Intracellular Measurements of Strain Transfer using Texture Correlation

1
Intracellular Measurements of Strain Transfer
using Texture Correlation

• Christopher L. Gilchrist, Sietske W.
  Witvoet-Braam, Farshid Guilak, and Lori A.
  Setton
• Duke University, Departments of Biomedical
  Engineering and Surgery, North Carolina USA
• Eindhoven University of Technology, Department
  of Biomedical Engineering, The Netherlands

2
Introduction

• Cells experience variety of mechanical stimuli
  in situ
• Fluid flow, hydrostatic pressures, mechanical
  deformations
• Mechanical inputs may regulate
• Survival, proliferation, differentiation,
  metabolism, motility
• Sensing of extracellular mechanical signals
  (i.e.mechanotransduction) may occur at
• Cell Membrane
• Cytoplasm
• Nucleus

3
Introduction

• Studying Intracellular Deformations
• Ligand-coated magnetic beads Maniotis et al,
  1997 Hu et al, 2003
• Single point load or displacement
• Cytoskeletal deformations due to fluid shear
  Helmke et al, 2003
• GFP-labeled intermediate filaments
• Cells subjected to substrate stretch
• Fluorescent beads injected into cytoplasm,
  nuclear dye to estimate overall nucleus
  deformation Caille et al 1998
• Track mitochondria to measure substrate strains
  Wall et al, 2006

Little known about strains transferred to
attached cells at scale of organelles (lt10mm2)
4
Objective

• Present Study
• Determine transfer of deformations from planar,
  elastic substrates to specific cellular regions
  (cytoplasm, nuclei) of adherent cells
• 2D image correlation
• uniaxial substrate stretch

Hypotheses H1 Strain is fully transferred from
substrate to cell cytoplasm and nucleus H2
Amount of strain transferred to cell depends on
cells alignment with direction of stretch
5
Methods Cell Culture
Intervertebral Disc

• Cells isolated enzymatically from anulus
  fibrosus (AF) region of porcine intervertebral
  disc
• AF cells exhibit elongated, fibroblast-like
  morphology in monolayer culture
• Cells cultured in monolayer for 4-10 days prior
  to experiments (F-12 media,10 FBS, antibiotics)

AF region
AF cells, monolayer culture
6
Methods Fluorescent Labeling

• Transparent, elastic silicone membranes (35mm x
  10mm, 250mm thickness) coated with
• Top surface Type I Collagen (40 mg/mL)
• Bottom surface fluorescent mspheres (2mm
  diameter, Ex/Em 535/575nm)
• AF cells seeded on membranes and allowed 48h to
  adhere/spread
• Cells labeled with fluorescent dyes specific for
  
• Cell Nuclei (SYTO 82, 10mM, Ex/Em541/560)
• Cell Mitochondria (Mitotracker, 1.5mM,
  Ex/Em644/655)

7
Methods Cell Stretch
Micrometer

• Membrane placed in custom-built uniaxial stretch
  device
• Reference images acquired of cells, substrate
• Stretch applied to membrane via micrometer
  (strain rate 0.002 s-1)
• Stretched images acquired

Media-filled Chamber
Microscope Objective
Coverslip
8
Methods Imaging

• Cell images
• Mitochondria and nucleus acquired on separate
  channels (63X 1.2NA, slice thickness 2mm mito,
  4mm nucleus)
• Multiple cells (4-6) imaged per membrane
• 36mm2 within each nucleus and cytoplasm analyzed
• Substrate images
• Objective focused on underside of adherent
  membrane
• 270mm X 270mm area

36mm2 area
10mm
10mm
9
Methods Strain Measurement

• Texture Correlation
• Digital image correlation utilizing inherent
  patterns of material image Bay 1995, Wang et al,
  2002
• 2D Texture Correlation (TC) Algorithm Gilchrist
  et al, 2004
• 4x4 grid of tracking points selected for tracking
  (2mm spacing)

• Grid point displacement in stretched image
  determined via 1st-order TC algorithm
• 2D Lagrangian finite strains (Exx, Eyy, Exy)
  calculated from smoothed displacement field

10
Substrate-to-Cell Strain Transfer

• Uniaxial stretch experiments (stretch levels of
  1.05-1.15) performed for 56 cells (10 membranes,
  4-6 cells/membrane)
• Strain Transfer Analysis
• Cell (cytoplasmic, nuclear) and substrate strains
  computed for each cell via texture correlation
• Strain Transfer Ratio (STR)
• H1 Is strain transferred fully from substrate to
  cytoplasm or nucleus? (STR 1?)

11
Results Cell Strains

• Substrate stretch, mean substrate strain Exx
  0.09
• Cytoplasmic strains heterogeneous, magnitudes
  similar to substrate
• Nuclear strains lower than substrate, more
  homogeneous

Box plot box25-75, whiskers90, linemedian
12
Results Cell Strains

• Transverse to applied stretch (Eyy)
• Cytoplasm levels approach underlying substrate
• Nucleus low levels of strain transferred

13
Results Strain Transfer

• STR values in direction of applied stretch
  (STRx)
• Cytoplasm 0.79 lt 1
• Nucleus 0.17 ltlt 1
• Transverse (STRy)
• Similar trends

Different from 0 and 1, plt0.0001
Paired t-test, plt0.0001
14
Results Strain Transfer
Dependence on Substrate Strain Magnitude

• Cytoplasmic strain correlated with substrate
  strain
• slope 1.02, r20.36, plt0.0001
• Nuclear strain not highly correlated with
  substrate strain
• slope 0.25, r2.08, plt0.0001
• Non-zero intercepts indicate offset or nonlinear
  behavior at low strains?

15
Results Strain Transfer
Dependence on Cell Alignment

• Cell alignment measured for each cell (n56)
• Cells binned into aligned and unaligned
  groups
• Strain transfer higher in cytoplasm of cells
  aligned with stretch direction

16
Discussion

• Cell Cytoplasm
• Cytoplasmic strains lower than underlying
  substrate (STRx 0.79)
• Regression suggests 11 relationship with
  substrate strain
• Results similar to recent findings Wall et al,
  2006 where 63 of cytoplasmic strain (STRx0.63)
  transferred to cell cytoplasm

17
Discussion

• Cell Cytoplasm
• Substantial strain heterogeneity within single
  cell, between cells attached to same substrate
• Variabilty in cytoskeletal filament organization
• Microtubule buckling Maniotis et al 1997 Chen
  and Ingber, 1999
• Active remodeling

• Tracking response of single cells may yield
  detailed understanding of strain transfer
  behaviors

18
Discussion

• Cell Nucleus
• Measurable deformation in response to substrate
  strain Maniotis et al 1997 Caille et al 1998,
  Hu et al 2005
• Strain levels less than substrate (0.17) and
  cytoplasm
• Higher nuclear stiffness Dong et al 1991 Guilak
  2000 Caille et al, 2002
• Limited or indirect linkage to substrate

19
Discussion

• Cell Orientation
• Cells aligned with direction of stretch ? higher
  STR for cell cytoplasm
• Findings suggest coupling of cell to substrate
  higher along cells long axis

• Correlate strain transfer with cells
  microstructural architecture via GFP-labeled
  cytoskeletal filaments Helmke et al, 2003,
  adhesion proteins Balaban et al, 2001

20
Applications of Cell Strain Measurement Method

• Examining individual roles of complex
  extracellular inputs on strain transfer (i.e.
  uncoupling inputs)
• ECM ligand Takai et al, 2005, ligand density,
  receptor
• Substrate stiffness Pelham and Wang, 1997
  Discher et al 2005
• Cell prestress Hu et al 2005
• Dimensionality Cukierman et al, 2003

21
Acknowledgments

• Duke University Medical Center
• Steve Johnson, Liufang Jing

Funding Sources National Institutes of Health
(AR47442 and T32 GM08555-09)