Biomedical Engineering Reference
In-Depth Information
explain the additional variation in
E
. We found that the relationship
E
1 0 0 6
provided
a best fit that explained the most variation in
E
(Figure 11.12a,
R
2
= 0.80). Similarly, the relationship
E
~
skew skew
−
.
−
.
SHG
TPF,matrix
0 7 1 8
provided a best fit that explained the most variation in
E
(Figure 11.12b,
R
2
= 0.83).
The observation of the linear and log-log plots of the multiple regressions shows that the nonlinear
model using the skewness parameters tends to overestimate
E
of gels with sparse matrix (days 0-3,
Figure 11.12a, inset), whereas the SC nonlinear model tends to overestimate
E
of uncross-linked gels
cultured for 16 days (day 15, Figure 11.12b, inset).
~
SC
−
.
SC
−
.
SHG
TPF,matrix
11.4.6 Summary
Microstructural parameters change systematically during cell-mediated gel contraction: pores become
smaller, fiber bundles become larger, and cells occupy holes in a dense three-dimensional collagen net-
work. For cellularized collagen gels, SHG image parameters such as skewness and SC change with colla-
gen fiber density in a predictable manner. Cellularized gel
E
can largely be predicted by the variation in
SHG and matrix-derived TPF image parameters (skewness and SC), which depend upon collagen fiber
and cross-link spatial patterns.
11.5 conclusions
MPM is a promising imaging tool that is currently being adapted for use with fiber-optic handheld
probes and scanning modules. It has become the premier technique for imaging of living cells and cul-
tures of excised and engineered tissues. There is a great deal of interest in the link between extracellular
matrix microstructure and bulk tissue mechanical properties. Researchers are beginning to apply the
knowledge of tissue microstructure derived from MPM for understanding the development of mechani-
cal properties in tissues such as cellularized and acellular silk, the visceral pericardium [103], and the
fibrous cap of atherosclerosis.
In this burgeoning research environment, there is a need to develop simple and robust methods for
data mining of MPM images to extract all mechanically relevant information. Such mechanically rele-
vant information includes the concentration of mechanically relevant species (e.g., collagen), the volume
fraction of the species, the pore size of polymer networks, the fluorescent cross-link content of tissues,
the diameter and length of fibers, and in general, the size and abundance of mechanically relevant struc-
tures, and the spatial distribution of species in three dimensions. This chapter focused on the measure-
ment of mechanically relevant image parameters from SHG signals of acellular and cellularized gels.
Collagen gels of similar concentrations may nonetheless possess varied microstructure and resulting
bulk mechanics. Signal area fractions, pore size measurements, and fiber diameter measurements from
SHG images robustly and effectively estimate the aspects of collagen network microstructure that influ-
ence the bulk shear moduli.
MPM imaging of cellularized gels contracting through a range of collagen concentrations from 4
to 200 mg/mL revealed problems with the robustness of signal image fraction and pore size measure-
ments. The optical resolution of LSM limits microstructural information extractable from SHG images,
especially from concentrated gels (>60 mg/mL). The pores for these gels were smaller than the pixel res-
olution limit (~0.3 μm
3
) and could not be quantified. Nevertheless, SHG signal and image parameters
are sensitive to a wide range of collagen network concentrations. MPM imaging of SHG signal provides
an unparalleled noninvasive approach for studying microstructure-mechanics and cell-matrix interac-
tions in collagen gel-based engineered tissues.
Acknowledgments
This work was supported, in part, by the National Heart, Lung, and Blood Institute (R01 HL067954, SCG), the
Air Force Office of Scientific Research (FA9550-04-1-0101). CBR was supported by a Kirschstein predoctoral
