Lack of Tissue-Specific Differences in the Distribution of Type I Collagen Fibril Morphologies
+1Wallace, J M; 1Chen Q; 1Fang M; 1Erickson B; 1Orr B G; 1Banaszak Holl, M M
+1University of Michigan, Ann Arbor, MI
[email protected]
INTRODUCTION
Type I collagen, the most abundant form of the most abundant protein
in mammals, forms the structural scaffolding upon which many tissues
are built. Based on the seminal work of Hodge and Petruska in 1963
(and earlier work by many others), the primary morphological
characteristic of Type I collagen fibrils, the D-periodic axial gap/overlap
spacing, was shown to be 67 nm based on theoretical models of a single
collagen fibril in isolation. In the 46 years since this assertion, x-ray
diffraction and electron microscopy work support this singular spacing
value within the error of the individual technique. However, given the
complexity of the collagen fibril itself and the range of tissues in which
these fibrils are incorporated, it is unlikely that a single spacing value for
all fibrils exists.
Because of the wide-ranging roles that collagen-based tissues play
and the number of diseases that can strike these tissues, measuring and
understanding true morphological features of the ultrastructure of
collagen-based tissues is imperative. The need for accurate quantitative
analytical methods to assess collagen’s nanoscale morphology and
mechanical integrity with as little disruption to the tissue as possible has
prompted us to study the collagen ultrastructure of various tissues in situ
using atomic force microscopy (AFM). Samples imaged using AFM can
remain intact and in their native state, implying that measured properties
are samples characteristics rather than artifacts of processing or imaging.
We hypothesized that using AFM, we could image and quantitatively
analyze the morphology of Type I collagen fibrils in fully intact and
mineralized bones and teeth as well as in non-mineralized tendon in situ
to learn more about the normal nanoscale properties of these materials.
METHODS
Male mice from a mixed background strain (Sv129/CD-1/C57BL/6S)
were maintained until 8 weeks of age (UCUCA protocol #09637). After
sacrifice, femora, mandibular incisors and tails were harvested, wrapped
in gauze soaked with saline and stored at -20°C. Before use, the
proximal and distal ends of each left femur were removed leaving
compact bone of the diaphysis. Mandibular incisors were cleaned of
soft tissue and surrounding bone. Bones (anterior side facing up) and
teeth (lateral side facing up) were mounted to a steel disk and a flat
polished surface was created using a 3 µm diamond suspension and a
0.05 µm alumina suspension. To remove extrafibrillar mineral, the
surface of each bone and tooth were treated for up to 15 minutes with
0.5M EDTA (pH=8), then vigorously rinsed with ultrapure water and
soaked at 4°C for at least 16 hours. Before imaging, each sample was
briefly sonicated to remove surface bound mineral. From each tail,
tendons were removed and homogenized in ultrapure water to disrupt
the fascicle structure and release collagen fibrils. Fibril-containing
solutions were deposited onto freshly-cleaved mica and allowed to dry.
Samples were imaged in air using a PicoPlus 5500 AFM (line scan
rates of 2 Hz or lower, 512 lines per frame). Images were acquired in
tapping mode from 9 axial locations in each bone (5-10 fibrils per
location) and 3 axial locations in each tooth (15-20 fibrils per location).
35-50 fibrils from each tail sample were imaged in contact mode. At
each location, 3.5 µm x 3.5 µm amplitude (error) images were analyzed.
Two dimensional Fast Fourier Transforms (2D FFT) were performed on
individual fibrils and the first harmonic peak was analyzed to determine
the value of the Dperiodic spacing
(Figure 1).
To investigate
tissue-specific
differences in fibril
morphology, the Dperiodic spacing
Figure 1
was compared.
Values measured
from an individual sample were pooled, yielding an average value for
that sample. The values from dentin (n=5), bone (n=4) and tendon (n=4)
were then compared using One Way ANOVA with post-hoc Bonferroni
tests. To investigate differences in the distributions of fibril
morphology, a Kolmogorov-Smirnov (K-S) test was performed on the
cumulative density functions (CDF) calculated from the kernel density
of each distribution. For all investigations, a value of p < 0.05 was
considered significant.
RESULTS
Measurements within each sample (1 tooth, 1 bone, 1 tail) were
pooled to yield the mean fibril spacing for that sample. The overall
mean values within in each tissue type were 67.77 nm, 67.28 nm and
67.87 nm for dentin, bone and tendon, respectively. Figure 2a displays
these data as boxplots. The boxes represent the middle 50% of the data
and the whiskers depict the data extremes. The diamond is the mean and
the line within the box is the median of each group. The dashed
horizontal line in Figure 2a corresponds to the 67 nm value predicted by
the Hodge-Petruska model. The mean spacing was not significantly
different in any of the groups according to One Way ANOVA.
Figure 2b shows that a distribution of spacings existed in each group.
A comparison of different regions of the histograms suggests that
different populations of fibrils existed within the groups. For dentin,
85.9% of all fibrils had spacings within 1 standard deviation of the mean
(between 66 nm and 69 nm), compared with 77.4% of bone and 58.8%
of tendon fibrils for the same 66-69 nm region. Fibrils outside of this
range constituted 14.1%, 22.6% and 41.3% of fibrils in the dentin, bone
and tendon samples, respectively. However, when compared using a KS test (inset in Figure 2b), the CDF from none of the groups were
significantly different, indicating that no significant differences were
present in population distributions (Dentin vs. Bone, p=0.940; Dentin vs.
Tendon, p=0.305; Bone vs. Tendon; p=0.478).
Figure 2
DISCUSSION
In the current study, bone, dentin and tendon samples from the same
animals were analyzed. These 3 tissues were chosen for their
differences in function, as well as to compare mineralized tissues with a
tissue which lacks mineral in a normal physiological setting. The
presence of a distribution of D-period spacings in all 3 tissues is an
important observation which confirms that this type of distribution is
fundamental to Type I collagen morphology. Further, the observation
verifies that the presence of mineral is not responsible for the
distributions in bone and dentin.
Although no significant differences existed in mean D-period spacing
or in the spacing distributions between the three tissues, slight variations
in the location of populations were noted (Figure 2b). The most
compelling of these differences existed between tendon and the
mineralized tissues. Specifically, 41.3% percent of the fibrils in the
tendon samples fell outside of a 66-69 nm range (chosen to contain ± 1
SD of the dentin samples) compared with smaller percentages for the
other tissues (14.1% and 22.6%). The most likely explanation for these
subtle differences may lie in tissue-specific differences in cross-linking.
Our data indicate that bone, dentin and tendon contain distributions of
Type I collagen morphologies and that these distributions were similar
between tissues. The concept of a distribution is often overlooked in
measurements of collagen, and the mean value for the D-periodic
spacing is reported without explanation. However, the influences that a
distribution of collagen morphologies could have on the mechanical
properties of a tissue are important, as are differences in these
distributions between tissues and in cases of diseases. These links
between morphology and mechanics will be investigated in future
studies.
Poster No. 562 • 56th Annual Meeting of the Orthopaedic Research Society