**Table 2.** Studies on ZHX2 expression, function, clinical significance, and molecular mechanism.



OS in patients with gastric cancer.

220

ZHX2-involved tumor occurrence, development, progression in various tumors, and potential molecular mechanisms are summarized in Figure 2. Tumor-suppressive functions of ZHX2 were supported by the following studies, with most from hepatocellular carcinoma (HCC). Elevating ZHX2 expression can suppress HCC cell multiplication, colony formation, and mice tumor size. Conversely, reducing ZHX2 expression by siRNA significantly enhanced cell proliferation and colony formation [37]. Mechanistic studies demonstrated that tumor inhibition was through the suppression of Cyclins A and E via a direct binding between ZHX2 and the promoter domain of Cyclins A and E. The inverse correlation between ZHX2 and Cyclins A and E was further detected from liver cancer tissues. The overexpression of ZHX2, including nuclei and the cytoplasm, did not differ in tumor tissues and adjacent non-tumor tissues. Interestingly, a lower nuclear ZHX2 expression and a higher cytoplasmic ZHX2 expression were found in HCC tissues, but a higher nuclear ZHX2 expression and lower cytoplasmic ZHX2 expression were found in adjacent non-tumor tissue. This suggests that a reduced nuclear expression and increased cytoplastic expression of ZHX2 could play roles in hepatocarcinogenesis. This was further supported by cellular fragment analysis, showing that the low nuclear expression of ZHX2 was correlated with a high Cyclins A and E and Ki-67 expression level. Moreover, the patients with reduced ZHX2 nuclear expression have poor tumor differentiation, larger sizes, and a significantly shorter survival time [37]. This indicated the critical tumor-inhibitory roles of nuclear ZHX2 in HCC progression. Epigenetic analysis results have revealed that a reduced expression of ZHX2 was due to the hypermethylation of the ZHX2 promoter [38]. carcinoma (HCC). Elevating ZHX2 expression can suppress HCC cell multiplication, col‐ ony formation, and mice tumor size. Conversely, reducing ZHX2 expression by siRNA significantly enhanced cell proliferation and colony formation [37]. Mechanistic studies demonstrated that tumor inhibition was through the suppression of Cyclins A and E via a direct binding between ZHX2 and the promoter domain of Cyclins A and E. The inverse correlation between ZHX2 and Cyclins A and E was further detected from liver cancer tissues. The overexpression of ZHX2, including nuclei and the cytoplasm, did not differ in tumor tissues and adjacent non‐tumor tissues. Interestingly, a lower nuclear ZHX2 ex‐ pression and a higher cytoplasmic ZHX2 expression were found in HCC tissues, but a higher nuclear ZHX2 expression and lower cytoplasmic ZHX2 expression were found in adjacent non‐tumor tissue. This suggests that a reduced nuclear expression and increased cytoplastic expression of ZHX2 could play roles in hepatocarcinogenesis. This was further supported by cellular fragment analysis, showing that the low nuclear expression of ZHX2 was correlated with a high Cyclins A and E and Ki‐67 expression level. Moreover, the patients with reduced ZHX2 nuclear expression have poor tumor differentiation, larger sizes, and a significantly shorter survival time [37]. This indicated the critical tumor‐ inhibitory roles of nuclear ZHX2 in HCC progression. Epigenetic analysis results have revealed that a reduced expression of ZHX2 was due to the hypermethylation of the ZHX2 promoter [38].

ZHX2‐involved tumor occurrence, development, progression in various tumors, and potential molecular mechanisms are summarized in Figure 2. Tumor‐suppressive func‐ tions of ZHX2 were supported by the following studies, with most from hepatocellular

*Int. J. Mol. Sci.* **2022**, *23*, x FOR PEER REVIEW 10 of 24

**Figure 2.** ZHX2 and cancers: biological functions and potential signaling networking. **Figure 2.** ZHX2 and cancers: biological functions and potential signaling networking.

Figure 2 A previous study demonstrated that the ZHX2 protein was involved in the postnatal repression of alpha fetoprotein (AFP) in mice [39]. Other studies have shown

Figure 2 A previous study demonstrated that the ZHX2 protein was involved in the postnatal repression of alpha fetoprotein (AFP) in mice [39]. Other studies have shown that ZHX2 expression in HCC was inversely correlated with serum AFP levels in HCC patients [14], suggesting a repressive role of ZHX2 on AFP. This was supported by in vitro experiments, showing that ZHX2 overexpression significantly decreased AFP secretion in HCC cells and that reducing the expression of ZHX2 restored APF expression in LO2 and SMMC7721 cells. Luciferase-based assays revealed that the repression of AFP by ZHX2 is regulated by the AFP promoter and requires intact HNF1 binding sites [40]. Besides the repression of AFP, ZHX2 also repressed the cancer biomarkers pyruvate kinase M1/2 (PKM) and hexokinase 2 (HK2) in HCC [16]. Moreover, ZHX2 suppressed oncogenic activation of glypican 3 (GPC3) in HCC [41]. Special attention has been given to GPC3 because of its diagnostic potential in HCC [61]. However, ZHX2 expression was negatively associated with GPC3 from HCC tissues and in cultured liver cell lines. In these studies, the increased expression of ZHX2 markedly reduced GPC3 expression, while the downregulation of ZHX2 by small molecules elevated GPC3 expression levels in vitro-cultured cells. Mechanistic studies have revealed that the suppressive action of ZHX2 is achieved through binding to the GPC3 promoter [41].

Epidemiology and clinical studies have demonstrated that virus-related hepatocellular carcinoma, particularly hepatitis B virus (HBV)-related HCC, is one of the main types of HCC [62]. During carcinogenesis of HCC, ZHX2 also plays a tumor-inhibitory role through dysregulating HBV X protein (HBx). In turn, HBx plays an important part in HBV-related HCC, activating miR-3188 and Notch signaling via CREB, and then inhibiting ZHX2 expression [42]. Conversely, the interaction between ZHX2 and NFYA can reduce Notch1. It has been reported that miR-3188 plays an oncogenic role and is highly expressed in HBV transgenic mice liver and HepG2.215 cells. Knockdown of miR-3188 inhibits cell proliferation, migration, invasion, and mouse tumor growth [42]. Thus, HBx-miR-3188- ZHX2-Notch1 is considered as the major signaling pathway during carcinogenesis and development in HBV-associated HCC. Song et al. also reported that ZHX2 is an HBVrelated tumor suppressor gene in HCC [43]. ZHX2 expression is significantly decreased in tumor tissues of HBV-positive HCC and livers of HBV-transgenic mice. Further studies confirmed that ZHX2 expression levels and tumor-suppressive activities were inhibited by HBV-encoded proteins, particularly HBx [43]. Mechanistic studies showed that ZHX2 expression levels were affected by miR-155 via its seed sites in the ZHX2 3UTR. In fact, miR-155 levels were significantly increased in HBx-overexpressing HCC cell lines, HBVpositive HCC tissues, and HBV-transgenic mouse livers. Interestingly, ZHX2 levels were upregulated when miRNA-155 levels were blocked in vitro [43]. These findings suggest that HBV-mediated HCC-promoting properties could be attributed to the miR-155-induced silencing of ZHX2, and they suggest a novel therapy for HBV-related HCC by targeting the miR-155 pathway.

As an HCC-associated tumor suppressor, ZHX2 is found to be hepatoprotective by reducing liver lipid levels in a high-fat diet-fed rodent model [63]. In addition, ZHX2 could suppress lipoprotein lipase and indirectly prevent HCC formation [64]. It is known that lipogenesis plays an important role in carcinogenesis and the development of HCC. An experimental study has shown that the overexpression of ZHX2 in HCC cells significantly inhibited de novo lipogenesis [44]. ZHX2 expression was inversely correlated with the expression of SREBP1c. Further studies have revealed that ZHX2 inhibited liver tumorigenesis and development by inhibiting SREBP1c, which regulates de novo lipogenesis. The therapeutic strategy provided additional evidence showing that fatostatin, as a SREBP1c suppressor, attenuated liver tumorigenesis in Zhx2 conditional deletion in the liver of mouse models. Mechanistically, ZHX2 upregulated miR-24–3p expression at transcription levels, further leading to the degradation of SREBP1c targeted by miR-24–3p. Conclusively, these data provided a novel mechanism for ZHX2 inhibiting HCC progression, and targeting the ZHX2/SREBP1c axis could be a novel treatment method of liver cancer.

Numerous studies have proved that non-alcoholic fatty liver disease (NAFLD) is closely associated with hepatocellular carcinoma (HCC). ZHX2 was found to be decreased in NAFLD-HCC liver tissue [45]. This study showed that increasing ZHX2 expression disturbed lipid homeostasis and hepatocytes' lipid deposition and prevented exogenous lipids uptake via the suppression of lipid lipase (LPL), resulting in the inhibition of HCC cell proliferation. Vice versa, increasing LPL expression in HCC cells could reverse ZHX2 mediated cell proliferation inhibition, tumor growth in a xenograft mouse model, lipid metabolism, and tumor formation in mouse liver. Moreover, in an HCC cohort study, immunohistochemical staining demonstrated a negative correlation between ZHX2 and LPL expression. Taken together, ZHX2 protects hepatocytes from cell growth and NAFLD-HCC progression through regulating lipid deposition and the transcriptional repression of LPL [45].

ZHX2 expression levels have been significantly reduced in liver cancer stem cells (CSCs) from different origins [46]. Manipulating the deficiency of ZHX2 could enhance tumor progression and liver CSC stemness. Elevating ZHX2 expression attenuated liver CSCs transformation from the initiation of tumor formation, self-renewal, and sorafenibresistance. Mechanical studies have demonstrated that ZHX2 suppresses liver CSCs via epigenetic regulation, that is, by suppressing the KDM2A-related demethylation of histone H3 lysine 36 (H3K36), because the H3K36 promoter has multiple binding sites of stemnessassociated transcription factors, including NANOG, SOX4, and OCT4. Clinical studies have found that a lower expression of ZHX2 and a higher KDM2A expression were associated with a shorter survival time, which was linked to the transcriptionally repressing KDM2A by ZHX2 in HCC. These findings have improved our understanding of the molecular mechanisms of HCC relapse and drug resistance [46].

Therapeutic studies have also demonstrated that ZHX2 improved the cytotoxicity of chemotherapeutic drugs by inhibiting multidrug resistance 1 (MDR1) via an interaction with NFYA in liver tumor cells [48]. Firstly, an inverse correlation of ZHX2 and MDR1 expression levels was observed in HCC tissues by immunohistochemical staining. Secondly, luciferase reporter assays showed that ZHX2 repressed the promoter activity of the MDR1. However, the knockdown of NFYA or mutation of the NFY binding site eliminated the ZHX2-mediated repression of MDR1 at a transcriptional level. This suggests that ZHX2 was interacting with NFYA, thus, reducing NFY binding to the MDR1 promoter.

However, it has been reported that the ZHX2 protein was overexpressed in liver cancer tissues, and that its upregulation expression was related to poor differentiation and cancer metastasis [47], suggesting an oncogene role of ZHX2 in HCC.

In lung cancer, ZHX2 exhibits tumor suppressor functions. For example, ZHX2 expression was significantly reduced in the human lung cancer cell lines [49]. Compared to the control groups, cell growth, moving, and invasion were dramatically inhibited by ZHX2, in which cell apoptosis and apoptosis-related proteins were increased. In addition, ZHX2 inhibited tumor growth in terms of the reductions in growth rate, tumor size and weight, and PCNA-positive cell numbers. Mechanically, ZHX2 acts as a tumor suppressor in lung cancer by inhibiting the p38MAPK signaling pathway.

In glioma cells, it has been reported that ZHX2 interacts with HNRNPD and further regulates vasculogenic mimicry (VM) formation through the linc00707/miR-651–3p/SP2 pathway [50]. The expression levels of ZHX2 and miR-651–3p were remarkedly reduced, while HNRNPD, linc00707, and specific protein 2 (SP2) expression were remarkedly increased in glioma. Moreover, increasing the ZHX2 and miR-651–3p expression or decreasing the expression of HNRNPD, linc00707, and SP2 suppressed glioma cells' proliferation, migration, invasion, and VM formation. Interestingly, by reducing the HNRNPD expression, it caused an increase in ZHX2 mRNA stability, and ZHX2 negatively regulated the linc00707 expression by binding to its promoter region. Firstly, linc00707 binds with miR-651–3p, and the latter binds to the SP2 mRNA 30 untranslated region (30UTR) and negatively regulates its expression. Then, SP2 binds to the promoter regions of MMP2, MMP9, and VE- cadherin, which are VM formation-related proteins, and therefore, play a role in enhancing transcription for the regulation of glioma cell VM formation.

It was also reported that ZHX2 inhibited the progression of thyroid cancer [51]. The decreased ZHX2 expression in thyroid cancer tissues was correlated with poor outcomes. The knockdown of ZHX2 significantly promoted thyroid cancer cell migration. S100 calciumbinding protein (S100A14) was highly expressed in human thyroid cancers, showing a negative correlation with ZHX2. Mechanistically, ZHX2 binds to the promoter region of the S100 to decrease its transcription. The inhibition of S100A14 attenuated thyroid cancer cell metastasis induced by ZHX2 knockdown in cultured cells and in animal models.

Multiple myeloma (MM) is an incurable hematological malignancy. Compared to the low-risk or indolent disease, ZHX2 expression was remarkably reduced in the high-risk or active disease. The multiple myeloma patients with lower ZHX2 expression have worse outcomes [54]. Other studies have also reported that increasing the ZHX2 expression could improve the response to high-dose chemotherapy in multiple myeloma [52,53]. In contrast, another study by Jiang et al. reported that multiple myeloma patients with a higher ZHX2 expression showed poorer clinical outcomes, and the knockdown of ZHX2 in cancer cells caused MM cells to be more sensitive to Bortezomib (BTZ) by regulating the nuclear translocation of NF-κB, also affecting NF-κB and the corresponding target genes' expression at the mRNA levels [55].

ZHX2 has also been reported as a tumor suppressor in Hodgkin lymphoma, where the recurrent breakpoint at 8q24 targets ZHX2. This aberration broke the far upstream activation elements of ZHX2 and decreased ZHX2 expression [56]. The gene expression profiling results indicated the regulating function of ZHX2 on differentiation and apoptosis. STAT1 (signal transducer and activation of transcription 1) and several STAT1 target genes were included, suggesting that ZHX2 functions as a tumor suppressor in Hodgkin lymphoma [56]. Further studies demonstrated the transcriptional deregulation of ZHX2 in B cell malignancies. Two transcription factors, homeodomain protein msh homeobox 1 (MSX1) and bZIP protein X-box binding protein 1 (XBP1), were shown to directly regulate ZHX2 expression. Multiple mechanisms might be involved in the suppression of ZHX2 expression in Hodgkin lymphoma cell lines. These include the loss of activation elements of ZHX2 upstream and a decreased expression of activators MSX1 and XBP1 [57].

ZHX2 can drive tumorigenesis in clear cell renal cell carcinoma (ccRCC). The loss of VHL usually upregulates ZHX2 levels, in particular, the nuclear expression of ZHX2 in ccRCC tumors. Mechanistic studies showed that VHL-upregulating ZHX2 was achieved through prolyl hydroxylation and proteasomal degradation signaling, supported by an experiment showing that the inhibition of prolyl hydroxylation and the degradation of proteasome could increase ZHX2 expression. The engineered deletion of ZHX2 caused decreased anti-apoptosis-associated multiple gene expression and inhibited cancer cell growth, metastasis, and metabolism. Moreover, ZHX2 promoted NF-κB activation and drove renal carcinogenesis. These studies reveal the oncogenic functions of ZHX2 in renal cancer and uncover a potential therapeutic target for ccRCC [21]. The ccRCC study of Zhu et al. found an additional mechanism. ZHX2-driven cell proliferation and migration were achieved though the activation of the MEK/ERK1/2 signaling pathway and the increasing VEGF expression, and led to Sunitinib resistance through regulating self-protective autophagy, providing a new insight for advanced ccRCC treatment [58].

Triple-negative breast cancer (TNBC) belongs to a more aggressive breast cancer subtype with a higher mortality rate; therefore, it is an urgent requirement to find a novel treatment strategy. Studies have demonstrated that ZHX2 is amplified or increased in TNBC patient tissues. Functional studies showed that manipulating the deletion of ZHX2 inhibits TNBC tumor growth and metastasis. Chromosomal immunoprecipitation sequencing results have shown that ZHX2 binds and transcriptionally activates HIF1α (hypoxia-inducible factor alpha), finally promoting gene expression. AP2B1, COX20, KDM3A, and PTGES3L were regulated by both ZHX2 and HIF1α—their overexpression could partially rescue ZHX2 depletion-caused cell growth in TNBC cells. These findings

strongly suggest that these target genes synergistically help the oncogene function of ZHX2. Genomic studies further showed that three residues (R491, R581, and R674) on ZHX2 are important for ZHX2 to exert its transcriptional activity. It was suggested that ZHX2 acts as an oncogene via activating the HIF1α pathway in TNBC, and ZHX could be used as a novel therapeutic target for TNBC [59].

It has been reported that ZHX2 forms heteromeric complexes with ZHX3 [8,16]. Their effect on breast cancer prognosis is similar. Both ZHX2 and ZHX3 expressions were remarkably reduced in TNBC tissues, and a high mRNA expression of ZHX2 and ZHX3 were closely associated with a better prognosis of breast cancer patients, especially in luminal A subtype breast cancer. In addition, the mRNA expression of ZHX2 and ZHX3 were also positively associated with the estrogen receptor and progesterone receptor, but ZHX2 mRNA expression was inversely correlated with HER2 in breast cancers [19].

In gastric cancer (GC), ZHX2 also exerts an oncogenic function [60]. An increase in ZHX2 expression was closely correlated with clinical characteristics; for instance, a higher ZHX2 expression predicted worse outcomes and immune infiltrating. An in vitro experimental study showed that ZHX2 overexpression can facilitate gastric cancer cell proliferation and migration and can suppress cell apoptosis. These findings suggest ZHX2 as a prognostic biomarker, and immune infiltration could be associated with the effect of ZHX2 on gastric cancer.

By contrast, gastric cancer patients with a decreased ZHX2 and ZHX3 mRNA expression showed better outcomes [31]. Reduced ZHX2 and ZHX3 expression was remarkably correlated with a longer survival time in the subgroups of GC patients without lymph node metastases or distant metastasis, suggesting that the low expression of ZHX2 and ZHX3 predicts a better prognosis for early-stage gastric cancer patients. Subgroup analyses also found that the downregulation of ZHX2 was predictive of an improved survival time in HER2-positive gastric cancer patients, but not in HER2-negative gastric cancer patients. In addition, the reduced expression of ZHX2 was associated with a favorable overall survival rate in gastric cancer patients who only received surgery, but not in the patients who received additional therapies [31]. Taken together, these data suggest that ZHX2 and ZHX3 act as oncogenes in gastric cancer.

The expression levels of ZHX2 in nature killer (NK) cells have shown an important clinical significance [65]. Kaplan–Meier analysis of overall survival from a TCGA dataset showed that a lower ZHX2 expression level in NK cells was associated with better prognosis in HCC patients. The genetical deleting ZHX2 gene improves IL-15-mediated NK cell activity, maturation, and cell viability in vitro, resulting in better antitumor immunity. Therapeutically, the transfer of ZHX2-deficient NK cells inhibited hepatoma homograft tumor growth and metastasis. These studies reveal a novel regulatory function of ZHX2 in NK cell maturation and its therapeutic potential by enhancing NK cell-mediated cancer surveillance [65].

#### *2.3. ZHX3 and Cancer*

The biological functions of ZHX3 in cancer have not been well characterized, but limited studies have shown that ZHX3 is a tumor suppressor in HCC, non-small cell lung cancer, breast cancer, and renal cancer (Table 3). It has also been reported that an increased ZHX3 expression is associated with the progression of bladder carcinoma and gastric cancer, indicating that ZHX3 is involved in oncogenic programs (Table 3). The clinical significance of ZHX3 and the underlying potential mechanism are summarized in Figure 3.

ZHX3 was first identified as a suppressor of the AFP gene in HCC and was a good candidate for the tumor suppressor present at 16q22 [16]. As with ZHX2, ZHX3 is expressed at very low levels in HCC cells compared with rat hepatocytes. ZHX3 repressed the transcription of the luciferase reporter gene that was fused to the promoters of PKM and HK2 [16]. This suggests that the loss of expression of ZHX3 might be a critical factor during hepatocellular carcinogenesis.

*Int. J. Mol. Sci.* **2022**, *23*, 11167


**Table 3.** Studies on ZHX3 expression, function, clinical significance, and molecular mechanism.

ZHX3 was first identified as a suppressor of the AFP gene in HCC and was a good candidate for the tumor suppressor present at 16q22 [16]. As with ZHX2, ZHX3 is ex‐ pressed at very low levels in HCC cells compared with rat hepatocytes. ZHX3 repressed the transcription of the luciferase reporter gene that was fused to the promoters of PKM ZHX3 expression was also remarkably decreased in non-small cell lung cancer (NSCLC) in comparison to the adjacent non-tumor tissues [66]. Interestingly, a lower ZHX3 expression in the tumor had a significantly greater risk of lymph node metastasis and was associated with a poorer survival time. Therefore, ZHX3 was an independent factor affecting metastasis and in predicting the 5-year overall survival rate among NSCLC patients.

and HK2 [16]. This suggests that the loss of expression of ZHX3 might be a critical factor during hepatocellular carcinogenesis. ZHX3 expression was also remarkably decreased in non‐small cell lung cancer (NSCLC) in comparison to the adjacent non‐tumor tissues [66]. Interestingly, a lower ZHX3 expression in the tumor had a significantly greater risk of lymph node metastasis and was associated with a poorer survival time. Therefore, ZHX3 was an independent factor affecting metastasis and in predicting the 5‐year overall survival rate among NSCLC patients. A recent study on renal carcinoma showed a tumor-suppressive role for ZHX3 [18]. As seen with ZHX1, ZHX3 showed a lower expression in cancers, and was associated with a high-risk of lymph node metastasis and worse outcomes. Further analysis of mRNA co-occurrence using cBioPortal data (www.cbioportal.org) showed opposite relationships between the expressions of ZHX1 and ZHX3 and some well-known oncogenes [18]. For instance, ZHX3 expression was negatively associated with the expressions of N-myc, STAT Interactor (NMI), and actin-related protein 2/3 complex subunit 5 (ARPC5). Both NMI and ARPC5 play critical roles in initiating cancer formation and facilitating cancer cell proliferation, migration, and metastasis [67,68].

A recent study on renal carcinoma showed a tumor‐suppressive role for ZHX3 [18]. As seen with ZHX1, ZHX3 showed a lower expression in cancers, and was associated with a high‐risk of lymph node metastasis and worse outcomes. Further analysis of mRNA co‐ occurrence using cBioPortal data (www.cbioportal.org) showed opposite relationships between the expressions of ZHX1 and ZHX3 and some well‐known oncogenes [18]. For instance, ZHX3 expression was negatively associated with the expressions of N‐myc, In breast cancer, a negative ZHX3 expression was correlated with lymph node metastasis, poor differentiation, advanced tumor stage, and positive estrogen receptor expression [19]. The immunohistochemical (IHC) staining analysis showed that the patients with decreased ZHX3 protein levels had a poor prognosis. The increased ZHX3 expression was associated with good outcomes in breast cancer patients, indicating that ZHX3 might act as a prognostic biomarker for breast cancer patients [19].

STAT Interactor (NMI), and actin‐related protein 2/3 complex subunit 5 (ARPC5). Both NMI and ARPC5 play critical roles in initiating cancer formation and facilitating cancer

pression [19]. The immunohistochemical (IHC) staining analysis showed that the patients

In breast cancer, a negative ZHX3 expression was correlated with lymph node me‐

cell proliferation, migration, and metastasis [67,68].

ZHX3 expression can also be used as a prognostic indicator for gastric cancer [31]. IHC staining confirmed that gastric cancer patients with a high ZHX3 expression have a worse prognosis [31].

In a recent study [20], ZHX3 was first screened in a urothelial carcinoma of the bladder (UCB) as a critical oncogenic factor associated with poor prognosis—this was based from The Cancer Genome Atlas dataset and a large cohort of UCB clinical samples. ZHX3 promoted the migration and invasion of UCB cell in vitro and in vivo. Mechanistically, ZHX3 repressed the expression of regulator of G protein signaling 2 (RGS2). Meanwhile, as a target, ZHX3 was regulated by tripartite motif 21 (TRIM21), which mediates its ubiquitination and subsequent degradation. These results indicate that ZHX3 is an oncogene and has a therapeutic potential for UCB [20].

#### *2.4. ZHXs and Cancer Patient Survival*

The association between ZHX family member expression levels and the survival time in multiple types of cancers were refined by us using the comprehensive datasets from the GEPIA (Gene Expression Profiling Interactive Analysis, http://gepia.cancer-pku. cn/). We selected overall survival (OS) or disease-free survival (DFS) as survival time indicators. Figure 4A shows that the cancer types with a low ZHX1, ZHX2, or ZHX3 expression have better OS or DFS, including ZHX1 in adrenocortical carcinoma and liver hepatocellular carcinoma, ZHX2 in bladder urothelial carcinoma, and ZHX3 in bladder urothelial carcinoma, esophageal carcinoma, uveal melanoma, mesothelioma, and ovarian serous cystadenocarcinoma. In contrast, the types of cancers with high ZHX1, ZHX2, or ZHX3 expressions have better OS or DFS, as shown in Figure 4B, including ZHX1 in cholangial carcinoma and kidney renal clear cell carcinoma, ZHX2 in brain lower grade glioma, mesothelioma and sarcoma, and ZHX3 in renal clear cell carcinoma and thyroid carcinoma. These results again have proven the roles of the ZHX genes as a double-edged sword in cancer patient survival.

(**A**)

**Figure 4.** *Cont*.

**Figure 4.** ZHX1, ZHX2, and ZHX3 gene expression level and survival analysis in 12 types of cancers. These panels were generated using the GEPIA database (http://gepia.cancer‐pku.cn/) on 06/02/2022. (**A**): Patients with low ZHX1, ZHX2, or ZHX3 expressions have better OS or DFS. The words in red stood for ZHX1, ZHX2, and ZHX3 high expressions, the words in blue stood for ZHX1, ZHX2, and ZHX3 low expressions. (**B**): Patients with high ZHX1, ZHX2, or ZHX3 expres‐ sions have better OS or DFS. GEPIA performs overall survival (OS) or disease‐free survival (DFS) analysis based on ZHX1, ZHX2, and ZHX3 gene expressions. The cox proportional hazard ratio and 95% confidence interval information can also be included in the survival plot. "Median" was used in group cutoff. Cutoff‐ high (%) indicates that samples with expression levels higher than this threshold are considered as the high‐expression cohort and samples with expression levels lower than this threshold are considered the low‐expression cohort. Cancer types of abbreviations: ACC, adrenocortical carcinoma; BLCA, bladder urothelial carcinoma; CHOL, cholangial carcinoma; ESCA, esophageal carcinoma; KIRC, kidney renal clear cell carcinoma; LGG, brain lower grade glioma; LIHC, liver hepatocellular carcinoma; MESO, mesothelioma; OV, ovarian serous cystadenocarcinoma; SARC, sarcoma; THCA, thyroid carcinnooma; UVM, uveal melanoma. **Figure 4.** ZHX1, ZHX2, and ZHX3 gene expression level and survival analysis in 12 types of cancers. These panels were generated using the GEPIA database (http://gepia.cancer-pku.cn/) on 06/02/2022. (**A**): Patients with low ZHX1, ZHX2, or ZHX3 expressions have better OS or DFS. The words in red stood for ZHX1, ZHX2, and ZHX3 high expressions, the words in blue stood for ZHX1, ZHX2, and ZHX3 low expressions. (**B**): Patients with high ZHX1, ZHX2, or ZHX3 expressions have better OS or DFS. GEPIA performs overall survival (OS) or disease-free survival (DFS) analysis based on ZHX1, ZHX2, and ZHX3 gene expressions. The cox proportional hazard ratio and 95% confidence interval information can also be included in the survival plot. "Median" was used in group cutoff. Cutoff-high (%) indicates that samples with expression levels higher than this threshold are considered as the high-expression cohort and samples with expression levels lower than this threshold are considered the low-expression cohort. Cancer types of abbreviations: ACC, adrenocortical carcinoma; BLCA, bladder urothelial carcinoma; CHOL, cholangial carcinoma; ESCA, esophageal carcinoma; KIRC, kidney renal clear cell carcinoma; LGG, brain lower grade glioma; LIHC, liver hepatocellular carcinoma; MESO, mesothelioma; OV, ovarian serous cystadenocarcinoma; SARC, sarcoma; THCA, thyroid carcinnooma; UVM, uveal melanoma.

#### **3. Conclusions and Perspectives 3. Conclusions and Perspectives**

Currently, studies on the ZHX family members in cancer have demonstrated tumorsuppressive and oncogenic roles in carcinogenesis and progression by power BI tools (Figure 5). These opposing effects are also seen in the roles of ZHXs in the prediction of outcomes and responses to therapy. These differences are most likely specific to cancer type. However, the studies are still limited to a few cancer types. The biological functions of ZHXs in cancers of the respiratory system, gastrointestinal tract, head and neck, and reproductive system are largely unknown, and thus, require further investigation. Moreover, neither ZHX-related signaling pathways nor upstream and downstream regulators in carcinogenesis and therapy are clear, and thus, need further elucidation. Currently, studies on the ZHX family members in cancer have demonstrated tumor‐ suppressive and oncogenic roles in carcinogenesis and progression by power BI tools (Fig‐ ure 5). These opposing effects are also seen in the roles of ZHXs in the prediction of out‐ comes and responses to therapy. These differences are most likely specific to cancer type. However, the studies are still limited to a few cancer types. The biological functions of ZHXs in cancers of the respiratory system, gastrointestinal tract, head and neck, and re‐ productive system are largely unknown, and thus, require further investigation. Moreo‐ ver, neither ZHX‐related signaling pathways nor upstream and downstream regulators in carcinogenesis and therapy are clear, and thus, need further elucidation.

*Int. J. Mol. Sci.* **2022**, *23*, x FOR PEER REVIEW 20 of 24

**Figure 5.** A comprehensive illustration of ZHXs as a double‐edged sword in cancers: an oncogene or tumor suppressor, underlying mechanisms, and biological functions. **Figure 5.** A comprehensive illustration of ZHXs as a double-edged sword in cancers: an oncogene or tumor suppressor, underlying mechanisms, and biological functions.

**Author Contributions:** Conceptualization, Y.G., Y.B. and W.Y.; Methodology, Y.G.; Data collection, Y.G.; Formal analysis, Y.G. and H.Z.; Data curation, Y.G. and Z.H.; Writing—original draft prepa‐ ration, Y.B. and W.Y.; Funding acquisition, Y.G., Y.B. and W.Y. All authors have read and agreed to the published version of the manuscript. **Author Contributions:** Conceptualization, Y.G., Y.B. and W.Y.; Methodology, Y.G.; Data collection, Y.G.; Formal analysis, Y.G. and H.Z.; Data curation, Y.G. and Z.H.; Writing—original draft preparation, Y.B. and W.Y.; Funding acquisition, Y.G., Y.B. and W.Y. All authors have read and agreed to the published version of the manuscript.

**Funding:** This project was funded in part by the Doctoral Research Startup Fund from Mudanjiang Medical University (2021‐MYBSKY‐064 to Y.B. and 2021‐MYBSKY‐065 to Y.G.), and by the Cancer Center Support Grant from the National Cancer Institute (P30 CA016058) and the Clinical and Translational Science Shared Resource at Ohio State University Comprehensive Cancer Center. **Funding:** This project was funded in part by the Doctoral Research Startup Fund from Mudanjiang Medical University (2021-MYBSKY-064 to Y.B. and 2021-MYBSKY-065 to Y.G.), and by the Cancer Center Support Grant from the National Cancer Institute (P30 CA016058) and the Clinical and Translational Science Shared Resource at Ohio State University Comprehensive Cancer Center.

**Institutional Review Board Statement:** Not applicable. **Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable. **Informed Consent Statement:** Not applicable.

**Data Availability Statement:** Not applicable. **Data Availability Statement:** Not applicable.

**Conflicts of Interest:** The authors declare no conflict of interest. **Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**

