Articles

Contact | Print version | Recommend article | RSS-Feed

Confocal Detection of NBT/BCIP In Situ Hybridization
Samples by Reflection Microscopy

Gáspár Jékely1,2* and Detlev Arendt1
1Developmental Biology Unit, European Molecular Biology Laboratory, Heidelberg, Germany
2Max Planck Institute for Developmental Biology, Tübingen, Germany

*Corresponding author

Confocal laser scanning reflection microscopy (CLSM) in combination with whole-mount in situ hybridization (WMISH) is a novel method to visualize gene expression patterns in three dimensions in whole embryos. The method uses conventional NBT/BCIP precipitation by alkaline phosphatase in WMISH samples and confocal detection by reflection microscopy. WMISH protocols can also be combined with fluorescent counterstainings such as DAPI, antibodies, GFP lines, and fluorescent in situ probes. Such double and triple labeling combined with confocal microscopy allows expression profiling at a cellular resolution in cell types or embryological structures of interest. Whole-mount reflection CLSM will thus greatly facilitate large-scale, cellular resolution expression profiling in vertebrate and invertebrate model organisms.

Introduction

The determination of gene expression patterns in three dimensions with cellular resolution is an important goal in developmental biology [1]. The most widely used method, in situ hybridization, conventionally uses digoxigenin-11-UTP-labeled RNA probes and detection with an alkaline phosphatase-coupled anti-digoxigenin antibody. Alkaline phosphatase activity is visualized with 5-bromo-4-chloro-3-indolyl-phosphate (BCIP) and nitro-blue-tetrazolium (NBT), which yield a black-purple precipitate. The sensitivity and precision of alkaline phosphatase-NBT/BCIP staining is very high and is often the preferred method over fluorescent in situ techniques. However, such samples are normally documented only by bright-field microscopy. Our reflection method allows the confocal imaging of NBT/BCIP samples and brings several advantages that greatly facilitate expression profiling in whole embryos [2].

Materials and Methods

Microscopy

Confocal images were taken with either a Leica TCS SP2 or a Leica TCS SPE confocal microscope using a 40x oil-immersion objective and a 1–2 airy unit pinhole opening. For reflection imaging a 633 nm or a 635 nm laser was used at maximum intensity with the spectral detector set to 630–640 nm. We also used a Zeiss LSM 510 system for reflection imaging that has no spectral detector. In this case we used a 633 nm excitation laser, a HFT UV/488/568/633 main dichroic beam splitter and no emission filter, or a long-pass emission filter with a wavelength block below 633 nm (e.g., LP 585). For the spectral measurements we used the spectral detector of a Leica TCS SP2 confocal microscope and moved a 5-nm-wide sliding window at 1-nm intervals. Fluorescent signals were detected using the appropriate laser lines and detector settings. 3D-projections of confocal stacks were done using ImageJ and Imaris 5.5. Bright-field images were taken on a Zeiss Axiophot microscope using DIC optics.

Whole-mount in situ hybridization

WMISH was perfomed using standard protocols [2,3] on Platynereis and zebrafish embryos fixed in 4% paraform­aldehyde in PTW (PBS + 0.1% Tween-20) for 2 hours. NBT/BCIP staining was combined with fluorescent antibody staining and fluorescent tyramide WMISH using published protocols.

Results and Discussion

When looking at WMISH samples in a confocal microscope we found that the NBT/BCIP precipitate can be detected using the reflection mode. The signal we detected came from the black-purple precipitate, as is evident by the comparison of bright-field and confocal images (Figure 1). The signal is generated by the reflection of the laser light from the sample, as shown by illuminating WMISH samples with six different laser lines and measuring the spectral properties of light emitt­ed by the sample. When the sample was illuminated with any of the six laser lines we always detected the maximum signal with a wavelength that corresponded to the wavelength of illumination (Figure 2a). These data show that we detected reflection, not fluorescence, from the NBT/BCIP precipitate.

Using the reflection method we were able to visualize the whole volume of a gene’s expression pattern even for broadly expressed genes. The method is sensitive enough to visualize single cells or sub-cellular structures with localized RNA, such as axonal projections.

We tested the reflection method on WMISH samples from three species (Platynereis, zebrafish and Drosophila) and obtained similar results [2]. We also tested whether confocal detection of NBT/BCIP can be combined with the detection of other fluorescent markers to allow cellular resolution co-localization studies. We performed double labeling experiments by combining NBT/BCIP staining for one gene with anti-acetylated tubulin immunostaining or with fluorescent WMISH for another gene. Using anti-acetylated tubulin immunostaining we co-localized the expression of a cell-type-specific sensory receptor with the apical sensory dendrite of the cell and the axonal scaffold (Figure 2b). We also combined NBT/BCIP staining and anti-GFP antibody staining in a transgeniczebrafish line expressing GFP in the lateral line primordium and detected cellu-lar co-localization by combining reflection and fluorescent detection (Figure 3). Our method greatly enhances the sensitivity and ease of performing such double-labeling experiments and also allows co-expression studies for weakly expressed genes that work poorly in fluorescent WMISH [1].

The reflection method has some limitations that should be taken into consideration when designing experiments. Depending on the nature of the embryos under study, light could also be reflected by certain anatomical structures, such as yolk or chaetae, giving a background signal. Another potential problem arises when samples are strongly stained with NBT/BCIP in a broad domain. Such strong signal can partly absorb the exciting light and the emitted fluorescence, resulting in a “shadowing effect” deeper in the tissue [2]. In such cases one has to select embryos that are stained weaker, to minimize the shadowing effect. The choice of fluorophore also affects the quality of the results, since shadowing is stronger at lower wavelengths (e.g., when using DAPI and a 405-nm laser).

Conclusions

Whole-mount confocal laser scanning reflection microscopy is a powerful alternative to fluorescent techniques to perform cellular resolution expression profiling, fluorescent counter-staining, and 3D volume rendering of gene expression patterns. 

References

1. Denes A et al. (2007) Cell 129:277–288

2. Jékely G, Arendt D (2007) Biotechniques 42:751–755

3. Tessmar-Raible K et al. (2005) Biotechniques 39:460, 462, 464

This article was originally published in Biochemica 4/2007, pages 12-14. ©Springer Medizin Verlag 2007