Extracellular Forces Cause the Nucleus to Deform in a Highly Controlled Anisotropic Manner

Abstract

Physical forces arising in the extra-cellular environment have a profound impact on cell fate and gene regulation; however the underlying biophysical mechanisms that control this sensitivity remain elusive. It is hypothesized that gene expression may be influenced by the physical deformation of the nucleus in response to force. Here, using 3T3s as a model, we demonstrate that extra-cellular forces cause cell nuclei to rapidly deform (<1 s) preferentially along their shorter nuclear axis, in an anisotropic manner. Nuclear anisotropy is shown to be regulated by the cytoskeleton within intact cells, with actin and microtubules resistant to orthonormal strains. Importantly, nuclear anisotropy is intrinsic, and observed in isolated nuclei. The sensitivity of this behaviour is influenced by chromatin organization and lamin-A expression. An anisotropic response to force was also highly conserved amongst an array of examined nuclei from differentiated and undifferentiated cell types. Although the functional purpose of this conserved material property remains elusive, it may provide a mechanism through which mechanical cues in the microenvironment are rapidly transmitted to the genome.

Figure 1. Extracellular forces applied to cell nuclei reveal anisotropic deformations.

(a) Force application over the central nuclear region was carried out using AFM while simultaneously tracking nuclear deformation using LSCM. Scale bar is 10 μm. (b). Process of strain measurement. Nuclei stained with Hoechst 33342 allowed for observation of nuclear deformation. Shown is a 3T3 fibroblast nucleus prior to loading (red), and following force application (green), image is 12 μm wide. Nuclear outlines were extracted: before (red) and during (green) the deformation. Outlines were fit to an ellipse in order to extract the change in area and strain along the major (εmajor) and minor (εminor) axes over time, where ε = ΔL/L for each respective axis. (c) Area expansion following 5 s of 0 nN (◻) n = 15, 10 nN (○) n = 32, and 20 nN (▴) n = 29. (d) Percent strain over time in the major axis, and (e) the minor nuclear axis. Axial strains reveal a clear mechanical anisotropy. Shown is mean ± s.e.m.

Figure 2. Cytoskeletal inhibition affects nuclear deformation.

(a–d) Immunofluorescent images of fixed 3T3 cells showing DNA (blue), actin (red), and MTs (green). (a) Untreated 3T3 cells (n = 32), (b) CytD (n = 24), (c) Noco (n = 24), and (d) NocoCytD (n = 21) treated cells. Scale bar is 20 μm. (e) Change in projected nuclear area as a function of time for untreated (black), Noco (red), CytD (blue), and NocoCytD (green) treated 3T3s. Corresponding strains shown for (f) the major axis (ε major) and (g) the minor axis (ε minor) over 5 s of a 10 nN load. Shown is mean ± s.e.m.

Figure 3. Nuclear anisotropy is intrinsic.

(a) Mean anisotropic stretch ratio (SR) is shown for untreated and treated 3T3s. Anisotropy was demonstrated in all cases (SR < 1). Noco treatment demonstrated significantly less anisotropic behavior, whereas 3T3’s treated with either CytD or NocoCytD resulted in significantly more anisotropic behavior than untreated (intact) cells (*P < 0.05, **P < 0.01, ***P < 0.001, with t-test). (b) Plot of anisotropic stretch ratio versus circularity. Nuclear shape (prior to deformation) is significantly less circular (more elliptical) for cells lacking an intact actin network. Untreated 3T3 (black), Noco (red), CytD (blue), NocoCytD (green), and isolated nuclei (grey). Shown is mean ± s.e.m. (c) Mean plateau strains (ε_p) from fits to Kelvin-Voigt model. All cell populations, except those treated with Noco, exhibited significantly greater strain in their minor axes relative to their major nuclear axes (P < 0.05, with paired t-test). Shown is significance with respect to untreated (intact) cells (*P < 0.05, ***P < 0.001, ^†P < 0.0001, ^††P < 0.00001, with t-test). (d) Typical isolated 3T3 nucleus (blue). Forces applied to elongated isolated nuclei (n = 23) by an AFM tip (shadow) demonstrated reduced SR (increased anisotropic nuclear deformation). Scale bar is 20 μm.

Figure 4. Role of nuclear architecture in anisotropy.

(a) DIC images of live untreated 3T3 cells (n = 30) (left) and cells treated with TSA (n = 34) (right). Scale bar is 10 μm. (b) Plots of mean plateau strains (ε_p) are shown. (c) Plot of mean stretch ratio and circularity for untreated (black) and TSA-treated (pink) 3T3s. (d) Over-expression of lamin-A in 3T3s, shown by live-cell image transiently expressing EGFP-lamin-A in the nuclear envelope. DNA in blue, lamin-A EGFP (green). Scale bar is 12.5 μm. (e) Plots of mean plateau strains (ε_p) are shown for untreated (black) and Lamin-A over-expressing 3T3s (green). (f) Plot of stretch ratio and circularity for untreated (n = 22) and lamin-A transfected cells (green) (n = 18). (*P < 0.05, **P < 0.01, with t-test for strains and stretch ratios, ^†P < 0.05, ^††P < 0.01, with t-test for circularity).

Figure 5. Nuclear anisotropy is consistent across cell types.

Plot of anisotropic stretch ratio (SR) versus circularity for various established cell lines: 3T3 (n = 32), HFF (n = 32), CHO (n = 17), MDCK (n = 30), HeLa (n = 26), C2C12 (n = 20), D3 mESC (n = 10), and primary cells including fibroblasts from BALB/c mice (n = 15), and fibroblasts from C57BL/10 (n = 30) and Dmd^mdx (n = 22) mice, as well as myoblasts from C57BL/10 (n = 16) and Dmd^mdx (n = 9) mice. Established cell types demonstrated a clear nuclear anisotropy, as demonstrated by SR consistently <1. Deformations associated with Dmd cell types were more isotropic than wild-type cells, possibly due to a lack of lamin A/C and altered chromatin organization. Shown is mean ± s.e.m.

Figure 6. Intrinsic nuclear anisotropy is regulated by cytoskeletal resistance.

(a) Shown is a schematic of a 3T3 cell deformed by an AFM tip, with an enlarged depiction of the resultant nuclear strain along the major and minor axes. (b) Depicted is the nuclear response to force in relation to the observed mechanics. Anisotropic strain occurred in isolated nuclei, corresponding to a larger resistance to deformation in the major axis, opposed to its minor axis (represented by effective internal spring constants, k_i*, in the minor and major axes). For an intact cell, nuclear deformations are regulated by the cytoskeleton. In response to an apically applied force, actin resists deformations in the minor axis, while microtubules resist strain along the major axis. Resistance to deformation along the major axis is greater than the minor axis, as shown by effective spring constants k*, allowing the cytoskeleton to promote the inherent anisotropy.