A tumor microenvironment model coupled with a mass spectrometry system to probe the metabolism of drug-loaded nanoparticles

Ling Lin *ab, Yajing Zheng ac, Zengnan Wu ac, Wei Zhang ab and Jin-Ming Lin b
aCAS Key Laboratory of Standardization and Measurement for Nanotechnology, CAS Center for Excellence in Nanoscience, National Center for Nanoscience and Technology, Beijing 100190, People's Republic of China. E-mail: linling@nanoctr.cn
bUniversity of Chinese Academy of Sciences, Beijing 100049, People's Republic of China
cBeijing Key Laboratory of Microanalytical Methods and Instrumentation, MOE Key Laboratory of Bioorganic Phosphorus Chemistry & Chemical Biology, Department of Chemistry, Tsinghua University, Beijing 100084, People's Republic of China

Received 17th June 2019 , Accepted 19th July 2019

First published on 22nd July 2019


This study describes a novel integrated microfluidic platform with microvessel network channels which are connected to an electrospray ionization-mass spectrometer (ESI-MS), allowing real-time probing of nanoparticle (NP) drug delivery in vascular niche around tumors and understanding the drug delivery efficacy in a realistic tumor microenvironment.


In recent years, many new technologies have been developed to provide new cancer treatments.1,2 Nanoparticles (NPs) for drug delivery have great potential for revolutionizing cancer treatment, due to their high selectivity and efficacy.3–5 Although improvements in delivery efficiency have been reported using specifically designed NPs, only about 5% of the drugs can be delivered to the target position.6,7 This limited delivery is thought to be caused by complex transport processes after systemic administration and pathophysiological barriers around tumors.8,9 For better understanding of these difficulties, the prerequisite is to develop an in vitro cell co-culture model to mimic the tumor microenvironment, allowing real-time monitoring of NP behaviour and online quantitative analysis of drug metabolites.

Microfluidic technology had a dramatic impact on many fields of physiologically relevant microenvironments, due to the microstructure and precise control of the chemical environment of cells.10–12 Compared with traditional cell culture, rational design of microchannels can better simulate the microenvironment of cell co-culture, such as real-time exchange of diffusible factors between different cell culture chambers.13–15 However, microscopic observation can hardly obtain critical information, such as drug transfer efficiency and the accumulation of tumor target sites. To unveil the drug delivery by NPs, real-time detection of cell apoptosis and drug metabolites in co-cultured TCs and ECs is highly required.

Moreover, drug analysis is applied to the study of not only drug quality and stability,16 but also all stages of new drug development.17 It is essential to assess the drug transportation.18 We aim to evaluate drug transfer efficiency and the accumulation of tumor target sites based on temporal order of drug metabolite detection. Integrating the advantages of mass spectrometry technology and tumor microenvironment chips could throw light on the dark corners for the investigation of NP performance in cancerous tissues. However, limited studies have developed an advanced microfluidic platform with functions of cell co-culture and analysis of drug metabolites to help the field of nano-medicine and the nanoparticle drug screening process.

In this study, we present an integrated microfluidic device with a tumor vascular network model to simulate NP drug delivery in vascular niche around tumors. This microfluidic drug delivery platform creates an in vivo tumor microenvironment encompassing the cell co-culture component and the metabolic detection component. In the cell co-culture component, cell proliferation and NP drug-induced apoptosis under different culturing conditions have been examined. The metabolite detection component, integrated with mass spectrometry, illustrated the chronological order of the NPs drug (paclitaxel) metabolites secreted by Human cervical carcinoma cell line (HeLa) coculture with Human umbilical vein endothelial cells (HUVECs). This integrated bionic microfluidic device has been proved as an attractive and powerful tool to mimic vessel structures approximately and monitor cells, collect the resulting paclitaxel metabolites released by the cells, and investigate the NP drug delivery within an engineered tumor microenvironment.

In order to recapitulate the complex microenvironment of tumor vasculature, the model was designed to mimic a pair of capillary vessels where tumor tissues are sandwiched as shown in Fig. 1a. The tumor vasculature supplies sufficient cytokines and signalling for HeLa cells and NPs will be transported through this microstructure to the cancer cells. As shown in Fig. 1b, this biomimetic system was composed of three parallel channels on a single chip: two side HUVEC culture chambers to simulate the blood vessels that could deliver the cell culture medium and drug solutions, one internal HeLa cell culture chamber to simulate the tumor interstitium for anti-tumor drug targeting positions and two symmetrically distributed microvessel network channels connecting the external and internal cell culture chambers. The design and fabrication details are illustrated in the ESI (Fig. S1). The NPs will be transported through this microvessel structure to the HeLa cells. Specially, these irregular connection channels will allow free exchange of the signalling of soluble factors between co-cultured HeLa cells and HUVECs, as well as minimization of cell migration and mixing.19–21 These compartmentalized microchannels in the microdevice enable more precise mimicking of the perivascular niche of the tumor microenvironment, allowing dynamical observation of the changes in cells, as well as delivery of drugs and nutrients to both cells (Fig. S2, ESI). The microfluidic network designed in this chip-ESI-MS platform provided important functions in continuous cell culture and drug screening.


image file: c9cc04628c-f1.tif
Fig. 1 Schematic of the integrated microdevices for mimicking a tumor-microenvironment. (a) A minimal functional unit of a solid tumor, a pair of capillary vessels, NP drug can be introduced from vessels to tumour cells. (b) HeLa cells and endothelial cells were co-cultured, and relative metabolites were analyzed by MS.

The cellular interactions between HeLa cells and HUVECs play key roles in regulating the tumor microenvironment. Thus, in this study, the long-term viability of co-cultured HeLa cells and HUVECs inside the cell co-culture was investigated. The HeLa cells were seeded into the center main channel chamber, while the HUVECs were seeded into the two side channels (check cell culture and stain details in the ESI). The cell culture medium continuously flowed into the microchannels (5 µL h−1) from the inlet of the two chambers containing HUVECs, across the connection microchannels, and out from the outlet of the chamber containing HeLa cells, which could partially mimic the tumor microenvironment. The viability of co-cultured cells inside the microfluidic system was determined by calcein-AM/EthD-1 staining at the predetermined time. After three days of co-culture of HeLa cells and HUVECs, both kinds of cells showed high viability over 95% (Fig. S3, ESI). These results demonstrated the successful achievement of cell co-culture in this system.

In this work, a vessel network model on a microfluidic device was constructed to investigate paclitaxel-NP cell cytotoxicity differentiation with a free paclitaxel treated cell group as a control group. The cytotoxicity of paclitaxel in co-cultured HeLa cells and HUVECs by spiking varying concentrations of paclitaxel into the cell culture medium was investigated. The cell co-culture loaded with HeLa cells and HUVECs was placed inside the incubator, and the cell culture medium spiked with paclitaxel was continuously injected into the microchip from the inlet of the HUVEC chamber at 5 µL h−1 respectively. After co-culture of HeLa cells and HUVECs for 1–3 days, we observe drug-induced (5.00 ng mL−1 paclitaxel) cytotoxicity of co-cultured HeLa cells, indicated by the decreased green fluorescence of cells stained with calcein-AM (Fig. S4, ESI).

The 50% inhibitory concentration (IC50) values of paclitaxel against co-cultured HeLa cells and HUVECs were investigated for incubation times from 12 h to 72 h (Fig. 2a and b). For the following experiments, a cell culture medium with 5.0 ng mL−1 paclitaxel was used. Next, we compared the cytotoxicity of nanoparticle paclitaxel and free paclitaxel to HeLa cells under co-culture in the microchip. Under the same culture conditions (5.0 ng mL−1 paclitaxel), HeLa cells treated with the nanoparticle paclitaxel showed much lower viability than those treated with the free paclitaxel over a time period of three days. As shown in Fig. 2c, the viability of HeLa cells treated with nanoparticle paclitaxel was reduced to 62.6% after 12 h. In contrast, the viability of HeLa cells treated with free paclitaxel was reduced to 85.6% after 12 h. This observation implied that nanoparticle paclitaxel could efficiently increase the cytosolic concentration of paclitaxel, allowing more paclitaxel to transport into the nucleus of HeLa cells. After observation for 72 hours, the viability of HeLa cells treated with nanoparticle paclitaxel decreased from 96.2% to 23.3%, while that of HeLa cells treated with free paclitaxel decreased from 98.2% to 68.8%. The results confirmed that the cytotoxicity of HeLa cells was greatly enhanced by the paclitaxel-loaded nanoparticles under the co-culture conditions.


image file: c9cc04628c-f2.tif
Fig. 2 Cytotoxicity of NPs paclitaxel in cocultured Hela cells and HUVECs. (a and b) Time-dependent viability of cocultured Hela cells (a) and HUVECs (b) treated with paclitaxel at different concentrations from 0 to 1000 ng mL−1. (c) Long-term viability of co-cultured Hela cells treated with 5.00 ng mL−1 paclitaxel, with nanoparticles paclitaxel or free paclitaxel. Data are shown as mean ± standard error of the mean (SEM) for n = 3. **p < 0.01; ***p < 0.001.

Reactive oxygen species (ROS) are constantly produced during cellular function; they are chemically reactive and toxic to cells at high levels. High concentration of ROS could prompt tumor development and progression. Glutathione (GSH) is one of the major cellular antioxidants that protect cells against cytotoxicity by ROS.

Here, two kinds of ROS and GHS were generated by paclitaxel-nanoparticles and free paclitaxel respectively induced in HeLa cell coculture with HUVECs inside the microchip. Confocal fluorescence observation and quantitative results of paclitaxel-treated HeLa cells indicate a low expression level of ROS when free paclitaxel induced cells after three days, while the level of ROS generated by paclitaxel-nanoparticle induced HeLa cells is significantly increased (Fig. 3a and c). Moreover, the expression level of GSH by free paclitaxel induced HeLa cells under coculture conditions was much higher than that by paclitaxel-nanoparticle induced HeLa cells (Fig. 3b and d). During the treatment of paclitaxel drugs, a high level of ROS and a low level of GHS indicate that the induction of paclitaxel-nanoparticles is beneficial to restrain the redox system of HeLa cells and thus reduce the survival rate of tumor cells faster. Due to the presence of cell resistance, it is difficult to accumulate free drugs in cells. However, paclitaxel delivered by nanoparticles can avoid drug resistance genes and enter cells rapidly and in large quantities. It provided better performance for the pharmacodynamic function.


image file: c9cc04628c-f3.tif
Fig. 3 Detection of ROS and GSH in paclitaxel-treated HeLa cells. (a and b) Confocal fluorescence observation of ROS (a) and GSH (b) generated by paclitaxel-nanoparticle or free paclitaxel treated (5 ng mL−1) co-cultured HeLa cells. (c and d) Quantitative results of the expression of ROS (c) and GSH (d) analyzed from fluorescence images. Data are shown as mean ± standard errors of the mean (SEM) for n = 3. **p < 0.01; ***p < 0.001.

To analyze the pharmacokinetics, we compared the drug metabolism of paclitaxel-nanoparticles and free paclitaxel inside the cell. In pharmacokinetic analysis, time-dependent drug metabolism is of great significance. In our study, a microfluidic device was coupled with an ESI-Q-TOF MS system to achieve the quantitative detection of paclitaxel. During continuous cell culture treated with paclitaxel-nanoparticle or free paclitaxel medium, the cell metabolites could be online-detected through connecting by ESI-Q-TOF MS. Using this method, we first constructed the calibration curve of paclitaxel through spiking different concentrations of paclitaxel-NPs into the co-culture medium and performing chip-ESI-MS. As shown in Fig. 4a, paclitaxel (0.5 µ mL−1) with [M − H]+ = 876.842 was detected after the corresponding extraction procedure. The MS/MS spectra of m/z 308.991 were presented to provide additional structure information. The molecular ion peak intensity (Y) increased linearly with paclitaxel-nanoparticles (X) in the range of 0.0–5.0 ng mL−1, and the fitting formula is written as Y = 265[thin space (1/6-em)]757x − 7939 (R2 = 0.9996) (Fig. S9a, ESI); while the relationship between the molecular ion peak intensity (Y) and free paclitaxel concentration (X) is written as Y = 289[thin space (1/6-em)]769x − 147[thin space (1/6-em)]279 (R2 = 0.9994), as shown in Fig. S9b (ESI). In the pharmacokinetic analysis, time-dependent drug absorption and dose-dependent drug metabolism are of great significance.22 For this reason, we performed an online quantitative analysis of time-dependent drug absorption kinetics compared with NP-paclitaxel and free-paclitaxel in HeLa cells using the micro-vessel-ESI-MS system. As shown in Fig. 4b, the amount of paclitaxel was obviously decreased with increasing incubation time because of absorption into the cells. Among them, the amount of paclitaxel-nanoparticles in cell co-culture medium decreased to 13.92% over 72 h. In comparison, the amount of free paclitaxel under the same conditions decreased to 47.54% over 72 h. The low level of paclitaxel by NP-paclitaxel in culture medium implies either a higher uptake of paclitaxel by cells or the decreased pumping of paclitaxel. Moreover, we also studied the NP-paclitaxel dose-dependent drug metabolism of HeLa cells by incubating them in concentrations that ranged from 0 to 10 µg mL−1 for 72 h. As shown in Fig. 4c, the intensity of paclitaxel in the co-culture medium increased with the increasing concentrations of NP-paclitaxel from 0 to 0.5 µg mL−1, since drug-induced cell apoptosis can decrease the cell viability when concentrations of paclitaxel increase after 72 h. The strength of paclitaxel in the cell culture medium treated with NP-paclitaxel was lower than that of the free-paclitaxel after 72 h treatment. This result confirms that NP-paclitaxel could efficiently increase the cytosolic concentration of paclitaxel, allowing more paclitaxel to transport into the nucleus of cells. Consequently, the microvessel-ESI-MS platform developed here provides a fast and sensitive analysis for absorption of NP-drugs and their metabolites.


image file: c9cc04628c-f4.tif
Fig. 4 Detection of paclitaxel metabolites by microvessel-ESI-MS. (a) Mass spectra of NP-paclitaxel metabolites from Hela cells incubated on a microfluidic chip, the structure was further identified with corresponding MS/MS spectra. (b) Kinetics of paclitaxel metabolites in cell culture medium under treated by NP-paclitaxel and free-paclitaxel. (c) The accumulation of paclitaxel metabolites in the coculture medium of cells incubated with varied concentrations of NP-paclitaxel from 0 to 10 ng mL−1 for 72 h on-chip. Data are shown as mean ± standard error of the mean (SEM) for n = 3. **p < 0.01; ***p < 0.001.

In summary, a novel integrated microfluidic platform with microvessel network channels was successfully developed to study drug nanoparticle cytotoxicity in vitro. The transport process from blood vessels to target tumors on them offers an advanced approach to investigate drug screening in a more biomimetic microenvironment. This multifunctional vessel-chip-MS platform allows real-time monitoring of paclitaxel-induced apoptosis, ROS, and GSH of cocultured HeLa cells and HUVECs in vitro. The system has also demonstrated the capability of selective and quantitative analysis of cell-based drug absorption and metabolites with high stability, sensitivity, and repeatability. The NP-paclitaxel could efficiently increase the cytosolic concentration of paclitaxel, allowing more paclitaxel to be transported into the nucleus of HeLa/HUVECs. NP-paclitaxel was also proved as a fast nano delivery system, showing great promise for cancer therapy. In conclusion, the microdevice has provided a platform for understanding drug delivery efficacy in a realistic tumor microenvironment, and shown immense potential for efficient drug screening in vitro.

This work was supported by the National Natural Science Foundation of China (No. 21804026 and 21727814).

Conflicts of interest

There are no conflicts to declare.

Notes and references

  1. J. M. Greene, D. Levy, S. P. Herrada, M. M. Gottesman and O. Lavi, Cancer Res., 2016, 76, 2882–2890 CrossRef CAS PubMed.
  2. F. Guo, P. Li, J. B. French, Z. Mao, H. Zhao, S. Li, N. Nama, J. R. Fick, S. J. Benkovic and T. J. Huang, Proc. Natl. Acad. Sci. U. S. A., 2015, 112, 43–48 CrossRef CAS PubMed.
  3. E. Blanco, H. Shen and M. Ferrari, Nat. Biotechnol., 2015, 33, 941–951 CrossRef CAS PubMed.
  4. Y. Barenholz, J. Controlled Release, 2012, 160(2), 117–134 CrossRef CAS.
  5. J. Shi, P. W. Kantoff, R. Wooster and O. C. Farokhzad, Nat. Rev. Cancer, 2017, 17(1), 20–37 CrossRef CAS.
  6. Y. H. Bae and K. Park, J. Controlled Release, 2011, 153, 198–205 CrossRef CAS.
  7. K. Park, ACS Nano, 2013, 7, 7442–7447 CrossRef CAS PubMed.
  8. A. Ozcelikkale, S. Ghosh and B. Han, Mol. Pharmaceutics, 2013, 10, 2111–2126 CrossRef CAS PubMed.
  9. L. Y. Ye, W. Chen, X. L. Bai, X. Y. Xu, Q. Zhang, X. F. Xia, X. Sun, G. G. Li, Q. D. Hu, Q. H. Fu and T. B. Liang, Cancer Res., 2016, 76, 818–830 CrossRef CAS PubMed.
  10. C. Zheng, X. Zhang, C. Li, Y. Pang and Y. Huang, Anal. Chem., 2017, 89, 3710–3715 CrossRef CAS.
  11. Q. Huang, S. Mao, M. Khan, L. Zhou and J.-M. Lin, Chem. Commun., 2018, 54, 2595–2598 RSC.
  12. C. Lin, L. Lin, S. Mao, L. Yang, L. Yi, X. Lin, J. Wang, Z.-X. Lin and J.-M. Lin, Anal. Chem., 2018, 90, 10326–10333 CrossRef CAS.
  13. W. Liu, J. Xu, T. Li, L. Zhao, C. Ma, S. Shen and J. Wang, Anal. Chem., 2015, 87, 9752–9760 CrossRef CAS PubMed.
  14. L. Lin, Q. Chen and J. Sun, Trends Anal. Chem., 2018, 99, 66–72 CrossRef CAS.
  15. N. Li, W. Zhang, L. Lin, Z. He, M. Khan and J.-M. Lin, Chem. Commun., 2018, 54, 11435–11438 RSC.
  16. T. Lammers, F. Kiessling, W. E. Hennink and G. Storm, J. Controlled Release, 2012, 161(2), 175–187 CrossRef CAS.
  17. L. C. Wienkers and T. G. Heath, Nat. Rev. Drug Discovery, 2005, 4(10), 825–833 CrossRef CAS.
  18. C. Wang, T. Stylianopoulos, J. Cui, J. Martin, V. P. Chauhan, W. Jiang, Z. Popović, R. K. Jain, M. G. Bawendi and D. Fukumura, Proc. Natl. Acad. Sci. U. S. A., 2011, 108(6), 2426–2431 CrossRef.
  19. J. Zhang, J. Wu, H. Li, Q. Chen and J.-M. Lin, Biosens. Bioelectron., 2015, 68, 322–328 CrossRef CAS.
  20. H. Wei, H. Li, S. Mao and J.-M. Lin, Anal. Chem., 2011, 83, 9306–9313 CrossRef CAS.
  21. L. Lin, X. Lin, L. Lin, Q. Feng, T. Kitamori and J.-M. Lin, Anal. Chem., 2017, 89(18), 10037–10044 CrossRef CAS PubMed.
  22. R. M. Savic, D. M. Jonker, T. Kerbusch and M. O. Karlsson, J. Pharmacokinet. Pharmacodyn., 2007, 34, 711–726 CrossRef CAS.

Footnote

Electronic supplementary information (ESI) available. See DOI: 10.1039/c9cc04628c

This journal is © The Royal Society of Chemistry 2019