siRNA 种子区化学修饰对脱靶效应的抑制作用（二）

👀

3 Methods 方法

3.1 Designing and Preparing siRNAs with DNA or 2′-OMe Modifications in the Seed Region 设计和制备在种子区域具有 DNA 或 2'-OMe 修饰的 siRNA

Highly functional siRNAs for mammalian cells (see Note 1) can be designed using the web-based online software, siDirect 2.0. In this section, the procedure for selecting highly functional siRNA for human vimentin is shown as an example.

对于哺乳动物细胞(见注释 1)，可以使用基于网络的在线软件 siDirect 2.0 设计高功能的 siRNA。本节以选择人维门蛋白高功能 siRNA 为例展示该过程。

1. Open http://siDirect2.RNAi.jp/ with any appropriate web browser. 打开网页浏览器，访问 siDirect 2.0 网站：http://siDirect2.RNAi.jp/

2. Input the cDNA sequence of vimentin in a nucleotide sequence box. Or enter the accession number of human vimentin (NM_003380) into the upper blank box.将 vimentin cDNA 序列粘贴到核苷酸序列输入框中。或者，在顶部文本框中输入人类 vimentin 的基因序列编号 (NM_003380)。

3. Click “retrieve sequence” to get the nucleotide sequence from GenBank (https://www.ncbi.nlm.nih.gov/genbank） .

   点击 “retrieve sequence” 按钮，从 GenBank 数据库 (https://www.ncbi.nlm.nih.gov/genbank ) 获取核苷酸序列。

4. Click the “Options: [click here]” button to set optional parameters if necessary (see Note 2).根据需要，点击“Option:[click here]”按钮，设置可选参数。

5. Click “design siRNA ,” and you can get result.点击 “design siRNA” 按钮，即可获得 siRNA 设计结果。

6. Synthesize siRNAs shown in Fig. 3 with or without DNA and 2′-OMe modifications in the siRNA seed region chemically, respectively (see Note 3).参照 Fig. 3 的 siRNA 序列，分别化学合成未修饰的 siRNA 以及 siRNA 种子区引入 DNA 或 2'-OMe 修饰的 siRNA (见注释 3)

7. Mix the solutions containing same amounts of guide and passenger strand RNAs, and add the same volume of 2× annealing buffer, and anneal by incubating the mixture at 95 °C for 3 min, at 37 °C for 30 min, and at 25 °C for 60 min.将等摩尔的引导链和过客链 RNA 溶液混合，加入等体积的 2× 退火缓冲液，混合物按如下程序退火：95 °C 孵育 3 分钟，37 °C 孵育 30 分钟，25 °C 孵育 60 分钟。

8. Annealed products may be checked using 19% polyacrylamide gel electrophoresis in TBE buffer, which can separate 21-nt double-stranded siRNA with 2-nt 3′ overhangs from single-stranded RNA . 退火产物可以通过 TBE 缓冲液中的 19% 聚丙烯酰胺凝胶电泳进行检测。该方法可以将具有 2 个核苷酸 3' 末端突出结构的 21 核苷酸双链 siRNA 与单链 RNA 区分开来。

3.2 Preparing Luciferase Reporters with CM and SM Sequences 利用 CM 和 SM 序列制备荧光素酶报告基因

1. Both strands of DNA oligonucleotides for gCM and gSM reporters, or pCM and pSM reporters are mixed, respectively, and add the same volume of 2× annealing buffer. They are annealed by incubating at 95 °C for 3 min, at 37 °C for 30 min, and at 25 °C for 60 min.

   将 gCM 和 gSM 报告基因或 pCM 和 pSM 报告基因的 DNA 寡核苷酸双链分别混合，并加入等体积的 2× 退火缓冲液。 按以下步骤进行退火：95 °C 孵育 3 分钟，37 °C 孵育 30 分钟，25 °C 孵育 60 分钟。

2. Insert the annealed oligonucleotides into the XhoI/EcoRI restriction enzyme sites within 3′ UTR region of Renilla luciferase gene in psiCHECK-1, and name each plasmid vector, such as psiCHECKgCM_VIM270, psiCHECK-gSM_VIM270, psiCHECK-pCM_VIM270, and psiCHECK-pSM_VIM270, respectively (Fig. 3). RNA silencing activities of the guide strand and passenger strand are measured using psiCHECKgCM_VIM270 and psiCHECK-pCM_VIM270, respectively. The seed-dependent off-target effects of the guide strand and passenger strand are measured using psiCHECK-gSM_VIM270 and psiCHECK-pSM_VIM270, respectively (see Note 4).

   将退火后的寡核苷酸插入 psiCHECK-1 中 Renilla 荧光素酶基因的 3'UTR 区 XhoI/EcoRI 限制性酶切位点，并分别命名质粒载体，例如 psiCHECKgCM_VIM270、psiCHECK-gSM_VIM270、psiCHECK-pCM_VIM270 和 psiCHECK-pSM_VIM270 (见图 3)。 分别使用 psiCHECKgCM_VIM270 和 psiCHECK-pCM_VIM270 来测量引导链和乘客链的 RNA 沉默活性。 分别使用 psiCHECK-gSM_VIM270 和 psiCHECK-pSM_VIM270 来测量引导链和乘客链的种子依赖性脱靶效应 (见注释 4)。

3.3 RNA Silencing Activity Assay Using the Dual Luciferase Reporter System 利用双荧光素酶报告系统进行 RNA 沉默活性测定

RNA silencing activity is measured using dual luciferase reporter assay system.

利用双荧光素酶报告测定系统测量 RNA 沉默活性。

1. Inoculate a human HeLa cell suspension (1.0 × 105 cells/mL) into a well of 24-well culture plate 1 day before transfection.

   在 24 孔培养板的一个孔中接种人类 HeLa 细胞悬浮液（1.0 × 10^5^ 细胞/mL），在转染前一天进行接种。

2. Transfect 0.0005, 0.005, 0.05, 0.5, 5 nM of each siRNA (see Note 5), 0.1μg of pGL3-Control vector encoding the firefly luciferase gene, and 0.1μg of each psiCHECK-gCM_VIM270, psiCHECK-gSM_VIM270, psiCHECK-pCM_VIM270, and psiCHECK-pSM_VIM270 reporter construct simultaneously into HeLa cells using 2μL of Lipofectamine 2000.

   使用 2μL Lipofectamine 2000，将每个 siRNA 的 0.0005、0.005、0.05、0.5、5 nM（见 注释 5），0.1μg 编码萤火虫荧光素基因的 pGL3-Control 载体，以及 0.1μg 每个 psiCHECK-gCM_VIM270、psiCHECK-gSM_VIM270、psiCHECK-pCM_VIM270 和 psiCHECK-pSM_VIM270 报告构建同时转染到 HeLa 细胞中。

3. Lyse the transfected cells with 1× passive lysis buffer 24 h after transfection (see Note 6).

   在转染后 24 小时，用 1×被动裂解缓冲液裂解转染的细胞（见 注释 6）。

4. Measure luciferase activity using the Dual-Luciferase Reporter Assay System.

   使用双荧光素酶报告试剂盒测量荧光素活性。

5. Calculate the percentage of the Renilla luciferase activity normalized by firefly luciferase activity (Renilla luciferase activity/firefly luciferase activity × 100) for unmodified siRNA, and siRNA with DNA (see Note 7) or 2′-OMe (see Note 8) modification in the seed region.

   计算未修饰 siRNA，以及在种子区域具有 DNA（见 注释 7）或 2′-OMe（见 注释 8）修饰的 siRNA 的Renilla荧光素活性相对于萤火虫荧光素活性的百分比（Renilla荧光素活性/萤火虫荧光素活性 × 100）。

3.4 Microarray Analysis

1. Inoculate HeLa cells in each 2 wells of 24-well plate at 1 × 10^5^ cells/mL.

   在 24 孔培养板的每个孔中接种 1.0 × 10^5^ 个/mL 的 HeLa 细胞，每个样品重复 2 孔。

2. Transfect 50 nM of each siRNA into HeLa cells using 2μL of Lipofectamine 2000. Prepare mock-transfected cells treated with the transfection reagent in the absence of siRNA as a control.

   使用 2μL 的 Lipofectamine 2000 将 50 nM 的 siRNA 转染到 HeLa 细胞中。 作为对照，制备模拟转染的细胞，即仅用转染试剂处理而不添加 siRNA。

3. At 24 h post-transfection, purify total RNA from the transfected cells using an RNeasy kit.

   转染 24 小时后，使用 RNeasy 试剂盒从转染的细胞中纯化总 RNA。

4. Measure the concentration of purified RNA using a NanoDrop 2000 spectrophotometer and confirm its quality by a Bioanalyzer (see Note 9).

   使用 NanoDrop 2000 分光光度计测量纯化 RNA 的浓度，并通过 Bioanalyzer 确认其质量（见 注释 9）。

5. Synthesize cDNA from 1μg of each total RNA sample using an Agilent one-color spike mix kit.

   使用 Agilent 单色探针混合试剂盒从 1μg 每个总 RNA 样品合成 cDNA。

6. Hybridize the cDNA product to an Agilent SurePrint G3 human GE microarray according to the manufacturer’s protocol.

   根据制造商的方案，将 cDNA 产物杂交到 Agilent SurePrint G3 人类 GE 微阵列上。

7. Calculate the reduction of the expression levels of target vimentin mRNAs in the cells transfected with each siRNA relative to that in the mock-transfected cells, respectively.分别计算每个 siRNA 转染细胞中靶标波形蛋白 mRNA 表达水平相对于未转染细胞的降低。

8. Generate M (log ratio) and A (mean average) (MA) plot and cumulative distribution using microarray data of each siRNA by package in R-studio software to evaluate the RNA silencing activity (see Note 10) and off-target effect (see Note 11).

   使用 R-studio 软件中的包，通过微阵列数据生成M（对数比率）和A（平均值）（MA）图和累积分布，评估 RNA 沉默活性（见 注释 10）和非靶效应（见 注释 11）。

👀

4 Notes 注释

1. siRNAs satisfied the following four sequence conditions simultaneously are considered to be functional: (a) A/U at the 5′-terminus of the siRNA guide strand, (b) G/C at the 5′-terminus of the siRNA passenger strand, (c) AU richness at the 5′ one-third of the guide strand, (d) absence of a long GC stretch (>9 nt). Bioinformatics analysis revealed that 14.7% of all the 23-mer subsequences in human mRNA satisfy functional siRNA sequence rules (Fig. 2a, b).

   满足以下四个序列条件的 siRNA 同时被认为是功能性的：（a）siRNA 导引链的 5'-末端为 A/U，（b）siRNA 被动链的 5'-末端为 G/C，（c）导引链的前 1/3 位置富含 AU，（d）不存在长的 GC 连续序列（>9 nt）。生物信息学分析显示，人类 mRNA 中所有 23 个核苷酸子序列中有 14.7%满足功能性 siRNA 序列规则（见图 2a，b）。

2. Using options, other siRNA selection algorithms established by Reynolds et al. or Amarzguioui and Prydz can be selected. Furthermore, the highly effective siRNA with reduced seed-dependent off-target effect is selectable using T~m~ value in the seed-target duplex below 21.5 °C. T~m~ value is a good benchmark to express the thermodynamic stability of the duplex between siRNA seed region and target mRNA. Calculated T~m~ value of siRNA seed-target duplex in protein-free condition discriminates siRNA with almost weak off-target effect from that with strong off-target effect. Specificity of siRNA sequence in human, mouse, or rat can be also checked. Other options which impose the number of contiguous GC or AT nucleotides or GC content are also available.

   可以使用 Reynolds 等人或 Amarzguioui 和 Prydz 建立的其他 siRNA 选择算法。此外，可以使用T ~m~值在 21.5°C 以下的 siRNA 选择高效的 siRNA，以减少种子依赖性的非靶效应。T ~m~值是表达 siRNA 种子区域与靶 mRNA 之间的双链热力学稳定性的良好基准。在无蛋白条件下计算的 siRNA 种子-靶双链的T ~m~值可将几乎没有非靶效应的 siRNA 与具有强非靶效应的 siRNA 区分开。还可以检查人类、小鼠或大鼠 siRNA 序列的特异性。还提供了其他选项，例如连续 GC 或 AT 核苷酸的数量或 GC 含量。

3. DNA is a modified form of RNA with hydrogen at 2′-position of the pentose sugar. The 5′ one-third of siRNA is capable of replacement with DNA without substantial loss of RNAi activity. 2′-OMe is a well-known C2′-modified RNA that is known to increase stability in the serum and thermodynamic stability in base-pairing, and also considered to abrogate immunogenicity. 2′-OMe modification is tolerated at multiple positions in the siRNA guide strand because of its small size, comparable to the 2′-OH of natural RNA.

   DNA 是 RNA 的一种修改形式，在五碳糖的 2'-位置具有氢。siRNA 的前 1/3 位置可用 DNA 替换而不会显著降低 RNAi 活性。2'-OMe 是一种众所周知的 C2'-修饰 RNA，它可以增加在血清中的稳定性和碱基对稳定性，还被认为可以减少免疫原性。由于其与天然 RNA 的 2'-OH 大小相似，2'-OMe 修饰可在 siRNA 导引链的多个位置耐受。

4. In RNAi, a target mRNA is reduced by cleavage of target mRNA by AGO2. However, miRNA-mediated RNA silencing is induced by translational repression.

   在 RNAi 中，通过 AGO2 的靶 mRNA 切割来降低目标 mRNA。然而，miRNA 介导的 RNA 沉默是通过转译抑制诱导的。

5. It is preferable to perform reporter assays at a series of multiple concentrations of each siRNA to calculate the IC50 of its RNA silencing activity quantitatively.

   最好在每种 siRNA 的一系列多个浓度下执行报告基因分析，以定量计算其 RNA 沉默活性的IC50。

6. If the cell debris is formed from transfected cells, it is preferable to remove the debris clearly before addition of 1× passive lysis buffer to measure the luciferase activity correctly.

7. 如果由转染细胞形成细胞碎片，则最好在添加 1×无活性裂解缓冲液之前清除碎片，以正确测量荧光素酶活性。

8. Unmodified siVIM-270 repressed the Renilla luciferase activity derived from psiCHECK-gCM for siVIM-270 (psiCHECK-gCM_VIM270) in a dose-dependent manner (Fig. 3). The calculated IC50 was 2 pM (Table 1). DNA-modified siVIM-270 repressed the Renilla luciferase activity at the almost equivalent level to that of unmodified siVIM-270, particularly at high concentrations (0.5 or 5 nM), and their IC50 was 4 pM (Table 1). The off-target effects on the SM target of the guide strand were also measured using psiCHECK-gSM_VIM270 (Fig. 3). Unmodified siVIM-270 showed strong off-target effects (IC50 = 10 pM), but DNA-modified siVIM-270 reduced the off-target effects (IC50 = 295 pM) (Table 1). This may be because thermodynamic stability in base-pairing of DNA-RNA duplex is weaker than that of RNA duplex. In fact, the calculated Tm value of the RNA duplex between the seed region of unmodified siVIM-270 and target RNA was 31.4 °C, and that of DNA-RNA heteroduplex between the seed region of DNA-modified siVIM-270 and target RNA was 3.1 °C. Furthermore, it is possible that the passenger strand also induces unintended RNAi and seed sequence-dependent off-target effects. Then dual luciferase reporter assays were also performed using psiCHECK-pCM_VIM270 and psiCHECK-pSM_VIM270, respectively. However, almost no or little inhibitory effects on both CM and SM targets of the passenger strand were observed (Fig. 3). Because we used siRNA satisfying the four functional siRNA sequence conditions, such siRNA is expected to be easily unwound from the 5′ terminus of the guide strand, but not that of the passenger strand.

   未修改的 siVIM-270 以剂量依赖的方式抑制了从 psiCHECK-gCM_VIM270（siVIM-270 的 psiCHECK-gCM_VIM270 派生的Renilla荧光素酶活性）产生的Renilla荧光素酶活性（见图 3）。计算得到的IC50为 2 pM（表 1）。DNA 修饰的 siVIM-270 在高浓度（0.5 或 5 nM）下抑制了Renilla荧光素酶活性，其 IC50几乎与未修改的 siVIM-270 相当（表 1）。在 psiCHECK-gSM_VIM270 上也测量了引导链 SM 目标的非特定效应（见图 3）。未修改的 siVIM-270 显示出强烈的非特定效应（IC50 = 10 pM），但 DNA 修饰的 siVIM-270 减少了非特定效应（IC50 = 295 pM）（表 1）。这可能是因为 DNA-RNA 双链的碱基配对的热力学稳定性比 RNA 双链要弱。实际上，未修改的 siVIM-270 种子区域与目标 RNA 之间的 RNA 双链的计算Tm 值为 31.4°C，而 DNA 修饰的 siVIM-270 种子区域与目标 RNA 之间的 DNA-RNA 杂交双链的Tm 值为 3.1°C。此外，过客链也可能引发非预期的 RNA 干扰和种子序列依赖的非特定效应。然后，还使用 psiCHECK-pCM_VIM270 和 psiCHECK-pSM_VIM270 进行了双荧光素酶报告基因测定，分别用于观察过客链的 CM 和 SM 目标的抑制效果。然而，几乎未观察到对过客链的 CM 和 SM 目标的抑制作用（见图 3）。因为我们使用了满足四个功能性 siRNA 序列条件的 siRNA，所以预期这种 siRNA 易于从引导链的 5'末端解开，但不易于从过客链解开。

9. The 2′-OMe-modified siVIM-270 repressed the Renilla luciferase activity derived from psiCHECK-gCM for siVIM-270 (psiCHECK-gCM_VIM270) at the almost equivalent level to that of unmodified siVIM-270, and its IC50 was 10 pM (Table 1). The off-target effects on the SM target of the guide strand of 2′-OMe-modified siVIM-270 measured using psiCHECK-gSM_VIM270 was significantly reduced, even though this modification is well-known modification to enhance base-pairing stability. It is known that the phosphates of RNA oligonucleotides interact with the amino acid side chains of human AGO protein. Therefore, 2′-OMe-modified single-stranded RNA structure on the AGO protein was examined computationally (Fig. 4). The 2′-OMe modification in the siRNA seed region at position 3 was essentially stable on the AGO protein. However, RNA structure of the nucleotide adjacent to the 2′-OMe-modified nucleotide was changed from that of unmodified RNA (Fig. 4a-d), suggesting that 2′-OMe modification in the siRNA seed region at least at position 3 from 5′ end may disturb base-pairing between siRNA guide strand and its target or off-target mRNAs. But, as shown in Fig. 3, the expression of the CM target was significantly repressed, suggesting that the non-seed region may compensate the incomplete seed-target base pairing for inducing RNAi on the CM target.

   2′-OMe 修饰的 siVIM-270 在抑制源自 psiCHECK-gCM_VIM270 的Renilla荧光素酶活性方面几乎与未修改的 siVIM-270 相当，其 IC50 为 10 pM（表 1）。使用 psiCHECK-gSM_VIM270 测量的 2′-OMe 修饰的 siVIM-270 引导链 SM 目标的非特定效应显着减少，即使这种修饰是已知的增强碱基配对稳定性的修饰。众所周知，RNA 寡核苷酸的磷酸与人类 AGO 蛋白的氨基酸侧链相互作用。因此，通过计算分析了 AGO 蛋白上的 2′-OMe 修饰的单链 RNA 结构（见图 4）。在 AGO 蛋白上，siRNA 种子区域中位置 3 的 2′-OMe 修饰基本稳定。但是，与未修饰的 RNA 相比，2′-OMe 修饰的核苷酸相邻核苷酸的 RNA 结构发生了变化（见图 4a-d），表明至少从 5'端的位置 3 处的 siRNA 种子区域的 2′-OMe 修饰可能会干扰 siRNA 引导链与其靶向或非特定 mRNA 之间的碱基配对。但是，如图 3 所示，CM 目标的表达显着受到抑制，表明非种子区域可能会补偿不完整的种子-靶标碱基配对以诱导对 CM 靶标的 RNA 干扰。

10. Particularly, it is recommended that the value of RNA Integrity Number (RIN), a quantitative measurement of RNA quality, of total RNA sample for microarray application is high. High-quality samples with RIN values higher than 9 are best for microarray application, but the reliable results are also obtained from the sample with the value as low as 7.0.

    特别建议微阵列应用的总 RNA 样品的 RNA 完整性数值（RIN），这是 RNA 质量的定量测量值，应该很高。RIN 值高于 9 的高质量样品最适合于微阵列应用，但从 RIN 值低至 7.0 的样品中也可以获得可靠的结果。

11. Unmodified siVIM-270 repressed the expression level of vimentin mRNA to 11%. DNA- and 2′-OMe-modified siVIM-270 repressed vimentin mRNA expression at almost equivalent level to 13% and 19%, respectively (Fig. 5a).

    未修饰的 siVIM-270 将波形蛋白 mRNA 的表达水平降低到 11％。DNA 和 2'-OMe 修饰的 siVIM-270 将波形蛋白 mRNA 的表达水平降低到几乎等同于 13％和 19％的水平（见图 5a）。

12. The seed-dependent off-target effect of each siRNA using microarray data can be evaluated by analyzing the expression profiles of genes with seed matched sequences in their 3′UTRs, since siRNA off-target effect is mainly induced by the mechanism similar to miRNA-mediated RNA silencing. Off-target effects on mRNAs having complementary sequences with siRNA seed region in their 3′UTRs were examined. The obvious off-target effects were observed by the introduction of unmodified siVIM-270. However, DNA-modified siVIM-270 showed reduced off-target effects. On the other hand, 2′-OMe-modified siRNA showed little off-target effects (Fig. 5b-d).

13. 利用微阵列数据评估每个 siRNA 的种子依赖性非靶效应可以通过分析其 3'UTR 中具有种子匹配序列的基因的表达谱来进行，因为 siRNA 非靶效应主要是通过与 miRNA 介导的 RNA 沉默类似的机制诱导的。通过引入未修饰的 siVIM-270 观察到明显的非靶效应。然而，DNA 修饰的 siVIM-270 显示出减少的非靶效应。另一方面，2'-OMe 修饰的 siRNA 显示出很少的非靶效应（见图 5b-d）。

Table 1 IC50s of unmodified, DNA-modified, or 2′-OMe-modified siVIM-270 for RNAi activity or off-target effect measured by reporter assay未修饰、DNA 修饰或 2'-OMe 修饰的 siVIM-270 对于 RNAi 活性或通过报告基因测定测量的脱靶效应的IC50s

∞ 表示 IC50 远超过 5000 pM