Increased asymmetric and multi-daughter cell division in mechanically confined microenvironments

Abstract

As the microenvironment of a cell changes, associated mechanical cues may lead to changes in biochemical signaling and inherently mechanical processes such as mitosis. Here we explore the effects of confined mechanical environments on cellular responses during mitosis. Previously, effects of mechanical confinement have been difficult to optically observe in three-dimensional and in vivo systems. To address this challenge, we present a novel microfluidic perfusion culture system that allows controllable variation in the level of confinement in a single axis allowing observation of cell growth and division at the single-cell level. The device is capable of creating precise confinement conditions in the vertical direction varying from high (3 µm) to low (7 µm) confinement while also varying the substrate stiffness (E = 130 kPa and 1 MPa). The Human cervical carcinoma (HeLa) model with a known 3N+ karyotype was used for this study. For this cell line, we observe that mechanically confined cell cycles resulted in stressed cell divisions: (i) delayed mitosis, (ii) multi- daughter mitosis events (from 3 up to 5 daughter cells), (iii) unevenly sized daughter cells, and (iv) induction of cell death. In the highest confined conditions, the frequency of divisions producing more than two progeny was increased an astounding 50-fold from unconfined environments, representing about one half of all successful mitotic events. Notably, the majority of daughter cells resulting from multipolar divisions were viable after cytokinesis and, perhaps suggesting another regulatory checkpoint in the cell cycle, were in some cases observed to re-fuse with neighboring cells post-cytokinesis. The higher instances of abnormal mitosis that we report in confined mechanically stiff spaces, may lead to increased rates of abnormal, viable, cells in the population. This work provides support to a hypothesis that environmental mechanical cues influences structural mechanisms of mitosis such as geometric orientation of the mitotic plane or planes.

Figure 1. Microfluidic cell confinement device.

(a) Device schematic with inset of confinement assay. The posts are 20×80 µm spaced equally 40 µm apart. (b) In the unconfined state, the posts are raised from the glass substrate (top), upon applying pressure the confinement chamber is compressed such that the posts are in contact with the glass substrate. (c) Seeding cells in the confinement chamber (left), when the posts are lowered the cell is confined to 3 or 7 µm height and is squeezed out from the post area (middle). Cells spread and attach in the confined volume (right).

Figure 2. Abnormal cell divisions under mechanical confinement.

(a) In an unconfined device control, mitosis occurs within 150 minutes. (b) A confinement condition (Δy = 3 µm, E = 130 kPa) showing sequential tri-daughter divisions with viable daughter cells. The first tri-daughter division is seen at 100 minutes, with cell fusion labeled by black arrows. The second tri-polar division is seen at 1550 minutes along with the viable 2^nd daughter cell labeled in orange. (c) A confinement condition (Δy = 7 µm, E = 130 kPa) showing an asymmetric multi-daughter division of cytoplasm after mitosis (cell with arrow receives disproportionate cytoplasmic volume). (d) A confinement condition (Δy = 7 µm, E = 1 MPa) showing a tetra-daughter division with a cross geometry chromosome alignment (marked by black arrows at 0 minutes) and cell fusion at 60 minutes (labeled by black arrows). (e) A confinement condition (Δy = 7 µm, E = 1 MPa) showing a triangular chromosome arrangement (marked by black arrows) leading to cell death. All scale bars are 20 µm.

Figure 3. Daughter cell asymmetry and multi-daughter division increases with confinement and stiffness.

Overlaid cell traces for control and all experimental conditions are shown. (a) Cell traces are analyzed 5 minutes post-telophase. Average control divisions are shown. (b,c) Low confinement average traces show increased size and asymmetry in stiffer environments. Notably, multi-daughter division type (tri- vs. tetra-daughter) depends on microenvironment mechanical properties. (d) In high confinement states, the daughter cells are more spread out and highly asymmetric when compared to both low compression and control cases. (e) Area difference between daughter cells in each test case demonstrating asymmetry differences. Error bars are SEM. All scale bars are 20 µm.

Figure 4. Quantitative effects of confinement on cell-cycle abnormalities.

(a) Summary of experimental data in terms of abnormalities per cell cycle characterized by: multi-daughter divisions, asymmetric divisions, apoptotic divisions, and divisions lasting for more than 140 minutes. (b) Multi-daughter divisions per cell cycle characterized as divisions producing more than two daughter cells. Error bars are SEP. (c) Asymmetric divisions are characterized by daughter cells having a difference greater than 15% in area. Error bars are SEP. (d) Cell deaths per cell cycle are characterized by cells that cease activity and contract prior to completion of mitosis. Error bars are SEP. (e) Comparing the mitosis duration of the high confinement (Δy = 3 µm, E = 130 kPa) condition to control and both low confinement conditions. Error bars are SEM. (f) Summary of divisions in each case completing division within 140 minutes. For high confinement E = 1 MPa, no data (ND) was available as no completed mitosis events was observed. Error bars are SEP.

Figure 5. A penta-daughter cell division may lead to aneuploid daughter cells.

As a qualitative example, this cell is in a partially confined state between 3–7 µm. Extreme confinement conditions observed here can lead to aberrant mitotic orientation resulting in highly asymmetric divisions. These microscopic images show the time lapse of a cell undergoing a single 1→5 successful mitosis. Scale bar: 20 µm.