Enlibio,Your biological research partner
Xinqidi Biotech Co.,Ltd,Wuhan,China 2008-2022
R&D 14th year

Peptide nano-blanket impedes fibroblasts activation and subsequent formation of pre-metastatic niche

Issuing time:2022-05-31 15:03

Abstract

There is evidence to suggest that the primary tumor induces the formation of a pre-metastatic niche in distal organs by stimulating the production of pro-metastatic factors. Given the fundamental role of the pre-metastatic niche in the development of metastases, interruption of its formation would be a promising strategy to take early action against tumor metastasis. Here we report an enzyme-activated assembled peptide FR17 that can serve as a “flame-retarding blanket” in the pre-metastatic niche specifically to extinguish the “fire” of tumor-supportive microenvironment adaption. We show that the in-situ assembled peptide nano-blanket inhibits fibroblasts activation, suppressing the remodeling of the metastasis-supportive host stromal tissue, and reversing vascular destabilization and angiogenesis. Furthermore, we demonstrate that the nano-blanket prevents the recruitment of myeloid cells to the pre-metastatic niche, regulating the immune-suppressive microenvironment. We show that FR17 administration effectively inhibits the formation of the pulmonary pre-metastatic niche and postoperative metastasis, offering a therapeutic strategy against pre-metastatic niche formation.

Introduction

Though therapeutic outcomes and survival rate for patients with various cancers have been greatly improved in last decades, effective treatments for patients with metastatic cancer are still limited1. Though emerging technologies provided early detection of malignant transformation or even warning of high metastatic risk by biomarker screening before the actual occurrence of metastasis in clinic2,3,4,5,6, clinical treatment on metastasis prevention lags far behind. The contemporary therapeutic strategies against metastasis in clinic, including systemic chemotherapy, radiotherapy and immunotherapy, mainly focus on the later time period of metastasis development, or at least after the arrival and colonization of disseminated tumor cells to the distal organs, which have gained unsatisfied clinical outcomes7,8. Observations of the adjuvant chemotherapy resistance6 and treatment-related toxicities9 revealed the shortcoming of the currently available strategies10.

Growing evidence illustrated that an inflammatory, neovascularized, immunosuppressive, tumor-supportive microenvironment has emerged before the arrival and colonization of disseminated tumor cells, which is termed as pre-metastatic niche (PMN)11. Relevant studies revealed that complex interactions between multiple participators and alteration in regulative pathways energized the construction of PMN, such as primary tumor-derived cytokines and exosomes, myeloid-derived suppressor cells (MDSCs), and the tumor re-educated stromal environment including pre-metastasis associated fibroblasts, destabilized vasculature and extracellular matrix (ECM)12,13 (Fig. 1a). In 2005, Lyden et al. brought to light the recruitment of VEGFR+ myeloid progenitor cells to PMN by localized fibronectin (FN) deposition14, which would be over-produced by activated resident fibroblasts. This specific cell population and its subtypes were then unified and classified as MDSCs with its potent capability to suppress immune responses15, who make major contributions in developing immunosuppressive microenvironment via activation of nitric oxide (NO) signaling or reactive oxygen species (ROS) pathway16. What’s more, recent studies revealed the irritation of stromal cells especially fibroblasts in primary tumor and in distal site induced by tumor-derived secreted factors via STAT3 signaling17 and JNK signaling18pathways. It is reported that tumor-educated fibroblasts serve to reconstruct ECM, induce angiogenic and pro-inflammatory response of endothelial cells, preparing a tumor-supportive host stroma19, which gives a clue to the tipping point of PMN initializing as fibroblast activation.

Fig. 1: Schematic illustration of peptide nano-blanket impedes fibroblasts activation and subsequent formation of pre-metastatic niche.
figure 1

a Illustration of the pathological process of pre-metastatic niche (PMN) formation. The primary tumor produces pro-metastatic factors, such as tumor-derived secreted factors (TDSF), to induce fibroblast activation in metastatic destination organs. The tumor-educated activated fibroblasts serve to construct a metastasis-supportive host stroma, including extracellular matrix (ECM) deposition and reconstruction, angiogenesis and vascular leakage, as well as myeloid-derived suppressor cells (MDSC) recruitment. b After FR17 subcutaneous administration, the in-situ assembled peptide nano-blanket in PMN stromal microenvironment impedes fibroblasts activation so as to retard stromal and vessel pro-metastatic reconstruction, inhibiting MDSC recruitment and metastatic cascades.

Since PMN was considered as the foundation laid for circulating tumor cells colonization and one of the vital premises to develop metastasis in distant organs, we wonder if the early process of tumor metastasis can be terminated or even totally prevented by interrupting the formation of PMN. To view the entire process of PMN establishment and metastasis development as a progress of the occurrence and the spread of a huge forest fire, it would be more efficient to contain and beat out the local flame by preventing the formation of PMN.

Inspired by the self-assembled peptides found in many natural life processes, researchers have modified, designed and synthesized diverse self-assembled peptides applied as functional biomaterials for various applications20,21,22. In the application on anti-tumor and anti-metastasis therapy, enzyme-responsive self-assembly or ligand-receptor interactions-triggered morphology transforming of the peptide nanofibrils and hydrogel have been developed to induce cell death23,24,25, to restrict tumor cell invasion26, to serve as biocompatible drug delivery platforms27, to achieve specific targeting of tumor28 and imaging29 as well. One of the research branches, which has been widely applied, was based on the β-sheet regions of Amyloid-β (Aβ)30. As the original fragment of Aβ16-20, KLVFF has been applied as therapeutic agents26,31, imaging agents32 or delivery platform33 in the field of Alzheimer’s disease34 and cancer35. Further researches cut the peptide fragment down to dipeptide FF to obtain discrete nanotubes through self-assembly in aqueous solution36. Afterwards, felicitous modification and re-designment on FF with perfect biocompatibility has been reported with a wide range of applications37. For example, FFKF was developed to construct drug delivery system which can be completely degraded by cathepsin proteases27. Yang and his team employed FFFK to develop molecular hydrogel for codelivery of anti-cancer drugs38. In another work, the application of FFYK was explored in organelles targeting and cancer cell killing39. Some introduced naphthyl group on N-terminal of Phe to favor the intermolecular hydrophobic interactions of FFKY40.

Here we report an enzyme-activatable assembled peptide FR17 that can serve as a “flame-retarding blanket” at PMN site specifically, containing and suppressing the “fire blaze” of PMN formation to further develop into overt metastasis (Fig. 1b). As a substrate of matrix metalloproteinase 2 (MMP2), FR17 is designed to release self-assembled monomer FG8 to construct peptide nano-blanket in PMN stromal microenvironment. The self-assembly feature of FG8 is provided by its backbone, FFKY, taking advantages of both the self-assembly feature of diphenylalanine structural motif FF with the assistance of Y and the editable site provided by K to combine with hydrophilic fragment via enzyme-cleavable linker. The in-situ assembled peptide nano-blanket impedes fibroblasts activation so as to prevent metastatic cascades. In this work, we demonstrate the subsequent impact of inhibition on fibroblasts activation induced by FR17 intervene, revealing the underlying mechanism on cellular interactions among fibroblasts, vascular endothelial cells and extracellular components, and the intervention on PMN recruited MDSCs via in vitro and in vivo experiments.

Results

Peptide nano-blanket transformed from FR17

The MMP2-activatable self-assembled branched peptide (FR17, FFK(GPLGLAGG-YVDKR)Y) consists of (1) the backbone of a self-assembled peptide domain Phe-Phe-Lys-Tyr (FFKY), a variant of Phe-Phe (FF), which is derived from Aβ36; (2) thymopentin (TP5, Arg-Lys-Asp-Val-Tyr, RKDVY), a pentapeptide with perfect hydrophilic property and immune modulation effect, which is applied as the adjunctive therapeutic agent on cancer treatment in clinic to prevent postoperative infection and to activate immune response41,42. TP5 was conjugated to the side-chain of Lys (K) in the main-chain FFKY with the MMP2-cleavable peptide linker (Gly-Pro-Leu-Gly-Leu-Ala-Gly-Gly, GPLGLAGG)43, increasing the hydrophilic property of the entire peptide molecule. Furthermore, the sequence of TP5 in peptide FR17 was replaced by the scrambled pentapeptide of TP5 without immunomodulatory bioactivity, i.e. DVYKR, to form the scrambled group (sFD17, FFK(GPLGLAGG-RKYVD)Y) as peptide assembly control.

Peptides were synthesized via Fmoc solid-phase peptide synthesis technology and the peptide sequences were verified by mass spectra (Supplementary Figs. 1 and 2). When specifically cleaved by MMP2, both FR17 and sFD17 are able to release self-assembled monomer FG8 (FFK(GPLG)Y) (Supplementary Fig. 3), which would spontaneously fold into membrane-like structure, i.e. the peptide nano-blanket, both in neutral and weak acidic conditions, which simulates the pH condition of inflammatory microenvironment in PMN44,45 (Supplementary Fig. 4). The transmission electron microscopy (TEM) images (Fig. 2a) and the Cryo-TEM image (Supplementary Fig. 5a) showed the lamellar structure of the peptide self-assemblies formed by enzymatic degradation (Supplementary Fig. 5b) dispersed with an average diameter of ~500 nm (Fig. 2b). The suspected appearance of peptide nano-blanket has been observed in the intercellular substance in PMN lung in vivo (Supplementary Fig. 5c), and the morphology and the size of which ranges from hundreds to thousands nanometers depending on the intercellular space specifically at PMN sites. Besides, the aggregation-induced emission (AIE) effect was employed to monitor the spontaneous aggregation of FG8 in aqueous system (Supplementary Fig. 6). Fluorescent dots observed in the organs’ sections revealed the responsive-assembly of the peptide nano-blanket in PMN lung in vivo (Supplementary Fig. 7) with few aggregations in PMN liver and no observation of AIE effect in other organs (Supplementary Fig. 8). Taken together, these results indicated the self-assembly property of FG8 monomer, and MMP2-cleaved release of FG8 from FR17 or sFD17 to form the peptide nano-blanket both in vitro and in vivo.

Fig. 2: Enzyme-activated self-assembly of FR17 and all-atom molecular dynamics (MD) simulation of the self-assembly of FG8.

a TEM images of FR17 or sFD17 (500 µM) treated with MMP2 (scale bar = 200 nm), FG8 (500 µM, scale bar = 100 nm) and FFKY (500 µM, scale bar = 100 nm). The lamellar structure of the peptide nano-blanket was pseudo-colored in gold. b Size distribution of FR17 and sFD17 (500 µM) after MMP2 cleavage. c Molecular structure of FFKY. d The FFKY assemblies generated at t = 200 ns of MD simulation of 16 FFKY molecules in water (containing NaCl for charge neutralization). e Typical molecular cluster and the non-bonded interactions involved in FFKY assemblies in detail. f Molecular structure of FG8. g The FG8 assemblies generated at t = 200 ns of MD simulation of 16 FG8 molecules in water (containing NaCl for charge neutralization). h Typical molecular cluster and the non-bonded interactions involved in FG8 assemblies in detail. The dashed purple lines denote the π-π stacking. The dashed blue lines denote the hydrogen bonds. The nitrogen atoms are labeled in blue. And the oxygen atoms are labeled in red. Source data are provided as a Source Data file.

The peptide assembly relies on the non-bonded interactions, typically hydrogen bonds and π-π stacking, between adjacent peptide molecules. The self-assembling pattern of FG8 changed due to the branch modification with GPLG side-chain. The all-atom molecular dynamics (MD) simulation of FFKY-assemblies and FG8-assemblies gives a microscopic account of the impacts of branch modification on peptide interactions. The MD simulation revealed that there are more intermolecular hydrogen bonds involved in FG8 cluster, which are more ordered and mostly formed between Phe of adjacent FG8 molecules (Supplementary Fig. 9). While the intermolecular hydrogen bonds involved in FFKY-assemblies are less-formed and disordered (Fig. 2c–h).

The pathological process in the MCM-induced PMN model in vivo

In order to study the effect of FR17 administration on the process of PMN development, an in vivo PMN model induced by melanoma-conditioned media (MCM) has been established according to the previous research on PMN14. Briefly, tumor-derived factors secreted from primary tumor were replaced by MCM intraperitoneal injection to the mice for 10 consecutive days from day 1 to 10. On day 7, when the tumor-supportive microenvironment was successfully established in the lung, B16F10 melanoma cells were intravenously administrated as the simulation of circulative tumor cells (CTC) wandering through blood vessels and some would successfully colonize in the prepared “fire scene” in lung (Supplementary Fig. 10b).

The pathological process in the lung was assessed from various aspects during MCM-induced PMN establishment from day 3 to day 13 (Supplementary Figs. 10 and 11). Important cell derived molecular components and cytokines were closely monitored during the process. The higher expression of the ECM component FN was generated by MCM inducement. Moreover, matrix metalloproteinase 9 (MMP9) and vascular endothelial growth factor a (VEGFa) were up-regulated along with the pathological progress of lung PMN from day 0 to day 10 and reached a plateau on day 10 (Supplementary Fig. 10a). An accordant trend was found on MMP2 level as well (Supplementary Fig. 10d). Meanwhile, the immune cell population analysis during the pathological process in the lung of MCM-induced PMN model was conducted by flow cytometry, which drew our attention to the crime culprit-cell inducing PMN, MDSC, for the recruitment of MDSC increased through the timeline (Supplementary Figs. 11 and 12). It is reported that MDSC contributes a lot in developing the immunosuppressive microenvironment in PMN, preparing more suitable and fertile land for tumor cells to take root in refs. 13,46. Expression level of the major inflammatory mediator TGF-β1, possibly produced by MDSCs, increased gradually but sharply. Typical biomarkers produced by MDSC to exert immunosuppressive effect, in other words, the incriminating tools for the microenvironment re-education, were also detected, i.e. ROS, iNOS and arginase-1 (Supplementary Fig. 10a, c)16. Major characteristics of PMN also emerged in lung as time went by, such as activation of fibroblasts (Supplementary Fig. 10e, f), angiogenesis (Supplementary Fig. 10g) and increase of vascular permeability (Supplementary Fig. 10h)12. All these data indicated that the in vivo PMN model had been well-established, reflecting the pathological process of PMN development. With the overall support provided by PMN, it would consequently accelerate and aggravate metastasis in vivo (Supplementary Fig. 13).

FR17 administration interrupts the activation of fibroblast induced by tumor-derived factors

According to previous researches on tumor metastasis18 and our assessment on MCM-induced PMN mice model, the activation of the resident fibroblasts in distal tissues could be regarded as the tipping point of the beginning of PMN establishing, raising the alarm about laying the foundations of the potential metastasis. It is reported that tumor-educated fibroblasts would serve to construct a tumor-supportive host stroma via STAT3 signaling17 and JNK signaling18 pathways, promoting ECM degradation and reconstruction, inducing angiogenic and pro-inflammatory response of endothelial cells, recruiting VLA-4+ bone marrow-derived cells by localized FN deposition for niche formation19. Indeed, lysyl oxidase (LOX) cross-linked collagen surrounding pre-metastasis associated fibroblasts attracts CD11b+ myeloid cells invasion in destination organs, which corresponds to metastatic efficiency47.

Here in a cell model simulating the impact of secreted factors derived from primary tumor on lung fibroblasts in vitro (Fig. 3b), FR17 cultivation along with MCM stimulation showed an interruption on tumor-derived factors irritating mouse lung fibroblasts (MLF) (Fig. 3a). The treatment with FR17 or sFD17, which would form peptide nano-blanket assembled by the monomer FG8 that was released by enzyme-induced cleavage, successfully prevented the activation of fibroblasts by MCM (Supplementary Fig. 14). The inhibition on the expression of pathologic fibroblast biomarker alpha smooth muscle actin (αSMA) as well as the fibroblasts proliferation and cellular bio-functions (Fig. 3c–i) also substantiated the blocking of lung fibroblasts activity by FR17 or the scrambled control sFD17. The reverse of the increase in cytokines secretion, like MMP9, VEGF, FN, was determined by ELISA kit (Fig. 3d). The qPCR results further suggested that the peptide-assemblies might exert biological regulatory effect on cells while not just acting like a physical shield to wrap on the surface of cells (Fig. 3e). Besides, when irritated by tumor-derived factors to present the activated αSMA+ phenotype, the migration ability as well as the collagen gel contracting function of fibroblasts got promoted. By contrast, cultivation with FR17 minified these functional changes to a large extent (Fig. 3f–i). Administration of FR17 to the in vivo PMN model also gave great relief to the activation of the fibroblasts in lung (Fig. 3k). What’s more, the above data also suggested that the “flame-retarding” effect on fibroblast activation was almost completely contributed by peptide nano-blanket on account of the similar outcomes of the scrambled control sFD17 to that of FR17 treatment.

Fig. 3: FR17 interrupts the activation of fibroblast induced by tumor-derived factors.
figure 3

a Schematic drawing illustrates the in-situ assembled peptide nano-blanket interrupts tumor-derived factors-induced activation of fibroblast. b Illustration of the protocol of MCM stimulation and peptide treatment on MLF in vitro. c Cell proliferation of MLF after MCM stimulation and peptide treatment. n = 4. d Secretion of MMP9 (n = 4), VEGF (n = 4 for peptide treated groups and 3 for the control groups) and fibronectin (n = 4) in the culture media of MLF after stimulated by MCM, treated with or without peptide. e qPCR analysis of Acta2 (i.e., αSma), Mmp9, Vegfa, Fibronectin1 (Fn1) expression in MLF after MCM stimulation and peptide treatment. n = 3. For Mmp9: MCM free vs. MCM, p = 0.000000000011; MCM free vs. TP5, p = 0.00000000000056; MCM vs. sFD17, p = 0.0000000000037; MCM vs. FR17, p = 0.0000000000013; TP5 vs. sFD17, p = 0.00000000000017; TP5 vs. FR17, p = 0.000000000000071. f, g Migration assay to evaluate MLF migration ability after MCM stimulation and peptide treatment. Scale bar = 100 µm. n = 4. h, i Collagen gel contract assay to evaluate contracting function of MLF after MCM stimulation and peptide treatment. n = 3. j Expression level of fibronectin in the lung harvested from the PMN model mice administrated with different peptides on day 10. k Representative images and semi-quantification of αSMA+ fibroblasts in the lung harvested from mice administrated with different peptides on day 10. Scale bar = 50 µm. n = 6. Data are presented as mean ± SD for all column charts. One-way ANOVA followed by Tukey’s multiple comparisons test was employed for statistical evaluation for ce, g, i, k. Source data are provided as a Source Data file.

The stromal microenvironment, apart from fibroblasts, consisting of ECM and vasculature, could also be re-educated by tumor-derived factors48. For ECM would have gone through remodeling to rebuild a supportive niche for tumor colonization in PMN model, Western blot analysis, Masson staining and Sirius Red staining revealed the down-regulation in the expression of FN (Fig. 3j) and collagen (Supplementary Fig. 15a–c) in the lungs of FR17 group and sFD17 group, which would have likely been over-expressed by the activated fibroblasts otherwise. Besides, versican expression in the lung was also lower in the FR17 or sFD17 intervened groups than that of PMN control group and TP5 treated group (Supplementary Fig. 15d), and the elevated expression of which has been reported to contribute to angiogenesis in tumor49.

FR17 protects fibroblasts from activation to inhibit vascular leakage and angiogenesis

The activated fibroblasts in primary tumor, also known as cancer-associated fibroblasts (CAFs), have been reported to produce proangiogenic factors, such as VEGF, so as to promote tumor angiogenesis17. What’s more, the secretion of MMPs by CAFs induced tumor vascular leakage, exacerbating MDSCs and tumor cells intravasation into the vascular system. Therefore, we assumed that, as an important participator and vanguard in the construction of PMN, the activated fibroblast may also contribute to angiogenesis and the increased vascular permeability in PMN (Fig. 4a). Further exploration was carried out on mice vascular endothelial cells to find out whether fibroblasts promote angiogenesis and vascular permeability during the arousement induced by MCM (as sketched in Supplementary Fig. 16a). The tube forming assay demonstrated the proangiogenic capability of fibroblasts activated by MCM (Fig. 4b). Meanwhile, the transwell permeability assay (as illustrated in Supplementary Fig. 16b) to mimic the inner layer of blood vessel in vitro verified that the fibroblast-conditioned medium (FCM) collected from activated MLF increased vascular permeability (Fig. 4c), indicated directly by the dismission of endothelial adherence junctions mediated by vascular endothelial cadherin, VE-cadherin (Fig. 4d).

Fig. 4: FR17 administration protects fibroblasts from activation to inhibit vascular leakage and angiogenesis.
figure 4

a Illustration of the inhibition of vascular leakage and angiogenesis via the peptide nano-blanket protection on fibroblasts from tumor re-education. b Tube forming assay evaluating endothelial cells’ neovascularization when treated with conditional FCM. Scale bar = 100 µm. The semi-quantification was calculated from five random fields imaged per well from triplicate wells per group via ImageJ. Data are presented as mean ± SD. c Permeability of the endothelial cell layer in vitro when co-cultured with MCM and peptides pre-treated MLF. Data are presented as mean ± SD. n= 3. MCM free vs. MCM, p = 0.0104; MCM free vs. TP5, p = 0.0192; MCM vs. sFD17, p = 0.0072; MCM vs. FR17, p = 0.0168; TP5 vs. sFD17, p = 0.0132; TP5 vs. FR17, p = 0.0314. d The endothelial cell monolayer’s integrality after cultivated with conditional FCM, indicated by VE-cadherin. The white box and arrows indicate the adherens junction between endothelial cells. While the yellow box and arrows indicate the disruption of adherens junction. Scale bar = 50 µm. e Expression level of VEGFa, ANG2, MMP9 and MMP2 in the PMN lungs harvested from mice administrated with different peptides on day 10. f, g Vascular permeability in the PMN lung after peptide administration. Scale bar = 100 µm. Data are presented as mean ± SD. n = 6. Healthy vs. Ctrl, p = 0.0016; Healthy vs. TP5, p = 0.0013; Ctrl vs. sFD17, p = 0.0002; Ctrl vs. FR17, p = 0.0003; TP5 vs. sFD17, p = 0.0001; TP5 vs. FR17, p = 0.002. One-way ANOVA followed by Tukey’s multiple comparisons test was employed for statistical evaluation for b, c, f. Source data are provided as a Source Data file.

For all experiments conducted on endothelial cells, the different conditional FCM was obtained from fibroblasts pre-treated with FR17, sFD17 or TP5 on interrupting MCM stimulation separately. The conditional FCM pre-treated with FR17 or sFD17, rather than TP5, offset the increase in bEnd3 cell proliferation induced by the indirect stimulation of MCM (Supplementary Fig. 16c). The acceleration in neovascularization and the disruption on endothelial cell-cell connection to cause vascular leakage were made up by indirect FR17 or sFD17 treatment on MLF as indicated in Fig. 4b, c.

To examine the influence of FR17 intervention on the expression level of the proangiogenic factors and vascular remodeling enzymes in pulmonary PMN in general, Western blot analysis was carried out. When compared to PMN control group, the increased expression of VEGFa, ANG2, MMP9, MMP2 was reversed by FR17 treatment almost back to normal levels (Fig. 4e and Supplementary Fig. 17). In addition, the scrambled control sFD17, which does not possess the drug-bioactivity of TP5, exhibited similar inhibition result as FR17 while TP5 did not, suggesting the protective effect was contributed by the peptide-assemblies formed by the enzyme-activatable self-assembled monomer. Vascular leakage assay on PMN model in vivo verified that FR17 or sFD17 administration attenuated the enhancement of pulmonary vascular permeability (Fig. 4f, g).

Above data confirmed that FR17 can be employed as an in-situ spontaneously-assembled “flame-retarding blanket” to beat out the “flames” on fibroblasts caused by tumor-derived factors in PMN, preventing pro-metastatic angiogenesis and vascular destabilization.

FR17 administration impedes MDSC recruitment to pulmonary PMN and modulates the immuno-microenvironment

Given the fact that MDSC occupies a vital position in PMN construction and metastasis formation, MDSC has attracted wide concerns in recent years. It’s firstly reported by David Lyden’s team that VEGFR+ myeloid progenitor cells are recruited and formed cell clusters to the sites highly expressing FN, which is most likely produced by resident fibroblasts, in the distal tissues prior to the arrival of tumor cells14. The recruitment of MDSC was found to be related to the enhancement of ECM production and remodeling in PMN, including cross-linking collagen by LOX secreted by tumor cells47, FN deposition50, periostin (POSTN)-enrichment51. As MDSC plays a crucial role in developing the immunosuppressive and inflammatory microenvironment, some pioneers have demonstrated that blocking the recruitment of MDSC could be a promising strategy to suppress early metastasis by preventing the development of breeding ground for tumor52,53,54,55.

In our PMN model, the accumulation of MDSC in lung was decreased in FR17 and sFD17 groups, revealed by immunofluorescent staining (Fig. 5a) and flow cytometry analysis on day 10 (Fig. 5b). Moreover, focal enrichment areas of FN and the co-localization of LOX attracted an increasing number of CD11b+Gr+ MDSC in serial sections of the PMN lung comparing to healthy lung (Fig. 5a and Supplementary Fig. 18), consistent with the previous reports47. The immunofluorescent stains indicated that FR17 intervention successfully down-regulated the expression of LOX, FN and POSTN in the lung induced by tumor-derived factors to impede the construction of PMN, which probably resulted in the decrease in MDSC recruitment (Supplementary Fig. 18). To evaluate prevention of peptide administration on developing PMN immunosuppressive environment, protein level of TGF-β in PMN (Fig. 5c), the well-known immuno-modulator produced by MDSC to regulate the establishment of immunosuppressive tumor-supportive niche, as well as the pro-inflammatory cytokine IL-6 was detected (Fig. 5d). FR17 treatment successfully down-regulated TGF-β and IL-6 expression in lung, so as sFD17. In the meantime, TP5 administration exhibited partial abatement on the expression of these cytokines56, suggesting both the enzyme-responsive-assembled peptide nano-blanket and the immunoregulatory agent TP5 facilitated the normalization of immunosuppressive and inflammatory environment of PMN.

Fig. 5: FR17 administration prevents MDSC recruitment, pausing the development of the immunosuppressive microenvironment in PMN.
figure 5

a Serial sections of the lungs harvested from the model mice treated with different peptides. Serial sections show the distribution of lysyl oxidase (LOX) and Fibronectin (FN) and the co-location of CD11b+Gr1+ myeloid-derived suppressor cells (MDSC). Scale bar = 50 µm. b Recruitment of CD11b+Ly6g+ MDSC to the lungs of the model mice treated with different peptides on day 10. Data are presented as mean ± SD. n = 4 biologically independent mice for treatment groups, n = 3 for the control group. One-way ANOVA followed by Tukey’s multiple comparisons test was employed for statistical evaluation. c Expression of TGF-β1 in the lung harvested from PMN model mice administrated with different peptides. d IL-6 expression in the lung tissue fluid collected from PMN model mice administrated with different peptides. Data are presented as mean ± SD. n = 3 biologically independent mice. One-way ANOVA followed by Tukey’s multiple comparisons test was employed for statistical evaluation. e GO enrichment analysis of CD11b+Ly6g+ MDSC sorted from different treatment groups. RNA preparations were extracted from CD11b+Ly6g+ MDSCs sorted from lungs pooled from 10–12 mice per sample. The size of the dots corresponds to the number of genes per pathway, and the color indicates p value. Statistical significance was considered at least at p < 0.05. Source data are provided as a Source Data file.

To find out whether the impact on MDSC-induced immunosuppressive microenvironment in PMN was mainly contributed by TP5 or by the in-situ assembled peptide nano-blanket, cell transcriptomic analysis of MDSCs recruited to the lung was carried out on different treatment groups. Given the fact that CD11b+Ly6g+ MDSC takes the majority (over 90%) of MDSC population in the lung of MCM-induced PMN model (Supplementary Fig. 11a, b), CD11b+Ly6g+ MDSC were sorted on day 10 from pulmonary PMN after different treatments as the representative subset for further transcription analysis (Supplementary Fig. 19). GO enrichment analysis indicated that the regulation effect of FR17 might relate to the activation of immune response pathway, cytokine production involved in immune response, regulation of leukocyte and lymphocyte activation, leukocyte chemotaxis and migration (Fig. 5e). Relative enriched pathways provided possible comments on the underlying mechanisms, including: regulation on cytokine biosynthetic process, adaptive immune response based on somatic recombination immune receptors built from immunoglobulin super-family domains, interferon-γ-mediated signaling pathway and the impact on chemokine receptor binding as well as CXCR chemokine receptor binding (Supplementary Fig. 20a). From above, when compared with the differential gene enrichment pathway of sFD17 and TP5, we found that the regulation on immune cells chemotaxis and migration pathways was contributed by the in-situ peptide-assemblies, for the peptide nano-blanket would only form in FR17 or sFD17 treated lung while not in free TP5 treated group. And the Venn diagram and further enrichment analysis to put TP5 aside suggested these would correspond to the regulation on cell surface receptor signaling pathway by peptide-assemblies (Supplementary Fig. 20b, c). In addition, the regulation on leukocyte differentiation and T cell activation pathway was mainly contributed by TP5 (Supplementary Fig. 20d), which has been commonly accepted as one of the mechanisms for TP5 to exert immune regulation effect57,58,59.

Moreover, the in-situ assembled peptide nano-blanket formed by FR17 inhibited tumor cells migration as well (Supplementary Fig. 21c), with hardly any inhibition on cell viability (Supplementary Fig. 21a).

FR17 administration inhibits melanoma lung metastasis in vivo

First, we evaluated the anti-metastasis efficacy of FR17 on MCM-induced lung metastasis model. C57BL/6 mice were pre-induced by MCM administration for 10 days to set up the tumor-supportive PMN in lung. Tumor cells were then injected through the tail vein on day 7 to simulate the CTC wandering through blood vessels and eventually settled down in pulmonary PMN. As illustrated in Fig. 6a, peptide subcutaneous administration started from day 3 and ended at 2 weeks-post lung tumor inoculation. The metastasis inhibition efficacy was evaluated in terms of the tumor nodule number on day 28. FR17 effectively suppressed tumorigenesis and the development of metastasis in lung, compared to both the control and TP5 groups (Fig. 6b–d). Moreover, the good outcomes of sFD17 treatment which was close to that of FR17 group emphasized the main role of nano-blanket assembled by the enzyme-activated released monomer in intervening PMN construction. Therefore, the in-situ assembled nano-blanket also played a part in inhibiting metastasis development and growth. Preliminary safety evaluation on these model mice at the end of the experiment revealed no severe safety concerns of FR17 administration (Supplementary Figs. 22 and 23). The systemic immune modulation effect of TP5 was reflected on the recovery of thymus shrinking as well (Supplementary Figs. 22d and 23). What’s more, peptide administration prevented tumor cells infiltration into the spleen (Supplementary Fig. 23).

Fig. 6: FR17 administration inhibits lung metastasis in vivo by retarding PMN formation.
figure 6

a Schedule of MCM-induced lung metastasis model with peptide treatment. Peptide administration started from day 3 for metastasis prevention. b Number of the metastatic nodules in the lung of different treatment groups. n = 7. Data are presented as mean ± SD. One-way ANOVA followed by Tukey’s multiple comparisons test was employed for statistical evaluation. c Images of the lung metastasis harvested on day 28 from different treatment groups. d Representative images of Hematoxylin and Eosin staining of lung sections from different treatment groups. The lung metastasis is circled by yellow dotted lines. Scale bar = 200 µm. e Schedule of post-surgery metastasis model. Peptide administration started from day 7 to 25. f Representative in vivo bioluminescent images of mice without treatment or treated with FR17 or anti-PD1. n = 5. The gray patches represent dead mice in the control group. g Survival curves of the mice treated with FR17 or anti-PD1 or without treatment. n = 10. h The cumulative incidence of new pulmonary metastases in the mice treated with FR17 or anti-PD1 or without treatment. n = 10. i Semi-quantification of the in vivo bioluminescent signals in the lungs of the mice treated with FR17 or anti-PD1 or without treatment. n = 5. Data are presented as mean ± SD. Source data are provided as a Source Data file.

The anti-metastasis activity of FR17 was further verified on a post-surgery metastasis model52 (illustrated by Fig. 6e), which is closer to the clinical treatment of resectable melanoma followed-up with adjuvant therapy for patients at high risk of recurrence. Here we compared FR17 therapy with the rising-star of checkpoint inhibitors, PD1 antibody60,61, which was approved by FDA as adjuvant therapy on primary tumor excised with lymph node management or metastatic melanoma62. As demonstrated in Fig. 6f, the recurrence of lung metastasis was put off and the overall survival was greatly improved by FR17 therapy when compared with control (Fig. 6g–i). The median overall survival prolonged for 25.9% (from 29 days to 36.5 days), catching up with the outcomes of anti-PD1 treatment (from 29 days to 31.5 days). And the median lung metastasis-free survival prolonged from less than 19 days of both control group and anti-PD1 treated group to 23 days of FR17 treated group. At the first time-point for lung metastasis monitoring via bioluminescence imaging, lung metastasis has been observed in all of the 10 mice from the control group, 8 mice from anti-PD1 treated group while only 3 mice from FR17 treated group on day 19. The retard of the occurrence of lung metastasis after FR17 administration emphasized the importance of PMN intervention. What’s more, the relapse rate of the primary tumor post-resection was reduced from 10/10 in the control group to 4/10 in FR17 treated group (Supplementary Fig. 24).

Discussion

Though emerging technologies provided early detection of tumor metastasis63 or even warning of high metastatic risk before the actual occurrence of metastasis64, clinical treatment on metastasis prevention lags far behind. The contemporary therapeutic strategies against metastasis in clinic mainly focus on the later time period of macroscopic metastases development, or at least after the arrival and colonization of disseminated tumor cells to the distal organs, which have gained unsatisfied clinical outcomes65,66,67. As scientists further understanding the pathological process of tumor metastasis, the formation of PMN in distal organs arises by pro-metastatic factors produced by primary tumor has drawn more and more attention13.

With plenty of studies emphasized the vital role of MDSC in PMN25,51,52,53,54, little has been explored on the contribution of stromal cells to re-educate the stromal microenvironment for the further development of PMN. Therefore, we would like to turn a spotlight on the pre-metastasis associated fibroblasts, the bugler who sounds the charge for the army of metastasis. Here, we have explored the magical retarding effect of the PMN microenvironment responsive-assembled peptide nano-blanket on fibroblasts activation, impeding PMN development. The enzyme-activatable assembled peptide FR17 can be enzyme-cleaved to release the self-assembled monomer FG8 to construct a lamellar structure, which is named as peptide nano-blanket. Experiments demonstrated that FR17 administration not only beat out the “flame” set up on resident fibroblasts induced by tumor-derived factors, but also interrupted the subsequent PMN formation, including preventing pro-metastatic angiogenesis and vascular destabilization, and then intervening MDSCs’ recruitment as well as their bio-functions. Astonishingly, when treated with sFD17, tumor-induced fibroblast activation was also impeded by peptide-assemblies without the assistance of TP5 so as to arrest the following pro-metastatic pathological process. This finding illustrated that the “flame-retarding” effect on fibroblast activation could be contributed by the drug-free peptide nano-blanket alone, presenting a broad application prospect of drug-free peptide-assemblies to regulate PMN microenvironment and to prevent tumor distant metastasis. This work also elucidated the important role of pre-metastasis associated fibroblast in the complex interactions between major participators in PMN formation and metastasis development, suggesting that reprogramming or intervention on the key juncture could make a big difference on fighting against tumor metastasis.


Article classification: Biological abstract
Share to:
Tel:+86-027-87610298
Tel:+86-027-87610297
Add:Room A11-329, 1st Floor, No.1, SBI Venture Street, Optics Valley, East Lake
New Technology Development Zone, Wuhan, China.
Certificate NO.:U18Q28010569R0S