Tracking interactions between TAMs and CAFs mediated by arginase‐induced proline production during immune evasion of HCC

Synergistic changes between tumor‐associated macrophages (TAMs) and cancer‐associated fibroblasts (CAFs) aggravated immune evasion of hepatocellular carcinoma (HCC), however, the underlying molecular mechanisms remain elusive. Their continuous and dynamic interactions are subject to bioactive molecule changes. A real‐time and in situ monitoring method suitable for in vivo research of these processes would be indispensable but is scarce. In this study, a dual imaging strategy that tracing the TAMs and CAFs simultaneously was developed using a new arginase‐specific probe and established CAFs‐specific probe. The emerging roles of arginase in mediating CAFs activation in mice were explored. Results showed arginase up‐regulation in TAMs, followed by proline increase. Subsequently, proline produced by TAMs initiated the activation of CAFs. Through the JAK‐STAT signaling, CAFs up‐regulated the PD‐L1 and CTLA‐4, ultimately promoting immune evasion of HCC. This study revealed a new mechanism by which TAMs and CAFs collaborate in immune evasion, providing new targets for HCC immunotherapy.

The tumor microenvironment (TME) is the pathological environment in which immune evasion of HCC progression occurs. [4]The intercellular communication in TME is vital pathway for immune evasion. [5]To reveal the detailed molecular mechanisms of HCC immune evasion, it is prerequisite to explore the interplay of cells inside TME.Tumor-associated macrophages (TAMs) and cancerassociated fibroblasts (CAFs) are essential members of TME and exhibit various immunosuppressive functions. [6]mportantly, their mutual interactions in TME can induce recruitment and activation of each other, further enhancing immune evasion. [7]Although studies have reported the importance of CAFs and TAMs in tumor progression, the intercellular interaction mechanisms between them have not been fully revealed.Especially during the process of HCC immune evasion, the synergistic effect of CAFs and TAMs yet to be explored, and the detailed molecular mechanism remains elusive.
Arginase is a highly active enzyme that mostly expressed in TAMs. [8]Arginase consumes arginine in the TME, leading to T cell exhaustion, thus plays crucial roles in TAMs-mediated immune evasion. [9]Numerous studies have shown that arginase is overproduced in M2 macrophages and arginase is considered a specific marker of M2 macrophages.Overproduction of macrophage-specific arginase in TME might cause the elimination of arginine which result in the exhaustion of T cells. [10]In addition, arginase participates in regulating downstream pathways for the extracellular matrix (ECM) generation in the TME. [11]CAFs are responsible for ECM generation in the TME. [12]These findings suggest a hypothesis that TAMs might regulate the activity of CAFs through arginase, thereby affecting immune evasion of HCC.However, the underlying molecular mechanisms are mysterious.Cellular interplay between CAFs and TAMs is a continuous and dynamic process that is regulated by multiple active molecules.However, commonly used in vitro characterization methods such as western blot and polymerase chain reaction (PCR) are insufficient to explore the changes of active molecules in the physiological environment.The continuous and dynamic changes of arginase during the interaction between TAMs and CAFs are unrevealed.Fluorescence imaging technique would be a powerful tool for revealing the real-time and in situ changes of arginase in the interaction between TAMs and CAFs in vivo, but applicative imaging strategy is lacking. [13]herefore, in this study, we have developed a dual imaging strategy that tracing the dynamic changes of TAMs and CAFs in cells and HCC mice.To fulfil this goal, a new fluorescent probe TPEARG which emits bright fluorescence upon targeting arginase was constructed.This method enables in situ dynamic imaging of arginase changes in TAMs.By combining this imaging strategy with our established CAFs activation-specific imaging method, we have investigated the dynamic role of TAMs-derived arginase in promoting CAFs activation and facilitating immune evasion of HCC.The experimental results demonstrate that as the level of arginase in TAMs rises, the content of proline significantly increases.We have observed that the increased content of proline can significantly enhance the activation level of CAFs, which in turn affects the JAK-STAT pathway and enhances immune evasion of HCC.

Design strategy of TPEARG and optical properties of TPEARG
Rational selection of identifying group, regulation mechanism of fluorescence and appropriate fluorophore were made to fabric the fluorescent probe that fulfill our demand.8a] Then, according to the steric hindrance change after the probe bound to the active site domain of the arginase, aggregation-induced emission (AIE) principle was selected as the fluorescence regulation mechanism after recognition. [14]Tetraphenylethene (TPE) is an excellent AIE fluorophore with outstanding fluorescence imaging performance.TPE structure is compact with low steric hindrance, which is conducive to the probe entering the ectodomain of enzyme, so as to achieve high specificity and high selectivity imaging detection of enzymes. [15]In light of these considerations, we incorporated an arginine moiety to a TPE fluorophore via an amide linkage to construct the fluorescent probe TPEARG.The uncombined TPEARG emit weak fluorescence due to the bond rotation.Upon interaction with the arginase, the arginine could enter in the ectodomain of enzyme, resulting in the rotation restriction of TPE, which leading to about sixfold fluorescence increase at 450 nm (Figure 1A).
The detailed synthesis route and methods of TPEARG were described in the supporting information.The structure of TPEARG was characterized with 1 H NMR, 13 C NMR, and HRMS in the supporting information.The optical characterizations of TPEARG were examined in detail.Firstly, the AIE performances of TPEARG in various MeOH-H 2 O mixture were tested.As shown in Figure 1B, TPEARG was weakly fluorescent in MeOH because of the single molecular motion process of TPE.However, its fluorescence intensity enhanced greatly when the water fractions increased due to the formation of depositing aggregates.Moreover, the influence of viscosity on the AIE process was also measured.With the increasing of viscosity, fluorescence emission of TPEARG enhancement was observed resulted from the rotation restriction of TPE (Figure 1C).These data prove that we have successfully synthesized the TPEARG with outstanding AIE performances.
Nest, the recognition performances of TPEARG toward arginase were examined.As shown in the Figure 1D,E, upon interaction with sequential dosing of arginase, the fluorescence emission of TPEARG enhanced swiftly and gradually, which showed a linear manner (R 2 = 0.99).The limit of detection (LOD) was calculated as 0.04 U/mL.The selectivity of TPEARG to arginase and other proteases was further verified.As revealed in Figure 1F, TPEARG emitted low fluorescence when incubated with trypsase and other proteases than that with arginase.Moreover, widespread bioactive molecules like reactive oxygen species (ROS), metal ions and amino acids exhibit minimal interferences to the fluorescence of TPEARG (Figure 1G).The above experimental results show that TPEARG perform excellent sensitivity and selectivity toward arginase.Then, the fluorescence responses of TPEARG to arginase in the phosphate buffer saline (PBS) from pH 6.0 to pH 8.5 were further tested.Results indicated TPEARG's great photostability (Figure S1).The cytotoxicity and in vivo toxicity of TPEARG were measured.As shown in the Figure S2, TPEARG exhibited weak toxicity on cell proliferation and growth and organ development in mice.Hemolysis test in Figure S3A showed that TPEARG possessed minimal hemolytic activity (<3%) at tested concentrations (1 µM to 1 mM).To test the stability of TPEARG in blood, TPEARG was prepared into a solution at concentrations of 10 µM using mouse blood at room temperature.Then, fluorescence emission of TPEARG was measured after 1 h, 6 h, 12 h, and 24 h.The fluorescence intensity of TPEARG maintained stably low in the blood after prolonged incubation (Figure S3B).These above experimental results demonstrate that TPEARG has great potential for biological applications.

Characterizing the biding mode of TPEARG to arginase
Encouraged by the superior properties of TPEARG, molecular dynamics (MD) simulations were implemented to confirm the recognition mechanism.Arginase (PDB: 2CEV) was favorably docked into TPEARG (Figure 2A).Then a 10 ns MD simulation was performed based on Amber99SB force field using Gromacs 2018.4 software.Results exhibited that TPEARG molecules mainly bound in the cavity consisting of Asp126, Asn128, Ser133, Pro134, His139, Gly140, Ser176, Leu177, Asp178, Glu181, Thr240, and Glu271.The N atoms on the TPEARG form six hydrogen bonding interactions with Asp126, His139, Asp178, Thr240 and Glu271 in the protein, respectively (Figure 2B).And the benzene ring in the TPEARG molecule formed strong hydrophobic interactions with the hydrophobic amino acids (Asn128, Ser133, Pro134, Gly140, Ser176, Leu177, Glu181) around the pocket, which further enhanced the affinity between the TPEARG and arginase (Figure 2C,D).Root mean square deviation (RMSD) revealed that the average number of hydrogen bonds formed between arginase and TPEARG was 4.960 ± 0.732 (Figure S4).Accordingly, the interaction between arginase and TPEARG is stable, leading to an obverse fluorescence emission of TPEARG towards arginase.
Table 1 shows the molecular docking energy scores and predicted binding energies of TPEARG analysis with arginase.As can be seen from the table, TPEARG is expected to bind well to the arginase according to Autodock software, with a predicted binding energy of −6.73 kcal/mol.The data showed that the main contribution of binding energy came from intermolecular interactions, which contributed −10.31 kcal/mol.The intermolecular interaction mainly included van der Waals effect, hydrogen bonding, and the solvation energy.The sum of the above three interaction energies was −9.45 kcal/mol.The inhibition constant (Ki) was 11.66 uM.

Biological application of TPEARG at the cellular level
Afterward, due to the outstanding detection performances of TPEARG toward arginase, its biological applications were investigated in living cells in detail.6a,b] Thus, the RAW 264.7 cells were stimulated into M1 or M2 type using interferon-γ (IFN-γ)/lipopolysaccharide (LPS) or interleukin (IL-4), respectively.The RAW 264.7 cells treated with PBS (pH = 7.2) and BEC (an arginase inhibitor) were set as the control groups.Results exhibited that when M2 macrophages were incubated with BEC, the M2 features of the cells gradually disappeared, suggesting that arginase plays a key role in the maintenance of normal physiological function of M2 macrophages (Figure S5).Then, the cells were incubated with 1 µM TPEARG and imaged under laser   3A,B).In addition, when the arginase was inhibited by BEC, the cellular fluorescence intensity decreased significantly.These results suggest that TPEARG possess splendid potential in the detection of cellular arginase.To further explore the dynamic detection capability of TPEARG, we applied the TPEARG in M2 macrophages with sequential levels of arginase.M2 macrophages were pretreated with BEC for incremental time to obtain reduced expressions of arginase (Figure S6).Then, the TPEARG was used to observe the variation of cellular arginase.As displayed in Figure 3C,D, with the incubation time increased, the cellular fluorescence intensity decreased gradually.These results confirm that our established imaging method hold great potential in the dynamic detection of cellular arginase.

Observing the activation of CAFs induced by the arginase
8b,11] We supposed that the overproduction of arginase-induced proline in the TAMs would promote the activation of hepatic stellate cells (HSCs) to differentiate into CAFs, which further aggravated the immune evasion of HCC (Figure 4A).To verify our hypothesis, the proline concentrations in cell culture media with different levels of arginase were examined.As indicated in the Figure 4B, M2 macrophages secreted more proline than M1 macrophages, and the contents of proline decreased when the arginase was inhibited.Further, the proline concentrations depended on the expression of arginase (Figure 4C).These results suggest that the arginase profoundly impacted the produce of proline.Inspired by the above results, we further investigated whether the overproduced proline of TAMs could promote the activation of HSCs.LX-2 cells, a HSCs cell line, were incubated with dulbecco's modified eagle's medium (DMEM) (Control), DMEM from M2 macrophage, 10 µg/mL proline, 10 µg/mL aspartate, and 10 µg/mL cysteine, respectively.And the activation maker fibroblast activation protein (FAP) was imaged using our established probe Cy-FAP.In previous work, we developed Cy-FAP, a fluorescent imaging probe that specifically imaged the activation process of CAFs.When CAFs were incubated with Cy-FAP, the red fluorescence increased with the acceleration of CAFs activation. [16]Interestingly, compared to the cells treated with aspartate or cysteine, LX-2 cells incubated with proline exhibited nearly ninefold enhancement of fluorescence intensity (Figure 4D,E).Similarly, LX-2 cells co-cultured with medium extracted from M2 macrophages showed over sixfold intensity increase compared with the control, aspartate or cysteine group.When the arginase increases, the activation degree of CAFs was deeper as the fluorescence of Cy-FAP got brighter.These results indicate that elevated proline produced by the M2 macrophage is responsible for the activation of CAFs.
To further prove the contribution of arginase in inducing immune evasion which was mediated by CAFs activation, we measured the programmed death ligand 1 (PD-L1), IFNγ and IL-10 in each cell group.Experimental data revealed the up-regulation of PD-L1 and IFN-γ (anti-tumor cytokines) and down-regulation of IL-10 (tumor-promoting cytokines) in LX-2 cells that co-cultured with M2 macrophages was detected compared with the LX-2 cells or M2 macrophages (Figure S7).The above experimental results reveal that M2 macrophages can stimulate the differentiation of HSCs into CAFs by arginase-induced proline secretion, thereby promoting the immune evasion of HCC.

Activation of HSCs by arginase in mice
To further investigate the mechanism by which arginase promoted immune evasion of HCC cells, we imaged the changes of CAFs in mice with different levels of arginase expression.An orthotopic mouse model of HCC was constructed.The expression of arginase in mice was stimulated using LPS and inhibited by BEC.Flow cytometry was used to detect changes in the number of M2 macrophages in LPS-stimulated mice (Figure S8A).The results showed that the number of M2 macrophages in LPS-stimulated mice increased significantly compared with normal or HCC mice, which may be due to the increase of inflammatory burden in LPS-stimulated mice.Inside of the M2 macrophages with high expression of CD206, TPEARG emitted bright fluorescence, suggesting its excellent specificity toward M2 macrophages (Figure S8B).Subsequently, the mice were incubated with TPEARG and Cy-FAP to imaging the changes of arginase and CAFs.The results showed that the fluorescence intensity of arginase in tumor tissues of mice stimulated with LPS was significantly higher compared to the normal or HCC group.Nevertheless, the fluorescence intensity in mice decreased after inhibition of arginase with BEC (Figure 5A).TPEARG was adequate for monitoring changes of arginase in living mice.In addition, when the content of arginase in mouse tumors was increased, the CAFs exhibited subsequently enhanced activation levels.
When arginase was inhibited by the inhibitor, the amount of CAFs then decreased.
To further investigate the immune evasion caused by arginase change, we examined immune evasion related markers in each experimental group.The results exhibited that when the arginase increased, the number of CAFs also increased, which in turn led to the increase of immune evasion-promoting marker CTLA-4 (Figure 5B; Figure S9).Up-regulation of PD-L1, a widely recognized immunosuppressive factor, was also appeared in the HCC mice that arginase was high expressed (Figure 5C).Cytokines IL-10 that reported to contributed to the immunosuppressive effect exhibited up-regulation in the HCC mice that arginase was high expressed (Figure 5D).However, when the arginase was inhibited by the administration of BEC, these immunosuppressive factors were down-regulated, suggesting weaken immune evasion in HCC mice.Furthermore, pro-inflammatory effect like TNF-α and IFN-γ revealed down-regulation in the HCC mice that arginase was high expressed (Figure 5E,F).The above experimental data prove that TAMs in vivo stimulate the activation of CAFs via arginase-induced proline production, and then aggravate the immune evasion of HCC.

Signaling pathway underlying the arginase-mediated immune evasion
Ultimately, we determined genes and pathways associated with arginase-induced immune evasion.Mouse models of orthotopic HCC were constructed and treated with LPS as an arginase agonist, then the mRNA in mouse tumor tissues were tested and analyzed.As shown in Figure 6A, the expression of Oat, Scd2, Spp1, and Prelid2 genes related to arginine metabolism, lipid metabolism, and ECM production were elevated in the mice with high arginase expression, indicating elevated activation of HSCs in the tumor tissues of mice with high arginase expression.Further, genes with enhanced immune evasion functions, such as Map2k6 and Adipor2, were highly expressed when the arginase was highly expressed.Our previous work found JAK2-STAT3 signaling pathway through overproduced ROS.Here, we further analyzed the expression of signaling pathways associated with immune evasion in HCC mice.The IL-2-JAK2-STAT signaling pathway associated with oxidative stress was up-regulated when arginase was highly expressed (Figure 6B,C).From the above experimental results, we conclude a new mechanism of arginase promoting immune evasion from HCC: TAMs stimulate the activation of HSCs in the liver to disintegrate into CAFs through high expression of arginase, which in turn promote immune evasion if HCC by interfering with the JAK-STAT signaling pathway through the secretion of excessive ROS.

DISCUSSION
7a] However, the underlying mechanisms are unclear.Given that the interaction is based on the interaction between active molecules among the CAFs and TAMs, arginase, an immunoregulatory enzyme of TAMs, is involved in the ornithine cycle that plays a crucial role in the production of proline. [18]Meantime, proline serves as a major substrate for ECM synthesis by CAFs derived from HSCs. [19]hus, in this study, we set sights on exploring the unique roles of arginase during immune evasion.Tracing the change of arginase during the immune evasion process in the physiological environment would be beneficial to discovery the underlying mechanisms, but still challenging.Thus, we planned to develop an applicative detection strategy for real-time and in situ tracing of arginase in living cells and in vivo.
Currently, fluorescence imaging technique is a robust approach for the detection of biologically active molecules. [13,20]In this study, we designed and synthesized the first arginase-specific small molecule fluorescent probe, TPEARG, based on AIE principle.The microenvironmental changes after the enzyme identifies the probe molecule could be utilized by AIE principle.The advantage of this strategy lies in the ability to modulate the probe fluorescence independent of the enzyme-specific cut-off reaction.Consequently, we opted to employ the AIE principle in order to design arginase probes that rely on alterations in the microenvironment subsequent to probe identification of the enzyme's active site.TPEARG achieved good responses to arginase at different pH conditions.TPEARG recognition of arginase is not dependent on the deamination reaction of the guanidine group.After TPEARG targeted arginase via the arginine structure, the rotation of the tetraphenyl ethylene structure was restricted, resulting in luminescence.TPEARG can provide powerful assistance in the real-time and in-situ measurement of arginase change and distribution in cells and in vivo.
In this work, presented findings suggest that TAMs generate large amounts of proline by up-regulating the expression of arginase.We suspect this, in turn, may triggers the transformation of HSCs into CAFs within the HCC TME, thereby exacerbating the tumor cells' evasion of immune surveillance.Subsequently, experimental results demonstrated that proline could enhance the fluorescence intensity of Cy-FAP in CAFs, meaning an increased activation of CAFs.By combining the two imaging probes, the interaction between arginase expression in TAMs and the activation process of CAFs was observed for the first time.Proline secreted by TAMs could cause an increased activation of CAFs, while promoting CAFs to secrete more bioactive molecules with immunosuppressive functions.To this point, we find that TAMs produce large amounts of proline through high expression of arginase, which in turn stimulates CAFs to deepen the degree of immune evasion.
Furthermore, the detailed molecular mechanism of the process was explored.The JAK-STAT signaling pathway was found to be an important mechanism deeply involved in immune evasion from cancer cells in previous work. [16]Encouragingly, the results revealed that arginase medicated the down-regulation of JAK-STAT signaling pathway in CAFs via secretion of proline by TAMs.Then, increased secretion of PD-L1 and enhanced immune evasion of HCC were showed, which validated our previous conclusion.
In summary, by utilizing a new arginase-specific probe TPEARG and our established CAFs-specific probe Cy-FAP, we presented a dual imaging strategy for simultaneous observation of TAMs and CAFs.We applied this method in cells and in vivo to measure the changes of arginase and CAFs activation.Subsequently, using our established CAFs-specific imaging platform, we further explored the detailed molecular mechanism of arginase promoting immune evasion in HCC.Results showed that TAMs promoted the activation of CAFs by elevating proline levels through high expression of arginase, and CAFs deepened TAMs-induced immune evasion of HCC by secreting immunosuppressive active molecules.This approach we developed will provide a powerful tool for the study of arginase.Results has revealed a novel role played by arginase during immune evasion, which provided a new therapeutic target for HCC.

Materials and reagents
All chemical reagents were purchased from Shanghai Xianding biotechnology Co., Ltd.Analytical grade solvents were used without further purification.Mouse PD-L1 and TNFα ELISA kits were purchased from Absin Bioscience Inc. Raw264.7,LX-2 cells and all supplies for cell culture were purchased from Boster Biological Technology Co., Ltd.

Instruments
UV-2600 UV-vis spectrophotometer (Shimadzu, Japan) was utilized to analyze absorption spectra.The F-4600 HITACHI fluorescence spectrophotometer was used to detect the fluorescence spectra.Utilizing the Bruker maXis ultra-high resolution-TOF MS equipment, mass spectra were identified.Using Bruker NMR spectrometers, 1 H NMR spectra were measured at 400 MHz, while 13 C NMR spectra were recorded at 100 MHz.Using a Leica TCS SP8 confocal laser scanning microscope, fluorescence imaging of cells was performed.
Using the PerkinElmer IVIS Lumina III, in vivo NIR imaging of mice was measured.

Calculation of the LOD
The detection limit was calculated employing the following formula: Detection limit = 3σ/k, where σ is the standard deviation of blank measurement, k is the slope between the fluorescence intensity and the concentration of arginase.

4.6
Cell culture and cell type identification DMEM supplemented with 10% heat-inactivated fetal bovine serum and 1% penicillin and streptomycin was used to culture Raw 264.7 and LX-2 cells.Cells were grown at 37 • C in a humidified environment of 5% CO 2 .Medium replacements were made one or two days.LX-2 cells were collected using pancreatin and Raw 264.7 cells were collected by DMEM washing.
Protocol for stimulating M1 and M2 macrophages: RAW264.7 cells were incubated with 20 ng/mL INF-γ and 100 ng/mL LPS for 24 h to obtain M1 macrophages; 20 ng/mL IL-4 was utilized to treat RAW264.7 cells for 24 h to obtain M2 macrophages.M1 macrophages were found to be characterized by high expression of CD86 and low expression of CD206.In contrast, M2 macrophages were characterized by low expression of CD86 and high expression of CD206.To distinguish between these macrophage types, macrophages in each group were incubated with PE anti-mouse CD86 antibody and FITC anti-mouse CD206 antibody, and detected using flow cytometry.

Cytotoxicity assays
Raw 264.7 cells were seeded in two 96-well plates at a concentration of 1 × 10 5 and cultured for 12 h.In one 96-well plates, cells were cultured with 10 nM, 100 nM, 1 µM, 5 µM, 10 µM, 100 µM, 200 µM, 500 µM, and 1 mM TPEARG for 12 h.The medium volume of every well was 100 µL.Then, 10 µL CCK-8 solution was added to each well and incubated for another 1 h at 37 • C.After that, the absorbance at 450 nm was detected in a Triturus microplate reader.

In vivo toxicity assays
C57BL/6J mice (male, 4-6 weeks) were housed in a pathogen-free, temperature controlled facility with 12 h light/dark cycles.Animals received food and water ad libitum.
Mice were divided into two groups (6 mice per group).The mice in control group were given 200 µL PBS (pH = 7.4) solution via intravenous injection per day for 7 days.The mice in experimental group were given 200 µL 200 µM TPEARG PBS solution (pH = 7.4) via intravenous injection per day for 7 days.Their body weights were recorded every day.At the end of the experiment, major organs of heart, liver, spleen, lung and kidney in mice of each group were obtained and stained with hematoxylin and eosin (H&E) to assess toxicity.

4.9
Establishment of orthotopic mouse model of HCC C57BL/6J mice (male, 4-6 weeks) were subcutaneous injected with 1 × 10 7 HEPA 1-6 cells at the armpit of the right anterior limb.After 2-3 weeks, the mice were sacrificed by anesthetic overdose.The grown tumors were separated and cut into small pieces (1-2 mm 3 ).Then, these pieces of tumor were transplanted to the livers of other healthy mice through surgical operation.To preserve the activity of tumor pieces, surgical operation time should be completed within 30 min.

4.10
Fluorescence imaging of cells in vitro RAW 264.7 and LX-2 cells were seeded on the round coverslips in 6-well plates and then treated cells with different reagents.For arginase imaging, cells incubated with 100 µL 1 µM TPEARG PBS solution (1% DMF) for 30 min.Then the cells were washed with PBS 3 times, and imaged under the confocal laser scanning microscope with a blue channel (λ ex = 405 nm (intensity = 5%), λ em = 410-480 nm).For FAP imaging, cells were incubated with 100 µL 20 µM Cy-FAP PBS solution for 30 min.Then the cells were washed with PBS three times, and imaged under the confocal laser scanning microscope with a red channel (λ ex = 633 nm (intensity = 6%), λ em = 700-780 nm) for Cy-FAP.

Fluorescence imaging of mice tissue
The control group sham surgery group.Orthotopic mouse models of HCC were randomly divided into several groups (n = 5).Mice was injected with 50 µL 10 µM TPEARG PBS solution (1% DMF) and 50 µL 20 µM Cy-FAP PBS solution for 30 min via tail veins, respectively.Then liver tissues of each mouse were sliced using Leica CM1950.

F I G U R E 1
Optical characterization of TPEARG.(A) Structure and identification mechanism of TPEARG.The fluorescence intensity of TPEARG enhanced with the increasing of the solution (B) polarity and (C) viscosity.(D) When interacted with arginase, the TPEARG emitted fluorescence and the intensity increased with the concentrations of arginase.(E) Linearity relationship between the fluorescence intensity of TPEARG and concentrations of arginase.(F,G) Selectivity of TPEARG to several interferential factors. 1 µM TPEARG was introduced in these experiments.λ ex = 390 nm.The data are expressed as mean ± SD.

F I G U R E 2
Binding mode diagram of TPEARG and arginase (PDB: 2CEV).(A) Binding of TPEARG molecule on the hydrophobic surface of arginase, blue and orange indicate the hydrophilic and hydrophobic parts of the protein surface, respectively.(B) 2D binding mode of TPEARG molecule and arginase, green dashed line indicates hydrogen bonding interaction, purple solid line indicates ligand interaction, red gear-like indicates hydrophobic interaction.(C) Position of TPEARG molecule in the 3D structure of arginase.(D) 3D binding mode diagram of fluorescent probe molecule and arginase, green dashed line indicates hydrogen bonding interaction.TA B L E 1 Molecular docking scores.

F
I G U R E 3 Biological application of TPEARG in living cells.(A) 1 µM TPEARG was added in the RAW264.7 cells, M1 macrophages, M2 macrophages, and M2 macrophages pretreated with BEC (an arginase inhibitor, 200 µmol/L, 24 h), and imaged under the excitation of 400 nm.(B) Fluorescence intensity output of each group in (A).(C) Dynamic imaging detection of arginase using TPEARG in M2 macrophages pretreated with BEC (200 µmol/L) for incremental time.(D) Fluorescence intensity output of (C).The data are expressed as mean ± SD.F I G U R E 4 Activation of hepatic stellate cells (HSCs).(A) Hypothesis of the arginase in the TAM inducing the activation of HSCs to differentiate into cancer-associated fibroblasts (CAFs), which further aggravating the immune evasion.(B) Proline production by RAW 264.7 (Inactivated), M1 macrophage, M2 macrophage, and M2 macrophage incubated with BEC.(C) Measurement of proline secreted by M2 macrophage incubated with BEC for incremental time.(D) LX-2 cells, a HSCs cell line were incubated with DMEM (Control), DMEM from M2 macrophage, DMEM with 10 µg/mL proline, DMEM with 10 µg/mL aspartate, and 10 µg/mL cysteine, respectively.Then, the FAP in LX-2, a marker of CAFs, were imaged using 10 µM/mL Cy-FAP.(E) Fluorescence intensity output of (D).The data are expressed as mean ± SD. n = 3. Scale bar = 25 µm.

F I G U R E 5 F I G U R E 6
Activation of HSCs by arginase in mice.(A) Cy-FAP and TPEARG were applied in the normal mice (Control) and orthotopic mouse models of HCC treated with LPS (Arginase agonist) and BEC (Arginase inhibitor).(B) Western blot of CTLA4 in each mouse group.ELISA Kit test of the expressions of immune evasion markers (C) PD-L1, (D) TNF-α, (E) IL-10 and (F) IFN-γ.The data are expressed as mean ± SD. n = 3. Scale bar = 25 µm.Genes and pathways analysis in mice with arginase high expression.(A) Gene set enrichment analysis (GSEA) comparing gene-expression data from mouse tumor tissue and that treated with LPS.GSEA of (B) the Hallmark 'ECM receptor interaction' and (C) 'IL-2-JAK-STAT Signaling' pathway in mice tumor tissue.