Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Nano delivery of simvastatin targets liver sinusoidal endothelial cells to remodel tumor microenvironment for hepatocellular carcinoma

Nano delivery of simvastatin targets liver sinusoidal endothelial cells to remodel tumor... so-called capillarization, which amplifies  liver  damage Background by orchestrating the response of the liver microenviron- Hepatocellular carcinoma (HCC) is the second leading ment, permits and promotes HSCs activation, leading to cause of cancer-related deaths worldwide [1]. The etiol - liver fibrosis and carcinogenesis [8]. Reversal of capillar - ogy of HCC including hepatitis B virus infection, hepa- ized LSECs to differentiated LSECs promotes reversion titis C virus infection, alcoholic liver disease, metabolic of activated HSCs (aHSCs) to quiescence and induces syndrome and aflatoxin exposure usually causes injury senescence and apoptosis of aHSCs, which is critical and chronic inflammation in liver [2]. One unique feature for the treatment of both fibrosis and HCC. In addition, of HCC is its close association with chronic inflammation LSECs influence the composition of hepatic immune induced liver fibrosis. More than 80% of HCC develop in populations by distinct expression of adhesion mole- fibrotic or cirrhotic livers, suggesting an important role of cules and chemokines in diseased liver, which indicates liver fibrosis in the premalignant environment of the liver the crucial role of LSECs in immune-regulation [9]. The [3]. Besides radical treatment in early-stage HCC, there is characteristic of LSECs in contributing to HSC deacti- still an unmet clinical need for the majority of unresect- vation and immune response initiation makes them a able HCC. Unfortunately, most of them do not respond potential therapeutic target for HCC treatment. well to either multi-kinase inhibitors or immune-check- Statins, the 3-hydroxy-3-methylglutaryl CoA (HMG- point blocking antibodies, which are effective standard CoA) reductase inhibitors which decrease cholesterol cares for other cancers [4]. This failure is primarily due to synthesis, are usually used to treat dyslipidemia and car- the fibrotic microenvironment and the related immuno - diovascular disease. Recently, there is accumulating evi- suppressive factors in HCC [5]. Thus, stromal microen - dence that statins have beneficial effects in liver fibrosis vironment regression together with immunosuppressive and cirrhosis [10, 11]. Results of cohort studies have con- microenvironment remodeling will pose a potential strat- sistently found that liver cirrhotic patients who received egy to HCC therapy. statin treatment to reduce cholesterol had a lower risk Liver sinusoidal endothelial cells (LSECs) constitute of decompensation and death compared with patients a unique vascular bed in the liver, which receives the who did not receive statins [12–14].  Emerging clinical blood from both hepatic artery and portal veins into observational studies have also shown that statin use hepatic parenchyma [6]. LSECs are the first liver cell was associated with a reduced risk of HCC development type affected after liver  injury, and play a critical role in and related mortality [15–17]. Simvastatin, as the most the cellular crosstalk that regulates progressive chronic effective statin protecting the hepatic endothelium, was liver disease, which leads to fibrosis and carcinogenesis revealed to activate the transcription factor Kruppel-like [7]. Under physiological condition, differentiated LSECs factor (KLF2)-nitric oxide (NO) pathway to reverse LSEC function as gatekeeper to prevent hepatic stellate cells capillarization and mediate hepatic endothelial protec- (HSCs) from activation. While the LSEC differentiation tion [18, 19]. Mannose receptor is the typical scavenger usually lost prior to fibrosis during pathological process, Yu  et al. Journal of Nanobiotechnology (2022) 20:9 Page 3 of 16 receptors with high expression on LSECs, and mannan from NANOSOFT POLYMERS. Other synthetic mate- exclusively acts as a ligand to target the mannose recep- rials as previously reported were obtained from Sigma- tor, being ability to facilitate simvastatin into LSECs [25]. Aldrich MO, USA [20]. u Th s, the specific manipulation of this important switch in selected LSECs might be more effective than oral Antibodies administration of statin nowadays as a strategy to treat In vivo anti-mouse IgG, CD1d, CD8 and CD4 mAb were HCC. purchased from BioXcell (West Lebanon, NH). Primary Here, we demonstrate that LSEC capillarization is asso- and secondary antibodies used for immunoblotting, ciated with fibrotic HCC progression and poor survival. immunostaining and flow cytometry are listed in Addi - Simvastatin converts HSCs from activation to quiescence tional file 1: Table S1. by stimulating KLF2-NO signaling in LSECs and up-reg- ulates the expression of Chemokine (C-X-C motif ) ligand Murine HCC models establishment 16 (CXCL16) on LSECs. In fibrotic HCC mouse model, Four-week-old male C57BL/6 mice were purchased from targeted transfer of simvastatin into LSECs by nano- Jackson Laboratory. All animal regulations and proce- technology not only mitigates LSEC capillarization to dures were accepted by the Institutional Animal Care improve the stromal microenvironment, but also recruits and Use Committee of the University of North Caro- natural killer T (NKT) cells to suppress tumor progres- lina at Chapel Hill. For fibrotic HCC model, 4-week-old sion. Combined with anti-PD-L1 antibody, simvastatin male immune-competent C57BL/6 mice received intra- nanoparticles (NPs) evoke an improved therapeutic effect peritoneal injection of CCl (1 μL/g, diluted in 100 μL at hemi-splenic inoculated advanced-stage HCC. These olive oil) twice a week for 4  weeks. Age-matched mice findings unravel an immune-based therapeutic mecha - administered with olive oil served as control. Intrahe- nism of simvastatin and provide a novel therapeutic patic injection of 5 × 10 Hepa1-6-luc cells in 50 μL PBS strategy based on tumor microenvironment remodeling was performed at the end of the fourth week. Mice were against HCC. sacrificed at 3  weeks post-tumor cell inoculation or at humane endpoint. The tumors were isolated from the Materials and methods liver and measured with caliper. The volume was calcu - Human specimens lated with the formula of length × width × width/2, and Medical history from 26 non-fibrotic HCC patients and the average value of each group was equal to the sum of 38 fibrotic HCC patients were collected for survival anal - total volumes divided by the number of mice. The tumor ysis. Immunohistochemistry were performed in a sub- growth indicated by luciferase intensity was monitored set of patients with fibrotic HCC or non-fibrotic HCC. using IVIS (Perkin Elmer, CA) imaging every 2  days. All 64 patients with HCC were obtained from Shuguang Tumor and matched non-tumor liver tissues were col- Hospital affiliated to Shanghai University of Traditional lected for section analysis. For advanced-stage HCC Chinese Medicine (SUTCM) and approved by SHUTCM model, hemi-splenic injection of Hepa1-6 cells was Clinical Research Ethics Committee. performed after 4-week CCl exposure. To be specific, an incision located below the left rib cage was made to Cell culture exteriorize the spleen. The spleen was tied and cut into Mouse hepatoma cell line Hepa1-6, human hepatocel- two parts, each containing an intact vascular pedicle for lular carcinoma cell line Huh7, LSEC cell line SK-Hep1 each half of the spleen. The distal section of the spleen and HSC cell line LX2 were maintained in high glucose was inoculated with 1 × 10 Hepa1-6 cells in 150 μL PBS. Dulbecco’s modified Eagle Medium (DMEM, Gibco) sup - The half spleen containing inoculated cells were resected plemented with 10% bovine calf serum (BCS) (Hyclone), 5 min after inoculation allowing the cancer cells to enter 1% penicillin and streptomycin (Hyclone). Two μg/mL the portal vein. The other half of the spleen was returned puromycin was used additionally for Hepa1-6-luciferase to the cavity to keep the immune system competent. (luc) cells. All cells were grown at 37  °C in a humidified incubator with 5% CO . Synthesis and characterization of the drug‑loaded mannan modified NPs Reagents The PLGA copolymers were obtained by a previously Simvastatin with over 99% purity (Sigma-Aldrich MO, reported method [20, 21]. The PLGA-PEG and PLGA- USA) were used for the study. Mannan and L-NAME PEG-MAN were first synthesized. Mannan attachment were purchased from Sigma-Aldrich. DiD was purchased to the particle surface was achieved through covalent from ThermoFisher Scientific. Poly(lactic-co-glycolic acid attachment by conjugation chemistry of the COOH-ter- (PLGA)-Polyethylene glycol (PEG)-NH2 was purchased minus of PLGA and mannan [22]. Yu et al. Journal of Nanobiotechnology (2022) 20:9 Page 4 of 16 Gene expression profiling Two mg simvastatin was dissolved in 200  µL DMSO, SK-Hep1 cells in triplicate were treated with simvastatin and then mixed with 2  mg PLGA-PEG-MAN and 6  mg (20  μM) and control for 24  h. Total RNA was extracted PLGA-PEG in the drug solution. The drug-polymer mix - using Trizol reagent (Invitrogen, Carlsbad, CA) accord- ture was added dropwise to 10 mL deionized water while ing to the manufacturer’s protocol. RNA quality was stirring, and stirred for 4  h. Nanoparticles were purified assessed on an Agilent 2100 Bioanalyzer (Agilent Tech- by ultrafiltration using 10 k cutoff columns at 4000×g. nologies, Palo Alto, CA) and checked using RNase free The particle size and zeta potential of nanoparticle agarose gel electrophoresis. mRNAs were isolated and were measured by Malvern Nano-ZS (Malvern Instru- fragmented to about 200 base pairs length and reverse ments, UK). The morphology was analyzed by transmis - transcribed into cDNA using the QuantiTect Reverse sion electron microscopy (TEM). Five μL particles were Transcription kit (Qiagen). The cDNA fragments were added on 400-mesh carbon-filmed copper grids (Agar purified with QiaQuick PCR extraction kit (Qiagen, Scientific) for 2  min and stained with 2% (w/w) uranyl Venlo, Netherland), end repaired, poly(A) added, and acetate before imaging using TEM (JEOL JEM1230). ligated to Illumina sequencing adapters. The cDNA The drug encapsulation efficiency (EE) was calculated library products were size selected by agarose gel elec- by (Entrapped Drug/Drug Added)*100%. Loading capac- trophoresis, PCR amplified and sequenced using Illumina ity (LC) was calculated by (Entrapped Drug/Nanoparticle HiSeqTM4000 platform (Illumina, San Diego, CA). Tran- Weight)*100%. script-level expression analysis of sequencing data was performed using HISAT and StringTie software (http:// In vivo treatment studies ccb. jhu. edu/ softw are,shtml) [24]. Differential transcripts For intra-hepatic inoculation HCC model, the mice with of chemokines with stringent cutoff coefficient of less liver tumors were treated 3  days after the tumor cell than 0.05 were obtained to align to GO database (http:// inoculation. Five groups of mice were separately given www. geneo ntolo gy. org) for protein functional annotation PBS, blank PLGA NPs, simvastatin PLGA NPs (20  mg/ corresponding to immune population. kg), simvastatin PLGA-MAN NPs (20 mg/kg) by tail vein injection and simvastatin free drug oral gavage (40  mg/ MTT assay kg) every other day for totally 5 times. The anti-tumor Cell viabilities were assessed by MTT assay. 10^4 cells efficacy was regularly assessed using IVIS (Perkin Elmer, were seeded in 96-well plates per well overnight and sub- CA) imaging every 2  days and survival was recorded. jected to different treatments. Five mg/mL MTT (Alfa Some mice were sacrificed 3  days after treatment, the Aesar) reagent was added for 4  h at 37  °C, and then the tumors and matched non-tumor liver tissues were col- supernatant was discarded. The formazan was resus - lected and prepared for section study and immune effect pended in 100  μL of DMSO and absorbance was exam- evaluation. For immune cell depletion study, after Hepa1- ined by a spectrometer (Hidex Chameleon). 6-Luc intrahepatic inoculation, PBS, IgG, anti-CD1d, anti-CD4 or anti-CD8 antibody (100  µg) was intraperi- Histology toneally injected to mice every other day for totally 5 Liver tissues with tumor or major organs including injections 1  day before NPs treatment. For hemi-splenic hearts, livers, spleens, lungs, and kidneys were col- inoculation HCC model, intervention was applied 5 days lected and were fixed in 4% paraformaldehyde (PFA). after tumor cell inoculation. Four groups of mice were Fixed samples were paraffin-embedded, sectioned, and given separately by PBS, simvastatin PLGA-MAN NPs stained with hematoxylin and eosin (H&E) or Masson’s (20  mg/kg), anti-PD-L1 antibody (100  µg) or combina- trichrome at UNC histology facility. PFA-fixed sam - tion. NPs tail vein injection and anti-PD-L1 antibody ples were embedded with optimum cutting temperature intraperitoneal injection were performed every other day compound and sectioned at 8  μm thickness. For immu- for a total 5 times. nohistochemistry (IHC), sections were incubated with primary antibodies at 4  °C overnight, washed, and incu- Mouse liver sinusoidal endothelial cell isolation bated with horseradish peroxidase-conjugated secondary Mouse LSECs were isolated as previously described [23]. antibodies for 2  h at room temperature. Digital images In brief, livers were perfused through the portal vein were taken using brightfield light microscope (Olympus and digested with a collagenase solution. After mincing BX61). Immunofluorescence (IF) was performed using the liver, cells were filtered and centrifuged at 50×g to fluorescent antibodies and counterstained with Prolong remove hepatocytes. Non-parenchymal cells were then Diamond Antifade Mountant with DAPI (ThermoFisher separated by differential centrifugation using a Percoll Scientific). Antibodies are listed in Additional file  1: gradient. Kupffer cells were eliminated by plastic pre-cul - Table  S1. Apoptotic cells were stained with a terminal ture for 30 min. LSEC were collected for RNA extraction. Yu  et al. Journal of Nanobiotechnology (2022) 20:9 Page 5 of 16 deoxynucleotidyltransferase-mediated dUTP nick end Griess Reagent and color was developed in 10  min. The labeling (TUNEL) kit (Promega, Madison, WI). Images absorbance was measured at 540  nm and NO concen- were taken using laser-scanning confocal fluorescence tration was calculated referring to a standard curve. The microscope (Zeiss LSM 710). Liver samples fixed with experiments were done in triplicate. 4% PFA by perfusion through portal vein were sectioned using vibratome at UNC microscopy services laboratory Biodistribution and prepared for scanning electron microscopy (Zeiss 0.05% (wt) of lipophilic carbocyanine DiD (Ther - Supra 25 FESEM). Five random microscopic fields were moFisher) was used to formulate the DiD-labeled PLGA- selected and quantified by ImageJ software. Porosity was PEG or PLGA-PEG-MAN NP. In mice bearing fibrotic measured as percentage of LSEC surface occupied by tumor, 24  h following tail vein injection of DiD-labeled fenestrae in SEM in liver tissue. Fenestration frequency NP, major organs and liver with tumors were collected was calculated with total number of fenestrations divided and analyzed using IVIS (Perkin Elmer, CA) with the by total area of LSEC surface. excitation wavelength at 640 nm and the emission wave- length at 670 nm. Flow cytometric analysis Single-cell suspensions from tumor tissue were harvested Blood chemistry analysis in MACs buffer (1 × PBS + 2  mM EDTA + 0.5% BSA, fil - Seven days after 5 doses of nanoparticle injections, ter sterile), then subjected to conjugated staining with whole blood was obtained from the normal mice. Blood fluorescence. At least 10,000 live cells were subjected routine including RBC, WBC, PLT, HGB, and HCT was to flow cytometric analysis on a flow cytometer (Bec - tested using the whole blood. Aspartate aminotransferase ton Dickinson LSR II). Experimental data were analyzed (AST), alanine aminotransferase (ALT), urea nitrogen using FlowJo software. The antibodies used are listed in (BUN), and creatinine (CRE) in serum were analyzed as Additional file 1: Table S1. indicators of hepatic and renal function using the serum. Immunoblotting Statistical analysis Cells were lysed in RIPA lysis buffer with protease inhibi - GraphPad Prism 8 (GraphPad Software, La Jolla, Califor- tors. Total lysates were quantified by a BCA Protein Assay nia, USA) was used for statistical analysis. The independ - Kit (Biorad, CA). Thirty μg protein samples were used for ent Student t-test was used to compare data between immunoblotting analysis. After incubating with appro- the two groups. A one-way analysis of variance test with priate primary and secondary antibodies, the immuno- Bonferroni correction was used to compare data in mul- reactions were visualized with Western HRP substrate tiple groups of mice. Pearson correlation coefficient was (Thermo, Rockford, IL). The antibodies used are listed in calculated to show the correlation of two parameters. Additional file 1: Table S1. Two-way ANOVA with a Bonferroni post-hoc test was used to estimate the difference in multiple groups of mice Quantitative real‑time polymerase chain reaction (RT‑PCR) when considering two factors of time and treatment. assay Kaplan–Meier survival analysis was used to determine Total RNA was extracted from cells or the whole tumor the overall survival rates and tumor incidence-free rates, using an RNeasy microarray mini kit (Qiagen, Hilden, the differences were compared by the log-rank Mantel- Germany) and was reverse-transcribed to cDNA with Cox test. Data averages from each group are presented as an iScript cDNA synthesis kit (Bio-Rad, Hercules, CA). mean ± SD; *p < 0.05, **p < 0. 01, ***p < 0. 001. Quantitative PCR was performed in a 7500 RT-PCR system. The PCR primers are listed in Additional file  1: Tables S2 and S3. Results Liver fibrosis‑associated LSEC capillarization contributes Nitric oxide assay to aggressive HCC development Nitric oxide (NO) amount was assessed using Nitric To observe the role of fibrotic microenvironment dur - Oxide Colorimetric Assay Kit (BioVision, Milpitas, CA) ing HCC development, Hepa1-6 orthotopic HCC model in accordance to the manufacturer’s manual. Briefly, SK- was established using C57BL/6 mice pretreated with Hep1 cells were treated with different concentrations of CCl injection which induces chronic inflammation and simvastatin (0, 5, 20  μM) for 24  h and the supernatants fibrosis (Fig.  1A). Interestingly, tumor growth acceler- were collected. Equal aliquot of samples in every group ated more rapidly in fibrotic liver than the normal liver, were mixed with Nitrate Reductase Buffer and incubated as shown by the tumor growth curve (Fig.  1B) and bio- at room temperature for 1 h. The mixture was then added luminescence imaging of tumors, as well as liver tumor Yu et al. Journal of Nanobiotechnology (2022) 20:9 Page 6 of 16 Fig. 1 Liver fibrosis-associated LSEC capillarization contributes to aggressive HCC development. A Schematic diagram of liver fibrosis-associated HCC mouse model. B Tumor growth curve and C Bioluminescence imaging of tumor and representative liver tumor morphology at the endpoint. D Statistical analysis of tumor volume and weight. E Representative pictures of Masson trichrome staining, α-SMA immunofluorescence in mice non-fibrotic and fibrotic liver tissue. Scale bar stands for 50 μm. Yellow dotted line indicates the border between non-tumor and tumor tissue. The quantification of collagen density, α-SMA expression were shown. F LYVE1 and CD31 co-immunofluorescence in mice non-fibrotic and fibrotic liver tissue, scale bar stands for 50 μm. Scanning electron micrographs illustrating the morphological response of LSEC. Scale bar stands for 1 μm. The quantification of LYVE1 and CD31 expression were shown, as well as the porosity and fenestration on LSEC. G Percentage of α-SMA, CD31 fluorescence and tumor weight are pairwise positively correlated. H Kaplan–Meier overall survival and disease-free survival curves of HCC patients with non-fibrotic tumor (n = 26) or fibrotic tumor (n = 38). I Expressions of α-SMA and CD31 in non-fibrotic HCC and fibrotic HCC tissues as determined by immunohistochemistry. Scale bar stands for 50 μm.*p < 0.05, **p < 0.01, ***p < 0.001 morphology at the endpoint (Fig. 1C). Tumor weight and markers were highly expressed in peri-tumor and tumor volume are confirmed as well (Fig. 1D). tissues in C Cl -treated group as expected (Fig. 1E). Since To further investigate the fibrotic microenvironment LSECs play critical role during fibrosis, lymphatic vessel responsible for HCC development, we first evaluated the endothelial hyaluronan receptor  1 (LYVE1) and CD31 expression of α-smooth muscle actin (αSMA) and colla- which are the markers of differentiated and capillariza - gen which are key markers of aHSCs and fibrosis. Both tion LSECs, respectively, were assessed. Interestingly, Yu  et al. Journal of Nanobiotechnology (2022) 20:9 Page 7 of 16 high LYVE1 expression was observed in both peri-tumor (RNA-seq) in simvastatin treated SK-Hep1 cells, or not, and tumor tissues of non-fibrotic liver, while CD31 was to examine whether simvastatin affect the immune-regu - over-expressed in fibrotic liver (Fig.  1F). The porosity and lated genes in LSEC. Gene expression profiles analysis of fenestration on LSECs were much reduced in fibrotic RNA-seq data suggested that a set of chemokines, which liver compared to the non-fibrotic liver (Fig.  1F). We are classified in terms of the function on the regulation of further found that αSMA, CD31 expression level and the variety of leukocytes, were changed in SK-Hep1 cells tumor mass were significantly pairwise positively corre - treated with simvastatin compared with the untreated lated (Fig.  1G), while LYVE1 expression was negatively control cells. The investigation coupled with functional correlated with tumor mass. Thus, our results showed categorization of changed chemokine genes revealed that LSEC capillarization may contribute to aggressive HCC CXCL16 was identified as the significant regulator upon development. simvastatin treatment (Fig. 2F). Further quantitative PCR We next investigated the clinical relevance of LSEC and western blot detection confirmed that transcript capillarization in HCC patients with or without fibrosis/ level and protein expression of CXCL16 greatly increased cirrhosis. The Kaplan–Meier analysis revealed that HCC in SK-Hep1 treated with simvastatin in a dose-depend- patients with fibrosis were significantly correlated with ent manner (Fig.  2G, H and Additional file  1: Fig. S2C). shorter overall and disease-free survival rates (Fig.  1H). NKT cells were characterized with high expression of Consistent with the mouse model, the fibrotic HCC tis - CXCR6 (Fig.  2I), and CXCL16 as the ligand for CXCR6 sue exhibited high expression level of αSMA and CD31 exerts recruitment function on NKT cells. Further study (Fig.  1I). Thus, poor survival in fibrotic HCC patients is indicated that simvastatin alone could not influence the related to LSEC capillarization. activity of NKT cells, while simvastatin could upregulate the expression of CD69 and IFN-γ in NKT cells in the co- Simvastatin deactivates HSC via LSEC and up‑regulates culture with SK-Hep1, suggesting simvastatin evoked the CXCL16 expression on LSEC antitumor NKT immune response through the mediation Since simvastatin exerts beneficial effects in fibrosis of LSEC (Fig.  2J). Thus, simvastatin can not only deac - and HCC, we first examined the effect of simvasta - tivate HSCs via KLF2-NO signaling in LSECs, but also tin on three human cell lines which are SK-Hep1 LSEC induce CXCL16 overexpression in LSECs which recruits cell line, LX2 HSC cell line and Huh7 HCC cell line. We and activates NKT cells to the liver. found simvastatin almost had no cytotoxicity to all three cell lines (Additional file  1: Fig. S1). However, simvasta- Design of simvastatin LSEC‑targeted delivery NP platform tin increased both mRNA and protein expression levels Simvastatin is poor soluble in aqueous solution. The of KLF2 and endothelial nitric oxide synthase (eNOS) in PLGA polymer approved by FDA for drug delivery was SK-Hep1 cell line in a dose dependent manner (Fig.  2A, selected for encapsulating hydrophobic simvastatin due B and Additional file  1: Fig. S2A). The NO excretion was to the characteristic of both biodegradability and bio- up-regulated according to the activation of KLF2-eNOS compatibility. A unique characteristic of LSECs is their signaling induced by simvastatin (Fig. 2C). When lipopol- expression of high levels of scavenger receptors, which ysaccharide (LPS)-activated LX2 treated with simvasta- ensure the high endocytic capacity [7]. One of the most tin directly, no change of the activation marker αSMA extensively studied scavenger receptors on LSECs is and collagen was observed. Since simvastatin increased the mannose receptor [25]. Thus, mannan was selected NO excretion in SK-Hep1, we collected the conditioned as a ligand to target the mannose receptor on LSEC. medium of simvastatin treated SK-Hep1 cells and added The structure of synthetic PLGA-PEG-mannan was to cultured LX2 cells. The mRNA and protein expression confirmed by NMR. The H NMR spectrum of PLGA- levels of αSMA and collagen I were both down-regulated PEG-mannan showed that δ 4.639(H1), δ 3.51(H2), δ dose-dependently (Fig.  2D, E and Additional file  1: Fig. 5.20(H3)and δ 4.90(H4)are associated with CH of man- S2B). Further, an NOS inhibitor L-NAME was used to nan, CH2 of PEG, CH of lactic acid and CH2 of gly- verify the NO induced-HSC deactivation. Indeed, LX2 colic acid, respectively (Additional file  1: Fig. S3), which activation recovered after L-NAME intervention (Fig. 2D, indicated the success of synthesis. We utilized a solvent E and Additional file  1: Fig. S2B). Thus, simvastatin displacement process to load simvastatin (Fig.  3A). seemed to work on SK-Hep1 cells which then secreted PLGA NPs containing simvastatin presented a micelle- NO, or an NO-related factor, to deactivate LX2 cells. like structure with a diameter of approximately 100 nm In addition, since LSECs play critical roles in the com- revealed by transmission electron microscopy (Fig. 3B). position of hepatic immune populations by mediating the The particle size around 120  nm and zeta potential recruitment of leukocyte subsets through the expression around 3  mV were confirmed by particle sizing sys - of distinct chemokines, we performed RNA-sequencing tem (Fig.  3C, D). The drug encapsulation efficacy of Yu et al. Journal of Nanobiotechnology (2022) 20:9 Page 8 of 16 Fig. 2 Simvastatin deactivates HSC via LSEC and recruits NKT cells. A Quantitative RT-PCR analysis and B immunoblotting of KLF2 and eNOS in SK-Hep1 LSEC cells treated with the indicated concentrations of simvastatin for 24 h (n = 3). GAPDH acts as the loading control. C Extracellular concentration of NO upon different doses of simvastatin treatment (n = 3). D Quantitative RT-PCR analysis and, E immunoblotting of ACTA2 and COL1 in LPS-activated LX2 treated with the supernatant of SK-Hep1 upon different doses of simvastatin treatment with or without L-NAME (n = 3). GAPDH acts as the loading control. F Gene expression profiles of chemokines analyzed with RNA-seq data which derived from SK-Hep1 cells treated with simvastatin or control. Simvastatin-responsive chemokine genes were grouped in association with clusters of various immune cell-regulating functions. The p value of each comparison was indicated with colors. G Quantitative RT-PCR analysis and H immunoblotting of CXCL16 in SK-Hep1 cells treated with the indicated concentrations of simvastatin for 24 h (n = 3). GAPDH acts as the loading control. I CXCR6 expression was detected in NKT cells. J The expression of CD69 and IFN-γ was detected in NKT cells treated with control, simvastatin, the co-culture with SK-Hep1 cells, and the combination, respectively (n = 3). *p < 0.05, **p < 0.01, ***p < 0.001, n.s. = not significant Yu  et al. Journal of Nanobiotechnology (2022) 20:9 Page 9 of 16 simvastatin was approximately 85%, loading capacity Mice were treated by tail vein injection of PBS, blank was 6.5%. NPs, simvastatin un-targeted NPs (Sim NP) (20  mg/kg), NPs with mannan target tend to accumulate much simvastatin LSEC-targeted NPs (Sim NP-MAN) (20 mg/ more in fibrotic liver with tumors, but less in other kg) and simvastatin oral administration (40  mg/kg), main organs than the non-targeted NPs at 24  h post respectively (Fig.  4A). Notably, Sim NP-MAN impres- tail vein injection, which was tracked using bioimaging sively mitigated rapid tumor growth compared to PBS or (Fig.  3E, F). The uptake of NPs in fibrotic liver tissues blank NP-treated group, as shown by the lower increased was further examined by immunofluorescence stain - rate of tumorous luciferase intensity and tumor growth ing. As expected, mannan-coated particles encapsulat- curves. Either simvastatin oral administration or Sim ing DiD were captured by LSECs almost 4 times higher NPs injection exhibited partial inhibitory effect (Fig.  4B, compared to non-coated ones both in tumor and non- C). Besides, a significant survival prolongation from 23 tumor tissues (Fig. 3G). days in the control group to 43 days in the Sim NP-MAN treated group was also noticed (Fig.  4D). The livers with LSEC‑targeted delivery of simvastatin inhibits tumor tumor were excised at day 21 for inspection, liver tumor growth in mouse fibrotic HCC model load of the Sim NP-MAN treated group were signifi - To investigate the effect of targeted delivery of simvas - cantly reduced than those of the other groups (Fig.  4E). tatin in  vivo, the Hepa1-6 orthotopic HCC model was u Th s, LSEC-targeted delivery of simvastatin showed bet - established using C57BL/6 mice pretreated with CCl . ter anti-HCC effect than oral administration and un-tar - Three days after tumor cells inoculation, treatment was geted delivery. administrated every other day for continuous 5 times. Fig. 3 Simvastatin LSEC-targeted delivery NP platform was designed. A Efficient co-encapsulation of simvastatin into PLGA NP using a solvent displacement technique. B TEM images of PLGA NPs loaded with simvastatin cores. C Particle size of the PLGA NP. D Zeta Potential of the PLGA NP. E Images and F quantification of the DiD-loaded PLGA (DiD/PLGA-PEG) and mannan conjugated PLGA(DiD/PLGA-PEG-MAN) in liver with tumor and other major organs at 24 h after injection in mice bearing Hepa1-6 liver tumor. G Fluorescence imaging detecting DiD in PLGA-PEG or mannan-coated PLGA-PEG and LSEC marked with CD31, scale bar stands for 50 μm. *p < 0.05, **p < 0.01, ***p < 0.001, n.s. = not significant Yu et al. Journal of Nanobiotechnology (2022) 20:9 Page 10 of 16 Fig. 4 Simvastatin NPs suppress HCC development. A Tumor inoculation and treatment scheme. B Bioluminescence imaging of Hepa1-6 tumor bearing mice during tumor progression in various treatment groups. C Tumor growth curves and, D survival curves of Hepa1-6 tumor bearing mice in various treatment groups (n = 5). E Representative image of liver with Hepa1-6 tumor and quantification of liver weight at day 21 after inoculation (n = 3). Scale bar stands for 1 cm. *p < 0.05, **p < 0.01, ***p < 0.001, n.s. = not significant LSEC‑targeted delivery of simvastatin remodels stromal As CXCL16 is the ligand for CXCR6 which is expressed environment and recruits NKT cells in fibrotic HCC mostly on NKT cells, the cell frequencies of CXCR6 We next examined the effect of simvastatin delivered NKT cell in tumor tissues of various groups were evalu- with mannan-targeted NPs to remodel stromal environ- ated by flow cytometry. CXCR6 NKT cell frequency ment in the liver. HE staining revealed that the disorder increased from 0.82% in control group to 11.41% in structure of both tumor and non-tumor fibrosis tissue Sim NP-MAN treated group (Fig.  5F). Both CD69 and reversed in Sim NP-MAN treated group (Fig.  5A). The IFN-γ which are the activation markers of NKT cells collagen deposition reduced significantly as shown in were up-regulated (Fig.  5G). We further examined the Masson-trichrome staining. Accordingly, the expression critical cytokines in tumor tissues of various groups. of αSMA was approximately sevenfold down-regulated The increased expression of the immunostimulatory compared to control group (Fig. 5B). The over-expressed cytokines including CXCL16, IFN-γ, TNFα and the LSEC capillarization marker CD31 were remarkably decrease of immunosuppressive cytokines including down-regulated, while differentiated marker LYVE1 CXCL9, IL-6, TGFβ were noticed with Sim NP-MAN up-regulated in Sim NP-MAN treated group. In addi- treatment (Fig.  5H). Also, the TUNEL assay revealed tion, both porosity and fenestration frequency of LSECs increased number of apoptotic cells in tumor tissues after increased significantly (Fig.  5C). Further, we found the Sim NP-MAN treatment (Fig. 5I). To address the immune + + mRNA expression levels of Klf2 and Nos3 in LSECs mechanism, we further depleted NKT, CD4 or CD8 T extracted from liver tissues were up-regulated signifi - cells in mice HCC model (Fig. 5J). The tumor restraining cantly, which is consistent with the observation in  vitro, effect was abolished by treatment with anti-CD1d, anti- indicating the Klf2-NO signaling were activated by tar- CD8 or anti-CD4 antibody, whereas it was not affected geted delivery of simvastatin (Fig. 5D). Since simvastatin- when an isotype-matched IgG control was used. CD1d treated LSECs express CXCL16, we next examined the and CD4 antibody showed the most obvious abolishment co-localization of CXCL16 and LYVE1 in liver tissue. As which indicates that NKT cells played a more critical role expected, CXCL16 expression increased together with during the anti-tumor process than the CD8 T-cells LYVE1 after Sim NP-MAN treatment (Fig. 5E). (Fig. 5J). Taken together, the result suggested that stromal Yu  et al. Journal of Nanobiotechnology (2022) 20:9 Page 11 of 16 Fig. 5 Simvastatin NPs remodel tumor microenvironment in fibrotic HCC. A HE staining of tissue features from HCC mice with various treatments. Yellow dotted line indicates the border between normal liver and tumor tissue. The left side is non-tumor liver tissue and right side is tumor tissue. Representative images are shown. Scale bar stands for 50 μm. B Masson trichrome (MT ) staining shown together with αSMA fluorescence staining in liver tissues. Data are presented as mean ± SD. Scale bar in MT image stands for 20 μm, in IF image stands for 50 μm. C LYVE1 and CD31 expression by fluorescence staining shown together with LSEC feature using SEM in liver tissues. The quantification of porosity and fenestration of LSEC is shown (n = 5). Scale bar in IF image stands for 50 μm, in SEM image stands for 1 μm. D Quantitative RT-PCR analysis of Klf2 and Nos3 in LSECs extracted from liver tissues intervened with various treatments (n = 5). E The fluorescence staining of LYVE1 and CD31. Scale bar stands for 50 μm. + + F NKT cell frequency in liver tissues detected by flow cytometry (n = 5). G CD69 and IFN-γ levels of hepatic NKT cells detected by flow cytometry (n = 5). H mRNA expression of chemokines and cytokines detected by real-time PCR in tumors in various treatment groups (n = 5). I The average number of apoptosis cells per high-power field (HPF) detected by TUNEL immune fluorescence staining. Scale bar stands for 50 μm. J Scheme of immune cell depletion study and tumor growth curves of Hepa1-6 tumor treated with simvastatin NP after anti-CD1d, CD4or CD8 intraperitoneal injection (n = 5). *p < 0.05, **p < 0.01, ***p < 0.001, n.s. = not significant environment re-modulation and NKT cell recruitment Simvastatin NPs show no obvious toxicity might be the main reason for the antitumor effect of Sim To further evaluate the safety of simvastatin NPs for the NP-MAN in fibrotic HCC. development of both effective and translational therapy, biosafety-related toxicological serum and pathology Yu et al. Journal of Nanobiotechnology (2022) 20:9 Page 12 of 16 analyses were performed. The pathology revealed Discussion by H&E staining showed no significant morphologi - Our study indicated that LSEC capillarization accounts cal damage among major organs including heart, lung, for tumor progression and poor survival in fibrotic HCC. liver, spleen and kidney (Additional file  1: Fig. S4A). The Simvastatin functions as hepatic endothelium protec- body weight remained stable without severe weight loss tor stimulates KLF2-NO signaling in LSECs to regress between different groups during experiment (Additional HSC activation. Simvastatin also up-regulates CXCL16 file  1: Fig. S4B). The blood routine and hepatorenal func - in LSECs which can recruit NKT cells. In order to exert tion of mice in all groups remained in the normal range the dual effect of simvastatin on LSEC, we targeted- (Additional file  1: Fig. S4C, D). Thus, simvastatin NPs delivered simvastatin to LSECs by a PLGA-PEG-MAN showed no obvious toxicity in the mouse model. NP formulation. LSEC-targeted delivery of simvastatin not only alleviates LSEC capillarization to regress the stromal microenvironment, but also recruits NKT cells Simvastatin NPs together with PD‑L1 antibody achieve to remodel immunosuppressive microenvironment and the synergistic effect in late‑stage HCC inhibit tumor progression in murine HCC model (Fig. 7). Despite surveillance, HCC often presents in clinic at In advanced-stage HCC, combination therapy of simvas- an  advanced  stage  that only systemic therapy is feasi- tatin NP and anti-PD-L1 antibody achieves an improved ble. Currently, most studies on the mechanisms of HCC therapeutic effect. immune evasion have focused on the programmed death LSECs are highly specialized endothelial cells in receptor 1 (PD-1)/programmed death ligand 1 (PD-L1) liver, which lies between blood cells on the sinusoidal pathway [26]. PD-1 is an immunoinhibitory receptor side and hepatocytes and HSCs on the abluminal side expressed in activated T and B cells, as well as NKT cells. [28]. With the structure of fenestrae, LSECs represent PD-L1, the major ligand of PD-1, is expressed in various a permeable barrier allowing exchanges between two immune cells such as antigen-presenting cells, as well as sides. However, LSECs do not simply form a barrier but in endothelial cells [27]. The PD-L1 expression was higher exert instrumental effect in maintaining microenviron - in Hepa1-6 tumor tissue compared to normal tissue, and ment homeostasis and mediating immune response in simvastatin NPs showed no effect on PD-L1 expression liver. Differentiated LSECs are the gatekeeper to main - (Fig.  6A). To explore an effective strategy for late-stage tain HSC in their quiescent state, which are critical to HCC treatment, the hemi-spleen inoculation of Hepa1-6 suppress fibrogenesis and carcinogenesis [29]. LSECs based on C Cl -induced fibrosis model was used to test also influence the composition of hepatic immune the therapeutic effect of simvastatin NPs combined with populations by expression of adhesion molecules and PD-L1 antibody. In this aggressive HCC model, due chemokines [9]. The unique positioning, phenotype and to the dispersing of tumor cell to liver through splenic function make LSECs an attractive candidate for organ- artery and portal vein, 20% of lung metastasis can be specific therapy. In our study, simvastatin exerted dual observed and mice usually die around day 20 if without effect via LSECs to deactivate aHSCs and recruit NKT treatment. Five days after inoculation, mice were treated cells, leading to stromal regression and tumor sup- with PBS, Sim NP-MAN (20  mg/kg, tail vein injection), pression. Since LSECs are able to take up molecules by PD-L1 antibody (100  μg, intraperitoneal injection), or scavenger receptors, nanoparticles with target ligand both for totally 5 times every other day (Fig.  6B). Both can be an excellent way to apply therapy specific to Sim NP-MAN and PD-L1 antibody showed a slight effect LSECs. Several studies verified the therapeutic effect in tumor suppression in HCC mice, whereas the combi- of nanoparticles targeting LSECs in various liver dis- nation of both treatments inhibited tumor progression eases [30–32]. Mannan is one of the most extensively significantly (Fig.  6C, D). Severe tumor development used ligand to target mannose receptor on LSECs. In and lung metastasis could be observed at day 19 after this study, PLGA with mannan ligand showed high liver Hepa1-6 inoculation, while much smaller tumor foci and accumulation and LSEC uptake rate, leading to satisfac- no lung metastasis was found in group with combination tory therapeutic effect in murine HCC model. treatment (Fig.  6E). The overall survival in the combina - NKT cells which share phenotypic and functional fea- tion group extended almost 2 times of the control group tures with NK cells represent a subpopulation of T lym- (Fig.  6F). As expected, upon the same treatment of Sim phocytes. The liver is the organ with the highest amount NP-MAN, the prognosis of tumors in the advanced stage of NKT cells compared with other T-cell subpopulations HCC model was much worse than intrahepatic inocula- [33]. Indeed, NKT cells constitute about 30% of all lym- tion HCC model. However, Sim NP-MAN combined phocytes in the liver, pointing to a critical role in liver with PD-L1 antibody still achieved a satisfactory thera- disease [34]. Recently, several pieces of evidence showed peutic effect in the late-stage HCC model. that NKT cells exert important functions in antitumor Yu  et al. Journal of Nanobiotechnology (2022) 20:9 Page 13 of 16 Fig. 6 Combination of simvastatin NP and PD-L1 antibody shows synergistic efficacy in advanced HCC suppression. A PD-L1 level of HCC tissues from various treatments and normal liver were measured by flow cytometry. B Mice HCC model establishment and treatment scheme. C Bioluminescence imaging of Hepa1-6 tumor bearing mice in various treatment groups (n = 5). D Tumor growth curves. E Representative liver tumor morphology. Scale bar stands for 1 cm. F Survival curves (n = 5). **p < 0.01, ***p < 0.001, n.s. = not significant immunity. It was revealed that hepatic NKT cell accumu- Conclusions lation precedes HCC inhibition [35]. CXCR6-dependent In summary, the therapeutic effect stems from the dual Recruitment of NKT cells and CD4 T cells to the liver effect of simvastatin on tumor microenvironment in exerts important functions in tumor surveillance to HCC, as well as the effective PLGA-PEG-MAN NP inhibit hepatocarcinogenesis [36]. Our findings demon - encapsulation and LSEC-targeted delivery. The LSEC- strated that the CXCL16 expression on LSECs induced by targeted delivery of simvastatin notably exerts effect on simvastatin attracted C XCR6 NKT cells in the liver to stromal microenvironment regression and NKT cell inhibit tumor growth, which is consistent with previous recruitment in fibrotic HCC mouse model. Besides, the studies. IFN-γ is instrumental for NKT initiated tumor lower dose than oral administration exerts higher effi - immunity. Interestingly, it was found that NKT cells from ciency with no toxicity, which would do great benefit to cancer patients produce reduced amounts of IFN-γ than HCC patients in clinic whose liver function is vulnerable healthy subjects [37], while simvastatin targeted delivery with limited therapeutic tolerability. Our findings reveal to LSECs activated NKT cells and induced IFN-γ produc- an immune-based therapeutic mechanism of simvas- tion, which contributed to anti-tumor effect. tatin and offer a broad application of this drug to HCC patients based on tumor microenvironment remodeling. Yu et al. Journal of Nanobiotechnology (2022) 20:9 Page 14 of 16 Fig. 7 Schematic representation of the Nano delivery of simvastatin targeting LSECs to remodel HCC microenvironment Abbreviations ***p < 0.001. Figure S3. Synthesis of the LSEC-targeting PLGA-PEG. (A) HCC: Hepatocellular carcinoma; LSECs: Liver sinusoidal endothelial cells; The synthesis process of PLGA-PEG-mannan. (B) H-NMR spectra of the HSCs: Hepatic stellate cells; aHSCs: Activated hepatic stellate cells; HMG-CoA: synthesized particles PLGA-PEG-mannan. Comparing the integrated area 3-Hydroxy-3-methylglutaryl CoA; KLF2: Kruppel-like factor; NO: Nitric oxide; of peak g (mannan protons) and peak a (protons in PEG), the ratio of CXCL: Chemokine (C-X-C motif ) ligand; CXCR: Chemokine (C-X-C motif ) mannose to PEG is approximately 4.5. Figure S4. Simvastatin NPs show no receptor; NKT: Natural killer T; PD-1: Programmed death receptor 1; PD-L1: Pro- obvious toxicity in mice. (A) HE staining of major organs from mice with grammed death-1-ligand-1; NPs: Nanoparticles; PFA: Paraformaldehyde; H&E: various treatments. Scale bar stands for 50 μm. (B) Body weight changes Hematoxylin ad eosin; IHC: Immunohistochemistry; IF: Immunofluorescence; during treatment. (C) Mice blood routine test and (D) Hepatorenal func- TUNEL: Terminal deoxynucleotidyltransferase-mediated dUTP nick end labe- tion test after treatment (n = 5). n.s. = not significant. Table S1. Antibody ling; RT-PCR: Real-time polymerase chain reaction; DMSO: Dimethylsulfoxide; list. Table S2. Gene Primer list for real-time PCR. Table S3. Cytokine Primer TEM: Transmission electron microscopy; EE: Encapsulation efficiency; LC: Load- list for real-time PCR. ing capacity; AST: Aspartate aminotransferase; ALT: Alanine aminotransferase; BUN: Urea nitrogen; CRE: Creatinine; αSMA: α-Smooth muscle actin; LYVE1: Lymphatic vessel endothelial hyaluronan receptor 1; eNOS: Endothelial nitric Acknowledgements oxide synthase; LPS: Lipopolysaccharide; RNA-seq: RNA-sequencing; PLGA: The authors thank Eshelman School of Pharmacy, University of North Carolina, Poly(lactic-co-glycolic acid); PEG: Polyethylene glycol; MAN: Mannan. for logistic support and laboratory facilities. The authors thank Shuguang Hospital affiliated to Shanghai University of Traditional Chinese Medicine for providing the HCC tissue samples and related clinical data. Supplementary Information The online version contains supplementary material available at https:// doi. Authors’ contributions org/ 10. 1186/ s12951- 021- 01205-8. ZY and LH conceived study and designed procedures. ZY, JG, YL, ZL performed experiments and acquired data. ZY and MW analyzed and interpreted data. YG acquired clinical samples. ZY wrote the draft, LH revised the manuscript. Additional file 1: Figure S1. Simvastatin shows no obvious cytotoxicity. LH and YG supervised this work. All authors read and approved the final Cytotoxicity study of SK-Hep1, LX2 and Huh7 cells treated with simvastatin manuscript. at different concentrations (n = 3). Figure S2. Simvastatin inhibits the activity of HSC via LSEC. (A) The expression of KLF2 and eNOS was quanti- Funding fied in SK-Hep1 cells treated with indicated concentration of simvastatin This work was supported by NIH Grant (CA198999), National Natural Science for 24 h (n = 3). (B) The expression of α-SMA and collagen1 was quantified Foundation of China (82074154, 81774240), China Scholarship Council in LPS-activated LX2 treated with the supernatant of SK-Hep1 upon dif- (201808310047), Shanghai Rising-Star Program (17QA1403900), Training Plan ferent doses of simvastatin treatment with or without L-NAME (n = 3). (C) of Outstanding Young Medical Talents from Shanghai Municipal Health Bureau The expression of CXCL16 was quantified in SK-Hep1 cells treated with (2017YQ021), Youth Tip-top Talent program in Shanghai, Youth Xinglin Scholar the indicated concentrations of simvastatin for 24 h (n = 3). **p < 0.01, from Shanghai University of Traditional Chinese Medicine, Siming Scholar Yu  et al. Journal of Nanobiotechnology (2022) 20:9 Page 15 of 16 from Shanghai Shuguang Hospital (SGXZ-201904), and the “Talents Cultivation 13. Abraldes JG, Villanueva C, Aracil C, Turnes J, Hernandez-Guerra M, Program” of Jilin University. Genesca J, et al. Addition of simvastatin to standard therapy for the pre- vention of variceal rebleeding does not reduce rebleeding but increases Availability of data and materials survival in patients with cirrhosis. Gastroenterology. 2016;150(5):1160-70 All data generated or analyzed during this study are included in this published e3. article and its additional files. 14. Pose E, Napoleone L, Amin A, Campion D, Jimenez C, Piano S, et al. Safety of two different doses of simvastatin plus rifaximin in decompensated cirrhosis (LIVERHOPE-SAFETY ): a randomised, double-blind, placebo- Declarations controlled, phase 2 trial. Lancet Gastroenterol Hepatol. 2020;5(1):31–41. 15. Singh S, Singh PP, Singh AG, Murad MH, Sanchez W. Statins are associated Ethics approval and consent to participate with a reduced risk of hepatocellular cancer: a systematic review and This study was conducted in accordance with the mandates of the Helsinki meta-analysis. Gastroenterology. 2013;144(2):323–32. Declaration and the guidelines of the Clinical Research Ethics Committee 16. Kim G, Jang SY, Nam CM, Kang ES. Statin use and the risk of hepatocel- of SUTCM. All the patients enrolled in this study signed written informed lular carcinoma in patients at high risk: a nationwide nested case-control consent, and this study was approved by SUTCM Clinical Research Ethics Com- study. J Hepatol. 2018;68(3):476–84. mittee (Approval No. 2018-629-58-01). All mice experiments were approved 17. Simon TG, Duberg AS, Aleman S, Hagstrom H, Nguyen LH, Khalili H, et al. by the Institutional Animal Care and Use Committee of the University of North Lipophilic statins and risk for hepatocellular carcinoma and death in Carolina at Chapel Hill. patients with chronic viral hepatitis: results from a nationwide swedish population. Ann Intern Med. 2019;171(5):318–27. Consent for publication 18. Marrone G, Russo L, Rosado E, Hide D, Garcia-Cardena G, Garcia-Pagan Not applicable. JC, et al. The transcription factor KLF2 mediates hepatic endothelial protection and paracrine endothelial-stellate cell deactivation induced Competing interests by statins. J Hepatol. 2013;58(1):98–103. Leaf Huang is a consultant of PDS Biotechnology, Samyang Biopharmaceutical 19. Marrone G, Maeso-Diaz R, Garcia-Cardena G, Abraldes JG, Garcia-Pagan Co, Stemirna and Beijing Inno Medicine. All other authors declare that they JC, Bosch J, et al. KLF2 exerts antifibrotic and vasoprotective effects in have no competing interests. cirrhotic rat livers: behind the molecular mechanisms of statins. Gut. 2015;64(9):1434–43. Author details 20. Guo S, Lin CM, Xu Z, Miao L, Wang Y, Huang L. Co-delivery of cisplatin and Division of Pharmacoengineering and Molecular Pharmaceutics, Eshelman rapamycin for enhanced anticancer therapy through synergistic effects School of Pharmacy, University of North Carolina, Chapel Hill, NC 27599, USA. and microenvironment modulation. ACS Nano. 2014;8(5):4996–5009. Department of Liver Disease, Shuguang Hospital, Affiliated to Shanghai 21. Yu Z, Guo J, Hu M, Gao Y, Huang L. Icaritin exacerbates mitophagy and University of Traditional Chinese Medicine, Shanghai 201203, China. School synergizes with doxorubicin to induce immunogenic cell death in hepa- of Pharmaceutical Sciences, Jilin University, Changchun 130021, China. tocellular carcinoma. ACS Nano. 2020;14(4):4816–28. 22. Liu Q, Wang X, Liu X, Kumar S, Gochman G, Ji Y, et al. Use of polymeric Received: 19 September 2021 Accepted: 12 December 2021 nanoparticle platform targeting the liver to induce treg-mediated antigen-specific immune tolerance in a pulmonary allergen sensitization model. ACS Nano. 2019;13(4):4778–94. 23. Gracia-Sancho J, Lavina B, Rodriguez-Vilarrupla A, Garcia-Caldero H, Bosch J, Garcia-Pagan JC. Enhanced vasoconstrictor prostanoid production by References sinusoidal endothelial cells increases portal perfusion pressure in cirrhotic 1. Siegel RL, Miller KD, Jemal A. Cancer statistics, 2019. CA Cancer J Clin. rat livers. J Hepatol. 2007;47(2):220–7. 2019;69(1):7–34. 24. Pertea M, Kim D, Pertea GM, Leek JT, Salzberg SL. Transcript-level expres- 2. El-Serag HB. Hepatocellular carcinoma. N Engl J Med. sion analysis of RNA-seq experiments with HISAT. StringTie and Ballgown 2011;365(12):1118–27. Nat Protoc. 2016;11(9):1650–67. 3. Hernandez-Gea V, Toffanin S, Friedman SL, Llovet JM. Role of the 25. Canton J, Neculai D, Grinstein S. Scavenger receptors in homeostasis and microenvironment in the pathogenesis and treatment of hepatocellular immunity. Nat Rev Immunol. 2013;13(9):621–34. carcinoma. Gastroenterology. 2013;144(3):512–27. 26. Makarova-Rusher OV, Medina-Echeverz J, Duffy AG, Greten TF. The 4. Llovet JM, Zucman-Rossi J, Pikarsky E, Sangro B, Schwartz M, Sherman M, yin and yang of evasion and immune activation in HCC. J Hepatol. et al. Hepatocellular carcinoma. Nat Rev Dis Primers. 2016;14(2):16018. 2015;62(6):1420–9. 5. Buonaguro L, Mauriello A, Cavalluzzo B, Petrizzo A, Tagliamonte M. Immu- 27. Keir ME, Butte MJ, Freeman GJ, Sharpe AH. PD-1 and its ligands in toler- notherapy in hepatocellular carcinoma. Ann Hepatol. 2019;18(2):291–7. ance and immunity. Annu Rev Immunol. 2008;26:677–704. 6. Couvelard A, Scoazec JY, Dauge MC, Bringuier AF, Potet F, Feldmann G. 28. Muller WA. Transendothelial migration: unifying principles from the Structural and functional differentiation of sinusoidal endothelial cells endothelial perspective. Immunol Rev. 2016;273(1):61–75. during liver organogenesis in humans. Blood. 1996;87(11):4568–80. 29. Zimmermann HW, Bruns T, Weston CJ, Curbishley SM, Liaskou E, Li KK, 7. Poisson J, Lemoinne S, Boulanger C, Durand F, Moreau R, Valla D, et al. et al. Bidirectional transendothelial migration of monocytes across Liver sinusoidal endothelial cells: physiology and role in liver diseases. J hepatic sinusoidal endothelium shapes monocyte differentiation and Hepatol. 2017;66(1):212–27. regulates the balance between immunity and tolerance in liver. Hepatol- 8. Thomas H. LSEC stretch promotes fibrosis during hepatic vascular con- ogy. 2016;63(1):233–46. gestion. Nat Rev Gastroenterol Hepatol. 2019;16(5):262–3. 30. Zuo Y, Ren S, Wang M, Liu B, Yang J, Kuai X, et al. Novel roles of liver sinu- 9. Shetty S, Lalor PF, Adams DH. Liver sinusoidal endothelial cells— soidal endothelial cell lectin in colon carcinoma cell adhesion, migration gatekeepers of hepatic immunity. Nat Rev Gastroenterol Hepatol. and in-vivo metastasis to the liver. Gut. 2013;62(8):1169–78. 2018;15(9):555–67. 31. Carambia A, Freund B, Schwinge D, Bruns OT, Salmen SC, Ittrich H, et al. 10. Tsochatzis EA, Bosch J. Statins in cirrhosis-ready for prime time. Hepatol- Nanoparticle-based autoantigen delivery to Treg-inducing liver sinusoi- ogy. 2017;66(3):697–9. dal endothelial cells enables control of autoimmunity in mice. J Hepatol. 11. Pose E, Trebicka J, Mookerjee RP, Angeli P, Gines P. Statins: old drugs as 2015;62(6):1349–56. new therapy for liver diseases? J Hepatol. 2019;70(1):194–202. 32. Yu X, Chen L, Liu J, Dai B, Xu G, Shen G, et al. Immune modulation of liver 12. Abraldes JG, Albillos A, Banares R, Turnes J, Gonzalez R, Garcia-Pagan JC, sinusoidal endothelial cells by melittin nanoparticles suppresses liver et al. Simvastatin lowers portal pressure in patients with cirrhosis and metastasis. Nat Commun. 2019;10(1):574. portal hypertension: a randomized controlled trial. Gastroenterology. 33. Mossanen JC, Tacke F. Role of lymphocytes in liver cancer. Oncoimmunol- 2009;136(5):1651–8. ogy. 2013;2(11):e26468. Yu et al. Journal of Nanobiotechnology (2022) 20:9 Page 16 of 16 34. Zimmermann HW, Tacke F. Modification of chemokine pathways and immune cell infiltration as a novel therapeutic approach in liver inflam- mation and fibrosis. Inflamm Allergy Drug Targets. 2011;10(6):509–36. 35. Ma C, Han M, Heinrich B, Fu Q, Zhang Q, Sandhu M, et al. Gut microbi- ome-mediated bile acid metabolism regulates liver cancer via NKT cells. Science. 2018. https:// doi. org/ 10. 1126/ scien ce. aan59 31. 36. Mossanen JC, Kohlhepp M, Wehr A, Krenkel O, Liepelt A, Roeth AA, et al. CXCR6 Inhibits hepatocarcinogenesis by promoting natural killer T- and CD4(+) T-cell-dependent control of senescence. Gastroenterology. 2019;156(6):1877-89 e4. 37. Tahir SM, Cheng O, Shaulov A, Koezuka Y, Bubley GJ, Wilson SB, et al. Loss of IFN-gamma production by invariant NK T cells in advanced cancer. J Immunol. 2001;167(7):4046–50. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in pub- lished maps and institutional affiliations. Re Read ady y to to submit y submit your our re researc search h ? Choose BMC and benefit fr ? Choose BMC and benefit from om: : fast, convenient online submission thorough peer review by experienced researchers in your field rapid publication on acceptance support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year At BMC, research is always in progress. Learn more biomedcentral.com/submissions http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Nanobiotechnology Springer Journals

Nano delivery of simvastatin targets liver sinusoidal endothelial cells to remodel tumor microenvironment for hepatocellular carcinoma

Download PDF
16 pages

Loading next page...
 
/lp/springer-journals/nano-delivery-of-simvastatin-targets-liver-sinusoidal-endothelial-80kCBvb0sf

References (42)

Publisher
Springer Journals
Copyright
Copyright © The Author(s) 2021
eISSN
1477-3155
DOI
10.1186/s12951-021-01205-8
Publisher site
See Article on Publisher Site

Abstract

so-called capillarization, which amplifies  liver  damage Background by orchestrating the response of the liver microenviron- Hepatocellular carcinoma (HCC) is the second leading ment, permits and promotes HSCs activation, leading to cause of cancer-related deaths worldwide [1]. The etiol - liver fibrosis and carcinogenesis [8]. Reversal of capillar - ogy of HCC including hepatitis B virus infection, hepa- ized LSECs to differentiated LSECs promotes reversion titis C virus infection, alcoholic liver disease, metabolic of activated HSCs (aHSCs) to quiescence and induces syndrome and aflatoxin exposure usually causes injury senescence and apoptosis of aHSCs, which is critical and chronic inflammation in liver [2]. One unique feature for the treatment of both fibrosis and HCC. In addition, of HCC is its close association with chronic inflammation LSECs influence the composition of hepatic immune induced liver fibrosis. More than 80% of HCC develop in populations by distinct expression of adhesion mole- fibrotic or cirrhotic livers, suggesting an important role of cules and chemokines in diseased liver, which indicates liver fibrosis in the premalignant environment of the liver the crucial role of LSECs in immune-regulation [9]. The [3]. Besides radical treatment in early-stage HCC, there is characteristic of LSECs in contributing to HSC deacti- still an unmet clinical need for the majority of unresect- vation and immune response initiation makes them a able HCC. Unfortunately, most of them do not respond potential therapeutic target for HCC treatment. well to either multi-kinase inhibitors or immune-check- Statins, the 3-hydroxy-3-methylglutaryl CoA (HMG- point blocking antibodies, which are effective standard CoA) reductase inhibitors which decrease cholesterol cares for other cancers [4]. This failure is primarily due to synthesis, are usually used to treat dyslipidemia and car- the fibrotic microenvironment and the related immuno - diovascular disease. Recently, there is accumulating evi- suppressive factors in HCC [5]. Thus, stromal microen - dence that statins have beneficial effects in liver fibrosis vironment regression together with immunosuppressive and cirrhosis [10, 11]. Results of cohort studies have con- microenvironment remodeling will pose a potential strat- sistently found that liver cirrhotic patients who received egy to HCC therapy. statin treatment to reduce cholesterol had a lower risk Liver sinusoidal endothelial cells (LSECs) constitute of decompensation and death compared with patients a unique vascular bed in the liver, which receives the who did not receive statins [12–14].  Emerging clinical blood from both hepatic artery and portal veins into observational studies have also shown that statin use hepatic parenchyma [6]. LSECs are the first liver cell was associated with a reduced risk of HCC development type affected after liver  injury, and play a critical role in and related mortality [15–17]. Simvastatin, as the most the cellular crosstalk that regulates progressive chronic effective statin protecting the hepatic endothelium, was liver disease, which leads to fibrosis and carcinogenesis revealed to activate the transcription factor Kruppel-like [7]. Under physiological condition, differentiated LSECs factor (KLF2)-nitric oxide (NO) pathway to reverse LSEC function as gatekeeper to prevent hepatic stellate cells capillarization and mediate hepatic endothelial protec- (HSCs) from activation. While the LSEC differentiation tion [18, 19]. Mannose receptor is the typical scavenger usually lost prior to fibrosis during pathological process, Yu  et al. Journal of Nanobiotechnology (2022) 20:9 Page 3 of 16 receptors with high expression on LSECs, and mannan from NANOSOFT POLYMERS. Other synthetic mate- exclusively acts as a ligand to target the mannose recep- rials as previously reported were obtained from Sigma- tor, being ability to facilitate simvastatin into LSECs [25]. Aldrich MO, USA [20]. u Th s, the specific manipulation of this important switch in selected LSECs might be more effective than oral Antibodies administration of statin nowadays as a strategy to treat In vivo anti-mouse IgG, CD1d, CD8 and CD4 mAb were HCC. purchased from BioXcell (West Lebanon, NH). Primary Here, we demonstrate that LSEC capillarization is asso- and secondary antibodies used for immunoblotting, ciated with fibrotic HCC progression and poor survival. immunostaining and flow cytometry are listed in Addi - Simvastatin converts HSCs from activation to quiescence tional file 1: Table S1. by stimulating KLF2-NO signaling in LSECs and up-reg- ulates the expression of Chemokine (C-X-C motif ) ligand Murine HCC models establishment 16 (CXCL16) on LSECs. In fibrotic HCC mouse model, Four-week-old male C57BL/6 mice were purchased from targeted transfer of simvastatin into LSECs by nano- Jackson Laboratory. All animal regulations and proce- technology not only mitigates LSEC capillarization to dures were accepted by the Institutional Animal Care improve the stromal microenvironment, but also recruits and Use Committee of the University of North Caro- natural killer T (NKT) cells to suppress tumor progres- lina at Chapel Hill. For fibrotic HCC model, 4-week-old sion. Combined with anti-PD-L1 antibody, simvastatin male immune-competent C57BL/6 mice received intra- nanoparticles (NPs) evoke an improved therapeutic effect peritoneal injection of CCl (1 μL/g, diluted in 100 μL at hemi-splenic inoculated advanced-stage HCC. These olive oil) twice a week for 4  weeks. Age-matched mice findings unravel an immune-based therapeutic mecha - administered with olive oil served as control. Intrahe- nism of simvastatin and provide a novel therapeutic patic injection of 5 × 10 Hepa1-6-luc cells in 50 μL PBS strategy based on tumor microenvironment remodeling was performed at the end of the fourth week. Mice were against HCC. sacrificed at 3  weeks post-tumor cell inoculation or at humane endpoint. The tumors were isolated from the Materials and methods liver and measured with caliper. The volume was calcu - Human specimens lated with the formula of length × width × width/2, and Medical history from 26 non-fibrotic HCC patients and the average value of each group was equal to the sum of 38 fibrotic HCC patients were collected for survival anal - total volumes divided by the number of mice. The tumor ysis. Immunohistochemistry were performed in a sub- growth indicated by luciferase intensity was monitored set of patients with fibrotic HCC or non-fibrotic HCC. using IVIS (Perkin Elmer, CA) imaging every 2  days. All 64 patients with HCC were obtained from Shuguang Tumor and matched non-tumor liver tissues were col- Hospital affiliated to Shanghai University of Traditional lected for section analysis. For advanced-stage HCC Chinese Medicine (SUTCM) and approved by SHUTCM model, hemi-splenic injection of Hepa1-6 cells was Clinical Research Ethics Committee. performed after 4-week CCl exposure. To be specific, an incision located below the left rib cage was made to Cell culture exteriorize the spleen. The spleen was tied and cut into Mouse hepatoma cell line Hepa1-6, human hepatocel- two parts, each containing an intact vascular pedicle for lular carcinoma cell line Huh7, LSEC cell line SK-Hep1 each half of the spleen. The distal section of the spleen and HSC cell line LX2 were maintained in high glucose was inoculated with 1 × 10 Hepa1-6 cells in 150 μL PBS. Dulbecco’s modified Eagle Medium (DMEM, Gibco) sup - The half spleen containing inoculated cells were resected plemented with 10% bovine calf serum (BCS) (Hyclone), 5 min after inoculation allowing the cancer cells to enter 1% penicillin and streptomycin (Hyclone). Two μg/mL the portal vein. The other half of the spleen was returned puromycin was used additionally for Hepa1-6-luciferase to the cavity to keep the immune system competent. (luc) cells. All cells were grown at 37  °C in a humidified incubator with 5% CO . Synthesis and characterization of the drug‑loaded mannan modified NPs Reagents The PLGA copolymers were obtained by a previously Simvastatin with over 99% purity (Sigma-Aldrich MO, reported method [20, 21]. The PLGA-PEG and PLGA- USA) were used for the study. Mannan and L-NAME PEG-MAN were first synthesized. Mannan attachment were purchased from Sigma-Aldrich. DiD was purchased to the particle surface was achieved through covalent from ThermoFisher Scientific. Poly(lactic-co-glycolic acid attachment by conjugation chemistry of the COOH-ter- (PLGA)-Polyethylene glycol (PEG)-NH2 was purchased minus of PLGA and mannan [22]. Yu et al. Journal of Nanobiotechnology (2022) 20:9 Page 4 of 16 Gene expression profiling Two mg simvastatin was dissolved in 200  µL DMSO, SK-Hep1 cells in triplicate were treated with simvastatin and then mixed with 2  mg PLGA-PEG-MAN and 6  mg (20  μM) and control for 24  h. Total RNA was extracted PLGA-PEG in the drug solution. The drug-polymer mix - using Trizol reagent (Invitrogen, Carlsbad, CA) accord- ture was added dropwise to 10 mL deionized water while ing to the manufacturer’s protocol. RNA quality was stirring, and stirred for 4  h. Nanoparticles were purified assessed on an Agilent 2100 Bioanalyzer (Agilent Tech- by ultrafiltration using 10 k cutoff columns at 4000×g. nologies, Palo Alto, CA) and checked using RNase free The particle size and zeta potential of nanoparticle agarose gel electrophoresis. mRNAs were isolated and were measured by Malvern Nano-ZS (Malvern Instru- fragmented to about 200 base pairs length and reverse ments, UK). The morphology was analyzed by transmis - transcribed into cDNA using the QuantiTect Reverse sion electron microscopy (TEM). Five μL particles were Transcription kit (Qiagen). The cDNA fragments were added on 400-mesh carbon-filmed copper grids (Agar purified with QiaQuick PCR extraction kit (Qiagen, Scientific) for 2  min and stained with 2% (w/w) uranyl Venlo, Netherland), end repaired, poly(A) added, and acetate before imaging using TEM (JEOL JEM1230). ligated to Illumina sequencing adapters. The cDNA The drug encapsulation efficiency (EE) was calculated library products were size selected by agarose gel elec- by (Entrapped Drug/Drug Added)*100%. Loading capac- trophoresis, PCR amplified and sequenced using Illumina ity (LC) was calculated by (Entrapped Drug/Nanoparticle HiSeqTM4000 platform (Illumina, San Diego, CA). Tran- Weight)*100%. script-level expression analysis of sequencing data was performed using HISAT and StringTie software (http:// In vivo treatment studies ccb. jhu. edu/ softw are,shtml) [24]. Differential transcripts For intra-hepatic inoculation HCC model, the mice with of chemokines with stringent cutoff coefficient of less liver tumors were treated 3  days after the tumor cell than 0.05 were obtained to align to GO database (http:// inoculation. Five groups of mice were separately given www. geneo ntolo gy. org) for protein functional annotation PBS, blank PLGA NPs, simvastatin PLGA NPs (20  mg/ corresponding to immune population. kg), simvastatin PLGA-MAN NPs (20 mg/kg) by tail vein injection and simvastatin free drug oral gavage (40  mg/ MTT assay kg) every other day for totally 5 times. The anti-tumor Cell viabilities were assessed by MTT assay. 10^4 cells efficacy was regularly assessed using IVIS (Perkin Elmer, were seeded in 96-well plates per well overnight and sub- CA) imaging every 2  days and survival was recorded. jected to different treatments. Five mg/mL MTT (Alfa Some mice were sacrificed 3  days after treatment, the Aesar) reagent was added for 4  h at 37  °C, and then the tumors and matched non-tumor liver tissues were col- supernatant was discarded. The formazan was resus - lected and prepared for section study and immune effect pended in 100  μL of DMSO and absorbance was exam- evaluation. For immune cell depletion study, after Hepa1- ined by a spectrometer (Hidex Chameleon). 6-Luc intrahepatic inoculation, PBS, IgG, anti-CD1d, anti-CD4 or anti-CD8 antibody (100  µg) was intraperi- Histology toneally injected to mice every other day for totally 5 Liver tissues with tumor or major organs including injections 1  day before NPs treatment. For hemi-splenic hearts, livers, spleens, lungs, and kidneys were col- inoculation HCC model, intervention was applied 5 days lected and were fixed in 4% paraformaldehyde (PFA). after tumor cell inoculation. Four groups of mice were Fixed samples were paraffin-embedded, sectioned, and given separately by PBS, simvastatin PLGA-MAN NPs stained with hematoxylin and eosin (H&E) or Masson’s (20  mg/kg), anti-PD-L1 antibody (100  µg) or combina- trichrome at UNC histology facility. PFA-fixed sam - tion. NPs tail vein injection and anti-PD-L1 antibody ples were embedded with optimum cutting temperature intraperitoneal injection were performed every other day compound and sectioned at 8  μm thickness. For immu- for a total 5 times. nohistochemistry (IHC), sections were incubated with primary antibodies at 4  °C overnight, washed, and incu- Mouse liver sinusoidal endothelial cell isolation bated with horseradish peroxidase-conjugated secondary Mouse LSECs were isolated as previously described [23]. antibodies for 2  h at room temperature. Digital images In brief, livers were perfused through the portal vein were taken using brightfield light microscope (Olympus and digested with a collagenase solution. After mincing BX61). Immunofluorescence (IF) was performed using the liver, cells were filtered and centrifuged at 50×g to fluorescent antibodies and counterstained with Prolong remove hepatocytes. Non-parenchymal cells were then Diamond Antifade Mountant with DAPI (ThermoFisher separated by differential centrifugation using a Percoll Scientific). Antibodies are listed in Additional file  1: gradient. Kupffer cells were eliminated by plastic pre-cul - Table  S1. Apoptotic cells were stained with a terminal ture for 30 min. LSEC were collected for RNA extraction. Yu  et al. Journal of Nanobiotechnology (2022) 20:9 Page 5 of 16 deoxynucleotidyltransferase-mediated dUTP nick end Griess Reagent and color was developed in 10  min. The labeling (TUNEL) kit (Promega, Madison, WI). Images absorbance was measured at 540  nm and NO concen- were taken using laser-scanning confocal fluorescence tration was calculated referring to a standard curve. The microscope (Zeiss LSM 710). Liver samples fixed with experiments were done in triplicate. 4% PFA by perfusion through portal vein were sectioned using vibratome at UNC microscopy services laboratory Biodistribution and prepared for scanning electron microscopy (Zeiss 0.05% (wt) of lipophilic carbocyanine DiD (Ther - Supra 25 FESEM). Five random microscopic fields were moFisher) was used to formulate the DiD-labeled PLGA- selected and quantified by ImageJ software. Porosity was PEG or PLGA-PEG-MAN NP. In mice bearing fibrotic measured as percentage of LSEC surface occupied by tumor, 24  h following tail vein injection of DiD-labeled fenestrae in SEM in liver tissue. Fenestration frequency NP, major organs and liver with tumors were collected was calculated with total number of fenestrations divided and analyzed using IVIS (Perkin Elmer, CA) with the by total area of LSEC surface. excitation wavelength at 640 nm and the emission wave- length at 670 nm. Flow cytometric analysis Single-cell suspensions from tumor tissue were harvested Blood chemistry analysis in MACs buffer (1 × PBS + 2  mM EDTA + 0.5% BSA, fil - Seven days after 5 doses of nanoparticle injections, ter sterile), then subjected to conjugated staining with whole blood was obtained from the normal mice. Blood fluorescence. At least 10,000 live cells were subjected routine including RBC, WBC, PLT, HGB, and HCT was to flow cytometric analysis on a flow cytometer (Bec - tested using the whole blood. Aspartate aminotransferase ton Dickinson LSR II). Experimental data were analyzed (AST), alanine aminotransferase (ALT), urea nitrogen using FlowJo software. The antibodies used are listed in (BUN), and creatinine (CRE) in serum were analyzed as Additional file 1: Table S1. indicators of hepatic and renal function using the serum. Immunoblotting Statistical analysis Cells were lysed in RIPA lysis buffer with protease inhibi - GraphPad Prism 8 (GraphPad Software, La Jolla, Califor- tors. Total lysates were quantified by a BCA Protein Assay nia, USA) was used for statistical analysis. The independ - Kit (Biorad, CA). Thirty μg protein samples were used for ent Student t-test was used to compare data between immunoblotting analysis. After incubating with appro- the two groups. A one-way analysis of variance test with priate primary and secondary antibodies, the immuno- Bonferroni correction was used to compare data in mul- reactions were visualized with Western HRP substrate tiple groups of mice. Pearson correlation coefficient was (Thermo, Rockford, IL). The antibodies used are listed in calculated to show the correlation of two parameters. Additional file 1: Table S1. Two-way ANOVA with a Bonferroni post-hoc test was used to estimate the difference in multiple groups of mice Quantitative real‑time polymerase chain reaction (RT‑PCR) when considering two factors of time and treatment. assay Kaplan–Meier survival analysis was used to determine Total RNA was extracted from cells or the whole tumor the overall survival rates and tumor incidence-free rates, using an RNeasy microarray mini kit (Qiagen, Hilden, the differences were compared by the log-rank Mantel- Germany) and was reverse-transcribed to cDNA with Cox test. Data averages from each group are presented as an iScript cDNA synthesis kit (Bio-Rad, Hercules, CA). mean ± SD; *p < 0.05, **p < 0. 01, ***p < 0. 001. Quantitative PCR was performed in a 7500 RT-PCR system. The PCR primers are listed in Additional file  1: Tables S2 and S3. Results Liver fibrosis‑associated LSEC capillarization contributes Nitric oxide assay to aggressive HCC development Nitric oxide (NO) amount was assessed using Nitric To observe the role of fibrotic microenvironment dur - Oxide Colorimetric Assay Kit (BioVision, Milpitas, CA) ing HCC development, Hepa1-6 orthotopic HCC model in accordance to the manufacturer’s manual. Briefly, SK- was established using C57BL/6 mice pretreated with Hep1 cells were treated with different concentrations of CCl injection which induces chronic inflammation and simvastatin (0, 5, 20  μM) for 24  h and the supernatants fibrosis (Fig.  1A). Interestingly, tumor growth acceler- were collected. Equal aliquot of samples in every group ated more rapidly in fibrotic liver than the normal liver, were mixed with Nitrate Reductase Buffer and incubated as shown by the tumor growth curve (Fig.  1B) and bio- at room temperature for 1 h. The mixture was then added luminescence imaging of tumors, as well as liver tumor Yu et al. Journal of Nanobiotechnology (2022) 20:9 Page 6 of 16 Fig. 1 Liver fibrosis-associated LSEC capillarization contributes to aggressive HCC development. A Schematic diagram of liver fibrosis-associated HCC mouse model. B Tumor growth curve and C Bioluminescence imaging of tumor and representative liver tumor morphology at the endpoint. D Statistical analysis of tumor volume and weight. E Representative pictures of Masson trichrome staining, α-SMA immunofluorescence in mice non-fibrotic and fibrotic liver tissue. Scale bar stands for 50 μm. Yellow dotted line indicates the border between non-tumor and tumor tissue. The quantification of collagen density, α-SMA expression were shown. F LYVE1 and CD31 co-immunofluorescence in mice non-fibrotic and fibrotic liver tissue, scale bar stands for 50 μm. Scanning electron micrographs illustrating the morphological response of LSEC. Scale bar stands for 1 μm. The quantification of LYVE1 and CD31 expression were shown, as well as the porosity and fenestration on LSEC. G Percentage of α-SMA, CD31 fluorescence and tumor weight are pairwise positively correlated. H Kaplan–Meier overall survival and disease-free survival curves of HCC patients with non-fibrotic tumor (n = 26) or fibrotic tumor (n = 38). I Expressions of α-SMA and CD31 in non-fibrotic HCC and fibrotic HCC tissues as determined by immunohistochemistry. Scale bar stands for 50 μm.*p < 0.05, **p < 0.01, ***p < 0.001 morphology at the endpoint (Fig. 1C). Tumor weight and markers were highly expressed in peri-tumor and tumor volume are confirmed as well (Fig. 1D). tissues in C Cl -treated group as expected (Fig. 1E). Since To further investigate the fibrotic microenvironment LSECs play critical role during fibrosis, lymphatic vessel responsible for HCC development, we first evaluated the endothelial hyaluronan receptor  1 (LYVE1) and CD31 expression of α-smooth muscle actin (αSMA) and colla- which are the markers of differentiated and capillariza - gen which are key markers of aHSCs and fibrosis. Both tion LSECs, respectively, were assessed. Interestingly, Yu  et al. Journal of Nanobiotechnology (2022) 20:9 Page 7 of 16 high LYVE1 expression was observed in both peri-tumor (RNA-seq) in simvastatin treated SK-Hep1 cells, or not, and tumor tissues of non-fibrotic liver, while CD31 was to examine whether simvastatin affect the immune-regu - over-expressed in fibrotic liver (Fig.  1F). The porosity and lated genes in LSEC. Gene expression profiles analysis of fenestration on LSECs were much reduced in fibrotic RNA-seq data suggested that a set of chemokines, which liver compared to the non-fibrotic liver (Fig.  1F). We are classified in terms of the function on the regulation of further found that αSMA, CD31 expression level and the variety of leukocytes, were changed in SK-Hep1 cells tumor mass were significantly pairwise positively corre - treated with simvastatin compared with the untreated lated (Fig.  1G), while LYVE1 expression was negatively control cells. The investigation coupled with functional correlated with tumor mass. Thus, our results showed categorization of changed chemokine genes revealed that LSEC capillarization may contribute to aggressive HCC CXCL16 was identified as the significant regulator upon development. simvastatin treatment (Fig. 2F). Further quantitative PCR We next investigated the clinical relevance of LSEC and western blot detection confirmed that transcript capillarization in HCC patients with or without fibrosis/ level and protein expression of CXCL16 greatly increased cirrhosis. The Kaplan–Meier analysis revealed that HCC in SK-Hep1 treated with simvastatin in a dose-depend- patients with fibrosis were significantly correlated with ent manner (Fig.  2G, H and Additional file  1: Fig. S2C). shorter overall and disease-free survival rates (Fig.  1H). NKT cells were characterized with high expression of Consistent with the mouse model, the fibrotic HCC tis - CXCR6 (Fig.  2I), and CXCL16 as the ligand for CXCR6 sue exhibited high expression level of αSMA and CD31 exerts recruitment function on NKT cells. Further study (Fig.  1I). Thus, poor survival in fibrotic HCC patients is indicated that simvastatin alone could not influence the related to LSEC capillarization. activity of NKT cells, while simvastatin could upregulate the expression of CD69 and IFN-γ in NKT cells in the co- Simvastatin deactivates HSC via LSEC and up‑regulates culture with SK-Hep1, suggesting simvastatin evoked the CXCL16 expression on LSEC antitumor NKT immune response through the mediation Since simvastatin exerts beneficial effects in fibrosis of LSEC (Fig.  2J). Thus, simvastatin can not only deac - and HCC, we first examined the effect of simvasta - tivate HSCs via KLF2-NO signaling in LSECs, but also tin on three human cell lines which are SK-Hep1 LSEC induce CXCL16 overexpression in LSECs which recruits cell line, LX2 HSC cell line and Huh7 HCC cell line. We and activates NKT cells to the liver. found simvastatin almost had no cytotoxicity to all three cell lines (Additional file  1: Fig. S1). However, simvasta- Design of simvastatin LSEC‑targeted delivery NP platform tin increased both mRNA and protein expression levels Simvastatin is poor soluble in aqueous solution. The of KLF2 and endothelial nitric oxide synthase (eNOS) in PLGA polymer approved by FDA for drug delivery was SK-Hep1 cell line in a dose dependent manner (Fig.  2A, selected for encapsulating hydrophobic simvastatin due B and Additional file  1: Fig. S2A). The NO excretion was to the characteristic of both biodegradability and bio- up-regulated according to the activation of KLF2-eNOS compatibility. A unique characteristic of LSECs is their signaling induced by simvastatin (Fig. 2C). When lipopol- expression of high levels of scavenger receptors, which ysaccharide (LPS)-activated LX2 treated with simvasta- ensure the high endocytic capacity [7]. One of the most tin directly, no change of the activation marker αSMA extensively studied scavenger receptors on LSECs is and collagen was observed. Since simvastatin increased the mannose receptor [25]. Thus, mannan was selected NO excretion in SK-Hep1, we collected the conditioned as a ligand to target the mannose receptor on LSEC. medium of simvastatin treated SK-Hep1 cells and added The structure of synthetic PLGA-PEG-mannan was to cultured LX2 cells. The mRNA and protein expression confirmed by NMR. The H NMR spectrum of PLGA- levels of αSMA and collagen I were both down-regulated PEG-mannan showed that δ 4.639(H1), δ 3.51(H2), δ dose-dependently (Fig.  2D, E and Additional file  1: Fig. 5.20(H3)and δ 4.90(H4)are associated with CH of man- S2B). Further, an NOS inhibitor L-NAME was used to nan, CH2 of PEG, CH of lactic acid and CH2 of gly- verify the NO induced-HSC deactivation. Indeed, LX2 colic acid, respectively (Additional file  1: Fig. S3), which activation recovered after L-NAME intervention (Fig. 2D, indicated the success of synthesis. We utilized a solvent E and Additional file  1: Fig. S2B). Thus, simvastatin displacement process to load simvastatin (Fig.  3A). seemed to work on SK-Hep1 cells which then secreted PLGA NPs containing simvastatin presented a micelle- NO, or an NO-related factor, to deactivate LX2 cells. like structure with a diameter of approximately 100 nm In addition, since LSECs play critical roles in the com- revealed by transmission electron microscopy (Fig. 3B). position of hepatic immune populations by mediating the The particle size around 120  nm and zeta potential recruitment of leukocyte subsets through the expression around 3  mV were confirmed by particle sizing sys - of distinct chemokines, we performed RNA-sequencing tem (Fig.  3C, D). The drug encapsulation efficacy of Yu et al. Journal of Nanobiotechnology (2022) 20:9 Page 8 of 16 Fig. 2 Simvastatin deactivates HSC via LSEC and recruits NKT cells. A Quantitative RT-PCR analysis and B immunoblotting of KLF2 and eNOS in SK-Hep1 LSEC cells treated with the indicated concentrations of simvastatin for 24 h (n = 3). GAPDH acts as the loading control. C Extracellular concentration of NO upon different doses of simvastatin treatment (n = 3). D Quantitative RT-PCR analysis and, E immunoblotting of ACTA2 and COL1 in LPS-activated LX2 treated with the supernatant of SK-Hep1 upon different doses of simvastatin treatment with or without L-NAME (n = 3). GAPDH acts as the loading control. F Gene expression profiles of chemokines analyzed with RNA-seq data which derived from SK-Hep1 cells treated with simvastatin or control. Simvastatin-responsive chemokine genes were grouped in association with clusters of various immune cell-regulating functions. The p value of each comparison was indicated with colors. G Quantitative RT-PCR analysis and H immunoblotting of CXCL16 in SK-Hep1 cells treated with the indicated concentrations of simvastatin for 24 h (n = 3). GAPDH acts as the loading control. I CXCR6 expression was detected in NKT cells. J The expression of CD69 and IFN-γ was detected in NKT cells treated with control, simvastatin, the co-culture with SK-Hep1 cells, and the combination, respectively (n = 3). *p < 0.05, **p < 0.01, ***p < 0.001, n.s. = not significant Yu  et al. Journal of Nanobiotechnology (2022) 20:9 Page 9 of 16 simvastatin was approximately 85%, loading capacity Mice were treated by tail vein injection of PBS, blank was 6.5%. NPs, simvastatin un-targeted NPs (Sim NP) (20  mg/kg), NPs with mannan target tend to accumulate much simvastatin LSEC-targeted NPs (Sim NP-MAN) (20 mg/ more in fibrotic liver with tumors, but less in other kg) and simvastatin oral administration (40  mg/kg), main organs than the non-targeted NPs at 24  h post respectively (Fig.  4A). Notably, Sim NP-MAN impres- tail vein injection, which was tracked using bioimaging sively mitigated rapid tumor growth compared to PBS or (Fig.  3E, F). The uptake of NPs in fibrotic liver tissues blank NP-treated group, as shown by the lower increased was further examined by immunofluorescence stain - rate of tumorous luciferase intensity and tumor growth ing. As expected, mannan-coated particles encapsulat- curves. Either simvastatin oral administration or Sim ing DiD were captured by LSECs almost 4 times higher NPs injection exhibited partial inhibitory effect (Fig.  4B, compared to non-coated ones both in tumor and non- C). Besides, a significant survival prolongation from 23 tumor tissues (Fig. 3G). days in the control group to 43 days in the Sim NP-MAN treated group was also noticed (Fig.  4D). The livers with LSEC‑targeted delivery of simvastatin inhibits tumor tumor were excised at day 21 for inspection, liver tumor growth in mouse fibrotic HCC model load of the Sim NP-MAN treated group were signifi - To investigate the effect of targeted delivery of simvas - cantly reduced than those of the other groups (Fig.  4E). tatin in  vivo, the Hepa1-6 orthotopic HCC model was u Th s, LSEC-targeted delivery of simvastatin showed bet - established using C57BL/6 mice pretreated with CCl . ter anti-HCC effect than oral administration and un-tar - Three days after tumor cells inoculation, treatment was geted delivery. administrated every other day for continuous 5 times. Fig. 3 Simvastatin LSEC-targeted delivery NP platform was designed. A Efficient co-encapsulation of simvastatin into PLGA NP using a solvent displacement technique. B TEM images of PLGA NPs loaded with simvastatin cores. C Particle size of the PLGA NP. D Zeta Potential of the PLGA NP. E Images and F quantification of the DiD-loaded PLGA (DiD/PLGA-PEG) and mannan conjugated PLGA(DiD/PLGA-PEG-MAN) in liver with tumor and other major organs at 24 h after injection in mice bearing Hepa1-6 liver tumor. G Fluorescence imaging detecting DiD in PLGA-PEG or mannan-coated PLGA-PEG and LSEC marked with CD31, scale bar stands for 50 μm. *p < 0.05, **p < 0.01, ***p < 0.001, n.s. = not significant Yu et al. Journal of Nanobiotechnology (2022) 20:9 Page 10 of 16 Fig. 4 Simvastatin NPs suppress HCC development. A Tumor inoculation and treatment scheme. B Bioluminescence imaging of Hepa1-6 tumor bearing mice during tumor progression in various treatment groups. C Tumor growth curves and, D survival curves of Hepa1-6 tumor bearing mice in various treatment groups (n = 5). E Representative image of liver with Hepa1-6 tumor and quantification of liver weight at day 21 after inoculation (n = 3). Scale bar stands for 1 cm. *p < 0.05, **p < 0.01, ***p < 0.001, n.s. = not significant LSEC‑targeted delivery of simvastatin remodels stromal As CXCL16 is the ligand for CXCR6 which is expressed environment and recruits NKT cells in fibrotic HCC mostly on NKT cells, the cell frequencies of CXCR6 We next examined the effect of simvastatin delivered NKT cell in tumor tissues of various groups were evalu- with mannan-targeted NPs to remodel stromal environ- ated by flow cytometry. CXCR6 NKT cell frequency ment in the liver. HE staining revealed that the disorder increased from 0.82% in control group to 11.41% in structure of both tumor and non-tumor fibrosis tissue Sim NP-MAN treated group (Fig.  5F). Both CD69 and reversed in Sim NP-MAN treated group (Fig.  5A). The IFN-γ which are the activation markers of NKT cells collagen deposition reduced significantly as shown in were up-regulated (Fig.  5G). We further examined the Masson-trichrome staining. Accordingly, the expression critical cytokines in tumor tissues of various groups. of αSMA was approximately sevenfold down-regulated The increased expression of the immunostimulatory compared to control group (Fig. 5B). The over-expressed cytokines including CXCL16, IFN-γ, TNFα and the LSEC capillarization marker CD31 were remarkably decrease of immunosuppressive cytokines including down-regulated, while differentiated marker LYVE1 CXCL9, IL-6, TGFβ were noticed with Sim NP-MAN up-regulated in Sim NP-MAN treated group. In addi- treatment (Fig.  5H). Also, the TUNEL assay revealed tion, both porosity and fenestration frequency of LSECs increased number of apoptotic cells in tumor tissues after increased significantly (Fig.  5C). Further, we found the Sim NP-MAN treatment (Fig. 5I). To address the immune + + mRNA expression levels of Klf2 and Nos3 in LSECs mechanism, we further depleted NKT, CD4 or CD8 T extracted from liver tissues were up-regulated signifi - cells in mice HCC model (Fig. 5J). The tumor restraining cantly, which is consistent with the observation in  vitro, effect was abolished by treatment with anti-CD1d, anti- indicating the Klf2-NO signaling were activated by tar- CD8 or anti-CD4 antibody, whereas it was not affected geted delivery of simvastatin (Fig. 5D). Since simvastatin- when an isotype-matched IgG control was used. CD1d treated LSECs express CXCL16, we next examined the and CD4 antibody showed the most obvious abolishment co-localization of CXCL16 and LYVE1 in liver tissue. As which indicates that NKT cells played a more critical role expected, CXCL16 expression increased together with during the anti-tumor process than the CD8 T-cells LYVE1 after Sim NP-MAN treatment (Fig. 5E). (Fig. 5J). Taken together, the result suggested that stromal Yu  et al. Journal of Nanobiotechnology (2022) 20:9 Page 11 of 16 Fig. 5 Simvastatin NPs remodel tumor microenvironment in fibrotic HCC. A HE staining of tissue features from HCC mice with various treatments. Yellow dotted line indicates the border between normal liver and tumor tissue. The left side is non-tumor liver tissue and right side is tumor tissue. Representative images are shown. Scale bar stands for 50 μm. B Masson trichrome (MT ) staining shown together with αSMA fluorescence staining in liver tissues. Data are presented as mean ± SD. Scale bar in MT image stands for 20 μm, in IF image stands for 50 μm. C LYVE1 and CD31 expression by fluorescence staining shown together with LSEC feature using SEM in liver tissues. The quantification of porosity and fenestration of LSEC is shown (n = 5). Scale bar in IF image stands for 50 μm, in SEM image stands for 1 μm. D Quantitative RT-PCR analysis of Klf2 and Nos3 in LSECs extracted from liver tissues intervened with various treatments (n = 5). E The fluorescence staining of LYVE1 and CD31. Scale bar stands for 50 μm. + + F NKT cell frequency in liver tissues detected by flow cytometry (n = 5). G CD69 and IFN-γ levels of hepatic NKT cells detected by flow cytometry (n = 5). H mRNA expression of chemokines and cytokines detected by real-time PCR in tumors in various treatment groups (n = 5). I The average number of apoptosis cells per high-power field (HPF) detected by TUNEL immune fluorescence staining. Scale bar stands for 50 μm. J Scheme of immune cell depletion study and tumor growth curves of Hepa1-6 tumor treated with simvastatin NP after anti-CD1d, CD4or CD8 intraperitoneal injection (n = 5). *p < 0.05, **p < 0.01, ***p < 0.001, n.s. = not significant environment re-modulation and NKT cell recruitment Simvastatin NPs show no obvious toxicity might be the main reason for the antitumor effect of Sim To further evaluate the safety of simvastatin NPs for the NP-MAN in fibrotic HCC. development of both effective and translational therapy, biosafety-related toxicological serum and pathology Yu et al. Journal of Nanobiotechnology (2022) 20:9 Page 12 of 16 analyses were performed. The pathology revealed Discussion by H&E staining showed no significant morphologi - Our study indicated that LSEC capillarization accounts cal damage among major organs including heart, lung, for tumor progression and poor survival in fibrotic HCC. liver, spleen and kidney (Additional file  1: Fig. S4A). The Simvastatin functions as hepatic endothelium protec- body weight remained stable without severe weight loss tor stimulates KLF2-NO signaling in LSECs to regress between different groups during experiment (Additional HSC activation. Simvastatin also up-regulates CXCL16 file  1: Fig. S4B). The blood routine and hepatorenal func - in LSECs which can recruit NKT cells. In order to exert tion of mice in all groups remained in the normal range the dual effect of simvastatin on LSEC, we targeted- (Additional file  1: Fig. S4C, D). Thus, simvastatin NPs delivered simvastatin to LSECs by a PLGA-PEG-MAN showed no obvious toxicity in the mouse model. NP formulation. LSEC-targeted delivery of simvastatin not only alleviates LSEC capillarization to regress the stromal microenvironment, but also recruits NKT cells Simvastatin NPs together with PD‑L1 antibody achieve to remodel immunosuppressive microenvironment and the synergistic effect in late‑stage HCC inhibit tumor progression in murine HCC model (Fig. 7). Despite surveillance, HCC often presents in clinic at In advanced-stage HCC, combination therapy of simvas- an  advanced  stage  that only systemic therapy is feasi- tatin NP and anti-PD-L1 antibody achieves an improved ble. Currently, most studies on the mechanisms of HCC therapeutic effect. immune evasion have focused on the programmed death LSECs are highly specialized endothelial cells in receptor 1 (PD-1)/programmed death ligand 1 (PD-L1) liver, which lies between blood cells on the sinusoidal pathway [26]. PD-1 is an immunoinhibitory receptor side and hepatocytes and HSCs on the abluminal side expressed in activated T and B cells, as well as NKT cells. [28]. With the structure of fenestrae, LSECs represent PD-L1, the major ligand of PD-1, is expressed in various a permeable barrier allowing exchanges between two immune cells such as antigen-presenting cells, as well as sides. However, LSECs do not simply form a barrier but in endothelial cells [27]. The PD-L1 expression was higher exert instrumental effect in maintaining microenviron - in Hepa1-6 tumor tissue compared to normal tissue, and ment homeostasis and mediating immune response in simvastatin NPs showed no effect on PD-L1 expression liver. Differentiated LSECs are the gatekeeper to main - (Fig.  6A). To explore an effective strategy for late-stage tain HSC in their quiescent state, which are critical to HCC treatment, the hemi-spleen inoculation of Hepa1-6 suppress fibrogenesis and carcinogenesis [29]. LSECs based on C Cl -induced fibrosis model was used to test also influence the composition of hepatic immune the therapeutic effect of simvastatin NPs combined with populations by expression of adhesion molecules and PD-L1 antibody. In this aggressive HCC model, due chemokines [9]. The unique positioning, phenotype and to the dispersing of tumor cell to liver through splenic function make LSECs an attractive candidate for organ- artery and portal vein, 20% of lung metastasis can be specific therapy. In our study, simvastatin exerted dual observed and mice usually die around day 20 if without effect via LSECs to deactivate aHSCs and recruit NKT treatment. Five days after inoculation, mice were treated cells, leading to stromal regression and tumor sup- with PBS, Sim NP-MAN (20  mg/kg, tail vein injection), pression. Since LSECs are able to take up molecules by PD-L1 antibody (100  μg, intraperitoneal injection), or scavenger receptors, nanoparticles with target ligand both for totally 5 times every other day (Fig.  6B). Both can be an excellent way to apply therapy specific to Sim NP-MAN and PD-L1 antibody showed a slight effect LSECs. Several studies verified the therapeutic effect in tumor suppression in HCC mice, whereas the combi- of nanoparticles targeting LSECs in various liver dis- nation of both treatments inhibited tumor progression eases [30–32]. Mannan is one of the most extensively significantly (Fig.  6C, D). Severe tumor development used ligand to target mannose receptor on LSECs. In and lung metastasis could be observed at day 19 after this study, PLGA with mannan ligand showed high liver Hepa1-6 inoculation, while much smaller tumor foci and accumulation and LSEC uptake rate, leading to satisfac- no lung metastasis was found in group with combination tory therapeutic effect in murine HCC model. treatment (Fig.  6E). The overall survival in the combina - NKT cells which share phenotypic and functional fea- tion group extended almost 2 times of the control group tures with NK cells represent a subpopulation of T lym- (Fig.  6F). As expected, upon the same treatment of Sim phocytes. The liver is the organ with the highest amount NP-MAN, the prognosis of tumors in the advanced stage of NKT cells compared with other T-cell subpopulations HCC model was much worse than intrahepatic inocula- [33]. Indeed, NKT cells constitute about 30% of all lym- tion HCC model. However, Sim NP-MAN combined phocytes in the liver, pointing to a critical role in liver with PD-L1 antibody still achieved a satisfactory thera- disease [34]. Recently, several pieces of evidence showed peutic effect in the late-stage HCC model. that NKT cells exert important functions in antitumor Yu  et al. Journal of Nanobiotechnology (2022) 20:9 Page 13 of 16 Fig. 6 Combination of simvastatin NP and PD-L1 antibody shows synergistic efficacy in advanced HCC suppression. A PD-L1 level of HCC tissues from various treatments and normal liver were measured by flow cytometry. B Mice HCC model establishment and treatment scheme. C Bioluminescence imaging of Hepa1-6 tumor bearing mice in various treatment groups (n = 5). D Tumor growth curves. E Representative liver tumor morphology. Scale bar stands for 1 cm. F Survival curves (n = 5). **p < 0.01, ***p < 0.001, n.s. = not significant immunity. It was revealed that hepatic NKT cell accumu- Conclusions lation precedes HCC inhibition [35]. CXCR6-dependent In summary, the therapeutic effect stems from the dual Recruitment of NKT cells and CD4 T cells to the liver effect of simvastatin on tumor microenvironment in exerts important functions in tumor surveillance to HCC, as well as the effective PLGA-PEG-MAN NP inhibit hepatocarcinogenesis [36]. Our findings demon - encapsulation and LSEC-targeted delivery. The LSEC- strated that the CXCL16 expression on LSECs induced by targeted delivery of simvastatin notably exerts effect on simvastatin attracted C XCR6 NKT cells in the liver to stromal microenvironment regression and NKT cell inhibit tumor growth, which is consistent with previous recruitment in fibrotic HCC mouse model. Besides, the studies. IFN-γ is instrumental for NKT initiated tumor lower dose than oral administration exerts higher effi - immunity. Interestingly, it was found that NKT cells from ciency with no toxicity, which would do great benefit to cancer patients produce reduced amounts of IFN-γ than HCC patients in clinic whose liver function is vulnerable healthy subjects [37], while simvastatin targeted delivery with limited therapeutic tolerability. Our findings reveal to LSECs activated NKT cells and induced IFN-γ produc- an immune-based therapeutic mechanism of simvas- tion, which contributed to anti-tumor effect. tatin and offer a broad application of this drug to HCC patients based on tumor microenvironment remodeling. Yu et al. Journal of Nanobiotechnology (2022) 20:9 Page 14 of 16 Fig. 7 Schematic representation of the Nano delivery of simvastatin targeting LSECs to remodel HCC microenvironment Abbreviations ***p < 0.001. Figure S3. Synthesis of the LSEC-targeting PLGA-PEG. (A) HCC: Hepatocellular carcinoma; LSECs: Liver sinusoidal endothelial cells; The synthesis process of PLGA-PEG-mannan. (B) H-NMR spectra of the HSCs: Hepatic stellate cells; aHSCs: Activated hepatic stellate cells; HMG-CoA: synthesized particles PLGA-PEG-mannan. Comparing the integrated area 3-Hydroxy-3-methylglutaryl CoA; KLF2: Kruppel-like factor; NO: Nitric oxide; of peak g (mannan protons) and peak a (protons in PEG), the ratio of CXCL: Chemokine (C-X-C motif ) ligand; CXCR: Chemokine (C-X-C motif ) mannose to PEG is approximately 4.5. Figure S4. Simvastatin NPs show no receptor; NKT: Natural killer T; PD-1: Programmed death receptor 1; PD-L1: Pro- obvious toxicity in mice. (A) HE staining of major organs from mice with grammed death-1-ligand-1; NPs: Nanoparticles; PFA: Paraformaldehyde; H&E: various treatments. Scale bar stands for 50 μm. (B) Body weight changes Hematoxylin ad eosin; IHC: Immunohistochemistry; IF: Immunofluorescence; during treatment. (C) Mice blood routine test and (D) Hepatorenal func- TUNEL: Terminal deoxynucleotidyltransferase-mediated dUTP nick end labe- tion test after treatment (n = 5). n.s. = not significant. Table S1. Antibody ling; RT-PCR: Real-time polymerase chain reaction; DMSO: Dimethylsulfoxide; list. Table S2. Gene Primer list for real-time PCR. Table S3. Cytokine Primer TEM: Transmission electron microscopy; EE: Encapsulation efficiency; LC: Load- list for real-time PCR. ing capacity; AST: Aspartate aminotransferase; ALT: Alanine aminotransferase; BUN: Urea nitrogen; CRE: Creatinine; αSMA: α-Smooth muscle actin; LYVE1: Lymphatic vessel endothelial hyaluronan receptor 1; eNOS: Endothelial nitric Acknowledgements oxide synthase; LPS: Lipopolysaccharide; RNA-seq: RNA-sequencing; PLGA: The authors thank Eshelman School of Pharmacy, University of North Carolina, Poly(lactic-co-glycolic acid); PEG: Polyethylene glycol; MAN: Mannan. for logistic support and laboratory facilities. The authors thank Shuguang Hospital affiliated to Shanghai University of Traditional Chinese Medicine for providing the HCC tissue samples and related clinical data. Supplementary Information The online version contains supplementary material available at https:// doi. Authors’ contributions org/ 10. 1186/ s12951- 021- 01205-8. ZY and LH conceived study and designed procedures. ZY, JG, YL, ZL performed experiments and acquired data. ZY and MW analyzed and interpreted data. YG acquired clinical samples. ZY wrote the draft, LH revised the manuscript. Additional file 1: Figure S1. Simvastatin shows no obvious cytotoxicity. LH and YG supervised this work. All authors read and approved the final Cytotoxicity study of SK-Hep1, LX2 and Huh7 cells treated with simvastatin manuscript. at different concentrations (n = 3). Figure S2. Simvastatin inhibits the activity of HSC via LSEC. (A) The expression of KLF2 and eNOS was quanti- Funding fied in SK-Hep1 cells treated with indicated concentration of simvastatin This work was supported by NIH Grant (CA198999), National Natural Science for 24 h (n = 3). (B) The expression of α-SMA and collagen1 was quantified Foundation of China (82074154, 81774240), China Scholarship Council in LPS-activated LX2 treated with the supernatant of SK-Hep1 upon dif- (201808310047), Shanghai Rising-Star Program (17QA1403900), Training Plan ferent doses of simvastatin treatment with or without L-NAME (n = 3). (C) of Outstanding Young Medical Talents from Shanghai Municipal Health Bureau The expression of CXCL16 was quantified in SK-Hep1 cells treated with (2017YQ021), Youth Tip-top Talent program in Shanghai, Youth Xinglin Scholar the indicated concentrations of simvastatin for 24 h (n = 3). **p < 0.01, from Shanghai University of Traditional Chinese Medicine, Siming Scholar Yu  et al. Journal of Nanobiotechnology (2022) 20:9 Page 15 of 16 from Shanghai Shuguang Hospital (SGXZ-201904), and the “Talents Cultivation 13. Abraldes JG, Villanueva C, Aracil C, Turnes J, Hernandez-Guerra M, Program” of Jilin University. Genesca J, et al. Addition of simvastatin to standard therapy for the pre- vention of variceal rebleeding does not reduce rebleeding but increases Availability of data and materials survival in patients with cirrhosis. Gastroenterology. 2016;150(5):1160-70 All data generated or analyzed during this study are included in this published e3. article and its additional files. 14. Pose E, Napoleone L, Amin A, Campion D, Jimenez C, Piano S, et al. Safety of two different doses of simvastatin plus rifaximin in decompensated cirrhosis (LIVERHOPE-SAFETY ): a randomised, double-blind, placebo- Declarations controlled, phase 2 trial. Lancet Gastroenterol Hepatol. 2020;5(1):31–41. 15. Singh S, Singh PP, Singh AG, Murad MH, Sanchez W. Statins are associated Ethics approval and consent to participate with a reduced risk of hepatocellular cancer: a systematic review and This study was conducted in accordance with the mandates of the Helsinki meta-analysis. Gastroenterology. 2013;144(2):323–32. Declaration and the guidelines of the Clinical Research Ethics Committee 16. Kim G, Jang SY, Nam CM, Kang ES. Statin use and the risk of hepatocel- of SUTCM. All the patients enrolled in this study signed written informed lular carcinoma in patients at high risk: a nationwide nested case-control consent, and this study was approved by SUTCM Clinical Research Ethics Com- study. J Hepatol. 2018;68(3):476–84. mittee (Approval No. 2018-629-58-01). All mice experiments were approved 17. Simon TG, Duberg AS, Aleman S, Hagstrom H, Nguyen LH, Khalili H, et al. by the Institutional Animal Care and Use Committee of the University of North Lipophilic statins and risk for hepatocellular carcinoma and death in Carolina at Chapel Hill. patients with chronic viral hepatitis: results from a nationwide swedish population. Ann Intern Med. 2019;171(5):318–27. Consent for publication 18. Marrone G, Russo L, Rosado E, Hide D, Garcia-Cardena G, Garcia-Pagan Not applicable. JC, et al. The transcription factor KLF2 mediates hepatic endothelial protection and paracrine endothelial-stellate cell deactivation induced Competing interests by statins. J Hepatol. 2013;58(1):98–103. Leaf Huang is a consultant of PDS Biotechnology, Samyang Biopharmaceutical 19. Marrone G, Maeso-Diaz R, Garcia-Cardena G, Abraldes JG, Garcia-Pagan Co, Stemirna and Beijing Inno Medicine. All other authors declare that they JC, Bosch J, et al. KLF2 exerts antifibrotic and vasoprotective effects in have no competing interests. cirrhotic rat livers: behind the molecular mechanisms of statins. Gut. 2015;64(9):1434–43. Author details 20. Guo S, Lin CM, Xu Z, Miao L, Wang Y, Huang L. Co-delivery of cisplatin and Division of Pharmacoengineering and Molecular Pharmaceutics, Eshelman rapamycin for enhanced anticancer therapy through synergistic effects School of Pharmacy, University of North Carolina, Chapel Hill, NC 27599, USA. and microenvironment modulation. ACS Nano. 2014;8(5):4996–5009. Department of Liver Disease, Shuguang Hospital, Affiliated to Shanghai 21. Yu Z, Guo J, Hu M, Gao Y, Huang L. Icaritin exacerbates mitophagy and University of Traditional Chinese Medicine, Shanghai 201203, China. School synergizes with doxorubicin to induce immunogenic cell death in hepa- of Pharmaceutical Sciences, Jilin University, Changchun 130021, China. tocellular carcinoma. ACS Nano. 2020;14(4):4816–28. 22. Liu Q, Wang X, Liu X, Kumar S, Gochman G, Ji Y, et al. Use of polymeric Received: 19 September 2021 Accepted: 12 December 2021 nanoparticle platform targeting the liver to induce treg-mediated antigen-specific immune tolerance in a pulmonary allergen sensitization model. ACS Nano. 2019;13(4):4778–94. 23. Gracia-Sancho J, Lavina B, Rodriguez-Vilarrupla A, Garcia-Caldero H, Bosch J, Garcia-Pagan JC. Enhanced vasoconstrictor prostanoid production by References sinusoidal endothelial cells increases portal perfusion pressure in cirrhotic 1. Siegel RL, Miller KD, Jemal A. Cancer statistics, 2019. CA Cancer J Clin. rat livers. J Hepatol. 2007;47(2):220–7. 2019;69(1):7–34. 24. Pertea M, Kim D, Pertea GM, Leek JT, Salzberg SL. Transcript-level expres- 2. El-Serag HB. Hepatocellular carcinoma. N Engl J Med. sion analysis of RNA-seq experiments with HISAT. StringTie and Ballgown 2011;365(12):1118–27. Nat Protoc. 2016;11(9):1650–67. 3. Hernandez-Gea V, Toffanin S, Friedman SL, Llovet JM. Role of the 25. Canton J, Neculai D, Grinstein S. Scavenger receptors in homeostasis and microenvironment in the pathogenesis and treatment of hepatocellular immunity. Nat Rev Immunol. 2013;13(9):621–34. carcinoma. Gastroenterology. 2013;144(3):512–27. 26. Makarova-Rusher OV, Medina-Echeverz J, Duffy AG, Greten TF. The 4. Llovet JM, Zucman-Rossi J, Pikarsky E, Sangro B, Schwartz M, Sherman M, yin and yang of evasion and immune activation in HCC. J Hepatol. et al. Hepatocellular carcinoma. Nat Rev Dis Primers. 2016;14(2):16018. 2015;62(6):1420–9. 5. Buonaguro L, Mauriello A, Cavalluzzo B, Petrizzo A, Tagliamonte M. Immu- 27. Keir ME, Butte MJ, Freeman GJ, Sharpe AH. PD-1 and its ligands in toler- notherapy in hepatocellular carcinoma. Ann Hepatol. 2019;18(2):291–7. ance and immunity. Annu Rev Immunol. 2008;26:677–704. 6. Couvelard A, Scoazec JY, Dauge MC, Bringuier AF, Potet F, Feldmann G. 28. Muller WA. Transendothelial migration: unifying principles from the Structural and functional differentiation of sinusoidal endothelial cells endothelial perspective. Immunol Rev. 2016;273(1):61–75. during liver organogenesis in humans. Blood. 1996;87(11):4568–80. 29. Zimmermann HW, Bruns T, Weston CJ, Curbishley SM, Liaskou E, Li KK, 7. Poisson J, Lemoinne S, Boulanger C, Durand F, Moreau R, Valla D, et al. et al. Bidirectional transendothelial migration of monocytes across Liver sinusoidal endothelial cells: physiology and role in liver diseases. J hepatic sinusoidal endothelium shapes monocyte differentiation and Hepatol. 2017;66(1):212–27. regulates the balance between immunity and tolerance in liver. Hepatol- 8. Thomas H. LSEC stretch promotes fibrosis during hepatic vascular con- ogy. 2016;63(1):233–46. gestion. Nat Rev Gastroenterol Hepatol. 2019;16(5):262–3. 30. Zuo Y, Ren S, Wang M, Liu B, Yang J, Kuai X, et al. Novel roles of liver sinu- 9. Shetty S, Lalor PF, Adams DH. Liver sinusoidal endothelial cells— soidal endothelial cell lectin in colon carcinoma cell adhesion, migration gatekeepers of hepatic immunity. Nat Rev Gastroenterol Hepatol. and in-vivo metastasis to the liver. Gut. 2013;62(8):1169–78. 2018;15(9):555–67. 31. Carambia A, Freund B, Schwinge D, Bruns OT, Salmen SC, Ittrich H, et al. 10. Tsochatzis EA, Bosch J. Statins in cirrhosis-ready for prime time. Hepatol- Nanoparticle-based autoantigen delivery to Treg-inducing liver sinusoi- ogy. 2017;66(3):697–9. dal endothelial cells enables control of autoimmunity in mice. J Hepatol. 11. Pose E, Trebicka J, Mookerjee RP, Angeli P, Gines P. Statins: old drugs as 2015;62(6):1349–56. new therapy for liver diseases? J Hepatol. 2019;70(1):194–202. 32. Yu X, Chen L, Liu J, Dai B, Xu G, Shen G, et al. Immune modulation of liver 12. Abraldes JG, Albillos A, Banares R, Turnes J, Gonzalez R, Garcia-Pagan JC, sinusoidal endothelial cells by melittin nanoparticles suppresses liver et al. Simvastatin lowers portal pressure in patients with cirrhosis and metastasis. Nat Commun. 2019;10(1):574. portal hypertension: a randomized controlled trial. Gastroenterology. 33. Mossanen JC, Tacke F. Role of lymphocytes in liver cancer. Oncoimmunol- 2009;136(5):1651–8. ogy. 2013;2(11):e26468. Yu et al. Journal of Nanobiotechnology (2022) 20:9 Page 16 of 16 34. Zimmermann HW, Tacke F. Modification of chemokine pathways and immune cell infiltration as a novel therapeutic approach in liver inflam- mation and fibrosis. Inflamm Allergy Drug Targets. 2011;10(6):509–36. 35. Ma C, Han M, Heinrich B, Fu Q, Zhang Q, Sandhu M, et al. Gut microbi- ome-mediated bile acid metabolism regulates liver cancer via NKT cells. Science. 2018. https:// doi. org/ 10. 1126/ scien ce. aan59 31. 36. Mossanen JC, Kohlhepp M, Wehr A, Krenkel O, Liepelt A, Roeth AA, et al. CXCR6 Inhibits hepatocarcinogenesis by promoting natural killer T- and CD4(+) T-cell-dependent control of senescence. Gastroenterology. 2019;156(6):1877-89 e4. 37. Tahir SM, Cheng O, Shaulov A, Koezuka Y, Bubley GJ, Wilson SB, et al. Loss of IFN-gamma production by invariant NK T cells in advanced cancer. J Immunol. 2001;167(7):4046–50. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in pub- lished maps and institutional affiliations. Re Read ady y to to submit y submit your our re researc search h ? Choose BMC and benefit fr ? Choose BMC and benefit from om: : fast, convenient online submission thorough peer review by experienced researchers in your field rapid publication on acceptance support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year At BMC, research is always in progress. Learn more biomedcentral.com/submissions

Journal

Journal of Nanobiotechnology – Springer Journals

Published: Jan 4, 2022

Keywords: Liver sinusoidal endothelial cells; Hepatocellular carcinoma; Simvastatin; Nanoparticles; Tumor microenvironment remodeling

There are no references for this article.