A modified CUT&RUN-seq technique for qPCR analysis of chromatin-protein interactions

Abstract

Chromatin immunoprecipitation coupled with quantitative PCR (ChIP-qPCR) even with optimization may give low signal-to-background ratio and spatial resolution. Here, we adapted Cleavage Under Targets and Release Using Nuclease (CUT&RUN) (originally developed by the Henikoff group) to develop CUT&RUN-qPCR. By studying the recruitment of selected proteins (but amenable to other proteins), we find that CUT&RUN-qPCR is more sensitive and gives better spatial resolution than ChIP-qPCR. For complete details on the use and execution of this protocol, please refer to Skene et al. (2018) and Skene and Henikoff (2017).

Keywords: Cell Biology; Cell-based Assays; Chromatin immunoprecipitation (ChIP); Molecular Biology.

Graphical abstract

Figure 1

Schematic overview of CUT&RUN-qPCR (i) We use a Tus/Ter reporter system containing 6×Ter sequence and targeted it as a single copy to the Rosa26 locus of mouse chromosome 6 in mES cells. (ii) Transient expression of Tus-HA protein leads to its binding with the 6×Ter sequence. (iii) We use Anti-HA antibody as a primary antibody that binds the Tus protein, followed by proteins A/G fused with micrococcal nuclease (iv), to bind the primary antibody. (v) Chromatin is cleaved on either side of the 6×Ter sequence and released into solution. (vi) Eluted DNA is quantified and amplified by qPCR using appropriate controls.

Figure 2

Comparison of Tus enrichment at 6×Ter array at Rosa26 in mES cells using CUT&RUN-qPCR and ChIP-qPCR (A and B) Data shows signal at Tus/Ter RFB for Tus protein C-terminally tagged with HA (panel A) or Flag (panel B). Blue bars: Tus-HA or Tus-Flag enrichment using ChIP-qPCR. Purple bars: Tus-HA or Tus-Flag enrichment using CUT&RUN-qPCR. Numbers indicate distance in base pairs from the outer qPCR primer to the nearest edge of the 6×Ter array. Cartoon shows primer positions as red half-arrows. Orange triangles: Ter sites. Blue line: I-SceI restriction site. Data in CUT&RUN and ChIP figures show means of 2^{–ΔΔ CT} values, normalized to EV and β-actin control locus. Data show mean ± SD. Statistical analysis by Student’s two-tailed unpaired t-test (n = 3), assuming unknown variance. ∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001; ∗∗∗∗p < 0.0001; ns, not significant. Note more intense enrichment of Tus using CUT&RUN-qPCR than for ChIP-qPCR.

Figure 3

Comparison of Rad51 enrichment at Tus/Ter RFB and at I-SceI-induced DSB at Rosa26 in mES cells, using CUT&RUN-qPCR and ChIP-qPCR (A and B) Data shows Rad51 signals at Tus/Ter RFB (panel A) and I-SceI-induced DSB (panel B), both positioned at the Rosa26 locus in mES cells. Blue bar: Rad51 enrichment using ChIP-qPCR. Purple bar: Rad51 enrichment using CUT&RUN-qPCR. Numbers indicate distance in base pairs from the outer qPCR primer to the nearest edge of the 6×Ter array. Cartoon features as in Figure 1. Data in CUT&RUN and ChIP figures show means of 2^{–ΔΔ CT} values, normalized to EV and beta-actin control locus. Data show mean ± SD. Statistical analysis by Student’s two-tailed unpaired t-test (n = 3), assuming unknown variance. ∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001; ∗∗∗∗p < 0.0001; ns, not significant. Note greater sensitivity of CUT&RUN-qPCR Rad51 signal.

Figure 4

Comparison of BLM enrichment at Tus/Ter RFB at Rosa26 in mES cells, using CUT&RUN-qPCR and ChIP-qPCR Data shows BLM signal at Tus/Ter RFB positioned at the Rosa26 locus in mES cells, in cells that are wild type (Fancm^+/+) or null (Fancm^−/−) for Fancm. Gold bars indicate CUT&RUN-qPCR or ChIP-qPCR signals in cells transfected with empty vector (EV), to show background signal in absence of Tus. Blue bars: BLM signal using ChIP-qPCR. Purple bars: BLM signal using CUT&RUN-qPCR. Numbers indicate distance in base pairs from the outer qPCR primer to the nearest edge of the 6×Ter array. Cartoon features as in Figure 1. Data in CUT&RUN and ChIP figures show means of 2^{–ΔΔ CT} values, normalized to EV and beta-actin control locus. Data show mean ± SD. Statistical analysis by Student’s two-tailed unpaired t-test (n = 3), assuming unknown variance. ∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001; ∗∗∗∗p < 0.0001; ns, not significant. Note greater fold-enrichment of BLM signal in wild type cells using CUT&RUN-qPCR. Note also the ability of CUT&RUN-qPCR to detect a reduced BLM signal at Tus/Ter in Fancm^−/− cells, whereas ChIP-qPCR reveals no BLM signal in this setting.

Figure 5

Comparison of resolution of Tus signal at Tus/Ter RFB at Rosa26 in mES cells, using CUT&RUN-qPCR and ChIP-qPCR (A and B) Data shows signal at Tus/Ter RFB for Tus protein C-terminally tagged with HA (panel A) or Flag (panel B). Blue bars: Tus-HA or Tus-Flag enrichment using ChIP-qPCR. Purple bars: Tus-HA or Tus-Flag enrichment using CUT&RUN-qPCR. Numbers indicate distance in base pairs from the outer qPCR primer to the nearest edge of the 6×Ter array. qPCR primer positions are shown by red half-arrows. Orange triangles: Ter sites. Blue line: I-SceI restriction site. Data in CUT&RUN and ChIP figures show means of 2^{–ΔΔ CT} values, normalized to EV and β-actin control locus. Data show means ± SD. Statistical analysis by Student’s two-tailed unpaired t-test (n = 3), assuming unknown variance. ∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001; ∗∗∗∗p < 0.0001; ns, not significant. Note in CUT&RUN-qPCR data the intense Tus signal position -128 bp (i.e., where the inner PCR primer is immediately adjacent to the 6×Ter array), but no signal at position -443 bp (which lacks Ter sequences). In contrast, ChIP-qPCR reveals a positive Tus signal at both loci, including at position -443 bp, which lacks Ter sequences. This data shows that the spatial resolution of CUT&RUN-qPCR is superior to that of ChIP-qPCR.

Figure 6

Troubleshooting: avoiding adherence of ConcanvalinA beads to the wall of the Eppendorf tube (A and B) Use of more than 1 million cells per sample and of regular Eppendorf tube results in beads sticking on the wall of Eppendorf tube (panel A). However, use of 1 million cells and of low-binding Eppendorf tubes with proper resuspension of beads avoids this problem (panel B).

Figure 7

Tapestation analysis Cell transfected with Tus-HA expression plasmid processed for CUT&RUN using an anti-HA antibody. Tapestation shows cleaved fragments size distribution having a peak of 327 base pairs that match with the Tus binding region (197 base pairs -6× Ter array+126 base pairs- adapter size).