Nanocarrier-mediated targeting of chemotherapy-induced DPP4 enhances T cell infiltration and improves cancer immunochemotherapy

Shangyu Chen; Shichen Li; Zhangyi Luo; Yixian Huang; Hua Zhang; Yiqing Mu; Bei Zhang; Chien-Yu Chen; Ganqian Hou; Min Zhang; Song Li

doi:10.1016/j.bioactmat.2026.07.053

. 2026 Aug 4;67:95–113. doi: 10.1016/j.bioactmat.2026.07.053

Nanocarrier-mediated targeting of chemotherapy-induced DPP4 enhances T cell infiltration and improves cancer immunochemotherapy

Shangyu Chen ^a,^b,^c,¹, Shichen Li ^a,^b,^c,¹, Zhangyi Luo ^a,^b,^c, Yixian Huang ^a,^b,^c, Hua Zhang ^a,^b,^c, Yiqing Mu ^a,^b,^c, Bei Zhang ^a,^b,^c, Chien-Yu Chen ^a,^b,^c, Ganqian Hou ^a,^b,^c, Min Zhang ^b, Song Li ^a,^b,^c,^⁎

PMCID: PMC13463813 PMID: 42592018

Abstract

Cancer immunotherapy has transformed cancer treatment, yet its efficacy remains limited by a “cold” tumor immune microenvironment (TIME) characterized by poor T cell infiltration. Induction of immunogenic cell death (ICD) improves T cell infiltration by promoting the release of T-cell-recruiting chemokines such as CXCL10. However, the outcome may be limited by dipeptidyl peptidase IV (DPP4), a serine protease that degrades CXCL10 and related chemokines. We report that chemotherapeutic agents, particularly those capable of inducing strong ICD, transcriptionally upregulate DPP4 in cancer cells, establishing a negative feedback mechanism that dampens CXCL10-mediated immune responses. To overcome this limitation, we developed an enhanced triple-combination immunochemotherapy based on the codelivery of doxorubicin, a DPP4 inhibitor (Sitagliptin, Sitag), and a COX-2 inhibitor (5-ASA) using a 5-ASA-derivatized hyaluronic acid (HA) dendrimer nanocarrier (HASA). HA is a natural ligand for CD44 that is overexpressed on tumor and tumor endothelial cells, enabling precise targeting. In preclinical tumor models, this strategy enhanced T cell infiltration, antitumor immunity, and therapeutic efficacy while minimizing systemic toxicity, resulting in significant survival benefit when combined with anti-PD-1 therapy. Our work identifies chemotherapy-induced DPP4 upregulation as an adaptive immune-resistance feedback loop and establishes a targeted triple-drug nanoplatform that integrates chemotherapy, DPP4 inhibition, and COX-pathway modulation for enhanced immunochemotherapy.

Keywords: DPP4, Chemokines, Immunogenic cell death, Nanocarrier, Immunochemotherapy

Graphical abstract

Chemotherapy-induced immunogenic cell death promotes CXCL10 secretion but also induces DPP4 and COX-2, which attenuate CXCL10/CXCR3-mediated CD8⁺ T-cell recruitment and antitumor immunity. A CD44-targeted hyaluronic acid-dendrimer nanocarrier co-delivering doxorubicin (DOX) and sitagliptin (Sitag), with 5-ASA incorporated into the carrier, simultaneously preserves functional CXCL10 signaling and suppresses COX-2-associated immunosuppression, thereby enhancing CXCR3⁺ CD8⁺ T-cell infiltration and improving cancer immunochemotherapy.

Highlights

•
Immunogenic cell death (ICD)-inducing chemotherapy triggers a DPP4-mediated negative feedback response.
•
DPP4 truncates CXCL10 and impairs T cell infiltration.
•
HASA integrates CD44-targeted DOX/Sitag codelivery with 5-ASA functionality.
•
Dual inhibition of DPP4 and COX-2 enhances immunochemotherapy.

1. Introduction

Cancer immunotherapy, which leverages the body's immune system to combat malignancies, has transformed cancer treatment by enabling tumor recognition and elimination [[1], [2], [3]]. Strategies such as immune checkpoint inhibitors (ICIs) [4], chimeric antigen receptor (CAR) T-cell therapy [5], cancer vaccines [6], and cytokine-based treatments [7] have achieved remarkable anti-tumor responses in selected cancer patients. However, fewer than 40% of patients benefit from these treatments, largely due to a “cold” tumor immune microenvironment (TIME) marked by insufficient T cell infiltration, which restricts the broader clinical benefits of these strategies [[8], [9], [10]]– (see Scheme 1).

Scheme 1 — **Schematic illustration of nanocarrier-mediated enhancement of cancer immunochemotherapy. a)** DOX induces CXCL10 secretion but also upregulates DPP4, which truncates CXCL10 and limits CXCL10/CXCR3-mediated T-cell recruitment. HASA/Sitag/DOX enables CD44-targeted co-delivery of DOX and Sitag, thereby inhibiting DPP4-mediated CXCL10 truncation, preserving functional CXCL10 signaling, and enhancing CXCR3⁺ T-cell infiltration. b) Proposed mechanism showing that DOX directly induces DPP4 and COX-2. Sitag blocks DPP4-mediated CXCL10 truncation to promote CXCL10/CXCR3-dependent CD8⁺ T-cell infiltration, while 5-ASA suppresses COX-2-associated immunosuppressive signaling and contributes to enhanced antitumor immunity. Dashed arrows indicate a potential DPP4/NF-κB-related contribution to COX-2 induction. The figure is created with BioRender.com.

Effective immunotherapy relies on chemokine-driven T cell recruitment into solid tumors via chemotaxis [11]. Immunogenic cell death (ICD), a regulated form of cell death, amplifies antigen presentation and promotes antitumor immunity by inducing the release of T-cell-recruiting chemokines such as CXCL10 from dying tumor cells [[12], [13], [14]]. However, increasing evidence indicates that ICD alone, especially induced by chemotherapy, is often transient and insufficient to sustain effective T-cell activation and infiltration within the immunosuppressive tumor microenvironment [15]. This suggests the presence of counter-regulatory mechanisms that limit the chemokine-mediated immune response. While post-translational regulation of chemokine activity is well-documented, the feedback mechanisms governing their expression during chemotherapy-induced ICD and their impact on immunotherapy remain elusive. In addition, ICD-inducing chemotherapy such as doxorubicin (DOX) can activate DNA damage, inflammatory, and stress-response signaling pathways, including NF-κB- and p53-related programs, which may contribute to chemotherapy-induced transcriptional adaptation in tumor cells [[16], [17], [18], [19]].

Dipeptidyl peptidase IV (DPP4, also known as CD26), a ubiquitous type II transmembrane glycoprotein, regulates diverse physiological and pathological processes [[20], [21], [22]]. Functioning as a serine protease, DPP4 cleaves substrates bearing proline or alanine at the second N-terminal position, thereby degrading CXCL10 and related chemokines [23]. Barreira da Silva et al. demonstrated that inhibition of DPP4-mediated chemokine degradation can enhance lymphocyte trafficking by preserving bioactive CXCL10, suggesting its potential as a therapeutic target [24]. Although DPP4 inhibitors (DPP4i) such as sitagliptin (Sitag) are clinically well established and well tolerated in the treatment of type 2 diabetes, systemic, non-targeted DPP4 blockade in cancer settings may elicit distinct immunological consequences. Specifically, it has been reported to increase plasma levels of enzymatically active soluble DPP4 (sDPP4), which may affect the immune homeostasis in cancer patients [[25], [26], [27]]. Furthermore, the interplay between chemotherapy and DPP4 activity, particularly the feasibility of tumor-selective co-delivery of DPP4i with ICD-inducing chemotherapeutic agents to minimize the impact of both therapeutics on normal tissues, remains largely unexplored.

Here, we report that chemotherapeutic agents, particularly those capable of inducing strong ICD, upregulate DPP4 expression in tumors, revealing a chemotherapy-associated stress-adaptive feedback mechanism that counteracts CXCL10/CXCR3-mediated T cell recruitment. This response highlights the role of DPP4 in modulating T cell infiltration in cancer immunochemotherapy. To address this, we developed a CD44-targeted hyaluronic acid (HA)-dendrimer nanoplatform for synchronized delivery of DPP4i, chemotherapeutic agents, and the COX-pathway inhibitor. This targeted triple-combination strategy enhances tumor-localized delivery, improves T cell infiltration and antitumor immunity, and minimizes systemic toxicity in preclinical tumor models.

2. Results

2.1. DPP4 was induced by chemotherapeutic agents

The levels of DPP4 gene expression have previously been shown to vary among different cancer types, such as lung cancer, mesothelioma and melanoma. Similar results were shown in our analysis of the TCGA data: DPP4 gene expression was downregulated in breast cancer (BRCA) but was upregulated in pancreatic adenocarcinoma (PAAD) while no significant changes were seen in colon adenocarcinoma (COAD) (Fig. S1a). In addition, there is no significant correlation between the DPP4 gene expression levels and the clinical prognosis among these patients (Fig. S1b–d). Interestingly, further analysis of a patient cohort receiving chemotherapy shows a significantly negative correlation of the DPP4 expression levels with clinical prognosis (Fig. 1a–b), suggesting that DPP4 may play an important role in determining the response to chemotherapy. This chemotherapy-specific association may reflect the ability of chemotherapy to induce an immune cell-recruiting tumor microenvironment, in which DPP4-mediated truncation of CXCL10 more effectively impairs CXCR3-dependent T-cell recruitment.

To further elucidate a potential role of DPP4 in chemotherapy, we examined the basal DPP4 mRNA expression levels in several different types of cancer cell lines. As shown in Fig. 1c, the human cancer cell lines expressed higher levels of DPP4 mRNA compared to the murine cancer cell lines. And the breast cancer cell line 4T1 showed the lowest DPP4 mRNA expression level among these cell lines. We then examined the DPP4 mRNA expression levels in several murine (4T1 and CT26) and human (MM231 and HCT116) cancer cell lines following treatment with gemcitabine (GEM), cisplatin (CDDP), paclitaxel (PTX), oxaliplatin (Oxa), and DOX, respectively (Fig. 1d). Although the extent of DPP4 induction varied among the tested cancer cell lines and chemotherapeutic agents, agents known to induce strong ICD, including PTX, Oxa, and DOX, consistently elicited greater DPP4 upregulation than GEM or CDDP. These findings suggest that DPP4 induction is not restricted to a single chemotherapeutic drug but may represent a broader adaptive response to chemotherapy that is particularly evident under ICD-inducing conditions. We also confirmed the induction of DPP4 at the protein level in both murine and human cell lines by ELISA after DOX treatment (Fig. 1e).

Fig. 1f shows that DOX treatment induced the DPP4 expression at both mRNA and protein levels in a dose-dependent manner in 4T1 cells. In addition to the intracellular DPP4, the levels of soluble DPP4 (sDPP4) in culture medium were also significantly increased following DOX treatment. Similar results were observed in human MM231 cells (Fig. 1g). Consistent with these in vitro findings, Western blot analysis of tumor tissues further showed that chemotherapy treatment markedly increased DPP4 protein expression in vivo compared with the control group (Fig. S2). Given that ICD-inducing chemotherapy can activate multiple stress-responsive pathways, including NF-κB signaling, we further examined whether NF-κB inhibition affects DOX-induced DPP4 upregulation. Co-treatment with the NF-κB inhibitor BMS-345541 (BMS) markedly reduced DOX-induced DPP4 mRNA upregulation in both 4T1 and MM231 cells, whereas BMS alone had minimal effects on basal DPP4 mRNA expression (Fig. S3). These results suggest that NF-κB signaling may be involved in chemotherapy-induced DPP4 upregulation.

A number of endogenous substrates have been identified for DPP4 such as CXCL10, a proinflammatory chemokine that plays a crucial role in promoting T cell infiltration into solid tumors. In addition to decreasing effective concentrations of CXCL10, DPP4 action leads to the formation of truncated CXCL10 that acts as an antagonist rather than an agonist [27], further limiting T-cell and NK cell recruitment to tumors. Sitagliptin (Sitag) is a DPP4 selective inhibitor and has been shown to enhance antitumor activity through facilitating the CXCL10-mediated T cell infiltration into tumors [24]. However, little information is available on the effect of Sitag on the expression of CXCL10, alone and in combination with chemotherapy. Fig. 1h–i demonstrates that DOX alone induced CXCL10 mRNA and protein expression in a dose-dependent manner in both 4T1 and MM231 cell lines. The combination of DOX with Sitag further enhanced CXCL10 expression. This upregulation of CXCL10 was further supported by the increased amounts of secreted CXCL10 in the culture medium of 4T1 and MM231 cells (Fig. S4).

2.2. Biological consequence of DPP4 inhibition alone or in combination with chemotherapy

Sitag alone showed minimal effects on tumor cell proliferation at concentrations up to 5000 ng/mL (Fig. S5), consistent with previous reports [28]. Interestingly, in 4T1 cells, MTT analysis showed that Sitag enhanced DOX-mediated cytotoxicity (Fig. 2a), and dose-matrix analysis confirmed a synergistic interaction between Sitag and DOX, with a positive mean synergy score of 12.63 (Fig. 2b). Similar results were observed in MM231 cells, where Sitag also enhanced DOX cytotoxicity and produced a positive synergy score of 12.88 (Fig. S6). Analysis of the Cancer Therapeutics Response Portal (CTRP) further supported a positive association between DPP4 expression and DOX sensitivity (Fig. S7). Together, these results indicate that DPP4 inhibition enhances tumor cell sensitivity to DOX treatment.

Fig. 2 — Biological consequence of DPP4 inhibition alone or in combination with chemotherapy. a) Dose-response analysis of 4T1 cell viability following treatment with DOX alone or Sitag + DOX. b) Synergy score heatmap showing the combinational effect of Sitag and DOX across the indicated concentration ranges using SynergyFinder. **c-d)** qRT-PCR analysis of *Ptgs2* mRNA expression (**Left**) and representative Western blot of COX-2 protein expression (**Right**) in 4T1 (c) and MM231 (d) cells following DOX treatment, and corresponding densitometry analysis. **e-f)** qRT-PCR analysis of *Ptgs2* mRNA expression (**Left**) and representative Western blot of COX-2 protein expression (**Right**) in 4T1 control vector (CV) (e) and 4T1 DPP4 knockout (KO) (f) cells following Sitag, DOX, and Sitag + DOX treatments, and corresponding densitometry analysis. g) Representative Western blot of phospho-p65 (p-p65) in 4T1 CV, KO, and DPP4 re-expression (RE) cells, with corresponding densitometry analysis. h) Representative Western blot analysis of p-p65 in WT, KO, and RE cells following treatment with CT, Sitag, DOX, or Sitag + DOX. i) NF-κB luciferase reporter activity in 4T1 CV, KO, and RE cells. j) NF-κB luciferase reporter activity in WT, KO, and RE cells following treatment with saline (CT), BMS-345541 (BMS), DOX, or BMS + DOX. k) NF-κB luciferase reporter activity in WT, KO, and RE cells following treatment with CT, Sitag, DOX, or Sitag + DOX. **l-m)** qRT-PCR analysis of CXCL10 mRNA expression (**Left**) and ELISA analysis of CXCL10 protein expression (**Right**) in 4T1 (l) and MM231 (m) cells following treatment of CT, 5-ASA, and Sitag + DOX + 5-ASA triple combination (5-ASA + Sitag + DOX). Data are presented as mean ± s.e.m. Statistical analysis was performed using GraphPad Prism. Comparisons among three or more groups were analyzed using one-way ANOVA followed by Tukey's post hoc test. P < 0.05 was considered statistically significant. Sample sizes: n = 5 biological replicates per condition in **a-b**; n = 3 biological replicates per condition in **c-m**.

In addition to DPP4 induction, DOX treatment also upregulated the expression of cyclooxygenase-2 (COX-2), an enzyme whose overexpression has been associated with drug resistance, metastasis, recurrence, and immunosuppression in cancer [[29], [30], [31], [32]]. This upregulation was observed in both 4T1 and MM231 breast cancer cell lines at both mRNA and protein levels in a dose-dependent manner (Fig. 2c–d). Fig. 2e shows a trend of COX-2 induction following DPP4 inhibition by Sitag, while there was no difference in COX-2 induction between DOX alone group and the combination group of DOX with Sitag.

To further examine the role of DPP4 in this regulation, we generated DPP4 knockout (KO) 4T1 cells using lentiviral-mediated gene editing, followed by DPP4 re-expression to establish rescue cells. Efficient DPP4 depletion in KO cells and restoration of DPP4 expression in re-expression (RE) cells were confirmed by both protein expression and enzymatic activity analyses (Fig. S8). Importantly, DPP4 KO did not affect basal viability of 4T1 cells (Fig. S9). Interestingly, COX-2 induction by DOX, alone or in combination with Sitag, was almost abolished in DPP4 KO 4T1 cells (Fig. 2f). Consistently, analysis of tumor samples from BRCA patients in the TCGA database revealed a positive correlation between DPP4 and COX-2 expression levels (Fig. S10). Collectively, these findings suggest that DPP4 may be involved in DOX-mediated COX-2 induction and highlight the potential therapeutic benefit of combining COX-2 targeting with DPP4 inhibition and chemotherapy.

To elucidate the molecular basis underlying the involvement of DPP4 in DOX-induced COX-2 expression, we focused on the DPP4-NF-κB signaling axis, as DPP4 KO/KD has been reported to reduce NF-κB activity [33], and NF-κB is a well-established regulator of COX-2 expression. We therefore hypothesized that loss of DPP4 attenuates DOX-induced COX-2 expression by impairing NF-κB activation.

Consistent with this hypothesis, both total and phosphorylated NF-κB p65 (p-p65) levels were markedly reduced in DPP4 KO 4T1 cells (Fig. 2g). Re-expression of DPP4 in KO cells restored enzymatic activity to levels comparable to, or slightly higher than, those in CV cells (Fig. S8) and concomitantly restored both total p-65 and p-p65 levels (Fig. 2g). DOX treatment increased p-p65 levels in WT cells, with Sitag co-treatment showing a further upward trend that did not reach statistical significance (Fig. 2h). This observation is consistent with earlier findings that pharmacological DPP4 inhibition can potentiate NF-κB signaling [34]. In contrast, DPP4 KO attenuated DOX-induced p-p65 activation, while DPP4 re-expression restored this response (Fig. 2h).

An NF-κB reporter assay further supported the involvement of NF-κB signaling in this regulation. Both basal and DOX-induced NF-κB activities were reduced in DPP4 KO cells and restored by DPP4 re-expression (Fig. 2i–k). Consistently, DOX treatment significantly increased NF-κB reporter activity in WT cells, whereas this response was markedly attenuated in DPP4 KO cells and rescued in DPP4 re-expression (RE) cells (Fig. 2j–k). Sitag + DOX further enhanced NF-κB activation in DPP4-expressing WT and RE cells, whereas this effect was largely lost in DPP4 KO cells (Fig. 2k). Functionally, BMS, a selective inhibitor of IKK/NF-κB signaling, suppressed NF-κB reporter activity and attenuated DOX-induced COX-2 upregulation (Fig. 2j and Fig. S11). Together, these findings suggest that DPP4 status sustains NF-κB signaling, thereby contributing to chemotherapy-induced COX-2 upregulation.

Given the oncogenic and immunosuppressive roles of COX-2, we next asked whether pharmacologic inhibition of COX-2 could further potentiate the effects of DPP4 blockade during DOX treatment. 5-aminosalicylic acid (5-ASA), a clinically approved anti-inflammatory agent with COX-inhibitory activity and an established safety profile [35], was employed as a proof-of-concept pharmacological inhibitor. As shown in Fig. 2l–m, treatment with Sitag + DOX markedly increased CXCL10 expression in both 4T1 and MM231 cells at the mRNA and protein levels. Importantly, the addition of 5-ASA further enhanced CXCL10 expression compared with the Sitag + DOX combination alone, demonstrating an additional contribution of COX inhibition. This increase was further supported by elevated levels of secreted CXCL10 in the culture medium (Fig. S12). In addition, the triple combination showed the highest level of cytotoxicity compared to individual treatments (Fig. S13), suggesting that simultaneous targeting of DPP4 and COX-2, in combination with ICD-inducing chemotherapeutic agents, may represent a promising strategy to enhance antitumor efficacy.

2.3. Development of a 5-ASA-conjugated HA-dendrimer-based nanocarrier (HASA) for co-delivery of DPP4i and chemotherapy agents

Our data thus far suggests that DPP4 represents a promising therapeutic target for cancer, particularly in combination with chemotherapy. A key challenge in clinical application of this therapy is the effective and targeted codelivery of all three compounds specifically to tumor tissues. Dendrimers, which are highly branched, well-defined macromolecules capable of self-assembling into supramolecular nanoparticles (NPs) with a large internal void space, offer great potential for drug delivery [[36], [37], [38], [39], [40]]. Hyaluronic acid (HA), a well-known biodegradable polymer with CD44-targeting properties, was selected as the foundational building block for our dendrimer design [[41], [42], [43], [44]]. As shown in Fig. 3a and Fig. S14, we synthesized 5-ASA-conjugated HA-based dendrimers of different generations (HA-TG₁-ASA and HA-TG₂-ASA) and evaluated their tumor targeting efficiency.

Fig. 3 — Development and biophysical characterization of HASA-based nanocarrier. a) Synthesis scheme of HA-based dendrimer with different generations. b)¹H NMR spectra of HA, HA-TG_0.5, HA-TG₁, and HA-TG₁-ASA in D₂O. c) Near-infrared fluorescence (NIRF) *ex vivo* imaging of major organs and tumors from 4T1 tumor-bearing mice at 24 h following i.v. administration of different molecular weight dendrimers (HA_8k-TG₁-ASA, HA_25k-TG₁-ASA, and HA_1M-TG₁-ASA) and oxidative OHA_8k-TG₁-ASA with similar dendrimer structure. **d-e)** NIRF *ex vivo* imaging (d) and corresponding quantification of fluorescence intensity (e) of major organs and tumors treated with free Cy5.5 or HASA-Cy5.5. f) Ratio of tumor to liver at different time points following treatment of free Cy5.5 or HASA-Cy5.5. g) A schematic diagram of the protocol for the preparation of Sitag/DOX-co-loaded HASA NPs (HASA/Sitag/DOX). The figure is generated by BioRender.com. h) UV-Vis absorbance spectra of Sitag, DOX, HASA, HASA/Sitag, HASA/DOX, and HASA/Sitag/DOX in aqueous solution. Data are presented as mean ± s.e.m. Statistical analysis was performed using GraphPad Prism. Comparisons between two independent groups were analyzed using an unpaired two-tailed Student's t-test in **e-f**. P < 0.05 was considered statistically significant. Sample sizes: n = 3 biological replicates per condition.

The CD44-mediated tumor-targeting of HA-based nanocarriers is known to be influenced by various chemical modifications. These modifications include deacetylation, sulfation, and alterations to specific functional groups, such as carboxylic acid (-COOH) and primary alcohol (-OH) groups, with the degree of modification typically maintained below 40% [[45], [46], [47]]. This provided the rationale for choosing a 30% modification of HA in our study. Nonetheless, no studies seem to have been conducted to investigate the impact of preserving HA's structural integrity, particularly the intact sugar ring structure, on its CD44-mediated tumor-targeting.

As an initial step in our Structure-Activity Relationship (SAR) study, we investigated the impact of the chemical modifications at two distinct sites on HA, the vicinal dihydroxyl and COOH groups. We focused on first-generation (G₁) dendrimers conjugated with 5-ASA and labeled with Cy5.5 (Fig. S20a). For the modification of vicinal dihydroxyl group, we oxidized approximately 30% of the HA sugar ring structure to disrupt its integrity, yielding OHA_8k-TG₁-ASA. The synthesis route is illustrated in Fig. S18, and the chemical structure was validated by ¹H NMR (Fig. S19). However, near-infrared fluorescence (NIRF) imaging revealed minimal Cy5.5 signal in tumor tissues 24 h after administration of Cy5.5-labeled OHA_8k-TG₁-ASA, indicating its poor tumor-targeting ability (Fig. 3c). Consistently, in vitro uptake studies using both WT and CD44 KO 4T1 cells showed weak cell-associated Cy5.5 signals, irrespective of CD44 KO status. Assuming the CD44-targeting ability of HA-TG₁-ASA is defined as 100%, the calculated loss of CD44 targeting for OHA-TG₁-ASA was approximately 50% (Fig. S21a–b). These results quantitatively demonstrate that disrupting the structural integrity of the HA sugar ring compromises CD44-targeting efficiency and impairs CD44-mediated cellular uptake.

For the modification of COOH group, we preserved HA's structural integrity by directly conjugating dendrimer arms to its COOH groups, yielding HA-TG₁-ASA. The chemical structure of HA-TG₁-ASA was confirmed by ¹H NMR, which revealed modification of approximately 30% of the COOH groups (Fig. 3b). In stark contrast to OHA_8k-TG₁-ASA, HA-TG₁-ASA demonstrated significantly enhanced CD44-mediated tumor targeting (Fig. 3c). This improvement was further supported by in vitro studies showing an increased uptake of HA-TG₁-ASA in WT 4T1 cells compared to CD44 KO 4T1 cells (Fig. S21a–b), underscoring the importance of maintaining HA's structural integrity for effective tumor targeting.

The molecular weight (MW) of HA has been reported to affect the CD44 targeting efficiency of HA-modified NPs [[48], [49], [50]]. For example, in vitro studies of poly (lactic-co-glycolic acid) (PLGA)-based NPs have demonstrated that tumor cell uptake and selectivity are optimal with a HA of 800 kDa in MW, whereas they are significantly deteriorated with a HA of 1435 kDa [49]. To examine the impact of MW on the CD44 targeting efficiency of HA-modified dendrimer, 3 HA-TG₁-ASA of different HA MW (HA_8k-TG₁-ASA, HA_25k-TG₁-ASA, and HA_1M-TG₁-ASA) were synthesized (Fig. S15). NIRF imaging revealed that tumor accumulation decreased progressively as MW increased from 8 kDa to 1 MDa (Fig. 3c), with HA_8k-TG₁-ASA exhibiting the highest tumor-targeting efficiency. Consistently, in vitro uptake studies using 4T1 CD44 WT and 4T1 CD44 KO cells showed that all HA-TG₁-ASA variants exhibited higher uptake in CD44 WT cells than in CD44 KO cells, supporting CD44-dependent internalization (Fig. S21). Notably, increasing HA MW progressively reduced uptake in CD44 WT cells, whereas uptake in CD44 KO cells remained low and showed no significant difference among the variants (Fig. S21c). When the CD44-targeting capacity of HA_8k-TG₁-ASA was used as the reference, HA_25k-TG₁-ASA and HA_1M-TG₁-ASA exhibited approximately 28% and 50% loss of CD44 targeting, respectively (Fig. S21d). Collectively, our results demonstrate that the tumor-targeting capability of our HA-based dendrimers hinges on two key factors: preserving the structural integrity of the HA sugar ring and optimizing its MW, with 8 kDa HA conferring superior performance among the tested variants. These insights provide a foundation for designing more effective HA-based nanocarriers for tumor-targeted therapies.

Based on the above findings, we further synthesized the second-generation (G₂) dendrimer with low MW HA (HA_8k-TG₂-ASA, abbreviated as HASA, Fig. S16–17) to enhance the conjugation capacity for 5-ASA, aiming to improve its efficacy in cancer therapy. The biodistribution of Cy5.5-labeled HASA (HASA-Cy5.5) was initially evaluated in a 4T1 mouse tumor model using ex vivo NIRF imaging of excised tumors and other major organs or tissues. Strong and concentrated Cy5.5 signals were observed in tumor tissues at 24 h following i.v. injection of HASA-Cy5.5, whereas signals from free Cy5.5 dye in tumors were barely detectable (Fig. 3d–e). The quantitative tumor-to-liver ratios and fluorescence intensities in tumors and the liver at different time points following treatment with free Cy5.5 or HASA-Cy5.5 are shown in Fig. 3f and Fig. S22, respectively. HASA-Cy5.5 reached its highest tumor-to-liver ratio at 24 h, with higher tumor fluorescence and lower liver fluorescence than free Cy5.5, followed by a gradual decline in the ratio. In addition, serial blood sampling from tumor-bearing mice further demonstrated that HASA-Cy5.5 stayed in the blood significantly longer than free Cy5.5, with detectable blood signals remaining up to 72 h (Fig. S23). All these data indicate that HASA was stable in circulation and highly effective in targeting tumor tissues.

Fig. 3g illustrates the key components and process involved in developing the Sitag and DOX co-loaded nanocarrier, HASA/Sitag/DOX, which exhibited a low critical micelle concentration (Fig. S24). The UV-Vis spectrum (Fig. 3h) displays the characteristic absorbance peaks of free Sitag and DOX at approximately 266 and 480 nm, respectively. A slight red shift in absorption, likely attributed to van der Waals interactions between the polymer and the hydrophobic drugs, was observed in single drug-loaded systems (HASA/Sitag and HASA/DOX) as well as the dual drug co-loaded HASA/Sitag/DOX, confirming the successful encapsulation of the drugs into the HASA nanocarrier.

2.4. pH sensitive behavior of HASA nanocarrier

The HASA dendrimer readily self-assembled into NPs with an average diameter of 143.2 ± 2 nm and a surface charge of −25.6 ± 0.6 mV in PBS (Fig. 4a). Additionally, Sitag and DOX could be effectively encapsulated within HASA. In the co-loaded HASA/Sitag/DOX formulation, the final drug loading contents were 7.8 wt% for Sitag and 8.0 wt% for DOX, with corresponding drug loading efficiencies of 93.2% and 96.2%, respectively. Single-drug-loaded HASA/Sitag and HASA/DOX formulations showed comparable DLC/DLE values of 8.3 wt%/91.6% and 8.6 wt%/94.6%, respectively, accompanied by a slight reduction in particle size (Fig. 4a and Fig. S25-26). Notably, dual loading did not reduce the DLE of either drug, as both Sitag and DOX maintained high DLE values above 90% in the co-loaded formulation.

To further evaluate the stability of the co-loaded formulation under biologically relevant conditions, HASA/Sitag/DOX was incubated in 10% FBS-containing PBS or DMEM for up to 72 h. HASA/Sitag/DOX showed no obvious increase in hydrodynamic diameter and maintained a relatively stable zeta potential in both media, indicating good colloidal stability under serum-containing and cell culture conditions (Fig. S27a–b). Moreover, HASA/Sitag/DOX slowed a slow kinetics of DOX leakage in 10% FBS-containing PBS, with a cumulative DOX release of ∼40% over 72 h (Fig. S27c). These results support the stability and suitability of the co-loaded HASA/Sitag/DOX formulation for in vivo delivery.

Subsequently, the stimuli-responsive behavior of HASA and its corresponding drug release profile were investigated. As shown in Fig. 4b, our HASA dendrimer exhibits a potential pH sensitivity due to the highly branched tertiary amines in its arms and the COOH groups of 5-ASA. At physiological pH (7.4), the tertiary amine groups remain neutral, while some COOH groups are deprotonated, forming carboxylate anions. These deprotonated COOH groups can interact with the primary amine groups of drugs (e.g., DOX and Sitag) through electrostatic attraction. Additionally, hydrophobic interactions, such as π-π stacking, can further facilitate effective drug loading into the system. At acidic pH (pH < 5.5), the situation changes significantly. Most of the tertiary amine groups become protonated, leading to the swelling of the HASA dendrimer. Meanwhile, the COOH groups become neutral. At this point, electronic repulsion becomes the predominant interaction between the HASA dendrimer and the drugs, triggering drug release (Fig. 4b). This shift in interactions makes our HASA system particularly suitable for drug release in acidic cellular compartments, such as endosomes (pH ≈ 5.5).

Fig. 4c shows that HASA maintained a relatively small particle size (below 200 nm) at physiological pH but exhibited a significant increase in size (exceeding 200 nm) when pH dropped below 5.5, likely due to protonation effects. A similar trend was observed in Fig. 4d, where the surface charge of HASA underwent a noticeable shift. These findings suggest that HASA has a pK_a close to acidic pH (approximately 5.5), facilitating pH-responsive drug release. For the drug co-loaded HASA/Sitag/DOX formulation, the particle size remained around 120 nm at physiological pH (Fig. 4a and e). However, at acidic pH (5.5), dynamic light scattering (DLS) analysis revealed an additional peak with an average size of approximately 600 nm (Fig. 4e). TEM analysis further confirmed the pH-triggered structural change of HASA/Sitag/DOX. The NPs maintained relatively uniform spherical morphology at pH 7.4 but showed obvious morphological disruption and irregular aggregation at pH 5.5 (Fig. S28), providing direct morphological evidence for acid-triggered structural destabilization. The size increase at acidic pH was associated with an increase in DOX fluorescence intensity, likely due to the release of free DOX from HASA/Sitag/DOX (Fig. 4f).

To further investigate the pH-responsive drug release behavior, the release profiles of Sitag and DOX from various formulations over a period of 48 h under different pH conditions were analyzed using high-performance liquid chromatography (HPLC) (Fig. S29). As shown in Fig. 4g–h, both Sitag and DOX exhibited sustained and slow kinetics of release at physiological pH (7.4), with only ∼20% released within 12 h and a cumulative release of approximately 30% after 48 h. In contrast, both drugs showed an accelerated release at acidic pH compared to physiological pH. The cumulative drug release at pH 5.5 increased to around 40% for all formulations after 48 h. These results indicate that Sitag and DOX remain well-protected during circulation under physiological conditions but undergo pH-triggered release upon reaching the acidic tumor microenvironment, particularly after internalization and reaching endosomes.

2.5. Favorable pharmacokinetics profile and efficient tumor targeting of HASA nanocarrier

Following the initial demonstration of effective tumor targeting in a 4T1 orthotopic model (∼200 mm³, Fig. 3d), we further evaluated the efficiency of targeting to primary 4T1 tumors of different sizes (50-400 mm³). As shown in Fig. 5a and c, effective tumor targeting was seen in tumors as small as 50 mm³. Effective tumor targeting was also seen in several other subcutaneous murine tumor models, including CT26, Panc02, and MyC-Cap (Fig. 5b and d). Additionally, a preliminary study indicated that HASA/Sitag/DOX effectively accumulated in metastatic lung tumors established via i.v. injection of 4T1 tumor cells, while minimal signal was detected in normal lung tissue (Fig. 5e).

We further quantitatively analyzed the pharmacokinetics (PK) and tissue distribution of DOX using HPLC-fluorescence detection. Fig. 5f shows the blood concentration of DOX over time following i.v. administration of free DOX or HASA/Sitag/DOX in 4T1 tumor-bearing mice. DOX encapsulated in HASA/Sitag/DOX exhibited significantly higher AUC_inf and t_1/2 values than free DOX, indicating enhanced stability and prolonged circulation in the bloodstream (Fig. 5g). Furthermore, the incorporation of DOX into HASA/Sitag/DOX led to a remarkable increase in tumor accumulation, with approximately 7% of the injected dose (ID) detected in tumors 24 h post-intravenous administration (Fig. 5h–i). In addition, the liver distribution of DOX was reduced in the HASA/Sitag/DOX formulation compared to free DOX, suggesting improved tumor targeting and reduced off-target accumulation.

2.6. HA/CD44-mediated tumor targeting and transcytosis of HASA nanocarrier in vivo

Our previous study suggests [41,[51], [52], [53]] that, in addition to enhanced permeability and retention (EPR)-mediated passive targeting, CD44-mediated active targeting plays a crucial role in tumor accumulation and penetration of nanocarrier decorated with chondroitin sulfate, another natural ligand for CD44. To define the role of CD44 in the effective tumor targeting of our HASA nanocarrier, we examined the cellular uptake of HASA/Sitag/DOX by CD44 WT or CD44 KO 4T1 cells. As shown in Fig. S29, cell-associated DOX signals were significantly higher in WT cells than in CD44 KO cells. In addition, the DOX signals in WT cells were significantly reduced by the pretreatment with excess amounts of free HA, clearly demonstrating a role of HA/CD44 in mediating the cellular uptake of HASA/Sitag/DOX. To further investigate the contributions of HA/CD44-mediated active targeting in vivo, we established CD44 WT or CD44 KO tumors in CD44^+/+ or CD44^−/− mice. The tissue distribution of the Cy5.5-labeled HASA/Sitag/DOX was then analyzed using NIRF imaging (Cy5.5) and HPLC (DOX). As shown in Fig. 6a–b, CD44 KO in mice resulted in a significant reduction of Cy5.5 signals in CD44 WT tumors. CD44 KO in tumor cells also led to a marked decrease of Cy5.5 signals in tumors in CD44^+/+ mice. KO of CD44 in tumors also led to further reductions of the signals in tumor in CD44^−/− mice, albeit to a less extent. Similar results were shown in quantitative analysis of DOX distribution (Fig. 6c).

Fig. 6d shows the distribution of Cy5.5 signals in tumor sections from the different groups. Cy5.5 signals were widely distributed throughout CD44 WT tumors in CD44^+/+ mice. In contrast, in CD44 KO tumors grown in CD44^+/+ mice, there is a modest reduction in the overall Cy5.5 signals. In addition, Cy5.5 signals were predominantly confined to areas adjacent to blood vessels (marked by CD31). CD44 KO in mice led to a drastic reduction in Cy5.5 signals within tumors, with the most pronounced decrease observed in CD44^KO tumors. We further quantified the spatial distribution of HASA-Cy5.5 relative to CD31⁺ tumor vessels in tumor sections. The penetration index, calculated as the ratio of Cy5.5 fluorescence intensity in distal tumor regions to that in perivascular regions surrounding CD31⁺ vessels, was highest in CD44^+/+ mice bearing CD44 WT tumors and was markedly reduced when CD44 was absent in either the tumor or host compartment (Fig. S30). These results further support the essential role of CD44-mediated targeting in HASA tumor accumulation and penetration in vivo.

To elucidate a role of HA/CD44-mediated transcytosis across tumor endothelial cells (ECs) and tumor cells in the tumor accumulation and penetration of HA dendrimers, we conducted a transwell study using tumor ECs and CD44 WT or KO tumor cells. Quiescent human umbilical vein endothelial cells (HUVEC_q) have low levels of CD44 but become activated (HUVEC_a) to express a high level of CD44 [54]. HUVEC_a has been widely used as a model tumor EC. We also synthesized an inulin-based polymer (In-ASA) with a dendrimer structure similar to that of HASA but lacking CD44 binding activity (Fig. S18).

As illustrated in Fig. 6e, various Cy5.5-labeled formulations, including free Cy5.5, In-ASA-Cy5.5, and HASA-Cy5.5, were added to the upper chamber and fluorescence signals were measured in the lower chamber at different time intervals (Fig. 6f–h). HASA-Cy5.5 exhibited significantly higher fluorescence signals in the lower chamber at all time points, indicating effective transcytosis. Notably, CD44^KO 4T1 cells or HUVEC_q cells showed markedly lower signals compared to CD44^WT 4T1 cells or HUVEC_a cells. Furthermore, this process was significantly attenuated for In-ASA NPs. Similar findings were observed in cellular uptake experiments (Fig. S31), highlighting a critical role of HA/CD44 interactions in both cellular uptake and transcytosis of HASA dendrimer-based NPs.

We further evaluated the tumor penetration ability of these NPs using a 4T1 tumorsphere model (Fig. 6i–j). Confocal z-stack imaging showed that HASA-Cy5.5 effectively penetrated into the inner regions of CD44 WT 4T1 tumorspheres, whereas the non-CD44-targeted In-ASA-Cy5.5 showed limited penetration and was largely retained in the peripheral regions. Quantitative analysis using a penetration index, calculated as the ratio of Cy5.5 mean fluorescence intensity in the middle section to that in the peripheral section, further confirmed the superior penetration of HASA-Cy5.5 compared with In-ASA-Cy5.5 (Fig. S32). Importantly, this enhanced penetration was markedly reduced in CD44 KO 4T1 tumorspheres, reinforcing the essential role of HA/CD44-mediated uptake and transcytosis in tumor penetration. Collectively, our findings suggest that the enhanced tumor targeting and deep penetration of HASA dendrimer-based NPs are driven by the combined contribution of the EPR effect, active targeting of tumor ECs and tumor cells, and HA/CD44-mediated internalization and transcytosis.

One concern is the potential uptake of NPs by immune cells especially T cells, which could impact their functions. However, CD44 levels in T cells have been reported to be significantly lower than those in tumor cells [51]. Accordingly, HASA NPs were barely taken up by T cells in tumor tissues at 24 h following i.v. administration, suggesting limited direct interference with T-cell function. Consistent with their high expression levels of CD44, tumor cells and tumor ECs effectively took up HASA NPs. Interestingly, significant amounts of HASA NPs were also taken up by fibroblasts and macrophages (Fig. S33). Such uptake by tumor-associated stromal and myeloid cells may also influence the intratumoral distribution of HASA NPs and contribute to subsequent modulation of the TME.

2.7. Co-delivery of DPP4i and chemotherapeutic drugs led to improved anti-tumor efficacy and immunity

Our data so far suggests that the combination of Sitag with chemotherapy can enhance cytotoxicity towards tumor cells and promote the production of CXCL10. In addition, DOX and Sitag could be effectively co-loaded into HASA NPs that are highly effective in targeted delivery to tumors. To evaluate the therapeutic effect of this novel combination therapy, we first examined ICD-associated responses induced by DOX-containing HASA formulations. Flow cytometric analysis showed that HASA/DOX and HASA/Sitag/DOX increased DPP4 expression and cell-surface CRT exposure in both 4T1 and MM231 cells (Fig. S34a–b). Consistently, both formulations promoted HMGB1 release into the culture medium (Fig. S34c).

Time-course analysis further showed that DPP4 mRNA expression in 4T1 cells increased after DOX treatment and peaked at 48 h (Fig. S34d). Flow cytometric analysis at 6, 24, and 48 h revealed coordinated increases in DPP4 expression and CRT exposure in both 4T1 and MM231 cells following DOX-containing HASA treatment (Fig. S33e–h). These results suggest that DOX-induced ICD-associated stress is accompanied by DPP4 feedback upregulation, supporting the rationale for concurrent DOX delivery and DPP4 inhibition within the same therapeutic window. Importantly, CRT immunofluorescence staining of tumor tissues collected after treatment further confirmed increased ICD-associated CRT exposure in vivo, particularly in the HASA/DOX and HASA/Sitag/DOX groups (Fig. S35).

We next investigated the in vitro biological activities of HASA/Sitag/DOX. In 4T1 cells, HASA/Sitag/DOX markedly inhibited DPP4 activity, increased CXCL10 protein levels, and showed stronger cytotoxicity than the free triple-drug combination (Fig. S36). To further determine whether DPP4 inhibition preserved functional CXCL10/CXCR3 signaling, we collected conditioned media (CM) from treated tumor cells and applied them to isolated CD8⁺ T cells. CM from Sitag/DOX- and HASA/Sitag/DOX-treated tumor cells induced greater CXCR3 internalization than DOX-treated CM, as indicated by reduced surface CXCR3 MFI on CD8⁺ T cells (Fig. S37a–c). In a transwell chemotaxis assay, these CM also promoted greater CD8⁺ T-cell migration compared with DOX-treated CM (Fig. S37d). These results demonstrate that HASA/Sitag/DOX integrates DOX-induced ICD-associated responses with effective DPP4 inhibition, increased CXCL10 availability, preserved CXCL10/CXCR3 signaling, enhanced CD8⁺ T-cell chemotactic activity, and improved tumor cell killing in vitro.

Next, we examined the in vivo inhibitory activity of Sitag in tumor tissues following i.v. administration of Sitag/DOX co-loaded NPs. This involved i.p. injection of Gly-Pro-aminoluciferin into 4T1-GFP-luc tumor-bearing mice treated with CT, HASA/DOX, or HASA/Sitag/DOX. Conversion of Gly-Pro-aminoluciferin to aminoluciferin by the intratumor DPP4 followed by luciferase action leads to luminescence signals. As shown in Fig. S38a, strong luminescence signals were seen in luciferase-expressing tumors in the CT group, which reflects the basal level of DPP4 activity in the tumors. Treatment with HASA/DOX led to significant increases in the luminescence signals. This is consistent with earlier in vitro study showing induction of DPP4 expression by DOX (Fig. 1d–g). Notably, HASA/Sitag/DOX treatment significantly suppressed this elevated DPP4 activity. Quantitative analysis further confirmed this inhibitory effect (Fig. S38b), demonstrating the potent DPP4 inhibition achieved by HASA/Sitag/DOX in vivo.

Fig. 7a–b shows the anti-tumor efficacy of various treatments in 4T1 orthotopic model. Treatment with HASA dendrimer loaded with either Sitag or DOX alone resulted in only modest therapeutic effects. However, the co-delivery of Sitag and DOX via HASA dendrimer significantly enhanced the therapeutic efficacy. We further validated the antimetastatic efficacy of HASA/Sitag/DOX in a 4T1 lung metastasis model. Compared with CT, HASA/Sitag/DOX markedly reduced lung metastatic burden, as evidenced by representative lung images, ex vivo bioluminescence imaging, and reduced lung weights (Fig. S39). We also developed a pharmacologically “inert” control nanocarrier, HASA_inert, in which 5-ASA was replaced with a structurally similar analogue that lacks COX pathway-inhibitory activity. HASA_inert/Sitag/DOX showed reduced therapeutic efficacy compared with HASA/Sitag/DOX, supporting the contribution of 5-ASA to the overall treatment effect (Fig. S39).

Additionally, DPP4 activity analysis of tumor tissues further confirmed effective DPP4 inhibition following treatment with 5-ASA + Sitag + DOX, HASA/Sitag, and HASA/Sitag/DOX (Fig. 7c). Notably, HASA dendrimer-mediated drug delivery achieved superior and tumor-specific DPP4 inhibition compared to the free drug combination. Given that systemic DPP4 inhibition has been reported to increase circulating sDPP4 levels [55], we further evaluated plasma sDPP4 levels and DPP4 enzymatic activity after repeated treatment. HASA/Sitag/DOX did not significantly increase plasma sDPP4 levels compared with the control group, whereas plasma DPP4 enzymatic activity was markedly reduced (Fig. S40), indicating effective circulating DPP4 enzymatic inhibition without detectable systemic sDPP4 elevation. Lack of increased DPP4 activity in tumors treated with HASA/DOX over the control group is likely due to decreased viability/metabolic activity of tumor cells following repeated treatments, as well as the three-day gap after treatment cessation (Fig. 7c). These findings highlight the tumor-targeting capability of the HASA dendrimer, ensuring efficient DPP4 inhibition and improved therapeutic outcomes while minimizing overt systemic sDPP4 perturbation.

All treatments were well tolerated, with normal weight gains, unchanged serum biochemical indices (AST, ALT, and creatinine), and no obvious abnormalities in H&E-stained major organs compared with the CT group (Fig. S41–43). Repeated-dose maximum tolerated doses (MTD) studies in CD-1 mice (three injections, 100-300 mg/kg HASA or HASA/Sitag/DOX formulations) revealed no obvious impacts on mouse general wellbeing or histology in major organs (Fig. S44a–b), establishing an MTD ≥300 mg/kg. In addition, long-term serum biochemical monitoring showed no significant elevation of AST or ALT at 1, 14, or 28 days after HASA administration, suggesting the absence of delayed hepatotoxicity (Fig. S44c–d). Flow cytometric profiling of splenic immune cells showed minimal changes in the proportions of several immune cell populations including CD8⁺ cells, CD4⁺ T_reg, MDSC, and M2-like macrophages (Fig. S45–46), excluding measurable immunosuppression. Collectively, these results confirm the excellent biosafety of the HASA nanoplatform for further preclinical development.

The effective inhibition of DPP4 in vivo led to a significant upregulation of CXCL10 in the treatment groups, with the most pronounced effect observed in the HASA/Sitag/DOX group (Fig. 7d). Additionally, RNAseq data demonstrated the upregulation of multiple immune-related pathways in the triple combination group, including enhanced IL2-STAT5 signaling and a strengthened interferon response (Fig. S47). Fig. 7e–i shows the flow analysis of tumor-infiltrating immune cells following various treatments. Using a sequential gating strategy based on singlets, live cells, CD45⁺ leukocytes, and CD8⁺ T cells, we quantified both the infiltration and functional activation of tumor-infiltrating T cells (Fig. S45). Specifically, significant increases were seen in the numbers of CD45⁺ immune cells, CD8⁺ T cells as well as CD8⁺ CXCR3⁺ T cells in the tumors treated with HASA/Sitag or HASA/Sitag/DOX (Fig. S48a–b and Fig. 7f), indicating enhanced CXCL10-mediated T cell infiltration. Fig. 7g–h shows that formulations loaded with Sitag or DOX, particularly HASA/Sitag/DOX, also led to significant increases in the numbers of functional (GzmB⁺ and/or IFNγ⁺) CD8⁺ T cells. The free drug combination (5-ASA + Sitag + DOX) only showed a modest effect in increasing the number of functional CD8⁺ T cells, likely due to the limited delivery of the free drugs to the tumors.

To further validate the role of CD8⁺ T cells in the therapeutic response, we performed an in vivo CD8⁺ T-cell depletion study. HASA/Sitag/DOX significantly inhibited tumor growth, whereas CD8⁺ T-cell depletion markedly weakened this antitumor effect, with stable body weights observed across groups (Fig. S49a–b). Flow cytometric analysis confirmed effective depletion of tumor-infiltrating CD8⁺ T cells following anti-CD8 antibody treatment (Fig. S49c). CD8 immunohistochemistry (IHC) staining further showed increased CD8⁺ cells within tumor cell-rich regions in the HASA/Sitag/DOX group, whereas CD8⁺ staining was substantially reduced after anti-CD8 treatment (Fig. S50).

We further examined whether the increased PD-1 expression on CD8⁺ T cells was associated with an exhaustion-like phenotype (Fig. 7j–k). Although HASA/Sitag/DOX increased the frequency of PD-1⁺ CD8⁺ T cells, it did not significantly increase LAG-3⁺ or TIM-3⁺ CD8⁺ T cells, nor PD-1⁺LAG-3⁺ or PD-1⁺TIM-3⁺ double-positive CD8⁺ T-cell populations (Fig. S51). Together with the elevated GzmB⁺ and IFNγ⁺ CD8⁺ T cells (Fig. 7g–h), these results support that HASA/Sitag/DOX promotes functional CD8⁺ T-cell activation, with PD-1 upregulation likely reflecting activation-associated checkpoint feedback rather than dominant T-cell exhaustion.

Additionally, treatment with HASA/Sitag/DOX significantly increased the M1/M2-like macrophage ratio (Fig. 7i) while reducing the abundance of CD4⁺ T_reg and MDSC cells (Fig. S48c–d), further contributing to an improved anti-tumor immune response. Western blot analysis (Fig. S52) of tumor tissues revealed a significant upregulation of DPP4 following DOX treatment, either alone or with co-delivery of 5-ASA and Sitag via the HASA dendrimer, highlighting the crucial role of DPP4 inhibition in sustaining CXCL10-driven antitumor immunity. Consistent with the in vitro findings, ICH demonstrated upregulation of COX-2 expression in both the free-drug combination and HASA drug-loaded groups, reinforcing the need of including ASA to counteract the COX-2-mediated negative impact (Fig. S53).

In line with the activation-associated increase in PD-1 expression on tumor-infiltrating CD8⁺ T cells (Fig. 7j–k), and in the absence of a dominant TIM-3- or LAG-3-associated exhaustion-like phenotype (Fig. S50), we next evaluated whether anti-PD-1 therapy could further improve the therapeutic outcome of HASA/Sitag/DOX. Indeed, the combination of HASA/Sitag/DOX with an anti-PD1 antibody (α-PD-1) led to a remarkable improvement in anti-tumor activity, as evidenced by a significant prolongation of survival. Notably, in the HASA/Sitag/DOX + α-PD-1 group, 2 out of the 8 mice achieved complete tumor eradication five days after the fifth treatment (Fig. 7l–m).

In a preliminary study, HASA-mediated co-delivery of Sitag and DOX was more effective than NPs-mediated co-delivery of DOX and DPP4 siRNA in inhibiting tumor growth (Fig. S54), despite significant DPP4 downregulation by the siRNA formulation (Fig. S55). These findings suggest that pharmacological DPP4 inhibition and DPP4 depletion may not be functionally equivalent, potentially reflecting additional DPP4-associated functions beyond CXCL10 truncation.

3. Discussion

As a dipeptidyl peptidase that inactivates the activity of CXCL10, DPP4 plays an important role in regulating the trafficking of T cells. Inhibition of DPP4 has previously been shown to improve antitumor activity through enhancing CXCL10-mediated tumor infiltration of T cells [24]. In this study, we extended our understanding of DPP4 by showing that the expression of DPP4 was significantly upregulated by chemotherapeutic agents, potentially acting as a negative feedback mechanism that dampens the magnitude of immune response. Combination of DPP4 inhibition with chemotherapy via NPs-mediated co-delivery led to significantly enhanced therapeutic outcome and an improved tumor immune microenvironment.

The expression levels of DPP4 in cancer patients vary with the cancer types (Fig. S1). In addition, bioinformatics analysis of TCGA patient data shows no significant correlations between the DPP4 gene expression levels and the patient prognosis. Interestingly, in a cohort of TNBC patients receiving chemotherapy, the prognosis was negatively correlated with the DPP4 gene expression levels (Fig. 1a). Similar results were seen in another cohort of COAD patients receiving chemotherapy (Fig. 1b), suggesting a potential of combining chemotherapy with a DPP4 inhibitor for improved treatment. This chemotherapy-specific association may be explained by chemotherapy-induced remodeling of the tumor immune microenvironment. Chemotherapy can induce ICD and CXCL10/CXCR3-dependent T-cell recruitment, whereas high DPP4 expression may truncate CXCL10 and weaken this immune-recruiting response, thereby enhancing its pro-tumor effect in chemotherapy-treated tumors.

The transcriptional regulation of DPP4 is, so far, not well understood. Our data shows that chemotherapeutic drugs especially those that are potent in inducing ICD such as PTX, Oxa, and DOX significantly upregulates the expression of DPP4 expression at transcriptional level (Fig. 1d). This response was not limited to DOX or to a specific drug class, and increased DPP4 activity was also observed in tumor tissues after DOX treatment in vivo (Fig. S38), indicating that chemotherapy-induced DPP4 upregulation is a general response associated with ICD-inducing treatment. As ICD is associated with increased production of T cell-recruiting chemokines such as CXCL10, the DOX-mediated induction of DPP4 expression likely serves as a negative feedback mechanism to limit the level and activity of CXCL10. Combination of DOX and Sitag led to further increases in CXCL10 production both in vitro and in vivo (Fig. S36b and Fig. 7d), suggesting that DPP4 induction restrains the accumulation of functional CXCL10 following chemotherapy.

Although the upstream mechanism responsible for chemotherapy-induced DPP4 transcription remains to be fully elucidated, our preliminary studies using the NF-κB inhibitor BMS suggest that NF-κB signaling contributes to this response (Fig. S3). This observation is consistent with previous reports showing that NF-κB is activated during DOX-induced cellular stress and can directly regulate DPP4 transcription [56]. Elucidating the precise molecular mechanisms governing DPP4 induction following chemotherapy is beyond the scope of the present study and warrants future investigation. Nevertheless, our findings establish that chemotherapy-induced DPP4 upregulation functionally promotes CXCL10 truncation, thereby limiting CXCR3-dependent T-cell recruitment and attenuating anti-tumor immune responses.

COX-2 plays an important role in cancer development and progression, and its overexpression correlates to tumor growth, metastasis, immunosuppression, and resistance to treatment. COX-2 expression was significantly induced by DOX treatment (Fig. 2c–d), which is consistent with literature report [29,30,32]. Sitag treatment also increased the expression of COX-2 although it is not statistically significant. It is interesting to note that COX-2 induction by DOX, alone or in combination with Sitag, was essentially abolished in DPP4^−/− cells (Fig. 2e–f). Mechanistic studies suggest that this effect is regulated by the DPP4-NF-κB axis. DPP4 KO reduced basal and DOX-induced NF-κB activation, whereas DPP4 re-expression restored p65/p-p65 levels and NF-κB reporter activity (Fig. 2g–k). Furthermore, pharmacological inhibition of IKK/NF-κB signaling attenuated DOX-induced COX-2 expression (Fig. S11), supporting NF-κB as an important intermediary linking DPP4 to chemotherapy-induced COX-2 upregulation. Together, these findings suggest that chemotherapy-induced DPP4 and COX-2 are coordinated adaptive responses that may limit antitumor immunity, providing a rationale for combining chemotherapy with simultaneous DPP4 and COX-2 inhibition.

Building on these findings, we next investigated whether simultaneous targeting of DPP4 and COX-2 signaling could further enhance the antitumor effects of chemotherapy. Pharmacological inhibition of COX-2 signaling using 5-ASA further enhanced CXCL10 production when combined with DOX and Sitag (Fig. 2l–m). The triple combination treatment produced the strongest induction of CXCL10 and the greatest cytotoxicity against tumor cells, supporting the therapeutic potential of concurrently targeting DPP4 and COX-2 signaling during chemotherapy.

To translate the clinical translation of this combination strategy, HASA nanocarrier is designed for co-delivery of Sitag and DOX considering its highly branched structure and multiple mechanisms of carrier-drug interactions, contributing to high drug loading capacity and excellent colloidal stability. As part of SAR study, we showed that disruption of the HA sugar ring via oxidation led to a drastic decrease in the efficiency of CD44 targeting (Fig. 3c). In consistent with literature, modification of the COOH of HA by less than 40% well maintained the effectiveness of CD44 targeting [46]. The impact of MW on HA-mediated CD44 targeting appears to be context dependent. While HA of 800 kDa was shown to be optimal for HA-PLGA system [49], a much smaller HA was shown to be ideal for HA-dendrimer in this study. It should be noted the conclusion from the former study was derived from an in vitro study while our optimization study was conducted in vivo. It is likely that conjugation of a bulky dendrimer to HA shall decrease the absolute binding affinity of modified HA to CD44. However, such “optimal” decrease in binding affinity may preferentially decrease the interaction with LSECs compared to tumor ECs or tumor cells as shown in our recent study [51] with PEG modification of chondroitin sulfate, another CD44 ligand. Although further studies are needed to test this hypothesis, HASA showed efficient and selective tumor targeting in vivo, including in small subcutaneous tumors and lung metastases (Fig. 5a and e).

Importantly, the tumor-selective delivery achieved by HASA may provide benefits beyond enhanced therapeutic efficacy by mitigating the potential consequences of systemic DPP4 inhibition. In addition to cleaving chemokines such as CXCL10, DPP4 also has non-enzymatic immune functions, and systemic DPP4 inhibition has been reported to increase circulating sDPP4, with potential inflammatory relevance that may be context-dependent [55]. In this context, HASA/Sitag/DOX inhibited DPP4 activity in both tumors (Fig. 7c) and plasma (Fig. S40b), but did not increase circulating sDPP4 levels (Fig. S40), suggesting that HASA-mediated delivery can achieve effective DPP4 enzymatic inhibition without overt systemic sDPP4 perturbation. This may be attributed to the altered exposure pattern enabled by nanocarrier-mediated co-delivery, which favors Sitag accumulation in tumor-associated compartments rather than free systemic distribution [40,41]. Nevertheless, future translational development of HASA will require further evaluation of scalable manufacturing, batch-to-batch reproducibility, and potential immune reactions in humans.

The superior antitumor activity of HASA/Sitag/DOX is likely attributed to both enhanced tumor cell killing and the significantly improved tumor immune microenvironment following NPs-mediated delivery. The triple combination therapy was most effective in increasing the numbers of both total CD8⁺ T cells and functional (IFNγ⁺) CD8⁺ T cells (Fig. S48 and Fig. 7h). Although HASA/DOX treatment led to significant increases in the number of IFNγ⁺ T cells, there was only a modest increase in the total number of CD8⁺ T cells likely due to DOX-mediated DPP4 induction, which limits the magnitude of T cell recruitment into tumors. Similar results have been reported in previous studies with NPs-mediated chemotherapy delivery [51,57,58], highlighting the significance of combining chemotherapy with DPP4 inhibition.

An early study by Barreira da Silva et al. showed that DDP4 inhibition by Sitag effectively synergized with anti-CTLA4-or anti-PD1/anti-CTLA4-based immunotherapy but not anti-PD1 treatment [24]. This is different from TP-100, a pan-DPP inhibitor that strongly synergized with anti-PD1 alone [59]. Mechanistically, effective PD-1 therapy appears to heavily rely on the involvement of CXCL9/CXCR3 but not CXCL10/CXCR3 axis. While Sitag prevents DPP4-mediated truncation of CXCL9/10/11, TP-100 additionally blocks DPP8/9 activity, thereby providing broader protection of CXCL9 in addition to its effect on CXCL10. As a strong ICD inducer, DOX may induce the expression of several T cell-recruiting chemokines including CXCL9, CXCL10, and CXCL11. This, together with the immunomodulating effect of COX-2 inhibition, adds significantly to the overall changes in the tumor immune microenvironment as shown in our RNAseq data (Fig. S56). Therefore HASA/Sitag/DOX may hold potential to combine with anti-PD1 (Fig. 7l–m) or other immunotherapy approaches, which will be further evaluated in the future.

In summary, our study identifies chemotherapy-induced DPP4 upregulation as a previously underappreciated adaptive mechanism that restricts CXCL10 activity and limits T-cell recruitment into tumors. By combining DPP4 inhibition, COX-2 suppression, and ICD-inducing chemotherapy, this work shows that chemotherapy-elicited immune activation can be better preserved and translated into stronger antitumor immunity. These findings provide a mechanistic rationale for combining DPP4 inhibitors with ICD-inducing chemotherapeutic agents and highlight the potential of the HASA nanocarrier platform for improving cancer immunochemotherapy through tumor-directed DPP4 enzymatic inhibition while minimizing overt systemic DPP4-related perturbation.

4. Methods

4.1. Clinical data analysis

The expression levels of DPP4 across various cancer types and their correlation with survival outcomes were analyzed using the GEPIA2 web tool [60]. Clinical data and transcriptome profiles from TCGA-BRCA and TCGA-COAD datasets were utilized for this study. The BRCA-TNBC cohort was identified using immunohistochemistry (IHC) data from breast cancer patients. Patients who had undergone chemotherapy were selected from clinical records, forming the BRCA-TNBC-Chemo (n = 85) and COAD-Chemo (n = 145) subgroups. Primary tumor samples served as the main focus for all analyses.

Transcriptome data, obtained from the Genomic Data Commons (GDC), were annotated according to the human reference genome GRCh38 and quantified as gene-level expression in Fragments Per Kilobase of transcript per Million mapped reads (FPKM), with an expression threshold established at the 50th percentile. To assess potential drug responses, the Cancer Therapeutics Response Portal (CTRP) database was used to predict correlations between DPP4 expression and sensitivity to various anticancer agents.

4.2. Lentiviral infection and RNA interference

The Edit-R mouse DPP4- and CD44-EGFP All-in-one lentiviral sgRNA-CRISPR-Cas9 plasmids were purchased from Horizon Discovery Ltd. (Cambridge, UK). The psPAX2 and pMD2.G plasmids were kindly supplied by a collaborator. The DPP4 (NM_010074) mouse tagged ORF clone lentiviral particle (Myc-DDK-P2A-Puro, MR222697L3V) was purchased from OriGene Technologies, Inc. (MD, USA). The pLentipuro3/TO/V5-GW/EGFP-Firefly Luciferase plasmid was sourced from Addgene (MA, USA). Both murine and human DPP4 siRNA (siDPP4) and control siRNA (siCT) were obtained from Sigma-Aldrich (MO, USA), with their sequences detailed in Table S2.

DPP4 KO cell lines were generated using CRISPR technology. Cells were transduced with lentivirus packaged from LentiCRISPR-sgRHBDF1 plasmid, which expresses EGFP or confers puromycin resistance [61]. The sgRNA sequences targeting RHBDF1 were derived from the GeCKO v2 Mouse CRISPR KO Pooled Library for mRHBDF1. DPP4 RE cell lines were generated according to the lentiviral transduction protocol provided by the manufacturer'.

Successful KO and RE cells were isolated through EGFP⁺ sorting and enriched for the DPP4 KO population or by puromycin selection. Successful editing was also confirmed by measuring DPP4 enzymatic activity (see the DPP4 enzymatic activity section), verifying the expected reduction or restoration of DPP4 activity. A similar procedure was employed to generate CD44 KO cell lines, with KO validation confirmed via Western blot analysis showing the absence of CD44 protein.

For siDPP4-mediated RNA interference, cells were transfected with 40 nM siDPP4 using Lipofectamine™ RNAiMAX (Thermo Fisher Scientific, MA, USA) according to the manufacturer's instructions. For the reporter gene assay, 4T1 cells were transiently transfected with NF-κB reporter constructs that were previously described and validated [62]. Transfection was performed using Lipofectamine™ 3000 (Thermo Fisher Scientific, MA, USA) according to the manufacturer's protocol.

4.3. ELISA assay

Detection of in vitro murine DPP4 was conducted using the DPP4 ELISA Kit (Aviva Systems Biology, CA, USA) following the manufacturer's protocol. To assess chemokine levels in serum, blood samples were collected following various treatments and left to clot for 30 min. Serum was subsequently isolated by centrifugation at 14,000 rpm for 10 min. Serum CXCL10 concentrations were quantified using Duoset ELISA kits (R&D Systems, MN, USA).

4.4. DPP4 enzymatic activity

To assess in vitro DPP4 enzymatic activity, 4T1 cancer cells were treated with various formulations for 24 h, followed by cell digestion to obtain lysates. For evaluating DPP4 activity in blood or tumor homogenates, plasma was collected from mice via bleeding and tumors were excised, weighed, and processed in PBS supplemented with protease inhibitor cocktail (Roche). Soluble fractions were extracted, and DPP4 activity across all samples was measured using the DPPIV-Glo™ Protease Assay (Promega, WI, USA). In 4T1-GFP-luc tumor-bearing mice, treatment was administrated with PBS (CT), HASA/DOX or HASA/Sitag/DOX 24 h prior to an intraperitoneal injection of 10 mM Gly-Pro-aminoluciferin (Promega, WI, USA). Bioluminescence imaging was performed 5 min post-injection using the IVIS 200 system (PerkinElmer, MA, USA).

4.5. Fabrication and physicochemical characterization of drug-loaded micelles

Blank micelles and micelles loaded or co-loaded with drugs (Sitag and/or DOX) were prepared via an ultrasonication method. Specifically, HA-dendrimers or HA-dendrimer-Cy5.5 were dissolved in H₂O or PBS and subjected to ultrasonication (100 W, 20 Hz, 8 s on/2 s off) for 5 min in an ice bath.

For HASA synthesis, dendrimer arms were conjugated to approximately 30% of the carboxyl groups of HA. In the final 5-ASA conjugation step, 5-ASA was added in excess at a molar feed ratio of approximately 3:1 relative to the terminal reactive groups of the dendrimer arms to promote efficient end-group modification. The final 5-ASA content in HASA was approximately 30 wt%. For drug-loaded NPs preparation, because HASA is a polymeric conjugate with a distribution of molecular weights, the formulation ratios for drug loading were reported as mass feed ratios. HASA/Sitag and HASA/DOX were prepared using a HASA mass feed ratio of 10:1. For the co-loaded HASA/Sitag/DOX formulation, Sitag and DOX were added at equal mass, with a HASA:Sitag:DOX mass feed ratio of 10:1:1.

For the preparation of HASAP/siDPP4/DOX micelles, HASA and polylysine-based cationic polymer (POP) were mixed in a H₂O/DMSO solution (50 μg/mL, 1 mL). Next, siRNA (5 μg) in deionized water was introduced to the mixture, followed by vortexing for 10 s. The resulting solution was purified using an Amicon Ultra centrifugal filter at 14,000 rpm for 5 min, with free drugs or siRNA removed by three cycles of deionized water addition and centrifugation. The concentrated micelles were then recovered for further experiments.

Drug loading capacity (DLC) and drug loading efficiency (DLE) of drugs were calculated according to the following equations: DLC (%) = weight of loaded drug/(weight of polymer + input drug) × 100%; DLE (%) = weight of loaded drug/weight of input drug × 100%. Drug concentrations were quantified using a fluorometer or HPLC, and particle sizes were measured via DLS (Nano-ZS, Mal-vern Instruments, Malvern, UK). UV-Vis absorbance spectra of the formulations were recorded with a Nanodrop Spectrophotometer (Thermo Fisher Scientific, MA, USA). The CMC of HASA/Sitag/DOX was evaluated using a DLS-based approach [63].

4.6. Pharmacokinetics and in vivo IVIS imaging

The PK profile of DOX in plasma after administration of HASA/Sitag/DOX was evaluated using HPLC-FLR. female Balb/c mice (n = 3) bearing 4T1 orthotopic breast tumors (∼350 mm³) were administered either free DOX or HASA/Sitag/DOX via tail vein injection at a DOX dose of 5 mg/kg. Blood was drawn into EDTA-containing tubes at specific intervals (5 min, 30 min, 1 h, 4 h, 12 h, 24 h post-injection), followed by centrifugation at 14,000 rpm for 10 min at 4°C to isolate 200 μL of plasma.

For DOX quantification, 100 μL plasma was combined with 50 μL daunorubicin (1 μg/mL in MeOH) as an internal standard, 250 μL of 12 mM phosphoric acid in PBS, and 600 μL of MeOH. The mixture was vortexed for 5 min, then spun at 14,000 rpm for 10 min at 4°C. The resulting supernatant (500 μL) was collected and analyzed by HPLC-FLR, using an elution mixture of 0.1% TFA:MeOH:MeCN (50:25:25), with excitation at 480 nm and emission at 580 nm.

For biodistribution analysis, mice were euthanized 24 h after receiving Cy5.5-conjugated HASA nanocarriers, and their major organs and tumors were harvested. Ex vivo imaging was performed using the IVIS 200 system. To quantify DOX accumulation, tumors and livers were weighed, resuspended in PBS (5 mL/g tissue), and homogenized. A 100 μL aliquot of each homogenate was processed following the same DOX extraction method as for plasma, with the extracted samples subsequently analyzed by HPLC-FLR.

4.7. Evaluation of tumor targeting efficiency of HASA nanocarrier

The tumor-targeting ability of HASA nanocarriers was assessed across multiple subcutaneous tumor models, including 4T1, CT26, Panc02, and MyC-Cap (prostate cancer). Additionally, primary 4T1 tumors of varying sizes were analyzed. For Lung metastasis assessment, Balb/c mice were injected with 4T1-GFP-luc cells (2 × 10⁵) via tail vein to establish a metastatic lung model, with tumor burden in the lungs verified by IVIS imaging prior to biodistribution analysis.

To further evaluate tumor specificity, HASA nanocarrier targeting efficiency was compared in two settings: (1) CD44 WT versus CD44 KO 4T1 tumors (s.c.) in WT Balb/c mice, and (2) CD44 WT versus CD44 KO Panc02 tumors (s.c.) in B6.129(Cg)-Cd44tm1Hbg/J (CD44^−/−) mice. Ex vivo imaging was conducted as previously outlined. Tumor tissues were then cryosectioned, stained with DAPI (to mark cell nuclei) and anti-CD31 antibody (to mark vascular ECs), and analyzed for fluorescence signals using a Keyence BZ-X800 fluorescence microscope.

4.8. Cellular uptake, transwell assay, and cell spheroid penetration

To evaluate cellular uptake, WT and CD44 KO cells were exposed to various Cy5.5-conjugated HA dendrimers, with or without prior pretreatment with free HA polymer. The fluorescence intensity of internalized Cy5.5-conjugated HA dendrimers was subsequently quantified by flow cytometry. The loss of CD44 targeting in OHA-TG₁-ASA was calculated based on the following metrics:

Targeting index (TI) = \frac{{Uptake}_{CD 44 WT}}{{Uptake}_{CD 44 KO}}

Percentage loss of targeting (% Loss) = (1 - \frac{{TI}_{OHA - TG 1 - ASA}}{{TI}_{HA - TG 1 - ASA}}) \times 100 %

For the transwell assay, 0.4 μm pore-size transwell inserts (Corning, #3470) were positioned in 6-well plates to create an in vitro multilayer cell model. WT or CD44 KO 4T1 cells, along with HUVEC cells, were seeded onto the apical surface of the inserts. The apical chamber contained 500 μL of growth medium, while the basolateral chamber held 2 mL [64]. The lower chamber was loaded with blank DMEM supplemented with free Cy5.5, In-ASA-Cy5.5, or HASA-Cy5.5 (1 μg/mL Cy5.5 equivalent). At designated time points following micelle addition, fluorescence intensity in the lower chamber medium was measured to evaluate the permeability of these formulations across the transwell.

To assess the penetration ability of HA dendrimers in 3D tumor models, WT or CD44 KO 4T1 cells were plated at 10,000 cells per well in a Nunclon Sphera 96-well U-bottom plate (Thermo Fisher) within a 6 μg/mL collagen I solution, promoting single spheroid formation per well [65]. After 96 h of incubation, dense spheroids were confirmed by microscopy. These spheroids were then treated with In-ASA-Cy5.5 or HASA-Cy5.5 for 18 h, followed by three gentle saline rinses. Z-stack imaging was performed using laser confocal microscopy (CLSM, FluoView 3000, Olympus, Japan) to assess micelle penetration depth.

4.9. In vivo therapeutic treatment

To evaluate anti-tumor efficacy in 4T1 orthotopic tumors, female Balb/c mice were inoculated with 50 μL of 1 × 10⁶ 4T1 cells into the mammary fat pads. The procedure involved lifting the fourth nipple with sterile tweezers and delivering the cells directly into the fat pad using a syringe needle (BD Biosciences, CA, USA). When tumors grew to approximately 50 mm³, mice were randomized into treatment groups (n = 5) and received intravenous (i.v.) injections of either PBS (CT), HASA, 5-ASA + Sitag + DOX, HASA/Sitag, HASA/DOX, or HASA/Sitag/DOX every three days for five doses (HASA: 50 mg/kg, Sitag: 5 mg/kg, DOX: 5 mg/kg). Tumor dimensions (length, L; width, W) and body weights were recorded every three days, with tumor volume (V) calculated as (L × W²)/2. Mice were observed until natural death or euthanized if tumors reached 2000 mm³, in accordance with ethical standards.

To investigate the potential synergy of HASA/Sitag/DOX with anti-PD-1, treatment began when tumors reached ∼200 mm³. Mice were divided into groups (n = 8) and administered PBS (CT), α-PD-1, HASA/Sitag/DOX, or a combination of HASA/Sitag/DOX + α-PD-1 every three days for five doses (HASA: 50 mg/kg, Sitag: 5 mg/kg, DOX: 5 mg/kg, α-PD-1: 200 μg per dose). Monitoring continued for approximately three months, with euthanasia performed if tumors exceeded 2000 mm³, in accordance with institutional guidelines for maximum tumor size.

4.10. In vivo toxicity study

After concluding the in vivo study, blood was drawn from treated mice, and plasma was separated by centrifugation at 14,000 rpm for 10 min. Hepatic and renal function were evaluated by measuring key biochemical markers, including ALT, AST, and creatinine levels. To establish the MTD of HASA/Sitag/DOX, mice were randomized into groups (n = 8) and administered increasing doses of the formulation. Body weight loss served as the primary indicator of toxicity, with a 10% reduction defined as the threshold [66].

4.11. Analysis of tumor-infiltrating immune cells by flow cytometry

The composition of immune cell populations within tumors following various treatments was assessed using flow cytometry, adhering to established protocols [57]. Briefly, one day after the final treatment, cell suspensions were prepared from spleens or tumors, and red blood cells were lysed. The resulting cells were stained with fluorescently labeled antibodies specific for immune markers, and Zombie dye was used to exclude non-viable cells. Multi-color flow cytometry was performed to assess the infiltration of distinct immune cell subsets and to quantify the production of effector molecules by lymphocytes.

4.12. RNAseq analysis

Tumors were harvested for RNA-seq analysis one day following the final treatment. Sequencing was performed at an institutional core facility [67]. Reads were mapped to the mouse reference genome GRCm39 using STAR [PMID: 23104886]. Gene expression quantification and differential expression were performed by RSEM [PMID: 21816040] and Cuffdiff [PMID: 22383036] respectively. Subsequently, gene set enrichment analysis (GSEA) [68] was conducted to identify treatment-related alterations in functional pathways.

4.13. Statistical analysis

Data are presented as mean ± s.e.m. Statistical significance was assessed using the log-rank (Mantel–Cox) test for survival analyses. Comparisons among three or more groups were performed using one-way ANOVA followed by Tukey's post hoc test, while comparisons between two independent groups were performed using an unpaired two-tailed Student's t-test, as specified in the figure legends. A P value < 0.05 was considered statistically significant. All statistical analyses and graphical representations were generated using GraphPad Prism 10.1.0 (GraphPad Software).

Data availability

Data from the CTRP can be accessed through the Cancer Therapeutics Response Portal (https://portals.broadinstitute.org/ctrp.v2.1/). Data from the TCGA can be accessed through the GDC data portal (https://portal.gdc.cancer.gov/). The bulk messenger RNAseq data mapped to the mouse genome (GRCm38: https://www.ncbi.nlm.nih.gov/assembly/GCF_000001635.20/) are available in the NCBI to Gene Expression Omnibus. The remaining data are available within the Article or Supplementary Information.

Ethics approval and consent to participate

All animal experiments were conducted in accordance with institutional guidelines and were approved by the Institutional Animal Care and Use Committee (IACUC) of the authors’ institution (Protocol No. 24075313). Animals were housed under pathogen-free conditions following AAALAC guidelines, with controlled temperature (22°C), humidity (45%), and a 14/10 h light/dark cycle, and were provided ad libitum access to food and water.

CRediT authorship contribution statement

Shangyu Chen: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Supervision, Validation, Visualization, Writing – original draft, Writing – review & editing. Shichen Li: Data curation, Formal analysis, Investigation, Methodology, Validation, Visualization. Zhangyi Luo: Investigation, Validation. Yixian Huang: Investigation, Validation. Hua Zhang: Investigation, Validation, Visualization. Yiqing Mu: Investigation, Validation, Visualization. Bei Zhang: Methodology, Validation, Visualization. Chien-Yu Chen: Investigation, Methodology. Ganqian Hou: Investigation, Validation. Min Zhang: Investigation, Methodology, Validation. Song Li: Conceptualization, Funding acquisition, Resources, Supervision, Writing – review & editing.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

This work was supported by the fund from National Institute of Health grants R01CA223788, R01CA278608, R01CA270623 (to SL), and the David and Betty Brenneman Scholar Fund (to SL).

Footnotes

Peer review under the responsibility of editorial board of Bioactive Materials.

^{Appendix A}

Supplementary data to this article can be found online at https://doi.org/10.1016/j.bioactmat.2026.07.053.

Appendix A. Supplementary data

The following is the Supplementary data to this article:

Multimedia component 1

mmc1.docx^{(97.3MB, docx)}

References

1.Hegde P.S., Chen D.S. Top 10 challenges in cancer immunotherapy. Immunity. 2020;52:17–35. doi: 10.1016/j.immuni.2019.12.011. [DOI] [PubMed] [Google Scholar]
2.Waldman A.D., Fritz J.M., Lenardo M.J. A guide to cancer immunotherapy: from T cell basic science to clinical practice. Nat. Rev. Immunol. 2020;20:651–668. doi: 10.1038/s41577-020-0306-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
3.Gupta B., Kim J.O. Recent progress in cancer immunotherapy approaches based on nanoparticle delivery devices. J. Pharm. Investig. 2021;51:399–412. [Google Scholar]
4.Meng L., Wu H., Wu J., et al. Mechanisms of immune checkpoint inhibitors: insights into the regulation of circular RNAS involved in cancer hallmarks. Cell Death Dis. 2024;15:3. doi: 10.1038/s41419-023-06389-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
5.Lim W.A., June C.H. The principles of engineering immune cells to treat cancer. Cell. 2017;168:724–740. doi: 10.1016/j.cell.2017.01.016. [DOI] [PMC free article] [PubMed] [Google Scholar]
6.Fan T., Zhang M., Yang J., Zhu Z., Cao W., Dong C. Therapeutic cancer vaccines: advancements, challenges and prospects. Signal Transduct. Targeted Ther. 2023;8:450. doi: 10.1038/s41392-023-01674-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
7.Chulpanova D.S., Kitaeva K.V., Green A.R., Rizvanov A.A., Solovyeva V.V. Molecular aspects and future perspectives of cytokine-based anti-cancer immunotherapy. Front. Cell Dev. Biol. 2020;8:450. doi: 10.3389/fcell.2020.00402. [DOI] [PMC free article] [PubMed] [Google Scholar]
8.Cercek A., Lumish M., Sinopoli J., et al. PD-1 blockade in mismatch repair-deficient, locally advanced rectal cancer. N. Engl. J. Med. 2022;386:2363–2376. doi: 10.1056/NEJMoa2201445. [DOI] [PMC free article] [PubMed] [Google Scholar]
9.Sharma P., Hu-Lieskovan S., Wargo J.A., Ribas A. Primary, adaptive, and acquired resistance to cancer immunotherapy. Cell. 2017;168:707–723. doi: 10.1016/j.cell.2017.01.017. [DOI] [PMC free article] [PubMed] [Google Scholar]
10.Topalian S.L., Taube J.M., Anders R.A., Pardoll D.M. Mechanism-driven biomarkers to guide immune checkpoint blockade in cancer therapy. Nat. Rev. Cancer. 2016;16:275–287. doi: 10.1038/nrc.2016.36. [DOI] [PMC free article] [PubMed] [Google Scholar]
11.Weber M., Hauschild R., Schwarz J., et al. Interstitial dendritic cell guidance by haptotactic chemokine gradients. Science. 2013;339:328–332. doi: 10.1126/science.1228456. [DOI] [PubMed] [Google Scholar]
12.Kroemer G., Galluzzi L., Kepp O., Zitvogel L. Immunogenic cell eeath in cancer therapy. Annu. Rev. Immunol. 2013;31:51–72. doi: 10.1146/annurev-immunol-032712-100008. [DOI] [PubMed] [Google Scholar]
13.Krysko D.V., Garg A.D., Kaczmarek A., Krysko O., Agostinis P., Vandenabeele P. Immunogenic cell death and DAMPs in cancer therapy. Nat. Rev. Cancer. 2012;12:860–875. doi: 10.1038/nrc3380. [DOI] [PubMed] [Google Scholar]
14.Ahmed A., Tait S.W.G. Targeting immunogenic cell death in cancer. Mol. Oncol. 2020;14:2994–3006. doi: 10.1002/1878-0261.12851. [DOI] [PMC free article] [PubMed] [Google Scholar]
15.Galluzzi L., Buqué A., Kepp O., Zitvogel L., Kroemer G. Immunogenic cell death in cancer and infectious disease. Nat. Rev. Immunol. 2017;17:97–111. doi: 10.1038/nri.2016.107. [DOI] [PubMed] [Google Scholar]
16.Alba Dalmases I.G., Menendez Silvia, Arpí Oriol, Corominas Josep Maria, Servitja Sonia, Tusquets Ignasi, Bigas Anna, Eroles Pilar, Furriol Jessica, Lluch Anna, Rovira Ana, Albanell Joan, Rojo Federico. Deficiency in p53 is required for doxorubicin induced transcriptional activation of NF-кB target genes in human breast cancer. Oncotarget. 2014;5:196–210. doi: 10.18632/oncotarget.1556. [DOI] [PMC free article] [PubMed] [Google Scholar]
17.Fucikova J., Kepp O., Kasikova L., et al. Detection of immunogenic cell death and its relevance for cancer therapy. Cell Death Dis. 2020;11:1013. doi: 10.1038/s41419-020-03221-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
18.Huang H., Ji B., Huang Y., et al. Advanced hierarchical computational modeling-based rational development of platinum (II) nanocomplex to improve lung cancer therapy. Adv. Funct. Mater. 2025;35 doi: 10.1002/adfm.202411334. [DOI] [PMC free article] [PubMed] [Google Scholar]
19.Luo Z., Huang Y., Chen S., et al. Delivery of PARP inhibitors through 2HG-incorporated liposomes for synergistically targeting DNA repair in cancer. Cancer Lett. 2024;604 doi: 10.1016/j.canlet.2024.217268. [DOI] [PubMed] [Google Scholar]
20.Bongers J., Lambros T., Ahmad M., Heimer E.P. Kinetics of dipeptidyl peptidase IV proteolysis of growth hormone-releasing factor and analogs. Biochim. Biophys. Acta. 1992;1122:147–153. doi: 10.1016/0167-4838(92)90317-7. [DOI] [PubMed] [Google Scholar]
21.Klemann C., Wagner L., Stephan M., von Hörsten S. Cut to the chase: a review of CD26/dipeptidyl peptidase-4's (DPP4) entanglement in the immune system. Clin. Exp. Immunol. 2016;185:1–21. doi: 10.1111/cei.12781. [DOI] [PMC free article] [PubMed] [Google Scholar]
22.Wang K., Zhang Z., Hang J., et al. Microbial-host-isozyme analyses reveal microbial DPP4 as a potential antidiabetic target. Science. 2023;381:eadd5787. doi: 10.1126/science.add5787. [DOI] [PubMed] [Google Scholar]
23.Mortier A., Van Damme J., Proost P. Overview of the mechanisms regulating chemokine activity and availability. Immunol. Lett. 2012;145:2–9. doi: 10.1016/j.imlet.2012.04.015. [DOI] [PubMed] [Google Scholar]
24.Barreira da Silva R., Laird M.E., Yatim N., Fiette L., Ingersoll M.A., Albert M.L. Dipeptidylpeptidase 4 inhibition enhances lymphocyte trafficking, improving both naturally occurring tumor immunity and immunotherapy. Nat. Immunol. 2015;16:850–858. doi: 10.1038/ni.3201. [DOI] [PubMed] [Google Scholar]
25.Baggio L.L., Varin E.M., Koehler J.A., et al. Plasma levels of DPP4 activity and sDPP4 are dissociated from inflammation in mice and humans. Nat. Commun. 2020;11:3766. doi: 10.1038/s41467-020-17556-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
26.Riva A., Laird M.E., Casrouge A., et al. Truncated CXCL10 is associated with failure to achieve spontaneous clearance of acute hepatitis C infection. Hepatology. 2014;60 doi: 10.1002/hep.27139. [DOI] [PubMed] [Google Scholar]
27.Casrouge A., Decalf J., Ahloulay M., et al. Evidence for an antagonist form of the chemokine CXCL10 in patients chronically infected with HCV. J. Clin. Investig. 2011;121:308–317. doi: 10.1172/JCI40594. [DOI] [PMC free article] [PubMed] [Google Scholar]
28.Jaiswal P., Tripathi V., Assaiya A., et al. Anti-cancer effects of sitagliptin, vildagliptin, and exendin-4 on triple-negative breast cancer cells via mitochondrial modulation. Biocell. 2022;46:2645–2657. doi: 10.2174/1568009622666220207101212. [DOI] [PubMed] [Google Scholar]
29.FitzGerald G.A. COX-2 and beyond: approaches to prostaglandin inhibition in human disease. Nat. Rev. Drug Discov. 2003;2:879–890. doi: 10.1038/nrd1225. [DOI] [PubMed] [Google Scholar]
30.Bell C.R., Pelly V.S., Moeini A., et al. Chemotherapy-induced COX-2 upregulation by cancer cells defines their inflammatory properties and limits the efficacy of chemoimmunotherapy combinations. Nat. Commun. 2022;13:2063. doi: 10.1038/s41467-022-29606-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
31.Lacher S.B., Dörr J., de Almeida G.P., et al. PGE2 limits effector expansion of tumour-infiltrating stem-like CD8+ T cells. Nature. 2024;629:417–425. doi: 10.1038/s41586-024-07254-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
32.Hashemi Goradel N., Najafi M., Salehi E., Farhood B., Mortezaee K. Cyclooxygenase-2 in cancer: a review. J. Cell. Physiol. 2019;234:5683–5699. doi: 10.1002/jcp.27411. [DOI] [PubMed] [Google Scholar]
33.Liu C., Xu J., Fan J., Liu C., Xie W., Kong H. DPP-4 exacerbates LPS-induced endothelial cells inflammation via integrin-α5β1/FAK/AKT signaling. Exp. Cell Res. 2024;435 doi: 10.1016/j.yexcr.2023.113909. [DOI] [PubMed] [Google Scholar]
34.Li R., Zeng X., Yang M., et al. Antidiabetic DPP-4 inhibitors reprogram tumor microenvironment that facilitates murine breast cancer metastasis through interaction eith cancer cells via a ROS-NF-кB-NLRP3 axis. Front. Oncol. 2021;11 doi: 10.3389/fonc.2021.728047. [DOI] [PMC free article] [PubMed] [Google Scholar]
35.Beiranvand M. A review of the biological and pharmacological activities of mesalazine or 5-aminosalicylic acid (5-ASA): an anti-ulcer and anti-oxidant drug. Inflammopharmacology. 2021;29:1279–1290. doi: 10.1007/s10787-021-00856-1. [DOI] [PubMed] [Google Scholar]
36.Wei T., Chen C., Liu J., et al. Anticancer drug nanomicelles formed by self-assembling amphiphilic dendrimer to combat cancer drug resistance. Proc. Natl. Acad. Sci. 2015;112:2978–2983. doi: 10.1073/pnas.1418494112. [DOI] [PMC free article] [PubMed] [Google Scholar]
37.Scott R.W.J., Wilson O.M., Crooks R.M. Synthesis, characterization, and applications of dendrimer-encapsulated nanoparticles. J. Phys. Chem. B. 2005;109:692–704. doi: 10.1021/jp0469665. [DOI] [PubMed] [Google Scholar]
38.Boas U., Heegaard P.M.H. Dendrimers in drug research. Chem. Soc. Rev. 2004;33:43–63. doi: 10.1039/b309043b. [DOI] [PubMed] [Google Scholar]
39.Chen S., Miao H., Jiang X., Sun P., Fan Q., Huang W. Starlike polymer brush-based ultrasmall nanoparticles with simultaneously improved NIR-II fluorescence and blood circulation for efficient orthotopic glioblastoma imaging. Biomaterials. 2021;275 doi: 10.1016/j.biomaterials.2021.120916. [DOI] [PubMed] [Google Scholar]
40.Chen S., Pan Y., Chen K., et al. Increasing molecular planarity through donor/side-chain engineering for improved NIR-IIa fluorescence imaging and NIR-II photothermal therapy under 1064 nm. Angew. Chem. Int. Ed. 2023;62 doi: 10.1002/anie.202215372. [DOI] [PubMed] [Google Scholar]
41.Luo Z., Chen C.-Y., Li S. Improving tumor targeting and penetration for nanoparticle-mediated cancer therapy. Small Methods. 2025;9 doi: 10.1002/smtd.202401860. [DOI] [PubMed] [Google Scholar]
42.Ganesh S., Iyer A.K., Morrissey D.V., Amiji M.M. Hyaluronic acid based self-assembling nanosystems for CD44 target mediated siRNA delivery to solid tumors. Biomaterials. 2013;34:3489–3502. doi: 10.1016/j.biomaterials.2013.01.077. [DOI] [PMC free article] [PubMed] [Google Scholar]
43.Ganesh S., Iyer A.K., Gattacceca F., Morrissey D.V., Amiji M.M. In vivo biodistribution of siRNA and cisplatin administered using CD44-targeted hyaluronic acid nanoparticles. J. Contr. Release. 2013;172:699–706. doi: 10.1016/j.jconrel.2013.10.016. [DOI] [PMC free article] [PubMed] [Google Scholar]
44.Chen S., Zhang W., Jiang X., et al. Bright NIR-II fluorescent small-molecule nanoparticles with reduced intermolecular interaction for targeted in vivo inflammation imaging. ACS Appl. Polym. Mater. 2021;3:5236–5242. [Google Scholar]
45.Bhattacharya D.S., Svechkarev D., Souchek J.J., et al. Impact of structurally modifying hyaluronic acid on CD44 interaction. J. Mater. Chem. B. 2017;5:8183–8192. doi: 10.1039/C7TB01895A. [DOI] [PMC free article] [PubMed] [Google Scholar]
46.Kwon M.Y., Wang C., Galarraga J.H., Puré E., Han L., Burdick J.A. Influence of hyaluronic acid modification on CD44 binding towards the design of hydrogel biomaterials. Biomaterials. 2019;222 doi: 10.1016/j.biomaterials.2019.119451. [DOI] [PMC free article] [PubMed] [Google Scholar]
47.Chen S., Sun B., Miao H., et al. NIR-II dye-based multifunctional telechelic glycopolymers for NIR-IIa fluorescence imaging-guided stimuli-responsive chemo-photothermal combination therapy. ACS Mater. Lett. 2020;2:174–183. [Google Scholar]
48.Zhong L., Liu Y., Xu L., et al. Exploring the relationship of hyaluronic acid molecular weight and active targeting efficiency for designing hyaluronic acid-modified nanoparticles. Asian J. Pharm. Sci. 2019;14:521–530. doi: 10.1016/j.ajps.2018.11.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
49.Della Sala F., Silvestri T., Borzacchiello A., Mayol L., Ambrosio L., Biondi M. Hyaluronan-coated nanoparticles for active tumor targeting: influence of polysaccharide molecular weight on cell uptake. Colloids Surf. B Biointerfaces. 2022;210 doi: 10.1016/j.colsurfb.2021.112240. [DOI] [PubMed] [Google Scholar]
50.Sun P., Hu D., Chen P., et al. Anti-quenching NIR-II excitation phenylboronic acid modified conjugated polyelectrolyte for intracellular peroxynitrite-enhanced chemo-photothermal therapy. Adv. Sci. 2024;11 doi: 10.1002/advs.202309446. [DOI] [PMC free article] [PubMed] [Google Scholar]
51.Chen Y., Huang Y., Li Q., et al. Targeting Xkr8 via nanoparticle-mediated in situ co-delivery of siRNA and chemotherapy drugs for cancer immunochemotherapy. Nat. Nanotechnol. 2023;18:193–204. doi: 10.1038/s41565-022-01266-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
52.Luo Z., Huang Y., Batra N., et al. Inhibition of iRhom1 by CD44-targeting nanocarrier for improved cancer immunochemotherapy. Nat. Commun. 2024;15:255. doi: 10.1038/s41467-023-44572-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
53.Chen S., Zhou P., Yang W., et al. Stealth-like polysarcosine-modified nanoparticles with low dye doses and long blood circulation for efficient breast cancer pulmonary metastasis imaging. Biomater. Sci. 2023;11:4644–4651. doi: 10.1039/d3bm00200d. [DOI] [PubMed] [Google Scholar]
54.Griffioen A.W., Coenen M.J.H., Damen C.A., et al. CD44 is involved in tumor angiogenesis; an activation antigen on human endothelial cells. Blood. 1997;90:1150–1159. [PubMed] [Google Scholar]
55.Varin E.M., Mulvihill E.E., Beaudry J.L., et al. Circulating levels of soluble dipeptidyl peptidase-4 are dissociated from inflammation and induced by enzymatic DPP4 inhibition. Cell Metab. 2019;29:320–334.e325. doi: 10.1016/j.cmet.2018.10.001. [DOI] [PubMed] [Google Scholar]
56.Ng S.-C., Kao S.-W., Chen C.J., et al. HIF-1α attenuates ferroptosis-associated dermal fibroblast senescence via modulation of NF-κB-DPP4 signaling. J. Transl. Med. 2026 doi: 10.1186/s12967-026-08445-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
57.Chen Y., Xia R., Huang Y., et al. An immunostimulatory dual-functional nanocarrier that improves cancer immunochemotherapy. Nat. Commun. 2016;7 doi: 10.1038/ncomms13443. [DOI] [PMC free article] [PubMed] [Google Scholar]
58.Huang H., Huang Y., Chen Y., et al. A novel immunochemotherapy based on targeting of cyclooxygenase and induction of immunogenic cell death. Biomaterials. 2021;270 doi: 10.1016/j.biomaterials.2021.120708. [DOI] [PMC free article] [PubMed] [Google Scholar]
59.Adams S., Miller G.T., Jesson M.I., Watanabe T., Jones B., Wallner B.P. PT-100, a small molecule dipeptidyl peptidase inhibitor, has potent antitumor effects and augments antibody-mediated cytotoxicity via a novel immune mechanism. Cancer Res. 2004;64:5471–5480. doi: 10.1158/0008-5472.CAN-04-0447. [DOI] [PubMed] [Google Scholar]
60.Tang Z., Kang B., Li C., Chen T., Zhang Z. GEPIA2: an enhanced web server for large-scale expression profiling and interactive analysis. Nucleic Acids Res. 2019;47:W556–W560. doi: 10.1093/nar/gkz430. [DOI] [PMC free article] [PubMed] [Google Scholar]
61.Sanjana N.E., Shalem O., Zhang F. Improved vectors and genome-wide libraries for CRISPR screening. Nat. Methods. 2014;11:783–784. doi: 10.1038/nmeth.3047. [DOI] [PMC free article] [PubMed] [Google Scholar]
62.Chai X., Guo Y., Jiang M., et al. Oestrogen sulfotransferase ablation sensitizes mice to sepsis. Nat. Commun. 2015;6:7979. doi: 10.1038/ncomms8979. [DOI] [PMC free article] [PubMed] [Google Scholar]
63.Yongvongsoontorn N., Chung J.E., Gao S.J., et al. Carrier-enhanced anticancer efficacy of sunitinib-loaded green tea-based micellar nanocomplex beyond tumor-targeted delivery. ACS Nano. 2019;13:7591–7602. doi: 10.1021/acsnano.9b00467. [DOI] [PubMed] [Google Scholar]
64.Lu W., Xiong C., Zhang R., et al. Receptor-mediated transcytosis: a mechanism for active extravascular transport of nanoparticles in solid tumors. J. Contr. Release. 2012;161:959–966. doi: 10.1016/j.jconrel.2012.05.014. [DOI] [PMC free article] [PubMed] [Google Scholar]
65.Cong Z., Zhang L., Ma S.-Q., Lam K.S., Yang F.-F., Liao Y.-H. Size-transformable hyaluronan stacked self-assembling peptide nanoparticles for improved transcellular tumor penetration and photo-chemo combination therapy. ACS Nano. 2020;14:1958–1970. doi: 10.1021/acsnano.9b08434. [DOI] [PubMed] [Google Scholar]
66.van Berlo D., Woutersen M., Muller A., Pronk M., Vriend J., Hakkert B. 10% body weight (gain) change as criterion for the maximum tolerated dose: a critical analysis. Regul. Toxicol. Pharmacol. 2022;134 doi: 10.1016/j.yrtph.2022.105235. [DOI] [PubMed] [Google Scholar]
67.Pertea M., Kim D., Pertea G.M., Leek J.T., Salzberg S.L. Transcript-level expression analysis of RNA-seq experiments with HISAT, StringTie and Ballgown. Nat. Protoc. 2016;11:1650–1667. doi: 10.1038/nprot.2016.095. [DOI] [PMC free article] [PubMed] [Google Scholar]
68.Subramanian A., Tamayo P., Mootha V.K., et al. Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles. Proc. Natl. Acad. Sci. 2005;102:15545–15550. doi: 10.1073/pnas.0506580102. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Multimedia component 1

mmc1.docx^{(97.3MB, docx)}

Data Availability Statement

[bib1] 1.Hegde P.S., Chen D.S. Top 10 challenges in cancer immunotherapy. Immunity. 2020;52:17–35. doi: 10.1016/j.immuni.2019.12.011. [DOI] [PubMed] [Google Scholar]

[bib2] 2.Waldman A.D., Fritz J.M., Lenardo M.J. A guide to cancer immunotherapy: from T cell basic science to clinical practice. Nat. Rev. Immunol. 2020;20:651–668. doi: 10.1038/s41577-020-0306-5. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib3] 3.Gupta B., Kim J.O. Recent progress in cancer immunotherapy approaches based on nanoparticle delivery devices. J. Pharm. Investig. 2021;51:399–412. [Google Scholar]

[bib4] 4.Meng L., Wu H., Wu J., et al. Mechanisms of immune checkpoint inhibitors: insights into the regulation of circular RNAS involved in cancer hallmarks. Cell Death Dis. 2024;15:3. doi: 10.1038/s41419-023-06389-5. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib5] 5.Lim W.A., June C.H. The principles of engineering immune cells to treat cancer. Cell. 2017;168:724–740. doi: 10.1016/j.cell.2017.01.016. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib6] 6.Fan T., Zhang M., Yang J., Zhu Z., Cao W., Dong C. Therapeutic cancer vaccines: advancements, challenges and prospects. Signal Transduct. Targeted Ther. 2023;8:450. doi: 10.1038/s41392-023-01674-3. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib7] 7.Chulpanova D.S., Kitaeva K.V., Green A.R., Rizvanov A.A., Solovyeva V.V. Molecular aspects and future perspectives of cytokine-based anti-cancer immunotherapy. Front. Cell Dev. Biol. 2020;8:450. doi: 10.3389/fcell.2020.00402. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib8] 8.Cercek A., Lumish M., Sinopoli J., et al. PD-1 blockade in mismatch repair-deficient, locally advanced rectal cancer. N. Engl. J. Med. 2022;386:2363–2376. doi: 10.1056/NEJMoa2201445. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib9] 9.Sharma P., Hu-Lieskovan S., Wargo J.A., Ribas A. Primary, adaptive, and acquired resistance to cancer immunotherapy. Cell. 2017;168:707–723. doi: 10.1016/j.cell.2017.01.017. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib10] 10.Topalian S.L., Taube J.M., Anders R.A., Pardoll D.M. Mechanism-driven biomarkers to guide immune checkpoint blockade in cancer therapy. Nat. Rev. Cancer. 2016;16:275–287. doi: 10.1038/nrc.2016.36. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib11] 11.Weber M., Hauschild R., Schwarz J., et al. Interstitial dendritic cell guidance by haptotactic chemokine gradients. Science. 2013;339:328–332. doi: 10.1126/science.1228456. [DOI] [PubMed] [Google Scholar]

[bib12] 12.Kroemer G., Galluzzi L., Kepp O., Zitvogel L. Immunogenic cell eeath in cancer therapy. Annu. Rev. Immunol. 2013;31:51–72. doi: 10.1146/annurev-immunol-032712-100008. [DOI] [PubMed] [Google Scholar]

[bib13] 13.Krysko D.V., Garg A.D., Kaczmarek A., Krysko O., Agostinis P., Vandenabeele P. Immunogenic cell death and DAMPs in cancer therapy. Nat. Rev. Cancer. 2012;12:860–875. doi: 10.1038/nrc3380. [DOI] [PubMed] [Google Scholar]

[bib14] 14.Ahmed A., Tait S.W.G. Targeting immunogenic cell death in cancer. Mol. Oncol. 2020;14:2994–3006. doi: 10.1002/1878-0261.12851. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib15] 15.Galluzzi L., Buqué A., Kepp O., Zitvogel L., Kroemer G. Immunogenic cell death in cancer and infectious disease. Nat. Rev. Immunol. 2017;17:97–111. doi: 10.1038/nri.2016.107. [DOI] [PubMed] [Google Scholar]

[bib16] 16.Alba Dalmases I.G., Menendez Silvia, Arpí Oriol, Corominas Josep Maria, Servitja Sonia, Tusquets Ignasi, Bigas Anna, Eroles Pilar, Furriol Jessica, Lluch Anna, Rovira Ana, Albanell Joan, Rojo Federico. Deficiency in p53 is required for doxorubicin induced transcriptional activation of NF-кB target genes in human breast cancer. Oncotarget. 2014;5:196–210. doi: 10.18632/oncotarget.1556. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib17] 17.Fucikova J., Kepp O., Kasikova L., et al. Detection of immunogenic cell death and its relevance for cancer therapy. Cell Death Dis. 2020;11:1013. doi: 10.1038/s41419-020-03221-2. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib18] 18.Huang H., Ji B., Huang Y., et al. Advanced hierarchical computational modeling-based rational development of platinum (II) nanocomplex to improve lung cancer therapy. Adv. Funct. Mater. 2025;35 doi: 10.1002/adfm.202411334. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib19] 19.Luo Z., Huang Y., Chen S., et al. Delivery of PARP inhibitors through 2HG-incorporated liposomes for synergistically targeting DNA repair in cancer. Cancer Lett. 2024;604 doi: 10.1016/j.canlet.2024.217268. [DOI] [PubMed] [Google Scholar]

[bib20] 20.Bongers J., Lambros T., Ahmad M., Heimer E.P. Kinetics of dipeptidyl peptidase IV proteolysis of growth hormone-releasing factor and analogs. Biochim. Biophys. Acta. 1992;1122:147–153. doi: 10.1016/0167-4838(92)90317-7. [DOI] [PubMed] [Google Scholar]

[bib21] 21.Klemann C., Wagner L., Stephan M., von Hörsten S. Cut to the chase: a review of CD26/dipeptidyl peptidase-4's (DPP4) entanglement in the immune system. Clin. Exp. Immunol. 2016;185:1–21. doi: 10.1111/cei.12781. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib22] 22.Wang K., Zhang Z., Hang J., et al. Microbial-host-isozyme analyses reveal microbial DPP4 as a potential antidiabetic target. Science. 2023;381:eadd5787. doi: 10.1126/science.add5787. [DOI] [PubMed] [Google Scholar]

[bib23] 23.Mortier A., Van Damme J., Proost P. Overview of the mechanisms regulating chemokine activity and availability. Immunol. Lett. 2012;145:2–9. doi: 10.1016/j.imlet.2012.04.015. [DOI] [PubMed] [Google Scholar]

[bib24] 24.Barreira da Silva R., Laird M.E., Yatim N., Fiette L., Ingersoll M.A., Albert M.L. Dipeptidylpeptidase 4 inhibition enhances lymphocyte trafficking, improving both naturally occurring tumor immunity and immunotherapy. Nat. Immunol. 2015;16:850–858. doi: 10.1038/ni.3201. [DOI] [PubMed] [Google Scholar]

[bib25] 25.Baggio L.L., Varin E.M., Koehler J.A., et al. Plasma levels of DPP4 activity and sDPP4 are dissociated from inflammation in mice and humans. Nat. Commun. 2020;11:3766. doi: 10.1038/s41467-020-17556-z. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib26] 26.Riva A., Laird M.E., Casrouge A., et al. Truncated CXCL10 is associated with failure to achieve spontaneous clearance of acute hepatitis C infection. Hepatology. 2014;60 doi: 10.1002/hep.27139. [DOI] [PubMed] [Google Scholar]

[bib27] 27.Casrouge A., Decalf J., Ahloulay M., et al. Evidence for an antagonist form of the chemokine CXCL10 in patients chronically infected with HCV. J. Clin. Investig. 2011;121:308–317. doi: 10.1172/JCI40594. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib28] 28.Jaiswal P., Tripathi V., Assaiya A., et al. Anti-cancer effects of sitagliptin, vildagliptin, and exendin-4 on triple-negative breast cancer cells via mitochondrial modulation. Biocell. 2022;46:2645–2657. doi: 10.2174/1568009622666220207101212. [DOI] [PubMed] [Google Scholar]

[bib29] 29.FitzGerald G.A. COX-2 and beyond: approaches to prostaglandin inhibition in human disease. Nat. Rev. Drug Discov. 2003;2:879–890. doi: 10.1038/nrd1225. [DOI] [PubMed] [Google Scholar]

[bib30] 30.Bell C.R., Pelly V.S., Moeini A., et al. Chemotherapy-induced COX-2 upregulation by cancer cells defines their inflammatory properties and limits the efficacy of chemoimmunotherapy combinations. Nat. Commun. 2022;13:2063. doi: 10.1038/s41467-022-29606-9. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib31] 31.Lacher S.B., Dörr J., de Almeida G.P., et al. PGE2 limits effector expansion of tumour-infiltrating stem-like CD8+ T cells. Nature. 2024;629:417–425. doi: 10.1038/s41586-024-07254-x. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib32] 32.Hashemi Goradel N., Najafi M., Salehi E., Farhood B., Mortezaee K. Cyclooxygenase-2 in cancer: a review. J. Cell. Physiol. 2019;234:5683–5699. doi: 10.1002/jcp.27411. [DOI] [PubMed] [Google Scholar]

[bib33] 33.Liu C., Xu J., Fan J., Liu C., Xie W., Kong H. DPP-4 exacerbates LPS-induced endothelial cells inflammation via integrin-α5β1/FAK/AKT signaling. Exp. Cell Res. 2024;435 doi: 10.1016/j.yexcr.2023.113909. [DOI] [PubMed] [Google Scholar]

[bib34] 34.Li R., Zeng X., Yang M., et al. Antidiabetic DPP-4 inhibitors reprogram tumor microenvironment that facilitates murine breast cancer metastasis through interaction eith cancer cells via a ROS-NF-кB-NLRP3 axis. Front. Oncol. 2021;11 doi: 10.3389/fonc.2021.728047. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib35] 35.Beiranvand M. A review of the biological and pharmacological activities of mesalazine or 5-aminosalicylic acid (5-ASA): an anti-ulcer and anti-oxidant drug. Inflammopharmacology. 2021;29:1279–1290. doi: 10.1007/s10787-021-00856-1. [DOI] [PubMed] [Google Scholar]

[bib36] 36.Wei T., Chen C., Liu J., et al. Anticancer drug nanomicelles formed by self-assembling amphiphilic dendrimer to combat cancer drug resistance. Proc. Natl. Acad. Sci. 2015;112:2978–2983. doi: 10.1073/pnas.1418494112. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib37] 37.Scott R.W.J., Wilson O.M., Crooks R.M. Synthesis, characterization, and applications of dendrimer-encapsulated nanoparticles. J. Phys. Chem. B. 2005;109:692–704. doi: 10.1021/jp0469665. [DOI] [PubMed] [Google Scholar]

[bib38] 38.Boas U., Heegaard P.M.H. Dendrimers in drug research. Chem. Soc. Rev. 2004;33:43–63. doi: 10.1039/b309043b. [DOI] [PubMed] [Google Scholar]

[bib39] 39.Chen S., Miao H., Jiang X., Sun P., Fan Q., Huang W. Starlike polymer brush-based ultrasmall nanoparticles with simultaneously improved NIR-II fluorescence and blood circulation for efficient orthotopic glioblastoma imaging. Biomaterials. 2021;275 doi: 10.1016/j.biomaterials.2021.120916. [DOI] [PubMed] [Google Scholar]

[bib40] 40.Chen S., Pan Y., Chen K., et al. Increasing molecular planarity through donor/side-chain engineering for improved NIR-IIa fluorescence imaging and NIR-II photothermal therapy under 1064 nm. Angew. Chem. Int. Ed. 2023;62 doi: 10.1002/anie.202215372. [DOI] [PubMed] [Google Scholar]

[bib41] 41.Luo Z., Chen C.-Y., Li S. Improving tumor targeting and penetration for nanoparticle-mediated cancer therapy. Small Methods. 2025;9 doi: 10.1002/smtd.202401860. [DOI] [PubMed] [Google Scholar]

[bib42] 42.Ganesh S., Iyer A.K., Morrissey D.V., Amiji M.M. Hyaluronic acid based self-assembling nanosystems for CD44 target mediated siRNA delivery to solid tumors. Biomaterials. 2013;34:3489–3502. doi: 10.1016/j.biomaterials.2013.01.077. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib43] 43.Ganesh S., Iyer A.K., Gattacceca F., Morrissey D.V., Amiji M.M. In vivo biodistribution of siRNA and cisplatin administered using CD44-targeted hyaluronic acid nanoparticles. J. Contr. Release. 2013;172:699–706. doi: 10.1016/j.jconrel.2013.10.016. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib44] 44.Chen S., Zhang W., Jiang X., et al. Bright NIR-II fluorescent small-molecule nanoparticles with reduced intermolecular interaction for targeted in vivo inflammation imaging. ACS Appl. Polym. Mater. 2021;3:5236–5242. [Google Scholar]

[bib45] 45.Bhattacharya D.S., Svechkarev D., Souchek J.J., et al. Impact of structurally modifying hyaluronic acid on CD44 interaction. J. Mater. Chem. B. 2017;5:8183–8192. doi: 10.1039/C7TB01895A. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib46] 46.Kwon M.Y., Wang C., Galarraga J.H., Puré E., Han L., Burdick J.A. Influence of hyaluronic acid modification on CD44 binding towards the design of hydrogel biomaterials. Biomaterials. 2019;222 doi: 10.1016/j.biomaterials.2019.119451. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib47] 47.Chen S., Sun B., Miao H., et al. NIR-II dye-based multifunctional telechelic glycopolymers for NIR-IIa fluorescence imaging-guided stimuli-responsive chemo-photothermal combination therapy. ACS Mater. Lett. 2020;2:174–183. [Google Scholar]

[bib48] 48.Zhong L., Liu Y., Xu L., et al. Exploring the relationship of hyaluronic acid molecular weight and active targeting efficiency for designing hyaluronic acid-modified nanoparticles. Asian J. Pharm. Sci. 2019;14:521–530. doi: 10.1016/j.ajps.2018.11.002. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib49] 49.Della Sala F., Silvestri T., Borzacchiello A., Mayol L., Ambrosio L., Biondi M. Hyaluronan-coated nanoparticles for active tumor targeting: influence of polysaccharide molecular weight on cell uptake. Colloids Surf. B Biointerfaces. 2022;210 doi: 10.1016/j.colsurfb.2021.112240. [DOI] [PubMed] [Google Scholar]

[bib50] 50.Sun P., Hu D., Chen P., et al. Anti-quenching NIR-II excitation phenylboronic acid modified conjugated polyelectrolyte for intracellular peroxynitrite-enhanced chemo-photothermal therapy. Adv. Sci. 2024;11 doi: 10.1002/advs.202309446. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib51] 51.Chen Y., Huang Y., Li Q., et al. Targeting Xkr8 via nanoparticle-mediated in situ co-delivery of siRNA and chemotherapy drugs for cancer immunochemotherapy. Nat. Nanotechnol. 2023;18:193–204. doi: 10.1038/s41565-022-01266-2. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib52] 52.Luo Z., Huang Y., Batra N., et al. Inhibition of iRhom1 by CD44-targeting nanocarrier for improved cancer immunochemotherapy. Nat. Commun. 2024;15:255. doi: 10.1038/s41467-023-44572-6. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib53] 53.Chen S., Zhou P., Yang W., et al. Stealth-like polysarcosine-modified nanoparticles with low dye doses and long blood circulation for efficient breast cancer pulmonary metastasis imaging. Biomater. Sci. 2023;11:4644–4651. doi: 10.1039/d3bm00200d. [DOI] [PubMed] [Google Scholar]

[bib54] 54.Griffioen A.W., Coenen M.J.H., Damen C.A., et al. CD44 is involved in tumor angiogenesis; an activation antigen on human endothelial cells. Blood. 1997;90:1150–1159. [PubMed] [Google Scholar]

[bib55] 55.Varin E.M., Mulvihill E.E., Beaudry J.L., et al. Circulating levels of soluble dipeptidyl peptidase-4 are dissociated from inflammation and induced by enzymatic DPP4 inhibition. Cell Metab. 2019;29:320–334.e325. doi: 10.1016/j.cmet.2018.10.001. [DOI] [PubMed] [Google Scholar]

[bib56] 56.Ng S.-C., Kao S.-W., Chen C.J., et al. HIF-1α attenuates ferroptosis-associated dermal fibroblast senescence via modulation of NF-κB-DPP4 signaling. J. Transl. Med. 2026 doi: 10.1186/s12967-026-08445-y. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib57] 57.Chen Y., Xia R., Huang Y., et al. An immunostimulatory dual-functional nanocarrier that improves cancer immunochemotherapy. Nat. Commun. 2016;7 doi: 10.1038/ncomms13443. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib58] 58.Huang H., Huang Y., Chen Y., et al. A novel immunochemotherapy based on targeting of cyclooxygenase and induction of immunogenic cell death. Biomaterials. 2021;270 doi: 10.1016/j.biomaterials.2021.120708. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib59] 59.Adams S., Miller G.T., Jesson M.I., Watanabe T., Jones B., Wallner B.P. PT-100, a small molecule dipeptidyl peptidase inhibitor, has potent antitumor effects and augments antibody-mediated cytotoxicity via a novel immune mechanism. Cancer Res. 2004;64:5471–5480. doi: 10.1158/0008-5472.CAN-04-0447. [DOI] [PubMed] [Google Scholar]

[bib60] 60.Tang Z., Kang B., Li C., Chen T., Zhang Z. GEPIA2: an enhanced web server for large-scale expression profiling and interactive analysis. Nucleic Acids Res. 2019;47:W556–W560. doi: 10.1093/nar/gkz430. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib61] 61.Sanjana N.E., Shalem O., Zhang F. Improved vectors and genome-wide libraries for CRISPR screening. Nat. Methods. 2014;11:783–784. doi: 10.1038/nmeth.3047. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib62] 62.Chai X., Guo Y., Jiang M., et al. Oestrogen sulfotransferase ablation sensitizes mice to sepsis. Nat. Commun. 2015;6:7979. doi: 10.1038/ncomms8979. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib63] 63.Yongvongsoontorn N., Chung J.E., Gao S.J., et al. Carrier-enhanced anticancer efficacy of sunitinib-loaded green tea-based micellar nanocomplex beyond tumor-targeted delivery. ACS Nano. 2019;13:7591–7602. doi: 10.1021/acsnano.9b00467. [DOI] [PubMed] [Google Scholar]

[bib64] 64.Lu W., Xiong C., Zhang R., et al. Receptor-mediated transcytosis: a mechanism for active extravascular transport of nanoparticles in solid tumors. J. Contr. Release. 2012;161:959–966. doi: 10.1016/j.jconrel.2012.05.014. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib65] 65.Cong Z., Zhang L., Ma S.-Q., Lam K.S., Yang F.-F., Liao Y.-H. Size-transformable hyaluronan stacked self-assembling peptide nanoparticles for improved transcellular tumor penetration and photo-chemo combination therapy. ACS Nano. 2020;14:1958–1970. doi: 10.1021/acsnano.9b08434. [DOI] [PubMed] [Google Scholar]

[bib66] 66.van Berlo D., Woutersen M., Muller A., Pronk M., Vriend J., Hakkert B. 10% body weight (gain) change as criterion for the maximum tolerated dose: a critical analysis. Regul. Toxicol. Pharmacol. 2022;134 doi: 10.1016/j.yrtph.2022.105235. [DOI] [PubMed] [Google Scholar]

[bib67] 67.Pertea M., Kim D., Pertea G.M., Leek J.T., Salzberg S.L. Transcript-level expression analysis of RNA-seq experiments with HISAT, StringTie and Ballgown. Nat. Protoc. 2016;11:1650–1667. doi: 10.1038/nprot.2016.095. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib68] 68.Subramanian A., Tamayo P., Mootha V.K., et al. Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles. Proc. Natl. Acad. Sci. 2005;102:15545–15550. doi: 10.1073/pnas.0506580102. [DOI] [PMC free article] [PubMed] [Google Scholar]

PERMALINK

Nanocarrier-mediated targeting of chemotherapy-induced DPP4 enhances T cell infiltration and improves cancer immunochemotherapy

Shangyu Chen

Shichen Li

Zhangyi Luo

Yixian Huang

Hua Zhang

Yiqing Mu

Bei Zhang

Chien-Yu Chen

Ganqian Hou

Min Zhang

Song Li

Abstract

Graphical abstract

Highlights

1. Introduction

Scheme 1.

2. Results

2.1. DPP4 was induced by chemotherapeutic agents

Fig. 1.

2.2. Biological consequence of DPP4 inhibition alone or in combination with chemotherapy

Fig. 2.

2.3. Development of a 5-ASA-conjugated HA-dendrimer-based nanocarrier (HASA) for co-delivery of DPP4i and chemotherapy agents

Fig. 3.

2.4. pH sensitive behavior of HASA nanocarrier

Fig. 4.

2.5. Favorable pharmacokinetics profile and efficient tumor targeting of HASA nanocarrier

Fig. 5.

2.6. HA/CD44-mediated tumor targeting and transcytosis of HASA nanocarrier in vivo

Fig. 6.

2.7. Co-delivery of DPP4i and chemotherapeutic drugs led to improved anti-tumor efficacy and immunity

Fig. 7.

3. Discussion

4. Methods

4.1. Clinical data analysis

4.2. Lentiviral infection and RNA interference

4.3. ELISA assay

4.4. DPP4 enzymatic activity

4.5. Fabrication and physicochemical characterization of drug-loaded micelles

4.6. Pharmacokinetics and in vivo IVIS imaging

4.7. Evaluation of tumor targeting efficiency of HASA nanocarrier

4.8. Cellular uptake, transwell assay, and cell spheroid penetration

4.9. In vivo therapeutic treatment

4.10. In vivo toxicity study

4.11. Analysis of tumor-infiltrating immune cells by flow cytometry

4.12. RNAseq analysis

4.13. Statistical analysis

Data availability

Ethics approval and consent to participate

CRediT authorship contribution statement

Declaration of competing interest

Acknowledgments

Footnotes

Appendix A. Supplementary data

References

Associated Data

Supplementary Materials

Data Availability Statement

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases