The consequences of mutations on gene expression are often assessed qualitatively by in situ hybridization and generally scored visually, which is subjective and biased by the researcher's expectations. This method provides a more consistent way to compare in situ experiments by quantifying image intensities and removes that bias. This can also be easily adapted to other model systems that use in situ hybridization as the reader of a gene expression.
Start by imaging the in situ hybridization-or ISH-stained embryos. Prepare a glycerol solution and mix it to homogenize. Transfer the embryos to the glycerol solution with a three-milliliter Pasteur pipette and leave them to settle for at least five minutes.
Prepare and label enough PCR tubes to transfer the ISH-stained embryos after imaging, then use a three-milliliter Pasteur pipette to add 100%glycerol to the bottom of the well of a glass depression slide. Transfer a single ISH-stained embryo to the glass slide and orient it under a stereo microscope equipped with a digital camera and bottom and top illumination. Using the first embryo, adjust the illumination and exposure time at the desired magnification.
Image as many embryos as required and label each image with a unique number. After imaging, transfer each embryo to a PCR tube labeled with the same number and if needed aspirate the excess glycerol from the PCR tubes. To extract the DNA, add 40 to 75 microliters of an alkaline lysis buffer such as HoTSHOT to each tube.
Incubate the tubes at 95 degrees Celsius for approximately 30 minutes and then cool them to four degrees Celsius. Add an equal volume of neutralization buffer and proceed with genotyping for the mutation of interest. To perform image analysis, select and open the image in Fiji and invert it by clicking Edit and then Invert, then convert the image type to 8-bit by clicking the Image tab, selecting Type, and selecting 8-bit.
Select the polygon tool and manually draw the region of interest or ROI on the image around the region that contains the ISH signal to be measured. Press T to open the ROI Manager and select the region of interest, then click Measure. Copy the mean value from the Results window to a spreadsheet.
To measure the background intensity, move the ROI to a region in the embryo with no staining and click Add in the ROI Manager. Select the background region and click Measure, then record the mean intensity value in the spreadsheet. Obtain the mean pixel intensity of the NC2 hybridization signal by subtracting the mean intensity of the background from that of the stained region.
After determining the mean pixel intensity of every sample, assign a genotype to each sample. If the genotype cannot be determined, exclude the sample from subsequent analysis. Sort the mean sample intensity values according to genotype and proceed with statistical tests.
The utility of this technique has been demonstrated on ISH for dnmt3bb. 1 in embryos from a runx1 heterozygous incross. A decrease in dnmt3bb.
1 expression was detected in runx1 homozygous mutants while heterozygous embryos showed no significant differences in dnmt3bb. 1 expression. When attempting to determine the effect of lmo4 mutation on runx1 expression, this analysis did not show significant differences in expression between wild type and homozygous lmo4 mutants.
However, a small decrease in runx1 expression was detected by single embryo qPCR. In about 0.4%of runx1-probed embryos, the ISH had a high background that resulted in a negative number when it was subtracted from the signal. In such cases, the embryos were excluded from analysis.
Four different regions on the embryos have been tested to optimize background corrections. While there was a relatively stable difference in intensity of the ROI in any of the background areas, R3 always showed higher intensity than the ROI and should not be used for background correction. R1 and R4 proved to be the most appropriate areas for background correction.
It is important to find the best conditions for the imaging and keep them unchanged throughout the experiment. We also recommend quantifying the pixel intensity before assigning a genotype to each sample.
