br Among the current available assays

Among the current available assays, conventional Northern blotting and microarray analysis generally have insufficient sensi-tivity for miRNAs at low expression level [7,8]. Furthermore, both of them only provide a limited degree of qualitative results [9]. Small difference in low-abundant miRNA levels cannot be easily dis-cerned [10]. Therefore, strategies for amplifying the detection have become a key component to achieve high sensitivity and resolution. Currently, amplification methods for nucleic Dehydroepiandrosterone fall into three categories, including target amplification, probe amplification and signal amplification [11]. Quantitative RT-PCR (qRT-PCR) is a typical target amplification method and is often considered as the gold standard for miRNA quantification [12]. This method copies the target miRNA exponentially and thus the short length of primer may lead to high nonspecific amplification background, which normally limits quantification accuracy, especially for extremely low copies of miRNAs [13]. In the past few years, several innovative probe amplification approaches have emerged [14e16]. However, nonspecific effect is still the issue because the amplification object changes solely from the target miRNA to its complementary probe. Therefore, signal amplification strategies that can circumvent copying of nucleic acid sequence are coming into view, for example, branched probes [11,17]. Among those potential branched probes, dendrimers are highly branched macromolecules with nanometer-scale dimensions [18]. Multiple reporter molecules can be attached to the branches of dendrimers and released in certain circum-stances. This so-called self-immolative process can amplify initial signal by a factor of four, eight, sixteen or more, depending on the number of branches in dendrimer [19]. In most cases, colored molecules were employed as reporters, but they were susceptible to various factors [20,21]. In the present study, peptide was selected as reporter molecule and ultimately quantified using liquid chromatography-tandem mass spectrometry (LC-MS/MS) (i.e., quasi-targeted proteomics [22]), which can offer sensitive and ac-curate detection [23].
In the present study, we circumvented copying of target nucleic acid sequence and designed a novel DNA-peptide dendrimer (DPD) probe for amplification of target miRNA signal (Fig. 1). This den-drimer probe contains three functional domains, including sub-strate peptides containing eight reporter peptides and tryptic cleavage sites, branched dendrimer scaffold and a piece of DNA sequence for complementary binding of the target miRNA. Prior to hybridization, the miRNA was biotinylated and immobilized on streptavidin agarose. Afterward, the probe was hybridized with the target miRNA (i.e., miR-21). After the treatment with trypsin, the reporter peptide was liberated and finally quantified using LC-MS/ MS. In this way, the signal of miRNA was converted and amplified into the mass response of the reporter peptide molecules (i.e., 
quasi-targeted proteomics [22,24]). In particular, design, prepara-tion and characterization of the DPD probe were paid careful thought and evaluation. Lastly, miR-21 level in 3 breast cancer cell lines was quantified, and so as well as 102 pairs of human breast primary tumors and adjacent normal tissue samples. The obtained results were compared with the quantitative reverse transcription PCR (qRT-PCR) results. The miR-21 expression in tissue was also evaluated according to histopathological features, molecular sub-types and prognosis of breast cancer.
2. Materials and methods
2.1. Chemicals and reagents
For detailed information, please see the supplementary material.
2.2. Cell culture and tissue collection
Breast cell lines MCF-10 A and breast cancer cell lines MCF-7/WT were obtained from the ATCC and grown routinely in DMEM media with 1% penicillin/streptomycin and 10% fetal bovine serum under a 5% CO2 atmosphere at 37 C. MCF-7/ADR (Keygen Biotech, Nanjing, China) were cultured in a RPMI 1640 media (with L-glutamine and sodium pyruvate). Cells were subcultured at a split ratio of 1:3 every 5 days using 0.25% trypsin and maintained by the addition of fresh medium. MCF-7/ADR cells was cultured with 1 mg mL 1 DOX and periodically reselected to keep a highly drug-resistant cell population. The cells was incubated without DOX for 48 h before experiments. Cell counting was performed with a hemocytometer.