Dickkopf-related protein 1 as a biomarker of local immune status and worse prognosis of Oral squamous cell carcinoma

Abstract

BACKGROUND:

Oral squamous cell carcinoma (OSCC) is an infiltrative malignancy characterized by a significantly elevated recurrence rate. Dickkopf-related protein 1 (DKK1), which plays an oncogene role in many cancers, acts as an inhibitor of the Wingless protein (Wnt) signaling pathway. Currently, there is a lack of consensus regarding the role of DKK1 in OSCC or its clinical significance.

OBJECTIVE:

To examine the role and effect of DKK1 in OSCC.

METHODS:

The identification of differentially expressed genes (DEGs) in OSCC was conducted by utilizing databases such as The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO). A comprehensive analysis of gene expression profile interactions (GEPIA) and Kaplan-Meier curve were conducted to investigate the associations among DEGs, patient survival and prognosis in individuals with OSCC. The biological function of DKK1 in OSCC was investigated by using molecular biology approaches.

RESULTS:

The expression of DKK1 was found to be upregulated in OSCC tissues at various stages. High levels of DKK1 expression exhibited a positive correlation with the overall survival (OS) and progression-free survival (PFS) rates among OSCC patients. DKK1 knockdown suppressed the proliferation and induced apoptotic response in OSCC cells. Moreover, DKK1 exerted a positive regulatory effect on HMGA2 expression, thereby modulating cell growth and apoptosis in OSCC. The expression of DKK1 was found to be positively correlated with the infiltration of immune cells in patients with OSCC. Additionally, higher levels of CD4⁺ T cells were associated with improved 5-year survival rates.

CONCLUSION:

DKK1 is a prognostic biomarker for patients with OSCC.

Keywords

1. Introduction

Oral cancer is one of the ten most frequent cancers with an incidence of 350,000 new cases every year [1, 2, 3]. In the oral cancer, OSCC accounts for more than 95% of cases [4]. Its survival rate has not been improved in the past 20 years [5, 6, 7]. Standard treatment strategies for OSCC consist of surgical resection and radiotherapy along with chemotherapy before and post-surgery [8, 9]. Nevertheless, the treatment efficacy and 5-year OS of OSCC remain mediocre due to local relapse coupled with metastasis, especially in the lung [10]. Most metastatic OSCC patients die within one year of diagnosis [11]. Hence, identifying new precise biomarkers for targeted treatment along with the prognostic assessment of OSCC patients is critical.

Dickkopf-related (DKK) proteins have various roles in different cancers. The DKK1 is a secreted protein implicated in a variety of developmental and physiological processes. The DKK1 is an essential agent in the DKK family [12] and is differently expressed in many malignancies, like non-small cell lung cancer cell [13], colorectal adenoma-carcinoma [14], hepatocellular carcinoma [15], and lymphoma [16]. Moreover, DKK1 can affect the prognosis and survival in HNSCC [17]. Hence, we hypothesized that DKK1 might also play a role in OSCC. However, there is no evidence showing the association between DKK1 and OSCC and the related mechanisms.

Herein, we observed the upregulation of DKK1 in OSCC and evaluated the relationship between the levels of DKK1 with clinicopathological characteristics and prognosis by mining datasets. The biological role of DKK1 in OSCC was assessed by long- and short-term cell growth. The possible molecular processes of DKK1 were analyzed via GSEA along with TCGA datasets. Besides, we also investigated the relationship of DKK1 with tumor-invading immune cells. Overall, we revealed the association of DKK1 with immune infiltration and the related mechanisms to be used as a predictive marker for the prognosis of OSCC patients.

2. Materials and methods

2.1 Patient collection and data extraction

Data in a total of 101 OSCC patients at different stages were collected from December 2017 to February 2021 at the Department of Oral and Maxillofacial Surgery, The Second Affiliated Hospital of Harbin Medical University. Informed consent was obtained from all participants. All protocols were approved by the Institutional Ethics Committee of Harbin Medical University (KY2017-057). The patient information is presented in Table 1.

Table 1
Patient information

Characteristics	High DKK1 expression¹, $n$ (%)		Low DKK1 expression¹, $n$ (%)		$P$ value² ( $X^{2}$ )
Gender
Male	32	(31.68)	26	(25.74)	0.169	(1.896)
Female	23	(22.77)	20	(19.80)
Age
$\leqslant$ 40	5	(4.95)	5	(4.95)	0.504	(0.612)
$>$ 40	49	(48.51)	51	(50.49)
Alcohol history
Yes	22	(21.78)	15	(14.85)	0.740	(0.110)
No	42	(41.58)	30	(29.70)
Smoking history
Yes	29	(28.71)	23	(22.77)	0.841	(0.040)
No	30	(29.70)	19	(18.81)
Differentiation
Well	10	(9.90)	12	(11.88)	0.008	(7.779)
Moderately	30	(29.70)	32	(31.68)
Poorly/undifferentiated	15	(14.85)	2	(1.98)
TNM stage
I	7	(6.93)	9	(8.91)	0.009	(8.938)
II	13	(12.87)	16	(15.84)
III	27	(26.76)	19	(18.81)
IV	7	(6.93)	3	(2.97)
Infiltrate depth
T1 $+$ T2	25	(24.75)	21	(20.79)	0.861	(0.031)
T3 $+$ T4	30	(29.70)	25	(24.75)
Lymph node involvment
Negative	26	(25.57)	35	(34.65)	0.003	(8.840)
Positive	29	(28.71)	11	(10.89)

¹A total score of $\geqslant$ 4 indicated high DKK1 expression, while a total score of $<$ 4 meant low DKK1 expression. ²P value is determined by chi-square test. Clinicopathological features were based on UICC.

The data used in the current study was collected from the public TCGA-HNSC database. Data were retrieved from the Genomic Data Commons (GDC) (https://portal.gdc.cancer.gov/) and included the following information: copy number variation (CNV), expression (Exp) data gathered in HTSeq-FPKM, HTSeq-Counts, as well as somatic mutations data. We also retrieved the survival data and clinical phenotype. We enrolled all the patients with oral cavity lesion sites (buccal mucosa, mandible, maxilla, tongue, floor of mouth, jaw, palate, gum, and lip), and excluded the patients without these clinical data.

2.2 Expression of DKK1 and survival analysis based on TCGA

Tumor Immune Estimation Resource (TIMER) is a web-based data resource containing the 10,009 expression patterns of 23 cancers of TCGA. Here, the TIMER database was used to validate DKK1 levels in different cancers. Gene expression profile interaction analysis (GEPIA), another platform that houses survival and clinicopathological data from various cancers based on TCGA, was used to analyze the association between DKK1 and OS and disease-free survival (DFS) in OSCC and to verify the expression of other prognostic genes in OSCC.

2.3 Comparative analysis of DKK1 expression

The UALCAN data resource (http://ualcan.path.uab.edu) was used to perform a comparative analysis of DKK1 expression. UALCAN is an easy-to-use interactive online tool for assessing cancer transcriptome data (TCGA and MET500 transcriptome sequencing). It can be used to discover markers or conduct in silico verification of possible target genes. Besides, its tools allow users to perform comparative analyses of query gene(s) between malignant and non-malignant samples, as well as diverse cancer molecular sub-types consisting of tumor stages, individual age, gender, or other clinicopathological parameters.

2.4 Calculation of tumor-infiltrating immune cells

The Cell type Identification by Estimating Relative Subsets of RNA Transcripts (CIBERSORT) deconvolution algorithm was used to determine tumor-infiltrating immune cells (TIICs) in OSCC samples. Comparative expression profiles of 547 genes in the samples were explored using CIBERSORT to determine the fraction of 22 TIICs in each instance. Expression patterns between 15 non-malignant tissues and 149 OSCC tissues were extracted from TCGA, and TIIC fractions in all samples were determined using R (v.3.6.1) according to the CIBERSORT algorithm. Then, TIIC fractions in non-malignant hepatic and OSCC tissues were stratified into two subclasses depending on the median levels of DKK1 and represent it in bar plots.

2.5 Enrichment analysis of DKK1 co-expressed genes

Pearson correlation and R were used to analyze the relationships in the TCGA profile. The functions of DKK1 co-expressed genes were assessed using KEGG analysis along with DAVID web tools.

2.6 Establishment of protein-protein interactions (PPI) network and Identification of hub genes

The STRING database was used to create a PPI network and determine the association between DKK1 co-expressed genes. An interaction score of 0.15 was used. Besides, the quality of the PPI network was enhanced using the Cytoscape web tool based on the cross-talk data abstracted from the STRING database. The genes established to directly cross-talk with DKK1 were regarded as the central hub genes, while those that presented cross-talk directly with the central hub genes were defined as subordinate hub genes.

2.7 Immunohistochemistry

An immunohistochemistry analysis was performed on paraffin-embedded slices of OSCC tissue at various stages. In brief, slices were dried at 62^∘C for 30 minutes to 1 hour. After deparaffinization twice with xylene for 20 minutes, they were soaked for 3 to 5 minutes in 100%, 95%, 90%, 80% and 70% ethyl alcohol, respectively. The slices were cooled at room temperature after being treated for 5 minutes with citrate buffer (pH 6.0) at 120^∘C. Following two washings with PBS, incubation with 3% H₂O₂ and three washings again, slices were blocked in 5% bovine serum albumin for 1 h, then incubated overnight at 4^∘C with 5% bovine serum albumin with primary antibodies against DKK1. As the slices were incubated at room temperature for 1 h on the following day, they were reheated for 1 hour and washed three times for 10 minutes each. Color development was then performed using triaminobenzidine tetrahydrochloride as a chromogen, followed by 10 minutes of counterstaining with hematoxylin. Following staining, the slices were dehydrated with increasing concentrations of ethanol and xylene before being photographed with a BX51 microscope (Japan).

2.8 Cell culture and transfection

Human OSCC cell line SCC-4 was acquired from cell bank of the Chinese Academy of Sciences (Shanghai, China). SCC-4 cells were cultivated in DMEM (Dulbecco’s modified Eagle medium, Gibco, USA) enriched with 10% FBS and 1% penicillin-streptomycin solution (Gibco) at 37^∘C and 5% CO₂ conditions.

2.9 Cell Counting Kit-8 (CCK-8) assay

Cell transfects (2 $\times$ 10³ cells/well in 200 $\mu$ L of complete culture medium) were inoculated in 96-well plates. The proliferation of the cells was assessed every 24 h on days 0, 1, 2 and 3. CCK-8 (10 $\mu$ L/well, Cat No: MA0218-1, Meilunbio, Dalian, China) was added to every well at each time point. After a two-hour incubation at 5% CO₂ and 37^∘C, the absorbance at 450 nm was read with a Sunrise Microplate Reader (Tecan Group, Ltd).

2.10 TUNEL assay

DNA breakage was explored using the TUNEL kit (Cat No: WLA030b, Wanleibio, Shenyang, China). After 24-hour cultivation, 30-minute fixation using 4% formaldehyde and rinsing with PBS, the cells were inoculated for 60 mins with the FITC dUTP and TdT enzyme reaction solution in the dark. Then, after 10-minute DAPI staining and three times washing with PBS, a microscope (Olympus) was employed to assess the apoptosis at 495 and 519 nm. ImageJ software (1.48 V) was used to analyze the fluorescence intensity to determine the rate of apoptosis.

2.11 Acridine orange/ethidium bromide (AO/EB) staining

After stable transfection with NC or DKK1 and 3-time washing with PBS, the cells were blocked with AO and EB (1:1) at 20 $\mu$ L for 2–5 min. Stained cells were visualized using immunofluorescence microscopy.

2.12 Statistical analyses

The Kaplan-Meier (K-M) analysis was based on the “survival” R package and the heatmap on the “heatmap” R package. Data from more than two groups were analyzed using the Wilcoxon rank-sum test with nonparametric statistical assumptions and one-way analysis of variance (ANOVA). The relationships between two different variables were accessed by the analysis of Spearman correlation. Data that conformed to Gaussian distribution was analyzed by the $t$ -test and ANOVA followed by Dunnett’s post-hoc test. A $P<$ 0.05 was considered statistically significant. All the statistical analysis was conducted using R software.

3. Results

3.1 Impact of DKK1 on the prognosis of various cancers

The TIMER database was used to explore the expression profile of DKK1 based on TCGA datasets. DKK1 was significantly upregulated in many cancers, such as stomach adenocarcinoma (STAD), lung squamous cell carcinoma (LUSC), liver hepatocellular carcinoma (LIHC), cholangiocarcinoma (CHOL), head and neck squamous cell carcinoma (HNSC), esophageal carcinoma (ESCA), colon adenocarcinoma (COAD), and thyroid carcinoma (THCA) (Fig. 1A). However, DKK1 was significantly downregulated in kidney renal papillary cell carcinoma (KIRP), bladder urothelial carcinoma (BLCA), kidney chromophobe (KICH), breast invasive carcinoma (BRCA), and prostate adenocarcinoma (PRAD). Also, DKK1 was analyzed for survival using GEPIA. In HNSC, both the OS and DFS were significantly different (Fig. 1B and C). Therefore, OSCC was selected as the primary target to examine the potential functions of DKK1.

Figure 1.

Expression and survival analysis of DKK1 in numerous cancers. (A) DKK1 expression in cancers. $P<$ 0.05 (^*), $P<$ 0.01 (^**), $P<$ 0.001 (^***). (B) The DFS and OS in the DKK1-high-expressed and low-expressed patients.

Figure 2.

Relationship of DKK1 expression with clinicopathological features and the diagnostic value of DKK1 in HNSC. Correlations of DKK1 expression with (A) The expression of DKK1 in OSCC, (B) AUC area of DKK1 in OSCC, (C) N stage, (D) Clinical stage, (E) Histologic grade, (F) Grade, (G) Age, (H) N stage, (I) M stage, (J) Race, (K) Alcohol history, (L) Lymph node invasion.

Figure 3.

DKK1 effect on cell growth and apoptosis. (A) The expression of DKK1 in OSCC tissue at different stages. (B) The expression of DKK1 after transfection with DKK1-siRNA. (C) The cell growth was tested after transfection with DKK1-si in SCC-4 cells with CCK-8 method. (D) TUNEL assay. (E) AO/EB staining. $n=$ 3–6. ^{*P <} 0.05 compared with the control group.

Figure 4.

The fraction of 22 TIICs in normal and OSCC tissues. The TIIC in the normal tissues (A) and OSCC tissues (B) were shown. Relationship of DKK1 levels with (C) M0 macrophages, (D) CD8 T cells, (E) follicular helper T cells, and (F) regulatory T cells.

Figure 5.

Enrichment of DKK1 co-expressed genes in OSCC. (A-B) Heatmaps showing the top 50 genes co-expressed and positively (A) or negatively (B) related to DKK1. Enrichment analysis of (C) KEGG and GO (D) BP, (E) CC, and (F) MF of DKK1 co-expressed genes.

Figure 6.

PPI network and the correlation between DKK1 and the hub genes. (A) PPI network and the 19 hub genes interacting with DKK1. (B) Relationship of DKK1 expressions with the 19 hub genes.

Figure 7.

DKK1 positively regulates the expression of HMGA2 in oral cancer cells. (A) Western blots showing the protein levels of HMGA2 in SCC-4 cells with different expression levels of DKK1. (B) Correlation analysis of mRNA expression levels of DKK1 and HMGA2 in TCGA datasets. (C) The mRNA expression of DKK1 after transfect with HMGA2-OE. (D) Cell viability of SCC4-HMGA2-OE cells with DKK1 silencing or not by CCK-8. (E) AO/EB staining in SCC-4-DKK1-OE cells with or without HMGA2 silencing. $n=$ 3–6 independent experiment. ^{*P <} 0.05 compared with the control group.

3.2 Association of DKK1 expression and various clinical indicators

Next, we explored the association of DKK1 expression levels in OSCC using the TCGA database. Our data showed that DKK1 was increased in OSCC patients compared with para-cancerous tissue (normal group, Fig. 2A). DKK1 showed a great area under the curve (AUC, 0.800) in OSCC (Fig. 2B). And then, relationship of DKK1 at age, gender, regional lymph node involvement, as well as tumor grade in OSCC was shown in Fig. 2C–L. Regardless of the clinical indicator, the DKK1 expression in OSCC was remarkably up-regulated. IHC results also confirm our previous results, DKK1 increased in OSCC at a stage-manner (Fig. 3A). These results indicated that DKK1 had an important role in OSCC progress.

3.3 Inhibition of DKK1 can suppress the growth and induce apoptosis in OSCC cells

To assess the mechanisms of DKK1 in OSCC, we first analyzed the expression of DKK1 after treatment with DKK1 siRNA. The protein levels of DKK1 were significantly downregulated in SCC-4 cells (Fig. 3B). Then, we assessed the viability of SCC-4 cells by CCK-8. The DKK1 knockdown inhibited the growth of SCC-4 cells (Fig. 3C). Moreover, the AO/EB and Tunel assays were used to test the role of DKK1 in the apoptosis of OSCC cells. The silencing of DKK1 induced the apoptosis of OSCC cells (Fig. 3D and E). These observations suggested that DKK1 significantly participated in the apoptosis and growth of OSCC cells.

3.4 The relationship between DKK1 expression and TIICs

The results above showed that DKK1 had an essential role in OSCC progress. However, its mechanisms remained unclear. Thus, to study how DKK1 modified the pathological process of OSCC, we collected 149 OSCC tumor samples and 15 paired normal tissue samples in TCGA database. The CIBERSORT algorithm was used to analyze the expression profile of TCGA for 22 TIICs. In normal tissues, the infiltration of CD8 T cells significantly increased in the DKK1 low-expression group (Fig. 4A). In contrast, the infiltration of NK resting cells significantly decreased in the DKK1 low-expression group (Fig. 4A). In OSCC tumor tissues, the infiltrations of M0 macrophages and CD4 T memory resting cells were observed in the DKK1 high expression group (Fig. 4B).

Meanwhile, CD8 T cells, follicular helper T cells, and modulatory T cells were remarkably increased in the low expression group (Fig. 4B). Besides, we observed that, in OSCC tumor samples, the infiltration degree of M0 macrophages ( $R=$ 0.094, $p=$ 0.25) increased with DKK1 expression. However, the infiltration degree of CD8 T cells ( $R=-$ 0.32, $p=$ 5.5e-05), follicular helper T cells ( $R=-$ 0.21, $p=$ 0.01), and regulatory T cells ( $R=-$ 0.3, $p=$ 0.00024) significantly decreased with increasing expressions of DKK1 (Fig. 4C–F). Overall, DKK1 and its related genes are linked to tumor invading immune cells in OSCC.

3.5 Enrichment of DKK1 co-expressed genes

Furthermore, we analyzed DKK1 co-expressed genes using Pearson correlations on TCGA database in R. The top 50 positively and negatively correlated genes were selected (Fig. 5A and B). Then, we used the DAVID database for enrichment analysis of these 100 co-expressed genes. Co-expressed genes were mainly significantly enriched in ribosome biogenesis in eukaryotes, salivary secretion, and mineral absorption pathways (Fig. 5C). The biological processes primarily involved included rRNA processing and transcription, protein stabilization, and positive regulation of cell cycle arrest (Fig. 5D). The cellular components were mainly composed of the nucleolus, small-subunit processome, and nucleoplasm (Fig. 5E). The enriched molecular functions mainly included poly(A) RNA binding, CTP binding, and nitric-oxide synthase regulator activity (Fig. 5F).

3.6 Construction of PPI network and recognition of hub genes

Four genes (HMGA2, FOXN1, KL, and ID2) with high connection were directly related to DKK1 and were defined as central hub genes. TESC, E2F4, ZAP70, KRT2, ALDH2, FXYD2, ALDH18A1, PAX1, IGF2BP2, IDH2, HSP90AB1, HSP90AA1, TRPV6, GZMM, and CALML5 were directly related to the central hub genes and lower-level hub genes of DKK1. Based on TCGA, we analyzed the levels of central and lower-level hub genes and the expression levels of DKK1 in OSCC tumor samples (Fig. 6A and B, Fig. S1).

The survival analysis was conducted with the 19 hub genes in OSCC tumor samples in TCGA. Four hub genes presented prognostic values (Fig. S2A, C, E, and G). Thus, a ROC model was established for these four hub genes. The AUC was over 0.5 (Fig. S2B, D, F, and H). These results showed that DKK1 might have a modulatory role in the polarization of HMGA2, FOXN1, KL, and ID2, thereby contributing to OSCC.

3.7 DKK1 positively regulates the expression of HMGA2 in OSCC cells

Moreover, the four hub genes (HMGA2, FOXN1, KL, and ID2) showed the greatest fold changes (Fig. S2). HMGA2 showed a good AUC (0.957) in OSCC (Fig. S2A). Western blot results also suggested that HMGA2 is down-regulated after transfect DKK1 inhibition in CAL-27 cells (Fig. 7A). Meanwhile, a strong correlation of HMGA2 and DKK1 expression was observed in TCGA datasets ( $P<$ 0.001, $r=$ 0.463, Fig. 7B). We observed that DKK1 inhibition downregulated HMGA2 expression in CAL-27 cells. Meanwhile, overexpressed DKK1 (DKK1-OE) exhibited increased HMGA2 expression, which was consistent with the observation at protein level. However, there is no change of DKK1 mRNA expression after transfecting with HMGA2-OE (Fig. 7C). These results indicated that HMGA2 expression was controlled by DKK1. After the suppression of DKK1 in SCC-4 cells by HMGA2 overexpression, the cell viability and apoptosis assays provided evidence that DKK1 promoted OSCC by regulating HMGA2 expression (Fig. 7D and E).

4. Discussion

OSCC is a frequent neoplasm of head and neck cancers [18, 19]. Despite the remarkable advancements in the diagnosis and therapy of OSCC in the past decades, the prognosis of OSCC patients remains poor [20]. Aberrant gene expression can be engaged in tumorigenesis and is linked to the prognosis of patients [21]. Nevertheless, the molecular mechanisms of OSCC remain unclear. Herein, we first showed that the expression of DKK1 was dramatically elevated in OSCC tissues and was linked to tumor grade, TNM stage, infiltration depth, lymph node metastasis, as well as the vital status of OSCC patients. Besides, DKK1 inhibition repressed cell growth and triggered apoptosis in OSCC cells. Meanwhile, DKK1 overexpression was linked to poor prognosis. These data indicated that DKK1 is an oncogene and promotes OSCC.

Firstly, we found that DKK1 was significantly upregulated in COAD, CHOL, BLCA, OSCC, READ, STAD, and UCEC based on the TIMER analysis. The GEPIA data showed significant differences in the DFS and OS between the high and low DKK1 expression groups in OSCC, indicating that DKK1 played a potential role in OSCC pathology. Simultaneously, the expression of DKK1 in the HNSC tumor group was upregulated and was different among clinical groups (Fig. 2).

Furthermore, DKK1 has been demonstrated as an oncogene in many cancers, especially squamous cell carcinoma. For example, Gao et al. showed positive correlations between DKK1 and cell growth and poor survival in head and neck squamous cell carcinoma [17]. Moreover, Shi et al. suggested that DKK1 regulates cell growth and is a novel prognostic biomarker for laryngeal squamous cell carcinoma [22]. Thus, we evaluated the expression of DKK1 in OSCC by IHC. We showed that DKK1 was upregulated in OSCC in a staged manner (Fig. 3A). On the other hand, DKK1 affected cell growth and apoptosis of SCC-4 cells, a significant part of the OSCC progress. Our results indicated that DKK1 knockdown could inhibit the growth and induce the apoptosis of SCC-4 cells (Fig. 3). These findings indicated that DKK1 inhibition can suppress cell growth and induce apoptosis. Hence, DKK1 plays an essential role in OSCC progress.

Previous research has suggested that TIICs can be used to estimate the status of lymph nodes and cancer patients OS [23, 24]. Here, we found a remarkable reduction in Mast resting cells in the high DKK1 group of OSCC compared with the increased proportion in the high DKK1 group of non-malignant tissues, suggesting that the elevated DKK1 might inhibit the expression of resting Mast cells in OSCC. M0 macrophages, similar to activated Mast cells and CD4 memory sleeping T cells [25], did not differ between the high and low DKK1 groups in normal tissues but increased in the high DKK1 group in OSCC tissues. Besides, activated Mast cells and M0 macrophages were positively linked to the expression of DKK1. In contrast, follicular helper T cells and naive B cells were significantly negatively correlated with their presentation. The typing, as well as quantity conversion of TIICs, in non-malignant and cancer tissues, demonstrated the important role of DKK1 in modulating the tumor immune microenvironment of OSCC but the deeper mechanisms should be further investigated.

Then, we used Spearman correlation data on TCGA in R to calculate DKK1 co-expressed genes and the top 50 negative and positive correlations were selected. Next, we analyzed these 100 co-expressed gene enrichment to understand the mechanisms of DKK1. The results showed that the co-expressed genes were mainly enriched in bacterial invasion of epithelial cells, modulation of the actin cytoskeleton, focal adhesion, and were primarily involved in biological processes such as extracellular matrix organization, actin filament bundle assembly, and hemidesmosome assembly, most of which are thought to be significant regulatory factors in some cancers [26, 27, 28]. The cellular components were mainly composed of plasma membrane, extracellular exosome, focal adhesion, and stress fiber, strongly linked to tumor infiltration and metastasis [29, 30, 31]. The molecules mainly presented functions such as receptor binding, ion channel binding, and integrin binding.

The molecular modulatory cascades by which DKK1 regulates OSCC is not fully known. Hence, a PPI network was constructed using DKK1 co-expressed genes and screened 19 hub genes that cross-talked with DKK1. The results showed that four hub genes (HMGA2, FOXN1, KL, and ID2) presented prognostic values (Fig. S2), and a ROC model was established for these genes. Aranda et al. [32] showed that FOXN1 is an essential immune effector responsible for eliminating hyperploid cancer cells. In immunocompromised patients with pneumonia, Gülbudak [33] suggested that the KL family has a major role in infection and colonization processes. Moreover, Liu et al. indicated that ID2 promotes early-stage breast cancer progress by modulating cancer stemness [34]. Here, the AUC for these genes was over 0.5. Additionally, high expression of DKK1 can promote the differentiation of macrophages into TAMs, thereby enhancing OSCC progress and leading to a poor prognosis. HMGA2 and PRTN3 proteins participate in the progression of vulvar squamous cells [35]. Herein, we found that HMGA2 was a direct downstream target of DKK1. HMGA2 rescued DKK1 cell function effects of OSCC during in vitro experiments.

Our manuscript describes the oncogene DDK1 and its biological functions in OSCC. However, there are still limitations to our study. Firstly, bioinformatic studies initially relied on data from multiple historical datasets. However, it is imperative to collect prospective data from a clinical cohort in order to develop more reliable clinical applications. The validation process will require further studies involving experimental verification and the identification of potential immune-related mechanisms.

5. Conclusion

We demonstrated DKK1 as a pan-cancer gene with remarkable prognostic significance in many cancers, especially OSCC. DKK1 might be an independent predictor of OSCC. Moreover, DKK1 knockdown inhibits cell growth and induces apoptosis. Hence, DKK1 might impact OSCC progress by modulating the tumor immune microenvironment, as well as participating in cell growth-linked cascades. Finally, further studies are required to advance the present findings.

Funding

This study was supported by the Medical Wisdom Research Fund by the Heilongjiang Sunshine Health Foundation (H21L0803), the Heilongjiang Academy of Medical Science (201703), and the National Science Foundation of Heilongjiang Province (LH2021H033).

Supplementary data

The supplementary files are available to download from https://dx-doi-org.web.bisu.edu.cn/10.3233/THC-230527.

Footnotes

Acknowledgments

None to report.

Conflict of interest

The authors declare that they have no conflict of interest.

References

Minervini

Franco

Marrapodi

Crimi

Badnjevic

Cervino

, et al. Correlation between temporomandibular disorders (TMD) and posture evaluated trough the diagnostic criteria for temporomandibular disorders (DC/TMD): A systematic review with meta-analysis. J Clin Med. 2023; 12(7).

Minervini

Franco

Marrapodi

Ronsivalle

Shapira

Cicciu

. Prevalence of temporomandibular disorders in subjects affected by Parkinson disease: A systematic review and metanalysis. J Oral Rehabil. 2023; 50(9): 877-85.

D’Souza

Addepalli

. Preventive measures in oral cancer: An overview. Biomed Pharmacother. 2018; 107: 72-80.

Peng

Cheng

Zhang

Fan

Mao

. circRNA_0000140 suppresses oral squamous cell carcinoma growth and metastasis by targeting miR-31 to inhibit Hippo signaling pathway. Cell Death Dis. 2020; 11(2): 112.

Saidak

Lailler

Testelin

Chauffert

Clatot

Galmiche

. ASO Visual Abstract: Contribution of Genomics to the Surgical Management and Study of Oral Cancer. Ann Surg Oncol. 2021.

Shi

Xue

Jiang

Qiu

. Glaucocalyxin A induces apoptosis and autophagy in tongue squamous cell carcinoma cells by regulating ROS. Cancer Chemother Pharmacol. 2021.

Zhong

Zhang

Hong

Sun

Chen

, et al. [Corrigendum] FNDC3B promotes epithelialmesenchymal transition in tongue squamous cell carcinoma cells in a hypoxic microenvironment. Oncol Rep. 2021; 45(6).

Samolyk-Kogaczewska

Sierko

Zuzda

Gugnacki

Szumowski

Mojsak

, et al. PET/MRI-guided GTV delineation during radiotherapy planning in patients with squamous cell carcinoma of the tongue. Strahlenther Onkol. 2019; 195(9): 780-91.

Galli

Bondi

Canevari

Tulli

Giordano

Di Santo

, et al. High-risk early-stage oral tongue squamous cell carcinoma, when free margins are not enough: Critical review. Head Neck. 2021.

10.

Fan

Wang

Zhang

Deng

Jiang

Liang

, et al. NUPR1 promotes the proliferation and metastasis of oral squamous cell carcinoma cells by activating TFE3-dependent autophagy. Signal Transduct Target Ther. 2022; 7(1): 130.

11.

Fan

Liu

Yang

Jin

Xiao

. LncRNA NOP14-AS1 Promotes Tongue Squamous Cell Carcinoma Progression by Targeting MicroRNA-665/HMGB3 Axis. Cancer Manag Res. 2021; 13: 2821-34.

12.

Lyros

Lamprecht

Nie

Thieme

Gotzel

Gasparri

, et al. Dickkopf-1 (DKK1) promotes tumor growth via Akt-phosphorylation and independently of Wnt-axis in Barrett’s associated esophageal adenocarcinoma. Am J Cancer Res. 2019; 9(2): 330-46.

13.

Sun

Chen

Wang

. Comparison of the diagnostic value of CEA combined with OPN or DKK1 in non-small cell lung cancer. Oncol Lett. 2020; 20(3): 3046-52.

14.

Liu

Sun

Zhao

Zhang

, et al. Dickkopf-1 expression is down-regulated during the colorectal adenoma-carcinoma sequence and correlates with reduced microvessel density and VEGF expression. Histopathology. 2015; 67(2): 158-66.

15.

Fezza

Moussa

Aoun

Haber

Hilal

. DKK1 promotes hepatocellular carcinoma inflammation, migration and invasion: Implication of TGF-beta1. PLoS One. 2019; 14(9): e0223252.

16.

van Andel

Kocemba

Spaargaren

Pals

. Aberrant Wnt signaling in multiple myeloma: Molecular mechanisms and targeting options. Leukemia. 2019; 33(5): 1063-75.

17.

Gao

Xiao

Guo

Shen

Cheng

, et al. Elevated DKK1 expression is an independent unfavorable prognostic indicator of survival in head and neck squamous cell carcinoma. Cancer Manag Res. 2018; 10: 5083-9.

18.

Solomon

Young

Rischin

. Head and neck squamous cell carcinoma: Genomics and emerging biomarkers for immunomodulatory cancer treatments. Semin Cancer Biol. 2018; 52(Pt 2): 228-40.

19.

Bakst

Glastonbury

Parvathaneni

Katabi

Yom

. Perineural invasion and perineural tumor spread in head and neck cancer. Int J Radiat Oncol Biol Phys. 2019; 103(5): 1109-24.

20.

Zhang

Liu

Yang

. Identi fi cation of key genes and pathways in tongue squamous cell carcinoma using bioinformatics analysis. Med Sci Monit. 2017; 23: 5924-32.

21.

Zheng

Zhang

Shen

Zhang

Lin

. LncRNA-MALAT1 is a promising biomarker for prognostic evaluation of tongue squamous cell carcinoma. Eur Arch Otorhinolaryngol. 2020; 277(11): 3155-60.

22.

Shi

Gong

Zhou

Tian

Wang

. Dickkopf-1 is a novel prognostic biomarker for laryngeal squamous cell carcinoma. Acta Otolaryngol. 2014; 134(7): 753-9.

23.

Zhang

Kang

Guo

Sun

Lin

, et al. Tumor-infiltrating immune cells act as a marker for prognosis in colorectal cancer. Front Immunol. 2019; 10: 2368.

24.

Zhao

Cheng

Gai

Zhang

. SPOCK2 serves as a potential prognostic marker and correlates with immune infiltration in lung adenocarcinoma. Front Genet. 2020; 11: 588499.

25.

Wang

Zhang

Gao

Tang

, et al. Profiles of immune cell infiltration and immune-related genes in the tumor microenvironment of colorectal cancer. Biomed Pharmacother. 2019; 118: 109228.

26.

Al Absi

Wurzer

Guerin

Hoffmann

Moreau

Mao

, et al. Actin cytoskeleton remodeling drives breast cancer cell escape from natural killer-mediated cytotoxicity. Cancer Research. 2018; 78(19): 5631-43.

27.

Insua-Rodriguez

Oskarsson

. The extracellular matrix in breast cancer. Advanced Drug Delivery Reviews. 2016; 97: 41-55.

28.

Lin

Huang

Gunda

Sun

Chellappan

, et al. Fascin controls metastatic colonization and mitochondrial oxidative phosphorylation by remodeling mitochondrial actin filaments. Cell Reports. 2019; 28(11): 2824-36 e8.

29.

Park

Burckhardt

Lazcano

Solis

Isogai

, et al. Mechanical regulation of glycolysis via cytoskeleton architecture. Nature. 2020; 578(7796): 621-6.

30.

Shen

Cao

Wang

Zeng

Liu

, et al. Hippo component YAP promotes focal adhesion and tumour aggressiveness via transcriptionally activating THBS1/FAK signalling in breast cancer. Journal of Experimental & Clinical Cancer Research: CR. 2018; 37(1): 175.

31.

Wortzel

Dror

Kenific

Lyden

. Exosome-mediated metastasis: Communication from a distance. Developmental cell. 2019; 49(3): 347-60.

32.

Aranda

Chaba

Bloy

Garcia

Bordenave

Martins

, et al. Immune effectors responsible for the elimination of hyperploid cancer cells. Oncoimmunology. 2018; 7(8): e1463947.

33.

Gulbudak

Ozturk

Kuyugoz

Tezcan Ulger

. Investigation of pneumocystis jirovecii infection and colonization in immunocompromised patients with pneumonia. Mikrobiyol Bul. 2020; 54(4): 583-95.

34.

Liu

Pandey

Sharma

Xing

Chittiboyina

, et al. ID2 and GJB2 promote early-stage breast cancer progression by regulating cancer stemness. Breast Cancer Res Treat. 2019; 175(1): 77-90.

35.

Fatalska

Rusetska

Bakula-Zalewska

Kowalik

Zieba

Wroblewska

, et al. Inflammatory proteins HMGA2 and PRTN3 as drivers of vulvar squamous cell carcinoma progression. Cancers (Basel). 2020; 13(1).

Dickkopf-related protein 1 as a biomarker of local immune status and worse prognosis of Oral squamous cell carcinoma

Abstract

BACKGROUND:

OBJECTIVE:

METHODS:

RESULTS:

CONCLUSION:

Keywords

1. Introduction

2. Materials and methods

2.1 Patient collection and data extraction

Table 1 Patient information

2.3 Comparative analysis of DKK1 expression

2.4 Calculation of tumor-infiltrating immune cells

2.5 Enrichment analysis of DKK1 co-expressed genes

2.6 Establishment of protein-protein interactions (PPI) network and Identification of hub genes

2.7 Immunohistochemistry

2.8 Cell culture and transfection

2.9 Cell Counting Kit-8 (CCK-8) assay

2.10 TUNEL assay

2.11 Acridine orange/ethidium bromide (AO/EB) staining

2.12 Statistical analyses

3. Results

3.1 Impact of DKK1 on the prognosis of various cancers

3.3 Inhibition of DKK1 can suppress the growth and induce apoptosis in OSCC cells

3.4 The relationship between DKK1 expression and TIICs

3.5 Enrichment of DKK1 co-expressed genes

3.6 Construction of PPI network and recognition of hub genes

3.7 DKK1 positively regulates the expression of HMGA2 in OSCC cells

4. Discussion

5. Conclusion

Funding

Supplementary data

Footnotes

Acknowledgments

Conflict of interest

References

Table 1
Patient information