Influence of semicrystalline order on the second-harmonic generation efficiency in the anisotropic bands of myocytes
The influence of semicrystalline order on the second-harmonic generation (SHG) efficiency in the anisotropic bands of Drosophila melanogaster sarcomeres from larval and adult muscle has been investigated. Differences in the semicrystalline order were obtained by using wild-type and mutant strains containing different amounts of headless myosin. The reduction in semicrystalline order without altering the chemical composition of myofibrils was achieved by observing highly stretched sarcomeres and by inducing a loss of viability in myocytes. In all cases the reduction of semicrystalline order in anisotropic bands of myocytes resulted in a substantial decrease in SHG. Second-harmonic imaging during periodic contractions of myocytes revealed higher intensities when sarcomeres were in the relaxed state compared with the contracted state. This study demonstrates that an ordered semicrystalline arrangement of anisotropic bands plays a determining role in the efficiency of SHG in myocytes.
Inﬂuence of semicrystalline order on the second-harmonic
generation efﬁciency in the anisotropic bands of myocytes
Catherine Greenhalgh, Nicole Prent, Chantal Green, Richard Cisek, Arkady Major, Bryan Stewart,
and Virginijus Barzda
The inﬂuence of semicrystalline order on the second-harmonic generation (SHG) efﬁciency in the aniso-
tropic bands of Drosophila melanogaster sarcomeres from larval and adult muscle has been investigated.
Differences in the semicrystalline order were obtained by using wild-type and mutant strains containing
different amounts of headless myosin. The reduction in semicrystalline order without altering the
chemical composition of myoﬁbrils was achieved by observing highly stretched sarcomeres and by
inducing a loss of viability in myocytes. In all cases the reduction of semicrystalline order in anisotropic
bands of myocytes resulted in a substantial decrease in SHG. Second-harmonic imaging during periodic
contractions of myocytes revealed higher intensities when sarcomeres were in the relaxed state compared
with the contracted state. This study demonstrates that an ordered semicrystalline arrangement of
anisotropic bands plays a determining role in the efﬁciency of SHG in myocytes. © 2007 Optical Society
OCIS codes: 170.3880, 180.5810, 190.4160.
With the advancement of microscopic imaging technol-
ogy a growing number of nonlinear optical processes is
being explored for biological imaging.
excitation ﬂuorescence (MPF) microscopy is currently
the most commonly used nonlinear contrast mecha-
However, interest in second-harmonic gen-
eration (SHG), third-harmonic generation (THG), and
coherent anti-Stokes Raman scattering microscopy is
growing because of an increasing number of biological
and medical applications. Unlike MPF imaging, which
is prone to signal degradation due to ﬂuorophore pho-
tobleaching, harmonic generation microscopy is free of
or has a largely reduced photobleaching, owing to the
possibility of generating harmonics outside of the lin-
ear and nonlinear absorption regions.
the harmonics are produced intrinsically within the
sample, and as such do not require labeling. These
unique properties of harmonic generation allow the
imaging of biological samples for prolonged periods of
time with negligible tissue damage.
SHG microscopy has been used for several biological
applications including the imaging of collagen,
cently, muscle was imaged with high-resolution SHG
A second harmonic was found to be
generated from the anisotropic bands of sarcomeres,
excluding the bare region in the center resulting in a
double peak signal. When combined with THG mi-
croscopy, the somite interfaces from myocytes can be
seen in the third harmonic alongside the sarcomeres
visualized with SHG.
Additionally, Chu et al.
demonstrated that with a high NA objective and a
wavelength of 1230 nm, the isotropic bands could be
observed in the third harmonic while imaging the
anisotropic bands with the second harmonic. In ad-
dition to SHG from the anisotropic bands, THG re-
vealed cardiomyocyte mitochondria when imaged
with 1064 nm excitation,
and the trachea near the
Drosophila melanogaster larval muscle when imaged
at 800 nm.
Structurally, myocytes contain longitudinal myoﬁ-
brils that are mainly made up of titin, and myosin
and actin myoﬁlaments. Myoﬁbrils have a unique
All the authors are with the University of Toronto at Missis-
sauga, 3359 Mississauga Road, Mississauga, Ontario L5L1C6,
Canada. C. Greenhalgh, N. Prent, C. Green, R. Cisek, A. Major,
and V. Barzda (firstname.lastname@example.org) are with the Depart-
ment of Chemical and Physical Sciences, and B. Stewart is with the
Department of Biology.
Received 5 July 2006; revised 20 November 2006; accepted 22
November 2006; posted 29 November 2006 (Doc. ID 72649); pub-
lished 13 March 2007.
© 2007 Optical Society of America
1852 APPLIED OPTICS 兾 Vol. 46, No. 10 兾 1 April 2007
striated structure, which can be clearly observed with
polarization microscopy. The fundamental unit of the
myoﬁbril is called the sarcomere (see Fig. 1). The
sarcomeres are made up of two main regions that can
be clearly visualized in a polarization microscope: the
isotropic band (I-band) and the anisotropic band (A-
band). The A-band consists of the overlapping zones
of actin and myosin myoﬁlaments that exhibit bire-
fringence, while the regions containing actin and titin
have no birefringence and are labeled as I-bands (Fig.
1). The thick myosin ﬁlaments are made up of rods
formed from the chiral myosin heavy chains (MHC)
and heads made up of the light chains. In the trans-
verse direction, electron microscopy has shown that
formation of the sarcomeres exhibits hexagonal sym-
In the overlapping region between actin
and myosin, each myosin ﬁlament is surrounded by
six actin ﬁlaments, and each actin ﬁlament is sur-
rounded by three myosin ﬁlaments. The sarcomere
size, as measured between the Z-lines is ⬃2–3 m for
Drosophila melanogaster adults and ⬃5–8 m for
Hence Drosophila larval sarcomeres
are ideally suited for understanding the origin of
SHG from myocytes because they are two to three
times larger than typical mammalian skeletal and
heart myocytes. Furthermore, the variety of Drosoph-
ila mutants with different myosin content and al-
tered structural integrity of A-bands provides a good
model to study the inﬂuence of A-band structure on
the efﬁciency of SHG. Several studies on the SHG of
muscle have been carried out.
It has been
shown that mutations of myosin in Caenorhabditis
elegans leads to a signiﬁcant distortion of the striated
structure of myocytes, pointing out that myosin is
involved in the structures generating SHG.
further established that SHG requires myosin but
not actin, and evidence was provided that SHG does
not vary with the concentration or orientation of my-
The tensor elements of the second-order
nonlinear susceptibility of myoﬁbrils were elucidated
by imaging the sample at different orientations with
respect to the linear polarization of the laser excita-
The semicrystalline structure of the myosin
ﬁlaments has been modeled using hexagonal symme-
try. In a similar experiment it was found that the
orientation of the harmonophore responsible for SHG
appears to be similar to the pitch of the ␣-helix of the
myosin rod along the thick ﬁlament axis.
ing SHG during mechanical stretching of myoﬁbrils in
skeletal myocytes, a strong increase in SHG signal
intensity was observed with the sarcomere lengthen-
Structural changes during contraction were also
shown to effect the width of the SHG bands.
It has been well established that noncentrosym-
metric media are required for generation of the sec-
The noncentrosymmetric media
could result from chiral structures, bulk crystals with
no inversion symmetry, or molecules with a strong
hyperpolarizability (␤) arranged in noncentrosym-
metric structures or on surfaces that result in a break
in symmetry. For SHG to become a viable imaging
and diagnostic tool in myocyte research, the noncen-
trosymmetric media of anisotropic bands responsible
for the SHG must be well understood. The anisotropic
band of sarcomeres is complex, and a full character-
ization of the origin of the second harmonic from
within this region is still under debate; some attempts
to elucidate the molecular origin have suggested that
myosin rod domains are the key structures giving
while others have interpreted that the packed
myosin heads in sarcomere thick ﬁlaments are re-
sponsible for the large second-harmonic endoge-
In addition to the molecular origin of SHG, the
macro-organization of the molecules in the semicrys-
talline structure of anisotropic bands of sarcomeres
plays a crucial role for SHG efﬁciency.
SHG is not observed in a random suspension of mole-
cules, even when they have a high hyperpolarizability
␤, because of the destructive interference of coherent
radiation from the oppositely oriented molecules.
ordered structures, coherent second-harmonic radia-
tion from molecules interferes constructively over a
distance given by the phase-matching conditions, ren-
dering a strong SHG signal in the far ﬁeld. The large
cone of rays produced by the high NA microscopic ob-
jective usually ensures phase-matching conditions for
some propagation directions.
Here we examine the extent to which the semi-
crystalline ordering of anisotropic bands inﬂuences
SHG generation efﬁciency in the Drosophila mela-
nogaster myocytes. SHG imaging of mutant strains
suggests that the efﬁciency of the second harmonic
is strongly dependent on the extent of the semi-
crystalline arrangement of the anisotropic bands. Ob-
servations show that stress exerted on sarcomeres
attributable to contractions in neighboring myoﬁbrils
signiﬁcantly reduces the efﬁciency of the SHG. The
investigations show that disorder in the semicrystal-
line arrangement of the anisotropic bands radically
inﬂuences the SHG efﬁciency, and demonstrates a
decrease in the SHG efﬁciency during muscle con-
Fig. 1. Structural schematic of two sarcomeres. The main actin
(thin) and myosin (thick) ﬁlaments are shown. The anisotropic
band (A-band) is the region where myosin is located, while the
isotropic region (I-band) is occupied by actin and titin. The small
bare region in the center of the myosin, where there are no myosin
heads, is called the H-zone. The Z-line is located in the middle of
1 April 2007 兾 Vol. 46, No. 10 兾 APPLIED OPTICS 1853
2. Materials and Methods
A. Microscope Setup
Two femtosecond lasers were used as an excita-
tion source for microscopic imaging. A home-built ex-
tended cavity Ti:sapphire oscillator emitting ⬃25 fs
pulses at a 26.7 MHz repetition rate was used mainly
when MPF imaging of actin labeled with coumarin-
conjugated phalloidin was conducted (see Fig. 4 be-
low). The emission wavelength of this laser was
tunable between 780 and 840 nm, and no more than
1 nJ of incident energy per pulse was used for imag-
ing. For higher signal-to-noise ratio imaging, less
invasive excitation at 1042 nm was provided by a
diode-pumped Yb-ion-doped potassium gadolinium
tungstate (Yb:KGW) laser. The full description of the
laser is presented elsewhere.
The home-built oscil-
lator delivered ⬃200 fs pulses at a repetition rate of
14.6 MHz. Up to 2 nJ of incident energy per pulse
could be used for microscopic imaging without dam-
aging the sample. Both lasers permitted imaging for
long periods of time with no observable damage to the
The Ti:sapphire or Yb:KGW femtosecond oscillator
was coupled to a home-built microscope (see Fig. 2).
The multimodal nonlinear microscope has been pre-
viously described elsewhere.
Brieﬂy, 2D images
were obtained by rastering the laser beam by two
closed-loop galvanometric mirrors (GSI Lumonics,
VM500 Series), which are represented by SM in
Fig. 2. Typically a 20⫻ 0.75 NA microscope objective
(Zeiss) was used to focus the fundamental light into
the sample. For acquisition of high-resolution 3D vol-
umes, an oil immersion 1.3 NA objective (Zeiss) was
used. The 0.75 NA was most often used due to a
larger working distance permitting deeper imaging
into the tissue. In the forward direction, the har-
monics were collected with a home-built UV trans-
mitting objective, whereas the MPF signal was
collected in the backward direction with the same
excitation objective. To obtain 3D sectioning, the sam-
ple was translated axially via a piezoelectric transla-
tion stage (Dynamic Structures and Materials, FPA-
500) with an optical encoder (MicroE Systems) for
position feedback. The emitted photons were detected
with photomultiplier tubes (PMTs) (Hamamatsu
model or H5783P-01, 03, and -06). The data acquisi-
tion electronics featured a photon-counting detection
method with simultaneous three-channel recording
(NI-6602, National Instruments). Both scanning and
detection was controlled with a LabVIEW interface
(National Instruments) that was created speciﬁcally
for this imaging system. Acquisition rates were ap-
proximately 8 frames兾s. Typical acquisition times for
the whole 3D stack were of the order of several min-
utes, depending on the desired resolution and signal-
to-noise ratio. The SHG signal strength was sufﬁcient
for using individual frames, but in general, several
frames were averaged to improve the contrast and
signal-to-noise ratio. Three-dimensional stacks of im-
ages collected with the nonlinear microscope were
analyzed with ImageJ software, and 3D volume ren-
dering was performed with home-developed software
based on a VolumePro 500 (TeraRecon) rendering
card. Confocal ﬂuorescence images were acquired with
a laser scanning microscope (Zeiss LSM 510) using the
543 nm line of a He–Ne laser.
B. Sample Preparation
Drosophila melanogaster stocks were raised on Bloom-
ington medium at 25 °C and the yw strain was used
throughout these experiments. The following mutant
strains were also used (gift of S. Bernstein, San Diego
State University): Mhc
is an allele in which no my-
osin heavy chain is produced in the adult indirect
ﬂight muscles (IFMs); Y97 is a transgenic line that
expresses headless myosin heavy chain in the IFMs
in a wild-type background;Mhc
; Y97 is a strain that
expresses headless myosin heavy chain in the IFMs
in the Mhc
background. Third instar larvae or
adults were selected and dissected under Drosophila
saline. Adult IFMs were lightly ﬁxed in 4% formal-
dehyde for 3 min to aid dissection and visualization.
For actin labeling, dissected muscles were ﬁxed,
washed in phosphate buffered saline plus 0.1% Triton
X-100 (PBT) and then incubated overnight in a
1:1500 dilution of rhodamine-conjugated phalloidin
for the confocal scans or coumarin-conjugated phal-
loidin (Sigma) for the multiphoton scans with 800 nm
excitation. Adult samples were washed in PBT and
mounted on glass slides in Vectashield.
3. Results and Discussion
A. Structure of Drosophila Myocytes
To visualize the structure of Drosophila myocytes,
high-resolution images were collected at 1042 nm ex-
citation. Figure 3 presents a second-harmonic image
Fig. 2. Nonlinear multimodal microscope setup. The OE and OC
represent excitation and collection objectives, respectively; D1 and
D2 are dichroic mirrors, M1 is a mirror, and SM represents two
galvanometric scanning mirrors. Additionally, optical ﬁlters F1,
F2, and F3 are placed in front of the PMT detectors to isolate
wavelengths of interest. (See text for a more detailed description.)
1854 APPLIED OPTICS 兾 Vol. 46, No. 10 兾 1 April 2007
from both larval and adult muscle. In Fig. 3(a), the
larval muscle shows the periodic striated structure,
which is characteristic of the sarcomere anisotropic
bands. However, unlike the adult IFMs seen in Fig.
3(b) the sarcomere size is much larger. Drosophila
larval muscle has a sarcomere size ranging from 5 to
8 m, whereas adult Drosophila IFM have a typical
sarcomere size of ⬃2 m, which is comparable to
other myocytes, such as cardiomyocytes and skeletal
muscles (see Section 1 for examples). The large differ-
ence in size is attributable mainly to the wider I-band
observed in larval myocytes; the widths of the SHG
bands are approximately equal for both specimens.
The large sarcomere size of larval myocytes is ad-
vantageous for high-resolution imaging. Two- and
three-dimensional images of larval muscle in Figs. 3(a)
and 3(c) display more details than the corresponding
adult IFM myocyte image [see Figs. 3(b) and 3(d)] re-
corded under the same magniﬁcation. The large I-band
in the larval muscle enables visualization of the inter-
connections of the myoﬁbrils. Staircaselike structures,
shown in Fig. 3(b), are routinely observed in both lar-
val and adult myocytes. Both et al.
served similar structures in mouse skeletal muscle
Initially, the origin of the second harmonic from
Drosophila larvae muscle was veriﬁed as originat-
ing from the anisotropic band (A-band, see Fig. 1)
by comparing images acquired simultaneously with
SHG and MPF at 800 nm excitation. A similar tech-
nique has been successfully used to investigate SHG
in skeletal and cardiac myocytes.
ﬂuorescence from actin was achieved by labeling the
samples with coumarin-conjugated phalloidin. Fig-
ure 4 depicts the same region of myoﬁbril imaged
simultaneously with (a) SHG and (b) MPF . The most
intense ﬂuorescence signal corresponds to the Z-line of
the I-band. The lowest MPF intensity is seen in the
actin void region called the H-zone. The apparent MPF
signal in the H-zone comes into view because of point
spread function broadening effects from the neighbor-
ing actin ﬁlaments. Since the second harmonic is gen-
erated from speciﬁc endogenous structures, the signal
is spatially conﬁned and exhibits excellent contrast
Fig. 3. (Multimedia online; ao.osa.org) SHG microscopy images of
Drosophila melanogaster larval muscle [(a) and (c)] and adult indi-
rect ﬂight muscle [(b) and (d)]. Two-dimensional images of the (a)
larval and (b) adult muscle are presented. The stacks of images were
rendered for the (c) larval and (d) adult myocytes. A conventional
ﬂuorescence image of (larval) muscle with actin labeled by
rhodamine-conjugated phalloidin is shown in (e) for comparison.
Images (a)–(d) were taken with 1042 nm excitation using a 1.3 NA
oil immersion objective. Scale bar represents 5 m in (a)–(d) and
100 m in (e).
Fig. 4. Comparison of SHG and MPF signals from Drosophila
melanogaster larvae muscle. Typical images from Drosophila lar-
vae muscle recorded with (a) SHG and (b) MPF microscopy at
800 nm excitation. A proﬁle of each signal across the row of sar-
comeres, the line shown in (a) and (b), is presented in (c). The solid
curve with circles represents the SHG proﬁle; the dotted curve
with open triangles shows the MPF proﬁle. Images were taken
with ⬃800 nm excitation using a 0.75 NA air objective.
1 April 2007 兾 Vol. 46, No. 10 兾 APPLIED OPTICS 1855
and signal-to-noise ratio. The second-harmonic signal
exhibits periodic bandings with the double lines
characteristic of the anisotropic bands of sarco-
meres. In comparing the intensity proﬁles of SHG
and MPF along the row of sarcomeres, the signal
proﬁles are shown to anticorrelate [see Fig. 4(b)];
this provides evidence that the observed second
harmonic is being generated from the anisotropic
bands of the Drosophila larval sarcomeres. Previ-
ous studies have explored the origin of SHG in fur-
The remaining sections of this paper
focus on how the SHG intensity is affected by var-
ious structural alterations.
B. Effect of Mutations on the Second-Harmonic
Generation Intensity of Anisotropic Bands
For exploring how the structure of A-bands affects
the SHG efﬁciency, adult Drosophila IFMs from
wild-type and three different mutant strains were
used to investigate the SHG response. One strain of
, develops IFMs that contain no myo-
sin heavy chain which leads to a highly distorted
sarcomere structure. Additionally, we looked at a mu-
tant strain that has headless myosin in both a wild-
type (Y97) and a Mhc
Figures 5(a)–5(d) compare the wild-type adult Dro-
sophila myocytes imaged with second-harmonic gen-
eration microscopy [Fig. 5(a)] with the SHG from each
of the three mutant strains [Figs. 5(b)–5(d)]. For com-
parison, Figs. 5(e)–5(h) show typical confocal images of
the ﬂuorescence from actin labeled with rhodamine-
conjugated phalloidin for the four sample types. As
expected, no signiﬁcant SHG is generated in the Mhc
samples that do not contain myosin [Fig. 5(d)]; in ad-
dition, highly distorted actin structures can be seen in
the ﬂuorescence image [Fig. 5(h)]. This is in agreement
with a study by Plotnikov et al.
that observed a dra-
matic decrease in SHG when a disruption of the myo-
sin ﬁlaments was induced by incubating myocytes
with a low-ionic strength pyrophosphate solution. In-
terestingly, SHG is generated in both mutants that
contain headless myosin [Figs. 5(b) and 5(c)]; however,
the mutant in the Mhc
background exhibits a more
distorted structure compared with the mutant in the
wild-type background. As a control, THG from struc-
tures located in the myocytes was utilized (the im-
ages are not shown); however, here we do not directly
address the THG imaging. In Fig. 5 we see that in
both the SHG and the FL images, the structural dis-
order of the sarcomeres increases as the mutation
becomes more severe. This agrees with transverse
and longitudinal electron microscopy (EM) images of
the same muscle types.
Although both Y97 and Mhc
; Y97 mutants gen-
erate the second harmonic, their SHG intensities
are signiﬁcantly different. Figure 5(i) shows the
average SHG intensities for the different sample
types. The wild-type muscles, which have the high-
est ordered structure, have the highest efﬁciency of
the SHG. In looking at the three mutant strains,
one sees that as the severity of the mutation in-
creases the content of the native myosin decreases
and, consequently, the average SHG intensity de-
creases. The absence of myosin in the Mhc
results in diminution of the ordered sarcomere struc-
ture and an almost complete loss of the SHG signal.
Our observations correspond well with the previous
study by Campagnola et al.
where mutations in
MHCBofC. elegans resulted in a major distortion of
striated structure and a substantial decrease in the
There are three possible explanations for the
variations in the SHG intensity of different mu-
tants. First, if the molecular hyperpolarizability ␤
of native myosin is higher than the headless myo-
sin, a reduction in the concentration of native my-
osin could result in a decrease in SHG intensity.
Second, the molecular hyperpolarizabilities might be
similar for native and headless myosin, but the con-
centration of myosin is decreased with the severity
of mutation. Consequently, a lower concentration of
Fig. 5. Analysis of SHG from IFM myocytes of different Drosoph-
ila melanogaster mutants. (a)–(d) correspond to SHG images: (a)
wild-type, (b) Y97, (c) Mhc
; Y97, and (d) Mhc
. For comparison,
confocal images of similar preparations of Drosophila myocytes
labeled with rhodamine-conjugated phalloidin are also presented:
(e) wild-type, (f) Y97, (g) Mhc
; Y97, and (h) Mhc
contrast has been adjusted in (a)–(h) for structural visualization.
For intensity comparison, a bar graph (i) shows the average SHG
intensities for the wild-type and the different mutants. The inten-
sity of SHG was calculated by averaging the SHG signals from
various scans. Note that the background was thresholded, and zero
values were ignored in calculating the average signals. Images
were taken with 1042 nm excitation using a 0.75 NA air objective.
The bar in the images corresponds to 5 m.
1856 APPLIED OPTICS 兾 Vol. 46, No. 10 兾 1 April 2007
myosin would result in a weaker second-harmonic
signal. Third, the semicrystalline order is very im-
portant for bulk SHG generation. SHG is not observed
in random suspension of noncentrosymmetric mole-
cules even when they have a high hyperpolarizability
␤. Therefore disorder in A-bands would cause an SHG
intensity decrease whether or not any changes to the
molecular hyperpolarizibility occurred. The images
in Fig. 5 and the original structural study of different
mutations with EM
provides clear evidence that the
semicrystalline structure gets increasingly distorted
when comparing wild-type with Y97, Mhc
; Y97, and
mutants. This observation shows that the SHG
intensity decrease to a large extent is determined by
the disorder of the semicrystalline structure.
C. Effect of Induced Structural Changes on
To estimate the inﬂuence of semicrystalline order on
the SHG efﬁciency in the A-bands of sarcomeres, we
performed several experiments where structural order
in the A-bands has been altered without chemical al-
teration to the myosin thick ﬁlaments. Larvae muscles
can be imaged for extended periods of time with no
visible damage, and as mentioned previously, the large
sarcomere size permits a more detailed study of the
inner structure of the myocyte bands. The addition of
100 mM KCl changes the ion concentrations in the
myocyte and induces contraction of the myoﬁbrils,
while an excess of KCl can cause sample damage and
loss of viability. These induced changes can be used to
further understand the origin of the second-harmonic
intensity changes in myocytes.
Sustained contractions of many sarcomeres in a
myoﬁbril have been observed after the addition of
KCl. Figure 6 presents 2D SHG images over time
showing the evolution of a sustained contraction. A
localized shortening starting from one sarcomere and
gradually engaging more than ten neighboring sar-
comeres can be seen. A large stretching and deforma-
tion of the neighboring sarcomeres adjacent to the
sustained contraction region can also be observed.
The deformation of sarcomeres is induced because of
the stress generated by the sustained contraction re-
gion. The SHG intensity of the A-bands is diminished
around the sustained contraction region. Presum-
ably, the myosin molecules are still present in the
stretched A-bands, suggesting that SHG intensity is
largely determined by the semicrystalline order of the
A-bands. The sustained contraction region also shows
reduced SHG intensity compared with the region of
unaffected sarcomeres. The disorder of semicrystal-
line structure induced by the large shortening could
introduce a degree of randomness, which affects the
efﬁciency of second-harmonic generation.
Further evidence of the importance of a semicrys-
talline order in the A-bands of sarcomeres was ex-
amined during the loss of viability of larval muscle.
After the addition of excess KCl, which can impact
the viability, a small volume of sample was repeat-
edly scanned, approximately every 3 min. After
thresholding the background intensity, the average
SHG signals over the entire volume were obtained
and plotted in Fig. 7(a). The signal decreases by
⬃70% over a period of 90 min; samples imaged under
similar conditions without the addition of KCl do not
change signiﬁcantly 共⬍20%兲.
In examining a single optical section, initially [Fig.
7(b)] and after 90 min [Fig. 7(c)], one sees that 90 min
after the KCl treatment, the myocyte image still shows
similar muscle structure, despite the decreased SHG
signal. The myoﬁbrils, however, do appear broader af-
ter a loss of viability, and the staircaselike structures
are no longer prevalent. In rendering the sample vol-
umes both at t ⫽ 0 [Fig. 7(d)] and at t ⫽ 90 min [Fig.
7(e)] one can see the overall loss of order, which can
lead to a decrease in SHG. This result agrees well
with the previous observation of SHG signal loss at-
tributable to the disorder of semicrystalline structure
during uncoupler induced hypercontraction of cardi-
These examples show that the alter-
ation of semicrystalline order without altering the
chemical composition of myoﬁlaments crucially im-
pacts the SHG efﬁciency.
D. Effect of Natural Periodic Contractions on
Second-Harmonic Generation Intensity
When myocytes undergo contraction, structural
changes occur, including the conformational changes
of the myosin heads. To study the changes in SHG
intensity during contraction, spontaneous, rhythmic
contractions of larval muscle were examined. It was
found that spontaneous periodic contractions show a
Fig. 6. (Multimedia online; ao.osa.org) Effect of sustained con-
traction on the neighboring sarcomeres. The SHG images at var-
ious times during sustained contraction of Drosophila
melanogaster larvae muscle are shown. The elapsed times for each
image are indicated in the upper right corner. The region of sus-
tained contraction is circled by the oval, and the arrow points to the
middle of the highly deformed region. Images were taken with
⬃800 nm excitation using a 0.75 NA air objective.
1 April 2007 兾 Vol. 46, No. 10 兾 APPLIED OPTICS 1857
higher average SHG intensity of a row of sarcomeres
in the relaxed state than in the contracted state.
Figure 8 shows the normalized average intensity of
SHG over time along with the sarcomere length; the
SHG intensity increases as the sarcomere length in-
creases. A maximum 20% change in sarcomere length
was observed, while the maximum intensity ﬂuctua-
tion was 10%. The SHG appears more efﬁcient when
the myocyte is in a relaxed state. Similar results on
the dependence of SHG on sarcomere length were
observed in mouse skeletal myocytes that were
Since we are not altering
the myosin itself, these changes can be attributed to
the changes in the ordering of the semicrystalline
structure; therefore a decrease in the semicrystalline
order appears during contraction. A more homoge-
neous structure can be envisioned in the relaxed state
of the myocyte.
The second-harmonic response from Drosophila myo-
cytes was veriﬁed to be from the anisotropic bands of
sarcomeres. The analysis of mutant strains provided
strong evidence that a noncentrosymmetric semicrys-
talline arrangement has a crucial inﬂuence on the ef-
ﬁciency of SHG. A decrease in the semicrystalline
order of anisotropic bands during mechanical stretch-
ing and without chemical alteration of myosin di-
minished the SHG intensity. This presented further
evidence that the most signiﬁcant factor in the SHG
efﬁciency is the semicrystalline order of the anisotropic
bands of sarcomeres. Results show that, during the
functional contraction of muscle, the SHG efﬁciency
changes are likely attributable to small changes in the
organization of the anisotropic bands. Differences in
the semicrystalline order of the anisotropic band ob-
served via SHG microscopy may be useful as a diag-
nostic tool for muscular disorders. By understanding
the semicrystalline origin of the second harmonic in
myocytes and how changes in the crystallinity affect
the signal, SHG microscopy, in combination with other
nonlinear techniques, can become an invaluable tool
for studying muscular function.
The authors thank S. Bernstein of San Diego State
University for kindly providing the Drosophila mela-
nogaster myosin mutants. The research was sup-
ported by the Natural Science and Engineering
Research Council of Canada (NSERC), the Canadian
Foundation for Innovation (CFI), and the Ontario
Innovation Trust (OIT).
1. J. A. Squier and M. Muller, “High resolution nonlinear micros-
copy: A review of sources and methods for achieving optimal
imaging,” Rev. Sci. Instrum. 72, 2855–2867 (2001).
2. A. C. Millard, P. J. Campagnola, W. Mohler, A. Lewis, and
L. M. Loew, “Second harmonic imaging microscopy,” Methods
Enzymol. 361, 47– 69 (2003).
3. W. R. Zipfel, R. M. Williams, and W. W. Webb, “Nonlinear
magic: multiphoton microscopy in the biosciences,” Nat. Bio-
technol. 21, 1368 –1376 (2003).
4. C. K. Sun, “Higher harmonic generation microscopy,” Adv.
Biochem. Eng. Biotechnol. 95, 17–56 (2005).
Fig. 7. 共Multimedia online; ao.osa.org兲 The evolution of SHG in-
tensity in Drosophila melanogaster larval myocytes after addition
of KCl. (a) SHG intensity changes as the sample undergoes a loss
of viability. A 2D image of the myocyte is presented at (b) the initial
state and (c) after 90 min. The 3D rendered structures of the
muscle at (d) the initial state, and (e) after 90 min show signiﬁcant
structural changes. Images were taken with 1042 nm excitation
using a 0.75 NA air objective. Scale bar represents 15 m.
Fig. 8. Inﬂuence of periodic contractions in the Drosophila mela-
nogaster myocyte on the SHG efﬁciency. Changes in sarcomere
length during spontaneous periodic contractions are shown with
triangles, while SHG intensity changes are shown with squares.
Data were collected with 800 nm excitation using a 0.75 NA air
1858 APPLIED OPTICS 兾 Vol. 46, No. 10 兾 1 April 2007
5. J. X. Cheng and X. S. Xie, “Coherent anti-Stokes Raman scat-
tering microscopy: Instrumentation, theory, and applications,”
J. Phys. Chem. B 108, 827– 840 (2004).
6. V. Barzda, “Non-linear Contrast Mechanisms for Optical
Microscopy,” in Biophysical Techniques in Photosynthesis,
T. Aartsma and J. Matysik, eds. (Springer, 2006), to be pub-
7. K. König, “Multiphoton microscopy in life sciences,” J. Microsc.
200, 83–104 (2000).
8. L. Moreaux, O. Sandre, M. Blanchard-Desce, and J. Mertz,
“Membrane imaging by simultaneous second-harmonic gener-
ation and two-photon microscopy,” Opt. Lett. 25, 320 –322
9. M. Muller, J. Squier, K. R. Wilson, and G. J. Brakenhoff, “3D
microscopy of transparent objects using third-harmonic gener-
ation,” J. Microsc. 191, 266 –274 (1998).
10. I. Freund and M. Deutsch, “Second-harmonic microscopy of
biological tissue,” Opt. Lett. 11, 94 –96 (1986).
11. Y. C. Guo, P. P. Ho, H. Savage, D. Harris, P. Sacks, S. Schantz,
F. Liu, N. Zhadin, and R. R. Alfano, “Second-harmonic tomog-
raphy of tissues,” Opt. Lett. 22, 1323–1325 (1997).
12. P. J. Campagnola, A. C. Millard, M. Terasaki, P. E. Hoppe,
C. J. Malone, and W. A. Mohler, “Three-dimensional high-
resolution second-harmonic generation imaging of endoge-
nous structural proteins in biological tissues,” Biophys. J.
82, 493–508 (2002).
13. S. W. Chu, I. H. Chen, T. M. Liu, P. C. Chen, C. K. Sun, and
B. L. Lin, “Multimodal nonlinear spectral microscopy based
on a femtosecond Cr : forsterite laser,” Opt. Lett. 26, 1909–
14. N. Prent, R. Cisek, C. Greenhalgh, J. Aus der Au, J. Squier, and
V. Barzda, “Imaging individual chloroplasts simultaneously
with third- and second-harmonic generation and multiphoton
excitation ﬂuorescence microscopy,” in Photosynthesis: Funda-
mental Aspects to Global Perspectives, A. Van der Est and D.
Bruce, eds. (Allen Press, 2005), pp. 1037–1039.
15. T. Boulesteix, E. Beaurepaire, M. P. Sauviat, and M. C.
Schanne-Klein, “Second-harmonic microscopy of unstained liv-
ing cardiac myocytes: measurements of sarcomere length with
20-nm accuracy,” Opt. Lett. 29, 2031–2033 (2004).
16. V. Barzda, C. Greenhalgh, J. Aus der Au, S. Elmore,
J. H. G. M. Van Beek, and J. Squier, “Visualization of mito-
chondria in cardiomyocytes by simultaneous harmonic gen-
eration and ﬂuorescence microscopy,” Opt. Express 13,
17. S. W. Chu, S. Y. Chen, T. H. Tsai, T. M. Liu, C. Y. Lin, H. J.
Tsai, and C. K. Sun, “In vivo developmental biology study
using noninvasive multiharmonic generation microscopy,” Opt.
Express 11, 3093–3099 (2003).
18. S. W. Chu, S. Y. Chen, G. W. Chern, T. H. Tsai, Y. C. Chen,
B. L. Lin, and C. K. Sun, “Studies of x((2))兾x((3)) tensors in
submicron-scaled bio-tissues by polarization harmonics op-
tical microscopy,” Biophys. J. 86, 3914–3922 (2004).
19. C. Greenhalgh, R. Cisek, B. Stewart, and V. Barzda, “Dynamic
and structural visualization of muscle structure in Drosophila
with multimodal harmonic generation microscopy,” in Bio-
medical Optics Topical Meeting of OSA Proceedings (Optical
Society of America, 2006) (to be published).
20. H. E. Huxley, “Electron microscope studies of the organisation
of the ﬁlaments in striated muscle,” Biochim. Biophys. Acta 12,
21. J. Hanson and H. E. Huxley, “Structural basis of the cross-
striations in muscle,” Nature 172, 530 –532 (1953).
22. M. Both, M. Vogel, O. Friedrich, F. von Wegner, T. Kunsting,
R. H. A. Fink, and D. Uttenweiler, “Second-harmonic imaging
of intrinsic signals in muscle ﬁbers in situ,” J. Biomed. Opt. 9,
23. S. V. Plotnikov, A. C. Millard, P. J. Campagnola, and W. A.
Mohler, “Characterization of the myosin-based source for
second-harmonic generation from muscle sarcomeres,” Biophys.
J. 90, 693–703 (2006).
24. C. Greenhalgh, B. Stewart, R. Cisek, N. Prent, A. Major, and
V. Barzda, “Dynamic investigation of Drosophila myocytes
with second-harmonic generations microscopy,” in Biomedi-
cal Photonics, P. Mathieu, ed., Proc. SPIE 6343, 6343081–
25. J. A. Armstrong, N. Bloembergen, J. Ducuing, and P. S.
Pershan, “Interactions between light waves in a nonlinear
dielectric,” Phys. Rev. 127, 1918–1939 (1962).
26. R. W. Boyd, Nonlinear Optics, 2nd ed. (Academic, 2003).
27. S. W. Chu, I. H. Chen, T. M. Liu, C. K. Sun, S. P. Lee, B. L. Lin,
P. C. Cheng, M. X. Kuo, D. J. Lin, and H. L. Liu, “Nonlinear
bio-photonic crystal effects revealed with multimodal nonlin-
ear microscopy,” J. Microsc. 208, 190 –200 (2002).
28. A. Major, R. Cisek, C. Greenhalgh, N. Prent, and V. Barzda,
“A diode-pumped high-power extended cavity femtosecond
Yb:KGW laser: from development to applications in nonlin-
ear microscopy,” in Ultra-Short Pulse Optics: Lasers, Sources
and Applications, P. Mathieu, ed., Proc. SPIE 6343,
29. R. M. Cripps, J. A. Suggs, and S. I. Bernstein, “Assembly of
thick ﬁlaments and myoﬁbrils occurs in the absence of the
myosin head,” EMBO J. 18, 1793–1804 (1999).
30. C. Greenhalgh, V. Barzda, S. Elmore, J. van Beek, J. Aus der
Au, and J. Squier, “Visualization of nanocontractions in
cardiomyocytes by simultaneous detection of second- and
third-harmonic generation and multiphoton excitation ﬂuo-
rescence microscopy,” in Three-Dimensional and Multidi-
mensional Microscopy: Image Acquisition and Processing
XII, J.-A. Conchello, C. J. Cogswell, and T. Wilson, eds., Proc.
SPIE 5701, 128–135 (2005).
1 April 2007 兾 Vol. 46, No. 10 兾 APPLIED OPTICS 1859