Page 360 - IJB-10-3

International Journal of Bioprinting hNVU chip for brain modeling and drug screening

development, proliferation, invasion, and angiogenesis by on the expression of tumor-related genes, and their effects
regulating cell growth and cell–ECM interactions. 44,45 FGB on metabolism (Figure 6A).
is typically absent in a healthy adult brain but is associated
with brain disorders such as multiple sclerosis, Alzheimer’s 3.4.1. 5-FU and vorinostat showed tolerable
disease, brain injury, and ischemia/hypoxia. 46,47 This cytotoxicity with the hNVU chip
suggests that the cellular remodeling of the ECM exhibits To investigate the effects of the two drugs on angiogenesis, we
pathological features. Finally, DMD (dystrophin), another first conducted cytotoxicity assessments on brain vascular
glioma marker closely associated with cytoskeletal constituent cells (ECs, pericytes, and SF188 cells) using
proteins, was also upregulated (Figure 5B). planar cell culture conditions (Figure S12 in Supplementary
File). At blood concentrations, 5-FU (50 μg/mL) reduced
The heatmap (Figure S10 in Supplementary File) the viability of ECs and pericytes, while vorinostat (0.24 μg/
revealed the altered expression of transporters during mL, equal to 0.9 μM) significantly damaged the viability
the development of the hNVU chip. The downregulation of glioma cells and ECs. Specifically, 5-FU exhibited
of a glucose transport-related gene (SLC5A1) suggests stronger cytotoxic effects on ECs than vorinostat at blood
a disturbance in glucose metabolism. On the other concentrations, whereas vorinostat demonstrated stronger
hand, fatty acid (SLC27A1)- and glutamine (SLC38A1, cytotoxicity toward glioma cells. This suggests that both
SLC38A3)-related genes showed significant upregulation, drugs have the potential to disrupt angiogenesis, but they
indicating a potential enhancement of aerobic respiration may target different cell types. Subsequently, we examined
metabolism and increased neurotransmitter metabolism the cytotoxicity of the two drugs at blood concentrations
within the brain region. Notably, this gene expression on the overall 3D hNVU chip. Within 24 and 48 h, 5-FU
profiling was consistent with the trend of expression of and vorinostat reduced the viability to approximately
transporter-related genes in glioma lesions. 48-50 70%, showing that they were less cytotoxic than in two-
Taken together, these gene expression results suggest dimensional cell culture conditions (Figure 6B; Figure S13
that the brain region structure of the hNVU chip may in Supplementary File). This indicates a certain level of
possess pathological features such as tumor angiogenesis tolerance to the cytotoxic effects to hNVU caused by 5-FU
and altered transporters. and vorinostat at blood concentrations.
3.4. The evaluation of therapeutic effects 3.4.2. 5-FU and vorinostat can permeate across
demonstrates the reliability of the hNVU chip for the BBB
drug screening in pediatric brain tumors Furthermore, we evaluated the penetration of 5-FU (50 μg/
Therapy for pediatric brain tumors requires not only mL) and vorinostat (0.24 μg/mL) at blood concentrations
considering the effectiveness of killing tumor cells, across the BBB (Figure 6C and D). One hour after
antitumor angiogenesis, and alteration of tumor glucose perfusion administration through channels A and B, both
metabolism but also evaluating the effects on neuronal drugs were detected in wells C and D in the brain region.
cells to avoid cognitive impairments during treatment. The concentration of 5-FU in the brain region (channel
5-FU was applicable to colon cancer and breast cancer C) reached its peak (17.5 μg/mL) after 8 h (Figure 6C),
and possessed anti-angiogenic ability. However, 5-FU while vorinostat reached its peak (0.14 μg/mL) at 16 h
51
could cause cognitive impairments in clinical trials of (Figure 6D). The slower penetration rate of vorinostat
chemotherapy. Vorinostat has entered phase II clinical compared to 5-FU may be attributed to its relatively poor
trials for pediatric brain tumors, but our understanding permeability. 52,53 In view of the ability of 5-FU to cross the
about the metabolism of vorinostat is still limited. We BBB, the cytotoxicity results offer a plausible explanation
chose these two classic drugs, 5-FU and vorinostat, for for cognitive impairment observed in human patients
research to further test the capabilities of our model during chemotherapy. 54-56 It also confirms that vorinostat
in drug screening for pediatric brain tumor treatment. can penetrate the BBB and reach the brain region, justifying
Astrocytoma cell SF188 was introduced in the model to its application in clinical trials for the treatment of brain
16
construct an hNVU chip with tumor cells for the study of tumors. After administration, 5-FU and vorinostat
chemotherapeutic drugs (Figure S8B in Supplementary significantly reduced gene expression of model barrier tight
File). This study found that when astrocytoma cells were junctions (Figure S14 in Supplementary File), adhesion
introduced into the model, the BTB compromises the junctions (Figure S15 in Supplementary File), gap junctions
BBB integrity (Figure S11 in Supplementary File). We (Figure S16 in Supplementary File), cytoskeletons, and
administered these chemotherapeutic drugs through other molecules (Figure S17 in Supplementary File). The
the A and B channels of the chip and investigated their BBB was destroyed, and the drug increased the penetration
cytotoxicity, their ability to penetrate the BBB, their impact rate of the BBB (Figure S18 in Supplementary File). This

Volume 10 Issue 3 (2024) 352 doi: 10.36922/ijb.1684