Third-generation in situ hybridization chain reaction: multiplexed, quantitative, sensitive, versatile, robust

Abstract

In situ hybridization based on the mechanism of the hybridization chain reaction (HCR) has addressed multi-decade challenges that impeded imaging of mRNA expression in diverse organisms, offering a unique combination of multiplexing, quantitation, sensitivity, resolution and versatility. Here, with third-generation in situ HCR, we augment these capabilities using probes and amplifiers that combine to provide automatic background suppression throughout the protocol, ensuring that reagents will not generate amplified background even if they bind non-specifically within the sample. Automatic background suppression dramatically enhances performance and robustness, combining the benefits of a higher signal-to-background ratio with the convenience of using unoptimized probe sets for new targets and organisms. In situ HCR v3.0 enables three multiplexed quantitative analysis modes: (1) qHCR imaging - analog mRNA relative quantitation with subcellular resolution in the anatomical context of whole-mount vertebrate embryos; (2) qHCR flow cytometry - analog mRNA relative quantitation for high-throughput expression profiling of mammalian and bacterial cells; and (3) dHCR imaging - digital mRNA absolute quantitation via single-molecule imaging in thick autofluorescent samples.

Keywords: Automatic background suppression; In situ HCR v3.0; Multiplexed quantitative in situ hybridization; dHCR imaging; qHCR flow cytometry; qHCR imaging.

Fig. 1.

In situ HCR v3.0 using split-initiator probes. (A) HCR mechanism. Green stars denote fluorophores. Arrowhead indicates 3′ end of each strand. (B) Standard probes carry full HCR initiator I1 and generate amplified background if they bind non-specifically. Split-initiator probes P1 and P2 each carry half of HCR initiator I1 and do not generate amplified background if they bind non-specifically. (C) Two-stage in situ HCR protocol. Detection stage: probe sets hybridize to mRNA targets, unused probes are washed from the sample. Amplification stage: specifically bound probe pairs trigger self-assembly of a tethered fluorescent amplification polymer and unused hairpins are washed from the sample. Automatic background suppression throughout the protocol: any reagents that bind non-specifically do not lead to generation of amplified background. (D) Multiplexing timeline. The same two-stage protocol is used independent of the number of target mRNAs. HCR amplification is performed overnight for qHCR imaging and qHCR flow cytometry experiments (to maximize the signal-to-background ratio) and for 45-90 min for dHCR imaging experiments (to resolve individual molecules as diffraction-limited dots).

Fig. 2.

Test tube validation of split-initiator HCR suppression. (A) Agarose gel electrophoresis. Reaction conditions: hairpins H1 and H2 at 0.5 µM each (lanes 1-7); initiator I1, probes P1 and P2 (each carrying half of initiator I1; Fig. 1B), and/or DNA target at 5 nM each (lanes noted on the gel); 5× SSCT buffer; overnight reaction at room temperature. Hairpins H1 and H2 were labeled with Alexa 647 fluorophore (green channel); a dsDNA 1 kb ladder was pre-stained with SYBR Gold (red channel). (B) Quantification of polymer bands in A. See Figs S3 and S4 for additional data.

Fig. 3.

In situ validation of automatic background suppression with split-initiator probes in whole-mount chicken embryos. (A) Fluorescent background and (B) signal-to-background ratio as probe set size is increased by adding unoptimized probes: total of 5, 10 or 20 standard probes (v2.0) versus 5, 10 or 20 split-initiator probe pairs (v3.0). Any standard probes that bind non-specifically will generate amplified background, necessitating probe set optimization; split-initiator probes eliminate the potential need for probe set optimization by providing automatic background suppression. (C) Confocal micrographs in the neural crest of fixed whole-mount chicken embryos. Unoptimized probe sets: 20 standard probes (left) or 20 split-initiator probe pairs (right). See Fig. S7 (top) for the optimized standard probe set (Choi et al., 2016) with five probes. (D) Pixel intensity histograms for background and signal plus background (pixels in the depicted regions of C): overlapping distributions using unoptimized standard probes; non-overlapping distributions using unoptimized split-initiator probes. Embryos fixed at stage HH11. Target mRNA is Sox10. See Figs S5-S11 and Tables S10-S14 for additional data.

Fig. 4.

Multiplexed mRNA imaging in whole-mount chicken embryos with large unoptimized probe sets using in situ HCR v3.0. (A) Expression schematics for four target mRNAs in the head and neural crest: FoxD3, EphA4, Sox10 and Dmbx1. (B) Four-channel confocal micrograph. (C) Zoom of depicted region of B. (D) Four individual channels from C with signal-to-background ratio measurements (mean±s.e.m., n=3 embryos). Probe sets: 12-20 split-initiator probe pairs per target. Amplifiers: four orthogonal HCR amplifiers carrying spectrally distinct fluorophores. Embryo fixed at stage HH10. See Fig. S12 and Table S15 for additional data.

Fig. 5.

qHCR imaging: analog mRNA relative quantitation with subcellular resolution in an anatomical context. (A) Two-channel redundant detection of target mRNA EphA4 in a whole-mount chicken embryo. The target is detected using two probe sets, each initiating an orthogonal spectrally distinct HCR amplifier (Ch1, Alexa 546; Ch2, Alexa 647). (B) Confocal microscopy: 0.2×0.2 µm pixels. Probe sets: 20 split-initiator probe pairs per channel; no probe set optimization. Embryo fixed at stage HH10. (C) High accuracy and precision for mRNA relative quantitation in an anatomical context. Highly correlated normalized signal (Pearson correlation coefficient, r) for subcellular 2.1×2.1×2.7 µm voxels in the selected regions of B. Accuracy: linearity with zero intercept. Precision: scatter around the line. See Figs S18 and S19 and Table S16 for additional data.

Fig. 6.

qHCR flow cytometry: analog mRNA relative quantitation for high-throughput analysis of human and bacterial cells. (A) High signal-to-background ratio for transgenic target mRNAs. Data are mean±s.e.m.; n=55,000 HEK cells (top), n=18,000 E. coli cells (bottom). Probe sets: 12 split-initiator probe pairs; no probe set optimization. (B) High accuracy and precision for high-throughput mRNA relative quantitation. Two-channel redundant detection of endogenous target mRNAs. Each target mRNA is detected using two probe sets, each initiating an orthogonal spectrally distinct HCR amplifier (Ch1, Alexa 488; Ch2, Alexa 594). Highly correlated normalized signal (Pearson correlation coefficient, r), n=20,000 HEK cells (top), n=3400 E. coli cells (bottom). Accuracy: linearity with zero intercept. Precision: scatter around the line. Probe sets: 10 split-initiator probe pairs per channel for GAPDH, 18 split-initiator probe pairs per channel for fusA; no probe set optimization. See Figs S20-S28 and Tables S17-S24 for additional data.

Fig. 7.

dHCR imaging: digital mRNA absolute quantitation in cultured human cells and whole-mount chicken embryos. (A) Redundant detection of target mRNA BRAF in HEK cells. Probe sets: 23 split-initiator probe pairs per channel; no probe set optimization. Pixel size: 0.06×0.06 µm. (B) Redundant detection of target mRNA Dmbx1 in whole-mount chicken embryos. Probe sets: 25 split-initiator probe pairs per channel; no probe set optimization. Pixel size: 0.1×0.1 µm. Embryos fixed at stage HH8. (A,B) Each target mRNA is detected using two probe sets, each initiating an orthogonal spectrally distinct HCR amplifier (Ch1, Alexa 647; Ch2, Alexa 546 for A; Ch1, Alexa 647; Ch2, Alexa 594 for B). Representative field of view from confocal micrographs. Red circles: dots detected in Ch1. Green circles: dots detected in Ch2. Yellow circles: dots detected in both channels. Colocalization represents the fraction of dots in one channel that are detected in both channels (mean±s.e.m., n=3 slides for A, n=3 embryos for B). See Figs S29-S31 and Tables S25-S27 for additional data.

Fig. 8.

Comparison of probe concepts. Scheme A corresponds to in situ HCR v2.0 with standard probes. Scheme E corresponds to in situ HCR v3.0 with split-initiator probes. Scheme A is vulnerable to non-specific probe binding in stage 1, leading to amplified background in stage 2. Scheme B provides automatic background suppression throughout the protocol at the cost of introducing sequence dependence between the target and the HCR amplifier. Scheme C provides automatic background suppression in stage 1 but is vulnerable to non-specific bridge binding in stage 2, leading to amplified background in stage 3 [a weakness shared by the pre-amplification and amplification stages (stages 2 and 3) of four-stage bDNA methods (Wang et al., 2012)]. Scheme D provides automatic background suppression throughout the protocol at the cost of using a three-stage protocol. Scheme E offers all of the advantages and none of the disadvantages of schemes A-D, providing automatic background suppression throughout the protocol, avoiding sequence dependence between the HCR amplifier and the target mRNA, and employing a two-stage protocol. Arrowhead indicates the 3′ end of each strand.