Breast cancer has surpassed lung cancer to become the most common type of tumor [1]. Triple-negative breast cancer (TNBC), which accounts for 15% to 20% of all breast cancers [2], exhibits poor response to conventional endocrine therapy and targeted therapy for lacking a clearly defined therapeutic target. Currently, surgery and chemotherapy remain the primary clinical treatments [3]. It is worth mentioning that there are certain DNA damage repair disorders in TNBC, indicating increased sensitivity to cisplatin (Cis) [[4], [5], [6]]. And Cis-plus is an efficient combination chemotherapy choice for patients with metastatic TNBC [7]. In general, long-term use of Cis contributes to drug resistance, which is the main factor limiting its clinical application [8].

The intractable nature of TNBC is highly depend on a complex and diverse TME [9], among which abundant stromal elements, specially cancer-associated fibroblasts (CAFs) occupy a dominant position [10]. The CAFs act as sensory receptors and inducers of tumor activation signals to exert multiple effects, including cancer progression, metastasis and recurrence promotion [11]. In addition, CAFs are critically involved in Cis resistance in TNBC [[12], [13], [14]]. There are three “shields” set up by CAFs that prevent drugs from reaching and inhibiting the “domain” of tumor. Firstly, the dense extracellular matrix (ECM) programmed by CAFs avoids drugs attack and immune cells infiltration. The CAFs secrete several ECM components, including fibronectin and collagen that can solidify ECM [15,16]. Besides, the CAFs also recruit endothelial progenitor cells to the tumor site to establish the tumor-surrounding network of vasculature [17,18]. Rapid tumor proliferation leads to the formation of numerous new vessels. However, these new vessels exhibit abnormal growth and high leakage, which in turn increase the interstitial pressure at the tumor site. Consequently, the efficiency of drug delivery at the tumor site is reduced [19]. CD31 is regularly employed as a marker to assess tumor neovascularization [20]. Therefore, the dense stroma-vascular barrier appears to inhibit the infiltration of immune cells and affect anti-tumor drug penetration [21]. Secondly, epithelial mesenchymal transition (EMT) appears to play a crucial role in the occurrence, metastasis, and drug resistance of TNBC [22], and it has been reported that CAFs-derived cytokine such as recombinant human TGF beta 1 protein (TGF-β) induces the EMT process, which results in migration and invasion phenotypes, further promoting the metastasis of TNBC [23]. Also, degradation of the basement membrane by CAFs-generated matrix metalloproteinases (MMPs) is a favorable condition for metastasis [24]. Interestingly, CAFs-induced EMT increases cancer stem cell markers expression accompanied by extending self-renewal capability, which tumor growth and metastasis depend on. Thirdly, abnormal CAFs secrete a variety of immunosuppressive factors including TGF-β, nitric oxide and prostaglandin E2, which may organize and recruit immunosuppressive cells (myeloid-derived suppressor cells, regulatory T cells (Treg), and macrophages) to turn into tumor-promoting immunosuppressive environment for conniving tumors to escape from immune surveillance [25]. Similarly, cytotoxic T lymphocyte–mediated killing of tumor cells are suppressed [26,27]. It must be noted that this behavior forms a negative feedback loop to exhibit immunosuppressive effects and generate a tolerogenic environment.

Inspired by the above conditions on TNBC therapy, we here proposed a systemic treatment strategy for targeting CAFs to further wipe out three “shields”, thereby interrupting the vicious circle between CAFs and TNBC progression. We constructed polylactic-glycolic acid copolymer (PLGA) nanoparticles coated with anisamide (AA) modified red blood cell membrane (Rm) and encapsulated silybin (SLB), which named ARm@SNP. The AA is commonly used as a CAFs-targeting molecule in functionalized nano-drug delivery due to its high affinity to sigma-1 receptors [[28], [29], [30], [31], [32]], which is used as an anchor for locating the system accurately on CAFs, serving as the prerequisite for the subsequent inhibition of the “shielding” function of CAFs. The Rm coating makes the nanoparticles less susceptible to phagocytosis by the monocyte-macrophage system, and attributes to long-cycle features with sound safety [33,34]. The SLB, a natural active ingredient extracted from milk thistle seeds, is a flavonoid lignan compound [35] with anti-inflammation, anti-oxidation, and anti-fibrosis activities. Previous research has shown that SLB can inhibit the secretion of monocyte chemoattractant protein-1 (MCP-1), impede fibroblast activation, reduce immune cell recruitment [36], inhibit the production of fibronectin by CAFs and induce apoptosis [37]. Additionally, the migration and invasion of breast cancer, ovarian cancer and prostate cancer which induced by multi-pathways are effectively attenuated by SLB through the suppression of EMT [[38], [39], [40]]. Thus, the ARm@SNP was designed to initially target CAFs and subsequently normalize them by harnessing the pharmacological properties of SLB to weaken the stromal barrier, intervene in tumor invasion and metastasis, reverse the immunosuppressive microenvironment through multiple ways and heighten the anti-tumor efficacy of Cis against TNBC.

In order to enhance the ability of targeting CAFs and break the three “shields” shaped by CAFs, ARm@SNP was used in combination with Cis (ARm@SNP/Cis) to achieve an improved effect on treating TNBC. The effect of ARm@SNP on reversing Cis resistance was investigated in a conditioned medium (CM) induced 4T1 resistant model in vitro. The targeting ability of ARm-modified nanoparticles containing the fluorescent probe was investigated both in vitro and in vivo by high content microscopy and IVIS system. Furthermore, the anti-TNBC effect of ARm@SNP/Cis and the mechanism involved in getting over the triple “shields” were explored in a 4T1 tumor-bearing mice model with safety evaluated.