ELSEVIER

Contents lists available at ScienceDirect

Life Sciences

journal homepage: www.elsevier.com/locate/lifescie

2

LIFESCIENCES

HOXC11 functions as a novel oncogene in human colon adenocarcinoma and kidney renal clear cell carcinoma

Check for updates

Yuanbo Cuiª,*,1, Chunyan Zhangb,1, Yaping Wangª,1, Shanshan Maª, Wei Cao, Fangxia Guana,*

ª School of Life Sciences, Zhengzhou University, Zhengzhou 450001, China

b Department of Clinical Laboratory, Zhengzhou Central Hospital Affiliated to Zhengzhou University, Zhengzhou 450007, China

” Department of Translational Medicine Center, Zhengzhou Central Hospital Affiliated to Zhengzhou University, Zhengzhou 450007, China

ARTICLE INFO

Keywords:

Homeobox-C 11 Pan-cancer

The Cancer Genome Atlas Prognosis PPARY

ABSTRACT

Aims: Accumulating evidence has confirmed the involvement of the homeobox (HOX) gene family in carcino- genesis. HOXC11, belongs to the homeobox-C (HOXC) gene cluster, has been reported to play important roles in the development of several cancers. However, its expression and clinical value in pan-cancer remain elusive. Materials and methods: Bioinformatics analysis, CCK-8 assay, Flow cytometry and Western blot were used to analyze gene expression and patient survival, cell proliferation, cell apoptosis and protein level, respectively. Key findings: In this study, we comprehensively analyzed the expression profile and prognostic value of HOXC11 in human pan-cancer using online The Cancer Genome Atlas (TCGA) databases. HOXC11 was widely up-regu- lated in tumor tissues when compared with the normal tissues in pan-cancer across nine cancer types. In ad- dition, high mRNA level of HOXC11 predicted poor overall survival (OS) of patients with adrenocortical car- cinoma (ACC), colon adenocarcinoma (COAD), kidney renal clear cell carcinoma (KIRC), mesothelioma (MESO) and pancreatic adenocarcinoma (PAAD), respectively. By comparative analysis, we found that HOXC11 was up- regulated and closely correlated patient OS in COAD and KIRC. Functionally, down-regulation of HOXC11 in- hibited cell proliferation but promoted apoptosis of COAD and KIRC in vitro. Mechanistically, HOXC11 promoted cell proliferation of COAD and KIRC might by inactivating the peroxisome proliferator-activated receptor gamma (PPARY) signaling pathway.

Significance: Our findings suggest that HOXC11 may act as a tumor driving gene in COAD and KIRC.

1. Introduction

Homeobox (HOX) gene encodes a series of transcription factors that play critical roles in human pathological and physiological processes. The HOX family consists of four gene clusters termed HOXA, HOXB, HOXC and HOXD, which located on Chromosome 7, Chromosome 17, Chromosome 12 and Chromosome 2, respectively. HOX genes control numerous cancer related biological behaviors, such as cell differentia- tion, proliferation, metastasis and apoptosis [1-3]. HOXC gene cluster contains nine members, namely, HOXC4, HOXC5, HOXC6, HOXC8, HOXC9, HOXC10, HOXC11, HOXC12 and HOXC13.

Previous studies showed that dysregulated HOXC genes contributed to the initiation and development of various types of cancer and had certain clinical values. For example, up-regulation of HOXC5 inhibits

HeLa cell growth both in vitro and in vivo by decreasing Human telo- merase reverse transcriptase (hTERT) expression [4]. HOXC6 is over- expressed in glioma, cervical cancer and non-small cell lung cancer, knockdown of HOXC6 inhibits cancer progression via specific signaling pathways [5-7]. HOXC8-mediated up-regulation of matrix Gla protein (MGP) promotes cell proliferation, invasion and epithelial-mesench- ymal transition (EMT) in triple-negative breast cancer [8]. High ex- pression of HOXC9 is associated with worse patient overall survival in colorectal cancer [9]. Increased HOXC10 facilitates cell proliferation and predicts poor prognosis in glioblastoma [10]. These studies imply that the HOXC genes may participate in human tumorigenesis and be potentially useful for the prognostic evaluation of cancer patients.

HOXC11, a member of the HOXC gene cluster, was also previously reported to be involved in the development of several types of cancer.

* Corresponding authors. E-mail addresses: cuiyuanbo18@126.com (Y. Cui), fxguan@126.com (F. Guan).

1 These authors contributed equally to this work.

Fig. 1. The transcriptional level of HOXC11 in human pan-cancer. HOXC11 is dysregulated in 22 types of cancer (GEPIA). Red bar graph refers to tumor tissues (T) and blue bar graph refers to normal tissues (N). * p < 0.05. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

-

-

Expression-log_(TPM + 1)

+

2

-

o

(num(T)=77; num(N)=128)

ACC

BLCA (num(T)=404; num(N)=28)

BRCA (num(T)=1085; num(N)=291)

CESC (num(T)=306; num(N)=13)

COAD (num(T)=275; num(N)=349)

DLBC (num(T)=47; num(N)=337)

ESCA (num(T)=182; num(N)=286)

-

.

Expression-log,(TPM +1)

2

-

0

GBM (num(T)=163; num(N)=207)

HNSC

(num(T)=519; num(N)=44)

KICH (num(T)=66; num(N)=53)

KIRC (num(T)=523; num(N)=100)

KIRP (num(T)=286; num(N)=60)

LGG (num(T)=518; num(N)=207)

LIHC (num(T)=369; num(N)=160)

6

Expression -logy(TPM +1)

4

2

-

0

LUAD (num(T)=483; num(N)=347)

LUSC (num(T)=486; num(N)=338)

PAAD (num(T)=179; num(N)=171)

READ (num(T)=92; num(N)=318)

SKCM (num(T)=461; num(N)=558)

STAD (num(T)=408, num(N)=211)

TGCT (num(T)=137, num(N)=165)

UCS (num(T)=57, num(N)=78)

HOXC11 was found to promote cell proliferation in both clear cell renal cell carcinoma and non-small cell lung cancer [11,12]. But its expres- sion, clinical values and function in pan-cancer are still largely un- known. In this study, we report that HOXC11 is a potential therapeutic target and (or) prognostic biomarker in multiple types of cancer based on the comprehensive bioinformatics analysis using online TCGA da- tabases. Our results highlight the oncogenic role of HOXC11 in COAD and KIRC.

2. Materials and methods

2.1. Bioinformatics analysis

Three online TCGA analysis databases including GEPIA [13] (http://gepia.cancer-pku.cn/index.html), UALCAN [14] (http://ualcan. path.uab.edu/) and starBase v3.0 [15] (http://starbase.sysu.edu.cn/ index.php) were used to observe the transcriptional levels of HOXC11 in human pan-cancer. Three online TCGA analysis databases including GEPIA, starBase v3.0 and LinkedOmics [16] (http://www.linkedomics. org) were utilized to analyze the relationship between gene expression

and patient overall survival (OS). The co-expressed genes of HOXC11 in COAD and KIRC were obtained from LinkedOmics. Gene Set Enrich- ment Analysis of HOXC11-related genes in COAD and KIRC based on the Kyoto Encyclopedia of Genes and Genomes (KEGG) was analyzed by the WebGestalt [17] (http://www.webgestalt.org/) tool.

2.2. Cell culture and siRNA transfection

The human COAD cell line (SW480) and KIRC cell line (Caki-1) were all maintained in RPMI 1640 medium containing 10% FBS (Gibco, USA) and cultured in a cell incubator at 37 ℃ with 5% CO2. The specific siRNAs for HOXC11 (si-HOXC11) (5’-CGAUGUUUAACUCGGUCA ACC-3’; 5’-UUGACCGAGUUAAACAUCGUU-3’) and PPARY (si-PPARY) ( 5’-CAGUGGUUGCAGAUUACAAGU-3’; 5’-UUGUAAUCUGCAACCACU GGA-3’) as well as the negative control siRNAs (si-NC) were transfected into cell using the INTERFERin regent (Polyplus Transfection, France).

2.3. CCK-8 assay

Cell counting kit-8 (CCK-8) (Solarbio, China) was used to detect cell

Fig. 2. The prognostic values of HOXC11 in pan-cancer were analyzed from GEPIA database. Six types of cancer refers to ACC, CESC, COAD, KIRC, MESO and PAAD.

Overall Survival

Overall Survival

Overall Survival

1.0

Low HOXC11 TPM

1.0

Low HOXC11 TPM

1.0

High HOXC11 TPM

High HOXC11 TPM

Low HOXC11 TPM

High HOXC11 TPM

Logrank p=0

Logrank p=6.48-08

n(high)=37

Logrank p=0.021

0.8

n(high)=4734

P=0

n(low)=4740

0.8

p=6.4E-8

n(low)=35

0.8

n(high)=146

Percent survival

n(low)=146

Percent survival

Percent survival

0.6

0.6

0.6

0.4

0.4

0.4

0.2

0.2

0.2

P=0.021

0.0

Pan-cancer

0.0

ACC

0.0

CESC

0

100

200

300

0

50

100

150

0

50

100

150

200

Months

Months

Months

Overall Survival

Overall Survival

Overall Survival

1.0

Low HOXC11 TPM

1.0

High HOXC11 TPM

Low HOXC11 TPM

1.0

High HOXC11 TPM

Low HOXC11 TPM

Logrank p=0.021

0.8

n(high)=130

Logrank p=0.02

High HOXC11 TPM

n(high)=256

Logrank p=0.0082

n(low)=135

0.8

n(low)=254

0.8

n(high)=41

Percent survival

Percent survival

n(low)=40

Percent survival

p=0.0082

0.6

0.6

0.6

0.4

0.4

0.4

0.2

p= 0.021

0.2

P=0.02

0.2

0.0

COAD

0.0

KIRC

0.0

MESO

0

50

100

150

0

50

100

150

0

20

40

60

80

Months

Months

Months

Overall Survival

Overall Survival

Overall Survival

1.0

Low HOXC11 TPM

1.0

High HOXC11 TPM

Low HOXC11 TPM

1.0

High HOXC11 TPM

Low HOXC11 TPM

Logrank p=0.0043

Logrank p=0.012

High HOXC11

0.8

n(high)=89

n(low)=89

0.8

n(high)=192

Logrank p=0.018 n(high)=703

Percent survival

P

0.0043

p=0.012

n(low)=192

0.8

Percent survival

P=0.018

n(low)=706

Percent survival

0.6

0.6

0.6

0.4

0.4

0.4

0.2

0.2

0.2

0.0

PAAD

0.0

STAD

0.0

Six types of cancer

0

20

40

60

80

0

20

40

60

80

100

120

0

50

100

150

200

Months

Months

Months

proliferative ability. Briefly, transfected cells were seeded into 96-well plates, 10 µL of CCK-8 solution was added into each well. After in- cubation at 37 °℃ for 4 h, the absorbance of each well at 450 nm was detected by a SpectraMax microplate reader (Molecular Devices, USA).

2.4. Flow cytometry

The procedures were following the user’s instructions of the Annexin V-FITC/PI kit (7sea biotech, China). In brief, after transfection, cells were collected in a clean tube and resuspended in binding buffer. Annexin V-FITC and PI were added in turn to each tube. Then, cell apoptosis was examined by Flow cytometry.

2.5. Western blot

RIPA buffer containing protease inhibitor was utilized to extract total protein of each sample. After denatured at 100 ℃ for 10 min, total protein of each sample was used for SDS-PAGE, PVDF membrane transfer and blocking in 5% skim milk. Then, the membranes were incubated with diluted primary antibodies (GAPDH, Abcam, USA; HOXC11, Santa, USA; PPARY, Santa, USA) and secondary antibody, respectively. After incubated with ECL regent for seconds, the im- munoreactive protein bands were detected by the imaging system (Bio- Rad, USA).

HOXC11 vs overall_survival (0 Median) (1 > Median) Median=1.4227

HOXC11 vs overall_survival (0 Median) (1 > Median) Median=0 Sample Size:(N=481)

Sample Size:(N=79)

1.0

1.0

:

HỘINGHỊ :

0.8

0.8

p=1.0E-9

Survival Probability

0.6

p=7.6E-8

Survival Probability

0.6

0.4

0.4

0.2

0.2

0.0

ACC

0.0

CHOL

0

1000

2000

3000

4000

0

1000

2000

3000

4000

5000

6000

Time (Days)

Time (Days)

HOXC11 vs overall_survival (0 Median) (1 > Median) Median=4.35735

HOXC11 vs overall_survival (0 Median) (1 > Median) Median=4.45 Sample Size:(N=518)

Sample Size:(N=144)

1.0

1.0

HOẠCH

:

:

0.8

p=0.032

0.8

Survival Probability

Survival Probability

0.6

0.6

0.4

0.4

0.2

0.2

P=1.7E-4

0.0

GBM

0.0

KIRC

0

500

1000

1500

2000

2500

0

1000

2000

3000

4000

Time (Days)

Time (Days)

HOXC11 vs overall_survival (0 Median) (1 > Median) Median=4.56285 Sample Size:(N=172)

HOXC11 vs overall_survival (0 Median) (1 > Median) Median=0 Sample Size:(N=118)

1.0

HOẠCH

:

1.0

:

0.8

0.8

Survival Probability

Survival Probability

0.6

p=0.017

0.6

p=0.016

0.4

0.4

0.2

0.2

0.0

PAAD

0.0

THYM

0

500

1000

1500

2000

2500

0

1000

2000

3000

4000

Time (Days)

Time (Days)

HOXC11 vs overall_survival (0 Median) (1 > Median) Median=1.2008 Sample Size:(N=367)

Fig. 3. The prognostic values of HOXC11 in pan-cancer were analyzed from LinkedOmics database. High expression of HOXC11 predicts poor OS of ACC, CHOL, COAD, GBM, KIRC, MESO and PAAD, but well OS of THYM.

1.0

HORCH : :

0.8

Survival Probability

0.6

0.4

0.2

p=0.031

0.0

COAD

0

1000

2000

3000

4000

Time (Days)

HOXC11 vs overall_survival (0 Median) (1 > Median) Median=1.37075 Sample Size:(N=84)

1.0

HOXC !! :

p=0.0047

0.8

Survival Probability

0.6

0.4

0.2

0.0

MESO

0

500

1000

1500

2000

2500

Time (Days)

2.6. Statistical analysis

For the prognosis analysis of gene, patients with cancer were di- vided into low expression group and high expression group based on the median mRNA level of interest gene in each database, log-rank p < 0.05 was regarded to be statistically significant. Data of in vitro experiments are expressed as mean + SD. SPSS 19.0 was used for statistical analysis (student’s t-test and one-way ANOVA). p value < 0.05 was considered to be statistically significant.

3. Results

3.1. HOXC11 is highly expressed in human pan-cancer

Three online TCGA analysis databases were used to examine the expression profile of HOXC11 in human pan-cancer. The GEPIA data- base contains information of 33 types of cancer. In GEPIA, HOXC11 was found to be up-regulated in 19 (ACC, BLCA, BRCA, CESC, COAD, DLBC, ESCA, GBM, HNSC, KIRC, KIRP, LGG, LIHC, LUAD, LUSC, PAAD, READ, STAD and UCS) but down-regulated in 3 types of cancer (KICH, SKCM

Fig. 4. The prognostic values of HOXC11 in pan-cancer were analyzed from starBase v3.0 database. High expression of HOXC11 predicts poor OS of ACC, COAD, KIRC, MESO and PAAD.

group

group

1.00

Overall Survival for HOXC11 in ACC Cancer

Log-Rank p=6.1c-08

1.00

Overall Survival for HOXC11 in COAD Cancer

Low Num=40

low

Log-Rank p=0.019

Low Num=224

low

High Num=39

Hazard Ratio=11.31

high

High Num=223

Hazard Ratio=1.61

high

0.75

(low, 1)

0.75

(low,1)

Percent Survival

(high,1)

(high, 1)

p=6.1E-8

Percent Survival

0.50

0.50

0.25

0.25

p=0.019

ACC

COAD

0.00

0.00

0

50

100

150

0

50

100

150

Time(months)

Time(months)

group

group

1.00

Overall Survival for HOXC11 in KIRC Cancer

Log-Rank p=0.0056

1.00

Overall Survival for HOXC11 in MESO Cancer

Low Num=259 High Num=258

low

Log-Rank p=0.019

Low Num=43

low

Hazard Ratio=1.53

high

High Num=42

Hazard Ratio=1.77

high

0.75

(low,1)

0.75

(low,1)

Percent Survival

(high,1)

Percent Survival

p=0.019

(high, 1)

0.50

0.50

0.25

0.25

p=0.0056

KIRC

MESO

0.00

0.00

0

50

100

150

0

25

50

75

Time(months)

Time(months)

group

1.00

Overall Survival for HOXC11 in PAAD Cancer

Log-Rank p=0.01 Low Num=89 High Num=89 Hazard Ratio=1.71

high

low

0.75

(high,1)

Percent Survival

p=0.01

+ (low,1)

0.50

0.25

PAAD

0.00

0

25

50

75

Time(months)

and TGCT) as compared with the normal tissues (Fig. 1). The UALCAN database contains information of 24 types of cancer. In UALCAN, HOXC11 was found to be up-regulated in 14 (BLCA, BRCA, CESC, COAD, ESCA, HNSC, KIRC, KIRP, LUAD, LUSC, PAAD, READ, SARC and STAD) but down-regulated in 2 types of cancer (KICH and UCEC) as compared with the normal tissues (Supplementary Fig. 1). The starBase v3.0 database contains information of 32 types of cancer. In this

database, HOXC11 was found to be up-regulated in 10 types of cancer (BRCA, COAD, ESCA, HNSC, KIRC, KIRP, LIHC, LUAD, LUSC and STAD) but down-regulated in KICH as compared with the normal tissues (Supplementary Fig. 2). Then, we compared these data (Fig. 1, Sup- plementary Figs. 1 and 2) and found that HOXC11 was evidently dys- regulated in 10 types of cancer (BRCA, COAD, ESCA, KIRC, KICH, KIRP, LUAD, LUSC, HNSC and STAD) in the three databases (Fig. 5A).

Fig. 5. Data comparison was shown as Venn diagram. (A) Data comparison of Fig. 1, Supplementary Figs. 1 and 2. (B) Data comparison of Figs. 2, 3 and 4. (C) Data comparison of Fig. 1, Supplementary Figs. 1, 2, Figs. 2 and 3.

A

Figure1

Suppl Figure1

B

Figure2

Figure3

7

4

2

2

0

3

10

5

1

0

0

0

0

0

Suppl Figure2 V

Figure4 V

HOXC11 is up regulated in BRCA, COAD, KIRC, LUAD, KIRP, ESCA, STAD, HNSC and LUSC, but down-regulated in KICH

HOXC11 up-regulation predicts poor OS of ACC, COAD, KIRC, MESO and PAAD

C

Figure1

Suppl Figure1

2

2

0

5

0

0

0

1

0

0

8

0

Suppl Figure2

1

1

2

0

1

1

0

0

0

0

0

0

0

0

0

Figure3

2

1

0

1

Figure2

V

HOXC11 is up-regulated and predicts poor OS of COAD and KIRC

3.2. High expression of HOXC11 predicts poor overall survival in pan- cancer.

The relationship between HOXC11 expression and patient OS was analyzed by three independent online TCGA tools. First, we observed that high HOXC11 expression was an unfavorable factor for patient OS (p = 0) in pan-cancer (33 types of cancer, GEPIA) (Fig. 2). Results from GEPIA also showed that high expression of HOXC11 was closely asso- ciated with worse OS in ACC (p = 6.4E-8), CESC (p = 0.021), COAD

(p = 0.021), KIRC (p = 0.02), MESO (p = 0.0082) and PAAD (p = 0.0043) but with well OS in STAD (p = 0.012) (Fig. 2). In addi- tion, high HOXC11 expression was also an unfavorable factor for pa- tient OS (p = 0.018) in these 6 types of cancer (Fig. 2). Results from LinkedOmics showed that high expression of HOXC11 was closely as- sociated with worse OS in ACC (p = 7.6E-8), CHOL (p = 1.0E-9), COAD (p = 0.031), GBM (p = 0.032), KIRC (p = 1.7E-4), MESO (p = 0.0047) and PAAD (p = 0.017) but with well OS in THYM (p = 0.016) (Fig. 3). Results from starBase v3.0 showed that high expression of HOXC11 predicted poor OS in ACC (p = 6.1E-8), COAD (p = 0.019), KIRC (p = 0.0056), MESO (p = 0.019) and PAAD (p = 0.01) (Fig. 4). Then, we compared these results (Figs. 2, 3 and 4) and found that high ex- pression of HOXC11 indicated poor OS in 5 types of cancer (ACC, COAD, KIRC, MESO and PAAD) in the three databases (Fig. 5B).

3.3. HOXC11 functions as an oncogenic gene to promote cell proliferation of COAD and KIRC.

Next, the five datasets of Fig. 1, Supplementary Figs. 1, 2, Figs. 2 and 3 were compared. We found that HOXC11 was not only dramati- cally up-regulated but also closely associated with worse patient OS in COAD and KIRC (Fig. 5C). Therefore, we speculated that HOXC11 may work as an oncogene in these two cancer types. To confirm this, we examined the effects of HOXC11 knockdown on cell proliferation and apoptosis of COAD and KIRC in vitro. The results showed that down- regulation of HOXC11 could significantly suppress cell proliferation (Fig. 6A, C) but promote apoptosis (Fig. 6B, D) in SW480 and Caki-1 cells. These results indicate that HOXC11 functions as an oncogenic gene in both COAD and KIRC.

3.4. HOXC11 promotes cell proliferation of COAD and KIRC via down- regulating PPARY.

Next, we intend to investigate the potential mechanism of HOXC11 in the development of COAD and KIRC. To this end, we obtained the co- expressed genes of HOXC11 in COAD and KIRC from the LinkedOmics. The overall closely co-expressed genes of HOXC11 in COAD (Fig. 7A) and KIRC (Fig. 7D) were shown as volcano plots. The top-50 positively and negatively co-expressed genes of HOXC11 in COAD were shown as

Fig. 6. The effects of HOXC11 knockdown on cell proliferation and apoptosis in COAD and KIRC in vitro. (A, C) CCK-8 assay was used to detect the effect of HOXC11 knockdown on cell proliferation of SW480 (A) and Caki-1 (C) cells, *p < 0.05. (B, D) Flow cytometry was used to detect the effect of HOXC11 knockdown on cell apoptosis of SW480 (B) and Caki-1 (D) cells *p < 0.05.

A

2.0

SW480

B

si-NC

si-HOXC11

15

OD value (450 nm)

+ si-NC

0

si-NC

1.6

Q1-UL(0.08%)

01-UR(0.40%)

0

Q1-UL(0.52%)

01-UR(7.68%)

si-HOXC11

si-HOXC11

105

105

12

*

1.2

105

105

Apoptosis (%)

9

*

PIPE-A

PIPE-A

V

0.8

10

10ª

6

10ª

102

0.4

10

10’

3

0.0

01-LR(0.35%)

01-LR(0.48%)

0

Q1-LL(99.17%)

0

Q1-LL(91.32%)

0

1

Days

2

3

0

10

102

103

104

10°

10º

107

0

10

10ª

10

104

105

10

0

Annexin-V FITC-A

Annexin-V FITC-A

SW480

C

2.0

Caki-1

D

si-NC

si-HOXC11

20

si-NC

OD value (450 nm)

1.6

+ si-NC

0

Q1-UL(0.04%)

01-UR(0.68%)

Q1-UL(0.34%)

01-UR(9.87%)

si-HOXC11

si-HOXC11

100

10º

16-

*

1.2

105

105

12

*

PIPE-A

102 104

Apoptosis (%)

.:

PIPE-A

104

0.8

103

8

0.4

102

102

101

101

4

0.0

Q1-LL(99.02%)

Q1-LL(88.98%)

01-LR(0.81%)

0

1

2

0

01-LR(0.26%)

0

Days

3

0

10

10-

103

104

105

10°

10’

0

101

102

10’

10ª

105

10

0

Annexin-V FITC-A

Annexin-V FITC-A

Caki-1

Fig. 7. The co-expressed genes of HOXC11 in COAD. The overall closely co-expressed genes of HOXC11 in COAD (A) and KIRC (D) were shown as volcano plots. The positively (B) and negatively (C) co-expressed top-50 genes of HOXC11 in COAD were shown as heat maps. The positively (E) and negatively (F) co-expressed top-50 genes of HOXC11 in KIRC were shown as heat maps.

A

B

C

HOXC11 Association Result

8

Negative

Negative

-log10(pvalue)

Positive Top-50

8

Top-50

Positive

Z-Score Group

Z-Score Group

Os

×3

3

3

3

1

0

2

1

0

1

2

0

0

1

-1

-1

0

&

3

-1

<3

-1

0

-1.0

-0.5

0.0

0.5

1.0

1.5

2.0

-

Pearson Correlation Coefficient (Pearson test)

COAD

D

E

F

HOXC11 Association Result

8

Positive

80

Negative

Negative

Top-50

-log10(pvalue)

Positive

Top-50

Z-Score Group

Z-Score Group

0p

×3

4

>3

1

2

1

4

2

0

0

-1

-2

0

-1

0

-2

2

<3

4

M

<3

4

0

-1.5

-1.0

-0.5

0.0

0.5

1.0

1.5

2.0

Pearson Correlation Coefficient (Pearson test)

KIRC

heat maps (Fig. 7B-C), respectively. The top-50 positively and nega- tively co-expressed genes of HOXC11 in KIRC were shown as heat maps (Fig. 7E-F), respectively. GSEA, which based on the KEGG database, was used to explore the potential signaling pathways mediated by HOXC11 in both COAD and KIRC. The results displayed that HOXC11 were mainly involved in the regulation of fatty acid degradation (p = 0, FDR = 0, NES = - 2.25), glycolysis (p= 0, FDR = 0, NES =- 2.12) and PPAR (p = 0, FDR = 0.0009, NES = - 2.03) pathways in COAD (Fig. 8A). In KIRC, HOXC11 was mainly related to the fatty acid me- tabolism (p = 0, FDR = 0.001, NES = - 2.06), PPAR (p = 0, FDR = 0.009, NES = - 1.87) and mTOR (p = 0, FDR = 0.009, NES = - 1.87) pathways (Fig. 8B).

Given that HOXC11 may be involved in the same PPAR pathway in both COAD and KIRC. We next focused on observing the effect of PPAR signaling on HOXC11-mediated cell proliferation of COAD and KIRC. Result of Western blot showed that the protein level of PPARY was elevated in SW480 and Caki-1 cells after knockdown of HOXC11 (Fig. 9A). The effects of HOXC11 knockdown on cell proliferation, apoptosis and HOXC11 level were partially rescued by down-regulation of PPARY (Fig. 9A-E) (p < 0.05). Collectively, our data suggests that HOXC11 may fulfill its oncogenic role in both COAD and KIRC through inactivating the PPARy signaling.

4. Discussion

HOXC11, located on chromosome 12q13.13, was first reported to act as an essential gene for metanephric kidney induction [17]. In re- cent years, it has been shown that HOXC11 played a role in the

development and prognosis in several human cancers. In clear cell renal cell carcinoma, over-expression of HOXC11 promotes cancer cell pro- liferation and negatively correlates with patient prognosis [11]. In non- small cell lung cancer, HOXC11 cooperates with microRNA-1197 and regulates cell proliferation and migration [12]. But the expression profile and prognostic value of HOXC11 in pan-cancer remains largely unclear. Herein, we for the first time report that HOXC11 may be a prognostic biomarker for multiple cancer types and functions as a tumor driving gene in COAD and KIRC.

As far as we know, only a few studies have investigated the ex- pression and clinical significance of HOXC11 in several types of cancer. In breast cancer, HOXC11 was expressed at a lower level in cancer tissues than adjacent normal tissues [18], HOXC11 interacted with steroid receptor co-activator SRC-1 and promoted resistance to endo- crine therapy [19]. High expression of HOXC11 predicted poor patient prognosis in breast cancer and cervical cancer [19,20]. In this study, we observed the expression landscape and prognostic value of HOXC11 in pan-cancer across thirty-three cancer types, and found that HOXC11 was abnormally expressed in ten types of cancer and correlated with patient OS in five cancer types. Thus, further studies regarding the expression of HOXC11 in body fluid and its correlation with patient clinical characteristics may help in developing HOXC11 as a clinical biomarker for these malignancies.

Considering that HOXC11 was significantly up-regulated and clo- sely correlated with patient OS in both COAD and KIRC based on our bioinformatics analysis. We focused on explore the function and pos- sible regulatory pathway of HOXC11 in COAD and KIRC. By in vitro functional experiments, we found that down-regulation of HOXC11

Fig. 8. GSEA analysis of HOXC11 co-expressed genes in COAD and KIRC. (A) Three pathways (fatty acid degradation, glycolysis and PPAR) were closely correlated with HOXC11 in COAD. (B) Three pathways (fatty acid metabolism, PPAR and mTOR) were closely correlated with HOXC11 in KIRC. FDR refers to false discovery rate and NES refers to normalized enrichment score.

A

Enrichment plot: Fatty acid degradation

Enrichment plot: Glycolysis / Gluconeogenesis

Enrichment plot: PPAR signaling pathway

=

p=0 FDR = 0

W

2

P=0

=

p= 0

3

Enrichment Score

FDR = 0

FDR = 0.0009

NES =- 2.25

Enichmant Score

NES =- 2.12

Enrichment Score

NES =- 2.03

=

3

=

:

=

COAD

=

COAD

=

Farhad Ist metric

Fatty acid degradation pathway

Ramund Ist wine

COAD

Glycolysis pathway

PPAR signaling pathway

:

2

2

6

2100

“Rank in Ordered Dataset

8033

.

2000

“flank in Ordered Dataset

8033

.

2000

“Ranks in Ordered Dataset

8033

B

Enrichment plot: Fatty acid metabolism

Enrichment plot: PPAR signaling pathway

Enrichment plot: mTOR signaling pathway

P=0

P=0

8

P=0

:

FDR = 0.001

:

S

Enrichment Score

FDR = 0.009

4

FDR = 0.009

NES =- 2.06

Enrichtung Score

NES =- 1.87

Enrichment 365.

NES =- 1.87

=

:

:

:

=

=

KIRC

=

KIRC

=

Ranked Ist matric

KIRC

Fatty acid metabolism pathway Rank il Ordered Dataset

PPAR signaling pathway - Rank l’ Ordered Dataset

mTOR signaling pathway

*

*

.

.

Rank in Ordered Dataset

enhanced cancer cell apoptosis but inhibited cell proliferation. In pre- vious studies, HOXC11 was found to promote cell proliferation in both KIRC and non-small cell lung cancer [11,12]. Therefore, our study confirmed the role of HOXC11 in KIRC growth and provided first-hand evidence for the oncogenic function of HOXC11 in COAD.

Through GSEA enrichment analysis, HOXC11 was found to be mainly involved in the negative regulation of the same PPAR signaling pathway in both COAD and KIRC. PPARY, one member of the PPAR gene family, was reported to be the master regulatory factor in adipo- genesis [21]. Recent studies highlight its important role in human tu- morigenesis [22,23]. In COAD, PPARy was a necessary pathway in- volved in the antitumor activity of drugs such as embelin [24]. In KIRC, PPARy acted as a crucial mediator in the cancer progression [25]. We then speculated that HOXC11 contributed to cell proliferation of COAD and KIRC via regulating PPARy. To test this hypothesis, we performed a series of rescue experiments in vitro. Our data displayed that HOXC11 negatively regulated the expression of PPARY in both SW480 and Caki- 1 cells. The effects of HOXC11 knockdown on cell proliferation and apoptosis were rescued in part by down-regulation of PPARY in both SW480 and Caki-1 cells. These data suggests that HOXC11 may con- tribute to development of COAD and KIRC via PPARy signaling.

In summary, this study comprehensively analyzed the expression profile and prognostic values of HOXC11 in pan-cancer based on

bioinformatics analysis, and investigated the function of HOXC11 in COAD and KIRC through in vitro experiments. Our data provided some clues for the development of HOXC11 as a clinical biomarker for some cancer types, such as ACC, COAD, KIRC, MESO and PAAD. Our findings suggest that HOXC11 may function as a tumor driving gene in both COAD and KIRC. The specific role and the precise mechanism of HOXC11 in the initiation and progression of these two malignancies need to be further extensively explored.

Supplementary data to this article can be found online at https:// doi.org/10.1016/j.lfs.2019.117230.

Authors’ contributions

YBC conceived this study and wrote the manuscript. CYZ, YPW, SSM and WC participated in the in vitro experiments, draft writing and data analysis. FXG provided assistance for revising the manuscript. All au- thors read the manuscript and approved for publication.

Funding

This work was funded by the Key Scientific Research Projects of Institutions of Higher Learning in Henan Province (20A310018).

Fig. 9. Down-regulation of PPARy abolishes the effects of HOXC11 knockdown on cell proliferation and apoptosis in COAD and KIRC. (A) Protein levels of HOXC11 and PPARY in each group of SW480 and Caki-1 cells were examined by Western blot. (B, D) Cell proliferation of SW480 (B) and Caki-1 (D) was detected by CCK-8 assay, *p < 0.05. (C, E) Cell apoptosis of SW480 (C) and Caki-1 (E) was detected by flow cytometry, *p < 0.05.

A

SW480

Caki-1

1

2

3

1

2

3

HOXC11

1.0

0.51

0.67

1.0

0.41

0.77

1.si-NC

PPARY

1.0

2.72

1.72

1.0

2.50

1.08

2. si-HOXC11

GAPDH

3. si-HOXC11+si-PPARY

B

2.5

SW480

C

OD value (450 nm)

si-NC

si-HOXC11

si-HOXC11+

25

2.0

Si-NC

si-PPARY

SI-NC

si-HOXC11

si-HOXC11

3

si-HOXC11+si-PPARY

Q1-UL(0.35%)

01-UR(1.51%)

0

Q1-UL(0.22%)

01-UR(5.15%)

0

Q1-UL(0.21%)

01-UR(5.22%)

20

si-HOXC11+PPARY

1.5

8

105

8

Apoptosis (%)

6

105

3

15

1.0

PIPE-A

0

PIPE-A

0

O

109

109

PIPE-A

*

10ª

10

0.5

100

102

3

0

10”

0

5

0.0

:

0

Q1-LL(98.14%)

21-LR(0.00%)

0

Q1-LL(89.98%)

01-LR(4.66%)

0

Q1-LL(93.78%)

01-LR(0.78%)

0

1

Days

2

3

0

101

102

103

104

10%

10°

107

0

10

102

109

104

10%

10°

10

0

10

102

10

104 10

10

105

10°

107

0

Annexin-V FITC-A

Annexin-V FITC-A

Annexin-V FITC-A

SW480

D

2.5

Caki-1

E

si-HOXC11+

25

si-NC

OD value (450 nm)

si-NC

si-HOXC11

2.0

si-NC

si-PPARY

si-HOXC11

si-HOXC11

“2

si-HOXC11+si-PPAR

Q1-UL(0.19%)

01-UR(0.26%)

3

Q1-UL(0.05%)

01-UR(6.39%)

3

Q1-UL(0.80%)

01-UR(4.22%)

20

si-HOXC11+PPARY

8

10°

Apoptosis (%)

1.5

8

105

2

15

1.0

E

*

PIPE-A

PIPE-A

1.04

103

PIPE-A

10ª

10ª

10-

0.5

D

102

E

0

101

2

5

0.0

0

Q1-LL(99.50%)

01-LR(0.04%)

0

Q1-LL(90.68%)

01-LR(2.88%)

0

Q1-LL(94.84%)

01-LR(0.14%)

0

1

Days

2

3

0

10”

102

10ª

10ª

10%

10°

107

0

10”

102

10ª

104

105

10°

10’

0

101

102

10ª

10ª

105

10°

107

0

Annexin-V FITC-A

Annexin-V FITC-A

Annexin-V FITC-A

Caki-1

Acknowledgements

We thank all members for their assistance for our work.

Declaration of competing interest

No.

References

[1] Y. Wang, Y. Dang, J. Liu, X. Ouyang, The function of homeobox genes and LncRNAs in cancer, Oncol. Lett. 12 (3) (2016) 1635-1641.

[2] S. Bhatlekar, J.Z. Fields, B.M. Boman, Role of HOX genes in stem cell differentiation and cancer, Stem Cells Int. 2018 (2018) 3569493.

[3] B. Li, Q. Huang, G.H. Wei, The role of HOX transcription factors in cancer predis- position and progression, Cancers (Basel) 11 (4) (2019) 528.

[4] T. Yan, W.F. Ooi, A. Qamra, A. Cheung, D. Ma, G.M. Sundaram, C. Xu, M. Xing, L. Poon, J. Wang, Y.P. Loh, J.H.J. Ho, J.J.Q. Ng, M.K. Ramlee, L. Aswad, S.G. Rozen, S. Ghosh, F.A. Bard, P. Sampath, V. Tergaonkar, J.O.J. Davies, J.R. Hughes, E. Goh, X. Bi, M.J. Fullwood, P. Tan, S. Li, HoxC5 and miR-615-3p target newly evolved genomic regions to repress hTERT and inhibit tumorigenesis, Nat. Commun. 9 (1) (2018) 100.

[5] T.F. Yan, M.J. Wu, B. Xiao, Q. Hu, Y.H. Fan, X.G. Zhu, Knockdown of HOXC6 in- hibits glioma cell proliferation and induces cell cycle arrest by targeting WIF-1 in vitro and vivo, Pathol. Res. Pract. 214 (11) (2018) 1818-1824.

[6] Y. Wang, C. Wang, N. Liu, J. Hou, W. Xiao, H. Wang, HOXC6 promotes cervical cancer progression via regulation of Bcl-2, FASEB J. 33 (3) (2019) 3901-3911.

[7] Y. Yang, X. Tang, X. Song, L. Tang, Y. Cao, X. Liu, X. Wang, Y. Li, M. Yu, H. Wan, F. Chen, Evidence for an oncogenic role of HOXC6 in human non-small cell lung cancer, PeerJ 7 (2019) e6629.

[8] C. Gong, J. Zou, M. Zhang, J. Zhang, S. Xu, S. Zhu, M. Yang, D. Li, Y. Wang, J. Shi, Y. Li, Up-regulation of MGP by HOXC8 promotes the proliferation, migration, and EMT processes of triple-negative breast cancer, Mol. Carcinog. (2019 Jul 1),

https://doi.org/10.1002/mc.23079.

[9] M. Hu, W. Ou-Yang, D. Jing, R. Chen, Clinical prognostic significance of HOXC9 expression in patients with colorectal cancer, Clin Lab 65 (8) (2019).

[10] Y. Guan, Y. He, S. Lv, X. Hou, L. Li, J. Song, Overexpression of HOXC10 promotes glioblastoma cell progression to a poor prognosis via the PI3K/AKT signaling pathway, J. Drug Target. 27 (1) (2019) 60-66.

[11] Y.J. Liu, Y. Zhu, H.X. Yuan, J.P. Zhang, J.M. Guo, Z.M. Lin, Overexpression of HOXC11 homeobox gene in clear cell renal cell carcinoma induces cellular pro- liferation and is associated with poor prognosis, Tumour Biol. 36 (4) (2015) 2821-2829.

[12] B. Sun, J. Hua, H. Cui, H. Liu, K. Zhang, H. Zhou, MicroRNA-1197 downregulation inhibits proliferation and migration in human non-small cell lung cancer cells by upregulating HOXC11, Biomed. Pharmacother. 117 (2019) 109041.

[13] Z. Tang, C. Li, B. Kang, G. Gao, C. Li, Z. Zhang, GEPIA: a web server for cancer and normal gene expression profiling and interactive analyses, Nucleic Acids Res. 45 (W1) (2017) W98-W102.

[14] D.S. Chandrashekar, B. Bashel, S.A.H. Balasubramanya, C.J. Creighton, I.P. Rodriguez, B.V.S.K. Chakravarthi, S. Varambally, UALCAN: a portal for facil- itating tumor subgroup gene expression and survival analyses, Neoplasia 19 (8) (2017) 649-658.

[15] J.H. Li, S. Liu, H. Zhou, L.H. Qu, J.H. Yang, starBase v2.0: decoding miRNA-ceRNA, miRNA-ncRNA and protein-RNA interaction networks from large-scale CLIP-Seq data, Nucleic Acids Res. 42 (Database issue) (2014) D92-D97.

[16] S.V. Vasaikar, P. Straub, J. Wang, B. Zhang, LinkedOmics: analyzing multi-omics data within and across 32 cancer types, Nucleic Acids Res. 46 (D1) (2018) D956-D963.

[17] Y. Liao, J. Wang, E.J. Jaehnig, Z. Shi, B. Zhang, WebGestalt 2019: gene set analysis toolkit with revamped UIs and APIs, Nucleic Acids Res. 47 (W1) (2019) W199-W205.

[18] K. Makiyama, J. Hamada, M. Takada, K. Murakawa, Y. Takahashi, M. Tada, E. Tamoto, G. Shindo, A. Matsunaga, K. Teramoto, K. Komuro, S. Kondo, H. Katoh, T. Koike, T. Moriuchi, Aberrant expression of HOX genes in human invasive breast carcinoma, Oncol. Rep. 13 (4) (2005) 673-679.

[19] M. McIlroy, D. McCartan, S. Early, P. O Gaora, S. Pennington, A.D. Hill, L.S. Young, Interaction of developmental transcription factor HOXC11 with steroid receptor coactivator SRC-1 mediates resistance to endocrine therapy in breast cancer, Cancer

Res. 70 (4) (2010) 1585-1594.

[20] K.J. Eoh, H.J. Kim, J.Y. Lee, E.J. Nam, S. Kim, S.W. Kim, Y.T. Kim, Pregulation of homeobox gene is correlated with poor survival outcomes in cervical cancer, Oncotarget 8 (48) (2017) 84396-84402.

[21] A. Kulyté, K.H.M. Kwok, M. de Hoon, P. Carninci, Y. Hayashizaki, P. Arner, E. Arner, MicroRNA-27a/b-3p and PPARG regulate SCAMP3 through a feed-for- ward loop during adipogenesis, Sci. Rep. 9 (1) (2019) 13891.

[22] S. Kaur, A. Nag, G. Gangenahalli, K. Sharma, Peroxisome proliferator activated receptor gamma sensitizes non-small cell lung carcinoma to gamma irradiation induced apoptosis, Front. Genet. 10 (2019) 554.

[23] R. Frapolli, E. Bello, M. Ponzo, et al., Combination of PPARy Agonist Pioglitazone

and Trabectedin Induce Adipocyte Differentiation to Overcome Trabectedin Resistance in Myxoid Liposarcomas[J], Clinical Cancer Research 25 (24) (2019) 7565-7575.

[24] Y. Dai, L. Qiao, K.W. Chan, M. Yang, J. Ye, J. Ma, B. Zou, Q. Gu, J. Wang, R. Pang, H.Y. Lan, B.C. Wong, Peroxisome proliferator-activated receptor-gamma con- tributes to the inhibitory effects of Embelin on colon carcinogenesis, Cancer Res. 69 (11) (2009) 4776-4783.

[25] Y. Wu, T. Song, M. Liu, Q. He, L. Chen, Y. Liu, D. Ni, J. Liu, Y. Hu, Y. Gu, Q. Li, Q. Zhou, Y. Xie, PPARG negatively modulates Six2 in tumor formation of clear cell renal cell carcinoma, DNA Cell Biol. 38 (7) (2019) 700-707.