https://www.nature.com/articles/s41592-020-0746-7

Chemogenetic Control of Nanobodies

Key figures

• [Fig. 1]: Defines the LAMA design principle (cpDHFR insertion into nanobody CDRs) and demonstrates ligand-triggered loss of target binding with reversible kinetics and a steric-hindrance structural rationale.
• [Fig. 2]: Shows reversible, minutes-timescale control of GFP-fusion localization and function in live cells using TMP (including genome-edited proteins and phenotypic readouts).

1) Thesis (one sentence)

To address the lack of rapid, reversible small-molecule control over intracellular nanobody–target binding, in live cells using nanobody fusions, inserting circularly permuted DHFR into nanobody binding loops causes ligand-triggered dissociation and re-association of nanobody–target complexes by cpDHFR folding upon NADPH and DHFR inhibitor binding to sterically hinder target engagement, supported by in vitro binding/kinetics assays, X-ray structures, and live-cell imaging.

2) Evidence card (three bullets only)

• Strongest result: Mitochondria-anchored GFPLAMA can sequester a genome-edited checkpoint protein (Mad2L1-EGFP) to mitochondria and TMP reversibly restores function, shifting nuclear morphology outcomes and nocodazole-driven mitotic arrest behavior (Fig. 2j,k).
• Method enabler: cpDHFR insertion into GFP nanobody CDR3 yields single-digit-nM binding without ligands, while NADPH+TMP drives dissociation on seconds-to-minutes timescales and crystal structures suggest steric blockage as the mechanism (Fig. 1c–h; protein domain insertion + fluorescence titrations/kinetics + X-ray crystallography).
• Critical limitation: Switching requires tight geometric coupling to the nanobody fold—adding flexible GGS linkers between cpDHFR and the nanobody abolishes TMP-dependent control, implying insertion site and mechanical coupling are hard constraints for portability (Supplementary Fig. 6).

Optional

Quote bank (2–4 short excerpts)

• Quote 1: “We introduce an engineered nanobody whose affinity to green fluorescent protein (GFP) can be switched on and off with small molecules.” (Abstract, p.279)
• Quote 2: “LAMAs disrupt the binding of the nanobody to its target by exploiting the change in conformation of cpDHFR on binding of NADPH and DHFR inhibitors.” (Results, Fig. 1a, p.279)
• Quote 3: “The dissociation kinetics of the complexes could also be tuned using DHFR inhibitors with different affinities to DHFR.” (Results, p.279–280)

Key comparisons (1–3 lines)

• Compared to: optogenetic nanobody control via LOV insertion (reported modest affinity shift) or split-antibody light systems (reported irreversible activation), LAMAs provide a reversible small-molecule chemogenetic toggle.
• Win: repeated on/off cycling on minute timescales in vitro and in live cells using cell-permeable TMP.
• Tradeoff: full cellular sequestration requires an excess of GFPLAMA relative to GFP target, and not all nanobodies tolerate cpDHFR insertion at attempted positions.

Methods I might copy (protocol hooks)

• Construct design / Models: Generate LAMAs by inserting cpDHFR into nanobody CDR3 (e.g., GFPLAMAF98, GFPLAMAG97, GFPLAMAN95) and fuse GFPLAMAs to localization anchors (Ntom20 for mitochondria, Lyn11 for plasma membrane, NLS for nucleus) to reversibly relocalize EGFP; apply to genome-edited HeLa Kyoto NUP62-mEGFP and Mad2L1-EGFP lines for functional perturbation (Fig. 1b–e; Fig. 2c–k).
• Conditions / Instruments: In vitro titrations used NADPH 100 µM and TMP 500 µM in 50 mM HEPES, 50 mM NaCl, 0.5 mg/ml BSA, 0.05% Triton-X100, pH 7.3, read on a Spark 20M (excitation 470 nm; emission 535 nm); dissociation kinetics used wtGFP 200 nM with TMP 1 µM (then eDHFR 8 µM and TMP 50 µM) in presence of NADPH 100 µM (Fig. 1f,g; Methods); live-cell perturbations used TMP typically 10 µM with confocal imaging (e.g., Nikon Eclipse Ti2 spinning disk; Leica SP8; Zeiss LSM780) at 37 °C and 5% CO2 (Fig. 2b,d–f; Methods).
• Readout / Analysis: Quantify GFP binding by fluorescence emission titrations and TR–FRET (SNAP-Lumi4-Tb labeling; Spark 20M TR–FRET mode with 320 nm excitation, 480/520 nm emissions); quantify live-cell redistribution by nuclear ROI intensity over time in FIJI/ImageJ with TMP perfusion cycles; phenotype scoring includes percent polylobed nuclei and time to mitotic exit under nocodazole 330 nM with TMP washout/maintenance (Fig. 1j,l; Fig. 2e,f,j,k; Methods).

Open questions / Theoretical implications (2–5 bullets)

• Can a general rule predict which nanobody loops tolerate cpDHFR insertion without killing basal affinity, beyond “try CDR3 first”?
• How tunable is the switching window (baseline KD vs ligand-inhibited KD) by changing cpDHFR inhibitor affinity and intracellular ligand availability?
• Can the same “folded domain sterically blocks binding” logic be engineered to yield ON-switch behavior (ligand enabling binding) rather than OFF-switch disruption?
• What are the practical intracellular limits set by stoichiometry (need for excess LAMA) when the target is endogenous and not overexpressed?