in situ hybridization
Whole-mount in situ hybridization for the detection of RNA in C. elegans embryos
Geraldine Seydoux and Andrew Fire
Adapted from Seydoux, G. and Fire, A. (1995). Whole-mount in situ hybridization for the detection of RNA in C. elegans embryos. In C. elegans: Modern Biological Analysis of an Organism. Methods in Cell Biology (ed. H. Epstein and D. Shakes) Academic Press, San Diego.
Overview.
A mixed population of C. elegans embryos is attached to microscope slides,
permeabilized by freezing and fixed with methanol and formaldehyde. Embryos
are then incubated overnight with a digoxigenin-labelled single-stranded DNA
probe, followed by extensive washes to remove excess probe. Fluorescent or enzyme-linked
anti-digoxigenin antibodies are used to visualize the hybridized probe. The entire procedure
requires approximately 1.5 days from harvest of embryos to probe visualization.
This protocol is designed for embryos, but can also be used for larvae/adults.
Best results have been obtained when looking at RNAs expressed in the adult germline,
but some somatic RNAs have also given nice patterns.
Generality of the technique
We have used this protocol to analyse the expression pattern of 21 genes
expressed during embryogenesis (Seydoux and Fire, 1994). We find that the
protocol allows the visualization of RNA in well-preserved embryos from
the one-cell stage to the pretzel stage (see Fig. 1). In general, RNAs are
detected in the cytoplasm of cells, with the exception of embryonically
transcribed RNAs which can sometimes be detected in nuclei when they are
first transcribed in early blastomeres. Low abundance RNAs expressed after
the lima bean-stage can be difficult to detect; proteinase K digestion of
embryos prior to hybridization can be helpful in such cases (Pete Okkema, pers. comm.).
Although the protocol was developed to detect RNAs in embryos, preliminary results
suggest that it is also be applicable for detecting RNAs in larvae and adults
(G. S. and A. F., unpublished data).
lacZ fusion RNAs
Because of their great abundance, RNAs derived from chromosomally integrated lacZ
fusions are an excellent target for in situ hybridization. When first transcribed,
these RNAs accumulate in two nuclear foci, which may correspond to the sites of
transcription on the two homologous chromosomes that carry the array (Seydoux and Fire, 1994).
The appearance of these "double dots" can help determine the earliest onset of transcription
for a gene of interest. Double dots can occasionally also be seen for endogenous RNAs, but are
in general more difficult to detect. In contrast to endogenous RNAs that quickly accumulate
in the cytoplasm after their initial appearance in the nucleus, lacZ fusion RNAs remain
predominantly nuclear and appear quite labile until the 26-cell stage (Seydoux and Fire, 1994).
After that stage, lacZ fusion RNAs accumulate in the cytoplasm and become more stable,
often perduring longer than endogenous RNAs. This behavior of lacZ fusion RNAs may be
due to the long, intron-less coding region of the lacZ gene.
Background vs authentic staining.
A common problem associated with the protocol presented here is the high
incidence of non specific sticking of the probe to embryos. Often, up to
50% of all wells in an experiment exhibit some form of non-specific staining.
This problem may be due to variability in the permeabilization of embryos
introduced during the freeze-cracking step. Fortunately, this non-specific staining
is easily distinguished from authentic staining. Non-specific staining usually
appears within 10 minutes in the color reaction as dark purple patches on the surface
of embryos or in nuclei. The best way to distinguish authentic staining from non-specific
staining is to compare staining patterns obtained from both antisense and sense probes.
Any staining common to both probes is likely to be due to non-specific background.
In our experience, successful hybridization with sense probes yields embryos with
no staining at all.