Check for updates
Pan-Cancer Analysis of P3H1 and Experimental Validation in Renal Clear Cell Carcinoma
Yongjie Li1(D . Ting Wang2 . Feng Jiang3
Accepted: 19 December 2023 / Published online: 4 January 2024 @ The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2023
Abstract
Prolyl 3-hydroxylase 1 (P3H1) has been implicated in cancer development, but no pan- cancer analysis has been conducted on P3H1. In this study, for the first time, aspects asso- ciated with P3H1, such as the mRNA expression, any mutation, promoter methylation, and prognostic significance, the relationship between P3H1 and clinicopathological param- eters, drug sensitivity, and immune cell infiltration were investigated by searching several databases including The Cancer Genome Atlas (TCGA), Genotype-Tissue Expression (GTEx), cBioPortal, and The Tumor Immune Evaluation Resource (TIMER2.0) using bio- informatics tools. The findings indicate significant differential expression of P3H1 in most tumors when compared to normal tissues, with a strong association with clinical prognosis. A pan-cancer Cox regression analysis revealed that high P3H1 expression is significantly associated with low overall survival in patients with brain lower grade glioma, kidney clear cell carcinoma, adrenocortical cancer, liver hepatocellular carcinoma, mesothelioma, sar- coma, uveal melanoma, bladder urothelial carcinoma, kidney papillary cell carcinoma, kid- ney chromophobe, thymoma, and thyroid carcinoma. A negative correlation was observed between P3H1 DNA methylation and its expression. P3H1 is significantly associated with infiltrating cells, immune-related genes, tumor mutation burden, microsatellite instability, and mismatch repair. Finally, A significant correlation was found between P3H1 expres- sion and sensitivity to nine drugs. Thus, enhanced P3H1 expression is associated with poor prognosis in a variety of tumors, which may be due to its role in tumor immune regulation and tumor microenvironment. This pan-cancer analysis provides insight into the function of P3H1 in tumorigenesis of different cancers and provides a theoretical basis for further in-depth studies to follow.
Keywords P3H1 . Pan-cancer . Prognosis . Immune infiltration . Tumor microenvironment
☒ Yongjie Li 11753599@qq.com
1 School of Pharmacy, Shaoyang University, Shaoyang, Hunan, China
2 The Second Affiliated Hospital, Hengyang Medical School, University of South China, Hengyang, Hunan, China
3 Department of Nutrition, Taizhou Central Hospital, Taizhou, Zhejiang, China
Background
According to the most recent evaluation by the International Agency for Research on Can- cer, a subsidiary of the World Health Organization, the global incidence of new cancer cases has escalated to 19.29 million in 2020, with corresponding 9.96 million fatalities [1]. In the twenty-first century, cancer is expected to overtake cardiovascular disease as the leading cause of premature death. Alarmingly, the current data also show that for the first time, breast cancer has surpassed lung cancer as the most commonly diagnosed can- cer worldwide. The current treatment modalities for cancer include surgery, chemotherapy, radiotherapy, and tumor immunotherapy, which have been rapidly developing in recent years [2]. Despite the clinical success of these treatments, the prognosis and survival rates of cancer patients are still not optimistic due to drug resistance, adverse drug reactions, and individual differences [3, 4]. As a result, more new tumor biomarkers and therapeutic tar- gets are urgently needed for cancer diagnosis and treatment.
Collagen synthesis and assembly is performed by the enzyme encoded by the prolyl 3-hydroxylase 1 (P3H1) gene. This enzyme is a member of the family of collagen proline hydroxylases and is found in the endoplasmic reticulum [5]. P3H1 belongs to a family of gene products that also includes the isozymes prolyl 3-hydroxylase 2 (P3H2), -3 (P3H3), and -4 (P3H4) and cartilage-associated protein (CRTAP). P3H1, P3H2, and P3H3 all con- tain highly conserved 2-ketoglutarate, ascorbate, and Fe(II)-dependent dioxygenase struc- tural domains that hydroxylate specific proline residues [6, 7]. P3H is responsible for pro- line 3-hydroxylation, an important post-translational modification of collagen, and its loss of function contributes to the development of the disease [8]. Several investigations have linked osteogenesis imperfecta to P3H1 mutations [9-12]. Retinal tears and posterior vitre- ous detachment are attributed to mutations in the P3H1 gene, coding for P3H1, which is involved in post-translational modification of type I, type II, and type V collagen [13]. In addition, hearing is greatly affected in P3H1 gene-deficient mice [14]. However, evidence links certain P3H family members to an increased risk of cancers, such as bladder cancer [15], lung cancer [16], breast cancer [17], and renal clear cell carcinoma [18].
However, bioinformatics studies of the role of P3H1 in pan-cancer are lacking. There- fore, in this study for the first time, several publicly available free databases were searched, and the role of P3H1 in pan-cancer was systemically analyzed. P3H1 may have a role in multiple aspects of pan-cancer, including mRNA expression, clinical prognosis, genetic alterations, immune cell infiltration, and drug sensitivity (Supplemental Fig. 1). P3H1 is a biomarker for immune infiltration and prognosis and a viable target for tumor treatment. This study may give a theoretical foundation for a deeper understanding of the function of P3H1 in tumor immunotherapy.
Methods
Differential Expression Analysis and Data Processing
Data of P3H1 expression was gathered from the Genotype-Tissue Expression (GTEx) database (https://commonfund.nih.gov/GTEx) for 31 normal tissues and from the Cancer Cell Line Encyclopedia (CCLE) (https://sites.broadinstitute.org/ccle/) for 21 tumor cell lines [19, 20]. The degree of variation in P3H1 expression between normal and cancerous
tissues in 33 tumors was analyzed using The Cancer Genome Atlas (TCGA) (https://tcgad ata.nci.nih.gov/tcga/) datasets [21]. The TCGA and GTEx data were collected from the UCSC Xena database (https://xena.ucsc.edu/); data for tumor tissues were obtained from the TCGA, and those for normal tissues were from both the TCGA and the GTEx [22]. UCSC Xena is a visualization online platform that provides TCGA genomic data. Down- loading the pan-cancer TPM expression values from the UCSC Xena website, which includes standardized TCGA and GTEx RNA Seq data, allows for more reliable expression analysis of tumor and normal samples. We downloaded renal clear cell carcinoma RNA Seq data in TPM data format from the UCSC Xena website, including samples from GTEx normal tissues (28 cases), TCGA adjacent tissues (72 cases), TCGA tumor tissues (531 cases), all sourced from UCSC Xena database and processed through the Toil pipeline.
Clinical Staging and Survival Analysis
Clinical data from the TCGA database were used to draw correlations between P3H1 expression with clinicopathological staging and patient prognosis, primarily in terms of overall survival (OS). We examined OS for all 33 cancers using Cox regression analysis and displayed the findings using Forest plots and Kaplan-Meier curves. The R packages survminer and survival were employed for analysis and visualization.
P3H1 Gene Mutation Analysis
To analyze mutation in the P3H1 gene, the cBioPortal database (http://www.cbioportal. org) was selected [23]. Through this site, we obtained information on the frequency of P3H1 gene mutation, mutation type, and changes in copy number and obtained information on promoter methylation in pan-cancerous tissues from the TCGA database, using HM450 methylation data.
Enrichment Analysis of P3H1
We performed a gene set enrichment analysis (GSEA) using data from the Reactome data- base, on the possible molecular mechanism of P3H1 in 33 cancers. ClusterProfiler, an R tool, was used for the analysis and visualization.
Tumor Microenvironment (TME) Analysis and Estimation of STromal and Immune Cells in MAlignant Tumor (ESTIMATE) Analysis
High and low P3H1 gene expression groups were visually distinguished through one box plot per tumor was generated to visually distinguish between, and a heat map was used to illustrate the relationship between gene and pathway scores as part of a TME study [24]. Then, the R language “ESTIMATE” package was used to calculate stromal cell scores, immune cell scores, combined stromal and immune cell scores, and tumor purity were cal- culated for tumor tissues, and all correlation results were presented using heat maps.
P3H1 Expression and Immune Correlation Analysis
Heat maps using data on pan-cancer immune infiltration obtained from the TIMER2.0 database (http://timer.comp-genomics.org/) were utilized to illustrate the link between P3H1 expression and each immune cell [25]. P3H1 expression levels were also com- pared to those of other immune-related genes (such as major histocompatibility com- plex (MHC), immune activation genes, immune suppression genes, chemokines, and chemokine receptors) and used the “reshape2,” “RcolorBreyer,” and “ggplot2” packages to graphically represent our findings.
Gene Expression for Mismatch Repair (MMR), Microsatellite Instability (MSI), and Tumor Mutation Burden (TMB) are all Associated with P3H1 Expression
Using Perl scripts, TMB scores were computed and adjusted for total exon length. For all samples, MSI levels were calculated using information on somatic mutations obtained from TCGA. The R packages “reshape2” and “RcolorBreyer” were used to pre- pare heat maps to assess the correlation of P3H1 expression with TMB and MSI. P3H1 expression was correlated with MMR gene expression in pan-cancer using TCGA data. This included MLH1, MSH2, MSH6, PMS2, and EpCAM. We used the “reshape2” and “RcolorBreyer” packages included in the R software suite to create these charts.
P3H1-Related drug Sensitivity Analysis
Tumor cell IC50 and gene expression data were downloaded from the Genomics of Drug Sensitivity in Cancer (GDSC) (https://www.cancerrxgene.org/) database, the rela- tionship between P3H1 and drug IC50 was analyzed, and the correlation between P3H1 expression and IC50 for each drug separately plotted [26].
Quantitative Polymerase Chain Reaction (Q-PCR)
Human renal cancer cell lines Caki-1, OS-RC-2, 786-O, and 769-P, and the human renal proximal tubule cells HK-2 were obtained from the Cell Bank of the Chinese Academy of Sciences (Shanghai, China). We followed the protocol for RNA extraction, reverse transcription for cDNA synthesis, and real-time fluorescence quantitative PCR proce- dures. P3H1, forward 5’- GATCCAGGACAGGGTGCAG-3’, reverse 5’-GCTCATCCT TGGGCTTCGAT-3’; ß-actin forward 5’-TTCCTTCCTGGGCATGGAGTC-3’, reverse 5’-TCTTCATTGTGCTGGGTGCC-3’.
Western Blotting
Protein was extracted using chilled RIPA buffer containing a protease (phosphatase) inhibitor. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis was performed to separate proteins, which were then transferred to the polyvinylidene fluoride (PVDF) membrane. Primary antibody (P3H1, 1:1000) was diluted with antibody diluent, added to the membrane, and incubated at room temperature for 10 min, and then stored at 4 ℃ overnight after being sealed with 5% skim milk for 1 h. Goat anti-rabbit IgG (H+L)
conjugated with horseradish peroxidase was diluted 1:10,000 in 5% skim milk powder- TBST and gently shaken for 40 min at room temperature. Enhanced chemiluminescence reagent was added to the membrane and reacted for 2-5 min and placed in a developer for development and exposure. The developed images were visualized and saved for data analysis. After development, the integrated optical density values of the strips were read using the software Image J and the data were analyzed.
Results
P3H1 Expression in Multiple Cancers and Normal Tissues
A combine study of TCGA and GTEx datasets revealed that P3H1 was differentially expressed in 27 cancers. Compared with normal tissues, the expression of P3H1 was high in 18 cancers and low in 9 cancers (Fig. 1A). Based on the mean P3H1 expression level, we rated 33 tumors from highest to lowest, using data from the TCGA database (Fig. 1B). The expression of P3H1 varies in all types of cancer, with the highest expression in sarcoma (SARC) and the lowest in kidney chromophobe (KICH). Next, using the GTEx database, we analyzed the physiological levels of the P3H1 gene among different tissues (Fig. 1C) and found relatively low P3H1 expression in most other normal organs, but among the highest in the pituitary, the nerve, and the testis tissues. CCLE results showed frequently higher P3H1 expression in many tumor cell lines compared to normal tissues (Fig. 1D).
A
10
ns
.***
ns
ns
ns
=
5
P3H1 expression TPM
tumor_type
0
normal
tumor
5
-10
.
ACC
BLCA
BRCA
CESC
CHOL
COAD
DLBC
ESCA
GBM
HNSC
KICH
KIRC
KIRP
LAML
LGG
LIHC
LUAD
LUSC
MESO
OV
PAAD PCPG
PRAD
READ
SARC
SKCM
STAD
TGCT
THCA
THYM
UCEC
UCS
UVM
tissue
B
Mean expression of P3H1 in TCGA
C
Mean expression of P3H1 in GTEx
4
7,88
-
-
-
LAML
W
-
-
-
0.03
0.14
U
-
CHÓN
-
-
-
Adpine Tiasve
-
-
D
Mean expression of P3H1 in CCLE
8.54
-
-
CIAO/READ ESCA
-
THCA
4.75
-
BANG
1.26
-
-
-
WAD
-
Đ
A
B
C
Type
Tumor
Normal
Type
Tumor
Normal
Type
Tumor
Normal
.
7-
.***
-
P3H1 expression log2(TPM+0.001)
P3H1 expression log2(TPM+0.001)
P3H1 expression log2(TPM+0.001)
.6
6
6
5
5
5
A
4
4
3
3
·
2
:
3
Tumor
BLCA
Normal
Tumor
BRCA
Normal
Tumor
CHOL
Normal
D
E
F
Type
Tumor
Normal
Type
Tumor
Normal
Type
Tumor
Normal
…
-
P3H1 expression log2(TPM+0.001)
5
P3H1 expression log2(TPM+0.001)
61
P3H1 expression log2(TPM+0.001)
6
4
5
4
4
.
4
3
3
2
2
2
1
Tumor
COAD
Normal
Tumor
ESCA
Normal
Tumor
HNSC
Normal
G
H
Type
Tumor
Normal
Type
Tumor
Normal
Type
Tumor
Normal
71
P3H1 expression log2(TPM+0.001)
P3H1 expression log2(TPM+0.001)
6
P3H1 expression log2(TPM+0.001)
6
5
5
5
%
4
3
3
3
2
2
Tumor
KIRC
Normal
2
Tumor
KIRP
Normal
Tumor
LIHC
Normal
J
K
L
Type
Tumor
Normal
Type
Tumor
Normal
Type
Tumor
Normal
…
7
P3H1 expression log2(TPM+0.001)
6
P3H1 expression log2(TPM+0.001)
P3H1 expression log2(TPM+0.001)
5.0
6
4.5
.5
5
4.0
4
4
3.5
3
3
3.0
:
Tumor
LUAD
Normal
Tumor
LUSC
Normal
Tumor
PRAD
Normal
M
N
Type
Tumor
Normal
Type
Tumor
Normal
Type
Tumor
Normal
6-
.
P3H1 expression log2(TPM+0.001)
>5
P3H1 expression log2(TPM+0.001)
P3H1 expression log2(TPM+0.001)
6.0
4
5
5.5
4
5.0
4.5
2
3
4.0
Tumor
STAD
Normal
Tumor
THCA
Normal
Tumor
UCEC
Normal
P3H1 Expression Levels in Paired Tumors and Normal Tissues
We used the TCGA database and further compared the difference of P3H1 expression between paired tumor and paraneoplastic tissues in multiple cancers (Fig. 2A-2O). The data showed that compared to paraneoplastic tissues, P3H1 showed high expression in 15 cancers, including bladder urothelial carcinoma (BLCA), breast invasive carcinoma (BRCA), cholangiocarcinoma (CHOL), colon adenocarcinoma (COAD), esophageal car- cinoma (ESCA), head and neck squamous cell carcinoma (HNSC), kidney clear cell car- cinoma (KIRC), kidney papillary cell carcinoma (KIRP), liver hepatocellular carcinoma (LIHC), lung adenocarcinoma (LUAD), lung squamous cell carcinoma (LUSC), prostate
adenocarcinoma (PRAD), stomach adenocarcinoma (STAD), thyroid carcinoma (THCA), uterine corpus endometrioid carcinoma (UCEC).
An Analysis of the Correlation Between P3H1 Expression and Clinicopathology Across Cancers
Our analysis included patients with malignancies at different stages (I, II, III, and IV), and assessed how P3H1 expression correlated with clinicopathological characteristics.
The study found that P3H1 is expressed differently in various stages of several types of cancer (Fig. 3), including adrenocortical cancer (ACC), BLCA, COAD, ESCA, KIRC, LIHC, mesothelioma (MESO), STAD, and THCA. In addition, we use the UALCAN data- base to evaluate the expression of P3H1 protein in different tumor stages, and we found that P3H1 protein is highly expressed in different stages of some tumor types (Supplemen- tal Fig. 2).
An Analysis of the Correlation Between P3H1 Expression and Genetic Alterations
Figure 4A shows the pan-cancer analysis of the link between the amount of P3H1 gene expression and the number of copies. In pheochromocytoma and paraganglioma (PCPG) tumors, the expression of the P3H1 gene was linked to the number of copies, and this relationship was the most robust of any type of tumor ( **** p<0.0001). The
A
B
C
ns
2.5-
…
ns
ns
ns
P3H1 expression log2(TPM+0.001)
10
P3H1 expression log2(TPM+0.001)
P3H1 expression log2(TPM+0.001)
ns
10.0
ns
9
n$
.
ns
.
8
ns
ns
ns
7.5
.
6
6
5.0
3
4
2.5
2
Stage I
Stage II
ACC
Stage II
Stage IV
Stage I
Stage II
BLCA
Stage III
Stage IV
0
Stage I
Stage II
COAD
Stage III
Stage IV
D
E
F
0-
ns
ns
ns
ns
10
ns
P3H1 expression log2(TPM+0.001)
ns
P3H1 expression log2(TPM+0.001)
P3H1 expression log2(TPM+0.001)
ns
8
ns
ns
ns
10
.
ns
8
-
.
6
!
6
:
5
4
:
4
2
·
Stage I
Stage II
Stage III
Stage IV
Stage I
Stage II
Stage III
Stage IV
Stage I
Stage II
Stage III
Stage IV
ESCA
KIRC
LIHC
G
H
ns
n$
:
ns
ns
P3H1 expression log2(TPM+0.001)
ns
P3H1 expression log2(TPM+0.001)
P3H1 expression log2(TPM+0.001)
ns
9
ns
ns
8
.
ns
7.5
ns
ns
ns
ns
·
7
6
5.0
5
4
2.5
·
:
·
Stage I
Stage II
Stage III
MESO
Stage IV
2
Stage I
Stage II
STAD
Stage III
Stage IV
Stage I
Stage II
THCA
Stage III
Stage IV
A
B
label
non-significant
positive
KIRC, n = 522, r = 0.37(pearson), p.value= 0
0.5
P3H1 Relative linear copy-number values
A
0.5
043
0.4
4
Y
A
S
correlation
J
3
E
5
De
5
0.0
3
5
2
.2
0.2
0.2
A
1
1
I
-0.5
2
2
A
0.0
0
PCPG
UVM
DLBC
THYM
LGG
SARC
LUSC
KIRC
CHOL
LUAD
KIRP
KICH
BLCA-
LIHC
SKCM-
PAAD-
UCEC
ESCA
BRCA
MESO
TGCT
OV
HNSC
CESC
STAD
PRAD
ACC
READ
COAD
UCS-
GBM
LAML
THCA-
P3H1 expression log2(TPM+0.001)
2
4
6
8
Correlation between CNA and mRNA expression
C
D
label
negtive
non-significant
KIRC, n = 317, r = - 0.14(pearson), p.value= 0.0116
0.0
LO
0.0
OU
P3H1 Methylation (HM450)
0.75
0.
0.1
0.14
0.1
N
-0.2
0,1
0.2
A
correlation
A
0.2
02
2
2
2
5
8
0.3
8
2
05
0.50
0.4
9
DA
I
=
=
-0.6
0.25
a.7
.
LAML
GBM
KICH
KIRP
LUAD
THCA
UCEC
KIRC
READ
CESC
PAAD
THYM-
DLBC
TGCT
HNSC
BRCA
PRAD
SKCM-
BLCA
SARC
LUSC
MESO
UCS
PCPG-
LIHC
COAD
STAD
ESCA-
CHOL
LGG-
UVM-
A
ACC-
Ov-
P3H1 expression log2(TPM+0.001)
3
5
7
Correlation between Methylation and mRNA expression
relationship between the level of P3H1 gene expression and the level of methylation of its promoter are presented in Fig. 4C. In ovarian plasmacytoid cystic adenocarci- noma, a strongly negative relationship was noted between the level of gene expres- sion and the level of methylation of its promoter (*p <0.05). As shown in Figs. 4B and 4C, the expression of P3H1 in KIRC is positively correlated with its number of copies ( **** p <0.0001), and is negatively correlated with its promoter methylation (*p<0.05).
A Pan-Cancer Analysis and Determination of The Prognostic Value of P3H1
We examined the link between P3H1 expression and patient outcome using a pan-can- cer dataset and measured the OS. A Cox regression analysis of 33 cancers showed a significant association of P3H1 to OS in 12 cancers, including brain lower grade glioma (LGG), KIRC, ACC, LIHC, MESO, SARC, uveal melanoma (UVM), BLCA, KIRP, KICH, thymoma (THYM), and THCA (Fig. 5A). Interestingly, P3H1 was a high-risk gene in 10 of the 11 cancers with a strong link to OS, but not in THYM. Kaplan-Meier survival curves showed a significant association of higher P3H1 expres- sion and worse OS in ACC, BLCA, CHOL, COAD, KIRC, KIRP, LGG, LIHC, MESO, SARC, and UVM (Fig. 5B-5L). Kaplan-Meier analysis showed that these 11 types of cancer were more likely to have a worse outcome in case of increased P3H1 transcript levels.
A
B
C
D
pvalue
Hazard ratio
LGG
<0.001
2.140(1.736-2.639)
ACC P3H1 Survival
BLCA P3H1 Survival
CHOL P3H1 Survival
KIRC
<0.001
1.915(1.598-2.295)
ACC
<0.001
3.174(2.095-4.808)
1.00
1.00-
LIHC
<0.001
1.825(1.455-2.288)
MESO
<0.001
1.899(1.433-2.516)
0.25
Survival probability
SARC
<0.001
1.382(1.149-1.662)
Survival probability
UVM
0.001
2.545(1.435-4.513)
0.50
8 50
BLCA
0.002
1.277(1.097-1.487)
KIRP
0.005
1.958(1.226-3.128)
P < 0.0001
p= 0.0029
P=0.045
KICH
0.008
2.643(1.285-5.433)
:
THYM
0.014
0.233(0.073-0.745)
Tìmg (days)
-
.
Tive (days)
THCA
0.031
2.354(1.082-5.125)
Time (orys)
Number at risk
Number at risk
Number at risk
STAD
0.068
1.216(0.986-1.500)
-
17
203
45
16
0
LUAD
0.076
1.203(0.981-1.474)
Ne
8
M
4
2
,
H
A
GBM
0.104
1.203(0.963-1.503)
.
29
202
22
bu
1000
Time (days)
4000
1
0000
4000
5006
Teve (daya)
2000
CHOL
0.138
2.006(0.799-5.036)
READ
0.196
1.633(0.776-3.437)
COAD
0.221
1.202(0.895-1.615)
PCPG
0.281
0.502(0.143-1.758)
LUSC
0.373
1.085(0.907-1.298)
E
F
G
BRCA
0.407
1.100(0.879-1.376)
COAD P3H1 Survival
KIRC P3H1 Survival
KIRP P3H1 Survival
CESC
0.424
1.131(0.836-1.531)
OV
0.465
0.953(0.839-1.084)
UCS
1.155(0.771-1.730)
1.00-
180.
0.484
PRAD
0.523
1.625(0.366-7.212)
HNSC
0.688
1.029(0.894-1.185)
Survival probability
Survival probability
SKCM
0.818
1.021(0.856-1.217)
50
TGCT
0.830
1.149(0.324-4.077)
UCEC
0.833
1.047(0.686-1.597)
p= 0.034
p < 0.0001
0.25
p=0.016
PAAD
0.864
1.024(0.783-1.339)
ESCA
0.975
0.996(0.751-1.321)
3000
6000
LAML
3000
0.981
0.995(0.664-1.492)
Time (days)
Tima (daya)
DLBC
0.998
0.998(0.255-3.916)
Number at risk
Number at risk
Number at risk
a
142
43
0.062 0.125 0.259 0500 100 200 420
142
53
9
1.
265
2
131
17
OR
23
A
-
143
24
143 0
65
H
M
A
Hazard ratio
a
Time (days)
4000
23
M
1000
Time gaysığ
-
2000
Time (dayı)
4000
5000
6000
H
J
K
L
LGG P3H1 Survival
LIHC P3H1 Survival
MESO P3H1 Survival
SARC P3H1 Survival
UVM P3H1 Survival
1.00
Survival probability
M
Survival probability
Survival grobuesity
p < 0.0001
p=0.0011
p=0.0001
p= 0.01
p = 0.00050
Time (days()
Time (days()
-
-
Time (days)
Time (dart)
-
Number at risk
Number at risk
Number at risk
Number at risk
Number at risk
4
259
16
10
0
2
183
182
40
2
43
63
12
NA
P
o
131
20
23
O
131
66
13
3
2
A2
0
25
0
30
4
ES
:
.
Time (days)
4000
.
Tìmg (days()
-
.
.
Timo (days)
5000
6000
.
500
2500
GSEA Enrichment Analysis
A GESA enrichment analysis was done on the Reactome database to assess the functional relevance of P3H1 expression in a variety of tumor types (Fig. 6). Data from nine tumor types, including ACC, BRCA, COAD, ESCA, KIRC, LGG, LIHC, ovarian serous cystad- enocarcinoma (OV), and pancreatic adenocarcinoma (PAAD), showed the association of P3H1 expression to cell cycle and immune regulatory pathways. These included cell cycle- related pathways (cell cycle, mitotic, M-phase, etc.) and immune regulatory pathways (innate immune system, cytokine signaling in immune system, adaptive immune system, and so on).
Correlation Analysis of P3H1 Expression and TME
Next, we examined the relation between P3H1 expression and TME in pan-cancer. As shown in Fig. 7A, P3H1 expression in pan-cancer samples was highly linked with path- way scores, including the epithelial-mesenchymal transition (EMT), the pan-focal T cell receptor, the immune checkpoint, and other pathways. Next, we assessed the corre- lation between P3H1 expression levels and the aforementioned three scores (Fig. 7B), revealing that the stromal score, the ESTIMATE score, the immune score, and the tumor Purity. Except for LAML, we found that P3H1 expression was strongly linked with the stromal score and the ESTIMATE score. In addition, P3H1 expression in all tumor types excluding testicular germ cell tumor (TGCT), skin cutaneous melanoma
Cell Lyde, MIOIC
Diseases of glycosylation
Regulation of Insulin-like Growth Factor (IG+)
Cel Cycle
transport and uptake by Insulin- like Growin For
RHO GTPase Effectors
Glycosaminoglycan metabolism
Degesofdycostation Diseases of glycosylation
Axon guidance
Diseases of metabolism
Platelet degranulation
Nervous system development
Platelet activation, signaling and aggregation
Response to elevated platelet cytosolic Ca2+
Infectious disease
Metabolism of carbohydrates
Platelet activation, signaling and aggregation
Transcriptional Regulation by TP53
Muscle contraction
Immunoregulatory interactions between a Lymphoid
and a non-Lymphoid cel
Signaling by Rho GTPases, Miro GTPases and RHOBTB3
CDC42 GTPase cycle
p.adjust
Diseases of metabolism
Signaling by Rho GTPases
Axon guidance
p.adjust
Hemostasis
p.adjust
Cellular responses to stress
Nervous system development
COPI-mediated anterograde transport
Cellular responses to external stimuli
0.005257606
Signaling by Receptor Tyrosine Kinases
0.0/1200081
Metabolism of carbohydrates
0.01780613
Metabolism of amino acids and derivatives
Hemostasis
Unfolded Protein Response (UPR)
Diseases of signal transduction by growth factor
receptors and second messengers
Neutrophil degranulation
ER to Golgi Anterograde Transport
Post-translational protein modification
Signaling by Interleukins
Leishmania infection
Class I MHC mediated antigen processing &
presentation
Signaling by GPCR-
Signaling by Receptor Tyrosine Kinases
Adaptive Immune System
Innate Immune System
Cytokine Signaling in Immune system
RNA Polymerase Il Transcription
Cytokine Signaling in Immune system
Neutrophil degranulation
Developmental Biology
GPCR ligand binding
Innate Immune System
Membrane Trafficking
GPCR downstream signaling
Adaptive Immune System
Vesicle-mediated transport
Developmental Biology
Vesicle-mediated transport-
0.4 0.6 0.8
0.0
0.3
0.6
0.9
0.25 0.50 0.75
D
E
F
Extracellular matrix organization
Collagen formation
Mitotic Prometaphase
Mitotic Metaphase and Anaphase
Degradation of the extracellular matrix
Mitotic Anaphase
Cell Cycle Checkpoints
Post-translational protein phosphorylation Regulation of Insulin-like Growth Factor (IGF)
Mitotic Metaphase and Anaphase
Processing of Capped Intron-Containing Pre-mRNA
transport and uptake by Insulin-like Growth Factor Binding Proteins (IGFEPS)
Separation of Sister Chromatids
Cell Cycle, Mitotic
Glycosaminoglycan metabolism
Cell Cycle Checkpoints
Metabolism of RNA
Diseases of glycosylation
Cell Cycle, Mitotic
Cell Cycle
Cilium Assembly
M Phase
M Phase
Immunoregulatory interactions between a Lymphoid
p.adjust
Processing of Capped Intron-Containing Pre-mRNA
p.adjust
DNA Repair
p.adjust
and a non-Lymphoid cell
Cell Cycle
Extracellular matrix organization
Platelet degranulation
Metabolism of RNA
Response to elevated platelet cytosclic Ca2+ Platelet activation, signaling and aggregation
081942002
0012549-49
Transcriptional Regulation by TP53
mRNA Splicing - Major Pathway
0.007569141
Cytokine Signaling in Immune system
Axon guidance
Neutrophil degranulation
Signaling by Hedgehog
RHO GTPase Effectors
Translation
Leishmania parasite growth and survival-
Anti-inflammatory response favouring Leishmania
Nervous system development
Signaling by Interleukins
parasite infection
Metabolism of carbohydrates
Innate Immune System
Hemostasis
Signaling by Rho GTPases
Generic Transcription Pathway
Nervous system development
Signaling by Rho GTPases, Miro GTPases and RHOBTB3
RHO GTPase Effectors
Axon guidance
DNA Repair
Class I MHC mediated antigen processing &
Signaling by Receptor Tyrosine Kinases
Cytokine Signaling in Immune system
presentation
Adaptive Immune System
Adaptive Immune System
RHO GTPase cycle
Cellular responses to stress
0.3
0.6
0.9
0.2 0.4 0.6 0.8
0.3 0.6 0.9
G
H
I
Cell Cycle Checkpoints
Membrane Trafficking
Mitotic Anaphase
Vesicle-mediated transport
Platelet degranulation
Cell Cycle, Mitotic
PTEN Regulation
Response to elevated platelet cytosolic Ca2+
Cell Cycle
S Phase
Apoptosis
M Phase
Regulation of TP53 Activity
G1/S Transition
DNA Repair
Cell Cycle
Platelet activation, signaling and aggregation
RHO GTPase Effectors
G1/S Transition
Neutrophil degranulation
Transcriptional Regulation by TP53
Cell Cycle, Mitotic
Programmed Cell Death
Cellular responses to stress
p.adjust
Diseases of signal transduction by growth factor
p.adjust
Innate Immune System
padjust
Cellular responses to external stimuli
receptors and second messengers
PTEN Regulation
Axon guidance
0.005675834
Signaling by VEGF
0003963524
S Phase
Transcriptional Regulation by TP53 PIP3 activates AKT signaling
8000543211
Organelle biogenesis and maintenance
Mitotic G1 phase and G1/S transition
Cell Cycle
Nervous system development
Mitotic G1 phase and G1/S transition
Adaptive Immune System
Infectious disease
Apoptosis
Hemostasis
Signaling by Rho GTPases, Miro GTPases and RHOBTB3
Transcriptional regulation by RUNX2
Vesicle-mediated transport
Signaling by Rho GTPases
Antigen processing: Ubiquitination & Proteasome
Programmed Cell Death
PIP3 activates AKT signaling
degradation
Intracellular signaling by second messengers
Diseases of signal transduction by growth factor
Neddylation
Neutrophil degranulation
receptors and second messengers
Class I MHC mediated antigen processing &
Membrane Trafficking
presentation
Adaptive Immune System
Cytokine Signaling in Immune system
Generic Transcription Pathway
Innate Immune System
Developmental Biology
0.2 0.4 0.6 0.8
0.4 0.5 0.6 0.7 0.8
0.4
0.6
0,8
(SKCM), and LAML, was positively linked with the immune score. Finally, except for LAML, the expression of P3H1 was strongly inversely linked with tumor purity.
Tumor Immune Cell Infiltration and P3H1 Expression
With the obvious connection between P3H1 expression and immune infiltration levels, a pan-cancer investigation of this association using the TIMER2.0 database revealed a positive association of P3H1 expression with cancer-associated fibroblasts (CAFs) and macrophages (MMCs) in most malignancies (Fig. 8A), suggesting the role of these infiltrating immune cells in cancer progression. P3H1 expression in pan-cancerous tis- sues was positively connected with the quantity of infiltrating immune cells, especially CAFs and MMCs.
A
EMT2
Pan_F_TBRs
EMT3
Base_excision_repair
DNA_replication
correlation
0.75
DNA_damage_response
.***
..
0.50
0.25
Mismatch_Repair
0.00
-0.25
Nucleotide_excision_repair
-0.50
Immune_Checkpoint
Antigen_processing_machinery
…
CD_8_T_effector
…
EMT1
…
OV
UVM
PAAD
LGG
ACC
GBM
KIRC
KICH
COAD
LIHC
READ
BLCA
ESCA
MESO
LUAD
KIRP
THCA
STAD
UCEC
SKCM
HNSC
LAML
LUSC
BRCA
CESC
CHOL
TGCT
PCPG
DLBC
PRAD
UCS
SARC
THYM
B
StromalScore
correlation
…
…
ESTIMATEScore
**
0.4
ImmuneScore
:
**
0.0
TumorPurity
:
**
-0.4
READ
UVM
COAD
PCPG
BLCA
KICH
ESCA
PAAD
BRCA
LGG
HNSC
DLBC
KIRC
SARC
STAD
MESO
OV
UCS
THCA
GBM
CESC
THYM
LUSC
TGCT
UCEC
KIRP
PRAD
CHOL
LUAD
LIHC
SKCM
ACC
LAML
Correlation of P3H1 Expression Levels with Immune-Related Genes
MHC, immunological activation genes, immune suppression genes, chemokines, and chemokine receptors, as well as the relationship between P3H1 expression and these genes were also studied. In most tumor types, P3H1 was positively linked with a large group of immune-related genes (Figs. 9A-E).
Correlation Analysis of P3H1 Expression in Different Tumors with TMB, MSI and MMR
We assessed the relation of P3H1 expression levels with TMB, MSI, and MMR (includ- ing MLH1, MSH2, MSH6, PMS2, and EPCAM). Immune checkpoint inhibitor sensitivity is significantly correlated with all three above-mentioned factors. Figure 10A depicts the correlation of P3H1 expression with TMB in six cancers, ACC, COAD, KIRC, LUAD, HNSC, and LAML. P3H1 expression was associated with MSI in the remaining four can- cers: TGCT, THCA, LGG, and STAD (Fig. 10B). In the majority of tumors, the MMR gene expression was significantly and strongly correlated with P3H1 expression, excluding CHOL, uterine carcinosarcoma (UCS), and diffuse large B-cell lymphoma (DLBC). Most
A
B_cell_memory_CIBERSORT
B_cell_memory_CIBERSORT_ABS
B_cell memory XCELL
_ B cell naive_CIBERSORT
B_cell naive_CIBERSORT_ABS
B_cell plasma_CIBERSORT
B_cell_plasma_CIBERSORT_ABS
Class_switched_memory_B_cell_XCELL
Cancer_associated fibroblast EPIC
Cancer associated fibroblast MCPCOUNTER
Cancer_associated_fibroblast_XCELL
_cell_CD4_central_memory_XCELL cell_CD4_effector_memory_XCELL
_cell_CD4_memory_activated_CIBERSORT T_cell_CD4_memory_activated CIBERSORT_ABS
_cell_CD4_memory_resting_CIBERSORT
T_cell_CD4_memory_resting_CIBERSORT_ABS
Myeloid_dendritic_cell activated_CIBERSORT Myeloid_dendritic_cell activated CIBERSORT_ABS
B cell MCPCOUNTER
B cell plasma XCELL
cell CD4 memory XCELL
T cell_CD4_naive_CIBERSORT
cell CD4_naive_CIBERSORT_ABS
cell_CD4_naive_XCELL
T_cell_CD4_non_regulatory_QUANTISEQ
_cell_CD4_non_regulatory_XCELL
T cell_CD4 Th1_XCELL
cell_CD4_Th2_XCELL
T_cell_CD8_central memory_XCELL
cell_CD8_CIBERSORT
T cell_CD8_CIBERSORT_ABS
cell_CD8_effector_memory_XCELL
cell CD8 MCPCOUNTER
T_cell_CD8 naive XCELL
T_cell_CD8 QUANTISEQ
Myeloid_dendritic_cell_activated XCELL
Myeloid_dendritic_cell_MCPCOUNTER
Myeloid_dendritic_cell_QUANTISEO
Myeloid_dendritic_cell_resting_CIBERSORT
Myeloid_dendritic_cell resting_CIBERSORT_ABS
B_cell_naive_XCELL
B cell QUANTISEQ
T cell_CD8 EPIC
I_cell_CD8 TIMER
Myeloid_dendritic_cell TIMER
Myeloid_dendritic_cell_XCELL
Plasmacytoid_dendritic_cell_XCELL
cell_CD4_EPIC
_cell_CD4_TIMER
_cell_CD8_XCELL
Endothelial cell EPIC
Endothelial_cell MCPCOUNTER Endothelial_cell_XCELL
Eosinophil CIBERSORT_ABS
B cell EPIC
B_cell XCELL
Eosinophil CIBERSORT
cell_gamma_delta_CIBERSORT_ABS _cell_gamma_delta_XCELL
Eosinophil_XCELL
cell gamma delta_CIBERSORT
. B_cell_TIMER
UVM
-
8
XIX
X
X
X
IX
UCS
X
×
IXI
X
XI
X
X
X
54
XI
XIX
X
D
X
X
KIXIX
XIXI
UCEC XI
X
XIX
THYM
CIXIX
X
KIXIX
X
X
3
X
THCA
X
TGCT
×
X
X
×
4
X
EK
STAD .
X
X
X
X
SKCM
X
X
SARC
XIX
X
X
*
X
X
READ
X
XI
X
XIX
X
IX
X
X
PRAD
X
IX
PCPG
X
X
3
X
XIXI
X
PAAD
XIXIX
IXIXI
XIX
S
X
(XIXIXIXI
OV -
KIX
X
EXIXI
MESO
XIX
XIXIXIX
X
X
XIXI
X
E
X
LUSC -
2
X
4
LUAD -
X
LIHC
X
LGG
KIRP
KIRC .
KICH
XIX
X
X
XIX
X
X
X
XX
EX
HNSC.
38
GBM
X
XIXIX
IXIXIX
I
KI
ESCA-
ZIXIXE
D
M
XX
DLBC
XIXIXI
X
XI
X
X
[X]
X
X D
X
X
XIX
X
KIXIX
COAD
X
X
CHOL
.
X
XIXIX
> X
XXX XIX
IX
X
XIX
X
CESC
BRCA -
BLCA -
XS
X
X
ACC
X
[X] X
[X]
KIX
X
XXXXX
XXL
X
XXIX
UVM
XI
X
X
L
X
X
X
X
K
X
x
X
X
UCS
X
XIX
X
8
X
× X
X
X
X
X
X
UCEC
x
X
X
THYM
X
X
X
THCA
%
TGCT
X
x
X
8
Z
X
STAD
SKCM-
X
SARC
X
X
correlation
READ
X
X
X
X
×
K
X
X
0.4
PRAD
PCPG .
X
X
X
IX
X
0.2
PAAD
X
K
XIX
OV-
X
X
X
X
0.0
MESO X
X
XI
X
X
X X
1
X
X
X
X
3
X
X
X
3
XIX
LUSC -
X
-0.2
LUAD -
LIHC
-0.4
LGG
KIRP
X
E
X
KIRC
X
pvalue
KICH
X
X
X
K
X
K
X
R
X
X
X
×
HNSC -
X
p<0.05
GBM-
X
X
X
X
ESCA-
8 p≥0.05
X
DLBC
X
X
X
X
X
X
X
X
A
X
X
X
X
X
COAD
CHOL
X
X
X
[2
X
X
X
4
X
X
X
14
4
XX
XIX
X
X JE
X
CESC
y
X
BRCA
BLCA
ACC SZ
Z
X
M
%
X
1
X
X
Hematopoietic_stem_cell_XCELL
Macrophage_EPIC
Macrophage_MO_CIBERSORT.
Macrophage_MO_CIBERSORT_ABS
Macrophage_M1_CIBERSORT,
Macrophage_M1_CIBERSORT_ABS.
Macrophage_M1_QUANTISEQ.
Macrophage_M1_XCELL.
Macrophage_M2_CIBERSORT _
Macrophage_M2_CIBERSORT_ABS.
Macrophage_M2_QUANTISEQ.
Macrophage_M2_XCELL
Macrophage_Monocyte_MCPCOUNTER
Macrophage_TIMER
Macrophage_XCELL
Mast_cell_activated_CIBERSORT
Mast_cell_activated_CIBERSORT_ABS.
Mast_cell_resting_CIBERSORT
Mast_cell_resting_CIBERSORT_ABS.
Mast_cell_XCELL
Macrophage_Monocyte_MCPCOUNTER
Monocyte_CIBERSORT
Monocyte_CIBERSORT_ABS
Monocyte_MCPCOUNTER
Monocyte_QUANTISEQ.
Monocyte_XCELL
Neutrophil_CIBERSORT
Neutrophil_CIBERSORT_ABS
Neutrophil_MCPCOUNTER
Neutrophil_QUANTISEQ.
Neutrophil_TIMER.
Neutrophil_XCELL
NK_cell_activated_CIBERSORT
NK_cell_activated_CIBERSORT_ABS.
NK_cell_EPIC
NK_cell_MCPCOUNTER
NK_cell_QUANTISEQ.
NK_cell_resting_CIBERSORT
NK_cell_resting_CIBERSORT_ABS.
NK_cell_XCELL
T_cell_NK_XCELL
Common_lymphoid_progenitor_XCELL
Common_myeloid_progenitor_XCELL
Granulocyte_monocyte_progenitor_XCELL.
T_cell_follicular_helper_CIBERSORT
T_cell_follicular_helper_CIBERSORT_ABS,
T_cell_regulatory_Tregs_CIBERSORT
T_cell_regulatory_Tregs_CIBERSORT_ABS
T_cell_regulatory_Tregs_QUANTISEQ.
T_cell_regulatory_Tregs_XCELL
A
B
TAPBP
-
HLA-A
CD276
TAP2
TNFRSF4
HLA-B
STING1
HLA-DPB1
TNFRSF8
HLA-C
TNFSF4
HLA-E
TNFRSF18
HLA-F
correlation
CXCR4
HLA-DMB
-
0.6
CD70
HLA-DMA
0.4
TNFRSF25
TAP1
NT5E
0.2
HLA-DQB1
ENTPD1
0.0
HLA-DRB1
CXCL12
HLA-G
-0.2
ULBP1
-0.4
CD40
HLA-DOA
PVR
HLA-DQA1
TNFSF9
HLA-DRA
IL2RA
HLA-DQA2
MICB
HLA-DPA1
IL6
HLA-DOB
TNFRSF14
B2M
-
TNFRSF9
correlation
UVM
COAD
LGG
PAAD
OV
BLCA
READ
BRCA
UCEC
PRAD
KICH
ESCA
PCPG
LIHC
HNSC
KIRC
STAD
DLBC
GBM
KIRP
LAML
LUAD
SARC
LUSC
CESC
UCS
CHOL
MESO
THCA
ACC
SKCM
TGCT
THYM
CD86
0.75
LTA
0.50
CD80
0.25
ICOSLG
0.00
CD28
-0.25
TNFSF138
-0.50
CD27
VSIR
ICOS
CD48
TNFRSF13C
TNFSF14
C
KLRK1
TGFB1
TNFSF13
NECTIN2
TNFSF15
IL 10RB
TMIGD2
ADORA2A
CD40LG
CSF1R
TNFRSF13B
KLRC1
TGFBR1
TNFSF18
LAG3
BTNL2
HAVCR2
TNFRSF17
PDCD1
IL6R
PDCD1LG2
correlation
RAET1E
-
HHLA2
-
.
KDR
0.50
UVM
OV
COAD
PAAD
READ
KICH
BLCA
KIRC
STAD
LIHC
ESCA
LGG
GBM
BRCA
KIRP
THCA
PCPG
LUSC
HNSC
LUAD
ACC
UCEC
LAML
SARC DLBC
CESC
UCS
SKCM
PRAD
MESO
CHOL
TGCT
THYM
IL10
LGALS9
0.25
CTLA4
0.00
CD96
-0.25
IDO1
D
TIGIT
CL26
CD244
CCL2
VTCN1
CXCL 12
BTLA
CCL11
KIR2DL3
CCL13
CD274
CCL3
KIR2DL1
CXCL3
CD160
-
.
.
CXCL8
UVM
COAD
OV
READ
PAAD
LGG
BLCA
KICH
STAD
KIRC
GBM
LIHC
ESCA
BRCA
KIRP
PCPG
ACC
HNSC
UCEC
LUAD THCA
SARC
LUSC
LAML
PRAD
CESC
MESO
DLBC
CHOL
UCS
SKCM
TGCT
THYM
CXCL5
CCL7
CCL8
CXCL1
CXCL6
CCL18
CCL5
CCL23
CXCL 16
CCL27
correlation
C
CR10
CCL21
CX3CL
0.50
CXCR4
CXCL2
0.25
CXCR5
CCL14
0.00
CCR1
CCL4
-0.25
CXCR3
CCL22
-0.50
CCR5
XCL1
CXCR1
correlation
CXCL 14
CCR3
0.6
CXCL13
CCR2
0.4
CCL 17
CCR8
0.2
CCL20
CCR7
0.0
CXCL 10
-0.2
XCL2
CCR4
-0.4
CXCL9
CXCR6
CXCL 11
CXCR2
CCL25
XCR1
CCL24
CX3CR1
-
CCL19
CCR6
CCL 16
CCR9
.
-
.
-
CCL1
UVM
COAD
READ
OV
PAAD
KICH BLCA
LGG
STAD
KIRC
LIHC
KIRP
GBM
BRCA
THCA
ESCA
LUSC
PCPG
HNSC
UCEC
DLBC
CHOL
LUAD
LAML
CESC
UCS
TGCT
PRAD
SARC
SKCM
ACC
MESO
THYM
CCL28
CCL15
CXCL 17
-
UVM
COAD
OV
BLCA
READ
PAAD
KICH
KIRC
BRCA
GBM
STAD
ESCA
DLBC
LIHC
TGCT
KIRP
PCPG
THCA
LUSC
HNSC
LUAD
SARC
LGG
ACC
PRAD
UCEC
UCS
CESC
SKCM
MESO
CHOL
LAML
THYM
A
B
TMB
MSI
UCS
LAML ** DLBC
ACC*
KICH
PAAD
STAD. CHOL DLBC
TGCT*
CHOL
0.4
READ
THYM
0.3
READ
PCPG
0.2
COAD*
ESCA
0.15
UVM
THYM
KIRC*
PCPG
THCA*
0
0
HNSC*
LIHC
SKCM
COAD
.2
15
ESCA
-0.4
LUAD*
UCS
-0.3
LGG*
UVM
UCEC
KIRP
GBM
STAD
SARC
LAML
ACC
BLCA
OV
KICH
MESO
SKCM
THCA
OV
SARC
PAAD
LGG
UCEC
HNSC
GBM
PRAD
LIHC
CESC
MESO
CESC BRCA LUSC
TGCT
BRCA
LUAD
KIRP
BLCA
PRAD LUSC
KIRC
C
MSH6
correlation
0.6
MLH1
**
**
0.3
MSH2
**
0.0
PMS2
**
-0.3
EPCAM
OV
ACC
PAAD
GBM
LIHC
LAML
ESCA
UVM
SKCM
STAD
LUAD
CESC
UCEC
KIRP
COAD
READ
BLCA
CHOL
HNSC
KICH
LUSC
LGG
TGCT
THCA
PRAD
UCS
KIRC
MESO
DLBC
SARC
THYM
BRCA
PCPG
of these cancer types, such as MLH1, MSH2, and MSH6, showed a positive correlation with P3H1 (Fig. 10C).
Expression of P3H1 and Sensitivities to Different Drugs
P3H1 expression may be linked to the emergence of treatment resistance; therefore, we searched the GDSC database on the performance of the nine most important anticancer medications in relation to P3H1 expression. Among eight of the nine drugs tested (sapi- tinib, osimertinib, acetalax, afatinib, gefitinib, lapatinib, erlotinib, and AZD3759), P3H1 expression was significantly positively correlated with drug IC50, while in tozasertib, it was significantly negatively correlated (Fig. 11A-I). Thus, P3H1 is a promising candidate for usage as a therapeutic target in human medicine.
Experimental Validation of P3H1 Expression in KIRC
On combining the results of the above-mentioned bioinformatic analyses, we discovered that an increase in P3H1 expression in KIRC is strongly linked to a worse prognosis and an advanced clinicopathological stage. Also for our further in-depth study in the future,
A
B
C
Tozasertib, n = 48, r = - 0.35(spearman), p.value= 0.0154
Sapitinib, n = 800, r = 0.31(spearman), p.value= 0
Osimertinib, n = 749, r = 0.28(spearman), p.value= 0
-
6
6
A
=
6
4
IC50 value of Tozasertib
IC50 value of Sapitinib
IC50 value of Osimertinib
4
3
2
0
0
-2
0
11
1
1
Y
I
1
-3
IL
4
5
&
P3H1 expression
7
8
4
P3H1 expression
6
8
10
-4
4
6
8
P3H1 expression
10
D
E
F
Acetalax, n = 726, r = 0.28(spearman), p.value= 0
Afatinib, n = 801, r = 0.28(spearman), p.value= 0
Gefitinib, n = 749, r = 0.27(spearman), p.value= 0
9
7.5
=
6
5.0
5.0
IC50 value of Acetalax
IC50 value of Afatinib
IC50 value of Gefitinib
2.5
3
2.5
0.0
~
0
0.0
-2.5
=
=
-3
-5.0
=
-2.5
11
4
P3H1 expression
6
8
10
4
P3H1 expression
6
8
10
4
P3H1 expression
6
8
10
G
H
Lapatinib, n = 754, r = 0.23(spearman), p.value= 0
Erlotinib, n = 749, r = 0.22(spearman), p.value= 0
AZD3759, n = 757, r = 0.22(spearman), p.value= 0
5.0
2
= M
.
5.0
6
IC50 value of Lapatinib
IC50 value of Erlotinib
IC50 value of AZD3759
2.5
2.5
0.0
0.0
0
U
=
P
-2.5
-2.5
III
4
P3H1 expression
6
8
10
4
P3H1 expression
6
8
10
4
P3H1 expression
6
8
10
we experimentally validated the expression of P3H1 mRNA and protein in normal human kidney cells and different human kidney cancer cell lines. Consistent with our bioinformat- ics analysis, we found at the mRNA as well as protein levels, P3H1 expression was signifi- cantly upregulated in human renal cancer cell lines Caki-1, OS-RC-2, 786-O, and 769-P, compared to normal human kidney cells HK-2 (Fig. 12A-B).
Discussion
Cancer is the second deadliest disease after cardiovascular disease; hence, early detection and treatment are crucial to improving survival rates [27]. Bioinformatics-based pan-can- cer analysis provides a significant theoretical basis for the prevention and tailored treatment of, which has been possible because of the fast growth of bioinformatics techniques [28]. Mutations in P3H1 are closely associated with osteogenesis imperfecta, a human genetic disease, the underlying cause of which is mainly related to alterations in collagen [29, 30]. Researchers found that in patients with lupus nephritis, serum P3H1 could serve as a
A
B
Relative expression of ?3H1
15
Caki-1
OS-RC-2
786-0
10
HK-2
769-P
5
P3H1
Actin
0
HK-2
Caki-1
OS-RC-2
786-O
769-P
biomarker [31]. Most studies on P3H1 are currently in the non-oncology field, and research on the pathogenic role of P3H1 in tumors is considerably lacking. A meta-analysis of genome-wide and proteomic data based on algorithms identified P3H1 as a potential bio- marker for colorectal cancer [32]. In another study, P3H1 in patients was associated with the prognosis of hepatocellular carcinoma and breast cancer [33]. Zhang Yin et al. showed that the typical features of the intraepithelial neoplasia- (HIN-) adenocarcinoma sequence are associated with extracellular matrix-related biological processes. They suggested that P3H1 may be the key protein of this sequence and is associated with extracellular matrix remodeling and immunosuppressive state in colorectal cancer remodeling [34].
This was the first full study to show the level of P3H1 expression in pan-cancer. Among the 18 tumor types, ACC, BLCA, BRCA, CHOL, DLBC, ESCA, GBM, HNSC, KIRC, KIRP, LGG, LIHC, PAAD, SARC, SKCM, STAD, THYM, and UCS exhibited signifi- cantly raised P3H1 expression. According to the TCGA database, P3H1 is expressed to different degrees in all cancer types. The highest amount of P3H1 expression was noted in SARC, while KICH has the lowest. P3H1 is also expressed in many normal human tissues, with the highest levels being in the pituitary, nerves, and testes, in that order. Nonetheless, its expression in most other normal human tissues was low. The CCLE results showed that the amount of P3H1 expression was usually higher in tumor cell lines than in normal tis- sues. Also, the P3H1 expression was higher in some types of cancer than in normal cells. According to the TCGA database, certain cancers at advanced stages, regardless of TNM staging, exhibit elevated P3HI expression; these tumor types include ACC, BLCA, COAD, ESCA, KIRC, LIHC, MESO, STAD, and THCA. These findings suggest that P3H1 may function as an oncogene in most tumors. Cox analysis and Kaplan-Meier plots used in this study visualized OS-related prognosis in a pan-cancer analysis and revealed the association of P3H1 expression with an inferior prognosis in a variety of tumor types. According to these results, P3H1 can be used as a prognostic biomarker for certain types of malignan- cies; however, additional studies and more precise data are required. In the genetic altera- tion analysis, we observed a positive correlation of P3H1 gene expression with copy num- ber and negatively correlated with methylation levels, consistent with the previous finding of upregulated mRNA expression. In recent years, the relationship between DNA meth- ylation and tumors is well elucidated. The expression of oncogenes is suppressed by DNA
methylation [35]. In another report, the expression of miRNAs was regulated through DNA methylation [36]. Therefore, the relationship between P3H1 expression and DNA meth- ylation should be investigated in greater depth. Our GSEA enrichment analysis revealed that P3H1 expression in nine malignancies was associated with either cell cycle regula- tion or immune regulation. These nine tumor types include ACC, BRCA, COAD, ESCA, KIRC, LGG, LIHC, OV, and PAAD, which involve cell cycle-related pathways like cell cycle, mitosis, M-phase, etc., and immune regulatory pathways like innate immune sys- tem, cytokine signaling in the immune system, adaptive immune system, and many others. According to these findings, P3H1 possibly regulates the cell cycle and immune response in a complex manner. Tumor cells reside in the extracellular matrix, soluble molecules, and tumor stromal cells that make up the TME [37]. In TME, non-tumor components such as immune cells and stromal cells have been proposed to be useful for the diagnosis and prog- nosis of tumors [38]. The ESTIMATE algorithm can be used to quantify tumor immune and stromal scores by calculating the immune score. Immune and stromal scores are com- puted using this algorithm by analyzing specific gene expression profiles of immune cells and stroma cells to predict the level of non-tumor cell infiltration [39]. Except for LAML, P3H1 expression was positively linked with stromal score and ESTIMATE score. Further- more, except for TGCT, SKCM, and LAML, P3H1 expression was substantially positively linked with immune scores in other cancers. Finally, P3H1 expression was shown to be inversely associated to tumor purity in the vast majority of tumor types, with the exception of LAML. The correlation between P3H1 expression and immunological infiltration was investigated by employing the TIMER2.0 database. In pan-cancerous tissues, P3H1 expres- sion was positively connected with the quantity of invading immune cells, especially CAFs and MMCs. MHC, immune activation genes, immune suppression genes, chemokines, and chemokine receptors were also studied, in addition to the relationship between P3H1 expression and these genes. These results showed a favorable correlation between P3H1 expression and the expression of genes related to immunity in a wide variety of tumor types. Tumor immunotherapy is a new generation of tumor treatment that has rapidly developed in recent years and holds great promise for clinical application. Currently, the main clinically used tumor immunotherapies include immune checkpoint inhibitors, peri- patetic cellular immunotherapy, and cancer vaccines [40]. We therefore examined any potential correlation of P3H1 expression with TMB, MSI, and MMR (including MLH1, MSH2, MSH6, PMS2, and EPCAM); immune checkpoint inhibitor sensitivity is substan- tially correlated with all three. P3H1 expression was observed to correlate with TMB in six tumor types, including ACC, COAD, KIRC, LUAD, HNSC, and LAML. In addition, in the remaining four tumors, TGCT, THCA, LGG, and STAD, P3H1 expression was associ- ated with MSI. MMR gene expression was significantly and strongly correlated with P3H1 expression levels in the majority of tumors, excluding CHOL, UCS, and DLBC; the major- ity of these cancer types revealed a positive correlation of MLH1, MSH2, and MSH6 with P3H1. Finally, drug sensitivity analysis suggested that P3H1 may act as a potential anti- cancer target. Based on the above findings, we may hypothesize that P3H1 expression is strongly linked to immune infiltration of tumor cells and, therefore, represents a novel tar- get for the development of immune checkpoint inhibitors. Finally, for in vitro experimen- tal validation, we selected P3H1 expression in human kidney cancer cell lines; the results were consistent with our bioinformatics analysis. Although we integrated data from multi- ple databases as much as possible, our study still has certain limitations. For example, due to the current experimental conditions, we could not further validate our specific results through specific in vivo experiments.
Conclusions
In conclusion, our analysis indicates that P3H1 can be used as a prognostic biomarker for a variety of tumors, and its elevated expression is associated with unfavorable prognoses for the majority of these cancers. P3H1 expression is also closely associated with immune cell infiltration and immune-related genes. By elucidating the function of P3H1 in tumor development, future precision cancer therapies, and personalized immunotherapies need to be developed.
Supplementary Information The online version contains supplementary material available at https://doi. org/10.1007/s12010-023-04845-8.
Acknowledgements We acknowledge the TCGA, GTEx, CCLE, TIMER2.0, cBioPortal, and GDSC data- bases for free use.
Author Contributions Conceptualization, Yongjie Li; Formal analysis, Yongjie Li and Ting Wang; Software analyses, Ting Wang; Visualization, Feng Jiang; Writing - original draft, Yongjie Li; Writing - review & editing, Yongjie Li.
Funding Not applicable.
Data Availability The data sets used in this research are publicly available online.
Declarations
Ethics Statement All the experiments were conducted in accordance with the ethical guidelines of Shaoyang University.
Consent for Publication Not applicable.
Conflict of Interest The authors declare that there are no conflict of interests.
References
1. Sung, H., Ferlay, J., Siegel, R. L., Laversanne, M., Soerjomataram, I., et al. (2021). Global cancer sta- tistics 2020: Globocan estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA: A Cancer Journal for Clinicians, 71, 209-249. https://doi.org/10.3322/caac.21660
2. Yang, M., Oh, I. Y., Mahanty, A., Jin, W. L., & Yoo, J. S. (2020). Immunotherapy for glioblastoma: current state, challenges, and future perspectives, Cancers (Basel), 12. https://doi.org/10.3390/cance rs 12092334
3. Bisht, D., Arora, A., & Sachan, M. (2022). Role of dna de-methylation intermediate “5-hydroxym- ethylcytosine” in ovarian cancer management: A comprehensive review. Biomedicine & Pharmaco- therapy, 155, 113674. https://doi.org/10.1016/j.biopha.2022.113674
4. Liu, X., Chen, L., & Wang, T. (2022). Overcoming cisplatin resistance of human lung cancer by sinomenine through targeting the mir-200a-3p-gls axis, J Chemother, 1-10. https://doi.org/10.1080/ 1120009X.2022.2111490.
5. Vranka, J. A., Pokidysheva, E., Hayashi, L., Zientek, K., Mizuno, K., et al. (2010). Prolyl 3-hydroxy- lase 1 null mice display abnormalities in fibrillar collagen-rich tissues such as tendons, skin, and bones. Journal of Biological Chemistry, 285, 17253-17262. https://doi.org/10.1074/jbc.M110.102228
6. Wu, J., Zhang, W., Xia, L., Feng, L., Shu, Z., et al. (2019). Characterization of ppib interaction in the p3h1 ternary complex and implications for its pathological mutations. Cellular and Molecular Life Sci- ences, 76, 3899-3914. https://doi.org/10.1007/s00018-019-03102-8
7. Hudson, D. M., Weis, M., Rai, J., Joeng, K. S., Dimori, M., et al. (2017). P3h3-null and sc65- null mice phenocopy the collagen lysine under-hydroxylation and cross-linking abnormality of
ehlers-danlos syndrome type via. Journal of Biological Chemistry, 292, 3877-3887. https://doi.org/ 10.1074/jbc.M116.762245
8. Cabral, W. A., Fratzl-Zelman, N., Weis, M., Perosky, J. E., Alimasa, A., et al. (2020). Substitution of murine type i collagen a1 3-hydroxylation site alters matrix structure but does not recapitulate osteogenesis imperfecta bone dysplasia. Matrix Biology, 90, 20-39. https://doi.org/10.1016/j.mat- bio.2020.02.003
9. Tan, W., Ji, Y., Qian, Y., Lin, Y., Ye, R., et al. (2022). Mutational screening of skeletal genes in 14 chinese children with osteogenesis imperfecta using targeted sequencing. Journal of Immunology Research, 2022, 5068523. https://doi.org/10.1155/2022/5068523
10. Zhytnik, L., Duy, B. H., Eekhoff, M., Wisse, L., & Pals, G., et al. (2022) Phenotypic variation in vietnamese osteogenesis imperfecta patients sharing a recessive p3h1 pathogenic variant, Genes (Basel), 13. https://doi.org/10.3390/genes13030407
11. Nadyrshina, D., Zaripova, A., Tyurin, A., Minniakhmetov, I., Zakharova, E., & Khusainova, R. (2022). Osteogenesis imperfecta: search for mutations in patients from the republic of bashkorto- stan (russia), Genes (Basel), 13. https://doi.org/10.3390/genes13010124
12. Tuysuz, B., Elkanova, L., Uludag, A. D., Gulec, C., Toksoy, G., et al. (2022). Osteogenesis imper- fecta in 140 turkish families: Molecular spectrum and comparison of long-term clinical outcome of those with col1a1/a2 and biallelic variants. Bone, 155, 116293. https://doi.org/10.1016/j.bone.2021. 116293
13. Scollo, P., Snead, M. P., Richards, A. J., Pollitt, R., & Devile, C. (2018). Bilateral giant retinal tears in osteogenesis imperfecta. BMC Medical Genetics, 19, 8. https://doi.org/10.1186/s12881-018-0521-0
14. Pokidysheva, E., Tufa, S., Bresee, C., Brigande, J. V., & Bachinger, H. P. (2013). Prolyl 3-hydrox- ylase-1 null mice exhibit hearing impairment and abnormal morphology of the middle ear bone joints. Matrix Biology, 32, 39-44. https://doi.org/10.1016/j.matbio.2012.11.006
15. Chen, Z., Liu, G., Hossain, A., Danilova, I. G., Bolkov, M. A., et al. (2019). A co-expression net- work for differentially expressed genes in bladder cancer and a risk score model for predicting sur- vival. Hereditas, 156, 24. https://doi.org/10.1186/s41065-019-0100-1
16. Li, Y., Chen, Y., Ma, Y., Nenkov, M., Haase, D., & Petersen, I. (2018). Collagen prolyl hydroxy- lase 3 has a tumor suppressive activity in human lung cancer. Experimental Cell Research, 363, 121-128. https://doi.org/10.1016/j.yexcr.2017.12.020
17. Xue, W., Sun, C., Yuan, H., Yang, X., Zhang, Q., et al. (2022). Establishment and analysis of an individualized emt-related gene signature for the prognosis of breast cancer in female patients. Dis- ease Markers, 2022, 1289445. https://doi.org/10.1155/2022/1289445
18. Zhou, P., Liu, Z., Hu, H., Lu, Y., Xiao, J., et al. (2022). Comprehensive analysis of senescence char- acteristics defines a novel prognostic signature to guide personalized treatment for clear cell renal cell carcinoma. Frontiers in Immunology, 13, 901671. https://doi.org/10.3389/fimmu.2022.901671
19. Consortium G. (2013). The genotype-tissue expression (gtex) project. Nature Genetics, 45, 580- 585.https://doi.org/10.1038/ng.2653
20. Nusinow, D. P., Szpyt, J., Ghandi, M., Rose, C. M., Mcdonald, E. R., et al. (2020). Quantitative proteomics of the cancer cell line encyclopedia. Cell, 180, 387-402. https://doi.org/10.1016/j.cell. 2019.12.023
21. Tomczak, K., Czerwinska, P., & Wiznerowicz, M. (2015). The cancer genome atlas (tcga): An immeasurable source of knowledge. Contemp Oncol (Pozn), 19, A68-A77. https://doi.org/10.5114/ wo.2014.47136
22. Goldman, M. J., Craft, B., Hastie, M., Repecka, K., Mcdade, F., et al. (2020). Visualizing and inter- preting cancer genomics data via the xena platform. Nature Biotechnology, 38, 675-678. https:// doi.org/10.1038/s41587-020-0546-8
23. Cerami, E., Gao, J., Dogrusoz, U., Gross, B. E., Sumer, S. O., et al. (2012). The cbio cancer genom- ics portal: An open platform for exploring multidimensional cancer genomics data. Cancer Discov- ery, 2, 401-404. https://doi.org/10.1158/2159-8290.CD-12-0095
24. Zeng, D., Li, M., Zhou, R., Zhang, J., Sun, H., et al. (2019). Tumor microenvironment characteri- zation in gastric cancer identifies prognostic and immunotherapeutically relevant gene signatures. Cancer Immunology Research, 7, 737-750. https://doi.org/10.1158/2326-6066.CIR-18-0436
25. Li, T., Fu, J., Zeng, Z., Cohen, D., Li, J., et al. (2020). Timer2.0 for analysis of tumor-infiltrating immune cells. Nucleic Acids Research, 48, W509-W514. https://doi.org/10.1093/nar/gkaa407
26. Yang, W., Soares, J., Greninger, P., Edelman, E. J., Lightfoot, H., et al. (2013). Genomics of drug sensitivity in cancer (gdsc): A resource for therapeutic biomarker discovery in cancer cells. Nucleic Acids Research, 41, D955-D961. https://doi.org/10.1093/nar/gks1111
27. Lv, W., Shi, L., Pan, J., & Wang, S. (2022). Comprehensive prognostic and immunological analysis of cct2 in pan-cancer. Frontiers in Oncology, 12, 986990. https://doi.org/10.3389/fonc.2022.986990
28. Chen, F., Fan, Y., Cao, P., Liu, B., Hou, J., et al. (2021). Pan-cancer analysis of the prognostic and immunological role of hsf1: A potential target for survival and immunotherapy. Oxidative Medicine and Cellular Longevity, 2021, 5551036. https://doi.org/10.1155/2021/5551036
29. Erbas, I. M., Ilgun, G. D., Manav, K. Z., Koc, A., Unuvar, T., et al. (2022). Clinical, genetic character- istics and treatment outcomes of children and adolescents with osteogenesis imperfecta: A two-center experience. Connective Tissue Research, 63, 349-358. https://doi.org/10.1080/03008207.2021.19328 53
30. de Souza, L. T., Nunes, R. R., de Azevedo, M. O., & Maria, F. T. (2021). A new case of osteogen- esis imperfecta type viii and retinal detachment. American Journal of Medical Genetics. Part A, 185, 238-241. https://doi.org/10.1002/ajmg.a.61934
31. Tang, C., Fang, M., Tan, G., Zhang, S., Yang, B., et al. (2022). Discovery of novel circulating immune complexes in lupus nephritis using immunoproteomics. Frontiers in Immunology, 13, 850015. https:// doi.org/10.3389/fimmu.2022.850015
32. Gawel, D. R., Lee, E. J., Li, X., Lilja, S., Matussek, A., et al. (2019). An algorithm-based meta-anal- ysis of genome- and proteome-wide data identifies a combination of potential plasma biomarkers for colorectal cancer. Science and Reports, 9, 15575. https://doi.org/10.1038/s41598-019-51999-9
33. Xiong, C., Wang, G., & Bai, D. (2020). A novel prognostic models for identifying the risk of hepato- cellular carcinoma based on epithelial-mesenchymal transition-associated genes. Bioengineered, 11, 1034-1046. https://doi.org/10.1080/21655979.2020.1822715
34. Zhang, Y., Li, C. Y., Pan, M., Li, J. Y., Ge, W., et al. (2021). Exploration of the key proteins of high- grade intraepithelial neoplasia to adenocarcinoma sequence using in-depth quantitative proteomics analysis. J Oncol, 2021, 5538756. https://doi.org/10.1155/2021/5538756
35. He, W., Lin, S., Guo, Y., Wu, Y., Zhang, L. L., et al. (2022). Targeted demethylation at znf154 promo- tor upregulates znf154 expression and inhibits the proliferation and migration of esophageal squamous carcinoma cells. Oncogene. https://doi.org/10.1038/s41388-022-02366-y
6. Avram, E. G., Moatar, I. A., Miok, V., Baderca, F., Samoila, C., et al. (2022). Gene network analysis of the transcriptome impact of methylated micrornas on oral squamous cell carcinoma. Advances in Clinical and Experimental Medicine. https://doi.org/10.17219/acem/151911
37. Xu, Q., Lan, X., Lin, H., Xi, Q., & Wang, M., et al. (2022). Tumor microenvironment-regulating nano- medicine design to fight multi-drug resistant tumors, Wiley Interdisciplinary Reviews Nanomedicine and Nanobiotechnology, e1842. https://doi.org/10.1002/wnan.1842.
38. Luo, J., Xie, Y., Zheng, Y., Wang, C., Qi, F., et al. (2020). Comprehensive insights on pivotal prog- nostic signature involved in clear cell renal cell carcinoma microenvironment using the estimate algo- rithm. Cancer Medicine, 9, 4310-4323. https://doi.org/10.1002/cam4.2983
39. Yoshihara, K., Shahmoradgoli, M., Martinez, E., Vegesna, R., Kim, H., et al. (2013). Inferring tumour purity and stromal and immune cell admixture from expression data. Nature Communications, 4, 2612. https://doi.org/10.1038/ncomms3612
40. . Signorini, L., Delbue, S., Ferrante, P., & Bregni, M. (2016). Review on the immunotherapy strategies against metastatic colorectal carcinoma. Immunotherapy-Uk, 8, 1245-1261. https://doi.org/10.2217/ imt-2016-0045
Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.