog- nostic factor that contributed to the clinical treatment and prevention of tumors.
3.3 The relationship between MYL9 expression and patient clinicopathological findings in Kaplan-Meier plotter dataset |
In previous results, we found that the low-expression level of MYL9 was linked to the great prognosis of gastric and ovarian cancer patients in the Kaplan-Meier plotter dataset. Therefore, we attempted to explore the under- lying mechanisms by using the Kaplan-Meier plotter database and to assess the relationship between the ex- pression of MYL9 and the clinicopathological factors of cancer patients. As shown in Table 1, we found that the expression level of MYL9 correlated markedly with OS and PPS, and with sex, TNM stage, Lauren classification,
and HER2 status in gastric cancer except for mixed Lauren classification. We further found that the expres- sion level of MYL9 related to each N stage, corresponding to the degree of lymph node metastasis in gastric cancer patients. This lymph node metastasis was the most common type of metastasis in gastric cancer patients and was directly related to the prognosis of patients. As for the relationship between MYL9 and PPS in gastric can- cer, HR in Stage N was the highest, indicating that MYL9 expression may affect the prognosis of gastric cancer patients by affecting lymph node metastasis. As shown in Table 2, we found that the expression of MYL9 related markedly to OS and PFS and to stage and histology in ovarian cancer. In particular, as for the relationship be- tween MYL9 and OS in ovarian cancer, histology ex- hibited the highest HR, and for the relationship between MYL9 and PFS in ovarian cancer, Stage I showed the highest HR.
3.4 Correlation analysis between the expression of MYL9 and immune cell infiltrating |
Many studies showed that the occurrence and develop- ment of tumors were related to tumor-infiltrating im- mune cells in the immune microenvironment. Therefore, in this part, we further explored the relation of MYL9 expression among immune infiltration of CAFs. Through the EPIC, MCPCOUNTER, XCELL, and TIDE algo- rithms, the heatmap about the potential relationship between the infiltration level of CAFs and MYL9 gene expression was exhibited by using the TIMER2.0 tool in different tumors (Figure 3A). As shown in Figure 3B, the correlation between CAFs and MYL9 expression was shown by using a scatter diagram. We found that only in SARC, the expressions of MYL9 was negatively related with the infiltration level of CAFs based on the TIDE algorithm, while the expressions of MYL9 gene in other tumors were positively correlated with the expressions of CAFs, including BRCA-Her2 with the TIDE algorithm, BLCA with the MCPCOUNTER algorithm, BRCA-LumA with the TIDE, CESC with the TIDE algorithm, CHOL with the TIDE algorithm, COAD with TIDE algorithm, DLBC with the XCELL algorithm, ESCA with the MCPCOUNTER algorithm, HNSC-HPV+ with the MCPCOUNTER algorithm, KICH with the TIDE algo- rithm, KIRC with the MCPCOUNTER algorithm, KIRP with the TIDE algorithm, LGG with the MCPCOUNTER algorithm, LUAD with TIDE algorithm, UCS with the TIDE algorithm, UVM with the TIDE algorithm, LUSC with the TIDE algorithm, MESO with the TIDE algo- rithm, OV with the TIDE algorithm, PCPG with the
(A)
Lung cancer GSE11117 OS HR = 0.72 Cox P=0.036 Kaplan-Meier plot
(B)
(C)
Ovarian cancer GSE17260 PFS HR = 1.50 Cox P = 0.002 Kaplan-Meier plot
(D)
Ovarian cancer GSE9891 OS
HR = 1.45 Cox P = 0.001 Kaplan-Meier plot
Blood cancer GSE8970 OS HR = 0.60 Cox P = 0.047 Kaplan-Meier plot
High n= 22 Low n= 19
High n= 123
High n= 33
LOW n= 15 55
Low n= 77
High n= 9
LOW
n= 25
Probability
0.8
Probability
0.8
Probability
0.8
Probability
0.8
0.4
0.4
0.4
0.4
0.0
0.0
0.0
0.0
0
200
400
600
800
1000
0
50
100
150
200
0
20
40
60
0
200
400
600
800
1000
Days
Months
Months
Days
(E) Colorectal cancer GSE14333 DFS
(F)
Colorectal cancer GSE17536 DSS
(G)
Brian glioma GSE4412-GPL96 OS
(H)
Prostate cancer GSE16560 OS
HR = 1.37 Cox P = 0.006 Kaplan-Meier plot
HR = 1.37 Cox P = 0.032 Kaplan-Meier plot
HR = 1.29 Cox P= 0.023 Kaplan-Meier plot
HR=0.48 Cox P = 0.002 Kaplan-Meier plot
High n= 143 Low n= 83
High n= 77
Low n= 100
High n= 6
Low n= 44
High n= 173
Low n= 108
Probability
0.8
Probability
0.8
Probability
0.8
Probability
0.8
0.4
0.4
0.4
0.4
0.0
0.0
0.0
0.0
0
20
40
60
80
100
120
140
0
20
40
60
80
100
120
140
0
500
1000
1500
0
50
100
150
200
250
Months
Months
Days
Months
(1)
Breast cancer GSE9893 OS
(J)
Breast cancer GSE7390 RFS
(K)
Breast cancer GSE7390 DMFS
HR = 1.33 Cox P = 0.029 Kaplan-Meier plot
(L) Breast cancer GSE7849 DFS
HR = 1.47 Cox P = 0.009 Kaplan-Meier plot
HR = 1.28 Cox P =0.023 Kaplan-Meier plot
HR = 1.50 Cox P = 0.029 Kaplan-Meier plot
High n= 87 Low n= 68
High n= 177 LOW n= 21
High n= 159 Low n= 39
High n= 57
Low n= 19
Probability
0.8
Probability
0.8
Probability
0.8
Probability
0.8
0.4
0.4
0.4
0.4
0.0
0.0
0.0
0.0
0
50
100
150
0
2000
4000
6000
8000
0
2000
4000
6000
8000
0
50
100
150
Months
Days
Days
Months
(M)
Lung cancer FP MYL9 (201058_s_at)
(N)
Breast cancer PPS MYL9 (201058_s_at)
(O)
Ovarian cancer PFS MYL9 (201058_s_at)
(P)
Ovarian cancer OS MYL9 (201058_s_at)
0
HR = 1.25 (1.03 - 1.52)
0
A
HR = 1.35 (1.07 - 1.7)
0
9 -
HR = 1.3 (1.14 - 1.49)
logrank P = 0.023
logrank P = 0.012
HR = 1,42 (1.25 - 1.61)
logrank P = 3.3e-08
logrank P = 0.00012
0
0.8
40
00 0
Probability
0.6
Probability
0.6
Probability
0.6
Probability
0.6
0.4
0.4
0.4
0,4
N 8
Expression
V
0
Expression
2
0
Expression
Expression
low
low
low
0.0
high
high
low
high
0.
8
high
0
50
100
150
200
0
50
100
150
200
0
50
100
150
200
250
0
50
100
Time (months)
Time (months)
150
200
250
Number at risk
Time (months)
Number at risk
Time (months)
low high
632 350
248
122
26
18
5
0
low
high
245
213
59 48
18
13
5
2
Number at risk 821 614
low
high
56
76
17
3
:
:
Number at risk
18
6
low
1137
519
305
high
123
68
29
15
$
?
0
(Q)
Ovarian cancer PPS MYL9 (201058_s_at)
(R)
Gastric cancer FP MYL9 (201058_s_at)
(S)
Gastric cancer OS MYL9 (201058_s_at)
(T)
Gastric cancer PPS MYL9 (201058_s_at)
1.0
HR = 1.26 (1.03 - 1.53)
0
HR = 2.75 (2.13 -3.55)
0
HR = 2.77 (2.21 - 3.49)
0
logrank P = 0.023
logrank P = 8e-16
logrank P < 1E-16
HR = 3.71 (2.9 - 4.76)
logrank P < 1E-16
0.8
0
00 0
00 0
Probability
0.6
Probability
0.6
Probability
0.6
Probability
3
0.4
0.4
0.4
0
2 0
Expression
N 0
Expression
4
0
Expression
Ny 0
Expression
0 8
low
low
low
low
high
0
high
0 0.
high
0
high
0
50
100
150
200
0
50
100
150
0
50
100
150
0
20
40
60
80
Time (months)
Time (months)
Time (months)
Time (months)
Number at risk
low
195
587
60
133
12
Number at risk
18
2
:
Number at risk
207 433
2
30
low
high
244
631
142
high
2
0
156
o
0
Number at risk
40
1
low
32
2
1
high
242 256
17
23 1
high
MCPCOUNTER algorithm, STAD with the TIDE algo- rithm, TGCT with the TIDE algorithm, THYM with the TIDE algorithm, UCEC with the MCPCOUNTER algo- rithm, SKCM with the EPIC algorithm. To further in- vestigate the correlation between MYL9 expression and immune infiltrating, we employed the TIMER tool to analyze the correlation of B cell, CD8+T cell, CD4+T cell,
macrophage, neutrophil, DC with MYL9 expression, and results showed that the expression of MYL9 significantly correlated with the infiltration of macrophage in BLCA, with the infiltration of CD4+T cell and macrophage, neutrophil, DC in COAD, with the infiltration of mac- rophage in ESCA, with the infiltration of DC in GBM, with the infiltration of macrophage in HNSC, with the
| Clinicopathological characteristics | Overall survival (n = 875) | postprogression survival (n = 498) | ||||
|---|---|---|---|---|---|---|
| N | Hazard ratio | p value | N | Hazard ratio | p value | |
| Sex | ||||||
| Male | 544 | 2.96 (2.25-3.9) | 6.60E-16 | 348 | 3.83 (2.86-5.13) | 1.00E-16 |
| Female | 236 | 2.65 (1.77-3.97) | 8.30E-07 | 149 | 3.5 (2.22-5.53) | 1.30E-08 |
| Stage | ||||||
| I | 67 | 3.99 (1.27-12.56) | .011 | 31 | ‐ | ‐ |
| II | 140 | 2.3 (1.26-4.19) | .0052 | 105 | 2.75 (1.42-5.33) | .0018 |
| III | 305 | 2.7 (1.82-3.99) | 2.50E-07 | 142 | 4.2 (2.54-6.95) | 1.90E-09 |
| IV | 148 | 1.87 (1.21-2.9) | .0043 | 104 | 2 (1.26-3.18) | .0028 |
| Stage T | ||||||
| 2 | 241 | 2.11 (1.37- 3.23) | .00046 | 196 | 3.01 (1.92-4.72) | 4.70E-07 |
| 3 | 204 | 2.06 (1.42-3) | .00011 | 150 | 2.62 (1.72-3.98) | 3.00E-06 |
| 4 | 38 | 3.48 (1.36-8.93) | .0061 | 29 | 1.89 (0.67-5.32) | .22 |
| Stage N | ||||||
| 0 | 74 | 3.09 (1.19-7.98) | .015 | 41 | 7.85 (2.09-29.43) | .00034 |
| 1 | 225 | 3.45 (2.22-5.35) | 4.10E-09 | 169 | 5.25 (3.23-8.55) | 1.60E-13 |
| 2 | 121 | 3.1 (1.91-5.04) | 1.70E-06 | 105 | 3.26 (1.94-5.47) | 2.70E-06 |
| 3 | 76 | 2.15 (1.23-3.78) | .0064 | 63 | 2.27 (1.24-4.15) | .0061 |
| 1+2+3 | 422 | 2.73 (2.06-3.63) | 5.00E-13 | 337 | 3.37 (2.5-4.54) | 1.00E-16 |
| Stage M | ||||||
| 0 | 444 | 2.5 (1.87-3.33) | 1.00E-10 | 342 | 3.81 (2.78-5.22) | 1.00E-16 |
| 1 | 56 | 2.21 (1.2-4.07) | .0095 | 36 | 3.38 (1.48-7.68) | .0024 |
| Lauren classification | ||||||
| Intestinal | 320 | 3.03 (2.14-4.28) | 5.10E-11 | 192 | 4.49 (2.94-6.84) | 2.80E-14 |
| Diffuse | 241 | 2.39 (1.67-3.41) | 8.70E-07 | 176 | 2.69 (1.8-4.03) | 5.10E-07 |
| Mixed | 32 | 3.79 (1.2-11.97) | 1.50E-02 | 16 | ‐ | ‐ |
| Differentiation | ||||||
| Poor | 165 | 1.63 (0.95-2.79) | .071 | 49 | 1.5 (0.73-3.06) | .27 |
| Moderate | 67 | 1.77 (0.92-3.39) | .081 | 24 | 0.62 (0.22- 1.78) | .37 |
| Well | 32 | 6.46 (2.13-19.56) | .00018 | 0 | ‐ | ‐ |
| HER2 status | ||||||
| HER2 negative | 532 | 2.59 (2-3.35) | 6.60E-14 | 334 | 3.38 (2.5- 4.57) | 1.00E-16 |
| HER2 positive | 343 | 1.81 (1.3-2.52) | 4.00E-04 | 164 | 4 (2.51-6.36) | 3.90E-10 |
Abbreviation: MYL9, myosin light chain 9.
infiltration of CD4+T cell and macrophage in HNSS- HPV , with the infiltration of macrophage in KIRP, with the infiltration of CD4+T cell, with the infiltration of CD4+T cell and macrophage in LIHC, with the infiltra- tion of macrophage, neutrophil, and DC in LUAD, with the infiltration of CD4+T cell, macrophage, neutrophil,
and DC in LUSC, with the infiltration of macrophage, neutrophil, and DC in PAAD, with the infiltration of CD4+T cell and macrophage in READ, and, with the infiltration of CD4+T cell, macrophage, and DC in STAD (Figure S8). All data of this part indicated that MYL9 expression correlated with immune infiltration in
| Clinicopathological characteristics | Overall survival (n = 1656) | Progression-free survival (n = 1435) | ||||
|---|---|---|---|---|---|---|
| N | Hazard ratio | p value | N | Hazard ratio | p value | |
| Stage | ||||||
| I | 74 | 2.54 (0.68-9.45) | .15 | 96 | 3.91 (1.36-11.24) | .006 |
| I+ II | 135 | 1.99 (0.75-5.29) | .16 | 163 | 1.84 (1.01-3.33) | .042 |
| II | 61 | 3.38 (0.73-15.57) | .1 | 67 | 1.46 (0.65-3.25) | .36 |
| II + III | 1105 | 1.29 (1.09-1.53) | .003 | 986 | 1.49 (1.28-1.73) | 1.7E-07 |
| II + III + IV | 1281 | 1.34 (1.12-1.59) | .001 | 1148 | 1.54 (1.34-1.77) | 7.8E-10 |
| III | 1044 | 1.27 (1.07-1.51) | .006 | 919 | 1.49 (1.28-1.74) | 3.7E-07 |
| III + IV | 1220 | 1.26 (1.07-1.47) | .004 | 1081 | 1.54 (1.34-1.77) | 2E-09 |
| IV | 176 | 2.81 (1-7.86) | .042 | 162 | 1.88 (1.29-2.74) | .001 |
| Histology | ||||||
| Endometrioid | 37 | 7.65 (1.26-46.43) | .009 | 51 | 3.75 (1.45-9.7) | .004 |
| Serous | 1207 | 1.29 (1.1-1.51) | .001 | 1104 | 1.61 (1.39-1.86) | 9.3E-11 |
| TP53 mutation | ||||||
| Mutated | 509 | 1.35 (1.03-1.76) | .026 | 483 | 1.43 (1.14-1.78) | .002 |
| Wild type | 94 | 1.61 (0.94-2.77) | .08 | 84 | 1.75 (1.02-3.02) | .04 |
| Debulk | ||||||
| Optimal | 801 | 1.52 (1.23-1.88) | 9.7e-05 | 696 | 1.49 (1.23-1.8) | 4.3E-05 |
| Suboptimal | 536 | 1.21 (0.96-1.51) | .099 | 459 | 1.25 (1-1.57) | .046 |
Abbreviation: MYL9, myosin light chain 9.
different tumors, which might be a potential mechanism to exert the effects on prognosis.
3.5 Correlations between the expression of MYL9 and immune cell markers |
To further investigate the potential relation between the expression level of MYL9 and infiltrating immune cells, we explored the relationships between the expression of MYL9 and the markers of several immune cells by using the GEPIA and TIMER tools. These markers were used to characterize immune cells, including CD8+ T and CD4+ T cell, B cell, M1/M2 macrophage, monocyte, neutrophil, NK, tumor-associated macrophage, and DC in STAD and COAD. We also analyzed the different functional T cells such as Treg, Th1, Th2, Th9, Th17, Th22, Tfh, and exhausted T cells. We found that in GAPIA, the MYL9 expression level obviously related to 61 out of 77 immune cell markers in STAD, and related to 47 out of 77 immune cell markers in the corresponding normal tissues. In COAD, the expression levels of MYL9
obviously correlated with 68 out of 77 immune cell markers and correlated with 23 out of 77 immune cell markers in corresponding normal tissues (Table S1). Elevated MYL9 expression was associated with increased DC infiltration in STAD and COAD, and consistent with this, the DC markers including CD1C, CD141, HLA- DPB1, HLA-DRA1, BDCA-4(NRP1), and CD11c(ITGAX) linked with the expression level of MYL9. This suggested that the expression level of MYL9 was closely related to the penetration of tumor DC. We additionally observed that there was a marked relation between the expression of MYL9 and the markers of Tregs and exhausted T cells including FOXP3, CCR8, CD25 (IL2RA), STAT5B, TGFß (TGFB1), PD1 (PDCD1), CTLA4, LAG3, TIM- 3(HAVCR2), GZMB in STAD and COAD (Table S1), in- dicating that MYL9 had a potential role in immune es- cape in STAD and COAD, although further research would be needed to demonstrate the mechanisms un- derlying such escape. Interestingly, MYL9 expression markedly correlated with TAM and M2 macrophage markers in STAD as well as the corresponding normal tissues; however, MYL9 expression significantly corre- lated with monocyte marker in STAD but the two were
(A)
(B)
Purity
Cancer associated fibroblast_TICE
Purity
noer associated fibroblast_MCPCOLINTE
Purity
ncer associated fibroblast_MCPCOUNT|
Purity
Cancer associated fibroblast_TIDE
25
Rho - - 0 372
Rho — 0.33
5.780 08
10
0
p > 0.05
Cancer associated fibroblast_EPIC
Cancer associated fibroblast_MCPCOUNTER
Cancer associated fibroblast_XCELL
Cancer associated fibroblast_TIDE
MYL9 Expression Level (log2 TPM)
MYL9 Expression Level (log2 TPM)
MYL9 Expression Level (log2 TPM)
MYL9 Expression Level (log2 TPM)
%
D
BRCA-Her2
S
00
p … 0.05
BLCA
BIRCA
BRCA-LumA
2
-
S
Partial_Cor
50-
-
-
1
0.25
0.50
0.75
1.00 -0.2
0.0
02
0.25
0.50
0.75
1.00 0
20000
40000
40000
0.25
0.50
0.75
1.00
10000 20000 30000 40000 50000
0.25
0.50
0.75
1.00
-02
0.2
0
Purity
Infiltration Level
Purity
Infiltration Level
Purity
Infiltration Level
Purity
Infiltration Level
-1
MYLS Expression Level (log2 TPM)
Purity
Cancer associated foroblast_TIDE
MYL9 Expression Level (log2 TPM)
Purity
Cancer associated fibroblast_TIDE
MYL9 Expression Level (log2 TPM)
Purity
acer associated fibroblast_MCPCOUNT
MYL9 Expression Level (log2 TPM)
Purity
Cancer associated fibroblast_TIDE
ACC (n=79)
POR59111
2.5
12
0
9.224-11
BLCA (n=408)
10.0
BRCA (n=1100)
0.0
0
BRCA-Basal (n=191)
1.5
CESC
CHO
COAD
BRCA-Her2 (n=82)
:-
CESC
BRCA-LumA (n=568)
BRCA-LumB (n=219)
5.0
6
CESC (n=306)
*
*
0.25
0.50
0.75
1.00
-0.2
00
02
0.4
0.25
0.50
0.75
1.000.2
0.0
0.1
02
0.25
0.50
0.76
1.00
3
5000
10000
1500
0.25
0.50
0.75
1.00
00
02
CHOL (n=36)
Purity
Infiltration Level
Purity
Infiltration Level
Punty
Infiltration Level
Purity
Infiltration Level
COAD (n=458)
DLBC (n=48)
MYL9 Expression Level (log2 TPM)
Purity
Cancer associated fibroblast_XCELL
MYL9 Expression Level (log2 TPM)
Purity
noer associated fibroblast_MCPCOLINTE
MYL9 Expression Level (log2 TPM)
Purity
ncer associated fibroblast_MCPCOUNTE
MYL9 Expression Level (log2 TPM)
0
Purky
Cancer associated fibroblast_TIDE
ESCA (n=185)
1-61
WHEN
.
5
9. 4564-01
· 00000 1.00 0 Purity 0.75 Infiltration Level 0 90 20000 Purity 25 PROPON 5.0 + 0.50 Purity Infiltration Level 1.00 -Q. 0 0.2 0.75 Purity a A UVM 9 o D.8 1,0-0.3 0.1 0 2 MYL9 Expression Level (log2 TPM) MYL9 Expression Level (log2 TPM) Purity Infiltration Level Purity - 0 6 MYL9 Expression Level (log2 TPM) 1 4
y
GBM (n=153)
HNSC (n=522)
0
2
1
HNSC-HPV- (n=422)
DLBC
ESCA
HNSC-HPV-
-
HNSC-HPV+ (n=98)
+
KICH (n=66)
_
3
A
KIRC (n=533)
1
..
%
*
KIRP (n=290)
0.25
0.50
0.75
1.00.00
0.05
0.15
0.25
0.50
0 75
0
10000
20000
30000
0.25
60000
0.2
0.4
0.6
0.8
1.0
-0.1
0.0
0.1
Infiltration Level
Purity
Infiltration Level
Purity
Infiltration Level
0.2
LGG (n=516)
Purity
LIHC (n=371)
LUAD (n=515)
MYL9 Expression Level (log2 TPM)
Purity
acer associated faroblant_MCPCOUNT
p-1.66-06
MYL9 Expression Level (log2 TPM)
Purity
Cancer associaned fibroblast_TICE
ncer associated fibroblast_MCPCOUNT!
MYL9 Expression Level (log2 TPM)
Purity
Cancer asociated fibroblast_TIDE
LUSC (n=501)
MORFO.LA
5923
MESO (n=87)
Q
::
S
OV (n=303)
25
**
PAAD (n=179)
1
0
KIR
LUAD
PCPG (n=181)
0
PRAD (n=498)
7.5
=
READ (n=166)
?
SARC (n=260)
0.25
0.50
0.75
1.00
5000
10000 15000 20000
0.25
2.5
0.25
0.50
0.75
2000
4000
0.25
0.50
0.75
D.O
02
0.4
SKCM (n=471)
Purity
Infiltration Level
Purity
1.00 0
Infiltration Level
Purity
1.00 -0.2
Infiltration Level
SKCM-Metastasis (n=368)
SKCM-Primary (n=103)
MYL9 Expression Level (log2 TPM)
Purity
Cancer associated fibroblast_TIDE
Cancer associated fibroblast_TIDE
MYL9 Expression Level (log2 TPM)
Purity
Cancer associated fibroblast_TIDE
MYL9 Expression Level (log2 TPM)
Purity
Cancer associaned fibroblast_TIDE
STAD (n=415)
TGCT (n=150)
0
.
+
8
9
THCA (n=509)
**
5
THYM (n=120)
A
LUSC
MESO
UCEC (n=545)
UCS (n=57)
6
-
.
:
UVM (n=80)
5
.
0.4
0.6
0.25
0.50
0.75
1.00
-0.1
0.0
0.1
02
0 00
0.25
0.50
0.75
1.00
02
0.0
02
Infiltration Level
0.25
0.50
0 75
2
Purity
Purity
Purity
1.00-0.2
Infiltration Level
Infiltration Level
Cancer associated fibroblast_TIDE
MYL9 Expression Level (log2 TPM)
Purity
noer associated fibroblast_MCPCOUNTE
MYL9 Expression Level (log2 TPM)
Purity
Cancer associated fibroblast_TIDE
MYL9 Expression Level (log2 TPM)
Purity
Cancer associated fibroblast_TIDE
H
P
2-
12-
PCPO
STAD
C
-
TGCT
A
=
1
:
0.4
08
1.0
-02
0.0
02
0.25
0.30
0.75
1.00 0
5000
10000
15000
0.25
0.50
0,75
1.00
02
00
02
0.4
0.25
0.50
4.75
1.00
Infiltration Level
Purity
Purity
-0.2
0.0
0.2
Infiltration Level
Infiltration Level
Infiltration Level
0.4
Purity
Purity
MYL9 Expression Level (log2 TPM)
Purity
Cancer associaned fibroblast_TIDE
MYL9 Expression Level (log2 TPM)
Purity
ancer associated fibroblast_MCPCOUNTE
MYL9 Expression Level (log2 TPM)
Purity
Cancer associated fibroblast_EPIC
MYL9 Expression Level (log2 TPM)
Purity
Cancer insocianed fibroblast_EPIC
2-
# 4.708-09
..
C
10-
THYM
UÇEC
SKOM
1
U
SAR
5.0
2
9
5
.
5
1
$
A
5%
.
V
0.25
Purity
0.75
1.00
-0.2
Infiltration Level
2
0.00
0.25
0.50
0.75
1.00 0
3000
000
9000
0.25
0.50
0.75
1.00 0.00
0.25
0.50
1.00
0.25
0.50
.75
Purity
Infiltration Level
Purity
Infiltration Level
0.75
Purity
1.000.00
0.25
Infiltration Level
1.00
(A)
(B)
(C)
(D)
Moneyte
TAM
M1 macrophage
M2 macrophage
MYLA
MYLS
MYLO
MYL#
MYLS
Expression Level fog? TPM)
Expression Level fog? TPM)
6
Expression Level fog2 TPM)
.. 8211
STAD
Expression Level (og2 1PM)
Expression Level fog2 TPhd)
Expression Level (og2 TPM)
Expression Level (log2 TPM)
Expression Level (og2: TPM)
Expression Level (og2 TPMD
Expression Level (og2 TPMD
Bor # 8. 451
1
W
PROSÍ
4
Expression Level (og2 TPM)
Expression Level (og2 TPM)
Expression Level Blog2 TPM)
Expression Level (Jog2 TPM
Expression Level (og2 TPM)
Expression Level (og& TPM)
Expression Level (log2 TPM)
Expression Level (log2 TPM)
Expression Level [ogž TPM)
Expression Level (og2 TPM)
(E)
Moncyte
(F)
TAM
(G)
(H)
M1 macrophage
M2 macrophage
MYLD
MYLS
MYLD
COAD
Level [og& TPM)
Expression Level [og2: TPM)
Expression Level (og2 TPMD
Expression Level dag2 TPLt)
Expression Level (og2 TPMt)
Expression Level (og2 TPMS)
Expression Level (og2 TPLt)
Expression Level Blog2 TIPA!)
Expression Level pog2 TPM)
Expression Level pog2 TIPA!)
=100%-13
.
o
-
Pras
7
1
2
Expression Level (logů TPM)
Expression Level fog2 TPLf)
Expression Level (og2 TPM)
Expression Level (og? TPut)
Expression Level dag? TPut)
1
Expression Level (og2 TPM)
Expression Level (log2 TPM)
Expression Level (log2 TPM)
Expression Level [og2: TPM)
Expression Level [og2: TPM)
irrelevant in corresponding normal tissues. In COAD, we found in particular that MYL9 expression significantly correlated with the M2 macrophage marker, but the two were irrelevant in corresponding normal tissues. So, we used the TIMER tool to further analyze the relation be- tween MYL9 expression and monocyte, TAM, M1 mac- rophage, and M2 macrophage markers, and this result brought into correspondence with the results in GAPIA. As shown in Figure 4, the expression level of MYL9 ob- viously related to monocyte markers (CD86, CD115), TAM markers (CCL2, IL10), M1 macrophage marker (IRF5), and M2 macrophage markers (CD163, VSIG4, MS4A4A) in STAD (Figure 4A-D) and COAD (Figure 4E-H). Therefore, we speculate that MYL9 may have the function of regulating the polarization of macrophages.
3.6 The prognosis analysis of MYL9 gene expression and immune infiltration cells in different tumors |
In previous results, we elaborated that the expression levels of the MYL9 gene were significantly correlated with the immune infiltrating in CAF. In this part, we attempted to explore the prognosis effects of MYL9 and CAF in the TCGA data set of different tumors. As shown in Figure 5A, a heatmap presented the normalized coefficient of the infiltrate for each Cox proportional hazard model across multiple cancer types. To move forward a single step, we found that in CESC, the CS of the low MYL9 expression and low CAF group was markedly longer than the low MYL9 expression and high CAF group; however, the CS of the high MYL9 expres- sion and high CAF group was obviously shorter than the high MYL9 expression and low CAF group (Figure 5B). As shown in Figure 5C, interestingly, the CS of the high MYL9 expression and high CAF group was longer than the high MYL9 expression and low CAF group in HNSS- HPV+, but there was no significant difference between groups of KIRP (Figure 5D). In LGG, there was longer CS in the low MYL9 expression and low CAF group than the low MYL9 expression and high CAF group, and, the CS of the high MYL9 expression and low CAF fibroblast group was longer than the high MYL9 expression and high CAF group (Figure 5E). As shown in Figure 5F, we found that the CS of the low MYL9 expression and low CAF group was significantly longer than the low MYL9 expression and high CAF group in SARC, but there was no significant difference between groups of READ (Figure 5G). The data of this part indicated that MYL9 expression and CAF played an important role in CS of cancer patients, and their effects were different or
adverse on different types of tumors, which would pro- vide potential and novel targeting for clinical cancer di- agnosis and therapy, expecting to improve the prognosis of cancer patients.
We further studied the effects of the immune in- filtrating of B cell, CD8+T cell, CD4+T cell, Macrophage, Neutrophil, DC, and the expression of MYL9 on CS in different cancers (Figure S9), and results showed that MYL9 expression was significantly correlated with prognosis in ACC (Figure 5H), CD8+T cell infiltration and MYL9 expression were obviously correlated with prognosis in BLCA (Figure 5I), B cell, CD8+T cell, CD4+T cell, Macrophage, Neutrophil, DC, and the ex- pression of MYL9 were significantly correlated with prognosis in LGG (Figure 5J), and, Neutrophil infiltra- tion and MYL9 expression was markedly correlated with prognosis in MESO (Figure 5K). This suggested that MYL9 played a strong role in regulating immune cell infiltration, with a particularly strong effect on CD8+T cell infiltration in BLCA, on B cell, CD8+T cell, CD4+T cell, macrophage, neutrophil, DC infiltration in LGG, and on Neutrophil infiltration in MESO.
3.7 | The enrichment analysis of MYL9-related gene
To further study the molecular mechanism of MYL9 in tumorigenesis, we attempted to screen out targeted MYL9 binding proteins and the related genes of MYL9 expression for pathway enrichment analysis. As shown in Figure 6A, through employing the STRING tool, we obtained a total of 50 MYL9-binding proteins, and showed the interaction network of these proteins. In addition, we used the GEPIA2 tool to combine all tumor expression data of TCGA and obtained the top 100 genes that correlated with MYL9 expression. Then, the top 6 gene of those correlated with MYL9 expression were further analyzed by the GEPIA2 tool for correlation, and the results were shown by scatter diagram (Figure 6B), including TAGLN (transgelin), CNN1 (calponin 1), LMOD1 (leiomodin 1), TPM2 (tropomyosin 2), KCNMB1 (potassium calcium-activated channel subfamily M reg- ulatory beta subunit 1), and JPH2 (junctophilin 2). The corresponding heatmap data also showed a positive cor- relation between MYL9 and the above six genes in most detailed cancer types (Figure 6C). Using the Venn dia- grams, an intersection analysis of the above two groups showed five common members, namely, MYH11 (myosin heavy chain 11), SPEG (striated muscle enriched protein kinase), MYLK (myosin light chain kinase), ACTA2 (actin alpha 2), and ACTG2 (actin gamma 2) (Figure 6D). To further study the pathogenesis of MYL9 gene in
(A)
(B)
(C)
(D)
Zscore
Cancer associated fibroblast_MOPCOUNTER
4.4
Cancer associated fibroblast_XCELL
1:Low Gene Expression + Low Cancer associated fibroblast_TIDE
+4-#|Low Gene Expression + Low Cancer associated fibroblast_XCELL
+++++++++BLOW Gene Expression + Low
0.0
Cancer associated fibroblast_EPIC
Cancer associated fibroblast_TIDE
1.0
1.0
1.0
2:Low Gene Expression + High Cancer associated fibroblast_TIDE
2Low Gene Expression + High Cancer associated fibroblast_XCELL
-3.4
3:High Gene Expression + Low Cancer associated fibroblast_TIDE
L:High High
3.FighiGene Expression + Low Cancer associated fibroblast XCELL
2:Low Gene Expression + High Cancer associated fibroblast_EPIC
3:High Low
Cumulative Survival
0.8
Cumulative Survival
0.8
4:High Gene Expression + High Cancer associated fibroblast_XCELL
gh Gene Expression + High Cancer associated fibroblast_EPIC
Cumulative Survival
0.8
XX
p> 0.05
p … 0.05
0.6
0.6
0.6
ACC (n= 79)
BLCA (n=408)
0.4
0.4
0.4
BRCA (n=1100)
BRCA-Basal (n=191)
BRCA-Her2 (n=82)
BRCA-LumA (n=568)
0.2
0.2
0.2
BRCA-LumB (n=219)
CESC
KIRP
CESC (n=306)
HNSC-HPV+
CHOL (n=36)
2 1; HR=3.89, p = 0.00531
COAD (n=458)
0.0
4 vs 3: HR=3.45, p = 0.0136
0.0
2 1; HR=2.3, p = 0.274
4 vs 3: HR=0.0436. p = 0.0112
0.0
2 1: HR=0.495, p = 0.354
4 vs 3; HR=6.83, p = 0.0931
DLBC (nm48)
ESCA (n= 185)
0
50
100
150
0
20
40
60
80
100
0
50
100
150
200
GBM (n=153)
HNSC (n=522)
(E)
Time to Follow-Up (months)
(F)
Time to Follow-Up (months)
Time to Follow-Up (months)
HNSC-HPV-(n=422)
(G)
HNSC-HPV+ (n=98)
KICH (n=66)
9
1.0
KIRC (n=533)
1:Low Low
1.0
+++ ++- Low Gene Expression + Low Cancer associated fibroblast_XCELL
XIXI
XIXIX
-
1:Low Gene Expression + Low Cancer associated fibroblast_TIDE
2:Low High Cancer associated fibroblast_TIDE
KIRP (n=290)
3.High + Low
2:Low Gene Expression + High Cancer associated fibroblast_XCELL
Gene Expression + High Cancer associated fibroblast_TIDE
3:High + Low
3:High
High associated
+ Low Cancer
LGG (n=516)
XIXI
4:High
Gene Expression + High Cancer associated fibroblast_XCELL
LIHC (n=371)
0.8
0.8
4:High Gene Expression + High Cancer associated fibroblast_XCELL
LUAD (n=515)
Cumulative Survival
Cumulative Survival
Cumulative Survival
0.8
LUSC (n=501)
MESO (n=87)
X
OV (n=303)
0.6
0.6
0.6
PAAD (n=179)
PCPG (n=181)
XIX
PRAD (n=498)
READ (n=166)
0.4
0.4
0.4
XIX
SARC (n=260)
SKCM (n=471)
SKCM-Metastasis (n=368)
SKCM-Primary (n=103)
0.2
0.2
0.2
STAD (n=415)
LGG
SARC
READ
TGCT (n=150)
THCA (n=509)
2 vs 1: HR=2.13, p = 0.0199
2 vs 1; HR=0.524. p = 0.0312
2 vs 1: HR=5.88e+62, p = 0.997
ΤΗΥΜ (n=120)
0.0
4 vs 3: HR=3.89, p = 4.84€-05
0.0
4 vs 3: HR=0.729, p = 0.339
0.0
4 vs 3: HR=6880. p = 0.07
UCEC (n=545)
UCS (n=57)
0
50
100
150
0
50
100
150
0
20
40
60
80
100
120
UVM (n=50)
Time to Follow-Up (months)
Time to Follow-Up (months)
Time to Follow-Up (months)
(H)
B Cell
CDB+ T Cell
CD4+ T Cell
Macrophage
Neutrophil
Dendrisc Cel
MYL9
1.0
Log-rank P = 0.396
Log-rank P = 0.055
Log-rank P = 0.724
Log-rank P = 0.175
Log-rank P = 0.475
Log-rank P = 0.601
Log-rank P = 0.013
0.8
5
0.6-
0.4
(1)
1.00
Log-rank P = 0.457
Log-rank P = 0.006
Log-rank P = 0.889
Log-rank P = 0.107
Log-rank P = 0.901
Log-rank P = 0.523
Log-rank P = 0.044
0.75
0.50
BLCA
Cumulative Survival
0.25
0.00
Level
Low (Bottom 50%)
(J)
1.00
Log-rank P = 0
Log-rank P = 0.01
Log-rank P = 0
Log-rank P = 0
Log-rank P = 0
Log-rank P = 0.001
Log-rank P = 0.01
High (Top 50%)
0.75
0.50
8
0.25
(K)
1.00
Log-rank P = 0.14
Log-rank P = 0.519
Log-rank P = 0.865
Log-rank P = 0.225
Log-rank P = 0.001
Log-rank P = 0.164
Log-rank P = 0.03
0.75
0.50
MESO
0.25
0.00
0
25
50
75
0
25
50
75
0
25
50
75
0
25
Time to Follow-Up (months)
50
75
0
25
50
75
0
25
50
75
0
25
50
75
tumors, we combined the two datasets to perform KEGG and GO enrichment analyses. As shown in Figure 6E, The bubble diagram of KEGG data suggested that “tight junction,” “vascular smooth muscle contraction,” “focal adhesion,” and “dilated cardiomyopathy” might be in- volved in the effect of MYL9 on tumor pathogenesis, and “Oxytocin and cGMP-PKG signaling pathway” might be
involved in the majority signaling pathway for tumor- igenesis mechanism. In the GO data set, we found that “muscle contraction,” “myosin filament,” “focal adhe- sion,” and “calmodulin binding” might be involved in the effect of MYL9 on tumor pathogenesis (Figure 6F). The combined information of the two databases manifested that the effects of MYL9 gene on oncogenesis and
(A)
(B)
OMICH
=
p-value = 0 R = 0.80
.
=
p-value = g
.
p-value = @
p-value = 0
p=value = D
-
R = 0.86
R-0.86
p-value = D R = 0.82
1
R = 0.85
R = 0.84
log?(TAGLN TPM)
log2(CNN1 TPM)
#
log2(LMOD1 TPM)
1
.
log2(TPM2 TPM)
log2(KCNMB1 TPM)
log2(JPH2 TPM)
.
.
.
*
$
-
1
,
¿
1
A
.
\
P
·
1
.
·
·
#
«
a
o
.4
2
2
-
=
9
2
15
1
1
·
.
P
2
·
4
·
0
”
w
0
3
1
log2(MYL9 TPM)
log2(MYL9 TPM)
log2(MYL9 TPM)
log2(MYL9 TPM)
log2(MYL9 TPM)
9
14
ASTH
AGTL7B
log2[MYL9 TPM)
POTCES
AGTLMA
(C)
8
p > 0.05
POTE!
p … 0.05
PONDHE
UVM (n=80)
UCS (n=57)
UCEC (n=545)
THYM (n=120)
THCA (n=509)
TGCT (n=150)
STAD (n=415)
SKCM-Primary (n=103)
SKCM-Metastasis (n=368)
SKCM (n=471)
SARC (n=260)
READ (n=166)
PRAD (n=498)
PCPG (n=181)
PAAD (n=179)
OV (n=303)
MESO (n=87)
LUSC (n=501)
LUAD (n=515)
LIHC (n=371)
LGG (n=516)
KIRP (n=290)
KIRC (n=533)
KICH (n=66)
HNSC-HPV+ (n=98)
HNSC-HPV-(n=422)
HNSC (n=522)
GBM (n=153)
ESCA (n=185)
CHOL (n=36)
BRCA-LumB (n=219)
BRCA-LumA (n=568)
BRCA-Basal (n=191)
POTICO
DLBC (n=48)
COAD (n=458)
CESC (n=308)
BRCA-Her2 (n=82)
BRCA (n=1100)
BLCA (n=408)
ACC (n=79)
Spearman_Cor
1
ACTATI
Q
MYLÉ
AGTLA
-1
X
CNN1
X
JPH2
ACTA
X
KCNMB1
LMOD1
TAGLN
X
X
TPM2
(D)
(E)
(F)
myosin filament
.
Tight junction
·
muscle myosin complex
Dilated cardiomyopathy
cell junction assembly
correlated
Hypertrophic cardiomyopathy (HCM)-
actin-dependent ATPase activity
Vascular smooth muscle contraction
microfilament motor activity
count
Cardiac muscle contraction
count
actin filament-based movement
5.0
·
10
MYH11
Arrhythmogenic right ventricular cardiomyopathy (ARVC)
muscle filament sliding
7.5
stress fiber
.
15
Viral myocarditis
.
10.0
structural constituent of muscle
· 20
SPEG
Bacterial invasion of epithelial cells
sarcomere
MYLK
Adrenergic signaling in cardiomyocytes
-log 10(pvalue)
myosin complex
-log10(pvalue)
95
5
45
Focal adhesion
9
platelet aggregation
20
Adherens junction
6
muscle contraction
16
ACTA2 ACTG2
Oxytocin signaling pathway
3
Z disc
12
Regulation of actin cytoskeleton
motor activity
CGMP-PKG signaling pathway
actin filament binding
8
Salmonella infection
calmodulin binding
Leukocyte transendothelial migration
actin binding
”
actin cytoskeleton
5
10
15
20
interacted
Fold enrichment
focal adhesion
30
60
90
Fold enrichment
progression mainly involved muscle contraction and fo- cal adhesion, which will provide a direction for clinical diagnosis and treatment of cancer patients.
4 DISCUSSION |
It has been reported that the myosin family was sub- divided into two groups: Class I (small, mostly mono- meric motors) and Class II (dimeric myosins similar to skeletal muscle myosin; this class includes both muscle and nonmuscle myosins).13 Class II myosin, also known as traditional myosin, is a hexameric molecule composed of two heavy chains and two pairs of light chains, in- cluding the essential light chain and the regulatory light chain.14 Myosin II activity is mediated primarily by posttranslational phosphorylation of MYL9 (also known as MLC2, MRLC1, or MLC-2C) and by the opposite ac- tivity of MLC kinase and an MLC phosphatase.15
MYL9 played an important role not only in onco- genesis but also could be used as a sensitive biomarker for tumor diagnosis. For example, in colon cancer, MYL9
expression was downregulated in tumor tissues com- pared with normal tissues and the area under the curve (AUC) of MYL9 in diagnosis for colon cancer was 0.826, indicating statistical significance.16 A study of MYL9 in nonsmall cell lung cancer (NSCLC) showed that the ex- pression levels of and MYL9 were significantly lower in cancer tissue than those in the paraneoplastic and nor- mal tissues, suggesting that low MYLK and MYL9 ex- pressions might be associated with the development of NSCLC.17 Studies exhibited that MYL9 was involved in regulating breast cancer invasion,18 and MYL9 depletion did not influence cell cycle progression or induce cell death in MDA-MB-231 cells but obviously reduced in- vasiveness.19 In the present study, we found that the expression levels of the MYL9 gene were also down- regulated in COAD compared with the corresponding normal tissues, the OS (p = . 0061) and DFS (p = . 019) of the low MYL9 expression group were longer than the high MYL9 expression group, which were in ac- cordance the result about the DFS of colon cancer pa- tients in PrognoScan database and suggested that MYL9 could be used as a biomarker for COAD diagnosis and
therapy. The OS of low MYL9 gene expression group in LGG (p =. 0015) and MESO (p =. 0054) was obviously longer than that in the MYL9 high expression group, while it became shorter in ACC (p = . 025). Interestingly, the DFS of low MYL9 gene expression group in THYM (p =. 013) was longer than the high MYL9 gene expres- sion group. Therefore, the value of MYL9 for diverse tumor diagnosis and therapy was different, which de- served further investigation. The results from Prognoscan and Kaplan-Meier plotter databases showed that MYL9 expression correlated significantly with prognosis in lung cancer, breast cancer, colorectal cancer, brain glioma, prostate cancer, especially ovarian cancer and gastric cancer. We further used the Kaplan-Meier plotter tool to study the relationship between MYL9 expression and different clinicopathological factors in gastric and ovar- ian cancer, finding that MYL9 expression may affect the prognosis of gastric cancer patients by affecting lymph node metastasis.
It is well known that tumor infiltrating immune cells, as an important part of the tumor microenvironment, are closely related to tumor genesis, development, or me- tastasis.20,21 All stromal cells are clustered in the tumor microenvironment, while cancer-related fibroblasts (CAFs) are the most abundant and play a key role in cancer progression.22 In this study, we explored the correlation between MYL9 gene expression and CAFs in all types of cancer in the TCGA dataset, finding that there is a significant positive correlation between the two in all kinds of tumor except SARC. In addition, we fur- ther employed the “Outcome” model of TIMER2.0 tool to analyze the effects of CAFs, age, gender, race, purity, and MYL9 on prognosis correlation by setting up multi- variable Cox Proportional Hazard Models. Results showed that the level of MYL9 expression and the degree of CAFs infiltration affected the CS of cancer patients, and they played different or adverse roles in different tumors. Recent studies showed that the CD69-My19 sys- tem in immune responses could be used as a new ther- apeutic target for intractable inflammatory disorders and tumors.23 In this study, we also investigated the effects of the immune infiltrating of B cell, CD8+T cell, CD4+T cell, macrophage, neutrophil, DC, and the expression of MYL9 on CS in different cancers, and results indicated that MYL9 played a strong role in regulating immune cell infiltration, with a particularly strong effect on CD8+T cell infiltration in BLCA, on B cell, CD8+T cell, CD4+T cell, macrophage, neutrophil, DC infiltration in LGG, and on neutrophil infiltration in MESO. Furthermore, we found that MYL9 expression correlated with the markers of monocyte, TAM, M1 and M2 macrophage, Tregs, and exhausted T cells, indicating that MYL9 may be capable of regulating the polarization of macrophages and may
play a role in immune escape in STAD and COAD. Therefore, our data manifested that the possibility of MYL9 combined with immune infiltration cells could be used as a new therapeutic target for patients with tumors.
Our previous data manifested that the MYL9 gene was associated with the prognosis in certain tumors. Subsequently, we integrated information on MYL9 binding components and genes associated with MYL9 expression in all tumors and performed a series of en- richment analyses to identify the potential role of muscle contraction and local adhesion in cancer etiology or pathogenesis. However, we just used public databases of Oncomine, TCGA, GEO, PrognoScan, and, Kaplan-Meier plotter datasets to illustrate the effects of the MYL9 gene on different tumors. When encountering problems in a particular tumor in clinical practice, we still need more evidence in cell and animal levels for detailed mechanism studies.
5 CONCLUSION |
In conclusion, our first pan-cancer analysis of MYL9 demonstrated statistical correlations of MYL9 expression with clinical prognosis, immune infiltrating across mul- tiple cancers, which aids in understanding the role of MYL9 in tumorigenesis from the perspective of clinical cancer samples. MYL9 can serve as a prognostic sig- nature in pan-cancer and is associated with immune infiltrating.
AUTHOR CONTRIBUTIONS
Data curation, formal analysis, investigation, methodol- ogy, software, visualization, and writing-original draft: Minghe Lv. Project administration: Xue Chen. Super- vision: Xue Chen. Validation: Lumeng Luo. Writing- review and editing: Minghe Lv and Xue Chen.
DATA AVAILABILITY STATEMENT
The datasets generated and analyzed during the current study are available in TCGA, GEO, Oncomine, GTEx, GEPIA 2, TIMER, TIMER2.0, PrognoScan, Kaplan-Meier Plotter, CPTAC, STRING, DAVID, KEGG, and GO datasets.
ORCID
Minghe Lv [ http://orcid.org/0000-0001-9342-773X
REFERENCES
1. Blum A, Wang P, Zenklusen JC. SnapShot: TCGA-analyzed tumors. Cell. 2018;173(2):530.
2. Wang Z, Jensen MA, Zenklusen JC. A Practical Guide to The Cancer Genome Atlas (TCGA). Methods Mol Biol. 2016;1418: 111-141.
3. Barrett T, Wilhite SE, Ledoux P, et al. NCBI GEO: archive for functional genomics data sets-update. Nucleic Acids Res. 2013;41(Database Issue): D991-D995.
4. Clough E, Barrett T. The gene expression omnibus database. Methods Mol Biol. 2016;1418:93-110.
5. Lu Q, Li J, Zhang M. Cargo recognition and cargo-mediated regulation of unconventional myosins. Acc Chem Res. 2014; 47(10):3061-3070.
6. Krendel M, Mooseker MS. Myosins: tails (and heads) of functional diversity. Physiology (Bethesda). 2005;20:239-251.
7. Conti MA, Adelstein RS. Nonmuscle myosin II moves in new directions. J Cell Sci. 2008;121(Pt 1):11-18.
8. Newell-Litwa KA, Horwitz R, Lamers ML. Non-muscle myo- sin II in disease: mechanisms and therapeutic opportunities. Dis Model Mech. 2015;8(12):1495-1515.
9. Vicente-Manzanares M, Ma X, Adelstein RS, Horwitz AR. Non-muscle myosin II takes centre stage in cell adhesion and migration. Nat Rev Mol Cell Biol. 2009;10(11):778-790.
10. Cermák V, Kosla J, Plachý J, Trejbalová K, Hejnar J, Dvorák M. The transcription factor EGR1 regulates metastatic potential of v-src transformed sarcoma cells. Cell Mol Life Sci. 2010;67(20): 3557-3568.
11. Gilles L, Bluteau D, Boukour S, et al. MAL/SRF complex is involved in platelet formation and megakaryocyte migration by regulating MYL9 (MLC2) and MMP9. Blood. 2009;114(19): 4221-4232.
2. Sturge J, Wienke D, Isacke CM. Endosomes generate localized Rho-ROCK-MLC2-based contractile signals via Endo180 to pro- mote adhesion disassembly. J Cell Biol. 2006;175(2):337-347.
13. Berg JS, Powell BC, Cheney RE. A millennial myosin census. Mol Biol Cell. 2001;12(4):780-794.
14. Nevitt C, Tooley JG, Schaner TC. N-terminal acetylation and methylation differentially affect the function of MYL9. Biochem J. 2018;475(20):3201-3219.
15. Sandquist JC, Swenson KI, Demali KA, Burridge K, Means AR. Rho kinase differentially regulates phosphorylation of non- muscle myosin II isoforms A and B during cell rounding and migration. J Biol Chem. 2006;281(47):35873-35883.
16. Yan Z, Li J, Xiong Y, Xu W, Zheng G. Identification of can- didate colon cancer biomarkers by applying a random forest approach on microarray data. Oncol Rep. 2012;28(3): 1036-1042.
7. Tan X, Chen M. MYLK and MYL9 expression in non-small cell lung cancer identified by bioinformatics analysis of public expression data. Tumour Biol. 2014;35(12):12189-12200.
18. Jurmeister S, Baumann M. Balwierz A, et al. MicroRNA-200c represses migration and invasion of breast cancer cells by targeting actin-regulatory proteins FHOD1 and PPM1F. Mol Cell Biol. 2012;32(3):633-651.
19. Medjkane S, Perez-Sanchez C, Gaggioli C, Sahai E, Treisman R. Myocardin-related transcription factors and SRF are required for cytoskeletal dynamics and experimental me- tastasis. Nat Cell Biol. 2009;11(3):257-268.
20. Fridman WH, Galon J, Dieu-Nosjean MC, et al. Immune in- filtration in human cancer: prognostic significance and disease control. Curr Top Microbiol Immunol. 2011;344:1-24.
21. Steven A, Seliger B. The Role of Immune Escape and immune cell infiltration in breast cancer. Breast Care (Basel). 2018; 13(1):16-21.
22. Chen X, Song E. Turning foes to friends: targeting cancer-associated fibroblasts. Nat Rev Drug Discov. 2019;18(2): 99-115.
23. Kimura MY, Koyama-Nasu R, Yagi R, Nakayama T. A new therapeutic target: the CD69-My19 system in immune re- sponses. Semin Immunopathol. 2019;41(3):349-358.
SUPPORTING INFORMATION
Additional supporting information may be found in the online version of the article at the publisher’s website.
How to cite this article: Lv M, Luo L, Chen X. The landscape of prognostic and immunological role of myosin light chain 9 (MYL9) in human tumors. Immun Inflamm Dis. 2022;10:241-254. doi:10.1002/iid3.557