A DAAM1 3′-UTR SNP mutation regulates breast cancer metastasis through affecting miR-208a-5p-DAAM1-RhoA axis

Background Dishevelled-associated activator of morphogenesis 1 (DAAM1) is a member of microfilament-related formins and mediates cell motility in breast cancer (BrCa). However, the genetic mutation status of DAAM1 mRNA and its correlation with pathological characteristics are still unclearly. Methods: A patient cohort and BrCa cells were recruited to demonstrate the role of functional SNP in microRNA-208a-5p binding site of DAAM1 3′-UTR and underlying mechanism in BrCa metastasis. Methods A patient cohort and BrCa cells were recruited to demonstrate the role of functional SNP in microRNA-208a-5p binding site of DAAM1 3′-UTR and underlying mechanism in BrCa metastasis. Results The expression and activation of DAAM1 increased markedly in lymphnode metastatic tissues. A genetic variant (rs79036859 A/G) was validated in the miR-208a-5p binding site of DAAM1 3′-UTR. The G genotype (AG/GG) was a risk genotype for the metastasis of BrCa by reducing binding affinity of miR-208a-5p for the DAAM1 3′-UTR. Furthermore, the miR-208a-5p expression level was significantly suppressed in lymphnode metastatic tissues compared with that in non-lymphnode metastatic tissues. Overexpression of miR-208a-5p inhibited DAAM1/RhoA signaling pathway, thereby leading to the decrease of the migratory ability. Conclusion Overall, the rs79036859 G variant of DAAM1 3′-UTR was identified as a relevant role in BrCa metastasis via the diversity of miR-208a-5p binding affinity. Electronic supplementary material The online version of this article (10.1186/s12935-019-0747-8) contains supplementary material, which is available to authorized users.


Background
Dishevelled-associated activator of morphogenesis 1 (DAAM1) is a member of microfilament-related formins and is involved in cell motility through mediating Wnt signaling pathway [1][2][3]. In the cytoplasm, DAAM1 exists in an autoinhibited state by intramolecular interaction between its N-terminal GBD and C-terminal DAD domains. When Dishevelled 2 binds to DAAM1, leading to disrupting the interaction between the GBD and DAD that mediates DAAM1 auto-inhibition, DAAM1 will exposure FH1 and FH2 domains to polymerize straight actin filaments [2,4]. DAAM1 is essential for Wnt-11/ Frizzled-induced activation of RhoA and Xenopus gastrulation [2]. DAAM1 directly collaborates to fascin in actin filaments and thus controls the formation of filopodia [5]. Our previous studies find that active DAAM1 is involved in Wnt5a/Dishevelled 2 and Collagen/Integrin αvβ3 signaling pathways, resulting in the elevation of the migratory and haptotatic ability of breast cancer (BrCa) cells [3,6]. However, the genetic mutation status of DAAM1 mRNA and its correlation with pathological characteristics are still unknown in BrCa patients.
Single nucleotide polymorphisms (SNPs) located in untranslated region (UTR) have been reported to be associated with dysregulation of genes expression. A recent global transcriptional network study identifies mutations at somatic expression quantitative trait locus (eQTL) located 5′-UTR of DAAM1, directly regulating the expression of DAAM1 mRNA [7]. Nevertheless, noncoding mutations in the 3′-UTR of DAAM1 has not been reported. An increasing evidence revealed that functional 3′-UTR SNPs are participated in the progression of multiple cancers [8][9][10][11]. Most 3′-UTR of mRNAs function as the target sequences of microRNAs (miRNAs) by base pairing, thereby leading to the degradation of mRNAs and decrease of their translation [12,13]. SNPs in the miRNA binding sites in the 3′-UTRs of target genes represent their differential binding affinities for corresponding miRNAs [14][15][16].
Here, we demonstrate that DAAM1 is highly expressed in lymphnode metastatic BrCa tissues. We also elucidate the functional role of SNP rs79036859 in the miR-208a-5p binding site of DAAM1 3′-UTR and its involvement in BrCa metastasis. Overall these data identify miR-208a-5p/DAAM1 axis as a potential therapeutic target in limiting BrCa metastasis and reducing death from this disease.

Clinical information
157 BrCa patients were recruited by the First Affiliated Hospital with Nanjing Medical University and Affiliated Cancer Hospital to Nanjing Medical University (NJMU) from 2015 to 2018. All cases had been diagnosed with BrCa by a pathologist on the basis of hematoxylin-eosin (HE) staining. Relevant clinicopathological characteristics record for each case were collected by review of patients' medical files. Ethical approval for the study was obtained from the Clinical Research Ethics Committee, NJMU. Pathologic staging was determined by a pathologist based on AJCC Cancer Staging Manual 8th classification criteria. All the participants voluntarily joined this study with informed contents.

Selection of SNPs and TaqMan genotyping
A next-generation sequencing of metastatic BrCa tissues (data not published) revealed some SNPs, including rs79036859 and rs45476291, locating in the 3′-UTR of DAAM1. After a review of dbSNP database (https :// www.ncbi.nlm.nih.gov/snp) and PolymiRTS Database 3.0 (http://compb io.uthsc .edu/miRSN P), we selected the candidate SNP (rs79036859) suggested as a transcriptional regulation factor for the 3′-UTR of DAAM1. SNP genotyping was conducted by allelic discrimination using the TaqMan SNP Genotyping Assays according to the manufacturer's instructions (Applied Biosystems). Specific primers (TAT CTC CTG AAA GAG ATA AGA, GTT TTT CCA ACA ACT CCA GT) and FAM/VIC-labeled TaqMan probes (FAM-labeled CAA ACA AAC AAA AAA AGCT TGC AAA ATA TTTT, VIC-labeled CAA ACA AAC AAA AAA GGCT TGC AAA ATA TTTT) were designed and supplied by Synbio Technologies (Soochow, China). The PCR conditions were as follows: initiation at 98 °C for 10 min, followed by 40 cycles of denaturation at 95 °C for 30 s and annealing/extension at 60 °C for 60 s. PCR application was undergoing in a StepOnePlus Real-Time PCR System (Applied Biosystems).
To detect the active level of DAAM1, we employed GST-RhoA beads as bait. The activate level of DAAM1 was detected by the Pulldown assays and subsequently immunoblotted with anti-DAAM1 antibody (Cat. 14876-1-AP, ProteinTech) [6]. SDS-PAGE and Western blotting were performed using the above methods.

Boyden chamber assays
Migratory ability of cancer cells was tested in a modified Boyden chamber (Cat. 3422, Costar, Corning, NY). The detail protocol was described as previously [6].

Immunofluorescence and actin cytoskeleton staining
Cells were placed on glass cover-slides and subjected to actin cytoskeleton staining. The detail protocol was described as previously [17]. The images were captured by a laser scanning confocal microscope (Zeiss LSM710, Oberkochen, Germany).

RhoA GTPase activation assays (G-LISA small GTPase activation assays)
Total protein lysates extracted from BrCa cells were turned to measure RhoA activity by using RhoA GTPase activation assays (Cat. BK121, Cytoskeleton Inc., Denver, CO). RhoA activation was described as previously [6].

Kaplan-Meier plotter analysis
DAAM1 (Affy ID: 216060_s_at) mRNA expression data and survival information (progression free survival, overall survival, post progression survival and distant metastasis free survival) were collected from Kaplan-Meier plotter (http://kmplo t.com). Then, the prognostic significance of DAAM1 mRNA was analyzed in BrCa. Kaplan-Meier survival plots were generated with survival curves compared by log-rank test.

Statistical analysis
All statistical analyses were calculated using SPSS 25.0 software (Chicago, IL). Most of the data were analyzed by Student's t-test or one-way ANOVA followed by Dunnett's multiple posthoc tests. All data are presented as means ± SDs of five independent experiments if not noted. The associations between DAAM1 mRNA expression levels, genotypes distribution of DAAM1 SNP, along with miR-208a-5p expression status and clinicopathological characteristics were performed using Pearson's Chi squared test. The correlations between DAAM1 mRNA expression and miR-208a-5p expression or between DAAM1 mRNA level and protein level were assessed by Spearman's correlation analysis.

DAAM1 expression positively correlates with lymphnode metastasis and associates with prognosis in BrCa
Our previous studies reported that DAAM1 regulates the re-organization of microfilaments for oriented the migration and haptotaxis of BrCa cells [3,6]. However, it is still limited evidence demonstrating that DAAM1 is correlated with tumor metastasis in BrCa patients. To investigate whether DAAM1 acts as a metastatic promoter in BrCa, the immunohistochemistry (IHC) were performed to test the DAAM1 expression in 46 BrCa cases. The sections highly and moderately expressed DAAM1 accounted for a majority of lymphnode metastatic cases (Fig. 1a, b).
Because of limited amount of IHC sections, we further examined the transcriptional level of DAAM1 in 157 BrCa samples. We firstly determined that DAAM1 mRNA expression was positively correlated with DAAM1 protein expression (Fig. 1c). When we compared DAAM1 mRNA expression in lymphnode metastatic tumor tissues and non-lymphnode metastatic tumor tissues along with distance metastatic tumor tissues and non-distance metastatic tumor tissues, the expression of DAAM1 mRNA was significantly increased in metastatic tissues (Fig. 1d, e, Table 1). Furthermore, we tested DAAM1 expression and DAAM1 activity in fresh BrCa tissues (Fig. 1f ). We found that both the expression and the activation of DAAM1 in lymphnode metastatic tumor tissues were significantly higher than these in nonlymphnode metastatic tumor tissues (Fig. 1f, g).
Moreover, both progression free survival (PFS) and overall survival (OS) rate of BrCa patients with high expression of DAAM1 mRNA was significantly worse than that of patients with low DAAM1 expression according to survival information from Kaplan-Meier plotter (Fig. 1h, i). Overall, the clinical data reveals that DAAM1 functions as a metastatic promoter and associates with prognosis in BrCa.

G genotype of rs79036859 enhances the risk of metastasis in BrCa
In view of a fact that DAAM1 showed the increased expression and activation level in lymphnode metastatic tumor tissues, we speculated that the regulation of genomic and/or transcriptional level may be heterogeneous in individual BrCa tissues. After a review of dbSNP database and PolymiRTS Database 3.0 [18], SNP rs79036859 was predicted as the regulatory element for the post-transcription of DAAM1 expression (Fig. 2a).
To further verify the potential effect of the DAAM1 3′-UTR SNP rs79036859 in BrCa, we performed a casebased method to assess the function of rs79036859 in tumor metastasis. A metastatic cases-control study including 157 BrCa cases (121 cases without lymphnode metastasis and 36 cases with lymphnode metastasis) was proceeded. BrCa patients with the G (AG/GG) allele genotypes had a prominently higher risk of metastasis compared with the AA genotype carriers (OR = 3.90, 95% CI 1.77-8.59; P < 0.001; Table 2). These findings demonstrate that G genotype in rs79036859 indicates a high risk of metastasis in BrCa.
(See figure on next page.) Fig. 1 The expression of DAAM1 are relative with the high rate of lymphnode metastasis and associated with prognosis in BrCa. a Representative microphotographs revealing low, medium, and high DAAM1 expression using immunohistochemical (IHC) staining. Brown, DAAM1. Blue, haematoxylin. Bar = 100 μm. Objective lens, magnification, ×10; numerical aperture, 0.75. b DAAM1 protein expression intensity proportion based on lymphnode metastasis. N0, without lymphnode metastasis. N1-N3, with lymphnode metastasis. c Correlation between the expression levels of DAAM1 protein and DAAM1 mRNA in BrCa tissues. d, e The expression of DAAM1 mRNA according to lymphnode metastasis and distance metastasis. M0, without distance metastasis. M1, with distance metastasis. f The lysates of tumor tissues were assayed for the expression levels of total and active DAAM1 by Western blotting and pulldown assay using a GST-RhoA as a bait, respectively. β-actin was used as the loading control.
To assess whether rs79036859 A/G was related to DAAM1 expression in clinical samples, we detected DAAM1 mRNA levels in cancer tissues using quantitative PCR in 157 cases and examined DAAM1 protein expression by IHC in 46 patients. The results showed that DAAM1 mutant tumor tissues (AG/GG genotypes) expressed higher levels of DAAM1 mRNA and protein than DAAM1 wildtype tumor tissues (AA genotype) (Fig. 3e, f ). MiR-208a-5p levels were measured in the above cancer tissues and shown non-significant differences between the wildtype and mutant genotype groups (Fig. 3g). Taken together, these results demonstrated that rs79036859 in DAAM1 3′-UTR decreases the binding affinity with miR-208a-5p and eliminates miR-208a-5pmediated decrease of DAAM1 expression in mammalian cells.

MiR-208a-5p inhibits the RhoA activity and disrupts the formation of microfilament in BrCa cells
DAAM1 is reported to be involved in Wnt5a-induced and collagen-induced signaling pathways [3,6]. RhoA is a direct downstream target of DAAM1, which functions in actin assemblage [1,3,6]. We investigated the fact that whether miR-208a-5p inhibited the activation of RhoA and the inhibition was reversed by the transfection of miR-208a-5p inhibitor (anti-sense of miR-208a-5p). MiR-208a-5p overexpression notably suppressed RhoA activation, that could be reversed by miR-208a-5p inhibitor overexpression (Fig. 5a). Next, fluorescent phalloidin was stained to display the arrangement of microfilaments in BrCa cells. MiR-208a-5p overexpression obviously blocked the microfilament formation, and miR-208a-5p inhibitors largely enhanced the microfilament assemblage in MDA-MB-231 cells (Fig. 5b-d). Thus, the findings illuminated that active RhoA induced formation of microfilament is disrupted by miR-208a-5p through downregulating DAAM1. Collectively, the findings suggested that the formation of microfilament is disrupted by miR-208a-5p overexpression via targeting DAAM1/ RhoA signaling in BrCa cells (Fig. 5e).

Discussion
The first finding in this study is that the high level of DAAM1 expresses in lymphnode metastatic BrCa tissues. These clinicopathological and biochemical characteristics imply DAAM1 is likely to function as a metastatic promoter in BrCa, which is consistent with our previous findings verified on the cellular levels [3,6,17]. Protein is translated from RNA, which is strictly conformed to the genetic central dogma [19]. The mRNA modification exists extensively in eukaryotic cells and leads to the diversity of protein expression. Zhang et al. has analyzed the big data from the Cancer Genome Atlas (TCGA) and identified novel mutations at eQTL situated in DAAM1 promotor (− 91 bp), which are able to increase DAAM1 mRNA expression levels [7]. In this study, we focus on the polymorphism of 3′-UTR of DAAM1 gene. Patients with the G (GG/GA) allele genotypes have the prominently higher risk of metastasis in BrCa than AA genotype of SNP rs79036859. These results demonstrate that G genotype of rs79036859 in DAAM1 3′-UTR enhances the risk of metastasis in BrCa.
The most common post-transcriptional regulation is that miRNAs bind to 3′-UTR of target gene and induce the degradation of whole mRNAs. We find that miR-208a-5p targets to the 3′-UTR of DAAM1 gene. Wang et al. reported the downregulation of miR-208a/miR-208b significantly suppresses the harmful effect of extracellular vesicles on hypoxia/reoxygenation damage in cardiac myoblasts [20]. The regulatory feedback loop of miR-208a-SOX2/β-catenin-LIN28-let-7a-DICER1 mediates the renovation of BrCa stem cells [21]. MiR-208a is capable of promoting gastric cancer progression by suppressing SFRP1 and downregulating MEG3 [22]. Expression analysis demonstrate that miR-208a-5p is significantly suppressed in lymphnode metastatic tumor samples than that in non-lymphnode metastatic tumor samples, indicating the potential role of miR-208a-5p on anti-tumor metastasis via DAAM1 signaling.
DAAM1, an element of cellular actin cytoskeleton, transduces Wnt/Dishevelled signaling to RhoA and then assemble microfilaments in mammalian cells [2,3,23]. Actin polymerization is the key step of the assemblage of microfilament in the process of cancer metastasis [24]. The functional and dynamic microfilaments are generally provided a possible route for the degradation of extracellular matrix and available for cancer cells escaping from the primary lesion [24,25]. Our previous studies find that active DAAM1 is elevated by the treatment of Wnt5a or type IV collagen and participates in the tumor cell migration and haptotaxis [3,6]. Here, we demonstrate that miR-208a-5p overexpression decreases DAAM1 mRNA and protein expression levels, resulting in the decrease of the migratory ability of BrCa cells. We also find that miR-208a-5p downregulates the RhoA activity, disrupts the formation of microfilaments of BrCa cells. We summarize the potential DAAM1 signaling pathways in Fig. 5e. Thus, miR-208a-5p/ DAAM1/RhoA axis is the potential therapeutic target in regulating microfilament assemblage and cell migration.

Conclusions
These results indicate that SNP rs79036859 G variant of DAAM1 3′-UTR contributes to the likelihood of BrCa metastasis via miR-208a-5p binding capacity. Cell migration is inhibited by miR-208a-5p overexpression via targeting DAAM1/RhoA signaling in BrCa cells. Overall these data identify miR-208a-5p/ DAAM1/RhoA axis as the novel therapeutic target in limiting BrCa metastasis.