兽研简要总结
乳腺肿瘤在雌性犬中常见，雌激素在其生长和行为中可发挥重要作用。一些肿瘤的雌激素受体表达降低，但其原因尚未完全明了。在本研究中，我们检测了两种小调控分子miR-18a和miR-19a是否可通过犬ESR1 3′UTR的选定区域影响报告基因活性。微小RNA可结合信使RNA并降低基因活性。我们发现miR-18a和miR-19a均可降低含有犬ESR1 3′UTR选定区域的报告基因活性。当预测的miRNA靶区域被删除时，未观察到该效应。阻断这些miRNA也增加了报告基因活性。这些发现表明miR-18a和miR-19a可通过犬ESR1 3′UTR的选定区域影响报告基因活性，并为犬乳腺癌症的进一步研究提供了基础。
关键词：犬乳腺肿瘤，miR-18a，miR-19a，雌激素受体1（ESR1），荧光素酶报告基因试验，3′非翻译区，转录后调控
1. 引言
乳腺癌是女性中最常被诊断的癌症之一，乳腺肿瘤在未绝育雌性犬中也很常见[1,2,3]。两个物种共有若干重要的生物学和临床特征，包括自发性肿瘤发生、显著的异质性和激素依赖性生长[4]。由于这些相似性，犬乳腺肿瘤（CMTs）常被用作自发性比较模型，以理解女性乳腺肿瘤发生和激素驱动的癌症进展[2,5]。
雌激素受体有两种类型——雌激素受体α（ERα）和雌激素受体β（ERβ），分别由ESR1和ESR2基因编码。ERα和ERβ具有一些相似功能，但根据组织和细胞类型的不同，它们也可产生不同效应。在某些组织中，ERβ可拮抗ERα介导的信号传导[6]。ERα和ERβ均已在犬乳腺组织和肿瘤中检测到，尽管其表达在正常、良性和恶性组织中存在差异[7,8]。本研究聚焦于由ESR1编码的ERα。当雌二醇与ERα结合时，形成二聚体并进入细胞核，在核内调控参与细胞生长、分化和存活的基因。基于ERα、PR和HER2表达谱，CMTs已被分为类似luminal A、luminal B、HER2阳性和三阴性乳腺癌的表型[3,9]。然而，这些犬分组可能并不完全对应于已建立的人类乳腺癌亚型。在人类乳腺癌中，表达ERα的肿瘤通常对内分泌治疗反应更好，包括芳香化酶抑制剂和选择性雌激素受体调节剂如letrozole和tamoxifen[10,11]。ERα表达降低或异常的肿瘤往往更具侵袭性，并在两个物种中与较差预后相关[10,12]。
在人类乳腺癌中，ERα缺失可通过多种机制发生，包括启动子甲基化、ESR1突变和转录后调控。然而，CMTs中ERα下调的机制尚不清楚，犬ESR1的转录后调控仍不明确。
微小RNA（miRNA）是小型、高度保守的非编码RNA，长约18–25个核苷酸，通过结合靶mRNA 3′UTR中的互补序列在转录后调控基因表达[13,14]。异常miRNA表达在许多癌症中常见，包括CMTs，并可影响细胞生长、凋亡、侵袭和转移[5,14,15,16]。其中，miR-18a和miR-19a——均为miR-17–92簇的一部分——与乳腺肿瘤进展相关[17,18]。在人类乳腺癌和CMTs中，较高的miR-18a表达与ESR1/ERα表达降低相关[19,20,21]。miR-19a也被报道在早期乳腺癌和CMTs中升高[15,22,23]。若干人类乳腺癌研究表明，miRNA介导的ESR1抑制可能促进内分泌治疗耐药[21,24,25]。由于CMTs中的miRNA表达模式与人类乳腺癌中报道的模式具有相似性，miR-18a和miR-19a可能同样影响犬的雌激素受体信号传导[5,26]。
我们基于研究组先前研究的发现选择了miR-18a和miR-19a。这两种miRNA均通过细胞系和外泌体miRNA谱分析被鉴定，并且与正常犬乳腺上皮细胞（CMECs）相比，在CMT细胞及其分泌的外泌体中水平更高[15,22]。在随后一项针对患有乳腺癌的犬的临床研究中，循环miR-18a被确定为潜在的预后生物标志物，较高水平与淋巴管侵袭相关[18]。生物信息学分析还确定犬ESR1转录本为这两种miRNA的预测靶标[22,27]。然而，表达模式和靶标预测并不能确立功能性调控相互作用。据我们所知，这些miRNA对含有犬ESR1 3′UTR区域的报告基因的影响此前尚未被报道。
在本研究中，我们检测了 miR-18a 和 miR-19a 是否可通过犬 ESR1 3′UTR 的选定区域影响乳腺肿瘤细胞中的报告基因活性。我们测定了 ESR1、miR-18a 和 miR-19a 的内源性表达。使用包含选定野生型 ESR1 3′UTR 区域的荧光素酶报告基因检测以及相应的突变型 ESR1 3′UTR 报告基因（在预测的 miR-18a 和 miR-19a 种子匹配区域内具有缺失）来评估 miRNA 依赖性报告基因活性。我们还使用了 miRNA 表达载体和 anti-miR 来评估这些效应的特异性。
2. 材料与方法
2.1. 细胞培养
本研究使用了犬乳腺肿瘤（CMT）细胞系 CMT12、CMT27、CMT28 和 CMT47。选择这些细胞系是因为它们是常用的 CMT 模型，并已用于先前的 miRNA 表达研究 [15,22]。这些细胞系最初来源于不同犬的乳腺肿瘤，如前所述 [28,29]。CMT12 来源于贵宾犬的腺癌，CMT27 来源于德国牧羊犬的癌，CMT28 来源于混种犬的腺癌，CMT47 来源于迷你雪纳瑞的腺癌 [22,28]。犬乳腺上皮细胞（CMEC）作为正常乳腺上皮参考细胞，来源于一只无乳腺癌病史的雌性比格犬尸检时的乳腺组织 [29,30]。CMT 细胞系使用于第 20 至 50 代之间。CMEC 代表单一原代细胞培养物，使用于第 15 代以下。我们使用原始冻存库存的标签和记录来鉴定每个细胞系。培养期间定期检查细胞形态，每个细胞系均保持其独特外观。本研究中未进行额外的分子鉴定。细胞在补充有 L-谷氨酰胺（Quality Biological，Gaithersburg，MD，USA）、10% 胎牛血清（Atlas Biologicals，Fort Collins，CO，USA）和 1% 青霉素-链霉素（Sigma-Aldrich，St. Louis，MO，USA）的 Dulbecco 改良 Eagle 培养基（DMEM）中生长，并维持于 37 °C、5% CO2 的加湿培养箱中。细胞在达到约 70–80% 汇合度后常规传代。所有细胞均在标准透气盖 75 cm2 组织培养瓶（VWR Avantor，Radnor，PA，USA）中操作。细胞培养物在冻存时通过电子显微镜检查支原体污染。本研究期间未进行常规支原体检测。涉及犬细胞系的工作按照奥本大学机构动物护理和使用委员会批准的方案进行。
2.2. RNA 提取
按照制造商说明，使用 miRNeasy 试剂盒配合 QIAzol 裂解试剂和无 RNase 的 DNase（Qiagen，Germantown，MD，USA）提取总 RNA。细胞在 70–80% 汇合度时收集，直接在 QIAzol 中裂解。RNA 使用 miRNeasy 旋转柱纯化，用无 RNase 水洗脱，并储存于 −80 °C 直至进一步使用。使用 NanoDrop 分光光度计（Thermo Fisher Scientific，Waltham，MA，USA）评估 RNA 浓度和纯度，仅 A260/A280 比值在 1.8 至 2.0 之间的样本用于 cDNA 合成和 qPCR。
2.3. cDNA 合成
对于 mRNA 靶标，包括 ESR1 和参考基因 GAPDH，使用 High-Capacity RNA-to-cDNA 试剂盒（Thermo Fisher Scientific，Waltham，MA，USA）合成 cDNA。每个反应包含 1 µg 总 RNA，于 37 °C 孵育 60 分钟，随后在 95 °C 热灭活 5 分钟。所得 cDNA 以未稀释形式储存于 −20 °C，并在 qPCR 前立即进行两倍稀释。
对于 miRNA 分析，使用 Qiagen miRCURY LNA miRNA RT 试剂盒合成 cDNA。每个反应包含约 200 ng 总 RNA，于 42 °C 孵育 60 分钟，随后在 95 °C 热灭活 5 分钟。所得未稀释 cDNA 储存于 −20 °C，并在 qPCR 前立即进行 60 倍稀释。
2.4. RT-qPCR
使用上述稀释的 cDNA，mRNA 和 miRNA 靶标的 RT-qPCR 反应分别在 Bio-Rad CFX96 实时 PCR 检测系统（Bio-Rad Laboratories，Hercules，CA，USA）上进行。每个反应设双复孔。所有运行均包含无逆转录（no-RT）和无模板对照，以评估基因组 DNA 和试剂污染。
为评估 ESR1 的 mRNA 表达，使用 PowerUp SYBR Green Master Mix（Thermo Fisher Scientific，Waltham，MA，USA）配合基因特异性引物（表1）配制反应体系。GAPDH 引物序列基于既往发表的研究[9,31]，而 ESR1 引物则使用 NCBI Primer-BLAST 设计，以确保扩增子跨越外显子-外显子连接处。所有引物均由 IDT（Integrated DNA Technologies，Coralville，IA，USA）合成。ESR1 扩增子通过常规 PCR 后进行 Sanger 测序（MGH DNA Core，Massachusetts General Hospital，Boston，MA，USA）进行序列验证，以确认靶标特异性。ESR1 的退火条件通过系列稀释曲线进行优化，最终反应在 60.5 °C 下运行，获得干净、特异的扩增和一致的 Ct 值。反应按以下循环条件运行：50 °C 2 分钟，95 °C 2 分钟，随后 36 个循环的 95 °C 1 秒和 60.5 °C 30 秒。包含熔解曲线步骤以确认扩增子特异性。GAPDH 用作内源性参照基因以标准化 ESR1 mRNA 表达。我们选择 GAPDH 是因为其已在既往犬乳腺肿瘤细胞系研究中使用[31]。然而，我们未使用 geNorm 或 NormFinder 正式比较其与其他参照基因的稳定性。
Table 1. Primer sequences and amplicon sizes used for canine mRNA RT-qPCR.
| Canine Gene | Forward Primer (5′ → 3′) | Reverse Primer (5′ → 3′) | Amplicon Size (bp) |
|---|---|---|---|
| GAPDH | ATGGTGAAGGTCGGAGTCAA | ATCACCCCATTTGATGTTGG | 261 |
| ESR1 | ATGTGTCCAGCTACCAACC | GTCTTTCCGTATCCCGCC | 117 |
对于 miRNA 表达，使用 miRCURY LNA miRNA PCR Kit（Qiagen，Germantown，MD，USA）配合 cfa-miR-18a-5p（Assay ID：YP00204207）、cfa-miR-19a-3p（Assay ID：YP00205862）和 U6 snRNA（Assay ID：YP02119464）的 LNA 引物检测进行 RT-qPCR。循环条件为 95 °C 2 分钟，随后 40 个循环的 95 °C 10 秒和 56 °C 1 分钟。U6 snRNA 用作 miRNA 标准化的内源性参照，如既往 CMT 研究[32]所报道。ΔCt 值相对于 U6 Ct 计算。相对表达使用 ΔCt 法计算，倍数变化值使用 2−ΔΔCt 法得出。
扩增效率由系列稀释 cDNA 生成的标准曲线斜率计算。扩增效率为：GAPDH 104.6%，ESR1 105.3%，miR-18a 105.1%，miR-19a 90.4%，U6 91.9%。相应的标准曲线及 R2 值见补充图 S1 和 S2。
2.5. miRNA 结合位点的计算机预测
在设计荧光素酶构建体之前，使用 miRDB 数据库（http://mirdb.org；首次访问于 2023 年 6 月 24 日，并于 2023 年 9 月 14 日复核）检查犬 ESR1 3′UTR 中预测的 miRNA 结合位点。miRDB 预测评分范围为 50 至 100，评分高于 80 通常被认为是高置信度靶标[27]。cfa-miR-18a 和 cfa-miR-19a 均出现在预测调控因子前列，预测评分分别为 99 和 96。还使用 TargetScanHuman 8.0（TargetScanHuman 8.0；首次访问于 2023 年 6 月 24 日，并于 2023 年 9 月 14 日复核）检查犬 ESR1 3′UTR（NM_001286958.2）中预测的 miR-18a 和 miR-19a 种子匹配位点的位置和保守性[33,34]。选择犬作为物种，分别搜索 miR-18 和 miR-19。位点位置和保守性分类直接取自 TargetScan，并报告为“保守”或“保守性差”。
基于这些结果，选择含有预测结合位点的 ESR1 3′UTR 片段克隆入荧光素酶报告载体，以实验评估其调控活性。
2.6. 质粒和 miRNA 构建体
本研究中使用的 ESR1 3′UTR 报告构建体和前体 miRNA 表达载体购自 GeneCopoeia（Rockville，MD，USA）。含有选定野生型 ESR1 3′UTR 区域的报告构建体（CS-miT740T-MT05-01）包含一段 859 bp 的插入片段，该片段源自犬 ESR1 3′UTR（NM_001286958.2），克隆于 pEZX-MT05 载体中 Gaussia 荧光素酶报告基因下游。该插入片段包含三个选定的非重叠 ESR1 3′UTR 区域，含有预测的 miR-18a 和 miR-19a 种子匹配位点。片段1为 308 bp，含有预测的 miR-18a 结合区域；片段2为 307 bp，含有预测的 miR-19a 结合区域；片段3为 244 bp，含有两个 miRNA 的重叠预测结合区域。以下将该构建体称为 ESR1-WT 报告体。“ESR1-WT”指所选区域的未修饰序列，并不表示完整的天然 ESR1 3′UTR。
使用相应的突变型 ESR1 3′UTR 报告构建体（CS-miT740T-MT05-02）作为阴性对照。相对于 ESR1-WT 报告体，该构建体在所选 ESR1 3′UTR 区域内含有三个缺失：报告插入片段核苷酸 151–158 处的 GCACCTTA、核苷酸 459–465 处的 TTTGCAC，以及核苷酸 766–772 处的 GCACCTT。这些缺失去除了三个预测的 miRNA 种子匹配位点，并将插入片段长度从 859 bp 缩短至 837 bp。以下将该构建体称为 ESR1-MUT 报告体。
将CMT28细胞用针对cfa-miR-18a（CS-miR738T-MR04-01）和cfa-miR-19a（CS-miR739T-MR04-01）的前体miRNA载体转染，用于报告基因检测。miR-18a载体含有eGFP报告基因，而miR-19a载体含有mCherry报告基因。使用含有eGFP的GeneCopoeia miRNA对照载体（miR-C；CmiR0001-MR04）作为miRNA载体转染的阴性对照。前体miRNA和荧光报告蛋白由同一载体在同一启动子下表达。因此，GFP和mCherry荧光提供了相应miRNA载体表达的间接证据，但不能直接证实成熟miR-18a或miR-19a水平的升高。对于回补实验，使用含有反义序列的miArrest™合成miRNA抑制剂（GeneCopoeia），靶向miR-18a-5p（HmiR-AN0255-SN-5）和miR-19a-3p（HmiR-AN0293-SN-5）。两种抑制剂均为单链、2′-O-甲基修饰的RNA寡核苷酸。尽管这些抑制剂是针对人miR-18a-5p和miR-19a-3p设计的，但这两种miRNA的成熟序列在人和犬中完全相同。因此，这些抑制剂也与cfa-miR-18a-5p（MIMAT0009832）和cfa-miR-19a-3p（MIMAT0006650）互补。回补实验中未包含非靶向抗miR对照。关于报告基因构建体、miRNA表达载体、抑制剂和ESR1 3′UTR序列的更多信息见补充表S1–S3。
2.7. 质粒制备
按照制造商说明，将ESR1 3′UTR报告基因构建体和miRNA表达质粒转化至Mix & Go! DH5α感受态大肠杆菌（Zymo Research，美国加利福尼亚州尔湾）。转化后的菌落于含氨苄青霉素的2YT培养液中，在37 °C振荡培养过夜。按照制造商说明，使用ZymoPURE Midiprep试剂盒（Zymo Research，美国加利福尼亚州尔湾）纯化质粒DNA。质粒浓度通常为1.5至3 µg/µL，A260/A280比值在1.8至2.0之间。
2.8. 瞬时转染与荧光素酶报告基因检测
将CMT28细胞以约1.5 × 105个/mL（500 µL/孔）接种于24孔板中，使其达到60–80%汇合度。选择CMT28细胞进行功能性报告基因检测，是因为其内源性miR-18a和miR-19a表达相对较低，且基于其生长和培养特性非常适合可重复的转染。对于回补实验，首先使用Lipofectamine RNAiMAX（Thermo Fisher Scientific，美国马萨诸塞州沃尔瑟姆）将5 pmol抗miR寡核苷酸转染至细胞。约18 h后，将转染培养基更换为新鲜完全培养基。
初次抗miR转染24小时后，使用Lipofectamine 3000（Thermo Fisher Scientific，美国马萨诸塞州沃尔瑟姆）在Opti-MEM中单独转染0.5 µg报告质粒，或共转染0.25 µg报告质粒与0.25 µg相应miRNA表达载体或miR-C对照载体。因此，每孔质粒DNA总量维持在0.5 µg。
质粒转染48小时后，收集条件培养基，并使用GeneCopoeia双发光检测试剂盒测定Gaussia荧光素酶活性。使用Synergy HTX多模式微孔板读数仪（BioTek，美国佛蒙特州威努斯基；Gen5软件（版本2.09.1））记录发光。Gaussia荧光素酶活性以分泌型碱性磷酸酶（SEAP）活性进行归一化，以校正转染效率。所有转染和荧光素酶报告基因实验均在三次独立实验中进行，每次实验中每种条件设两个转染孔。每孔的发光值进行双复孔测定。
2.9. 荧光成像
使用EVOS荧光成像系统（Thermo Fisher Scientific，美国马萨诸塞州沃尔瑟姆）确认CMT28细胞转染miRNA表达载体后荧光报告蛋白的表达。分别使用GFP和RFP滤光块检测eGFP和mCherry荧光，物镜为10×。GFP图像在60%照明强度下拍摄，曝光时间为250 ms；RFP图像在60%照明强度下拍摄，曝光时间为750 ms。图像仅用于确认荧光报告基因的表达。
2.10. 统计分析
RT-qPCR实验在四次独立实验中进行，每次实验中每种条件设两个技术重复。统计分析前，在每个实验内对技术重复取平均值。统计分析在R（版本4.3）中进行。
对于qPCR实验，ESR1表达以GAPDH为内参进行归一化，而miR-18a和miR-19a表达以U6为内参进行归一化。ΔCt值计算为目标基因Ct值与相应内参基因Ct值之差。ΔCt值使用lmerTest包以线性混合效应模型进行分析，将细胞系作为固定效应纳入，将实验批次作为随机截距纳入。估计边际均值（EMMs）使用emmeans包获得，每个CMT细胞系使用Dunnett校正对比与CMEC进行比较。相对倍数变化根据模型估计的差异使用2−ΔΔCt方法计算，以CMEC作为参照。模型残差的正态性使用残差Q-Q图和Shapiro–Wilk检验评估，方差齐性使用Levene检验评估。
荧光素酶报告基因实验进行了三次独立运行，每次运行中每种条件设两个转染孔。每次独立实验运行在不同日期使用新的细胞培养板和单独的转染进行。每孔进行双复孔测量。不完整的GLuc/SEAP读数对予以排除，有效的技术测量值在统计分析前于每次运行内取平均值。独立实验运行作为统计单位（n = 3）。Gaussia荧光素酶活性以SEAP活性进行归一化，仅报告基因条件设为1作为参照。由于仅报告基因活性用于归一化，故未纳入统计比较。报告基因构建体（ESR1-WT或ESR1-MUT）、miRNA处理（miR-C、miR-18a或miR-19a）及其交互作用的效应使用双因素重复测量ANOVA分析，实验运行作为重复测量。每种miRNA处理与miR-C的比较使用双侧配对t检验进行。miR-18a和miR-19a的效应在ESR1-WT与ESR1-MUT之间也使用配对t检验进行比较。拯救实验使用双侧配对t检验分析，比较每种miRNA处理与相应的miRNA加anti-miR处理。两项ESR1-WT与ESR1-MUT比较使用Holm校正。报告效应估计值、95% CI和精确p值。模型残差使用Q-Q图和Shapiro–Wilk检验评估。p值 < 0.05视为具有统计学显著性。运行水平的报告基因数据和效应估计值及95% CI见补充表S4。
3. 结果
3.1. 犬乳腺细胞系
本研究纳入四种肿瘤来源的犬乳腺细胞系CMT12、CMT27、CMT28和CMT47，以及一个正常原代犬乳腺上皮细胞群。所有培养物均呈贴壁生长并具有 distinct 的细胞形态。代表性相差显微镜图像见图1。
[图：本研究所用犬乳腺细胞系的代表性相差显微镜图像。（A）CMEC，（B）CMT12，（C）CMT]
3.2. 犬乳腺细胞系中ESR1的表达
由于ESR1 mRNA表达在乳腺肿瘤间可能存在差异，我们首先以CMEC作为正常犬乳腺上皮参照，评估了CMT细胞系中内源性ESR1 mRNA表达。细胞系中ESR1 mRNA表达通过RT-qPCR评估，以GAPDH为内参基因。ΔCt计算为Ct（ESR1）–Ct（GAPDH）。ΔCt值使用混合效应模型分析，实验运行作为随机效应纳入。CMT细胞系与CMEC之间ESR1 mRNA表达存在显著差异（p < 0.001；图2A）。CMT12的ΔCt值低于CMEC，表明ESR1 mRNA表达较高（p = 0.0039），而CMT27、CMT28和CMT47的ΔCt值较高，表明ESR1 mRNA表达较低（分别为p = 0.0152、p = 0.0009和p = 0.0002）。倍数变化值（相对表达）使用2−ΔΔCt方法计算，以CMEC为参照。CMT12中ESR1 mRNA表达约为CMEC的14.9倍。在CMT27、CMT28和CMT47中，ESR1 mRNA表达分别约为CMEC的9倍、26倍和70倍降低（图2B）。总体而言，四种肿瘤来源细胞系中有三种的ESR1 mRNA表达远低于CMEC，而CMT12显示ESR1表达显著较高。
[图：犬乳腺细胞系中ESR1 mRNA表达差异很大。（A）CMEC、CMT12、CMT2中ESR1 mRNA的ΔCt值]
3.3. 犬ESR1 3′UTR中预测的miR-18a和miR-19a种子匹配位点
TargetScan分析在犬ESR1 3′UTR（NM_001286958.2）中鉴定出保守和低保守种子匹配位点（图3）。对于miR-18a，保守位点位于1817–1824和4047–4053，预测的低保守位点位于650–656 [33,34]。对于miR-19a，保守位点位于3585–3591和4044–4050 [33,34]。这些预测的结合区域用于指导功能实验中ESR1 3′UTR报告基因构建体的设计。
[图：犬ESR1 3′非翻译区（3′UTR）中预测的miR-18a和miR-19a种子匹配位点。示意图]
3.4. 犬乳腺细胞系中miR-18a的表达
在ESR1表达分析之后，以U6 snRNA作为内源性对照，在相同细胞系中评估了miR-18a的表达。miR-18a表达在各细胞系间差异显著（F (4, 12) = 3.68，p = 0.0352；图4A）。Dunnett比较显示，CMT27中miR-18a表达比CMEC高8.49倍（校正p = 0.0172；图4B）。CMT12显示5.10倍升高，但该差异不显著（校正p = 0.0735）。CMT28和CMT47中表达也较高，但这些差异不显著（校正p分别为0.7372和p = 0.1872）。
[图：miR-18a和miR-19a在犬乳腺细胞系中的表达。（A）miR-18a的ΔCt值和（B）模型估计的]
模型估计的倍数变化值（2−ΔΔCt）：miR-18a：CMEC = 1.00；CMT12 = 5.10；CMT27 = 8.49；CMT28 = 1.79；CMT47 = 3.597。miR-19a：CMEC = 1.00；CMT12 = 3.759；CMT27 = 8.84；CMT28 = 1.052；CMT47 = 1.754。
3.5. miR-19a在犬乳腺细胞系中的表达
随后以U6作为归一化对照，在相同细胞系中评估了miR-19a的表达。miR-19a表达在各细胞系间差异显著（F (4, 12) = 6.15，p = 0.0063；图4C）。Dunnett校正比较显示，CMT27中miR-19a表达比CMEC高8.84倍（校正p = 0.0049；图4D）。CMT12、CMT28和CMT47与CMEC相比无显著差异（校正p分别为0.0885、p = 0.9985和p = 0.6588）。
总体而言，miR-18a和miR-19a表达的最明显升高见于CMT27。基于这些表达模式和预测的ESR1结合位点，我们接下来检验了这些miRNA是否可通过ESR1 3′UTR的选定区域影响报告基因活性。
3.6. ESR1 3′UTR报告基因和miRNA表达载体
为检验miR-18a和miR-19a对ESR1 3′UTR选定区域的影响，我们使用了为功能分析设计的荧光素酶报告基因和miRNA表达质粒构建体。ESR1-WT报告基因包含犬ESR1 3′UTR的选定片段，其中包括生物信息学分析所鉴定的预测miR-18a和miR-19a种子匹配区域。相应的ESR1-MUT报告基因在预测种子匹配区域内含有缺失，用作阴性对照。这些ESR1 3′UTR片段被置于Gaussia荧光素酶报告基因下游，从而可在miRNA表达后评估报告基因活性的变化。
同时，使用GeneCopoeia前体表达载体表达miR-18a和miR-19a，并将荧光报告基因整合到构建体中以可视化载体表达。miR-C对照和miR-18a表达载体含有eGFP报告基因，而miR-19a表达载体含有mCherry报告基因。转染后使用荧光成像确认荧光报告蛋白的表达。报告基因和miRNA表达构建体的总体设计总结于图5。
[图：用于ESR1 3′UTR报告基因试验和miRNA表达的质粒构建体。（A）Gaussia荧光素酶r的示意图]
3.7. miRNA载体转染后荧光报告基因表达的确认
在进行荧光素酶报告基因试验之前，使用荧光成像确认CMT28细胞中miRNA表达载体的荧光报告蛋白表达。未转染细胞未显示可检测的GFP或mCherry荧光，确认了成像条件下的背景信号极低。转染miR-C-eGFP对照载体的细胞显示清晰的GFP荧光。同样，转染miR-18a-eGFP和miR-19a-mCherry的细胞分别显示可检测的GFP和mCherry信号（图6）。由于前体miRNA和荧光报告蛋白在同一启动子下表达，荧光提供了相应前体miRNA载体表达的间接证据。未直接测量成熟miR-18a和miR-19a的水平。
[图：荧光成像确认CMT28细胞中miRNA载体的荧光报告蛋白表达。图（A]
3.8. miR-18a和miR-19a降低ESR1-WT报告基因活性
在CMT28细胞中使用ESR1-WT报告基因测试了miR-18a和miR-19a的作用。与miR-C对照相比，miR-18a显著降低报告基因活性（平均配对差值 = −0.397，95% CI：−0.717至−0.077；p = 0.0334；图7A）。与miR-C相比，miR-19a也显著降低报告基因活性（平均配对差值 = −0.428，95% CI：−0.609至−0.248；p = 0.0094；图7B）。这些结果表明，miR-18a和miR-19a均可通过犬ESR1 3′UTR的选定区域降低报告基因活性。
[图：miR-18a和miR-19a对CMT28细胞中ESR1-WT报告基因的影响。CMT28细胞转染ESR1-WT re]
3.9. miR-18a和miR-19a不降低ESR1-MUT报告基因活性
随后使用ESR1-Mut报告基因检验预测的种子匹配区域对miRNA效应是否重要。与ESR1-WT不同，与miR-C相比，miR-18a未显著改变ESR1-MUT报告基因活性（平均配对差值 = 0.078，95% CI：−0.197至0.353；p = 0.3467；图8A）。同样，miR-19a未显著改变报告基因活性（平均配对差值 = 0.004，95% CI：−0.305至0.312；p = 0.9617；图8B）。
[图：miR-18a和miR-19a对CMT28细胞中ESR1-MUT报告基因的影响。CMT28细胞转染ESR1-MUT]
双向重复测量方差分析显示，报告基因构建体与miRNA处理之间存在显著交互作用（F (2, 4) = 17.89，p = 0.0101）。直接比较了ESR1-WT与ESR1-MUT报告基因之间的miRNA效应。miR-18a（效应平均差 = −0.475，95% CI：−0.891至−0.058；原始p = 0.0391；Holm校正p = 0.0492）和miR-19a（效应平均差 = −0.432，95% CI：−0.729至−0.135；原始p = 0.0246；Holm校正p = 0.0492）均存在显著差异。这些结果提示，报告基因活性的降低取决于ESR1 3′UTR所选区域内预测的miRNA响应区域。
3.10. Anti-miR对ESR1-WT报告基因活性的影响
随后使用Anti-miR检测阻断每种miRNA是否能增加报告基因活性。与单独使用miR-18a相比，加入anti-miR-18a显著增加了报告基因活性（平均配对差 = 0.205，95% CI：0.165至0.246；p = 0.0021；图9A）。与单独使用miR-19a相比，anti-miR-19a也显著增加了报告基因活性（平均配对差 = 0.279，95% CI：0.049至0.509；p = 0.0347；图9B）。这些结果表明，anti-miR处理对ESR1-WT报告基因活性产生了显著但部分恢复的作用。总之，这些发现进一步支持miRNA对ESR1-WT报告基因的依赖性调控。
[图：anti-miR处理对CMT28细胞中ESR1-WT报告基因活性的影响。CMT28细胞转染了ESR1-W]
4. 讨论
本研究检测了犬乳腺细胞系中ESR1 mRNA的表达以及miR-18a和miR-19a对所选ESR1 3′UTR报告基因区域的影响。在四个肿瘤来源细胞系中的三个中，ESR1表达显著降低。miR-18a和miR-19a的表达在各细胞系间存在差异，仅在CMT27中观察到显著升高。这些表达模式在所有细胞系中并未显示这些miRNA与ESR1之间一致的负相关关系。通过表达分析、靶点预测和功能性报告基因试验，本研究提供了基于报告基因的证据，表明miR-18a和miR-19a可通过犬ESR1 3′UTR的所选区域抑制报告基因活性。
在本研究所用细胞系中也观察到ESR1表达的显著差异。CMT12显示较高的ESR1表达，而CMT27、CMT28和CMT47显示低得多的表达。这种差异与人类乳腺癌中报道的雌激素受体表达异质性相似[35]，后者与肿瘤行为和治療反应的差异相关[36]。CMT27、CMT28和CMT47中ESR1 mRNA低表达提示ESR1表达降低可能是这些CMT来源细胞系的共同特征，这与先前关于犬侵袭性、激素非依赖性疾病报道一致[37,38]。
miR-18a和miR-19a的表达模式在所有细胞系中并未显示与ESR1表达一致的负相关关系。两种miRNA在CMT27中均强烈升高，而该细胞系也显示低ESR1表达。在CMT12中，miR-18a和miR-19a表达数值上高于CMEC，但这些差异无统计学显著性，而ESR1表达仍保持高水平。因此，仅miR-18a和miR-19a只能部分解释ESR1表达的差异。其他调控机制可能参与ESR1 mRNA表达的控制。Abbate等人先前的研究报道，在高增殖性犬乳腺癌中miR-18a表达升高伴随ESR1 mRNA和ERα免疫表达降低[20]。这与本研究中CMT27观察到的模式相似，但其他细胞系中缺乏一致的负相关表达模式表明miR-18a与ESR1之间的关系可能在不同CMT模型中有所变化，且可能更为复杂。
计算预测将ESR1鉴定为miR-18a和miR-19a的潜在靶点。miRDB数据库对两种miRNA均给出了高预测评分[27]。由于仅凭预测无法确认调控作用，我们使用犬ESR1 3′UTR报告基因检测了它们的作用。
报告基因试验显示，miR-18a和miR-19a降低了ESR1-WT报告基因的活性。这与人类癌症研究中将miR-17–92簇成员与ERα表达和ESR1信号降低相关联的报道一致[19,39]。相反，两种miRNA均未显著降低含预测种子匹配区域缺失的突变型报告基因的活性。两种miRNA的效应在ESR1-WT和ESR1-MUT报告基因之间也存在显著差异。与单独使用miRNA相比，用相应anti-miR阻断每种miRNA部分增加了报告基因活性。这些发现表明，观察到的抑制作用取决于犬ESR1 3′UTR中所选的miRNA响应区域。据我们所知，这是首个功能性报告基因证据，表明miR-18a和miR-19a可通过犬ESR1 3′UTR的所选区域在CMT细胞中降低报告基因活性。
在人类和犬中，ERα表达降低常见于更具侵袭性的乳腺肿瘤，且ER阴性肿瘤往往生长更快，对激素治疗的应答较差[8,40]。我们的报告基因研究结果表明，miR-18a和miR-19a有可能参与CMTs中ESR1的调控，这与这些miRNA在人类系统中调控ESR1的实验证据一致[39]。然而，这需要在内源性水平上进行验证。这些miRNA也曾被研究作为肿瘤分类或预后的潜在生物标志物[41]，尽管这需要在更大规模的犬群中进行验证。
了解CMTs中ESR1的调控可能有助于未来的诊断和治疗研究。如果miRNA介导的抑制导致ESR1表达降低，miRNA谱可能有助于区分不同的肿瘤模式或亚型。然而，本研究并未确立miR-18a和miR-19a的任何诊断或治疗用途。这些可能的应用需要进一步验证。
本研究的一个局限性是所检测的犬乳腺细胞系数量较少，且功能报告基因实验仅使用了CMT28。此外，CMEC参考细胞来源于单只犬，可能不能代表正常犬乳腺组织间的生物学变异。尽管这些细胞系在ESR1和miRNA表达上显示出明显差异，但仍需要使用更多细胞系、原代CMT样本和体内模型进行研究，以确定相同模式是否更广泛地出现。此外，荧光素酶报告基因实验是专门为检测所选miR-18a和miR-19a应答的ESR1 3′UTR区域而设计的。因此，报告基因实验证明了对报告基因构建体的效应，但并未确定这些miRNA是否在完整肿瘤细胞中调控内源性ESR1 mRNA或ERα蛋白。荧光提供了前体miRNA载体表达的间接证据。报告基因实验中未直接评估成熟miR-18a、miR-19a水平及抑制剂效率。未纳入非靶向抗miR对照；因此，不能排除额外寡核苷酸或转染过程的非特异性效应。报告基因实验仅包括三次独立实验运行，这限制了估计值的精确度。因此，应谨慎解读结果。其他miRNA、RNA结合蛋白或表观遗传机制也可能导致所观察到的细胞系间ESR1表达差异。
总体而言，miR-18a和miR-19a通过犬ESR1 3′UTR的选定区域抑制了报告基因活性。ESR1-WT和ESR1-MUT报告基因对miRNA处理的应答存在显著差异。抗miR处理也部分增加了ESR1-WT报告基因活性。这些发现支持所选miRNA应答区域的参与。这种调控在内源性ESR1表达中的重要性仍有待确定。未来研究应检验抑制内源性水平的miR-18a和miR-19a是否能增加或恢复内源性ESR1 mRNA水平。还应研究miR抑制对细胞行为的影响。
5. 结论
本研究表明，miR-18a和miR-19a可降低含有犬ESR1 3′UTR选定区域的报告基因的活性。ESR1-WT和ESR1-MUT报告基因对miRNA处理的应答存在显著差异。在miR-18a或miR-19a存在的情况下，加入相应的抗miR能够部分恢复ESR1-WT报告基因的活性。需要进一步的功能实验来证明相同的调控有助于在其他细胞系和原代肿瘤细胞或体内模型中管理内源性ESR1 mRNA或ERα蛋白水平。需要此类进一步验证才能将这些miRNA确立为重要的生物标志物或可靶向分子。
补充材料
以下支持信息可在以下网址下载：https://www.mdpi.com/article/10.3390/vetsci13101071/s1，图S1：(A) GAPDH和(B) ESR1的RT-qPCR标准曲线；图S2：(A) miR-18a、(B) miR-19a和(C) U6的RT-qPCR标准曲线；表S1：ESR1 3′UTR荧光素酶报告质粒和前体miRNA表达载体；表S2：本研究中使用的抗miR寡核苷酸；表S3：荧光素酶报告基因实验中使用的选定非连续ESR1 3′UTR片段；表S4：运行水平荧光素酶报告基因数据及效应估计值（含95%置信区间）。
[内容请查看原文]
作者贡献
概念构思，R.C.B.和D.G.；方法学，R.C.B.和D.G.；软件，D.G.；验证，D.G.；形式分析，D.G.；调查研究，D.G.；资源，R.C.B.；数据管理，D.G.；写作——初稿准备，D.G.；写作——审阅与编辑，R.C.B.和D.G.；可视化，D.G.；监督，R.C.B.；项目管理，R.C.B.；资金获取，R.C.B.。所有作者均已阅读并同意稿件的发表版本。
机构审查委员会声明
动物研究方案经奥本大学机构动物护理与使用委员会（IACUC）批准（PRN 2009-1633，日期2009年10月19日）。
知情同意声明
已获得参与本研究的所有受试者的知情同意。
数据可用性声明
本研究呈现的原创性贡献已包含在文章/补充材料中。进一步咨询可联系通讯作者。
利益冲突
作者声明无利益冲突。
资助声明
本研究未获得外部资助。
脚注
[内容请查看原文]
参考文献
[内容请查看原文]
关联数据
补充材料
[内容请查看原文]
数据可用性声明
本研究呈现的原创性贡献已包含在文章/补充材料中。进一步咨询可联系通讯作者。
查看英文原文▾
Simple Summary
Mammary tumors are common in female dogs, and estrogen can play an important role in their growth and behavior. Some tumors have reduced estrogen receptor expression, but the reasons for this are not fully understood. In this study, we examined whether two small regulatory molecules, miR-18a and miR-19a, could affect reporter activity through selected regions of the canine ESR1 3′UTR. MicroRNAs can bind to messenger RNA and reduce gene activity. We found that both miR-18a and miR-19a reduced activity of a reporter containing selected regions of the canine ESR1 3′UTR. This effect was not seen when the predicted miRNA target regions were deleted. Blocking the miRNAs also increased reporter activity. These findings suggest that miR-18a and miR-19a can affect reporter activity through selected regions of canine ESR1 3′UTRs and provide a basis for further studies in canine mammary cancer.
Keywords: canine mammary tumor, miR-18a, miR-19a, estrogen receptor 1 (ESR1), luciferase reporter assay, 3′ untranslated region, post-transcriptional regulation
1. Introduction
Breast cancer is one of the most frequently diagnosed cancers in women, and mammary tumors are also common in intact female dogs [1,2,3]. Both species share several important biological and clinical features, including spontaneous tumor development, marked heterogeneity, and hormone-dependent growth [4]. Because of these similarities, canine mammary tumors (CMTs) are often used as a spontaneous comparative model to understand mammary tumorigenesis and hormone-driven cancer progression in women [2,5].
There are two types of estrogen receptors—estrogen receptor alpha (ERα) and estrogen receptor beta (ERβ), encoded by the ESR1 and ESR2 genes, respectively. ERα and ERβ have some similar functions, but they can also have different effects depending on the tissue and cell type. In some tissues, ERβ can oppose ERα mediated signaling [6]. Both ERα and ERβ have been detected in canine mammary tissues and tumors, although their expression varies among normal, benign, and malignant tissues [7,8]. This study focuses on ERα, encoded by ESR1. When estradiol binds to ERα, it forms a dimer and moves into the nucleus, where it regulates genes involved in cell growth, differentiation, and survival. Based on ERα, PR, and HER2 expression profiles, CMTs have been grouped into phenotypes resembling luminal A, luminal B, HER2-positive, and triple-negative breast cancer [3,9]. However, these canine groups may not fully correspond to the established human breast cancer subtypes. In human breast cancer, tumors that express ERα usually respond better to endocrine therapies, including aromatase inhibitors and selective estrogen receptor modulators such as letrozole and tamoxifen [10,11]. Tumors with reduced or abnormal ERα expression tend to be more aggressive and are associated with poorer outcomes in both species [10,12].
In human breast cancer, ERα loss can occur through several mechanisms, including promoter methylation, ESR1 mutations, and post-transcriptional regulation. However, the mechanisms underlying ERα downregulation in CMTs are not well understood, and the post-transcriptional regulation of ESR1 in dogs remains unclear.
MicroRNAs (miRNAs) are small, highly conserved non-coding RNAs, about 18–25 nucleotides long, that regulate gene expression after transcription by binding to complementary sequences in the 3′UTR of target mRNAs [13,14]. Abnormal miRNA expression is common in many cancers, including CMTs, and can influence cell growth, apoptosis, invasion, and metastasis [5,14,15,16]. Among these, miR-18a and miR-19a—both part of the miR-17–92 cluster—have been linked to mammary tumor progression [17,18]. In both human breast cancer and CMTs, higher miR-18a expression is associated with reduced ESR1/ERα expression [19,20,21]. miR-19a has also been reported to be elevated in early breast cancer and CMTs [15,22,23]. Several studies in human breast cancer suggest that miRNA-mediated suppression of ESR1 may contribute to resistance to endocrine therapy [21,24,25]. Since miRNA expression patterns in CMTs share similarities with those reported in human breast cancer, miR-18a and miR-19a may similarly influence estrogen receptor signaling in dogs [5,26].
We selected miR-18a and miR-19a based on findings from previous studies by our research group. Both miRNAs were identified by cell-line and exosomal miRNA profiling and were found at higher levels in CMT cells and their secreted exosomes compared with normal canine mammary epithelial cells (CMECs) [15,22]. In a later clinical study of dogs with mammary carcinoma, circulating miR-18a was identified as a potential prognostic biomarker, with higher levels associated with lymphatic invasion [18]. Bioinformatic analyses also identified the canine ESR1 transcript as a predicted target of both miRNAs [22,27]. However, expression patterns and target predictions do not establish a functional regulatory interaction. To our knowledge, the effects of these miRNAs on a reporter containing canine ESR1 3′UTR regions have not previously been reported.
In this study, we examined whether miR-18a and miR-19a can affect reporter activity through selected regions of the canine ESR1 3′UTR in mammary tumor cells. We measured endogenous expression of ESR1, miR-18a, and miR-19a. Luciferase reporter assays containing the selected wild-type ESR1 3′UTR regions and a corresponding mutant ESR1 3′UTR reporter having deletions within the predicted miR-18a and miR-19a seed-match regions were used to assess miRNA-dependent reporter activity. We also used miRNA expression vectors and anti-miRs to assess specificity of these effects.
2. Materials and Methods
2.1. Cell Culture
Canine mammary tumor (CMT) cell lines CMT12, CMT27, CMT28, and CMT47 were used in this study. These cell lines were selected because they are commonly used CMT models and have been used in previous studies of miRNA expression [15,22]. These cell lines were originally derived from mammary tumors from individual dogs, as described previously [28,29]. CMT12 was derived from adenocarcinoma in a poodle, CMT27 from a carcinoma in a German Shepherd, CMT28 from an adenocarcinoma in a mixed-breed dog, and CMT47 from an adenocarcinoma in a Miniature Schnauzer [22,28]. Canine mammary epithelial cells (CMECs), used as normal mammary epithelial reference cells, were derived from mammary glands of one female Beagle at necropsy with no prior history of mammary carcinoma [29,30]. The CMT cell lines were used between passages 20 and 50. The CMECs represented a single primary cell culture and were used below passage 15. We identified each cell line using the labels and records of the original frozen stocks. Cell morphology was regularly checked during culture, and each cell line maintained its distinct appearance. No additional molecular authentication was performed in this study. Cells were grown in Dulbecco’s Modified Eagle Medium (DMEM) supplemented with L-glutamine (Quality Biological, Gaithersburg, MD, USA), 10% fetal bovine serum (Atlas Biologicals, Fort Collins, CO, USA), and 1% penicillin-streptomycin (Sigma-Aldrich, St. Louis, MO, USA), and were maintained at 37 °C in a humidified incubator with 5% CO2. Cells were routinely passaged after reaching approximately 70–80% confluence. All cells were handled in standard vented-lid 75 cm2 tissue-culture flasks (VWR Avantor, Radnor, PA, USA). The cell cultures were examined for Mycoplasma contamination by electron microscopy at the time of cryopreservation. Routine Mycoplasma testing was not performed during the present study. Work involving canine cell lines was performed under protocols approved by the Auburn University Institutional Animal Care and Use Committee.
2.2. RNA Extraction
Total RNA was extracted using the miRNeasy kit with QIAzol lysis reagent and RNase-free DNase (Qiagen, Germantown, MD, USA) according to the manufacturer’s instructions. Cells were collected at 70–80% confluence and lysed directly in QIAzol. RNA was purified using miRNeasy spin columns, eluted in RNase-free water, and stored at −80 °C until further use. RNA concentration and purity were assessed using a NanoDrop spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA), and only samples with A260/A280 ratios between 1.8 and 2.0 were used for cDNA synthesis and qPCR.
2.3. cDNA Synthesis
For mRNA targets, including ESR1 and the reference gene GAPDH, cDNA was synthesized using the High-Capacity RNA-to-cDNA Kit (Thermo Fisher Scientific, Waltham, MA, USA). Each reaction contained 1 µg of total RNA and was incubated at 37 °C for 60 min, followed by heat inactivation at 95 °C for 5 min. The resulting cDNA was stored undiluted at −20 °C and diluted twofold immediately before qPCR.
For miRNA analysis, cDNA was synthesized using the Qiagen miRCURY LNA miRNA RT Kit. Each reaction contained approximately 200 ng of total RNA and was incubated at 42 °C for 60 min, followed by heat inactivation at 95 °C for 5 min. The resulting undiluted cDNA was stored at −20 °C and diluted 60-fold immediately before qPCR.
2.4. RT-qPCR
Using the diluted cDNA as described above, RT-qPCR reactions for mRNA and miRNA targets were performed separately on a Bio-Rad CFX96 Real-Time PCR Detection System (Bio-Rad Laboratories, Hercules, CA, USA). Each reaction was run in duplicate. No-reverse-transcription (no-RT) and no-template controls were included in all runs to assess genomic DNA and reagent contamination.
To assess mRNA expression of ESR1, and GAPDH reactions were prepared using the PowerUp SYBR Green Master Mix (Thermo Fisher Scientific, Waltham, MA, USA) with gene-specific primers (Table 1). The GAPDH primer sequences were based on a previously published study [9,31], while the ESR1 primers were designed using NCBI Primer-BLAST to ensure the amplicon spanned an exon–exon junction. All primers were synthesized by IDT (Integrated DNA Technologies, Coralville, IA, USA). The ESR1 amplicon was sequence-verified by conventional PCR followed by Sanger sequencing (MGH DNA Core, Massachusetts General Hospital, Boston, MA, USA) to confirm target specificity. Annealing conditions for ESR1 were optimized using a serial dilution curve, and the final reactions were run at 60.5 °C, yielding clean, specific amplification and consistent Ct values. The reaction was run under the following cycling conditions: 50 °C for 2 min, 95 °C for 2 min, followed by 36 cycles of 95 °C for 1 s and 60.5 °C for 30 s. A melt-curve step was included to confirm amplicon specificity. GAPDH was used as the endogenous reference gene to normalize the ESR1 mRNA expression. We selected GAPDH because it has been used in previous studies of canine mammary cancer cell lines [31]. However, we did not formally compare its stability with other reference genes using geNorm or NormFinder.
Table 1. Primer sequences and amplicon sizes used for canine mRNA RT-qPCR.
| Canine Gene | Forward Primer (5′ → 3′) | Reverse Primer (5′ → 3′) | Amplicon Size (bp) |
|---|---|---|---|
| GAPDH | ATGGTGAAGGTCGGAGTCAA | ATCACCCCATTTGATGTTGG | 261 |
| ESR1 | ATGTGTCCAGCTACCAACC | GTCTTTCCGTATCCCGCC | 117 |
For miRNA expression, RT-qPCR was performed using the miRCURY LNA miRNA PCR Kit (Qiagen, Germantown, MD, USA) with LNA primer assays for cfa-miR-18a-5p (Assay ID: YP00204207), cfa-miR-19a-3p (Assay ID: YP00205862), and U6 snRNA (Assay ID: YP02119464). Cycling conditions were 95 °C for 2 min, followed by 40 cycles of 95 °C for 10 s and 56 °C for 1 min. U6 snRNA was used as the endogenous reference for miRNA normalization, as reported in a previous CMT study [32]. ΔCt values were calculated relative to U6 Ct. Relative expression was calculated using the ΔCt method, and fold-change values were derived using the 2−ΔΔCt method.
Amplification efficiencies were calculated from the slopes of standard curves generated using serially diluted cDNA. The amplification efficiencies were 104.6% for GAPDH, 105.3% for ESR1, 105.1% for miR-18a, 90.4% for miR-19a, and 91.9% for U6. The corresponding standard curves with R2 values are provided in Supplementary Figures S1 and S2.
2.5. In Silico Identification of miRNA Binding Sites
Before designing the luciferase constructs, the canine ESR1 3′UTR was examined for predicted miRNA binding sites using the miRDB database (http://mirdb.org; initially accessed on 24 June 2023 and rechecked on 14 September 2023). miRDB prediction scores ranged from 50 to 100, and scores above 80 were generally considered high-confidence targets [27]. Both cfa-miR-18a and cfa-miR-19a appeared among the top predicted regulators, with prediction scores of 99 and 96, respectively. TargetScanHuman 8.0 (TargetScanHuman 8.0; initially accessed on 24 June 2023 and rechecked on 14 September 2023) was also used to examine the location and conservation of predicted miR-18a and miR-19a seed-match sites within the canine ESR1 3′UTR (NM_001286958.2) [33,34]. Dog was selected as the species, and miR-18 and miR-19 were searched separately. Site locations and conservation categories were taken directly from TargetScan and reported as “Conserved” or “Poorly conserved.”
Based on these results, the ESR1 3′UTR segments containing the predicted binding sites were selected for cloning into the luciferase reporter vector to experimentally assess their regulatory activity.
2.6. Plasmids and miRNA Constructs
The ESR1 3′UTR reporter constructs, and precursor miRNA expression vectors used in this study were obtained from GeneCopoeia (Rockville, MD, USA). The reporter construct containing selected wild-type ESR1 3′UTR regions (CS-miT740T-MT05-01) contained an 859 bp insert derived from the canine ESR1 3′UTR (NM_001286958.2) cloned downstream of the Gaussia luciferase reporter gene in the pEZX-MT05 vector. The insert contained three selected non-overlapping ESR1 3′UTR regions containing predicted miR-18a and miR-19a seed-match sites. Fragment 1 was 308 bp and contained the predicted miR-18a binding region, Fragment 2 was 307 bp and contained the predicted miR-19a binding region, and Fragment 3 was 244 bp and contained overlapping predicted binding regions for both miRNAs. Hereafter, this construct is referred to as the ESR1-WT reporter. “ESR1-WT” refers to the unmodified sequences of the selected regions and does not indicate the complete native ESR1 3′UTR.
A corresponding mutant ESR1 3′UTR reporter construct (CS-miT740T-MT05-02) was used as a negative control. Relative to ESR1-WT reporter, this construct contained three deletions within the selected ESR1 3′UTR regions: GCACCTTA at nucleotides 151–158, TTTGCAC at nucleotides 459–465, and GCACCTT at nucleotides 766–772 of the reporter insert. These deletions removed the three predicted miRNA seed-match sites and reduced the insert length from 859 to 837 bp. Hereafter, this construct is referred to as the ESR1-MUT reporter.
CMT28 cells were transfected with precursor miRNA vectors for cfa-miR-18a (CS-miR738T-MR04-01) and cfa-miR-19a (CS-miR739T-MR04-01) for the reporter assays. The miR-18a vector contained an eGFP reporter, whereas miR-19a vector contained an mCherry reporter. A GeneCopoeia miRNA control vector (miR-C; CmiR0001-MR04) containing eGFP was used as the negative control for miRNA vector transfection. The precursor miRNA and fluorescent reporter protein were expressed from the same vector under the same promoter. Therefore, GFP and mCherry fluorescence provided indirect evidence of expression of the corresponding miRNA vectors but did not directly confirm increased levels of mature miR-18a or miR-19a. For rescue experiments, miArrest™ synthetic miRNA inhibitors, containing antisense sequences, (GeneCopoeia) targeting miR-18a-5p (HmiR-AN0255-SN-5) and miR-19a-3p (HmiR-AN0293-SN-5) were used. Both inhibitors were single-stranded, 2′-O-methyl-modified RNA oligonucleotides. Although these inhibitors were designed for human miR-18a-5p and miR-19a-3p, the mature sequences of both miRNAs are identical in humans and dogs. Therefore, the inhibitors are also complementary to cfa-miR-18a-5p (MIMAT0009832) and cfa-miR-19a-3p (MIMAT0006650). A non-targeting anti-miR control was not included in the rescue experiments. Additional information on the reporter constructs, miRNA expression vectors, inhibitors and ESR1 3′UTR sequences is provided in Supplementary Tables S1–S3.
2.7. Plasmid Preparation
Mix & Go! DH5α competent E. coli (Zymo Research, Irvine, CA, USA) was transformed with the ESR1 3′UTR reporter constructs and miRNA expression plasmids according to the manufacturer’s instructions. Transformed colonies were grown overnight in 2YT broth containing ampicillin at 37 °C with shaking. Plasmid DNA was purified using the ZymoPURE Midiprep Kit (Zymo Research, Irvine, CA, USA) according to the manufacturer’s instructions. Plasmid concentrations typically ranged from 1.5 to 3 µg/µL, with A260/A280 ratios between 1.8 and 2.0.
2.8. Transient Transfection and Luciferase Reporter Assay
CMT28 cells were seeded in 24-well plates at ~1.5 × 105 cells/mL (500 µL/well) and allowed to reach 60–80% confluence. CMT28 cells were selected for the functional reporter assays because they showed relatively low endogenous miR-18a and miR-19a expression and were well suited for reproducible transfection based on their growth and culture characteristics. For rescue experiments, cells were first transfected with 5 pmol of anti-miR oligonucleotides using Lipofectamine RNAiMAX (Thermo Fisher Scientific, Waltham, MA, USA). After approximately 18 h, the transfection medium was replaced with fresh complete medium.
Twenty-four hours after the initial anti-miR transfection, cells were transfected with either 0.5 µg of reporter plasmid alone or co-transfected with 0.25 µg of reporter plasmid and 0.25 µg of the corresponding miRNA expression vector or miR-C control vector using Lipofectamine 3000 (Thermo Fisher Scientific, Waltham, MA, USA) in Opti-MEM. The total amount of plasmid DNA was therefore maintained at 0.5 µg per well.
Forty-eight hours after plasmid transfection, conditioned media were collected, and Gaussia luciferase activity was measured using the GeneCopoeia Dual Luminescence Assay Kit. Luminescence was recorded using a Synergy HTX multimode microplate reader (BioTek, Winooski, VT, USA; Gen5 software (version 2.09.1)). Gaussia luciferase activity was normalized to secreted alkaline phosphatase (SEAP) activity to account for transfection efficiency. All transfection and luciferase reporter experiments were performed in three independent experimental runs, with two transfected wells per condition in each run. Luminescence from each well was measured in duplicate.
2.9. Fluorescence Imaging
An EVOS fluorescence imaging system (Thermo Fisher Scientific, Waltham, MA, USA) was used to confirm expression of the fluorescent reporter proteins following transfection of the miRNA expression vectors in CMT28 cells. GFP and RFP filter cubes were used to detect eGFP and mCherry fluorescence, respectively, with a 10× objective lens. GFP images were taken at 60% illumination with an exposure time of 250 ms, while RFP images were taken at 60% illumination with an exposure time of 750 ms. The images were used only to confirm fluorescent reporter expression.
2.10. Statistical Analysis
RT-qPCR experiments were performed in four independent experimental runs, with two technical replicates per condition in each run. Technical replicates were averaged within each run before statistical analysis. Statistical analyses were performed in R (version 4.3).
For qPCR experiments, ESR1 expression was normalized to GAPDH, whereas miR-18a and miR-19a expression was normalized to U6. ΔCt values were calculated as the difference between the Ct value of the target and the corresponding reference gene. ΔCt values were analyzed using linear mixed-effects models with the lmerTest package, with cell line included as a fixed effect and experimental run included as a random intercept. Estimated marginal means (EMMs) were obtained using the emmeans package, and each CMT cell line was compared with CMEC using Dunnett-adjusted contrasts. Relative fold changes were calculated from the model-estimated differences using the 2−ΔΔCt method, with CMEC used as the reference. Normality of the model residuals was assessed using residual Q-Q plots and the Shapiro–Wilk test, and equality of variances was assessed using Levene’s test.
Luciferase reporter experiments were performed in three independent runs, with two transfected wells per condition in each run. Each independent experimental run was performed on a different day using a new cell culture plate and a separate transfection. Each well was measured in duplicate. Incomplete GLuc/SEAP read pairs were excluded, and valid technical measurements were averaged within each run before statistical analysis. The independent experimental run was treated as the statistical unit (n = 3). Gaussia luciferase activity was normalized to SEAP activity, and the reporter-only condition was set to 1 as the reference. Because reporter-only activity was used for normalization, it was not included in the statistical comparisons. The effects of reporter construct (ESR1-WT or ESR1-MUT), miRNA treatment (miR-C, miR-18a, or miR-19a), and their interaction were analyzed using a two-way repeated-measures ANOVA with experimental run treated as the repeated measure. Comparisons between each miRNA treatment and miR-C were performed using two-sided paired t-tests. The effects of miR-18a and miR-19a were also compared between the ESR1-WT and ESR1-MUT using paired t-tests. Rescue experiments were analyzed using two-sided paired t-tests comparing each miRNA treatment with the corresponding miRNA plus anti-miR treatment. Holm adjustment was used for the two ESR1-WT-versus-ESR1-MUT comparisons. Effect estimates, 95% CIs, and exact p values were reported. Model residuals were assessed using a Q-Q plot and the Shapiro–Wilk test. A p value < 0.05 was considered statistically significant. Run-level reporter data and effect estimates with 95% CIs are provided in Supplementary Table S4.
3. Results
3.1. Canine Mammary Cell Lines
This study included four tumor-derived canine mammary cell lines CMT12, CMT27, CMT28, and CMT47 and a normal primary canine mammary epithelial cell population. All cultures showed adherent growth and distinct cellular morphologies. Representative phase-contrast images are shown in Figure 1.
[图：Representative phase-contrast images of canine mammary cell lines used in this study. ( A ) CMEC, ( B ) CMT12, ( C ) CMT]
3.2. ESR1 Expression Across Canine Mammary Cell Lines
Because ESR1 mRNA expression can vary among mammary tumors, we first assessed endogenous ESR1 mRNA expression in the CMT cell lines using CMEC as the normal canine mammary epithelial reference. ESR1 mRNA expression in the cell lines was assessed by RT-qPCR using GAPDH as the reference gene. ΔCt was calculated as Ct (ESR1)–Ct (GAPDH). The ΔCt values were analyzed using a mixed-effects model with experimental run included as a random effect. ESR1 mRNA expression was significantly different between CMT cell lines and CMEC (p < 0.001; Figure 2A). CMT12 showed lower ΔCt values than CMEC, indicating higher ESR1 mRNA expression (p = 0.0039), whereas CMT27, CMT28, and CMT47 showed higher ΔCt values, indicating lower ESR1 mRNA expression (p = 0.0152, p = 0.0009, and p = 0.0002, respectively). Fold-change values (relative expression) were calculated using the 2−ΔΔCt method, with CMEC used as the reference. ESR1 mRNA expression in CMT12 was approximately 14.9 times higher than in CMEC. In CMT27, CMT28, and CMT47, ESR1 mRNA expression was approximately 9-, 26-, and 70-fold lower, respectively, compared with CMEC (Figure 2B). Overall, ESR1 mRNA expression was much lower in three of the four tumor-derived cell lines, while CMT12 showed significantly higher ESR1 expression.
[图：ESR1 mRNA expression varied widely across canine mammary cell lines. ( A ) ΔCt values for ESR1 mRNA in CMEC, CMT12, CMT2]
3.3. Predicted miR-18a and miR-19a Seed-Match Sites in the Canine ESR1 3′UTR
TargetScan analysis identified both conserved and poorly conserved seed-match sites within the canine ESR1 3′UTR (NM_001286958.2) (Figure 3). For miR-18a, conserved sites were located at 1817–1824 and 4047–4053, and a poorly conserved site was predicted at 650–656 [33,34]. For miR-19a, conserved sites were located at 3585–3591 and 4044–4050 [33,34]. These predicted binding regions were used to guide the design of the ESR1 3′UTR reporter constructs used in the functional assays.
[图：Predicted miR-18a and miR-19a seed-match sites within the canine ESR1 3′ untranslated region (3′UTR). Schematic represen]
3.4. miR-18a Expression Across Canine Mammary Cell Lines
Following ESR1 expression analysis, miR-18a expression was evaluated across the same cell lines using U6 snRNA as the endogenous control. miR-18a expression differed significantly among the cell lines (F (4, 12) = 3.68, p = 0.0352; Figure 4A). Dunnett comparison showed that miR-18a expression was 8.49-fold higher in CMT27 than in CMEC (adjusted p = 0.0172; Figure 4B). CMT12 showed a 5.10-fold increase, but this difference was not significant (adjusted p = 0.0735). Expression was also higher in CMT28 and CMT47, but these differences were not significant (adjusted p = 0.7372 and p = 0.1872, respectively).
[图：miR-18a and miR-19a expression across canine mammary cell lines. ( A ) ΔCt values for miR-18a and ( B ) model-estimated ]
Model-estimated fold-change values (2−ΔΔCt): miR-18a: CMEC = 1.00; CMT12 = 5.10; CMT27 = 8.49; CMT28 = 1.79; CMT47 = 3.597. miR-19a: CMEC = 1.00; CMT12 = 3.759; CMT27 = 8.84; CMT28 = 1.052; CMT47 = 1.754.
3.5. miR-19a Expression Across Canine Mammary Cell Lines
miR-19a expression was then evaluated across the same cell lines using U6 for normalization. miR-19a expression differed significantly among the cell lines (F (4, 12) = 6.15, p = 0.0063; Figure 4C). Dunnett-adjusted comparisons showed that miR-19a expression was 8.84-fold higher in CMT27 than in CMEC (adjusted p = 0.0049; Figure 4D). CMT12, CMT28, and CMT47 did not differ significantly from CMEC (adjusted p = 0.0885, p = 0.9985, and p = 0.6588, respectively).
Overall, the clearest increases in miR-18a and miR-19a expression were observed in CMT27. Based on these expression patterns and the predicted ESR1 binding sites, we next examined whether these miRNAs could affect reporter activity through selected regions of the ESR1 3′UTR.
3.6. ESR1 3′UTR Reporter and miRNA Expression Vectors
To examine the effects of miR-18a and miR-19a on the selected regions of the ESR1 3′UTR, we used luciferase reporter and miRNA expression plasmid constructs designed for functional analysis. The ESR1-WT reporter contained selected fragments of the canine ESR1 3′UTR that included the predicted miR-18a and miR-19a seed-match regions identified by bioinformatic analysis. A corresponding ESR1-MUT reporter contained deletions within the predicted seed-match regions and was used as a negative control. These ESR1 3′UTR fragments were placed downstream of the Gaussia luciferase reporter, allowing changes in reporter activity to be evaluated following miRNA expression.
In parallel, GeneCopoeia precursor expression vectors were used to express miR-18a and miR-19a, with fluorescent reporters incorporated into the constructs to visualize vector expression. The miR-C control and miR-18a expression vectors contained an eGFP reporter, whereas the miR-19a expression vector contained a mCherry reporter. Fluorescence imaging was used to confirm expression of the fluorescent reporter proteins following transfection. The overall design of the reporter and miRNA expression constructs is summarized in Figure 5.
[图：Plasmid constructs used for ESR1 3′UTR reporter assays and miRNA expression. ( A ) Schematic of the Gaussia luciferase r]
3.7. Confirmation of Fluorescent Reporter Expression Following miRNA Vector Transfection
Before performing the luciferase reporter assays, fluorescence imaging was used to confirm expression of the fluorescent reporter proteins from the miRNA expression vectors in CMT28 cells. Untransfected cells showed no detectable GFP or mCherry fluorescence, confirming minimal background signal under the imaging conditions. Cells transfected with the miR-C-eGFP control vector showed clear GFP fluorescence. Similarly, cells transfected with miR-18a-eGFP and miR-19a-mCherry showed detectable GFP and mCherry signals, respectively (Figure 6). Because the precursor miRNA and fluorescent reporter proteins were expressed under the same promoter, fluorescence provided indirect evidence of expression from the corresponding precursor miRNA vectors. Levels of mature miR-18a and miR-19a were not directly measured.
[图：Fluorescence imaging confirms expression of fluorescent reporter proteins from miRNA vectors in CMT28 cells. Panels ( A ]
3.8. miR-18a and miR-19a Reduce ESR1-WT Reporter Activity
The effects of miR-18a and miR-19a were tested using the ESR1-WT reporter in CMT28 cells. Compared with the miR-C control, miR-18a significantly reduced reporter activity (mean paired difference = −0.397, 95% CI: −0.717 to −0.077; p = 0.0334; Figure 7A). miR-19a also significantly reduced reporter activity compared with miR-C (mean paired difference = −0.428, 95% CI: −0.609 to −0.248; p = 0.0094; Figure 7B). These results showed that both miR-18a and miR-19a can reduce reporter activity through the selected regions of the canine ESR1 3′UTR.
[图：Effect of miR-18a and miR-19a on the ESR1 -WT reporter in CMT28 cells. CMT28 cells were transfected with the ESR1 -WT re]
3.9. miR-18a and miR-19a Do Not Reduce ESR1-MUT Reporter Activity
The ESR1-Mut reporter was then used to test whether the predicted seed-match regions were important for the miRNA effects. Unlike the ESR1-WT, miR-18a did not significantly change ESR1-MUT reporter activity compared with miR-C (mean paired difference = 0.078, 95% CI: −0.197 to 0.353; p = 0.3467; Figure 8A). Similarly, miR-19a did not significantly change reporter activity (mean paired difference = 0.004, 95% CI: −0.305 to 0.312; p = 0.9617; Figure 8B).
[图：Effect of miR-18a and miR-19a on the ESR1 -MUT reporter in CMT28 cells. CMT28 cells were transfected with the ESR1 -MUT ]
The two-way repeated-measures ANOVA showed a significant interaction between reporter construct and miRNA treatment (F (2, 4) = 17.89, p = 0.0101). The miRNA effects were directly compared between ESR1-WT and ESR1-MUT reporters. Significant differences were found for both miR-18a (mean difference in effect = −0.475, 95% CI: −0.891 to −0.058; raw p = 0.0391; Holm-adjusted p = 0.0492) and miR-19a (mean difference in effect = −0.432, 95% CI: −0.729 to −0.135; raw p = 0.0246; Holm-adjusted p = 0.0492). These results suggest that the reduction in reporter activity depends on the predicted miRNA-responsive regions within the selected regions of ESR1 3′UTR.
3.10. Effect of Anti-miRs on ESR1-WT Reporter Activity
Anti-miRs were then used to test whether blocking each miRNA could increase reporter activity. Addition of anti-miR-18a significantly increased reporter activity compared with miR-18a alone (mean paired difference = 0.205, 95% CI: 0.165 to 0.246; p = 0.0021; Figure 9A). Anti-miR-19a also significantly increased reporter activity compared with miR-19a alone (mean paired difference = 0.279, 95% CI: 0.049 to 0.509; p = 0.0347; Figure 9B). These results show that anti-miR treatment produced a significant but partial restoration of ESR1-WT reporter activity. Together, these findings provide further support for miRNA-dependent regulation of the ESR1-WT reporter.
[图：Effect of anti-miR treatment on ESR1 -WT reporter activity in CMT28 cells. CMT28 cells were transfected with the ESR1 -W]
4. Discussion
This study examined ESR1 mRNA expression in canine mammary cell lines and the effects of miR-18a and miR-19a on selected ESR1 3′UTR reporter regions. In three of the four tumor-derived cell lines, ESR1 expression was markedly reduced. miR-18a and miR-19a expression varied among the cell lines, with significant increases observed only in CMT27. The expression patterns did not show a consistent inverse relationship between these miRNAs and ESR1 across all cell lines. Using expression analysis, target prediction, and functional reporter assays, this study provides reporter-based evidence that miR-18a and miR-19a can repress reporter activity through selected regions of the canine ESR1 3′UTR.
Significant variation in ESR1 expression was also observed across the cell lines used in this study. CMT12 showed higher ESR1 expression, while CMT27, CMT28, and CMT47 showed much lower expression. This variation is similar to the heterogeneity in estrogen receptor expression reported in human breast cancer [35], which is associated with differences in tumor behavior and treatment response [36]. The low ESR1 mRNA expression in CMT27, CMT28, and CMT47 suggests that reduced ESR1 expression may be a common feature in these CMT-derived cell lines, consistent with earlier reports of aggressive, hormone-independent disease in dogs [37,38].
The expression patterns of miR-18a and miR-19a did not show a consistent inverse relationship with ESR1 expression across all cell lines. Both miRNAs were strongly elevated in CMT27, which also showed low ESR1 expression. In CMT12, miR-18a and miR-19a expression was numerically higher than in CMEC, but these differences were not statistically significant, while ESR1 expression remained high. Therefore, miR-18a and miR-19a alone can only explain part of the differences in ESR1 expression. Additional regulatory mechanisms likely contribute to the control of ESR1 mRNA expression. A previous study by Abbate et al. reported increased miR-18a expression together with reduced ESR1 mRNA and ERα immune-expression in highly proliferating canine mammary carcinomas [20]. This is similar to the pattern observed in CMT27 in our study, but lack of a consistent inverse expression pattern in the other cell lines indicates that the relationship between miR-18a and ESR1 may vary among CMT models and may be more complex.
Computational prediction identified ESR1 as a potential target of miR-18a and miR-19a. The miRDB database gave high prediction scores for both miRNAs [27]. Because prediction alone cannot confirm regulation, we tested their effects using the canine ESR1 3′UTR reporter.
The reporter assays showed that miR-18a and miR-19a reduced activity of the ESR1-WT reporter. This is consistent with reports from human cancer studies linking members of the miR-17–92 cluster with reduced ERα expression and ESR1 signaling [19,39]. In contrast, neither miRNA significantly reduced activity of the mutant reporter containing deletions in the predicted seed-match regions. The effects of both miRNAs were also significantly different between the ESR1-WT and ESR1-MUT reporters. Blocking each miRNA with the corresponding anti-miR partially increased reporter activity compared with the miRNA alone. These findings indicate that the observed repression depends on the selected miRNA-responsive regions of the canine ESR1 3′UTR. To our knowledge, this is the first functional reporter evidence that miR-18a and miR-19a can reduce reporter activity through selected regions of the canine ESR1 3′UTR in CMT cells.
Reduced ERα expression is commonly seen in more aggressive mammary tumors in both humans and dogs, and ER-negative tumors tend to grow more rapidly and respond poorly to hormone-based therapies [8,40]. Our reporter findings suggest that miR-18a and miR-19a have the potential to contribute to ESR1 regulation in CMTs, consistent with experimental evidence of ESR1 regulation by these miRNAs in human systems [39]. However, this requires validation at the endogenous level. These miRNAs have also been investigated as possible biomarkers for tumor classification or prognosis [41], although this would require validation in larger groups of dogs.
Understanding ESR1 regulation in CMTs could be useful for future studies of diagnosis and treatment. If miRNA-mediated repression contributes to reduced ESR1 expression, miRNA profiles may help distinguish different tumor patterns or subtypes. However, the present study does not establish any diagnostic or therapeutic use of miR-18a and miR-19a. These possible applications require further validation.
A limitation of this study is the small number of canine mammary cell lines examined and use of only CMT28 for the functional reporter experiments. In addition, the CMEC reference cells were derived from a single dog and may not represent the biological variation among normal canine mammary tissues. Although these cell lines showed clear differences in ESR1 and miRNA expression, studies using additional cell lines, primary CMT samples, and in vivo models will be needed to determine whether the same patterns occur more broadly. In addition, the luciferase reporter experiments were designed specifically to test the selected miR-18a and miR-19a responsive regions of the ESR1 3′UTR. The reporter assays therefore demonstrate an effect on the reporter construct but do not establish whether these miRNAs regulate endogenous ESR1 mRNA or ERα protein in intact tumor cells. Fluorescence provided indirect evidence of the precursor miRNA vector expression. Mature miR-18a, miR-19a levels and inhibitors efficiency were not directly assessed in the reporter experiments. A non-targeting anti-miR control was not included; therefore, non-specific effects of the additional oligonucleotide or transfection procedure cannot be excluded. The reporter assays included only three independent experimental runs, which limits the precision of the estimates. Therefore, the results should be interpreted with caution. Other miRNAs, RNA-binding proteins, or epigenetic mechanisms may also contribute to the differences in ESR1 expression observed among the cell lines.
Overall, miR-18a and miR-19a repressed reporter activity through selected regions of the canine ESR1 3′UTR. The response to miRNA treatment differed significantly between the ESR1-WT and ESR1-MUT reporters. Anti-miR treatment also partially increased ESR1-WT reporter activity. These findings support the involvement of the selected miRNA-responsive regions. The importance of this regulation in endogenous ESR1 expression remains to be determined. Future studies should examine whether inhibition of endogenous levels of miR-18a and miR-19a can increase or restore endogenous levels of ESR1 mRNA. The effects of miR inhibition on cellular behavior should also be investigated.
5. Conclusions
This study demonstrates that miR-18a and miR-19a can reduce the activity of a reporter containing selected regions of the canine ESR1 3′UTR. The response to miRNA treatment differed significantly between the ESR1-WT and ESR1-MUT reporters. Addition of the corresponding anti-miRs were able to partially re-store the activity of the ESR1-WT reporter in the presence of miR-18a or miR-19a. Further functional experiments will be needed to demonstrate that the same regulation helps manage endogenous ESR1 mRNA or ERα protein levels in additional cell lines and primary tumor cells or in vivo models. Such further validation will be required to establish these miRNAs as significant biomarkers or targetable molecules.
Supplementary Materials
The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/vetsci13101071/s1, Figure S1: RT-qPCR standard curves for (A) GAPDH and (B) ESR1; Figure S2: RT-qPCR standard curves for (A) miR-18a, (B) miR-19a, and (C) U6; Table S1: ESR1 3′UTR luciferase reporter plasmids and precursor miRNA expression vectors; Table S2: Anti-miR oligonucleotides used in this study; Table S3: Selected, non-contiguous ESR1 3′UTR fragments used in the luciferase reporter assays; Table S4: Run-level luciferase reporter data and effect estimates with 95% confidence intervals.
[内容请查看原文]
Author Contributions
Conceptualization, R.C.B. and D.G.; methodology, R.C.B. and D.G.; software, D.G.; validation, D.G.; formal analysis, D.G.; investigation, D.G.; resources, R.C.B.; data curation, D.G.; writing—original draft preparation, D.G.; writing—review and editing, R.C.B. and D.G.; visualization, D.G.; supervision, R.C.B.; project administration, R.C.B.; funding acquisition, R.C.B. All authors have read and agreed to the published version of the manuscript.
Institutional Review Board Statement
The animal study protocol was approved by the Institutional Animal Care and Use Committee (IACUC) of Auburn University (PRN 2009-1633 dated 19 October 2009).
Informed Consent Statement
Informed consent was obtained from all subjects involved in the study.
Data Availability Statement
The original contributions presented in this study are included in the article/Supplementary Materials. Further inquiries can be directed to the corresponding author.
Conflicts of Interest
The authors declare no conflict of interest.
Funding Statement
This research received no external funding.
Footnotes
[内容请查看原文]
References
[内容请查看原文]
Associated Data
Supplementary Materials
[内容请查看原文]
Data Availability Statement
The original contributions presented in this study are included in the article/Supplementary Materials. Further inquiries can be directed to the corresponding author.