CRISPR-Mediated Protein Tagging with Nanoluciferase to Investigate Native Chemokine Receptor Function and Conformational Changes

Abstract

G protein-coupled receptors are a major class of membrane receptors that mediate physiological and pathophysiological cellular signaling. Many aspects of receptor activation and signaling can be investigated using genetically encoded luminescent fusion proteins. However, the use of these biosensors in live cell systems requires the exogenous expression of the tagged protein of interest. To maintain the normal cellular context here we use CRISPR/Cas9-mediated homology-directed repair to insert luminescent tags into the endogenous genome. Using NanoLuc and bioluminescence resonance energy transfer we demonstrate fluorescent ligand binding at genome-edited chemokine receptors. We also demonstrate that split-NanoLuc complementation can be used to investigate conformational changes and internalization of CXCR4 and that recruitment of β-arrestin2 to CXCR4 can be monitored when both proteins are natively expressed. These results show that genetically encoded luminescent biosensors can be used to investigate numerous aspects of receptor function at native expression levels.

Keywords: ACKR3; CXCR4; G protein-coupled receptor; NanoBRET; NanoBiT; NanoLuc; endogenous.; fluorescent ligands; ligand binding; β-arrestin2.

Graphical abstract

Figure 1

Analysis of Protein Expression Following Genome Editing (A) CXCR4 mRNA expression in wild-type HEK293 cells or HEK293 clones expressing genome-edited NLuc/CXCR4, CXCR4/LgBiT, or CXCR4/LgBiT and ARRB2/SmBiT (dual). (B) CXCR4 mRNA expression in wild-type HeLa cells or HeLa clones expressing genome-edited NLuc/CXCR4. (C) ARRB2 mRNA expression in wild-type HEK293 cells or HEK293 clones expressing genome-edited ARRB2/SmBiT, or ARRB2/SmBiT and CXCR4/LgBiT (dual). Relative mRNA level, normalized to BM2 expression. Bars represent mean ± SEM of three cell passages of a single clone performed in triplicate. (D–G) Visualization of genome-edited receptor localization in HEK293 and HeLa cells using a bioluminescence LV200 Olympus microscope. (D) HEK293 and (E) HeLa cells expressing genome-edited NLuc/CXCR4, (F) HEK293 cells expressing genome-edited HiBiT/CXCR4 complemented with LgBiT and (G) HeLa cells expressing genome-edited NLuc/ACKR3. White arrow heads (D–F) indicate predominant expression at the plasma membrane of luciferase-tagged CXCR4, red arrow heads (G) indicate NLuc/ACKR3 expression in cytosolic compartments. Images were acquired by capturing total luminescence for 90 s. Scale bar represents 20 μm. See Figure S1.

Figure 2

Determination of the Binding Affinity of CXCL12-AF647 at NLuc/CXCR4 (A–D) NanoBRET saturation ligand binding curves obtained in (A) membrane preparations from HEK293 cells exogenously expressing NLuc/CXCR4 (B) live HEK293 cells expressing genome-edited NLuc/CXCR4 (C) live HeLa cells expressing genome-edited NLuc/CXCR4 or (D) live HEK293 cells expressing genome-edited HiBiT/CXCR4 complemented with LgBiT. Cells or membranes were incubated with increasing concentrations of CXCL12-AF647 in the absence (black circles) or presence (white circles) of AMD3100 (10 μM) for 1 h at 37°C. Data shown are mean ± SEM and are representative of three or four independent experiments performed in duplicate for (A and B) and (C and D), respectively. (E) Quantification of NLuc/CXCR4 expression by linear regression (F), as described in the STAR Methods, using membrane preparations made from HEK293 cells exogenously expressing NLuc/CXCR4 (NLuc/CXCR4 TG, black bar), HEK293 cells expressing genome-edited NLuc/CXCR4 (NLuc/CXCR4 HEK, gray bar), or HeLa cells expressing genome-edited NLuc/CXCR4 (NLuc/CXCR4 HeLa, white bar). Data shown are (F) mean ± SEM or (E) representative of five individual experiments performed in triplicate (see Figure S2).

Figure 3

Determination of the Binding Affinity of CXCL12-AF647 at NLuc/ACKR3 (A–C) NanoBRET saturation ligand binding curves obtained in (A) live HEK293 cells exogenously expressing NLuc/ACKR3, (B) membrane preparations from HEK293 cells exogenously expressing NLuc/ACKR3, and (C) live HeLa cells expressing genome-edited NLuc/ACKR3. Cells or membranes were incubated with increasing concentrations of CXCL12-AF647 in the absence (black circles) or presence (white circles) of CXCL11 (10 μM) for 1 h at 37°C. Data shown are mean ± SEM and are representative of five (A), four (B), and six (C) experiments performed in duplicate. (D) Quantification of NLuc/ACKR3 expression by linear regression, as described in the STAR Methods, using membrane preparations made from HEK293 cells exogenously expressing NLuc/ACKR3 (NLuc/ACKR3 TG, black bar) or HeLa cells expressing genome-edited NLuc/ACKR3 (NLuc/ACKR3 CRISPR, gray bar). Data shown are mean ± SEM of five individual experiments performed in triplicate.

Figure 4

Investigation of β-Arrestin2/SmBiT Recruitment to CXCR4/LgBiT in Genome-Edited HEK293 Cells HEK293 cells expressing (A and B) genome-edited β-arrestin2/SmBiT transiently transfected with CXCR4/LgBiT (CRISPR β-arr2/SmBiT), (C and D) genome-edited CXCR4/LgBiT transiently transfected with β-arrestin2/SmBiT (CRISPR CXCR4/LgBiT), (E and F) both genome-edited CXCR4/LgBiT and genome-edited β-arrestin2/SmBiT (Dual CRISPR), or (G and H) HEK293 cells expressing transiently transfected CXCR4/LgBiT and β-arrestin2/SmBiT (dual exogenous), were stimulated with 300 nM CXCL12 (A, C, E, and G) or increasing concentrations of CXCL12 (0.3–300 nM) (B, D, F, and H) in the absence (black squares and circles) or presence (white squares and circles) of 1 μM AM3100. Points represent mean ± SEM of six (C and D), seven (A, B, E, and F) or eight (G and H) individual experiments performed in triplicate. pEC₅₀ values stated were calculated from response at approximately 5 min after ligand addition. Baseline-corrected luminescence calculated as described in the STAR Methods (see Figures S3 and S4).

Figure 5

Using HiBiT-tagged CXCR4 to Investigate Cell Surface Expression and/or Conformational Changes (A–D) HEK293 cells expressing (A) exogenous or (B) genome-edited HiBiT/CXCR4 were incubated in the absence or presence of increasing concentrations of AMD3100 (black circles), IT1t (white circles), CXCL12 (black squares), or XAC (white squares) for 1 h at 37°C and luminescence measured 30 min following addition of purified LgBiT (10 nM) and furimazine (10 μM). Kinetic analysis of the change in luminescence mediated by addition of AMD3100 (1 μM, black circles), IT1t (1 μM, white circles), CXCL12 (1 μM, black squares), or XAC (1 μM, white squares) in (C) HEK293 cells expressing genome-edited HiBiT/CXCR4 or (D) using membrane preparations from HEK293 cells exogenously expressing HiBiT/CXCR4, both pre-incubated with 10 nM purified LgBiT. (E) AMD3100 (black circles), IT1t (white circles), CXCL12 (black squares), or XAC (white squares) concentration-response curves using membrane preparations from HEK293 cells exogenously expressing HiBiT/CXCR4 complemented with LgBiT. (F) Concentration-dependent inhibition of the AMD3100-mediated increase in luminescence by CXCL12 using membrane preparations from HEK293 cells exogenously expressing HiBiT/CXCR4 complemented with LgbiT and pre-incubated with 10 nM purified LgBiT. Points represent mean ± SEM of four (D), five (A, B, and, E), six (C), or eight (F) experiments performed in triplicate. Baseline-corrected luminescence calculated as described in the STAR Methods. **p < 0.01. Statistical analysis by one-way ANOVA with a Dunnett's multiple comparisons test (see Figures S5 and S6).

Figure 6

Investigation of the Effect of Protein Fusion on the Affinity of HiBiT-LgBiT Complementation (A–C) HEK293 cells expressing genome-edited HiBiT/CXCR4 (black symbols) or wild-type HEK293 cells (white squares) (A), membranes from HEK293 cells expressing genome-edited HiBiT/CXCR4 (black symbols) and wild-type HEK293 cells (white squares) (B), or purified HiBiT control protein (black symbols) were incubated with increasing concentrations of LgBiT in the absence (black circles) or presence (black squares) of AMD3100 (10 μM) (C). (D) HEK293 cells expressing genome-edited HiBiT/β₂-adrenoceptor (downward triangles) or wild-type HEK293 cells (white squares) were incubated with increasing concentrations of purified LgBiT. Points are mean ± SEM and are representative of five (C and D), six (B), or seven (A) experiments performed in triplicate.