English

• Tumor blockade therapy represents a promising strategy for tumor suppression, which hampers the normal physiological functions of malignant cells by disrupting the communication between the tumor and its surrounding microenvironments [1]. Rather than eliminating unwanted cells directly, it obstructs the essential supplies of nutrients, oxygen, and biomolecules, leading to reduced cell motility and decreased viability [2,3]. Generally, tumor blockade therapy involves vessel blockade [4-6], the introduction of artificial extracellular matrix [7,8], and biomineralization [9]. Among these, biomineralization is particularly noteworthy due to its biocompatibility and self-reinforcing capability using endogenous substances [10]. Notably, chemical tools that trigger biomineralization function at the peripheral area of tumor tissue. Thus, there is no need to navigate the high interstitial fluid pressure typically found within the tumor microenvironments or to cross endosomal barriers to reach the intracellular compartment, which circumvents the limitations that compromise the efficacy of many cell-targeting chemotherapeutics [11-13].

  In our previous studies, we initiated biomineralization around the tumor tissue by preparing bisphosphonate-containing polymers, including 1,2-distearoyl-sn‑glycero-3-phosphoethanolamine-N-poly(ethylene glycol)-alendronate (DPA) and dodecylamine-poly((γ-dodecyl-l-glutamate)-co-(l-histidine))-block-poly(l-glutamate-graft-alendronate) (BINP) [3,14,15]. It was proved that the biomineralized barrier around osteosarcoma indeed contributed to tumor growth suppression, pulmonary metastasis elimination, and bone erosion remission. Compared with DPA that was inserted into the cytomembrane without selectivity, the addition of a histidine component to the polypeptide-based BINP allowed it to respond to the mildly acidic microenvironments, endowing it with a tumor-selective property [16,17]. Specifically, tumor cells incubated with BINP at pH 6.5 had a calcium content 1.35 times higher after 12 h incubation than those incubated at pH 7.4. However, realizing specific tumor cell biomineralization remains a significant challenge.

  Bioorthogonal "click chemistry" encompasses a suite of reactions that proceed under physiological conditions without interfering with inherent biochemical processes [18,19]. Among these, the reaction between the dibenzocyclooctyne (DBCO) and azido group (N₃) is noted for its efficiency, selectivity, and catalyst-free properties [20]. Leveraging this principle, a metabolic glycoengineering strategy was adopted to facilitate targeted tumor cell biomineralization [21,22]. We designed a biomineralization-initiating polymer, dibenzocyclooctyne-poly(ethylene glycol)-alendronate (DBCO-PEG-ALN, DBPA), which consists of three functional components: DBCO group to mediate "click chemistry" reaction, the hydrophilic segment PEG, and the biomineralization-inducing component ALN. As depicted in Scheme 1, N₃ was tagged onto the targeted cells through the metabolism of mannose derivate Ac₄ManNAz for sialic acid biosynthesis. DBPA was then anchored efficiently to the cell membrane via a "click chemistry" reaction between DBCO and N₃. Subsequently, bisphosphonate groups attracted calcium ions (Ca²⁺) from the microenvironments, initiating biomineralization around tumor cells. This glycoengineering-assistant biomineralization ensures the selective blockade of cells marked with N₃, enhancing both the biosafety and therapeutic efficacy of blockade therapy.

  Scheme 1

  

  Scheme 1.  Glycoengineering-assistant tumor cell biomineralization.

  DownLoad: Full-Size Img PowerPoint

  DBPA was readily synthesized by reacting the N-hydroxysuccinimide (NHS) group of DBCO-PEG-NHS and the amine group of alendronate sodium (Scheme S1 in Supporting information). The successful attachment of ALN onto DBPA was confirmed by proton nuclear magnetic resonance (¹H NMR) and carbon nuclear magnetic resonance (¹³C NMR) spectra (Figs. 1A and B). Compared with the ¹H NMR spectrum of DBCO-PEG-NHS in Fig. S1 (Supporting information), the emergence of new characteristic peaks—peak b (−CH₂CH₂CH₂−, 1.80, 1.90 ppm) and peak f (−NHCH₂CH₂−, 3.24 ppm)—in the ¹H NMR spectrum of DBPA demonstrated the successful grafting of ALN. The integration area of peak b indicated that the substitution ratio of NHS groups by ALN was about 95.8%. Additionally, the matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS) spectrum revealed that the molecular weight of DBPA was about 2600 Da (Fig. 1C). The disappearance of the peak corresponding to stretching vibration v_C at 1710–1750 cm⁻¹ in the Fourier transform-infrared (FT-IR) spectrum for DBPA compared with the spectrum of DBCO-PEG-NHS illustrated the complete substitution of NHS groups (Fig. 1D). Peaks at 1150 and 1060 cm⁻¹ could be ascribed to the stretching vibrations of v_P−O and v_C−O in the ALN component, respectively, proving the successful grafting of ALN onto DBPA.

  Figure 1

  

  Figure 1.  Chemical characterizations of DBPA. (A) ¹H NMR and (B) ¹³C NMR spectra of DBPA in D₂O. (C) MALDI-TOF MS spectrum of DBPA. (D) FT-IR spectra of DBPA and DBCO-PEG-NHS. Blue arrows represent alterations of characteristic absorption peaks.

  DownLoad: Full-Size Img PowerPoint

  Following the successful synthesis of DBPA, we sought to determine the optimal conditions for labeling 143B cell membrane with N₃. To this end, 143B cells were incubated in media supplemented with 50.0 µmol/L Ac₄ManNAz for 1, 2, or 3 days, respectively. After each period, the cells were exposed to DBCO cyanine5 (DBCO-Cy5) for 30 min. A control group of tumor cells, cultured with DBCO-Cy5 but without Ac₄ManNAz pretreatment, was also established for comparison. As indicated in Figs. 2A and B, the mean fluorescence intensity (MFI) of tumor cells showed no apparent differences between groups pretreated with Ac₄ManNAz for 1 or 2 days, whereas 3-day incubation significantly increased the amount of N₃ attached. Specifically, the MFI of 143B cells pretreated with Ac₄ManNAz for 3 days increased by 9.4% compared with those treated with DBCO-Cy5 only, confirming the successful "click chemistry" reaction between DBCO and N₃. Furthermore, as shown in Fig. 2C, the MFI was more remarkable in cells treated with 50.0 µmol/L than those treated with 25.0 µmol/L Ac₄ManNAz. Based on these findings, an Ac₄ManNAz concentration of 50.0 µmol/L and a 3-day metabolic labeling period were used for subsequent experiments. To visually confirm the successful labeling, 143B cells were examined using confocal laser scanning microscopy (CLSM). The microscopy results, depicted in Fig. 2D, showed bright red fluorescence of DBCO-Cy5 co-localizing with the green fluorescence of cytomembrane marker, demonstrating the existence of N₃ on the tumor cell surface.

  Figure 2

  

  Figure 2.  Metabolic glycan labeling of 143B cells with N₃. (A) Representative Cy5 histograms and (B) quantified MFI of 143B cells incubated without or with 50.0 µmol/L Ac₄ManNAz for 1, 2, or 3 days and then stained with DBCO-Cy5. 143B cells treated with PBS were used as the Control group. (C) MFI of 143B cells incubated with 25.0 or 50.0 µmol/L of Ac₄ManNAz for 3 days and then stained with DBCO-Cy5. (D) CLSM images of Ac₄ManNAz-pretreated 143B cells that were then incubated with DBCO-Cy5. Cell nuclei and membranes were stained with 4′,6-diamidino-2-phenylindole (DAPI) and 3,3′-dioctadecyloxacarbocyanine perchlorate (DIO), respectively. Scale bar: 20 µm. Statistical data are represented as mean ± SD (n = 3). NS: not significant. ****P < 0.0001. Student's t-test.

  DownLoad: Full-Size Img PowerPoint

  To assess the reaction activity between PEGylated-DBCO and N₃, DBCO-PEG-Cy5 (DBPC) was synthesized by reacting sulfo-Cy5-amine with DBCO-PEG-NHS. Following the same experimental design as that used for the DBCO-Cy5 labeling, the MFI of 143B cells treated successively with Ac₄ManNAz and DBPC (referred to as MDBPC group) was compared with that of cells without Ac₄ManNAz pretreatment. It turned out that the MFI of MDBPC group was more potent than that of the DBPC group, suggesting that the PEG grafting did not adversely affect the "click chemistry" between DBCO and N₃ (Figs. 3A and B).

  Figure 3

  

  Figure 3.  Attachment of PEGylated-DBCO onto N₃-labeled cytomembrane. (A) Representative Cy5 histograms and (B) quantified MFI values of 143B cells incubated with 50.0 µmol/L Ac₄ManNAz or PBS for 3 days and then stained with DBPC. (C) MFI of 143B cells incubated sequentially with Ac₄ManNAz or PBS and then DBPF, with or without trypan blue treatment. (D) CLSM images of Ac₄ManNAz- or PBS-pretreated 143B cells incubated with DBPC. Cell nuclei and membranes were stained with DAPI and DIO, respectively. Scale bar: 20 µm. Statistical data are represented as mean ± SD (n = 3). P < 0.05, ***P < 0.001, ****P < 0.0001. Student's t-test.

  DownLoad: Full-Size Img PowerPoint

  Furthermore, DBPA needed to adhere to the cytomembrane to facilitate biomineralization rather than being internalized by tumor cells. To differentiate between intracellular and surface-bound polymers, DBCO-PEG-FITC (DBPF) was synthesized. The localization of polymer was determined using trypan blue, which quenches FITC fluorescence exclusively in the extracellular compartment. As shown in Fig. 3C, there were no apparent differences in MFI before and after the trypan blue treatment of cells incubated with DBPF. By contrast, the MFI of cells in the MDBPF group incubated sequentially with Ac₄ManNAz and DBPF decreased by 19.7% after trypan blue quenching. Moreover, CLSM images illustrated the partial co-localization of cytomembrane (green fluorescence) and DBPC (red fluorescence) in the MDBPC group, whereas approximately all the DBPC was endocytosed by 143B cells after 30-min incubation without Ac₄ManNAz pretreatment (Fig. 3D).

  Following their anchorage to the cell membrane, the bisphosphonate groups of DBPA effectively chelated positive ions (predominantly Ca²⁺), which then facilitated the attraction of negative ions (primarily phosphate (PO₄³⁻) from the surrounding microenvironments. Scanning electron microscopy (SEM) equipped with an energy dispersive spectrometer (EDS) was used to visualize the pattern of Ca²⁺ deposition on the cytomembrane. As illustrated in Fig. 4A and Fig. S2 (Supporting information), the 143B cells in the MDBPA group (treated sequentially with Ac₄ManNAz for 3 days and with DBPA for 12 h) displayed a prominent Ca signal in the element mapping images. By contrast, a minimal Ca signal was observed on the cell surfaces in the control and DBPA groups. Semiquantitative analysis of the Ca mapping images further underscored the effectiveness of glycometabolism-assisted biomineralization, as detailed in Fig. 4B.

  Figure 4

  

  Figure 4.  Biomineralization of 143B cells assisted by metabolic glycoengineering. (A) Representative SEM images and Ca element mapping images of 143B cells incubated sequentially with Ac₄ManNAz or PBS for 3 days and DBPA for 12 h. 143B cells treated with PBS were used as the Control group. Scale bar: 5 µm. (B) Semiquantitative Ca signal intensities calculated from Ca element mapping images. (C) Alterations in relative Ca content of 143B cells with extension of incubation time. Statistical data are represented as mean ± SD (n = 5 for B; n = 3 for C). **P < 0.01, ***P < 0.001, ****P < 0.0001. Student's t-test.

  DownLoad: Full-Size Img PowerPoint

  Additionally, inductively coupled plasma mass spectrometry was conducted to quantify the cellular Ca content following various treatments (Fig. 4C). Consistent with the EDS findings, there was a significant increase in Ca content of 143B cells in the MDBPA group. However, cells in the DBPA group exhibited only a slight increase in Ca content, emphasizing the critical role of metabolic pre-conditioning in enhancing biomineralization. Moreover, the extent of ion deposition was found to be dependent on the concentration of Ca²⁺ available in the medium. Specifically, the Ca content of 143B cells in the MDBPA+Ca group that had been pretreated with Ac₄ManNAz for 3 days and then incubated for 12 h in a medium containing DBPA and 10.0 mmol/L Ca²⁺ was about three times higher than that of the Control group, compared with increases of 2.1-fold in the MDBPA group and only 1.3-fold in the DBPA group. These data confirm the influence of Ca²⁺ concentration on the effectiveness of biomineralization, demonstrating that higher levels of available Ca²⁺ significantly enhance biomineralization.

  Biomineralization significantly impaired tumor cell viability, migration, and invasion. To evaluate the cytotoxicity of biomineralization, three experimental groups were established, namely, the DBPA, MDBPA, and MDBPA+Ca groups (Fig. 5A). For comparative analysis, 143B cells were also treated with phosphate-buffered saline (PBS; Control group), 10 mmol/L Ca²⁺ (Ca group), or 50.0 µmol/L Ac₄ManNAz. The results indicated that neither a high Ca²⁺ concentration in the microenvironments nor Ac₄ManNAz supplementation adversely affected tumor cell viability relative to the Control group. As expected, the MDBPA and MDBPA+Ca groups exhibited higher cytotoxicity compared to the DBPA group, which could be attributed to the Ca deposition on the cell surface facilitated by "click chemistry". Notably, there was a marked reduction in cell viability in the MDBPA+Ca group, with a 67.4% decrease compared with the Control group; this was substantially higher than the reductions observed in the DBPA (29.1%) and MDBPA (44.3%) groups.

  Figure 5

  

  Figure 5.  Biomineralization hampered the viability and mobility of 143B cells. (A) Cell viability of 143B cell after 48 h of Ca deposition induced by DBPA. (B) Relative scratch widths after 0, 12, and 24 h. (C) Representative images taken at bottom of chamber membrane and (D) numbers of invasive cells after 24 h of culture. Scale bar: 100 µm. Statistical data are represented as the mean ± SD (n = 5 for A; n = 3 for B; n = 4 for D). P < 0.05, ***P < 0.001, ****P < 0.0001. Student's t-test.

  DownLoad: Full-Size Img PowerPoint

  Cell mobility is a critical characteristic of tumor malignancy [23]. Here, a scratch assay was employed to assess the impact of biomineralization on tumor cell migratory capabilities (Fig. 5B and Fig. S3 in Supporting information). The scratch width decreased by 27.2% after 24 h in the Control group, whereas the decrease in scratch width was less pronounced in the treatment groups (17.5% in the DBPA group, 10.4% in the MDBPA group, and only 5.7% in the MDBPA+Ca group). This gradient reduction in scratch width across the groups illustrated the effectiveness of biomineralization in curbing cell migration. Moreover, transwell assays showed weakened cell invasiveness after biomineralization, confirming that metabolic glycan labeling significantly mediated Ca deposition (Figs. 5C and D). Specifically, only 61.6% of cells in the MDBPA group traversed the Matrigel to reach the bottom of membrane after 24-h incubation compared with the Control group. Furthermore, the number of invasive cells decreased to 34.3% with an extra 10.0 mmol/L Ca²⁺ addition in the MDBPA+Ca group. By contrast, DBPA treatment moderately affected cell invasion capability, resulting in an invasion rate of 80.7%. These findings suggest that the biomineralization of tumor cells, particularly when intensified by a higher concentration of Ca²⁺, significantly reduced their mobility, thereby contributing to the therapeutic efficacy of the tumor blockade strategy.

  In conclusion, we have successfully developed a tumor-targeting biomineralization strategy facilitated by metabolic glycoengineering. By incorporating the N₃ into tumor cells, the facilely synthesized DBPA was efficiently anchored to the cytomembrane. The attached bisphosphonate groups on the cell surface efficiently attracted positive ions (primarily Ca²⁺) and negative ions (primarily PO₄³⁻). Over time, a biomineralized barrier formed around tumor cells, significantly impeding their viability and mobility. Thus, leveraging metabolic glycoengineering-assistant biomineralization, this approach emerges as a promising strategy for in vivo tumor suppression.

  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.

  CRediT authorship contribution statement

  Yang Liu: Conceptualization, Validation, Formal analysis, Investigation, Writing – original draft. Minglu Li: Conceptualization, Investigation, Writing – review & editing. Jianxun Ding: Conceptualization, Writing – review & editing, Supervision, Project administration, Funding acquisition. Xuesi Chen: Conceptualization, Writing – review & editing, Supervision, Project administration, Funding acquisition.

  Acknowledgments

  This work was financially supported by the National Natural Science Foundation of China (Nos. U23A20591 and 52273158) and the Science and Technology Development Program of Jilin Province (Nos. 20240101002JJ and 20210504001GH).

  Supplementary materials

  Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2024.110146.

  
  
• 1. [1]

     J. Ding, X. Feng, Z. Jiang, et al., Biomacromolecules 20 (2019) 4258–4271. doi: 10.1021/acs.biomac.9b01263

  2. [2]

     D. Zhao, X. Huang, Z. Zhang, et al., Wiley Interdiscip. Rev. Nanomed. Nanobiotechnol. 13 (2021) e1691.

  3. [3]

     Z. Jiang, Y. Liu, R. Shi, et al., Adv. Mater. 34 (2022) 2110094.

  4. [4]

     H.F. Dvorak, Cancer J. 21 (2015) 237–243.

  5. [5]

     D.W. Siemann, Cancer Treat. Rev. 37 (2011) 63–74.

  6. [6]

     S. Yang, Z. Tang, C. Hu, et al., Adv. Mater. 31 (2019) 1805955.

  7. [7]

     X.X. Hu, P.P. He, G.B. Qi, et al., ACS Nano 11 (2017) 4086–4096. doi: 10.1021/acsnano.7b00781

  8. [8]

     D.W. Zheng, S. Hong, Q.L. Zhang, et al., Nat. Commun. 11 (2020) 4907.

  9. [9]

     Y. Wu, X. Pan, H. Xie, L. Que, X. Tang, Front. Pharmacol. 14 (2023) 1335019.

  10. [10]

      X. Yang, S. Gao, B. Yang, et al., Adv. Sci. 10 (2023) 2302272.

  11. [11]

      N. Tang, H. Li, L. Zhang, et al., Angew. Chem. Int. Ed. 60 (2021) 6509–6517. doi: 10.1002/anie.202016122

  12. [12]

      R. Zhao, B. Wang, X. Yang, et al., Angew. Chem. Int. Ed. 55 (2016) 5225–5229. doi: 10.1002/anie.201601364

  13. [13]

      Y. Omidi, J. Barar, Bioimpacts 4 (2014) 55–67.

  14. [14]

      Y. Liu, Z. Jiang, S. Tong, et al., Adv. Mater. 35 (2023) 2203291. doi: 10.1002/adma.202203291

  15. [15]

      Y. Liu, Y. Zhuang, J. Ding, J. Funct. Polym. 36 (2023) 1–5.

  16. [16]

      Y. Liu, D. Li, J. Ding, X. Chen, Chin. Chem. Lett. 31 (2020) 3001–3014.

  17. [17]

      P. Zheng, Y. Liu, J. Chen, et al., Chin. Chem. Lett. 31 (2020) 1178–1182.

  18. [18]

      J.M. Baskin, C.R. Bertozzi, QSAR Comb. Sci. 26 (2007) 1211–1219. doi: 10.1002/qsar.200740086

  19. [19]

      Y. Ma, S. Yan, X. Xu, H. Cao, R. Wang, Chin. Chem. Lett. 35 (2024) 108645.

  20. [20]

      H. Wang, R. Wang, K. Cai, et al., Nat. Chem. Biol. 13 (2017) 415–424. doi: 10.1038/nchembio.2297

  21. [21]

      H. Wang, D.J. Mooney, Nat. Chem. 12 (2020) 1102–1114. doi: 10.1038/s41557-020-00587-w

  22. [22]

      H. Lin, H. Hong, L. Feng, et al., Chin. Chem. Lett. 32 (2021) 4041–4044.

  23. [23]

      M.D. Ryser, B.H. Min, K.D. Siegmund, D. Shibata, Proc. Natl. Acad. Sci. U. S. A. 115 (2018) 5774–5779. doi: 10.1073/pnas.1716552115

• 

  Scheme 1  Glycoengineering-assistant tumor cell biomineralization.

  下载: 全尺寸图片 幻灯片
  

  Figure 1  Chemical characterizations of DBPA. (A) ¹H NMR and (B) ¹³C NMR spectra of DBPA in D₂O. (C) MALDI-TOF MS spectrum of DBPA. (D) FT-IR spectra of DBPA and DBCO-PEG-NHS. Blue arrows represent alterations of characteristic absorption peaks.

  下载: 全尺寸图片 幻灯片
  

  Figure 2  Metabolic glycan labeling of 143B cells with N₃. (A) Representative Cy5 histograms and (B) quantified MFI of 143B cells incubated without or with 50.0 µmol/L Ac₄ManNAz for 1, 2, or 3 days and then stained with DBCO-Cy5. 143B cells treated with PBS were used as the Control group. (C) MFI of 143B cells incubated with 25.0 or 50.0 µmol/L of Ac₄ManNAz for 3 days and then stained with DBCO-Cy5. (D) CLSM images of Ac₄ManNAz-pretreated 143B cells that were then incubated with DBCO-Cy5. Cell nuclei and membranes were stained with 4′,6-diamidino-2-phenylindole (DAPI) and 3,3′-dioctadecyloxacarbocyanine perchlorate (DIO), respectively. Scale bar: 20 µm. Statistical data are represented as mean ± SD (n = 3). NS: not significant. ****P < 0.0001. Student's t-test.

  下载: 全尺寸图片 幻灯片
  

  Figure 3  Attachment of PEGylated-DBCO onto N₃-labeled cytomembrane. (A) Representative Cy5 histograms and (B) quantified MFI values of 143B cells incubated with 50.0 µmol/L Ac₄ManNAz or PBS for 3 days and then stained with DBPC. (C) MFI of 143B cells incubated sequentially with Ac₄ManNAz or PBS and then DBPF, with or without trypan blue treatment. (D) CLSM images of Ac₄ManNAz- or PBS-pretreated 143B cells incubated with DBPC. Cell nuclei and membranes were stained with DAPI and DIO, respectively. Scale bar: 20 µm. Statistical data are represented as mean ± SD (n = 3). P < 0.05, ***P < 0.001, ****P < 0.0001. Student's t-test.

  下载: 全尺寸图片 幻灯片
  

  Figure 4  Biomineralization of 143B cells assisted by metabolic glycoengineering. (A) Representative SEM images and Ca element mapping images of 143B cells incubated sequentially with Ac₄ManNAz or PBS for 3 days and DBPA for 12 h. 143B cells treated with PBS were used as the Control group. Scale bar: 5 µm. (B) Semiquantitative Ca signal intensities calculated from Ca element mapping images. (C) Alterations in relative Ca content of 143B cells with extension of incubation time. Statistical data are represented as mean ± SD (n = 5 for B; n = 3 for C). **P < 0.01, ***P < 0.001, ****P < 0.0001. Student's t-test.

  下载: 全尺寸图片 幻灯片
  

  Figure 5  Biomineralization hampered the viability and mobility of 143B cells. (A) Cell viability of 143B cell after 48 h of Ca deposition induced by DBPA. (B) Relative scratch widths after 0, 12, and 24 h. (C) Representative images taken at bottom of chamber membrane and (D) numbers of invasive cells after 24 h of culture. Scale bar: 100 µm. Statistical data are represented as the mean ± SD (n = 5 for A; n = 3 for B; n = 4 for D). P < 0.05, ***P < 0.001, ****P < 0.0001. Student's t-test.

  下载: 全尺寸图片 幻灯片
•