Visualization of Lipid Droplets in C. elegans by Light and Electron Microscopy - Methods in Cell Biology - page 42

3.2 VISUALIZATION OF LIPID DROPLET ASSOCIATED
PROTEINS BY LASER SCANNING CONFOCAL MICROSCOPY
3.2.1 Background
Visualization of fluorescent fusion proteins that are expressed at physiological levels
demands the use of highly sensitive microscopy systems. Attempts to visualize GFP::
DGAT-2 and mRuby::DGAT-2 in wide-field microscopy systems largely failed be-
cause of fluorescence bleaching upon prolonged excitation. The prevalent autofluor-
escent signals (from lysosome-related organelles, LROs) in the C. elegans intestine
present additional challenges ( Schroeder et al., 2007; Zhang, Trimble, et al., 2010 ).
This is because the 488-nm laser for excitation of the GFP in confocal systems will
also strongly excite autofluorescent material ( Fig. 3.2 A) ( Xu et al., 2012 ). Further-
more, the emission spectra of GFP and autofluorescence substantially overlap
( Fig. 3.2 A). We overcame these challenges by (1) establishment of reference emis-
sion spectra of GFP, mRuby, and autofluorescence on a Zeiss LSM710 laser scan-
ning confocal microscopy system; (2) linear unmixing of fluorescent signals using
the Zeiss Zen software.

3.2.2 Methods
The embryonic development of C. elegans occurs under the protection of an egg shell
in utero and after egg-laying. Hatched animals then progress through four larval
stages before becoming reproductive adults. We routinely imaged animals that are
at the fourth larval stage (L4). Progressive increase of intestinal autofluorescent sig-
nals in adult animals makes older animals more challenging to image. Live animals
are transferred to an agarose pad on top of a glass slide before being covered by a
glass cover slip. Two methods can be used to immobilize the animals during imag-
ing. One involves the use of 2 mM tetramisole (in 1 PBS), an agonist of acetylcho-

line receptors in C. elegans , which causes hypercontraction of body wall muscles.
Another method utilizes polystyrene nanoparticles (100 nm in diameter) ( Kim,
Sun, Gabel, & Fang-Yen, 2013 ). This method avoids the use of drugs and improves
the success rate of animal recovery after imaging. We use a Zeiss LSM710 confocal
system, equipped with a 40 , NA1.2 water C-Apochromat objective. A 488-nm la-

ser is used for excitation, and its output power is minimized to prevent bleaching and
phototoxicity. Detector gain is adjusted to avoid overexposure. Images are taken in
lambda mode that affords fluorescence detection in 32-channels from 421 to 723 nm
(in 9.7 nm increments) simultaneously. Spectrally resolved signals collected in each
channel are used to generate emission spectra for GFP and autofluorescence
( Fig. 3.2 A). Using the Zeiss Zen software, linear unmixing was performed to deter-
mine if a pixel is occupied by GFP or autofluorescent signals. The same procedure
can be performed when the mRuby::DGAT-2 fusion protein is imaged. In this case, a
561-nm laser is used for excitation.