Micron 42 (2011) 283–289

Contents lists available at ScienceDirect

Micron
journal homepage: www.elsevier.com/locate/micron

Collagen type I amide I band infrared spectroscopy

Benedicto de Campos Vidal ∗ , Maria Luiza S. Mello
Department of Anatomy, Cell Biology, and Physiology and Biophysics, Institute of Biology, University of Campinas (UNICAMP), 13083-863 Campinas (SP), Brazil

a r t i c l e i n f o a b s t r a c t

Article history: Collagen fiber structure and organization have been found to vary in different tendon types. Differences
Received 2 July 2010 have been reported in the FT-IR spectra of the amide I band of collagen-containing structures. In the
Received in revised form present study, the FT-IR spectral characteristics of the amide I band of the bovine flexor tendon and the
16 September 2010
extended rat tail tendon were compared by using the diamond attenuated total reflectance technique.
Accepted 25 September 2010
The objective was to associate FT-IR spectral characteristics in tendons with their different collagen fiber
supraorganization and biomechanical properties. Nylon 6 and poly-l-lysine were used as polyamide
Keywords:
models. Each of these materials was found to exhibit molecular order and crystallinity, as revealed by their
Amide I band
Birefringence
birefringence. The following FT-IR parameters were evaluated: amide I band profile, absorption peaks and
Collagen type I areas, and the 1655 cm−1 /1690 cm−1 absorbance ratio. The amide I area and the 1655 cm−1 /1690 cm−1
FT-IR absorbance ratio were significantly higher for the bovine flexor tendon, indicating that its collagen fibers
Tendons are richer in pyridinoline-type cross-linking, proline and/or hydroxyproline and H-bonding, and that
these fibers are more packed and supraorganizationally ordered than those in the rat tail tendon. This
conclusion is additionally supported by differences in collagen solubility and biochemical/biomechanical
properties of the tendons.
© 2010 Elsevier Ltd. All rights reserved.

1. Introduction 1.1. FT-IR: technique principles

In tendons, collagen bundles are highly organized suprastruc-
tures, although they are differently oriented as a function of the
tendon types and under different physiological and biomechan-
ical demands (Vilarta and Vidal, 1989; Józsa and Kannus, 1997;
Wagnusson et al., 2003; Kjaer, 2004; Ker, 2007; Benjamin et al.,
2008; Vidal and Mello, 2009; Nakagaki et al., 2010). Collagen I is the
predominant macromolecule in tendon collagen structures (Józsa
and Kannus, 1997).
Collagen fiber molecular order and supraorganization have
been extensively studied by polarization microscopy, including
birefringence studies, and by electron microscopy. Advances in
microscopic studies at the molecular level have been made, and
several approaches of the Fourier transform-infrared spectra (FT-
IR) have been reported for different collagen-containing structures
(Lazarev et al., 1985; Payne and Veis, 1988; Twardowski and
Anzenbacher, 1994; Singh, 2000; Camacho et al., 2001; Blank et al.,
2003; Bi et al., 2005; Jastrzebska et al., 2006; Sellaro et al., 2007;
Tiong et al., 2008) (Table 1).
Infrared (IR) spectroscopy is an analytical technique that detects
the vibration characteristics of chemical functional groups in a
sample. A chemical functional group tends to absorb IR radiation
in a specific wavenumber (cm−1 ) range, as bonds and groups of
bonds tend to vibrate at characteristic frequencies. The term Fourier
Transform-IR (FT-IR) refers to the fact that a Fourier transform
algorithm is required to turn the raw data into a spectrum. The
absorption of a beam of IR light passing through a sample is exam-
ined at all wavenumbers at once, and peaks at specific wavenumber
ranges are revealed.
With modern technology available for IR spectroscopy, includ-
ing the use of the Illuminat IR IITM modular FT-IR spectrometers
with light microscope devices, FT-IR spectra for solid samples can
be rapidly obtained. These spectra represent the molecular sig-
nature of a sample and allow the chemical functional bonds in
a molecule to be determined. Complex molecular structures lead
to more absorption bands and more complex spectra. FT-IR spec-
troscopy is thus useful for several analytical purposes including
those in the biological studies.
Detailed principles and applications of FT-IR have been exten-
sively presented elsewhere (Griffiths and Haseth, 2007).

1.2. FT-IR studies on collagen and related peptides

∗ Corresponding author. Tel.: +55 19 3521 6123; fax: +55 19 3521 6185. Collagen dry films show a characteristic FT-IR spectrum, with
E-mail address: camposvi@unicamp.br (B. de Campos Vidal). absorption bands of amide I at ∼1650 cm−1 , amide II at ∼1560 cm−1 ,

0968-4328/$ – see front matter © 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.micron.2010.09.010
284 B. de Campos Vidal, M.L.S. Mello / Micron 42 (2011) 283–289

Table 1 Other models intended to confirm in the assignment of FT-

Amide I band peak frequencies (C O stretching) of collagen and related models
IR spectral characteristics for protein macromolecules have been
according to different reports.
investigated (Bryan et al., 2007). The finding that the factors respon-
Material characteristics Peak frequencies References sible for the amide I fine structure in triple-helical polytripeptides
(cm−1 )
are also active in native collagen (Vigano et al., 2000; Bryan et al.,
Collagen 1655 Camacho et al. 2007) supports the analysis of results obtained for collagen. Thus,
Proteoglycan 1640 (2001) one must consider that, in the case of collagen supraorganization,
Collagen amide I 1654 Twardowski and a higher number of carbonyls is a source of spectral characteristics,
Elastin 1655 Anzenbacher such as amide peak intensity, which are dependent on the protein
(1994) and Singh
Collagen 1654 chain conformation (Lazarev et al., 1985; Bryan et al., 2007). For
(2000)
Elastin 1655 the synthetic peptide LAH4 , the amide I signal is centered around
Human heart valve
Jastrzebska et al. 1657 cm−1 , and the signal for the amide II is present at 1546 cm−1 ,
Aortic 1651
(2006) confirming the predominantly helical structure of this molecule
Mitral 1655
Tricuspid 1651 (Vigano et al., 2000).
Pulmonary 1651 When water molecules are bound to collagen molecules, with
Heart tissue proteins adopting a 1650 Liu et al. (1996) formation of hydrogen bridges, changes in the collagen FT-IR spec-
predominantly ␣-helical trum are introduced (Susi et al., 1971). The spectrum for collagen
configuration
Native rat and cod skin collagens 1630 (shoulder)–
then shows a conspicuous increase in absorptions for amide A and
1656 (central peak) B, 3320 cm−1 (–OH) and 3020 cm−1 (NH stretching) at rates of
Polyproline II in D2 O solution 1632 collagen humidity from 0% to 75%; the most pronounced changes
Lazarev et al.
(20 ◦ C) occur in the amide II region at ∼1550 cm−1 (Susi et al., 1971).
(1985)
Polyproline II dried film 1640
The methods for processing mineralized tissues for FT-IR have
Polyproline II after heating 1650
(Gly-Pro-Ala)n in D2 O solution 1630, 1656 used a peak of 1660 cm−1 /1690 cm−1 area ratio to detect collagen
(80 ◦ C) maturation, and the obtained results have been interpreted as a
Z-(Gly-Pro-Pro)8 -OMe dried film 1640, 1665, 1693 possible difference in soluble to insoluble collagen cross-linking
Z-(Gly-Pro-Gly)4 -OMe dried film 1637, 1655, 1687 (Aparicio et al., 2002). Other authors have proposed the use of a
Native monomeric collagen type 1650 Payne and Veis 1660 cm−1 /1690 cm−1 absorbance ratio instead of an area ratio, as
I at 4 ◦ C (1988) this would be more representative of the pyridinoline to dehydrodi-
Deconvolved native collagen at 1660
hydroxynorleucine (deH-DHLNL) ratio (Blank et al., 2003).
4 ◦ C (triple helix conformation)
4% aqueous gelatin The differences in the frequencies of the maximal FT-IR absorp-
4 ◦C 1641 tion peaks for collagen, collagen-containing tissues, and related
20 ◦ C 1635 models as reported by different authors (Table 1) may be due not
50 ◦ C 1631 only to the specificity of the materials but also to the type of instru-
Collagen
ments and operational methods used for measurements. In this
4 ◦C 1659
20 ◦ C 1659 aspect, the diamond attenuated total reflectance technique uses
50 ◦ C 1637 total internal reflection (ATR 36×, called the evanescent wave) and
Bovine pericardium collagen 1630–1660 generally avoids dispersion and Mie scattering, which are more
Sellaro et al. (2007)
Bovine pericardium collagen 1632
common for ARO objectives (IRRA). It thus provides a more appro-
after changes in helicity caused
by mechanically induced priate approach to the study of collagen I fibers in histological
molecular fatigue sections and mechanically raw-extended tendons.
Collagen ∼1635 Tiong et al.
(2008)

LAH4 peptide (synthetic helical 1657 Vigano et al.
structure) (2000)
and a set of three weaker bands that represent amide III vibration
modes centered at ∼1245 cm−1 (Payne and Veis, 1988).
The amide I band results from the stretching vibration of the
peptide carbonyl group (–C O); the deconvoluted native spectrum
of this band shows three components positioned at ∼1633 cm−1 ,
∼1643 cm−1 , and ∼1660 cm−1 . The ∼1660 cm−1 component is the
most intense (Payne and Veis, 1988). The amide I peak can be sep-
arated into three component peaks (1651 cm−1 , 1642 cm−1 , and
the most intense peak at 1632 cm−1 ) after mechanical stretching
(Payne and Veis, 1988; Sellaro et al., 2007).
Three peaks have also been reported for the amide I region
of polyproline (1650 cm−1 , 1640 cm−1 , and maximal absorption at
1632 cm−1 ) (Lazarev et al., 1985). In this case, decreasing values for
absorbances from 1600 to 1700 cm−1 have been found, and differ-
ences in peak intensity are related to the type of the oligopolymer
model, e.g., polyglycine II (Lazarev et al., 1985). For collagen struc-
ture, the absorption curve profile in an aqueous solution resembles
that of (Gly-Pro-Pro)n (Lazarev et al., 1985).
1.3. Tendon collagen: proposed study by FT-IR
While the bovine flexor tendon is usually subjected to intense
biomechanical stress, the rat tail tendon does not undergo exten-
sive muscular demand. Thus, in comparison, differences in their
collagen I amide I areas, absorptions and absorption ratio from
FT-IR are expected. Indeed, the dissolution of bovine tendons to
obtain collagen requires the presence of pepsin and HCl in an acetic
acid solution and more time than that required for the rat tail
tendon, which needs only a few minutes in 3% acetic acid in aque-
ous solution to induce fiber intumescence (Vidal, 1986, 1995a,b,
2003). These differences in tendon solubility may be caused by
differences in the degree of collagen cross-linking and may be rep-
resented by the FT-IR spectra of these tendons. FT-IR of naturally
supraorganized tendon collagen fibers could be an important step
for interpreting and understanding the role of collagen fibers in the
biomechanical properties of the tendons.
In the present study, the FT-IR spectral characteristics of the
bovine digital flexor tendon were compared to those of the
harsh-squashed extended rat tail tendon using the diamond atten-
uated total reflectance objective (ATR 36×). The objective was to
establish whether previously detected differences in the biochem-
ical/biomechanical properties of these tendons have an effect on
their molecular structure, which in turn may be reflected by their
 B. de Campos Vidal, M.L.S. Mello / Micron 42 (2011) 283–289 285

IR spectroscopic features. These materials were previously checked Table 2

Spectral absorption peaks of the amide I bands for tendon collagen fibers, nylon 6
for their molecular order by optical anisotropy (birefringence).
and poly-l-lysine.

Materials Peak wavenumbers (cm−1 )

2. Materials and methods
2.1. Materials
Cryo-sections of unfixed 10-mm long, 5-mm thick bovine flexor
tendons were prepared. Before freezing in liquid nitrogen, the frag-
ments were treated with glucose and a cryo-protective agent. The
sections were cut parallel to the tendon’s long axis and placed onto
slides, dried in the refrigerator, and transferred to a 37 ◦ C incubator
for complete drying.
Rat tail tendon fibers were obtained from adult male rats (the
animals were part of other experiments that did not interfere with
natural tendon characteristics and properties). Tails were separated
from the animal body and carefully rinsed, dipped in 3% hypochlo-
rite solution for 2 min, and washed in water obtained from a Q-3
Millipore System. The skin of the tail was removed manually; colla-
gen bundles were dissected out from the tendons and mechanically
extended and squashed to attain thin films of fibers as parallel in
orientation as possible. The sections were then dried similarly to
the sections obtained from bovine tendons.
As polyamide models, dried preparations of nylon 6 filaments
(Platinum XT, 0.15 mm in diameter) (Ottoni, Japan) and poly-
l-lysine hydrobromide (Sigma, St. Louis, USA) were used for
comparisons. The choice of nylon 6 as a model of an ordered
polyamide chain was based on the fact that there are international
libraries of its FT-IR spectra from which it is easy and secure to
obtain information on their band width, peak and sharpness char-
acteristics.
2.2. Optical anisotropy
Morphological observations and birefringence images of the
materials were obtained using an Olympus BX-51-P BX2 polarizing
microscope equipped with Image Pro-Plus 6 software as previously
described (Vidal, 2003; Vidal and Mello, 2009).
2.3. FT-IR
FT-IR investigation was performed using the Illuminat IR IITM
microspectrometer (Smiths Detection, Danbury, USA) with a liquid
nitrogen cooled mercury-cadmium-telluride (MCT) detector and
Grams/AI 8.0 spectroscopy software (Thermo Electron Corporation,
Waltham, USA). A diamond ATR objective (magnification 36×) was
employed. The spectroscopic tests (performance validation) for the
equipment used revealed a noise signal ratio (SNR) of 7929:1; the
collected spectra thus showed a low noise level.
The measurement site area was 50 ␮m2 per side, and the
absorbances of the samples and background were measured using
64 scans each. The absorption spectral range was collected between
4000 cm−1 and 650 cm−1 , with a spectral resolution (a measure of
how well closely spaced spectral features were distinguished) of
4 cm−1 . A 1655 cm−1 /1690 cm−1 absorbance ratio and amide I area
ratio representative of the ratio of reducible pyridinoline to dehy-
drodihydroxynorleucine (Aparicio et al., 2002; Blank et al., 2003)
were estimated.
Bovine flexor tendon 1630–1632 1640 1650 1657
Rat tail tendon 1631–1632 1642 1651 1657
Nylon 6 1631 1637
Poly-l-lysine 1618–1620 1636 1648
Nylon 6 and poly-l-lysine were used as polyamide-rich models.
3. Results
3.1. Optical anisotropy
Polarized light microscopy revealed optical anisotropic images
for the collagen fibers from bovine tendon sections and extended
fibers from rat tail tendon (Fig. 1a and b) as well as for nylon 6 and
poly-l-lysine hydrobromide filamentous structures (Fig. 1c and d).
All these materials exhibited intense positive birefringence, which
is a function of their highly ordered molecular structure.
3.2. FT-IR
Taking into consideration that a partial polarization of the inci-
dent infrared radiation might have occurred in the present study,
and that collagen films have been reported to affect the absorp-
tion of infrared radiation (possibly, partially polarized) (Coats et al.,
2003), tests were performed that excluded a possible linear dichro-
ism at amide I under the conditions of the present work (details not
shown).
In general, the analysis of the FT-IR spectrum for the bovine
tendon sections, limited in this study mostly to amide I, showed
that the absorption peak corresponding to the amide I band was
higher than that of the amide II band. Conversely, the absorption
peak of the amide II band was higher than that of the amide I band
in rat extended tail tendons (data not shown).
Differences between the amide I band absorption spectral pro-
file of the two tendon types are evident (Fig. 2). The amide I band for
the collagen fibers of bovine and rat tendons and nylon 6, but not
poly-l-lysine hydrobromide, displayed a maximal peak centered
at 1631–1632 cm−1 (Table 2). This situation was practically main-
tained in collagen fibers of the rat tail tendon and in nylon 6 but
not in the bovine flexor tendon collagen, after peak processing by
Grams/AI 8.0 software (Table 3).
FT-IR differences between collagen fibers of the bovine flexor
tendon and the rat tail tendon were also observed in the amide
I band areas and the 1655 cm−1 /1690 cm−1 absorbance ratio. The
values obtained for both parameters were significantly higher
in collagen fibers of the bovine flexor tendon than those of the
rat tail tendon, as determined by ANOVA-unstacked one way
(Tables 4 and 5).
In comparison to the bovine tendon collagen fibers, the amide I
band profile for nylon 6, a model for polyamide, showed steeper
Table 3
Spectral absorption peaks of the amide I bands for tendon collagen fibers, nylon 6
and poly-l-lysine. The profiles were processed for peak fitting to obtain three amide
I components in tendon fibers.
Materials Peak wavenumbers (cm−1 )

Bovine flexor tendon 1622–1624 1629 1641

2.4. Statistics Rat tail tendon 1631 1632 1659
Nylon 6 1630 1641
Calculations and the ANOVA test were performed with Poly-l-lysine 1615–1620 1639 1646
Minitab12TM software (State College, PA, USA). Nylon 6 and poly-l-lysine were used as polyamide-rich models.
286 B. de Campos Vidal, M.L.S. Mello / Micron 42 (2011) 283–289

Fig. 1. Birefringence image of collagen bundles in a bovine flexor tendon section (a), extended collagen fibers of a rat tail tendon (b), nylon 6 filaments (c) and poly-l-lysine
dry filaments (d). The long axis of the tendon was oriented at 45◦ with respect to the crossed polarizers. In the bovine flexor tendon (a), the variation of birefringence brilliance
is due to differences in the orientation of collagen fibers along the tendon axis. The brilliance (higher optical retardations) is more intense in the more areas in which the
fibers are oriented at 45◦ to the polarizers and the thicker areas of rat tail tendon extended fibers. Bars, 50 ␮m.

Table 4 branches and only one shoulder, although both materials had
Analysis of variance for comparison of the amide I band areas of bovine and rat
their absorption peak centered around 1631–1632 cm−1 (Fig. 3 and
tendon collagen fibers.
Tables 2 and 3). The nylon 6 spectrum was in complete agree-
Materials Mean SD F P ment with those of the nylon 6 USA Blue Line and nylon 6 found
Bovine flexor tendon 15.031 4.034 10.220 0.003 in “Fibers Diamond ATR(c) 2003 ATF Bureau, SensIR Technologies
Rat tail tendon 9.894 4.740 Nylon Fibers (Dewflex-Berkley)”.
n = 15. The absorption spectrum of poly-l-lysine hydrobromide shifted
to shorter wavenumbers than the spectra for bovine flexor
tendon collagen fibers and rat extended tail tendon colla-
Table 5
Analysis of variance for comparison of the 1655 cm−1 /1690 cm−1 absorbance ratio
gen fibers (Fig. 4 and Tables 2 and 3). The most obvious
of the amide I spectral band of bovine versus rat tendon collagen fibers. differences among the absorbances for poly-l-lysine hydrobro-
mide, bovine tendon fibers, and extended rat tail tendon fibers
Materials Mean Std F P
were found in the range of 3387–2900 cm−1 . Absorbances for
Bovine flexor tendon 3.600 0.4378 6.640 0.015 poly-l-lysine hydrobromide were higher and shifted to shorter
Rat tail tendon 3.287 0.2438
wavenumbers compared to those for bovine and rat collagen
n = 17. fibers.
 B. de Campos Vidal, M.L.S. Mello / Micron 42 (2011) 283–289
Fig. 2. FT-IR spectra of bovine flexor tendon (red line) and of extended rat tail tendon
fibers (black line). Y axis, IR absorbances; X axis, wavenumbers in cm−1 .
Fig. 3. FT-IR spectra of nylon 6 fibers (red line) and the bovine flexor tendon (black
line). Both materials show their highest peak at close wavenumbers. The curve for
nylon 6 fiber has conspicuously steeper branches. Y axis, IR absorbances; X axis,
wavenumbers in cm−1 .
4. Discussion
4.1. Optical anisotropy
The birefringence images obtained for the materials examined
with polarized light microscopy demonstrate that the tendon col-
lagen fibers as well as nylon 6 and poly-l-lysine, which were
Fig. 4. FT-IR spectra of poly-l-lysine dry filaments (red line) and bovine flexor
tendons (black line), showing clear differences. Y axis, IR absorbances; X axis,
wavenumbers in cm−1 .
287
subsequently examined by FT-IR, possessed high molecular order
and crystallinity. In the case of the tendon collagen fibers, this
observation is in agreement with our previous reports (Vidal,
1986, 1995a,b, 2003; Vidal and Mello, 2010) and it demon-
strates that the collagen fibers analyzed here did not undergo
denaturation. The macromolecular order demonstrated by optical
anisotropy for nylon 6 and poly-l-lysine was in agreement with
reports by others (Hamza et al., 1991; Fouda, 2002; Lee et al.,
2008).
4.2. Amide I FT-IR peak variation
The multiple structures of the collagen amide I are due to the
heterogeneity of its peptide C O groups in a triple helix, a factor
that directly influences the profile of the amide I band IR spectrum
(Lazarev et al., 1985).
Although the 1630–1635 cm−1 IR peak for collagen has been
attributed to its random coil form (Kaminska and Sionkowska,
1996), this suggestion is not supported by our current find-
ings or studies of other groups (Tiong et al., 2008). In tendons
where denaturation has not occurred and the random coil form
is not expected for collagen, as revealed in the present case by
optical anisotropy, the maximal absorption peak was detected
at ∼1631 cm−1 . The report stating that stretching increases IR
absorption at 1632 cm−1 (Sellaro et al., 2007) can be reconciled
by our findings. Constant biomechanical demands on tendons
require adaptation and modeling that are expected to introduce
changes in their supraorganizational properties, influencing amide
I band complexity in terms of three to four peaks from 1670 to
1631 cm−1 .
According to Liu et al. (1996), “the main amide I absorption in
infarcted tissue is shifted from 1650 to 1628 cm−1 , a shift indicative
of either a structural rearrangement of the existing tissue proteins
or the expression of a new set of proteins with different struc-
tural characteristics”. These authors also suggested that the peak
at 1628 cm−1 is highly characteristic of intermolecular hydrogen
bonding, based on a report by Jackson and Mantsch (1995). The
point of issue here is that, in the present study, the procedure of
peak fitting by the Grams/AI 8.0 software for amide I components
identified the 1629 cm−1 vibration peak as the higher and more
constant peak for bovine tendon sections. These findings thus sup-
port the conclusion that the bovine tendon collagen fibers have high
hydrogen bonding and helical structure organization. It is impor-
tant to note that collagen molecular conformation is peculiar and
not an ␣-helix.
A high molecular order that is fiber-oriented and packed has
already been demonstrated for tendon collagen (Vidal, 1986, 2003;
Vidal and Mello, 2009). In terms of composition, a peak for the
polyproline structure is located near 1633 cm−1 (Jackson and
Mantsch, 1995), excluding the possibility that an absorption peak
at this wavenumber is due to collagen denaturation. Considering
that the 1628–1633 cm−1 region is characteristic of the absorption
of imine carbonyls, it is plausible that in tendons the peak band
at ∼1631 cm−1 could be due to the hydroxyproline concentration
in the collagen fibers. This is in accordance with the idea that “a
growing percentage of imino acid residues results in an increased
intensity of the shoulder at 1630 cm−1 ” (see Fig. 9 in Lazarev et al.,
1985). Furthermore, the collagen molecule is unique with a protein
motif defined by the supercoiling of three polypeptide chains in a
polyproline II conformation (Bryan et al., 2007), which justifies the
above reasoning. In the polyamide models used in this study, band
peaks at 1631 cm−1 and 1618–1620 cm−1 emphasize that care must
be taken in interpreting the possibility that, under these conditions,
the peak at 1631 cm−1 could represent the uncoiling of the collagen
triple helix.
288 B. de Campos Vidal, M.L.S. Mello / Micron 42 (2011) 283–289
4.3. FT-IR differences between tendon collagen and polyamide
models
The nylon 6 FT-IR spectrum obtained in the present study
fit accurately with the spectra of nylon 6 USA Blue Line and
nylon 6 found in “Fibers Diamond ATR(c) 2003 ATF Bureau, Sen-
sIR Technologies Nylon Fibers (Dewflex-Berkley)”. In addition to
the birefringence images observed in the present study, the FT-IR
spectrum demonstrated a highly ordered molecular structure for
the nylon 6 fibers. According to Lee and co-workers (2008), the
amide groups in nylon 6 are oriented approximately perpendicular
to the linear polymer chain axis (the methylene units form linear
chains), a conclusion corroborated by the ␣-form of this crystalline
material.
The differences in FT-IR absorptions between the tendon col-
lagen fibers and nylon 6 are likely caused by differences in the
composition of their amide environment.
The differences in FT-IR absorptions in collagen fibers of bovine
tendon sections and extended rat tail tendons compared to poly-
l-lysine hydrobromide cannot be explained solely by material
thickness or degree of packing. The collagen molecule is a unique
triple helix, and poly-l-lysine hydrobromide is an ␣-helix; in addi-
tion, tendons have a differing distribution of amino acid residues
than poly-l-lysine.
In summary, although one band peak for both collagen and the
polyamide models occurs at ∼1631 cm−1 , the differences between
the collagen fiber amide I band and the nylon 6 and poly-l-lysine
bands are expected based on the fact that collagen fibers have a
much more complex structure than the two polyamide models
used.
4.4. Tendon composition and cross-linking
In tendons, collagen type I accounts for 97–98% of the dry mass
(Józsa and Kannus, 1997); collagen varies between 76% and 88% of
the dry weight of the tendon zone.
Under the present experimental conditions it is logical to accept
that the examined material was not denatured, as it maintained
its original supraorganization. Under such conditions, the amide
I spectra can reflect –C O complex stretching vibration and N–H
bending.
The amide I band areas and the 1660 cm−1 /1690 cm−1
absorbance ratio parameters used in this study are considered
representative of the ratio of reducible pyridinoline and dehydrodi-
hydroxynorleucine, although the wavenumber for calculation of
the absorbance ratio differed slightly from that proposed in the
literature (1660 cm−1 /1690 cm−1 ) (Twardowski and Anzenbacher,
1994; Blank et al., 2003). The choice of a numerator provided by
absorbance at 1655 cm−1 was a function of the spectral profiles
obtained.
Based on the statistical results for the amide I band areas and the
1655 cm−1 /1690 cm−1 absorbance ratio, we conclude that bovine
flexor tendon collagen fibers are richer in pyridinoline type cross-
linking, proline and/or hydroxyproline, and/or hydrogen bonding
than rat tail tendon collagen fibers. Bovine tendon fibers demon-
strated not only a larger amide I area ratio, which is in agreement
with a more ordered supraorganizational rearrangement, but also
a more packed set of fibers than the rat tail tendons. This is prob-
ably due to the fact that bovine flexor tendons are subjected to
greater biomechanical stress than rat tail tendons. Differences in
the organization of collagen macromolecules in these tendons are
therefore expected. One might speculate that aging could be an
important factor in causing the pyridinoline-rich cross-linking. In
spite of this, physical exercise is considered an effective factor in
delaying aging in rat tendon collagen, and tendons of trained ani-
mals are clearly stiffer than untrained animals (Vilarta and Vidal,
1989). Inter-fibrillar proteoglycans may also play a role in the
biomechanical properties of the tendon (Viidik et al., 1996; Singh,
2000; Boskey and Camacho, 2007) as proteoglycans are oriented
in tendons (Vidal, 1965, 1986, 2003; Mello and Vidal, 2003) and
participate in the control of collagen fiber diameter determined by
decorin (Kjaer, 2004; Boskey and Camacho, 2007).
5. Conclusion
In conclusion, the macromolecularly oriented tendon collagen
fibers analyzed in this study showed FT-IR amide band I width
and peaks consistent with the collagen fiber composition and its
supraorganizational characteristics and properties. The obtained
FT-IR amide I spectra were smooth and band peak fitting was log-
ical. Based on FT-IR profile signatures, absorption peaks, areas and
1655 cm−1 /1690 cm−1 absorbance ratio, the collagen fibers of the
bovine flexor tendon are richer in pyridinoline-type cross-linking,
proline and/or hydroxyproline and H-bonding, and more supraor-
ganizationally ordered and packed than those in the rat tail tendon.
This conclusion is in agreement with previously reported differ-
ences in the collagen solubility and biochemical/biomechanical
properties of these tendon types.
Acknowledgments
The supports of Fundação de Amparo à Pesquisa do Estado de
São Paulo (FAPESP, grants no. 2003/04597-0 and 2007/058251-8)
and of the Brazilian Council for Research and Development (CNPq)
are gratefully acknowledged.
References
Aparicio, S., Doty, S.B., Camacho, N.P., Pachalis, E.P., Spevak, L., Mendelsohn, R.,
Boskei, A.L., 2002. Optimal methods for processing mineralized tissues for
Fourier transform infrared microspectroscopy. Calcif. Tissue Int. 70, 422–429.
Benjamin, M., Kaiser, E., Milz, S., 2008. Structure–function relationships: a review. J.
Anat. 212, 211–228.
Bi, X., Li, G., Doty, S.B., Camacho, N.P., 2005. A novel method for determination of
collagen orientation in cartilage by Fourier transform infrared imaging spec-
troscopy (FT-IRIS). Osteoarthr. Cartil. 13, 1050–1058.
Blank, R.D., Baldini, T.H., Kaufman, M., Bailey, S., Gupta, R., Yershov, Y., Boskey, A.L.,
Coppersmith, S.N., Demant, P., Paschalis, E.P., 2003. Spectroscopically deter-
mined collagen Pyr/deH-DHLNL cross-link ratio and crystallinity indices differ
markedly in recombinant congenic mice with divergent calculated bone tissue
strength. Connect. Tissue Res. 44, 134–142.
Boskey, A., Camacho, N.P., 2007. FT-IR imaging of native and tissue-engineered bone
and cartilage. Biomaterials 28, 2465–2478.
Bryan, M.A., Brauner, J.W., Anderie, G., Flach, C.R., Brodsky, B., Mendelsohn, R., 2007.
FTIR studies of collagen model peptides: complementary experimental and
simulation approaches to conformation and unfolding. J. Am. Chem. Soc. 129,
7877–7884.
Camacho, N.P., West, P., Torzilli, P.A., Mendelsohn, R., 2001. FTIR microscopic imag-
ing of collagen and proteoglycan in bovine cartilage. Biopolymers 62, 1–8.
Coats, A.M., Hukins, D.W.L., Imrie, C.T., Aspden, R.M., 2003. Polarization artefacts
of an FTIR microscope and the consequences for intensity measurements on
anisotropic materials. J. Microsc. 211, 63–66.
Fouda, I.M., 2002. Evaluation of the form birefringence and other structural parame-
ters due to thermal annealing for nylon 6 fibers. J. Polym. Res.-Taiwan 9, 37–44.
Griffiths, P.R., Haseth, J.A., 2007. Fourier Transform Infrared Spectrometry. Wiley-
Interscience, Hoboken, USA.
Hamza, A.A., Fouda, I.M., El-Farahaty, K.A., Seisa, E.A., 1991. Optomechanical prop-
erties of fibres. 1. Optical anisotropy in stretched nylon-6 fibers. Polymer Test.
10, 83–90.
Jackson, M., Mantsch, H.H., 1995. The use and misuse of FTIR spectroscopy in the
determination of protein structure. CRC Crit. Rev. Biochem. Mol. Biol. 30, 95–120.
Jastrzebska, M., Zalewska-Rejdak, J., Mróz, I., Barwinski, B.I., Wrzalik, R., Kocot, A.,
Nożynski, J., 2006. Atomic force microscopy and FT-IR spectroscopy investiga-
tions of human heart valves. Gen. Physiol. Biophys. 25, 231–244.
Józsa, L.G., Kannus, P., 1997. Human Tendons Anatomy, Physiology, and Pathology.
Human Kinetics, Champaign, USA, 573 pp.
Kaminska, A., Sionkowska, A., 1996. Effect of UV radiation on the infrared spectra of
collagen. Polymer Degrad. Stab. 51, 19–26.
Ker, R.F., 2007. Mechanics of tendon, from an engineering perspective. Int. J. Fatigue
29, 1001–1009.
Kjaer, M., 2004. Role of the extracellular matrix in adaptation of tendon and skeletal
muscle to mechanical loading. Physiol. Rev. 84, 649–698.
 B. de Campos Vidal, M.L.S. Mello / Micron 42 (2011) 283–289
Lazarev, Y.A., Grishkovsky, B.A., Khromova, T.B., 1985. Amide I band of IR spectrum
and structure of collagen and related polypeptides. Biopolymers 24, 1449–1478.
Lee, K.-H., Kim, K.-W., Pesapane, A., Kim, H.-Y., Rabolt, J.F., 2008. Polarized FT-IR study
of macroscopically oriented electrospun nylon-6 nanofibers. Macromolecules
41, 1494–1498.
Liu, K.-Z., Jackson, M., Sowa, M.G., Ju, H., Dixon, I.M.C., Mantsch, H.H., 1996. Modifi-
cation of the extracellular matrix following myocardial infarction monitored by
FTIR spectroscopy. Biochim. Biophys. Acta 1315, 73–77.
Mello, M.L.S., Vidal, B.C., 2003. Experimental tendon repair: glycosaminoglycan
arrangement in newly synthesized collagen fibers. Cell Mol. Biol. 49, 579–585.
Nakagaki, W.R., Pimentel, E.R., Benevides, G.P., Gomes, L., 2010. The effect of age and
spontaneous exercise on the biochemical and biophysical properties of chicken
superficial digital flexor tendon. Connect. Tissue Res. 51, 265–273.
Payne, K.J., Veis, A., 1988. Fourier transform IR spectroscopy of collagen and
gelatin solutions: deconvolution of the amide I band for conformational studies.
Biopolymers 27, 1749–1760.
Sellaro, T.L., Hildebrand, D., Lu, Q.J., Vyavahare, N., Scott, M., Sacks, M.S., 2007. Effects
of collagen fiber orientation on the response of biologically derived soft tissue
biomaterials to cyclic loading. J. Biomed. Mater. Res. 80A, 194–205.
Singh, B.R., 2000. Infrared Analysis of Peptides and Proteins: Principles and Appli-
cations. Amer. Chem. Soc. Series 750. Oxford Univ. Press, Corby, UK, 240 pp.
Susi, H., Ard, J.S., Carroll, R.J., 1971. The infrared spectrum and water binding of
collagen as a function of relative humidity. Biopolymers 10, 1597–1604.
Tiong, W.H.C., Damodaran, G., Naik, H., Kelly, J.L., Pandit, A., 2008. Enhancing amine
terminals in an amine-deprived collagen matrix. Langmuir 24, 11752–11761.
Twardowski, J., Anzenbacher, P., 1994. Raman and IR Spectroscopy in Biology and
Biochemistry. Ellis Horwood Ltd., Chichester, UK.
Vidal, B.C., 1965. The part played by mucopolysaccharides in the form birefringence
of collagen. Protoplasma 59, 472–479.
289
Vidal, B.C., 1986. Evaluation of carbohydrate role in the molecular order of collagen
bundles. Microphotometric measurements of textural birefringence. Cell Mol.
Biol. 32, 527–531.
Vidal, B.C., 1995a. Crimp as part of a helical structure. C. R. Acad. Sci. Paris Sci. Vie/Life
Sci. 318, 173–178.
Vidal, B.C., 1995b. From collagen type I solution to fibers with helical pattern: a
self-assembly phenomenon. C. R. Acad. Sci. Paris Sci. Vie/Life Sci. 318, 831–
836.
Vidal, B.C., 2003. Image analysis of tendon superstructure using interference and
polarized light microscopy. Micron 34, 423–432.
Vidal, B.C., Mello, M.L.S., 2009. Structural organization of collagen fibers in chordae
tendinae as assessed by optical anisotropic properties and fast Fourier transform.
J. Struct. Biol. 167, 166–175.
Vidal, B.C., Mello, M.L.S., 2010. Optical anisotropy of collagen fibers of rat calcaneal
tendons: an approach to spatially resolved supramolecular organization. Acta
Histochem. 112, 53–61.
Vigano, C., Manciu, L., Buyse, F., Goormaghtigh, E., Ruysschaert, J.-M., 2000. Atten-
uated total reflection IR spectroscopy as a tool to investigate the structure,
orientation and tertiary structure changes in peptides and membrane proteins.
Biopolymers 55, 373–380.
Viidik, A., Nieksen, H.M., Skalicky, M., 1996. Influence of physical exercise on aging
rats: II. Life-long exercise delays aging of tail tendon collagen. Mech. Age. Devel.
88, 139–148.
Vilarta, R., Vidal, B.D., 1989. Anisotropic and biomechanical properties of tendons
modified by exercise and denervation—aggregation and macromolecular order
in collagen bundles. Matrix 9, 55–61.
Wagnusson, S.P., Hansn, P., Kjaer, M., 2003. Tendon properties in relation
to muscular activity and physical training. J. Med. Sci. Sports 13, 211–
223.