Manipulation of TAMs functions to facilitate the immune therapy effects of
immune checkpoint antibodies
Yang Liu , Shuang Liang , Dandan Jiang , Tong Gao , Yuxiao Fang , Shunli Fu , Li Guan
Zipeng Zhang , Weiwei Mu , Qihui Chu , Yongjun Liu *
, Na Zhang *
Department of Pharmaceutics, Key Laboratory of Chemical Biology (Ministry of Education), School of Pharmaceutical Sciences, Cheeloo College of Medicine, Shandong
University, 44 Wenhua Xi Road, Jinan, Shandong Province 250012, China
ARTICLE INFO
Keywords:
Checkpoint blockade immune therapy
MMP2 responsive
Nanogels
NF-κB pathway
M2 TAMs re-polarization
ABSTRACT
Immune checkpoint antibodies have emerged as novel therapeutics, while many patients are refractory. Re￾searchers had identified tumor-associated macrophages (TAMs) is the pivotal factor involved in immune resis￾tance and that manipulation of TAMs functions would improve the immunotherapies effectively. NF-κB pathway
was one of the master regulators in TAMs manipulation. Inhibition of NF-κB pathway could achieve both re￾polarization M2 TAMs and downregulation the expression of programmed cell death protein 1 (PD-1) ligand 1
(PD-L1) on TAMs to improve the effect of immunotherapies. Here, IMD-0354, inhibitor of NF-κB pathway was
loaded in mannose modified lipid nanoparticles (M-IMD-LNP). Then, PD-1 antibody and M-IMD-LNP were co￾loaded in matrix metalloproteinase 2 (MMP2) responsive and tumor target nanogels (P/ML-NNG). P/ML-NNG
could co-deliver drugs to tumor site, disintegrated by MMP2 and release drugs to different targets. Evaluation
of PD-1 expression, inhibition of NF-κB pathway, expression of PD-L1 on M2 TAMs and M2 TAMs re-polarization
demonstrated that P/ML-NNG could block the PD-1/PD-L1 and NF-κB pathways simultaneously. Evaluation of
CD4 + T cells, CD8 + T cells, Tregs, cytokines and antitumor immunity confirmed that IMD-0354 could improve
the immunotherapies effectively. Those results provided forceful references for tumor immunetherapy.
1. Introduction
Recent years, immune checkpoint antibodies have emerged as novel
therapeutic modalities, and programmed cell death protein 1 antibodies
(PD-1 mAb) have been extensively researched and adhibition in clinic
[1]. But in fact, due to the complexity of tumor immune mechanisms,
many patients are refractory to these immunotherapies [2]. The com￾bination of PD-1 mAb with other immune activators to improve the ef￾fect of PD-1 mAb has received wide attention [3–5]. At present, many
immune activators have been combined with PD-1 mAb and satisfactory
effects could be achieved, such as interleukin-2 (IL-2) and tumor anti￾gens, etc. [6,7]. With the continuous exploration, researchers had
identified tumor-associated macrophages (TAMs) is the pivotal factor
involved in immune resistance and that manipulation of TAMs functions
would improve the immunotherapies efficacy of PD-1 mAb effectively
TAMs have been considered as major player in tumor microenvi￾ronment (TME)，accounting for over 50% of tumor immune cells and
involved in almost all stages of tumor progression [9,10]. TAMs were
divided into tumor promoting (M2 TAMs) and anti-tumorigenic (M1
TAMs) in TME because of their plasticity [11]. M2 TAMs were the core of
TME immunosuppression and the marker of poor prognosis [12]. In
tumor progression, NF-κB pathway was deemed to the moderator among
the dominating moderators in TAMs polarization [13,14]. Inhibition of
NF-κB pathway could achieve both re-polarization M2 TAMs and
downregulation the expression of PD-1 ligand 1 (PD-L1) on TAMs
[15,16]. Moreover, re-polarization of M2 TAMs could reduce the
secretion of transforming growth factor-β (TGF-β) and interleukin-10
(IL-10), and the downregulation of PD-L1 expression on TAMs could
block the PD-1/PD-L1 pathway and further to reduce immune resistance
and facilitate checkpoint blockade immune therapy [17,18]. Notably,
IMD-0354 was a specific inhibitor of NF-κB pathway. And some research
studies have certified that IMD-0354 could realize the effective inhibi￾tion of NF-κB pathway [19]. And IMD-0354 could polarize M2 TAMs to
M1 TAMs effectively, which was proved by our previous works [20].
Herein, IMD-0354 was selected to combine with PD-1 mAb for
* Corresponding authors.
E-mail address: [email protected] (N. Zhang).
Contents lists available at ScienceDirect
Journal of Controlled Release
journal homepage: www.elsevier.com/locate/jconrel
Received 3 March 2021; Received in revised form 2 June 2021; Accepted 6 July 2021
Journal of Controlled Release 336 (2021) 621–634
622
improving the efficacy of immunotherapies. PD-1 mAb could block PD-
1/PD-L1 pathway and then reactivate T cells [21]. IMD-0354 could
inhibit the NF-κB pathway, and then re-polarize M2 TAMs to M1 TAMs
and downregulate the expression of PD-L1 on TAMs to reduce immune
resistance and facilitate checkpoint blockade immune therapy. In order
to co-deliver PD-1 mAb and IMD-0354 to different targets, matrix met￾alloproteinase 2 (MMP2) responsive nanogels were selected. So as to
improve the effect of tumor therapy, the multiple targeting factors were
introduced to realize the precise targeting delivery. NGR (asparagines￾glycine-arginine peptide) as the first-order target was modified on the
surface of MMP2 responsive nanogels, which could mediate the active
target to the tumor site. In TME, MMP2 responsive nanogels could be
disintegrated by MMP2, PD-1 mAb and mannose-mediated target were
used to deliver PD-1 mAb and IMD-0354 to T cells and M2 TAMs. This
combination could be a promising therapeutic strategy for facilitating
the immune therapy effects of immune checkpoint antibodies. (See
Scheme 1.).
For this study, M-DOPE was synthesized though α-D-Mannopyr￾anosylphenyl (MPITC) and 1,2-dioleoyl-sn-glycero-3-phosphoethanol￾amine (DOPE). 4-arm-PEG-NGR was prepared by NGR-SH and 4-arm￾PEG-Mal. Then mannose mediated targeted IMD-0354 lipid nano￾particles (M-IMD-LNP) were prepared by combining nano-precipitation
method with high pressure homogenization method. Furthermore, PD-1
mAb and M-IMD-LNP co-loaded MMP2 responsive nanogels (P/ML￾NNG) was prepared though Michael addition reaction between MMP2
sensitive peptides, 4-arm-PEG-Mal and 4-arm-PEG-NGR. In the present
study, the physicochemical properties and MMP2 sensitivity of P/ML￾NNG were investigated. Tumor accumulation and differ-targeting de￾livery efficiency to T cells and M2 TAMs was demonstrated by cellular
uptake, co-culture and bio-distribution experiments. The expression of
PD-1 on T cells was evaluated by confocal laser scanning microscopy
(CLSM) images and flow cytometric analysis. And expression of PD-L1
on M2 TAMs was evaluated flow cytometric analysis after treatment
with different groups. And the inhibition of NF-κB pathway was inves￾tigated by CLSM images of tumor cyro-sections of different groups. M2
TAMs re-polarization was evaluated by flow cytometric analysis in vitro
and in vivo. In order to verify the reduction of immune resistance and
facilitation of checkpoint blockade immune therapy, immune status in
C57BL/6 mice were investigated. Meantime, for the antitumor efficacy,
it was evaluated on B16 bearing C57BL/6 mice. All the outcomes
demonstrated that MMP2 responsive PD-1 mAb and IMD-0354 co￾loaded nanogels could be a creative method in tumor immunotherapy.
2. Materials and methods
2.1. Materials
PD-1 mAb (Mw = 150kD) and PD-L1 mAb (Mw = 150kD) was pur￾chased from BioLegend (US). IMD-0354 was provided by Selleckchem
Co. Ltd. 4-arm-PEG-Mal (Mw = 10kD) was provided by Shanghai Yayi
Biotechnology Co. Ltd. (Shanghai, China). MMP2 sensitive peptide
(GPLGIAGQG, SH-pep-SH, Mw = 760.7) was provided by Leon Biolog￾ical Technology Co. Ltd. (Nanjing, China). NGR peptide (GCNGRCGC,
pep-SH, Mw = 670) was purchased from Shanghai Apeptide Co. Ltd.
(Shanghai, China). Soya lecithin (phosphatidylcholine accounts for
95%, pH 5.0–7.0) and 1, 2-dioleoyl-sn-glycero-3-phosphoethanolamine
(DOPE) were provided by Shanghai Taiwan Pharmaceutical Co. Ltd.
(Shanghai, China). α-D-Mannopyranosylpheny (MPITC，accounts for
98%) and Bull Serum Albumin (BSA) were provided by Sigma Aldrich
(US). All other materials and reagents were of analytical reagent grade.
2.2. Cell lines
Human umbilical vein endothelial cells (HUVEC), mouse melanoma
cells (B16) and mouse macrophages (RAW264.7) were provided by
Chinese Academy of Sciences (China). B16 were incubated in RPMI-
1640, RAW264.7 and HUVEC were incubated in DMEM added 10%
FBS. The cells were incubated at 37 ◦C in the presence of 5% CO2.
2.3. Animals
The female C57BL/6 (age: 6–8 weeks) were provided by SPF Beijing
Biotechnology Co., Ltd. Rat (weight: 200–300 g) were provided by
Experimental Animal Center of Shandong University. All experiments
were carried out according to the Animal Management Rules of the
Ministry of Health of the People’s Republic of China and the Animal
Experiment Ethics Review of Shandong University (Approval No.
18014).
2.4. Synthesized of 4-arm-PEG-NGR and M-DOPE
Briefly, NGR-SH (8.0 mg) and 4-arm-PEG-Mal (40.0 mg) were dis￾solved in 0.1 M phosphate buffer and reacted. The whole reaction was in
N2 atmosphere. The product was purified though dialysis (MWCO =
3500 Da). The purified product was lyophilized for testing and subse￾quent experiments. The product structure was verified by 1
H NMR(D2O,
300 MHz).
Briefly, MPITC (5.1 mg) were dispersed ultrasonically in 3 mL
methanol, and triethylamine (4.5 mL) were added. And DOPE (10.0 mg)
was dispersed ultrasonically in 200 μL dichloromethane. Dichloro￾methane mixed solution was dropwise added to methanol mixed solu￾tion and then reacted. Rotating evaporation to remove the solvent, and
then the products were re-dissolved in dichloromethane (5 mL) and
transferred to separating funnel. An equal volume of 0.1 M HCl was
added to the separator funnel, and the unreacted trimethylamine was
washed. Then saturated NaCl solution was added and the lower organic
phase was retained. Anhydrous Na2SO4 was added, and the water could
be removed after standing for 2 h. Then removing the anhydrous Na2SO4
through filtration and removing the solvent through rotating evapora￾tion. At last, the product (M-DOPE) was dried in vacuum drying oven.
The product weight was calculated by the bottle weight difference, the
product structure was verified by 1
H NMR(CD3 OD, 300 MHz).
2.5. Preparation of M-IMD-LNP, ML-NNG and P/ML-NNG
First, lipid nanoparticles (IMD-LNP) without targeting effects were
prepared. The IMD-LNP was prepared by combining nano-precipitation
method with high pressure homogenization method. Briefly, DOPE and
soya lecithin (molar ratio of DOPE: soya lecithin was 1:14) were dis￾solved in 6 mL 0.1 M phosphate buffer (with 1.0% Tween-80). IMD-0354
(molar ratio of IMD-0354: the total lipid was 1:4) was dissolved in 0.6
mL methanol. Under stirring and ice bath, the methanol phase was
injected into phosphate buffer mixture solution. Then the mixture so￾lution was homogenized at 14.0 kpsi for 20 times by high pressure ho￾mogenizer. Finally, the organic solvent was removed by stirring for 2 h.
M-DOPE was used to prepare M-IMD-LNP instead of DOPE. Free IMD-
0354 was removed by filtering through 0.22 μm membrane filtration.
The MMP2 sensitive nanogels were prepared by chemical cross￾linking method. For the preparation of ML-NNG, solution 1, M-IMD-LNP
(0.5 mL), 4-arm-PEG-NGR (3.5 mg) and 4-arm-PEG-Mal (7.5 mg) were
dispersed in 0.3 mL 0.1 M phosphate buffer. Solution 2, the MMP2
sensitive peptide (1.2 mg) were dispersed in 0.2 mL 0.1 M phosphate
buffer. At 1000 r min− 1 and 37 ◦C water bath, the solution 2 was
dropwise added to solution 1 and the reaction was reacted for 0.5 h.
Then the reaction was reacted at 500 r min− 1 and 37 ◦C water bath for 1
h. The whole reaction was in N2 atmosphere. P/ML-NNG was obtained
when PD-1 mAb (1.0 mg) dispersed with4-arm-PEG-NGR (3.5 mg), 4-
arm-PEG-Mal (7.5 mg) and M-IMD-LNP (0.5 mL) in 0.3 mL 0.1 M
phosphate buffer. Free PD-1 mAb was removed though dialysis (MWCO
= 300 kD).
Y. Liu et al.
Journal of Controlled Release 336 (2021) 621–634
623
2.6. Characterizations of M-IMD-LNP, ML-NNG and P/ML-NNG
The average diameters, polymer dispersity index (PDI) and zeta
potentials of M-IMD-LNP, ML-NNG and P/ML-NNG were measured by
dynamic light scattering (DLS) (NanoZS90, Malvern Instrument U.K).
And the morphologies were expressed by transmission electron micro￾scopy (TEM) (Hitachi, Japan). Drug loading (DL) of IMD-0354 and PD-1
mAb were measured by HPLC (SPD-10Avp Shimadzu pump, LC-10Avp
Shimadzu UV–vis Detector) and enhanced BCA kit (Beyotime Biotech￾nology, China), respectively. DL were calculated by following
formulations:
DL% = Wdrug/( Wdrug + Wcarrier materials)
× 100%
W drug means the weight of IMD-0354 or PD-1 mAb encapsulated in
carrier materials; W carrier materials means the total weight of all carrier
materials.
In addition, circular dichroism was used to analyze the structure and
biological activity of PD-1 mAb. The PD-1 mAb was incubated by
simulating the steps in the preparation of P/ML-NNG. The PD-1 mAb
stored at 4 ◦C was used as control.
Physical stability of P/ML-NNG was evaluated at 4 ◦C. The changed
zeta potentials and average diameters were measured by DLS over a
period of 7 days (n = 3).
2.7. Enzymatic digestion assay
The enzymatic cleavage of MMP2 sensitive peptide was evaluated by
enzymatic digestion method using the collagenase IV. Briefly, MMP2
sensitive peptide (1.5 mg mL − 1
) was incubated with the collagenase IV
in pH 7.4 phosphate buffer for 24 h at 37 ◦C. Final concentrations (0, 50,
100，200, 300, 400 and 500 μg mL − 1
) of collagenase IV were used for
each sample, respectively. And then enzymatic cleavage time was
explored. MMP2 sensitive peptide (1.5 mg mL − 1
) was incubated with
collagenase IV (400 μg mL− 1
) at 37 ◦C for 0, 0.5, 1, 2, 3 and 4 h. The
digestion fragments were identified by HPLC.
In order to demonstrate cleavage specificity of MMP2, other enzymes
commonly found in body fluids (such as catalase, lipase) were incubated
with MMP2 sensitive peptide (1.5 mg mL − 1
) for 24 h and at 37 ◦C and
then measured by HPLC.
2.8. In vitro disintegration behaviors of P/ML-NNG
To evaluate the disintegration behaviors of P/ML-NNG, particle size
and morphology were examined by DLS and TEM. Briefly, P/ML-NNG
was incubated with or without 400 μg mL − 1 collagenase IV at 37 ◦C
for 3 h. Then the particle size and morphology of the two samples were
examined.
2.9. In vitro release of IMD-0354 and PD-1 mAb
The dynamic dialysis method was used to assess the release behavior
of IMD-0354 with or without 400 μg mL − 1 collagenase IV in vitro.
Briefly, 0.5 mL of each sample (IMD-0354, M-IMD-LNP and P/ML-NNG;
with or without 400 μg mL− 1 collagenase IV) was placed in pre-swelled
dialysis bags (MWCO = 8–14 kD) and incubated in 10 mL phosphate
buffer (pH 7.4, with 1% Tween-80, with or without 400 μg mL− 1
collagenase IV) as release medium at 37 ◦C. At predetermined time
point, 0.5 mL release medium was withdrawn and replaced with 0.5 mL
fresh medium. Sample was performed in three times. The released IMD-
0354 was determined by HPLC.
The dynamic dialysis method was also used to assess the release
behavior of PD-1 mAb with or without 400 μg mL− 1 collagenase IV in
vitro. Briefly, 0.5 mL of each sample (PD-1 mAb and P/ML-NNG; with or
without 400 μg mL− 1 collagenase IV) was placed in dialysis pre-swelled
bags (MWCO = 1000kD) and incubated in 10 mL phosphate buffer (pH
7.4, with or without 400 μg mL− 1 collagenase IV) as release medium at
37 ◦C. At predetermined time point, 0.5 mL release medium was with￾drawn and replaced with 0.5 mL fresh medium. Sample was performed
in three times. The released PD-1 mAb was determined by the enhanced
BCA kit.
2.10. Cytotoxicity assay
The cytotoxicity of P/ML-NNG and free IMD-0354 were determined
by MTT assay. Briefly, B16 cells were seeded in 96-well plates at 5 × 10 3
cells per well and grown for 12 h. Then samples were added and incu￾bated for 48 h. After 48 h, MTT were added and further incubated for 4
h. Finally, DMSO was added after removing supernatant in each well.
Cell viability was measured at 570 nm by microplate reader. (Multi￾skanTM Go, Thermo Fisher Scientific, US).
2.11. Hemolysis assay
Fresh blood samples were obtained from rat (Experimental Animal
Center of Shandong University). Briefly, the red blood cells (RBCs) were
washed and collected by centrifugation. The 2% RBCs suspension was
obtained by re-suspension in normal saline (NS). The 2.5 mL samples
(NS was set as negative control group; P/ML-NNG groups was designed
with a series concentration gradient, and the concentration of IMD-0354
was 10, 20, 30, 40, 50 μg ml − 1
; distilled water was set as positive control
group) were added into 2.5 mL 2% RBCs suspension and incubated for 3
h at 37 ◦C. Hemolysis of samples were observed and photographed.
Finally, the supernatant was removed by centrifugation, and the
absorbance of haemoglobin in supernatant was measured at 576 nm by
UV–Vis spectrophotometer (TU-1810, Persee Corporation, China) to
calculate the hemolysis rate. Hemolysis rate was calculated using
following formulations:
Hemolysis rate% = (
ODP/ML− NNG–ODNS)/(ODdistilled water − ODNS) × 100%
OD P/ML-NNG means the absorbance of haemoglobin in P/ML-NNG
groups; OD NS means the absorbance of haemoglobin in the NS group;
OD distilled water means the absorbance of haemoglobin in the distilled
water group.
2.12. Evaluation of active targeting and tumor accumulation
Cellular uptake experiments on HUVEC were carried out to evaluate
active targeting characterization of P/ML-NNG. P/ML-NNG was pre￾pared by using Nile Red instead of IMD-0354. Meanwhile, P/ML-NG
(without targeting effects mediated by NGR) was prepared used same
method as the negative control group, free Nile Red was used as positive
control group. Briefly, HUVEC were seeded in 12-well plates at 2.0 × 10 5 cells per well and grown for 12 h. Then media were replaced by fresh
media containing free Nile Red, P/ML-NG and P/ML-NNG and every
group was incubated for 1 h and 4 h, respectively. In addition, HUVEC
were pre-incubated with free NGR (1 mg mL − 1
) for 1 h to incubation
with P/ML-NNG for competitive inhibition experiments. Finally, the
cellular uptake ability was examed by microplate reader (Synergy HTX
Multi-Mode Microplate Reader, BioTek,USA). In order to quantify the
cellular uptake ability of P/ML-NNG, HUVEC was treated as the above
method. And then cells were digested, collected and washed. Quanti￾tative analysis was performed by CytoFLEX S (Beckman Coulter, USA).
Tumor accumulation of P/ML-NNG in vivo was evaluated by Real￾time fluorescence imaging system (IVIS) spectrum (Caliper Perki￾nElmer, Waltham, MA, USA). In this experiment, IR780 was used to
prepare P/ML-NG and P/ML-NNG. B16 bearing C57BL/6 mice were
selected for this experiment, and 1 × 10 6 B16 were injected into the
right axilla intradermally of the mice. When the tumor grows to a certain
size, mice were intravenous injections with IR780, P/ML-NG and P/ML￾NNG, respectively. At presupposed time point, mice were anesthetized
Y. Liu et al.
Journal of Controlled Release 336 (2021) 621–634
624
and placed into the instrument to image. After administration for 24 h,
the tumor, heart, liver, spleen, lung and kidney were dissected for ex vivo
imaging.
2.13. Evaluation of RAW264.7 induced through IL-4
To evaluate that IL-4 could induced RAW264.7 to M2 TAMs, CLSM
and flow cytometric assay were used. RAW264.7 were seeded in laser
confocal dishes at 5.0 × 10 5 per dish and grown for 12 H. IL-4 (15 ng
mL− 1
) was added and then incubated for another 12 h to stimulate TAMs
polarized into M2 TAMs. All TAMs were marketed with F4/80. M2 TAMs
were labeled by F4/80 and CD206 simultaneously. Finally, CLSM and
flow cytometric assay were used to imaging and quantitative analysis,
differently.
2.14. Evaluation of target delivery to M2 TAMs mediated by mannose
To evaluate cellular uptake ability on M2 TAMs, IL-4 induced
RAW264.7 was used to replace the M2 TAMs and Nile Red was used to
prepare IMD-LNP, M-IMD-LNP and P/ML-NNG. RAW264.7 were seeded
in laser confocal dishes at 5.0 × 10 5 per dish and grown for 12 H. IL-4
(15 ng mL− 1
) was added and then incubated for another 12 h to stim￾ulate TAMs polarized into M2 TAMs. Then media were removed, fresh
media containing Nile Red, IMD-LNP, M-IMD-LNP and P/ML-NNG
(preincubated with 400 μg mL − 1 Collagenase IV for 3 h) were added
and then incubated for 1 h or 4 h. In addition, RAW264.7 were pre￾incubated with free MPITC (1 mg mL − 1
) for 1 h to incubation with
M-IMD-LNP for competitive inhibition experiments. All TAMs were
marketed with F4/80. M2 TAMs were labeled by F4/80 and CD206
simultaneously. Finally, CLSM was used to imaging. In order to quantify
the cellular uptake ability, the arithmetic mean intensity was recorded
of the CLSM images.
2.15. Evaluation of co-delivery to different targets
Co-culture experiment was carried out to evaluate the co-delivery of
different targets in vitro. For this experiment, T cells from mouse spleens
and RAW264.7 induced by IL-4 were co-cultured. IL-4 induced
RAW264.7 was used to replace the M2 TAMs. Nile Red and FITC labeled
PD-1 mAb were used to prepare P/ML-NNG. Briefly, RAW264.7 were
seeded in laser confocal dishes at 5.0 × 10 5 per dishes and grown for 12
H. IL-4 (15 ng mL− 1
) was added and then incubated for another 12 h to
stimulate TAMs polarized into M2 TAMs. Then media were removed,
fresh media containing T cells (1.0 × 10 5 cells of every well) and P/ML￾NNG (preincubated with 400 μg mL − 1 Collagenase IV for 3 h) were
added into the above laser confocal dishes and incubated for 4 h. At the
predesigned time, T cells were sucked out and collected. M2 TAMs were
labeled by F4/80 and CD206 simultaneously. Then washed, imaged by
CLSM of those two kinds of cells, respectively.
2.16. Evaluation of PD-1 expression
The level of PD-1 expression on T cells and M2 TAMs were deter￾mined by using the immunofluorescence staining. T cells from mouse
spleens, and IL-4 induced RAW264.7 was used to replace the M2 TAMs.
Briefly, RAW264.7 were seeded in 12-well plates at 2.0 × 10 5 per well
and grown for 12 h. Then IL-4 (15 ng mL − 1
) was added and incubated
for another 12 h to stimulate TAMs polarized into M2 TAMs. M2 TAMs
were labeled by F4/80 and CD206 simultaneously. Then the labeled
macrophages were collected in flow tube. Meanwhile, the spleens of B16
bearing C57BL/6 mice were selected to extract T cells. When tumors
grew to a certain size, the spleen was dissected and removed. Then,
spleens were ground into single-cell suspension and lymphocytes were
isolated and labeled with CD3 antibody. Finally, CD3+ T cells were
selected by MofloAstrios EQ (Beckman Coulter, USA). Next, CD3+ T cells
from mouse spleens (2.0 × 10 5 cells) were collected in flow tube.
Finally, the collected RAW264.7 and T cells were labeled by PD-1
antibody, washed with PBS. Fluorescence intensities were measured
by CytoFLEX S. Subsequently, the expression of PD-1 was characterized
by CLSM. RAW264.7 were seeded in laser confocal dishes at 5 × 10 5
cells per dish and grown for 12 h. The following processing method was
the same as above. Finally, RAW264.7 was in laser confocal dishes and T
cells were placed in adhesion slides. The images were observed by CLSM
of RAW264.7 and T cells.
2.17. Evaluation the expression of PD-L1 on M2 TAMs
The PD-L1 expression on M2 TAMs was evaluated by flow cytometric
assay, and IL-4 induced RAW264.7 was used to replace the M2 TAMs.
Briefly, RAW264.7 were seeded in 12-well plates at 5.0 × 10 5 per well
and grown for 12 h. Then IL-4 (15 ng mL − 1
) was added and then
incubated for another 12 h to stimulate TAMs polarized into M2 TAMs.
Fresh media containing P/ML-NNG (preincubated with 400 μg mL− 1
Collagenase IV for 3 h), M-IMD-LNP, IMD-LNP and IMD-0354 solution
(the concentration of IMD-0354 was 1.0 μg mL − 1
) were added and
cultured for 24 h. After incubating, collected the cells, incubated with
PD-L1 antibody and then with goat anti-mouse IgG/Alexa Fluor 488
antibody. Finally, washed and measured by CytoFLEX S.
2.18. Evaluation of M2 TAMs re-polarization in vitro
CytoFLEX S was used to analyze the re-polarization of M2 TAMs in
vitro. Hence, RAW264.7 was used. RAW264.7 were seeded into 12-well
plates at 5 × 10 5 and then incubated overnight at 37 ◦C. IL-4 (15 ng mL − 1
) was added and cultured for another 12 h to stimulate TAMs polar￾ized into M2 TAMs. Fresh media containing P/ML-NNG (preincubated
with 400 μg mL − 1 Collagenase IV for 3 h), M-IMD-LNP, IMD-LNP and
IMD-0354 solution (IMD-0354 was 1.0 μg mL− 1
) were added and incu￾bated for 24 h. After that, the cells were marked with F4/80, CD206 and
CD86. Then cells were digested, collected and washed and measured by
CytoFLEX S. F4/80 + CD206 + were considered as M2 TAMs, F4/80 +
CD86 + were considered as M1 TAMs.
2.19. In vivo antitumor efficacy and the evaluation of P/ML-NNG
facilitated antitumor immunity
For the antitumor efficacy, it was evaluated on B16 bearing C57BL/6
mice and 1 × 10 6 B16 cells were injected into the right axilla intra￾dermally of the mice. When the tumor grew to a certain volume, the
mice were divided into following eight groups randomly (n = 6): (1)
saline; (2) Blank nanogels (Blank NG); (3) PD-1 mAb; (4) IMD-0354; (5)
PD-1 mAb and IMD-0354 mixed solution (PD + IMD); (6) PD-1 mAb
loaded nanogels (P-NNG); (7) M-IMD-LNP loaded nanogels (ML-NNG);
(8) P/ML-NNG. The dose of PD-1 mAb and IMD-0354 was 5.0 mg kg − 1
and 1.14 mg kg − 1
, respectively. The mice were intravenous injections
with different formulations every 3 days. The mice were treated for 12
days. The mice used for survival period studies were also treated as
above. The body weight and tumor volume were recorded every two
days. The tumor volume was measured with a Vernier caliper every
other day by using the formula (length × width 2
)/2, where length is the
longest dimension and width is the widest dimension. On 12th day, the
mice were sacrificed and tumors from each group were surgically
excised, rinsed with normal saline, wiped, weighed and photographed.
The in vivo re-polarization ability was evaluated by CytoFLEX S.
Tumors were collected, lapped and filtered to obtain the single cell
suspensions. Then single cell suspensions were centrifuged to obtained
cells, marked antibodies as used in vitro and analyzed by CytoFLEX S. T
cells and Tregs were also evaluated by CytoFLEX S and cytokines were
evaluated by ELISA assay. Briefly, tumor tissues were collected, lapped
and filtered to obtain the single cell suspensions. Lymphocytes were
obtained by using 40% Percoll solution and then marked with different
anti-mouse immunofluorescent antibody. For T cells, CD3, CD4, CD8
Y. Liu et al.
Journal of Controlled Release 336 (2021) 621–634
625
was used to mark. The cells with CD3 + CD4 + were recognized to be CD4 + T cells, the cells with CD3+ CD8+ were recognized to be CD8 + T cells.
For Tregs, CD4 and Foxp3 was used to mark. The cells with CD4 + Foxp3 + were recognized to be Tregs. For each group, the orbital bloods were
taken from the mice. IL-10, TGF-β, IL-12, TNF-α and IFN-γ was measured
by ELISA kits (Dakewe, Shenzhen, China) according to the operating
instructions, respectively.
2.20. Evaluation the inhibition of NF-κB pathway
To evaluate the inhibition of NF-κB pathway, tumor tissue sections
after treatment with different formulations were fixed with 4% form￾aldehyde. Then, tumors were stained with NF-κB p65 and imaged.
2.21. Evaluation the expression of PD-L1 on M2 TAMs in vivo
To investigate the inhibition of the expression of PD-L1 on M2 TAMs,
tumor tissue sections after treatment with different formulations were
fixed with 4% formaldehyde. Then, tumors were stained with PD-L1 and
imaged.
2.22. Histological evaluation
To investigate the tumor proliferation, apoptosis and toxicity to
major organs, tumors and major organs were fixed with 4% formalde￾hyde. Then, tumors were stained with Ki67, H&E and TUNEL respec￾tively to analyze its proliferation and apoptosis. The major organs were
stained with H&E to analyze its toxicity.
2.23. Hematological analysis
For each group, the orbital bloods were taken from the mice, and the
hematological data were investigated.
2.24. Statistical analysis
The statistical differences between different groups were calculated
by the Student’s t-test. p < 0.05 was considered statistically significant.
All values are expressed as means ± SD.
3. Results and discussion
3.1. Synthesized of 4-arm-PEG-NGR and M-DOPE
NGR is the specific ligand of CD13 which was overexpressed in tumor
vascular epithelial cells [22]. In order to active targeting to tumor site,
targeting ligand NGR peptide was jointed to 4-arm-PEG to form 4-arm￾PEG-NGR by the reaction through -SH and -Mal. In 1
H NMR spectra of 4-
arm-PEG-NGR, the characteristic peak of NGR at 1.4–2.0 ppm was found
(Fig. S1). Mannose could target to M2 TAMs by recognizing and binding
with CD206 receptor which was high expressed on M2 TAMs [23]. M￾DOPE was synthesized through the reaction of MPITC and DOPE. In 1
NMR spectra of M-DOPE, the chemical shift of 7.1–7.3 ppm was the
characteristic peak of phenyl group in MPITC and 4.5 ppm was the
characteristic peak of carbon‑carbon double bond (-C=C-) in DOPE
(Fig. S2). The above results proved the successful synthesis of 4-arm￾PEG-NGR and M-DOPE.
3.2. Preparation and characterizations of M-IMD-LNP, ML-NNG and P/
ML-NNG
Firstly, M-IMD-LNP with small size was prepared by combining
nano-precipitation method with high pressure homogenization method.
The average diameters and zeta potentials of M-IMD-LNP were 32.52 ±
2.61 nm and − 0.147 ± 0.048 mV, and which were measured by DLS.
Meanwhile, morphologies of M-IMD-LNP were spherical particles and
were evaluated by TEM (Fig. 1a). Secondly, M-IMD-LNP loaded MMP2
sensitive nanogels (ML-NNG) was developed though Michael addition
reaction between MMP2 sensitive peptides, 4-arm-PEG-Mal and 4-arm￾PEG-NGR. The average diameters, zeta potentials were 97.06 ± 2.89
nm, − 1.95 ± 0.74 mV, and they were spherical particles (Fig. 1b). The
change in particle size between M-IMD-LNP and ML-NNG proved the
successfully formation of M-IMD-LNP loaded nanogels, and the TEM
image also showed that the small nanoparticles were successfully loaded
in nanogels. The P/ML-NNG was prepared through the same method
with ML-NNG. And the average diameters, zeta potentials were 118.67
± 1.45 nm, − 0.91 ± 0.23 mV, they also were spherical particles
(Fig. 1c). Compared with the TEM image of ML-NNG, the small nano￾particles in the nanogel were not obvious in the TEM image of P/ML￾NNG, which we speculated was due to the presence of PD-1 mAb. In
addition, P/ML-NNG was stable at 4 ◦C for 7 days (Fig. S5). (See Scheme
1.)
As shown in Fig. S4, the circular dichroism spectrum of incubated
PD-1 mAb was similar to the control group, indicating that the structure
and biological activity of PD-1 mAb did not change significantly (p >
0.05), indicated that the steps in the whole system might not lead to
denature of PD-1 mAb.
3.3. MMP2 sensitive evaluation of P/ML-NNG and responsively release of
IMD-0354 and PD-1 mAb
In order to achieve the co-delivery of PD-1 mAb and IMD-0354 to
different targets, P/ML-NNG was developed and MMP2 sensitive peptide
was introduced as “linker” in nanogels due to the MMP2 was overex￾pressed in TEM [24]. We checked the literatures and research on MMP2
sensitive nanoparticles. In the literature, the concentration of the model
enzyme selected in the MMP2 sensitive drug release experiment was 1
mg mL − 1 [25]. Hence, we have setted a series of enzyme concentrations
(below 1 mg mL − 1
) to investigate the sensitivity of MMP2 sensitive
peptides. Firstly, the sensitivity of MMP2 sensitive peptide (1.5 mg
mL− 1
) was evaluated by incubated with collagenase IV at different
concentration gradients (0, 50, 100，200, 300, 400 and 500 μg mL− 1
and different incubation time (0, 0.5, 1, 2, 3 and 4 h) and then measured
by HPLC (Fig. 1d). As shown in Fig. 1d, the collagenase IV had no peak in
the detection range, while the other samples all had absorption peak,
indicated that the collagenase IV had no influence on the detection of
MMP2 sensitive peptide under the chromatographic conditions in this
experiment. From the absorption peaks of MMP2 sensitive peptide and
its disintegration products, we could know that MMP2 sensitive peptide
was fully disintegrated by collagenase IV at 400 μg mL − 1 in 3 h.
Therefore, the subsequent experiments were conducted with collage￾nase IV at 400 μg mL− 1 in 3 h. As shown in Fig. S3, the results showed
that MMP2 sensitive peptide could not be disintegrated by other en￾zymes, which proved that MMP2 sensitive peptide selected in this study
could be disintegrated specificity of MMP2.
The in vitro disintegration behaviors of P/ML-NNG were evaluated by
DLS (Fig. 1f) and TEM (Fig. 1g). After P/ML-NNG was incubated with or
without 400 μg mL − 1 collagenase IV at 37 ◦C for 3 h, the hydrodynamic
diameters were 66.53 ± 2.76 nm and 122.46 ± 4.13 nm, respectively.
And as shown in TEM images, P/ML-NNG was fully disintegrated. These
results illustrated that P/ML-NNG possessed MMP2 sensitivity.
To evaluate the responsively release of IMD-0354 and PD-1 mAb, in
vitro release profile was determined with or without collagenase IV. As
shown in Fig. 1h and Fig. 1i, the cumulative release of IMD-0354 and
PD-1 mAb from P/ML-NNG was 52.94 ± 1.59% and 91.24 ± 3.42%
when incubated with 400 μg mL − 1 collagenase IV in 48 h, respectively.
Cumulative release of drugs when incubated with collagenase IV were
higher (p < 0.001) than that without collagenase IV (22.01 ± 1.59% and
37.20 ± 5.48%) due to the triggered release by collagenase IV. Although
more drugs could be released when incubated with collagenase IV, the
release of IMD-0354 and PD-1mAb was only 22.01 ± 1.59% and 37.20
± 5.48% in the absence of collagenase IV at 48 h, respectively. These
Y. Liu et al.
Journal of Controlled Release 336 (2021) 621–634
626
(caption on next page)
Y. Liu et al.
Journal of Controlled Release 336 (2021) 621–634
627
results demonstrated that when P/ML-NNG was injected intravenously
into the mice and reached the tumor site, drug release could be achieved
under the condition of MMP2 overexpression.
3.4. Evaluation of active targeting and tumor accumulation mediated by
NGR
Active targeting mediated by NGR to tumor site was evaluated by
cellular uptake experiments on HUVEC in vitro (Fig. 2b). Nile red, a
fluorescence probe, was used as a substitute for hydrophobic drugs
which lacked the fluorescent group for cellular uptake experiments in
some studies [26]. Hence, Nile Red was used to prepare P/ML-NNG in
cellular uptake experiments. Meanwhile, P/ML-NG was developed
without 4-arm-PEG-NGR to serve as a control group, and competitive
inhibition experiments were added to study the active targeting. The
fluorescent images showed that at 1 h and 4 h, the red fluorescence
intensity of P/ML-NNG group was stronger than that of P/ML-NG group.
The flow cytometric analysis also illustrated that fluorescence intensity
of P/ML-NNG was significantly increased compared with P/ML-NG (**p
< 0.01，**p < 0.01) in HUVEC. In addition, the fluorescence intensity
of P/ML-NNG group was stronger than that of NGR + P/ML-NNG group.
These results suggested that P/ML-NNG could promote cellular uptake
mediated by NGR, specifically.
Tumor accumulation was evaluated by Real-time imaging in C57BL/
6 mice, and IR780 was used to prepare P/ML-NG (Fig. 2c-e). P/ML-NG
had the strongest tumor accumulation, and the fluorescence intensities
were 1.45 folds and 2.12 folds than P/ML-NG group and IR780 group,
respectively. These results suggest active targeting and enhanced tumor
accumulation capacities of P/ML-NG mediated by NGR.
3.5. Evaluation of RAW264.7 induced through IL-4
IL-4 induced RAW264.7 was evaluated by CLSM and flow cytometric
assay. As shown in Fig. S6, the CLSM images and flow cytometric
analysis indicated that IL-4 successfully stimulated RAW264.7 cells to
M2 TAMs.
3.6. Evaluation of target delivery to M2 TAMs mediated by mannose
Next, the target delivery to M2 TAMs of M-IMD-LNP mediated by
mannose was evaluated by cellular uptake experiments on RAW264.7 in
vitro and imaged by CLSM (Fig. 2f). Nile Red was used to prepare IMD￾LNP, M-IMD-LNP and P/ML-NNG, and IL-4 induced RAW264.7 was used
to replace the M2 TAMs [27]. As shown in Fig. 2f, the fluorescent images
and arithmetic mean intensity illustrated that fluorescence intensity of
M-IMD-LNP was significantly increased (***p < 0.001，***p < 0.001)
compared with without mannose group IMD-LNP. In addition, the
fluorescence intensity of M-IMD-LNP group was stronger than that of
Fig. 1. Characterization of M-IMD-LNP, ML-NNG, P/ML-NNG and MMP2 sensitive evaluation. (a-c) Scheme, sizes, zeta potentials and morphologies of (a) M-IMD￾LNP, (b) ML-NNG and (c) P/ML-NNG. (d) HPLC chromatograms of MMP2 sensitive peptide (1.5 mg mL− 1
) after incubated with enzyme at different concentrations
and time. (e) Scheme of MMP2 sensitive disintegrated of P/ML-NNG. (f) Size and (g) morphology of P/ML-NNG after incubated without or with 400 μg mL− 1
collagenase IV at 37 ◦C for 3 h. (h-i) The enzyme sensitive release profiles of (h) IMD-0354 and (i) PD-1 mAb in pH 7.4 PBS (***p < 0.001, compared with the group
of P/ML-NNG with enzyme).
Scheme 1. Schematic illustration of MMP2 responsive PD-1 mAb and IMD-0354 co-loaded nanogels (P/ML-NNG) for manipulation of TAMs functions to facilitate
the immune therapy effects of immune checkpoint antibodies. (a) Preparation of P/ML-NNG. (b) Delivery of P/ML-NNG. In vivo, P/ML-NNG could active targeting to
tumor tissue, and then disintegrated by MMP2 and release PD-1 mAb and M-IMD-LNP. PD-1 mAb and M-IMD-LNP could target to T cells and M2 TAMs, especially
block PD1/PD-L1 pathway and NF-κB pathway. Inhibition of NF-κB pathway could achieve both re-polarization M2 TAMs and downregulation the expression of PD￾L1 on M2 TAMs to reduce immune resistance and facilitate checkpoint blockade immune therapy.
Y. Liu et al.
Journal of Controlled Release 336 (2021) 621–634
628
(caption on next page)
Y. Liu et al.
Journal of Controlled Release 336 (2021) 621–634
629
MPITC+M-IMD-LNP group. These results proved that M-IMD-LNP could
enhance the cellular uptake of M2 TAMs through mannose mediated
targeting.
3.7. Evaluation of co-delivery to different targets
Subsequently, the differ-targeting delivery efficiency of P/ML-NNG
was evaluated by co-culture experiment and imaged by CLSM
(Fig. 2g). Nile Red and FITC labeled PD-1 mAb were used to prepare P/
ML-NNG. And T cells from mouse spleens and RAW264.7 induced by IL-
4 were co-cultured. IL-4 induced RAW264.7 was used to replace the M2
TAMs. As showed in CLSM images, the most orange fluorescence (FITC
labeled PD-1 mAb) was observed in the T cells and the most green
fluorescence (Nile Red) was observed in RAW264.7 of P/ML-NG (with
enzyme, the enzyme was 400 μgmL− 1 collagenase IV) group. These re￾sults indicated that PD-1 mAb and M-IMD-LNP could target T cells and
M2 TAMs separately after P/ML-NNG disintegrated by MMP2 in TME.
3.8. Evaluation of PD-1 expression
In order to verify that PD-1 mAb could specifically target and bind to
PD-1 on the T cells to block the PD-1/PD-L1 pathway, the expression of
PD-1 on the surface of T cells from mouse spleens and RAW264.7
induced by IL-4 was measured by CLSM and flow cytometric analysis
(Fig. 3a). Hence, the expression of PD-1 was evaluated of T cells and
RAW264.7 after immunofluorescence staining with the FITC-labeled
PD-1 mAb (Fig. 3a). The CLSM images and flow cytometric analysis
showed PD-1 was significantly higher expression on T cells than
RAW264.7 (***p < 0.001). These results illustrated when P/ML-NNG
disintegrated by MMP2 and release the PD-1 mAb and M-IMD-LNP,
most of thePD-1 mAb could target to T cells rather than M2 TAMs, and
then the PD-1/PD-L1 could be blocked effectively.
3.9. Evaluation the inhibition of NF-κB pathway
IMD-0354 was a specific inhibitor of NF-κB pathway. And some
research studies have certified that IMD-0354 could realize the effective
inhibition of NF-κB pathway. Hence, the inhibition of NF-κB pathway
was certified by CLSM of tumor cyro-sections after treatment with
different formulations (Fig. 3b). As shown in Fig. 3b, the NF-κB pathway
could be inhibited significantly in P/ML-NNG group mediated by IMD-
0354.
3.10. Evaluation the expression of PD-L1 on M2 TAMs in vitro and in
vivo
With the continuous exploration, it was found that the inhibition of
NF-κB pathway could down regulation the PD-L1 expression on M2
TAMs. Therefore, the PD-L1 expression on M2 TAMs was evaluated by
flow cytometric assay in vitro, and IL-4 induced RAW264.7 was used to
replace the M2 TAMs (Fig. 3c). The results shown more PD-L1 was
expressed on M2 TAMs, and the expression of PD-L1 could be effectively
inhibited by M-IMD-LNP and P/ML-NNG.
In addition, the expression of PD-L1 on M2 TAMs was proved by
imaging the immunofluorescence of tumor tissue sections after
treatment with different formulations (Fig. S7), the results were
consistent with the results in vitro.
3.11. Evaluation of M2 TAMs re-polarization in vitro and in vivo
IMD-0354 could repolarize M2 TAMs to M1 TAMs by inhibit NF-κB
pathway, which was proved by our previous works. RAW264.7 cells
were used as the TAMs polarization model in vitro in some publications
in journals such as ACS Nano, Nano letters, Biomaterials, Journal of
Controlled Release and Cancer Letters [28–30]. Hence, M2 TAMs
repolarization ability was certified on RAW264.7 after treatment with
different formulations for 24 h (Fig. 3d). F4/80 was used to mark TAMs.
F4/80 and CD206 was used to mark M2 TAMs. F4/80 and CD86 was
used to mark M1 TAMs. As measured by flow cytometric assay, it was
shown that M2 TAMs were reduced while the numbers of M1 TAMs were
increased in IMD-0354, IMD-LNP, M-IMD-LNP and P/ML-NNG (with
enzyme, the enzyme was 400 μg mL− 1 collagenase IV) groups compared
with IL-4 group. The ratio of M1/M2 was further calculated to further
illustrate the re-polarization of M2 TAMs (Fig. 3d). The increase of the
ratio indicated the enhancement of re-polarization. The ratio of M-IMD￾LNP and P/ML-NNG (with enzyme, the enzyme was 400 μg mL− 1
collagenase IV) groups was increased significantly, suggesting that M￾IMD-LNP and P/ML-NNG could repolarize M2 TAMs to M1 TAMs
efficiently.
M2 TAMs repolarization ability was evaluated by flow cytometric
assay in vivo (Fig. 3e). M1 TAMs (F4/80 + CD86 +) were most increased
in ML-NNG and P/ML-NNG groups. The ratio of M1/M2 in NML-NNG
and P/ML-NNG groups were same and mostly higher than other groups.
3.12. Evaluation of P/ML-NNG facilitated antitumor immunity
Inhibition of NF-κB pathway could reduce immune resistance and
facilitate checkpoint blockade immune therapy. The immunotherapy
efficacy was assessed on the infiltration of immunogenic cells. CD3 +CD4 +, CD3 +CD8 + T cells and Tregs were evaluated by flow cytometry
(Fig. 4a-c, i). The results showed that the amounts of Tregs in TME in P/
ML-NNG group was significantly lower than P-NNG group (p < 0.01)
and other groups, while the amounts of CD4 + T cells and CD8 + T cells in
P/ML-NNG group were significantly higher than P-NNG group (p <
0.001, p < 0.01) and other groups, respectively, which proved that P/
ML-NNG could effectively inhibit the proliferation of Tregs and pro￾mote infiltration of CD4+ T cells and CD8+ T cells to TME. Due to the
cytokines play a key role in the tumor immune microenvironment,
immunosuppressive cytokines TGF-β and IL-10 which could be secreted
by M2 TAMs and immune activating cytokines IFN-γ, TNF-α and IL-12
which could be secreted by M1 TAMs and T cells were evaluated by
ELISA assay (Fig. 4d-h). Meanwhile, the levels of INF-γ, IL-12 and TNF-α
in P/ML-NNG group were significantly higher than those in P-NNG
group (p < 0.01, p < 0.01, p < 0.001) and other groups, respectively,
while the levels of IL-10 and TGF-β were significantly lower than those
in the P-NNG group (p < 0.01, p < 0.01) and other groups, respectively.
The above results indicated that P/ML-NNG could reverse the immu￾nosuppressive state of TME, activate the anti-tumor immunity and
facilitate the immune therapy effects of PD-1 mAb.
Fig. 2. P/ML-NNG exhibited enhanced tumor accumulation and differ-targeting delivery efficiency to T cells and M2 TAMs, respectively. (a) Scheme of bio￾distribution and separated cell targeting process in tumor tissues. (b-e) Evaluation of active targeting and tumor accumulation mediated by NGR. (b) Fluores￾cence microscope images and flow cytometric analysis of cellular uptake of Nile Red loaded nanogels in HUVEC (n = 3, **p < 0.01, compared with the group of P/
ML-NG, scale bar: 20 μm). (c) In vivo imaging of mice after administration with IR780, IR780 loaded P/ML-NG and P/ML-NNG at different time intervals. (d) Ex vivo
imaging and (e) relative fluorescence intensity after mice were sacrificed after administration for 24 h (n = 3, **p < 0.01, compared with the group of P/ML-NG, ##p
< 0.01, compared with the group of IR780) (f) Evaluation of target delivery to M2 TAMs mediated by mannose. CLSM images and arithmetic mean intensity of
cellular uptake of Nile Red, Nile Red loaded IMD-LNP, M-IMD-LNP and P/ML-NNG in RAW264.7 (n = 3, ***p < 0.001, compared with the group of LNP, scale bar:
20 μm). (g) Evaluation of co-delivery to different targets. CLSM images of co-culture assay of T cells and RAW264.7 incubated with P/ML-NNG (IMD-0354 was
substituted by Nile Red; PD-1 mAb was labeled with FITC, scale bar: 20 μm). (For interpretation of the references to colour in this figure legend, the reader is referred
to the web version of this article.)
Y. Liu et al.
Journal of Controlled Release 336 (2021) 621–634
630
(caption on next page)
Y. Liu et al.
Journal of Controlled Release 336 (2021) 621–634
631
3.13. In vivo antitumor efficacy
For the antitumor efficacy, it was evaluated on B16 tumor bearing
C57BL/6 mice (Fig. 5a-f). The administration schedule was shown in
Fig. 5a. The administration dosages of PD-1 mAb and IMD-0354 were 5
mg kg − 1 and 1.14 mg kg − 1
, respectively. As shown in Fig. 5b, the tumor
Fig. 3. P/ML-NNG exhibited blocked of PD-1/PD-L1 pathway and inhibited of NF-κB pathway, simultaneously. (a) Evaluation of PD-1 expression to verify that PD-1
mAb could specifically target and bind to PD-1 on the T cells to block the PD-1/PD-L1 pathway. CLSM images and flow cytometric analysis of T cells and M2 TAMs
after immunofluorescence staining with the FITC-labeled PD-1 mAb. IL-4 induced RAW264.7 was used to replace the M2 TAMs. (n = 3, ***p < 0.001, compared with
the group of M2 TAMs, scale bar: 4 μm). (b) CLSM images of tumor cyro-sections after the mice were administrated with different groups to evaluate inhibition of NF-
κB pathway in vivo. TAMs were labeled by F4/80 (green), M2 TAMs were labeled by CD206 (red), NF-κB was labeled by NF-κB p65 (pink). (scale bar: 20 μm). (c) Flow
cytometric analysis of the expression of PD-L1 on M2 TAMs in vitro, IL-4 induced RAW264.7 was used to replace the M2 TAMs. (n = 3, ***p < 0.001, *p < 0.05,
compared with the group of IL-4). (d) Flow cytometric analysis and ratio of M1 TAMs to M2 TAMs after treatment with different samples in vitro (n = 3, ***p < 0.001,
**p < 0.01, *p < 0.05, compared with the group of IL-4, ##p < 0.01, #p < 0.05, compared with the group of IMD-0354). (e) Flow cytometric analysis and ratio of M1
TAM to M2 TAMs after treatment with different samples in vivo (n = 3, ***p < 0.001, **p < 0.01, compared with the group P/ML-NNG). (For interpretation of the
references to colour in this figure legend, the reader is referred to the web version of this article.)
Fig. 4. P/ML-NNG facilitated the immune therapy effects of immune checkpoint antibodies through inhibition of NF-κB pathway. (a) Percentage of CD4 + T cells in
tumor tissues after treatment with different formulations. (b) Percentage of CD8 + T cells in tumor tissues after treatment with different formulations. (c) Percentage
of Tregs in tumor tissues after treatment with different formulations. (d-h) The levels of different cytokines in peripheral blood. (n = 3, ***p < 0.001, **p < 0.01, *p
< 0.05, compared with the group of P/ML-NNG). (i) The characterization of lymphocytes phenotype. Flow cytometric analysis of the CD4 + T cells, CD8 + T cells and
Tregs phenotype in tumor tissues after treatment with different formulations.
Y. Liu et al.
Journal of Controlled Release 336 (2021) 621–634
632
Fig. 5. P/ML-NNG enhanced immunotherapy antitumor efficacy in B16 bearing C57BL/6 mice. (a) Schedule of intravenous injection with different formulations in
vivo (n = 6, black arrows represented the administration time). (b) Tumor volume changes (***p < 0.001, **p < 0.01, compared with P/ML-NNG group, n = 6). (c)
Photographs of tumors. (d) Tumor weights (***p < 0.001, **p < 0.01, *p < 0.05, compared with P/ML-NNG group, n = 6). (e) Body weight changes. (f) Immu￾nohistochemistry study results of tumor tissues (Ki67 400×, H&E 200× and TUNEL 20×).
Y. Liu et al.
Journal of Controlled Release 336 (2021) 621–634
633
volumes of P/ML-NNG group were significantly decreased compared to
NS groups (p < 0.001). At the end, isolated tumors of different groups
were photographed and weighed. In Fig. 5c and d, the results were
consistent with the results of Fig. 5b. These results indicated that the
tumor growth could be inhibited by P/ML-NNG effectively. As showed
in Fig. 5e, body weights were no evident reduction during the admin￾istration of different formulations, indicating that the developed for￾mulations with the low systemic toxicity. Ki67, H&E and TUNEL stains
were applied to evaluated cell apoptosis and proliferation (Fig. 5f). From
the H&E stain results, P/ML-NNG group shown most numerous cell
necrosis. About the Ki67 stain results, the brown represents cells pro￾liferation. The less proliferation of tumor cells was achieved in P/ML￾NNG group. About the TUNEL stain results, the green fluorescence
represents cells apoptosis. Massive cell apoptosis could be observed in P/
ML-NNG group. As shown in Fig. S9, PML-NNG group prolonged the
survival time of mice, effectively. Taken above results, it was revealed
that P/ML-NNG exhibited superior antitumor activity.
3.14. Preliminary safety evaluation
In order to preliminarily evaluate the safety of P/ML-NNG, the
cytotoxicity experiments (Fig. S11a), hemolysis (Fig. S11b) and H&E
stained images of histological sections (Fig. S11c) were investigated.
From Fig. S11a, we could know the cell viability of P/ML-NNG and free
IMD-0354 on B16 cells were all higher than 80% at the designated
concentrations (calculated based on IMD-0354), which illustrated that
no observe significant cytotoxicity of P/ML-NNG to B16 cells. As shown
in Fig. S11b, the hemolysis rates of P/ML-NNG were all less than 5% and
have no noticeable hemolysis. As shown in H&E stained images, the
main organs all showed no obvious lesions, suggesting low systemic
toxicity of P/ML-NNG. As shown in Fig. S10, the lymph and platelets
data of mice treated with P/ML-NNG showed no significant change
compared with the NS group and within the normal range, indicating the
low toxicity of P/ML-NNG. Taken above results, it was revealed that P/
ML-NNG exhibited low systemic toxicity.
4. Conclusion
In summary, the MMP2 responsive and tumor target nanogels (P/
ML-NNG) were developed, which were combined the IMD-0354 with
PD-1 mAb to manipulation of TAMs functions by inhibiting NF-κB
pathway and further to facilitate checkpoint blockade immune therapy.
P/ML-NNG could co-deliver PD-1 mAb and M-IMD-LNP to tumor site,
disintegrated by MMP2 and release drugs to target T cells and M2 TAMs,
separately. Furthermore, P/ML-NNG enhanced re-polarization of M2
TAMs and expression of PD-L1 on M2 TAMs, and futher reduced immune
resistance and facilitated PD-1 blockade immune therapy. Our studies
suggested that the combination of NF-κB pathway inhibitor with im￾mune checkpoint antibodies could be a promising therapeutic strategy
for facilitating the immune therapy effects.
Data availability
The authors declare that all data supporting the findings of this study
are available within the paper and its supplementary information, and
will be available on request.
Declaration of Competing Interest
The authors declare no conflict of interest.
Acknowledgements
This work was supported by National Natural Science Foundation of
China (81974498, 81773652) and the Young Scholar Program of
Shandong University (YSPSDU, 2017WLJH40). We thank Translational
Medicine Core Facility of Shandong University for consultation and in￾strument availability that supported this work. We were thankful to the
Microscopy Characterization Facility of Shandong University for the
help of CLSM.
Appendix A. Supplementary data
References
[1] A. Ribas, J.D. Wolchok, Cancer immunotherapy using checkpoint blockade,
Science 359 (2018) 1350–1355, https://doi.org/10.1126/science.aar4060.
[2] V. Longo, O. Brunetti, A. Azzariti, D. Galetta, P. Nardulli, F. Leonetti, N. Silvestris,
Strategies to improve cancer immune checkpoint inhibitors efficacy, other than
abscopal effect: a systematic review, Cancers (Basel) 11 (2019), https://doi.org/
10.3390/cancers11040539.
[3] L. Zhou, Q. Xu, L. Huang, J. Jin, X. Zuo, Q. Zhang, L. Ye, S. Zhu, P. Zhan, J. Ren,
T. Lv, Y. Song, Low-dose carboplatin reprograms tumor immune microenvironment
through STING signaling pathway and synergizes with PD-1 inhibitors in lung
cancer, Cancer Lett. 500 (2021) 163–171, https://doi.org/10.1016/j.
canlet.2020.11.049.
[4] P. Yang, H. Song, Y. Qin, P. Huang, C. Zhang, D. Kong, W. Wang, Engineering
dendritic-cell-based vaccines and PD-1 blockade in self-assembled peptide
Nanofibrous hydrogel to amplify antitumor T-cell immunity, Nano Lett. 18 (2018)
4377–4385, https://doi.org/10.1021/acs.nanolett.8b01406.
[5] C. Chen, A. Li, P. Sun, J. Xu, W. Du, J. Zhang, Y. Liu, R. Zhang, S. Zhang, Z. Yang,
C. Tang, X. Jiang, Efficiently restoring the tumoricidal immunity against resistant
malignancies via an immune nanomodulator, J. Control. Release 324 (2020)
574–585, https://doi.org/10.1016/j.jconrel.2020.05.039.
[6] R. Bommireddy, L.E. Munoz, A. Kumari, L. Huang, Y. Fan, L. Monterroza, C.
D. Pack, S. Ramachandiran, S.J.C. Reddy, J. Kim, Z.G. Chen, N.F. Saba, D.M. Shin,
P. Selvaraj, Tumor membrane vesicle vaccine augments the efficacy of Anti-PD1
antibody in immune checkpoint inhibitor-resistant squamous cell carcinoma
models of head and neck cancer, Vaccines (Basel) 8 (2020), https://doi.org/
10.3390/vaccines8020182.
[7] C. Xu, H. Hong, Y. Lee, K.S. Park, M. Sun, T. Wang, M.E. Aikins, Y. Xu, J.J. Moon,
Efficient lymph node-targeted delivery of personalized cancer vaccines with
reactive oxygen species-inducing reduced Graphene oxide Nanosheets, ACS Nano
14 (2020) 13268–13278, https://doi.org/10.1021/acsnano.0c05062.
[8] D. Muraoka, N. Seo, T. Hayashi, Y. Tahara, K. Fujii, I. Tawara, Y. Miyahara,
K. Okamori, H. Yagita, S. Imoto, R. Yamaguchi, M. Komura, S. Miyano, M. Goto, S.
I. Sawada, A. Asai, H. Ikeda, K. Akiyoshi, N. Harada, H. Shiku, Antigen delivery
targeted to tumor-associated macrophages overcomes tumor immune resistance,
J. Clin. Invest. 129 (2019) 1278–1294, https://doi.org/10.1172/JCI97642.
[9] T. Hou, T. Wang, W. Mu, R. Yang, S. Liang, Z. Zhang, S. Fu, T. Gao, Y. Liu,
N. Zhang, Nanoparticle-loaded polarized-macrophages for enhanced tumor
targeting and cell-chemotherapy, Nano-Micro Lett. 13 (2020), https://doi.org/
10.1007/s40820-020-00531-0.
[10] T. Wang, J. Zhang, T. Hou, X. Yin, N. Zhang, Selective targeting of tumor cells and
tumor associated macrophages separately by twin-like core-shell nanoparticles for
enhanced tumor-localized chemoimmunotherapy, Nanoscale 11 (2019)
13934–13946, https://doi.org/10.1039/c9nr03374b.
[11] F. Zhang, N.N. Parayath, C.I. Ene, S.B. Stephan, A.L. Koehne, M.E. Coon, E.
C. Holland, M.T. Stephan, Genetic programming of macrophages to perform anti￾tumor functions using targeted mRNA nanocarriers, Nat. Commun. 10 (2019)
3974, https://doi.org/10.1038/s41467-019-11911-5.
[12] B. Chen, A. Gao, B. Tu, Y. Wang, X. Yu, Y. Wang, Y. Xiu, B. Wang, Y. Wan,
Y. Huang, Metabolic modulation via mTOR pathway and anti-angiogenesis
remodels tumor microenvironment using PD-L1-targeting codelivery, Biomaterials
255 (2020) 120187, https://doi.org/10.1016/j.biomaterials.2020.120187.
[13] K. Taniguchi, M. Karin, NF-kappaB, inflammation, immunity and cancer: coming of
age, Nat. Rev. Immunol. 18 (2018) 309–324, https://doi.org/10.1038/
nri.2017.142.
[14] A. Sica, A. Mantovani, Macrophage plasticity and polarization: in vivo veritas,
J. Clin. Invest. 122 (2012) 787–795, https://doi.org/10.1172/JCI59643.
[15] T. Maeda, M. Hiraki, C. Jin, H. Rajabi, A. Tagde, M. Alam, A. Bouillez, X. Hu,
Y. Suzuki, M. Miyo, T. Hata, K. Hinohara, D. Kufe, MUC1-C induces PD-L1 and
immune evasion in triple-negative breast Cancer, Cancer Res. 78 (2018) 205–215,

https://doi.org/10.1158/0008-5472.CAN-17-1636.

[16] R. Noy, J.W. Pollard, Tumor-associated macrophages: from mechanisms to
therapy, Immunity 41 (2014) 49–61, https://doi.org/10.1016/j.
immuni.2014.06.010.
[17] D. Saha, R.L. Martuza, S.D. Rabkin, Macrophage polarization contributes to
Glioblastoma eradication by combination Immunovirotherapy and immune
checkpoint blockade, Cancer Cell 32 (2017) 253–267, e255, https://doi.org/10.10
16/j.ccell.2017.07.006.
[18] M. Yang, D. McKay, J.W. Pollard, C.E. Lewis, Diverse functions of macrophages in
different tumor microenvironments, Cancer Res. 78 (2018) 5492–5503, https://
doi.org/10.1158/0008-5472.CAN-18-1367.
Y. Liu et al.
Journal of Controlled Release 336 (2021) 621–634
634
[19] S. Wang, J. Li, J. Bai, J.M. Li, Y.L. Che, Q.Y. Lin, Y.L. Zhang, H.H. Li, The
immunoproteasome subunit LMP10 mediates angiotensin II-induced retinopathy in
mice, Redox Biol. 16 (2018) 129–138, https://doi.org/10.1016/j.
redox.2018.02.022.
[20] R. Yang, Z. Zhang, S. Fu, T. Hou, W. Mu, S. Liang, T. Gao, L. Guan, Y. Fang, Y. Liu,
N. Zhang, Charge and size dual switchable Nanocage for novel triple-interlocked
combination therapy pattern, Adv. Sci. (Weinh) 7 (2020) 2000906, https://doi.
org/10.1002/advs.202000906.
[21] N. Zhang, J. Song, Y. Liu, M. Liu, L. Zhang, D. Sheng, L. Deng, H. Yi, M. Wu,
Y. Zheng, Z. Wang, Z. Yang, Photothermal therapy mediated by phase￾transformation nanoparticles facilitates delivery of anti-PD1 antibody and
synergizes with antitumor immunotherapy for melanoma, J. Control. Release 306
(2019) 15–28, https://doi.org/10.1016/j.jconrel.2019.05.036.
[22] W.E. Berdel, S. Harrach, C. Brand, K. Brommel, A.F. Berdel, H. Hintelmann,
C. Schliemann, C. Schwoppe, Animal safety, toxicology, and pharmacokinetic
studies according to the ICH S9 guideline for a novel fusion protein tTF-NGR
targeting procoagulatory activity into tumor vasculature: are results predictive for
humans? Cancers (Basel) 12 (2020) https://doi.org/10.3390/cancers12123536.
[23] C. Ngambenjawong, H.H. Gustafson, S.H. Pun, Progress in tumor-associated
macrophage (TAM)-targeted therapeutics, Adv. Drug Deliv. Rev. 114 (2017)
206–221, https://doi.org/10.1016/j.addr.2017.04.010.
[24] Y. Liu, D. Zhang, Z.Y. Qiao, G.B. Qi, X.J. Liang, X.G. Chen, H. Wang, A peptide￾network weaved nanoplatform with tumor microenvironment responsiveness and IMD 0354
deep tissue penetration capability for Cancer therapy, Adv. Mater. 27 (2015)
5034–5042, https://doi.org/10.1002/adma.201501502.
[25] H. Yu, J. Chen, S. Liu, Q. Lu, J. He, Z. Zhou, Y. Hu, Enzyme sensitive, surface
engineered nanoparticles for enhanced delivery of camptothecin, J. Control.
Release 216 (2015) 111–120, https://doi.org/10.1016/j.jconrel.2015.08.021.
[26] C. Gong, C. Hu, F. Gu, Q. Xia, C. Yao, L. Zhang, L. Qiang, S. Gao, Y. Gao, Co￾delivery of autophagy inhibitor ATG7 siRNA and docetaxel for breast cancer
treatment, J. Control. Release 266 (2017) 272–286, https://doi.org/10.1016/j.
jconrel.2017.09.042.
[27] T. Wang, W. Mu, F. Li, J. Zhang, T. Hou, X. Pang, X. Yin, N. Zhang, “Layer peeling”
co-delivery system for enhanced RNA interference-based tumor associated
macrophages-specific chemoimmunotherapy, Nanoscale 12 (2020) 16851–16863,

https://doi.org/10.1039/d0nr04025h.

[28] C. Shi, T. Liu, Z. Guo, R. Zhuang, X. Zhang, X. Chen, Reprogramming tumor￾associated macrophages by nanoparticle-based reactive oxygen species
Photogeneration, Nano Lett. 18 (2018) 7330–7342, https://doi.org/10.1021/acs.
nanolett.8b03568.
[29] D. Yang, K. Liu, L. Fan, W. Liang, T. Xu, W. Jiang, H. Lu, J. Jiang, C. Wang, G. Li,
X. Zhang, LncRNA RP11-361F15.2 promotes osteosarcoma tumorigenesis by
inhibiting M2-like polarization of tumor-associated macrophages of CPEB4, Cancer
Lett. 473 (2020) 33–49, https://doi.org/10.1016/j.canlet.2019.12.041.
[30] X. Wei, L. Liu, X. Li, Y. Wang, X. Guo, J. Zhao, S. Zhou, Selectively targeting tumor￾associated macrophages and tumor cells with polymeric micelles for enhanced
cancer chemo-immunotherapy, J. Control. Release 313 (2019) 42–53, https://doi.
org/10.1016/j.jconrel.2019.09.021.
Y. Liu et al.