1. Introduction

The cytotoxicity of optical radiation on cells and tissues has been well-documented through both in vivo and in vitro studies ^1,2,3. Exposure to ultraviolet (UV) radiation invariably damages living cells within tissues, inducing DNA damage through photochemical stress. In particular, UVC radiation (100–280 nm) can be directly absorbed by DNA, adversely affecting genomic stability ⁴. Furthermore, UV radiation is known to generate free radicals that induce DNA strand breaks ^5,6. Considering that DNA damage disrupts normal biological functions, accelerates aging, and causes various diseases, effective strategies for DNA repair and protection are essential.

From a biomechanical perspective, recent studies have indicated that biophysical and biomechanical cues are critical factors in regulating cellular events such as proliferation, differentiation, apoptosis, and gene expression ^7,8,9,10. The nucleus, which is the largest and stiffest organelle in eukaryotic cells, is subjected to mechanical forces transmitted from the extracellular environment through the cytoskeleton ^11,12. It has been suggested that the nucleus itself acts as a cellular mechanosensor. For instance, changes in nuclear shape induced by controlled cell adhesion areas may influence the regulation of cell proliferation ^13,14. Our previous work demonstrated that nuclear deformation significantly inhibits cell migration, proliferation, and protein synthesis ^15,16. Consequently, we hypothesized that large-scale deformation of the nucleus would alter the intranuclear mechanical environment in association with chromatin DNA condensation, potentially enhancing the radiation resistance of DNA. In our earlier study using microfabricated substrates with micropillar arrays (several micrometers in diameter and height), we observed that amphibian-derived cells (Xenopus laevis cell line derived from a tadpole) cultured on these micropillar substrates exhibited significant nuclear deformation as they were compressed between the pillars, which subsequently suppressed UV-induced DNA strand breaks ¹⁷.

Beyond their biological significance, it is important to understand how artificial microstructured materials mechanically regulate nuclear morphology and intracellular responses for the rational design of implantable sensors, biointerfaces, and other artificial devices introduced into cells or tissues. Knowledge of the influence of microstructured surfaces on nuclear mechanics and genomic stability may contribute to improving long-term biocompatibility while enabling stable intracellular signal readouts in future biomedical devices.

In the present study, we extended this investigation to mammalian cells (embryonic rat aortic smooth muscle cells) to examine the effects of nuclear deformation on cellular UV radiation resistance. Specifically, we fabricated micropillar substrates with both circular and elliptical cross sections to compare the differences in mechanical transitions imposed by the pillar geometry. The cell nuclei exhibited dramatic deformation, which matched the surface topography of the micropillar arrays. We characterized the nuclear morphological changes and confirmed the localization of intranuclear DNA and deformation of the nuclear envelope using confocal microscopy. We also assessed the inhibitory effects of mechanical confinement on cell migration using live cell imaging. Finally, the UV-induced DNA damage was quantified using fluorescence imaging, and we discussed the influence of nuclear deformation by micropillar arrays on the radiation resistance of cells.

2. Materials and Methods

2.1. Micropillar Array Substrates

Micropillar array substrates were fabricated using a replica-molding technique ^18,19,20. Briefly, an array of photoresist posts was patterned onto a silicon wafer using standard photolithography to serve as a male mold. To create a female mold, poly(dimethylsiloxane) (PDMS) (Sylgard 184; Dow Corning, Midland, MI, USA) was poured over the male mold, cured at 70 °C for 6 h, and subsequently peeled off. The resulting female mold was silanized to facilitate the subsequent release of PDMS. Finally, the PDMS was cast into a female mold, cured at 100 °C for 1 h, and peeled off to obtain the final micropillar array substrate.

To investigate the effects of mechanical confinement on the cellular behavior, PDMS micropillar arrays with two distinct geometries were fabricated in a hexagonal arrangement. The circular pillars, based on our previous work ¹⁷, had a diameter, height, and center-to-center spacing of 3, 9, and 9 μm, respectively (Fig. 1(A)). The elliptical pillars featured a major axis of 7 μm and a minor axis of 3 μm (Fig. 1(B)). The height for both pillar types was maintained at 9 μm.

3. Results

Typical fluorescence images of A7r5 cells cultured on flat PDMS and PDMS micropillar substrates are shown in Fig. 2. On both the fibronectin-coated flat and pillar substrates, the cells spread completely and were characterized by well-developed linear actin filaments (Figs. 2(A)–(C)). Cell nuclei on the flat substrate had an elliptical shape and a smooth surface (Fig. 2(D)). In contrast, on both circular and elliptical micropillar substrates, the nuclei were fully inserted and squeezed between the pillars. These nuclei were significantly deformed according to the pillar geometry, appearing to be mechanically “trapped” within the pillar arrays (Figs. 2(E), (F), (H), and (I)).

Figure 3 shows representative confocal fluorescence microscopy images of intranuclear DNA, nuclear lamina morphology, and DNA damage sites in the cells on each substrate after UV irradiation. In the nuclei on the flat substrate, DNA appeared to be distributed throughout the nucleoplasm (Fig. 3(A)), while the nuclear lamina (Lamin A/C) formed a thin, relatively uniform envelope along the nuclear periphery (Fig. 3(D)). In contrast, in the nuclei trapped between the micropillars, a high DNA density was observed along the regions significantly indented by the lateral surfaces of the pillars (Figs. 3(B) and (C)). At these sites, the fluorescence intensity of the nuclear lamina increased (Figs. 3(E) and (F)). Regarding the DNA damage sites visualized by \(\gamma\)-H2AX, discrete foci were scattered within the nuclei in all groups, and no specific trend was identified. However, the distribution of these damage sites appeared to be relatively heterogeneous in the nuclei trapped between micropillars (Figs. 3(H) and (I)).

Quantitative morphological analysis of the cell nuclei, based on confocal fluorescence microscopy data, showed that the projected area, major axis length, and minor axis length of the nuclei on the flat PDMS substrates were \(337{\pm}81\) μm\(^2\), \(24.1{\pm}3.8\) μm, and \(16.8{\pm}2.4\) μm, respectively. In comparison, the nuclear projected area on both the circular and elliptical micropillar substrates was significantly reduced to approximately 60% of that on the flat PDMS substrates (Fig. 4(A)). Furthermore, the nuclear shape index decreased to below 0.5 on these pillar substrates, indicating a significant shift toward an elongated and irregular morphology (Fig. 4(B)). Regarding nuclear volume, the nuclei on the circular micropillar substrates showed a significant decrease to approximately 60% compared with those on the flat substrates. In contrast, the nuclear volume on the elliptical micropillar substrates remained comparable to that on the flat substrates (Fig. 4(C)). The fluorescence intensity of intranuclear DNA was significantly increased on both the circular and the elliptical micropillar substrates compared to those on the flat substrates (Fig. 4(D)).

Figure 5 shows the cell migration analysis of A7r5 cells on the PDMS micropillar substrates, where significant nuclear deformation occurred, in comparison with cells on the flat PDMS substrates. On the flat PDMS substrates, the cells moved in random directions (Fig. 5(A)) with a migration speed of approximately 10 μm/h (Fig. 5(D), light blue bar). In contrast, cell migration was markedly restricted on the circular micropillar substrates (Fig. 5(B)), and the migration speed was reduced by less than half (Fig. 5(D), orange bar). On elliptical micropillar substrates, the cell migration speed was also significantly reduced (Fig. 5(D), green bar) by approximately 60% compared with that on the flat substrates. However, in some cases, the cells exhibited linear movement, as if they had passed through the gaps between the pillars (Fig. 5(C)), suggesting that the inhibitory effect on cell migration was lower than that of the circular micropillars.

Finally, A7r5 cells on each substrate were cultured for 0.5 to 6 h following UV irradiation, and fluorescence imaging of intranuclear DNA and phosphorylated histone \(\gamma\)-H2AX was performed (Figs. 6(A)–(C)). In cells on the flat substrates, the fluorescence intensity of \(\gamma\)-H2AX gradually increased after UV irradiation, peaking at 2 h, and then showed a decreasing trend while maintaining relatively high intensity levels even after 6 h (Fig. 6(D), light blue circles). In contrast, in the nuclei deformed by the micropillars, on both the circular and elliptical substrates, the expression of \(\gamma\)-H2AX peaked at 0.5 h post-irradiation. Subsequently, the intensity gradually decreased, reaching levels comparable to those of the non-irradiated control group after 6 h (Fig. 6(D), red squares and green triangles). Notably, the inhibitory effect on DNA damage was most pronounced in the cell nuclei trapped between the circular micropillars (Fig. 6(D), red squares).

4. Discussion

In this study, the effects of nuclear deformation on the UV resistance of A7r5 cells were investigated using PDMS micropillar substrates. These results suggested that large-scale nuclear deformation mediated by mechanical trapping between micropillars significantly suppressed UV-induced DNA damage. These findings are consistent with our previous results using amphibian cells ¹⁷ and are the first to demonstrate that the inhibitory effect of nuclear deformation on UV-induced DNA damage also occurs in mammalian cells. Furthermore, micropillar substrates with circular and elliptical cross sections were fabricated to compare the differences in mechanical cell confinement and their respective inhibitory effects. The suppression of UV-induced DNA damage was more pronounced in cells in which the nuclei were trapped between the circular micropillars.

Mechanical trapping of the cell nucleus not only caused significant deformation of the nuclear lamina and heterogeneity in intranuclear DNA (Fig. 3) but also induced a reduction in nuclear size (Figs. 4(A) and (C)). This mechanical stimulation led to an increase in the fluorescence intensity of Hoechst-stained DNA, which was evident from the increased intranuclear DNA density (Fig. 4(D)). The effects of nuclear reduction and DNA condensation were particularly prominent in cells trapped between the circular micropillars. Versaevel et al. ¹⁴ demonstrated that controlling cell adhesion shapes by microcontact printing causes large-scale nuclear elongation, which increases the DNA fluorescence intensity, suggesting drastic condensation of intranuclear chromatin. Similar increases in the intranuclear DNA density associated with nuclear deformation may have occurred in the present study. However, because chromatin-specific markers were not examined, the present results do not directly demonstrate functional chromatin condensation or heterochromatin formation.

The mechanical confinement of the cell nucleus by the micropillars also significantly inhibited cell migration (Fig. 5). While no significant difference in the inhibitory effect on migration was observed between the circular and elliptical micropillar substrates, the mean migration speed was lower on the circular micropillars (Fig. 5(D), orange bar). Analysis of the micropillar dimensions revealed that the minimum distance between adjacent pillar surfaces was 3 μm for the circular pillars, whereas it ranged from 4 to 6 μm for the elliptical pillars (Fig. 1). Thus, it suggests that the circular micropillar substrates provide a stronger mechanical trapping effect on the cell nucleus. From an engineering perspective, the narrower confinement geometry of the circular micropillars provides fewer available migration pathways and imposes stronger local mechanical restriction on the nucleus than the elliptical micropillars. This geometrical difference is likely responsible for the enhanced nuclear deformation and the greater suppression of UV-induced DNA damage observed in the circular micropillar substrates. This higher degree of migration inhibition likely contributes to the promotion of nuclear volume reduction and increase of intranuclear DNA density.

Mechanical deformation of the cell nucleus on micropillar substrates may also inhibit UV-induced DNA damage, as estimated from the fluorescence intensity of \(\gamma\)-H2AX (Fig. 6). In particular, the observed trend, where \(\gamma\)-H2AX expression peaks within a relatively short period after UV irradiation and subsequently declines rapidly (Fig. 6(D), red squares and green triangles), suggests a favorable effect on the efficient repair of DNA damage. It is widely believed that UV radiation generates reactive radicals such as OH radicals through the radiolysis of water molecules within the cell, which serve as major contributors to DNA damage ²². Takata et al. ²³ investigated the effects of chromatin condensation on radiation-induced DNA damage using isolated chromatin and reported that the frequency of DNA double-strand breaks in condensed chromatin after exposure to ionizing radiation was significantly lower than that in decondensed chromatin. This suppression of damage is likely because condensed chromatin contains fewer water molecules per unit of chromatin, thereby generating fewer reactive radicals ²⁴. In the present study, such chromatin condensation is thought to be associated with the compressive stress applied to the nucleus by the micropillars. The mechanically increased DNA density observed in the confined nuclei may contribute to the reduced DNA susceptibility to reactive radicals. However, it is unclear whether this phenomenon reflects chromatin condensation epigenetically. Future studies should investigate whether the observed increase in DNA density is accompanied by epigenetic chromatin remodeling using established chromatin markers (e.g., H3K9me3, H3K27me3, HP1, H3K4me3, and H3K27ac). Further investigation of intranuclear chromatin mobility will help to elucidate the detailed mechanisms underlying nuclear deformation-induced UV radiation resistance.

A limitation of this study is that the potential negative effects of large-scale nuclear deformation were not fully eliminated. We previously demonstrated that although nuclear deformation and trapping using micropillar substrates do not affect cell viability ¹⁵, they can decrease the synthesis of proteins such as the actin cytoskeleton ¹⁶. Further research is necessary to clarify the comprehensive biological impact of nuclear deformation on micropillars. Another limitation is that the local UV irradiation conditions around the micropillar structures were not directly evaluated. Although all the samples were irradiated under identical macroscopic conditions, the micropillar geometry may locally influence the UV fluence through reflection, scattering, or partial shielding. Therefore, geometry-dependent optical effects cannot be completely excluded and should be quantitatively assessed in future studies using optical simulations or direct UV measurements.

5. Conclusion

This study demonstrated that the mechanical deformation of the cell nucleus, induced by PDMS-based micropillar arrays, significantly enhanced the UV resistance of DNA in mammalian cells. Mechanical confinement of the nucleus between the pillars led to a reduction in nuclear volume and increased intranuclear DNA density, which effectively suppressed UV-induced DNA damage. Our findings suggest that these inhibitory effects are driven by both the restriction of cell motility and an increase in intranuclear DNA density under mechanical stress. These results provide critical insights into the interplay between biomechanical factors and genomic integrity and offer a potential foundation for the development of new physical strategies to protect DNA from radiation damage.