Analysis of Lipid Droplets in Cardiac Muscle - Methods in Cell Biology - page 135

12. Morphometry analysis can be performed using collected 2D images. LD size
and mitochondria measurement are performed on 20-30 (heart) fields
containing longitudinally arrayed myofibrils, photographed from each section
at the 3200

magnification. Scanned micrographs are analyzed using ImageJ
software ( http://rsb.info.nih.gov/ij ) to manually generate masks of
mitochondrial and LD contours that are used for the calculation of
mitochondrial and LD areas maximum and total mitochondrial and LD numbers
( Wang et al., 2013 ).

8.2.4 Results and discussion
In the preparation of specimens for TEM, the most common stain for LDs is osmium
tetroxide ( Hayat, 1970 ). Various modifications of the osmium staining procedure can
enhance the electron density of LDs in TEM. These include ferrocyanide-reduced
osmium ( de Bruijn, Schellens, van Buitenen, & van der Meulen, 1980 )or
imidazole-buffered osmium tetroxide ( Angermuller & Fahimi, 1982 ) as well as
osmium-thiocarbohydrazine liganding, termed the OTO method ( Guyton, Bocan,
& Schifani, 1985 ). Inclusion of tannic acid and p -phenylenediamine as mordant after
the initial osmium fixation step has also been reported to enhance the staining of LDs
( Guyton & Klemp, 1988 ). However, some of these treatments may lead to artifactual
staining or disruption of other cellular components ( Neiss, 1983 ). Another important
common consideration when designing EM preparation protocols for LDs is the risk
of lipid loss and/or loss of the osmium stain during sample preparation. Depending
on the degree of lipid or osmium extraction during preparation, LDs may exhibit an
electron translucent center of various sizes, or, in the case of complete loss, may re-
semble empty vacuoles ( Neiss, 1983 ). Although the loss of lipids/osmium most
likely occurs during the dehydration steps, others have also reported that extraction
may occur at poststaining steps with other heavy metals such as uranyl acetate and
lead citrate ( Neiss, 1983 ). Although the exact chemical reactions of these extractions
of lipids and/or osmium are still unclear, it has been shown that the translucent center
of the LDs is a result of the loss of osmium. It may be speculated that osmium ox-
idation in the center of the LDs may not be complete and thus is more prone to be
extracted during the remaining of the embedding process. However, increasing the
osmium concentration or extending the time of osmium treatment does not seem to
increase the electron density of LDs ( Guyton & Klemp, 1988 ). The depth of pene-
tration may be a limiting factor.
Here, we compare LDs stain with or without additional poststain procedures: ura-
nyl acetate and lead citrate or OTO ( Fig. 8.3 ). LDs appear homogenously stained
using the TEM preparation method we described earlier ( Fig. 8.3 A and ( Wang
et al., 2013 )). Figure 8.3 A and B shows marked differences in the appearances of
LDs, mitochondria, and myofibrils of the ultrathin section collected from the same
specimen preparation with ( Fig. 8.3 B) or without ( Fig. 8.3 A) uranyl acetate (10 min)
and lead poststaining (5 min), a procedure routinely applied to TEM specimen to en-
hance the contrast ( Hayat, 1970 ). While membranes are well contrasted by using the