Biomedical Engineering Reference
In-Depth Information
of SHG images to microtubule structures with distinct morphology such as the mitotic spindles. In
less obvious situations, such as in neurons, simultaneous imaging of fluorescently labeled tubulin
(Campagnola et al., 2002) or post hoc immunohistochemistry of microtubule-associated proteins
(Dombeck et al., 2003, Kwan et al., 2008) revealed a tight relationship between the microtubule dis-
tribution and SHG intensity. It is possible that microtubule-associated proteins may contribute to the
observed signal, but this is unlikely to be the primary source because SHG is present in many cell types,
animal species, and brain and body regions that do not share the same composition of microtubule-
associated proteins. Second, the application of microtubule-depolymerizing drugs such as nocodazole
(Dombeck et al., 2003, Kwan et al., 2008, Barnes et al., 2010) and colchicine (Stoothoff et al., 2008) sig-
nificantly reduces the SHG intensity, whereas actin-specific agents such as cytochalasin D has no effect.
7.1 Why Study Microtubules with SHG imaging?
SHG imaging is an optical technique; so, it is noninvasive and can provide information at high spatial
and temporal resolution. Moreover, when applied to microtubules, SHG imaging has numerous unique,
useful features. Here, we will discuss these features and also some drawbacks with an emphasis on com-
paring SHG imaging to alternative methods for visualizing microtubules.
7.1.1 SHG intensity Reflects Microtubule Polarity
One unique property of SHG is coherent summation, where signal amplitudes from scattering units are
summed. Phase differences can enhance or reduce SHG intensity depending on whether the inference
is constructive or destructive. SHG signal is largest when the scatterers are oriented similarly, that is,
from tubulins that have the same orientation. As a result, SHG is sensitive to the microtubule orienta-
tion within the focal volume and therefore can be used as a probe of polarity with submicron spatial
resolution. However, the sign of the polarity, whether it is plus- or minus-end pointing, is not known.
The SHG intensity dependence on microtubule polarity can be best illustrated by comparing the
fluorescence and SHG images of a mitotic spindle (Figure 7.3). Green fluorescent protein (GFP)-labeled
tubulin proteins imaged with two-photon-excited fluorescence microscopy revealed a spatial pattern
that directly relates to the tubulin concentration. The fluorescence image showed the characteristic shape
of a mitotic spindle, consisting of two spindle poles with microtubules forming asters, most of which
met at the equator, where duplicated chromosomes were to be separated. In contrast, SHG intensity
depended on both the tubulin concentration and the microtubule polarity; so, it displayed a different
FIgurE 7.3
SHG and two-photon-excited fluorescence images of a
C. elegans
embryo during first mitosis. The
microtubules in the mitotic spindle were fluorescent because the cell expresses β-tubulin fused with GFP. SHG signals
were detected in a subset of microtubules; it was absent in the spindle equator, distal ends of the astral microtubules,
and the center of the spindle poles. This profile of SHG intensity is consistent with coherent summation and the qua-
dratic dependence on microtubule density. Scale bar = 10 μm. (Reprinted from
Biophysical Journal
, 82, Campagnola,
P. J. et al., Three-dimensional high-resolution second-harmonic generation imaging of endogenous structural proteins
in biological tissues. 493-508. Copyright 2002, with permission from Elsevier.)
