Cell-Specific Chemical Delivery Using a Selective Nitroreductase

Aug 16, 2018 - ... confirmed their stability in cells, and identified the best substrate for NTR. ... Enzyme-Activated Fluorogenic Probes for Live-Cel...
0 downloads 0 Views 6MB Size
Subscriber access provided by Kaohsiung Medical University

Article

Cell-Specific Chemical Delivery Using a Selective Nitroreductase–Nitroaryl Pair Todd D. Gruber, Chithra Krishnamurthy, Jonathan B. Grimm, Michael R Tadross, Laura M. Wysocki, Zev J. Gartner, and Luke D. Lavis ACS Chem. Biol., Just Accepted Manuscript • DOI: 10.1021/acschembio.8b00524 • Publication Date (Web): 16 Aug 2018 Downloaded from http://pubs.acs.org on August 16, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Biology

Cell-Specific Chemical Delivery Using a Selective Nitroreductase–Nitroaryl Pair Todd D. Gruber,1,a,* Chithra Krishnamurthy,1,2 Jonathan B. Grimm,1 Michael R. Tadross,1,b Laura M. Wysocki,1,c Zev J. Gartner,2 and Luke D. Lavis1,* 1

2

Janelia Research Campus, Howard Hughes Medical Institute, Ashburn, Virginia 20147 USA Department of Pharmaceutical Chemistry, University of California, San Francisco, California 94158 USA

*Corresponding authors

a

Current address: Christopher Newport University, 1 Avenue of the Arts, Newport News, Virginia 23606 b

c

Current address: Duke University, 101 Science Drive, Durham, North Carolina 27708

Current address: Wabash College, 301 W Wabash Avenue, Crawfordsville, Indiana 47933

*Corresponding authors: Todd D. Gruber, Christopher Newport University, 1 Avenue of the Arts, Newport News, VA 23606; email: [email protected]. Luke D. Lavis, Janelia Research Campus, Howard Hughes Medical Institute, 19700 Helix Drive, Ashburn, Virginia 20147; email: [email protected]. Keywords: fluorescence ・ imaging ・pharmacology ・enzymology

ACS Paragon Plus Environment

ACS Chemical Biology 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 30

Gruber et al. Abstract: The utility of small molecules to probe or perturb biological systems is limited by the lack of cell-specificity. ‘Masking’ the activity of small molecules using a general chemical modification and ‘unmasking’ it only within target cells overcomes this limitation. To this end, we have developed a selective enzyme–substrate pair consisting of engineered variants of E. coli nitroreductase (NTR) and a 2-nitro-N-methylimidazolyl (NM) masking group. To discover and optimize this NTR–NM system, we synthesized a series of fluorogenic substrates containing different nitroaromatic masking groups, confirmed their stability in cells, and identified the best substrate for NTR. We then engineered the enzyme for improved activity in mammalian cells, ultimately yielding an enzyme variant (enhanced NTR, or eNTR) that possesses up to 100-fold increased activity over wild-type NTR. These improved NTR enzymes combined with the optimal NM masking group enable rapid, selective unmasking of dyes, indicators, and drugs to genetically defined populations of cells.

2

ACS Paragon Plus Environment

Page 3 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Biology

Gruber et al. Small molecules are important tools for biological research. The ability to control the properties and activity of compounds through organic synthesis allows the generation of imaging agents such as fluorophores,1,2 sensors that bind analytes to measure the cellular environment,3,4 or pharmacological agents that interact with specific proteins to modulate cell-signaling pathways.5 Nevertheless, the use of small molecules is hampered by the absence of cellspecificity because most molecular probes act indiscriminately on all cells. This presents a major challenge when using these compounds in complex, multicellular organisms, which divide their physiological functions between multiple distinct and highly interacting cell types. One attractive solution to this problem is the use of selective enzyme–substrate pairs, where molecules are ‘masked’ with a disposable group that is stable to endogenous enzymes, but is rapidly removed by an exogenous enzyme that is expressed in defined subset of cells.6-8 This strategy combines the molecular specificity of small molecules with the cellular specificity of genetic manipulation. There are two strategies for developing selective enzyme–substrate pairs for use in mammalian cells. The first approach involves making an existing delivery strategy cell-selective; we accomplished this previously by development of a selective esterase–ester pair.8 The other strategy involves utilizing an enzyme that is orthogonal to mammalian biochemistry and identifying a small masking motif that is removed efficiently by the action of the enzyme. An attractive enzyme system for this strategy is E. coli nitroreductase (NTR), which is a flavoenzyme that reduces nitroaryl groups using NADH as a cofactor.9 NTRs have been used to activate fluorophores10-21 and toxins22-24 in cells and in vivo, but a general masking strategy has not been described. Here, we use fluorogenic compounds to identify, evaluate, and optimize selective nitroreductase–nitroaryl pairs that allow facile delivery of diverse small molecules to defined cells.

3

ACS Paragon Plus Environment

ACS Chemical Biology 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 30

Gruber et al.

RESULTS AND DISCUSSION Synthesis and reactivity of bis(nitroaryl)-fluoresceins. The literature on NTR-activated molecules is somewhat scattered despite the utilization of NTR enzymes in different areas of biology. The identification of bacterial NTR enzymes began in the 1940s with the discovery that compounds containing nitroaromatic groups could selectivity kill bacterial cells; this observation led to the development of new antibiotics such as nitrofurazone.25 In the 1970s, genetics led to the discovery of two genes in E. coli, nfsA and nfsB (nfs = nitrofurazone sensitivity), that encoded flavoproteins capable of reducing nitroaryl groups using NADH as a cofactor in the presence of oxygen,26 putting them in the category of oxygen-insensitive nitroreductases.27 In particular, the NTR encoded by nfsB has found utility in gene-directed enzyme-prodrug therapy (GDEPT)28 and targeted cell ablation in zebrafish.22,23 Several different nitroaryl groups have been used to mask fluorophores10-21 or luciferins29-31 to create indicators for nitroreductases but no systematic study of the reactivity of different nitroaryl groups with NTR has been performed. To add to the confusion, many of these same nitroaromatic motifs can be activated by endogenous oxygen-sensitive nitroreductases under extreme hypoxia;10,12,14 the absence of molecular oxygen allows multi-electron reduction of the nitroaryl to be catalyzed by these oxygen-sensitive reductases.27 Although oxygen-sensitive nitroreductases are widespread in both bacteria and multicellular organisms,27 oxygen-insensitive nitroreductases like the NTR encoded by nfsB do not appear to be found in higher eukaryotes.32 All of this prior work on NTR— antibiotics, GDEPT, targeted cell ablation, and selective unmasking in hypoxic tissue—supports the orthogonality of the oxygen-insensitive NTR encoded by nfsB in mammalian cells under normoxic conditions.

4

ACS Paragon Plus Environment

Page 5 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Biology

Gruber et al. To find the best general masking group for NTR encoded by nfsB, we synthesized a series of potential fluorogenic substrates based on 2ʹ,7ʹ-difluorofluorescein (Oregon Green), a photostable variant of fluorescein that is pH insensitive in the physiological range.33 Fluorescein-based substrates are useful as fluorogenic enzyme substrates because alkylation of the phenolic oxygens locks the molecule in a nonfluorescent lactone form resulting in low background signal. Although preparation of dialkylfluoresceins is complicated by competing alkylation the orthocarboxyl group, we circumvented this issue by using reduced ‘leuco’ fluorescein 1 as a synthetic intermediate to prepare novel substrates in a divergent fashion (Figure 1a, Supplementary Figure S1).34 Alkylation of 1 by different aryl bromides followed by oxidation with DDQ led to a high yield of the desired compounds. We prepared the following substrates based on commercially available nitroaryl groups with known NTR reactivity: bis(4-nitrobenzyl)-Oregon Green, (NBOG, 2); bis(5-nitrofuranyl)-Oregon Green, (NFOG, 3); bis(5-nitrothiophenyl)-Oregon Green, (NTOG, 4); bis(2-nitro-N-methyl imidazolyl)-Oregon Green, (NMOG, 5). All of these substrates show low visible absorption and fluorescence background due to the formation of the nonfluorescent lactone form of the fluorescein dye. We then tested this panel of substrates to find a structure that was unmasked rapidly by NTR but was stable to the endogenous intracellular reducing environment of mammalian cells. Measurement of enzyme kinetics with purified NTR enzyme showed the nitroimidazole 5 (NMOG) was the optimal substrate in vitro (Figure 1a,b) with a kcat/KM value of 8.1 ± 0.8 × 104 M–1s–1 and a Km value of 1.6 ± 0.2 µM. NFOG (3) and NTOG (4) showed lower reactivity with kcat/KM = 3.0 ± 0.5 × 103 M–1s–1 and ~6 × 102 M–1s–1, respectively, and the nitrobenzyl compound, NBOG (2), was largely unreactive. The 4-nitrobenzyl mask is arguably the most common NTR activated mask in the literature,10,11,13-17,19,21 making the higher reactivity of the

5

ACS Paragon Plus Environment

ACS Chemical Biology 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 30

Gruber et al. nitroimidazole mask particularly noteworthy. Evaluation in cells overexpressing NTR mirrored the in vitro experiments, with 5 showing the highest performance in HEK293 cells (Figure 1c). The intracellular NMOG unmasking requires NTR expression; mock transfection and incubation with 5 gave low cellular fluorescence (Figure 1c). Based on the rapid unmasking and low background fluorescence we selected NMOG (5) and its 2-nitro-N-methyl imidazolyl (NM) masking group for subsequent experiments. ‘Macro’ NTR engineering. We undertook a macro-engineering approach to improve the activity and expression of NTR in cultured neurons, reading out the enzymatic activity in cells using NMOG (5) and the enzyme expression levels using immunofluorescence (Figure 2a, Supplementary Figure S2). Fusion of NTR to a fluorescent protein improved expression, with mCherry giving a 3.5-fold increase in expression while preserving the activity per monomer. Because NTR is likely an obligate dimer35 we then explored different tandem-dimer configurations (tdNTR) to enhance stability and overall expression. Combining fusion of mCherry and the tandem dimer (tdNTR-mCherry) further enhanced both enzymatic expression and unmasking of NMOG fluorescence, with this construct showing a 12-fold increase in activity over wild-type enzyme (Figure 2a, Supplementary Figure S2). This fusion allows further testing of the cell-specificity of the unmasking using fluorescence microscopy. A co-culture of nitroreductase-expressing (NTR+) and untransfected (NTR–) cells incubated with NMOG (5) showed the Oregon Green fluorophore (green) is unmasked only in NTR+ cells expressing the tdNTR-mCherry fusion (red); NTR– cells bearing an epitope marker (blue) do not unmask the enzyme (Figure 2b-e, Supplementary Figure S3). Cell-specific delivery of a novel Ca2+ indicator. We deployed this system to target a smallmolecule calcium indicator to a specific cellular population. Existing calcium indicators, such as

6

ACS Paragon Plus Environment

Page 7 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Biology

Gruber et al. Fluo-4 (6, Figure 3a) have a fluorinated xanthene core structure similar to Oregon Green.4 However, 6 lacks the ortho-carboxyl group that is necessary for the molecule to adopt the fully colorless, nonfluorescent lactone form. We therefore synthesized a Fluo-4 derivative with the requisite carboxyl group, which we called “Fluo-4XL” (Fluo-4 carboxyl, 7, Figure 3a, Supplementary Figure S4). We were also curious if the introduction of an ortho-substituent on the pendant ring would improve the performance of the indicator as suggested by the patent literature.36 We compared solutions of 6 and 7 with matched absorption at 490.5 nm (A490.5 = 0.095) in buffers containing different concentrations of free Ca2+ (Figure 3b). Both indicators showed low fluorescence of the apo state (F0). We found the standard Fluo-4 to exhibit modest brightness in the Ca2+-bound state with a quantum yield (Φ) value of 0.14 (∆F/F0 = 150) and Kd = 390 nM (Figure 3b, Supplementary Figure S5a). The Fluo-4XL indicator (7) showed a slightly lower affinity (Kd = 610 nM), presumably due to the electron-withdrawing nature of the additional carboxyl group. However, Fluo-4XL exhibited a substantially brighter Ca2+-bound state with Φ = 0.40 and ∆F/F0 = 270; this high ∆F/F0 was also observed under two-photon excitation (Figure 3b, Supplementary Figure S5b). Delivery of small molecule indicators to specific cells in a mixed population has been hampered by the requisite carboxyl groups found in most ion chelator motifs, which require masking groups to cross the cellular membrane. For example, most small–molecule fluorescent Ca2+ indicators are designed to be trapped in live cells using acetoxymethyl (AM) esters that are unmasked by ubiquitous nonspecific cellular esterases. Targeting with HaloTag and SNAP-tag enables subcellular localization of indicators,37 but such compounds contain AM esters and can still load and be unmasked indiscriminately in all cells. In addition to substantially improving the indicator properties, the extra carboxyl group in Fluo-4XL (7) makes it possible to fully mask the

7

ACS Paragon Plus Environment

ACS Chemical Biology 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 30

Gruber et al. fluorophoric portion of the indicator, allowing unmasking of the dye only in defined NTR+ cells. We therefore synthesized a derivative of 7 bearing the nitroimidazole groups on the fluorophoric portion and standard acetoxymethyl (AM) groups on the BAPTA calcium-chelator moiety (8, Fluo-4XL NM/AM, Figure 3c). Although this cell-permeable derivative is likely loaded into all cells due to nonspecific AM ester hydrolysis, it is only fully unmasked in specific neurons expressing NTR in a co-culture with NTR– cells, which bore a surface epitope for detection. As with the simpler NMOG substrate (Figure 2b), we observed fluorescence (green) only in the NTR+ cells (red) with no signal in the NTR– cells (blue). The green fluorescence signal was Ca2+-sensitive and responded to field stimulation with the ability to detect single action potentials in cultured neurons (Figure 3d-f, Supplementary Figure S6, Movie S1). Cell-specific delivery of a masked cAMP analog. We next extended the NTR–NM targeting method to target the membrane-impermeant protein kinase A (PKA) activator Sp-8OHcAMPS (9), an analog of cyclic adenosine monophosphate (cAMP).38 We synthesized a NM-masked analog of Sp-8OH-cAMPS (10, Sp-(8-ONM)-cAMPS-NM, i.e., ‘Sp-NM’, Figure 4a, Supplementary Figure S7a). Because compound 9 is polar and cell impermeant, we surmised that addition of NM groups would allow 10 to diffuse through the cellular membrane, where it would be unmasked to yield trapped compound 9. To generalize the NTR–NM pair to additional cellular types, we tested this approach in MCF-10A cells, which were virally transduced with tdNTR and NTR expression validated by unmasking of NMOG (Figure S7b). cAMP analogs activate PKA signaling in MCF-10A cells, inhibiting phosphorylation of ERK.39 MCF-10A cells expressing either NTR or an empty plasmid were treated with either 10 or vehicle (DMSO/Pluronic F-127). The masked analog Sp-NM (10) activated cAMP signaling in NTR+ MCF-10A cells (MCF10A-tdNTR) but not in NTR– cells expressing the empty plasmid and a

8

ACS Paragon Plus Environment

Page 9 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Biology

Gruber et al. red fluorescent protein nuclear marker (MCF10A-H2B-RFP), as measured by increased phosphorylation of PKA substrates and decreased ERK phosphorylation (Figure 4b, Supplementary Figure S7c). To investigate cell-specific unmasking of 10 in mixed cultures, we utilized a heterogeneous wound-healing assay. MCF10A-tdNTR (NTR+) and MCF10A-H2B-RFP (NTR–) cells were seeded at confluence in two-well IBIDI cell culture inserts in order to create a cell free gap between two distinct cellular populations (Figure 4c). The cultures were starved for 18–24 h to synchronize the cells and limit proliferation. Upon insert removal, cultures were treated with vehicle or 100 µM Sp-NM 10 for 30 minutes, stimulated with assay media containing vehicle or 100 µM compound 10, and imaged over 24 h to measure migration of each cell population. Vehicle-treated cells migrated to close the gap over 18 h regardless of NTR expression, NTR+ cells treated with Sp-NM (10) showed substantially lower cell migration, and treatment with SpNM 9 did not inhibit migration in NTR– cells (Figure 4d and Movie S2). The difference in cell migration was particularly distinct in co-cultures of NTR+ and NTR– cells. The masked cAMP analog Sp-NM 10 inhibited migration specifically in the NTR+ population (Figure 4d, bottom panel, red nuclei), while NTR– (green nuclei) migrated freely. ‘Micro’ NTR engineering. The initial NTR fusion proteins showed improved expression and kinetics allowing unmasking of Fluo-4XL and Sp-8OH-cAMPS in specific cells. Both these compounds are cell impermeant once unmasked, however, allowing the use of relatively long incubation times. Other pharmacological agents are more cell permeable, which could necessitate a further improvement in enzyme kinetics. We thus engaged in a ‘micro’ engineering approach using directed evolution near the NTR enzyme active site. We used degenerate codons to mutagenize nitroreductase to all proteogenic amino acids at each of seven active site

9

ACS Paragon Plus Environment

ACS Chemical Biology 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 30

Gruber et al. positions40 and used NMOG (5) to read out enzymatic activity in E. coli lysates in a highthroughput fluorescence format using a microplate reader (Figure 5a). Improved variants were then overexpressed, purified, and assayed to determine kinetic parameters (Figure 5b, Supplementary Figure S8a). We found three hotspots for improving the enzyme: positions 68, 70, and 124. Tryptophan mutations at positions Y68 and F124 resulted in significant improvements in kcat value and position F70 could accept the most diversity with six variants showing improved enzyme kinetics. Incorporation of these mutations into the tdNTR-mCherry construct revealed that a single point mutant, F124W, showed the largest enhancement of kcat in vitro (Figure 5b) and in neurons (Figure 2a, Supplementary Figure S8b,c); this variant was previously identified as an improved catalyst for prodrug CB1954 activation.41 Tryptophan exhibits stronger π-stacking interactions with nitroaryl groups compared to other aromatic amino acids, which might explain these improved enzyme kinetics.42 Collectively, the ‘macro’ and ‘micro’-engineering efforts resulted in a tdNTR(F124W)-mCherry (enhanced NTR, or eNTR) that shows ~100-fold improvement in cellular activity (expression × kinetics) over the wild-type NTR enzyme (Figure 2a). Cell-specific delivery of NMDAR antagonists. We then used this new eNTR enzyme variant to target an antagonist of the NMDA receptor (NMDAR) based on MK-801 (11), which has a binding pocket that has been mapped on the interior face of the NMDAR protein.43 Acylation of MK-801 abolishes pharmacological activity and intracellular targeting of MK-801 has recently been accomplished in acute brain slice using our previously reported selective esterase–ester pair system.44 We first synthesized nitroimidazole-masked MK-801, compound 12 (Figure 5c, Supplementary Figure S9a), which released MK-801 (11) in a rapid reaction catalyzed by NTR that was enhanced by the F124W mutation (Figure S9b). Using a short

10

ACS Paragon Plus Environment

Page 11 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Biology

Gruber et al. incubation with compound 12 (10 min), we observed potent inhibition of NMDAR function measured in a mono-culture of eNTR+ neurons, and no inhibition in mono-cultured eNTR– cells (Figure 5d, green and magenta). However, experiments using a co-culture eNTR+ (identified by mCherry signal) and eNTR– (identified by extracellular HA epitope tag; see Methods) showed no differential effect on the two cell types (Figure 5d, black). MK-801 (11) is cell permeable (logD = +1.8) and our results suggest that eNTR releases sufficient MK-801 from eNTR+ cells to block NMDA receptors in all cell types; this “bystander effect” is observed even with very sparse expression of eNTR in a mixed culture (Supplementary Figure S10). To remedy this bystander effect, we examined the medicinal chemistry literature45 to find a derivative of MK-801 with lower membrane permeability. A hydroxylated derivative of MK-801 (MK-801-OH, 13) shows similar binding to NMDAR as the parent MK-801 (11) in membrane preparations (Ki of 11/13 = 56 nM/260 nM), but was expected to show substantially decreased potency in intact slice experiments based on similar hydroxylated MK-801 analogs.45 The disparity between the membrane and slice assay suggested that introduction of extra hydroxyl groups on MK-801 decreases membrane permeability, as the pharmacological agents only have to cross membranes in the slice experiments. We synthesized the NM-masked derivative of MK801-OH, compound 14 (Figure 5e). Repeating the neuronal cellular experiments with compound 14 showed a clear delineation in the NMDAR activity in eNTR+ and eNTR– cells, even in coculture experiments (Figure 5f). CONCLUSION Small molecules are important tools for biology, yet their application to complex systems is often limited by indiscriminate delivery of chemical probes to all cells. The combination of molecular genetics and synthetic organic chemistry promises the ability to selectively measure

11

ACS Paragon Plus Environment

ACS Chemical Biology 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 30

Gruber et al. and manipulate specific cell types in complex physiological environments using small molecules. Here, we provide a systematic evaluation of nitroaryl groups as substrates for the E. coli NTR enzyme encoded by nfsB (Figure 1), engineer the enzyme for improved activity against the best nitroimidazole mask (Figure 2, Figure 5), show the first cell-specific targeting of a smallmolecule calcium ion indicator (Figure 3) and cAMP analog (Figure 4) using enzymatic unmasking, and investigate how membrane-permeability can alter the bystander effect (Figure 5). This NTR–NM pair displays generality, allowing the masking of a diverse array of molecules and functional groups ranging from a relatively large (780.6 g/mol) Ca2+ indicator masked via an ether bond (Figure 3c), to a small (221.3 g/mol) NMDAR antagonist masked via a carbamate linkage (Figure 5). These results also further validate our approach of using fluorogenic enzyme substrates to identify and optimize selective enzyme–substrate pairs. The prospect of cell-specific pharmacology is particularly interesting as biological investigations move increasingly from monolithic populations of cells in a dish to more complex mixtures in culture, in tissue, or in vivo. The selective enzyme–substrate approach compliments approaches that utilize the HaloTag or SNAP-tag to tether drugs to the cell surface,46,47 allowing pharmacological manipulation of targets inside cells. Importantly, our data reveals some additional design principles for cell-type specific pharmacology approaches, as many standard drugs that work on intracellular targets might be too permeable and may alter nearby cells through the bystander effect. This problem can be solved in many ways depending on the nature of the system under investigation; we utilized highly polar compounds (e.g., cAMP analog 10), and revisited the drug discovery literature to find pharmacological agents (e.g., 13) that exhibit high potency in homogeneous assays with partially purified protein but low potency in cell-based assays. Other solutions to this problem could include using covalent inhibitors to allow much

12

ACS Paragon Plus Environment

Page 13 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Biology

Gruber et al. lower compound concentrations, or perfusing the sample to increase the dilution of drugs that escape target cells. Our engineered NTR–NM system can thus be readily applied to release a wide variety of masked molecules to interrogate diverse biological systems.

METHODS Substrate comparison. The initial velocity of fluorescence release (λex = 490 nm, λem = 520 nm) of substrates 2–5 (10 µM to 0.078 µM, except for compound 2 where the highest concentration was 5 µM due to solubility) upon addition of 1 µg/mL nitroreductase (Sigma) was measured in triplicate on a FlexStation 3 platereader (Molecular Devices) in 10 mM HEPES pH 7.3, 1% DMSO, 100 µM NADH, 100 nM FMN. Data were fit using GraphPad Prism software to the Michaelis–Menten equation to extract kinetic parameters. Vector construction. The E. coli nitroreductase (gene nfsB) sequence was used in nitroreductase vectors. See SI for full vector construction methods. HEK293 experiments. HEK293T/17 cells were transfected with pCAG-NfsB-IRESNLSmCherry using Lonza Nucleofection following manufacturer’s protocols. At 24 hours, substrates (11 µM) were added in Tyrode’s media with 0.1% DMSO and 0.02% pluronic-F127 (Sigma) and fluorescence was read as above. Neuronal assays of Fluo-4XL: Dissociated hippocampal neurons were separately nucleofected with either SYN-tdNfsB-mCherry or a control construct bearing an extracellular HA epitope tag. Nucleofected cells were combined to generate mixed co-cultures of NTR+ (mCherry-positive) and NTR– (surface HA positive) cells. Experiments were performed two weeks later. NTR– neurons were labeled live via immunofluorescence by incubating coverslips in primary mouse anti-HA (1:1000, Covance) followed by secondary Alexa 647 anti-mouse (1:1000, Molecular

13

ACS Paragon Plus Environment

ACS Chemical Biology 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 30

Gruber et al. Probes) antibodies (30 min each, in NbActiv4 media; 37 °C in the incubator). A 10 mM stock of Fluo-4XL NM/AM was prepared in fresh DMSO containing 20% pluronic F-127, and dissolved 1:1000 in NbActiv media (10 µM Fluo-4XL NM/AM final). Cells were incubated in this solution for 4 hours at 37 °C in the incubator. Cells were then rinsed 2× in a physiological buffer, and placed in a glass-bottom 24-well plate. Each well contained 500 µL of a physiological solution (in mM): 150 NaCl, 4 KCl, 2 MgCl2, 2 CaCl2, 10 HEPES, 10 glucose, pH 7.4, with 10 µM NBQX (to block AMPAR transmission), 10 µM CPP (to block NMDAR transmission), and 10 µM gabazine (to block GABAA transmission). Plates were transferred to an Olympus IX81 with 10X (0.4 NA) air objective lens, Cairn OptoLED illumination system, and a high-speed Andor Technology EMCCD camera (DU860_BV, 128×128 resolution, 500 frames/s, 1000 electron multiplying gain, 1× pre-amp gain, –60°C). A Grass S48 Stimulator (Grass Technologies) was used for field stimulation, as previously described.48 Analysis was performed in custom MATLAB scripts. Western blotting. MCF-10A cells were seeded at 2.5 × 104 cells/well in a 48-well plate and cultured for 24 h. Near confluent cells were starved in non-supplemented DMEM/F12 Media overnight and subsequently treated with Sp-NM 10 in DMSO mixed with an equivalent volume of Pluronic (20% solution in DMSO) for 30 minutes. To measure phospho-ERK inhibition, cells were then stimulated with 20 ng/mL EGF for five min and lysed immediately in 1× lysis buffer (Cell Signaling Technologies) supplemented with protease and phosphatase inhibitors (Roche), diluted with sample buffer (BioRad) and analyzed by gel electrophoresis and Western blotting. To measure PKA activation, cells were lysed immediately after treatment with Sp-NM 10, and lysates were analyzed by gel and Western blot.

14

ACS Paragon Plus Environment

Page 15 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Biology

Gruber et al. MCF-10A Migration Assay: Cells were plated in IBIDI µ-well 2-well inserts (IBIDI, Germany) at 105 cells/cm2. Cells were allowed to adhere for 6-10 h and then starved in non-supplemented DMEM/F12 media overnight. Inserts were removed to create a 500 µm cell free gap, and cells were treated with 100 µM Sp-NM 10 for 30 min. Migration was stimulated with the addition of DMEM/F12 media supplemented with 20 ng/mL EGF, 10 mg/mL insulin, and 2% horse serum. Wells were immediately imaged on an inverted microscope (Axiovert 200M; Zeiss, Germany) at 10× for 24 h at 30 min increments. Images were analyzed using Image J using a macro to quantify cell area and gap area (Supplementary Information). Screening of NTR variants. Mutations were introduced into pRSET-NfsB at individual amino acids using degenerate NNS codons at positions S40, T41, Y68, F70, N71, G120, and F124 using Quikchange Multi Lightning (Agilent). For each site, 144 (probability 88% for full coverage) colonies in BL21Gold(DE3) (Agilent) were picked for further analysis. Overexpression lysates were diluted 1:500 into PBS, and the reaction was initiated via addition of 0.2 µM NMOG (5) with 100 µM NADH (final concentrations) by a liquid handling platereader (Hamamatsu FDSS, green filter set). All colonies with rates of release of fluorescence greater than three standard deviations above wild-type NTR controls, as well as well as those with rates just under that threshold, were restreaked, reassayed, and sequenced. Masked MK-801 in cultured neurons. Dissociated hippocampal neurons were separately nucleofected with either SYN-tdNfsB(F124W)-mCherry (eNTR+) or a control construct bearing an extracellular HA epitope tag (eNTR–). Nucleofected cells were combined to generate mixed co-cultures of eNTR+ (mCherry-positive) and eNTR– (surface HA positive) cells. Experiments were performed two weeks later. eNTR– neurons were first labeled live via immunofluorescence by 30 min incubation in NbActiv media containing a primary mouse anti-HA (1:1000, Covance),

15

ACS Paragon Plus Environment

ACS Chemical Biology 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 30

Gruber et al. followed by 30 min incubation in NbActiv media containing a secondary Alexa 647 antibody along with 3 µM Fluo2-AM-LR calcium dye (TefLabs), all at 37 °C. Masked drug-containing NbActiv media was transferred to 24-well plates, and warmed for 30 minutes in a 37 °C incubator during the second round of antibody staining. Neurons were incubated in drugcontaining media for 10 minutes at 37 °C, then transferred to a glass-bottom 24-well plate containing 500 µL of a Mg-free resting solution consisting of (in mM): 150 NaCl, 4 KCl, 4 CaCl2, 10 HEPES, 10 glucose, pH 7.4, with 10 µM NBQX (to block AMPAR transmission), 1 µM TTX (to block action potentials), and 10 µM gabazine (to block GABAA transmission). For each coverslip, one field of view was selected and imaged on an inverted IX81 microscope in multi-channel time-lapse imaging mode: TexRd (mCherry), Cy5 (surface HA epitope), and FITC (Fluo dye) channels were imaged once every 6 seconds. Following 30 seconds of baseline imaging, 1000 µL of a stimulation solution was added to produce a final concentration of 50 µM NMDA + 50 µM glycine. Cells were imaged for 2 minutes. To minimize batch effects, each dose of each drugs was assayed consistently for 8 weeks of independent cultures. Cells were manually segmented and analyzed in custom MATLAB scripts. Bystander effect assay. Mixed co-cultures of eNTR+ and eNTR– dissociated hippocampal neurons, prepared as above, were incubated with Fura 2 AM/LR, followed by NM-MK-801 (12) for 10 minutes. Assay was performed by addition of 50 µM NMDA and 50 µM glycine in 0 mM Mg2+, 4 mM Ca2+ Tyrode’s media. For further information on synthetic methods, vector construction, culture methods, imaging assays, screening assays and other protocols, see SI Materials and Methods.

Associated Content

16

ACS Paragon Plus Environment

Page 17 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Biology

Gruber et al. Full details of synthesis and characterization of new materials, experimental methods, Figures S1-S10, and Movies S1 and S2 can be found online. This material is available free of charge via the internet at http://pubs.acs.org. Acknowledgements We thank B.C. Shields for assistance with cultured-neuron assays, A.C. Arnold and H. White for cell culture and initial microscopy studies, A.N. Tkachuk for assistance with molecular biology, D.J. Kim for use of the microscope for Ca2+ imaging/stimulation, J.J. Macklin for twophoton excitation of Fluo-4XL, and S.M. Sternson, D.B. Murphy, and P.H. Lee for contributive discussions (all at Janelia). Z.J.G. is a CZ Biohub Investigator and is supported in part by grants from the National Science Foundation (MCB-1330864 and DBI-1548297) and Department of Defense (W81XWH-13-1-0221). This work was supported by the Howard Hughes Medical Institute.

Competing Financial Interests The authors declare no competing financial interests.

17

ACS Paragon Plus Environment

ACS Chemical Biology 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 30

Gruber et al. Figure Legends Figure 1. Identification of NM as an optimal NTR-reactive mask. (a) Synthesis and apparent kinetic parameters (n = 3, mean ± SEM) of fluorogenic nitroreductase substrates 2–5 from leuco-fluorescein 1 (b) Michaelis-Menten plots of enzyme kinetics using substrates 2–5 with wild-type NTR (n = 3, mean ± SEM). Inset: baseline region expanded to show data for compounds 2-4 more clearly. (c) Fluorescence release of fluorogenic substrates (11 µM) in HEK293 cells transfected with NTR (n=3, dashed lines indicate SEM). Fluorescence from NBOG (2) signal is indistinguishable from mock transfected cells. 5* denotes incubation of NMOG (5) with mock transfected cells. Figure 2. Engineering of NTR. (a) Improved activity of optimized NTR constructs, as measured by fluorescence release in cultured neurons transfected with NTR. Data represents analysis of >300 cells from each of 3 or more independent cultures (mean ± SEM). (b) Dye release from NMOG (5) within cultured neurons (red: mCherry signal from tdNTR-mCherry transfected neurons; green: released fluorescent dye; blue: antibody-labeled HA epitope tag marking NTR negative neurons; scale bar = 100 µm). (c-e) Individual color channels of image in (b).

Figure 3. Synthesis and targeted delivery of Fluo-4XL. (a) Structure of Fluo-4 and Fluo-4XL (b) Comparison of the fluorescence emission spectra of Ca2+ indicators using samples with matched absorption. (c) Chemical structure of Fluo-4XL NM/AM (8). (d–f) Cell-specific loading of Fluo-4XL NM/AM 8 (4 h) in dissociated hippocampal neurons and resulting fluorescence signal traces upon field stimulation (AP: action potential, ∆F: maximum change in raw fluorescence values, “mCherry:Cy5 ratio”: mCherry signal divided by Cy5 signal from HA tag, ∆F/F0: ∆F divided by initial fluorescence (F0). Data represents 8 coverslips containing a total of 85 cells, with 42 designated NTR+ (mCherry:Cy5 ≥ 2.0) and 38 designated NTR– (mCherry:Cy5 ≤ 0.8). (d) Fluorescence change (∆F) as a function of mCherry:Cy5 ratio. (e) Fluorescence traces upon field stimulation over time (mean ± SEM, gradient from black = 160 AP to lightest gray = 1 AP, bottom trace = signal from NTR– cells). (f) Change in fluorescence over initial fluorescence (∆F/ F0) upon field stimulation (mean ± SEM). See also Movie S1, and Figure S6 for additional time points. Figure 4: The masked cAMP analog Sp-NM 10 activates PKA signaling in a cell specific manner. (a) Structure and activation of masked cAMP analog (Sp-NM 10). (b) Treatment with

18

ACS Paragon Plus Environment

Page 19 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Biology

Gruber et al. Sp-NM 10 promotes PKA substrate phosphorylation (left) and inhibits ERK phosphorylation (right) in NTR+ MCF-10A cells, but not in cells expressing empty plasmid (H2B-RFP). Data represent mean ± SEM of two independent experiments, each run in duplicate. Statistical significance was assessed using unpaired Student’s t test. ***P < 0.005. (c) Schematic describing a heterogeneous wound healing assay. (d) 100 µM Sp-NM 10 inhibits migration in homogeneous NTR+ cells (middle, right panel) but not homogeneous MCF-10A cells expressing the empty plasmid (top, right panel) over the period of 12 hours. In a heterogeneous culture of NTR+ and NTR– cells, Sp-NM 10 inhibits migration only in NTR+ (red nuclei) population (bottom, right panel). Yellow outlines represent cell edge at t = 0 h. Scale bar is 100 µm. (e) Quantification of cell migration in homogenous culture (top) and heterogenous culture (bottom). Data represent mean ± SEM of two independent experiments quantified at t = 12 h, each performed in triplicate, with each replicate containing 6–8 non-overlapping frames along the cell gap. Statistical significance was assessed using unpaired Student’s t test. ***P < 0.005 Figure 5. Cell-specific delivery of MK-801 using enhanced nitroreductase (eNTR). (a) Active site residues40 investigated by saturation (NNS) mutagenesis (yellow: FMN cofactor; green: bound nitrofuran antibiotic. PDB 1YKI.49 (b) Fold improvement in kcat value (NMOG, 5) for mutations identified though high-throughput screening (n = 3, mean ± SEM). (c) Chemical structure of NM-masked MK-801 (12) and release by nitroreductase. (d) Calcium response to masked MK-801 12 in eNTR+ and eNTR– co-cultured neurons (n = 8, mean ± SEM). (e) Chemical structure of NM–masked MK-801-OH (14) and release by nitroreductase. (f) Calcium response to masked MK-801-OH 14 in eNTR+ and eNTR– co-cultured neurons (n = 8, mean ± SEM). REFERENCES

[1] Lavis, L. D., and Raines, R. T. (2008) Bright ideas for chemical biology, ACS Chem. Biol. 3, 142-155. [2] Lavis, L. D., and Raines, R. T. (2014) Bright building blocks for chemical biology, ACS Chem. Biol. 9, 855-866.

19

ACS Paragon Plus Environment

ACS Chemical Biology 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 30

Gruber et al. [3] Grynkiewicz, G., Poenie, M., and Tsien, R. Y. (1985) A new generation of Ca2+ indicators with greatly improved fluorescence properties, J. Biol. Chem. 260, 3440-3450. [4] Gee, K. R., Brown, K. A., Chen, W. N., Bishop-Stewart, J., Gray, D., and Johnson, I. (2000) Chemical and physiological characterization of fluo-4 Ca(2+)-indicator dyes, Cell Calcium 27, 97-106. [5] van Hattum, H., and Waldmann, H. (2014) Chemical biology tools for regulating RAS signaling complexity in space and time, Chem. Biol. 21, 1185-1195. [6] Gao, W., Xing, B., Tsien, R. Y., and Rao, J. (2003) Novel fluorogenic substrates for imaging beta-lactamase gene expression, J. Am. Chem. Soc. 125, 11146-11147. [7] Ng, D. N., and Fromherz, P. (2011) Genetic targeting of a voltage-sensitive dye by enzymatic activation of phosphonooxymethyl-ammonium derivative, ACS Chem. Biol. 6, 444-451. [8] Tian, L., Yang, Y., Wysocki, L. M., Arnold, A. C., Hu, A., Ravichandran, B., Sternson, S. M., Looger, L. L., and Lavis, L. D. (2012) Selective esterase-ester pair for targeting small molecules with cellular specificity, Proc. Natl. Acad. Sci. U.S.A. 109, 4756-4761. [9] Searle, P. F., Chen, M. J., Hu, L., Race, P. R., Lovering, A. L., Grove, J. I., Guise, C., Jaberipour, M., James, N. D., Mautner, V., Young, L. S., Kerr, D. J., Mountain, A., White, S. A., and Hyde, E. I. (2004) Nitroreductase: a prodrug-activating enzyme for cancer gene therapy, Clin. Exp. Pharmacol. Physiol. 31, 811-816. [10] Cui, L., Zhong, Y., Zhu, W., Xu, Y., Du, Q., Wang, X., Qian, X., and Xiao, Y. (2011) A new prodrug-derived ratiometric fluorescent probe for hypoxia: high selectivity of nitroreductase and imaging in tumor cell, Org. Lett. 13, 928-931. [11] Huang, H. C., Wang, K. L., Huang, S. T., Lin, H. Y., and Lin, C. M. (2011) Development of a sensitive long-wavelength fluorogenic probe for nitroreductase: a new fluorimetric

20

ACS Paragon Plus Environment

Page 21 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Biology

Gruber et al. indictor for analyte determination by dehydrogenase-coupled biosensors, Biosens. Bioelectron. 26, 3511-3516. [12] Li, Z., Li, X., Gao, X., Zhang, Y., Shi, W., and Ma, H. (2013) Nitroreductase detection and hypoxic tumor cell imaging by a designed sensitive and selective fluorescent probe, 7-[(5nitrofuran-2-yl)methoxy]-3H-phenoxazin-3-one, Anal. Chem. 85, 3926-3932. [13] Shi, Y., Zhang, S., and Zhang, X. (2013) A novel near-infrared fluorescent probe for selectively sensing nitroreductase (NTR) in an aqueous medium, Analyst 138, 1952-1955. [14] Guo, T., Cui, L., Shen, J., Zhu, W., Xu, Y., and Qian, X. (2013) A highly sensitive longwavelength fluorescence probe for nitroreductase and hypoxia: selective detection and quantification, Chem. Commun. 49, 10820-10822. [15] Lee, M. K., Williams, J., Twieg, R. J., Rao, J., and Moerner, W. E. (2013) Enzymatic activation of nitro-aryl fluorogens in live bacterial cells for enzymatic turnover-activated localization microscopy, Chem. Sci. 42, 220-225. [16] Li, Z., He, X., Wang, Z., Yang, R., Shi, W., and Ma, H. (2015) In vivo imaging and detection of nitroreductase in zebrafish by a new near-infrared fluorescence off-on probe, Biosens. Bioelectron. 63, 112-116. [17] Li, Y., Sun, Y., Li, J., Su, Q., Yuan, W., Dai, Y., Han, C., Wang, Q., Feng, W., and Li, F. (2015) Ultrasensitive near-infrared fluorescence-enhanced probe for in vivo nitroreductase imaging, J. Am. Chem. Soc. 137, 6407-6416. [18] Ao, X., Bright, S. A., Taylor, N. C., and Elmes, R. B. P. (2017) 2-Nitroimidazole based fluorescent probes for nitroreductase; monitoring reductive stress in cellulo, Org. Biomol. Chem. 15, 6104-6108.

21

ACS Paragon Plus Environment

ACS Chemical Biology 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 30

Gruber et al. [19] Luo, S., Zou, R., Wu, J., and Landry, M. P. (2017) A probe for the detection of hypoxic cancer cells, ACS Sens. 2, 1139-1145. [20] Xu, S., Wang, Q., Zhang, Q., Zhang, L., Zuo, L., Jiang, J. D., and Hu, H. Y. (2017) Real time detection of ESKAPE pathogens by a nitroreductase-triggered fluorescence turn-on probe, Chem. Commun. 53, 11177-11180. [21] Zhou, Y., Bobba, K. N., Lv, X. W., Yang, D., Velusamy, N., Zhang, J. F., and Bhuniya, S. (2017) A biotinylated piperazine-rhodol derivative: a 'turn-on' probe for nitroreductase triggered hypoxia imaging, Analyst 142, 345-350. [22] Curado, S., Stainier, D. Y., and Anderson, R. M. (2008) Nitroreductase-mediated cell/tissue ablation in zebrafish: a spatially and temporally controlled ablation method with applications in developmental and regeneration studies, Nat. Protoc. 3, 948-954. [23] Mathias, J. R., Zhang, Z., Saxena, M. T., and Mumm, J. S. (2014) Enhanced cell-specific ablation in zebrafish using a triple mutant of Escherichia coli nitroreductase, Zebrafish 11, 85-97. [24] Copp, J. N., Mowday, A. M., Williams, E. M., Guise, C. P., Ashoorzadeh, A., Sharrock, A. V., Flanagan, J. U., Smaill, J. B., Patterson, A. V., and Ackerley, D. F. (2017) Engineering a multifunctional nitroreductase for improved activation of prodrugs and PET probes for cancer gene therapy, Cell Chem. Biol. 24, 391-403. [25] Scott, A. B., and Stillman, W. B. (1947) Series of nitrofuran compounds, US Patent 2,416,234. [26] McCalla, D. R., Kaiser, C., and Green, M. H. (1978) Genetics of nitrofurazone resistance in Escherichia coli, J. Bacteriol. 133, 10-16.

22

ACS Paragon Plus Environment

Page 23 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Biology

Gruber et al. [27] Peterson, F. J., Mason, R. P., Hovsepian, J., and Holtzman, J. L. (1979) Oxygen-sensitive and -insensitive nitroreduction by Escherichia coli and rat hepatic microsomes, J. Biol. Chem. 254, 4009-4014. [28] Denny, W. A. (2002) Nitroreductase-based GDEPT, Curr. Pharm. Des. 8, 1349-1361. [29] Porterfield, W. B., Jones, K. A., McCutcheon, D. C., and Prescher, J. A. (2015) A "caged" luciferin for imaging cell-cell contacts, J. Am. Chem. Soc. 137, 8656-8659. [30] Feng, P., Zhang, H., Deng, Q., Liu, W., Yang, L., Li, G., Chen, G., Du, L., Ke, B., and Li, M. (2016) Real-time bioluminescence imaging of nitroreductase in mouse model, Anal. Chem. 88, 5610-5614. [31] Yang, X., Li, Z., Jiang, T., Du, L., and Li, M. (2017) A coelenterazine-type bioluminescent probe for nitroreductase imaging, Org. Biomol. Chem. 16, 146-151. [32] De Oliveira, I. M., Bonatto, D., and Henriques, J. A. P. (2010) Nitroreductases: Enzymes with environmental, biotechnological and clinical importance, Current Research, Technology and Education Topics in Applied Microbiology and Microbial Biotechnology, 1008-1019. [33] Sun, W.-C., Gee, K. R., Klaubert, D. H., and Haugland, R. P. (1997) Synthesis of fluorinated fluoresceins, J. Org. Chem. 62, 6469-6475. [34] Wysocki, L. M., Grimm, J. B., Tkachuk, A. N., Brown, T. A., Betzig, E., and Lavis, L. D. (2011) Facile and general synthesis of photoactivatable xanthene dyes, Angew. Chem., Int. Ed. 50, 11206-11209. [35] Parkinson, G. N., Skelly, J. V., and Neidle, S. (2000) Crystal structure of FMN-dependent nitroreductase from Escherichia coli B: A prodrug-activating enzyme, J. Med. Chem. 43, 3624-3631.

23

ACS Paragon Plus Environment

ACS Chemical Biology 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 24 of 30

Gruber et al. [36] Diwu, Z., He, J., and Liao, J. (2008) Fluorescent ion indicators and their applications, US Patent 2008/0254498 A1. [37] Kamiya, M., and Johnsson, K. (2010) Localizable and highly sensitive calcium indicator based on a BODIPY fluorophore, Anal. Chem. 82, 6472-6479. [38] Otmakhov, N., and Lisman, J. E. (2002) Postsynaptic application of a cAMP analogue reverses long-term potentiation in hippocampal CA1 pyramidal neurons, J. Neurophysiol. 87, 3018-3032. [39] Bruzzone, A., Sauliere, A., Finana, F., Senard, J. M., Luthy, I., and Gales, C. (2014) Dosage-dependent regulation of cell proliferation and adhesion through dual beta2adrenergic receptor/cAMP signals, FASEB J. 28, 1342-1354. [40] Grove, J. I., Lovering, A. L., Guise, C., Race, P. R., Wrighton, C. J., White, S. A., Hyde, E. I., and Searle, P. F. (2003) Generation of Escherichia coli nitroreductase mutants conferring improved cell sensitization to the prodrug CB1954, Cancer Res. 63, 5532-5537. [41] Race, P. R., Lovering, A. L., White, S. A., Grove, J. I., Searle, P. F., Wrighton, C. W., and Hyde, E. I. (2007) Kinetic and structural characterisation of Escherichia coli nitroreductase mutants showing improved efficacy for the prodrug substrate CB1954, J. Mol. Biol. 368, 481-492. [42] An, Y., Bloom, J. W., and Wheeler, S. E. (2015) Quantifying the pi-stacking interactions in nitroarene binding sites of proteins, J. Phys. Chem. B 119, 14441-14450. [43] Huettner, J. E., and Bean, B. P. (1988) Block of N-methyl-D-aspartate-activated current by the anticonvulsant MK-801: selective binding to open channels, Proc. Natl. Acad. Sci. U.S.A. 85, 1307-1311.

24

ACS Paragon Plus Environment

Page 25 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Biology

Gruber et al. [44] Yang, Y., Lee, P., and Sternson, S. M. (2015) Cell type-specific pharmacology of NMDA receptors using masked MK801, eLife 4, e10206. [45] Thompson, W. J., Anderson, P. S., Britcher, S. F., Lyle, T. A., Thies, J. E., Magill, C. A., Varga, S. L., Schwering, J. E., Lyle, P. A., Christy, M. E., and et al. (1990) Synthesis and pharmacological evaluation of a series of dibenzo[a,d]cycloalkenimines as N-methyl-Daspartate antagonists, J. Med. Chem. 33, 789-808. [46] Broichhagen, J., Damijonaitis, A., Levitz, J., Sokol, K. R., Leippe, P., Konrad, D., Isacoff, E. Y., and Trauner, D. (2015) Orthogonal optical control of a G protein-coupled receptor with a SNAP-tethered photochromic ligand, ACS Cent. Sci. 1, 383-393. [47] Shields, B. C., Kahuno, E., Kim, C., Apostolides, P. F., Brown, J., Lindo, S., Mensh, B. D., Dudman, J. T., Lavis, L. D., and Tadross, M. R. (2017) Deconstructing behavioral neuropharmacology with cellular specificity, Science 356, 6333. [48] Wardill, T. J., Chen, T. W., Schreiter, E. R., Hasseman, J. P., Tsegaye, G., Fosque, B. F., Behnam, R., Shields, B. C., Ramirez, M., Kimmel, B. E., Kerr, R. A., Jayaraman, V., Looger, L. L., Svoboda, K., and Kim, D. S. (2013) A neuron-based screening platform for optimizing genetically-encoded calcium indicators, PLoS One 8, e77728. [49] Race, P. R., Lovering, A. L., Green, R. M., Ossor, A., White, S. A., Searle, P. F., Wrighton, C. J., and Hyde, E. I. (2005) Structural and mechanistic studies of Escherichia coli nitroreductase with the antibiotic nitrofurazone. Reversed binding orientations in different redox states of the enzyme, J. Biol. Chem. 280, 13256-13264.

25

ACS Paragon Plus Environment

ACS Chemical Biology

Gruber et al.

FIGURES

Figure 1

Figure 2 tdNTR(F124W)mCherry: eNTR

a Relative enzyme activity

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 26 of 30

tdNTRmCherry

10×

NTRmCherry

NTR

1× expression



activity per monomer



c

b

100×

3.5× 4.7× 1×

2.6×

d

4.7× 18×

e

26

ACS Paragon Plus Environment

Page 27 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Biology

Gruber et al.

Figure 3

27

ACS Paragon Plus Environment

ACS Chemical Biology 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 28 of 30

Gruber et al.

Figure 4

28

ACS Paragon Plus Environment

Page 29 of 30

Gruber et al.

a

G120

Y68

T41 S40

PDB 1YKI

Figure 5

c

N

e

O

O2N N

N

O

O 2N N

O

O N

N F124

NTR N71

OH

14

12

NTR

F70 HN

HN OH

11

13

Normalized Ca2+ response

d 6

3

WT S40A Y68M Y68W F70G F70L F70M F70S F70T F70V G120A F124M F124W

0

f eNTR– not co-cultured

1

eNTR– co-cultured eNTR+

0 0

1

10

100

Normalized Ca2+ response

b Fold kcat improvement

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Biology

eNTR– not co-cultured

1

eNTR– co-cultured eNTR+ 30×

0 0

10

100

1000

Concentration of 14 (µM)

Concentration of 12 (µM)

29

ACS Paragon Plus Environment

O2N N

N

O F

1 2 3 4 5

O

O O

N

NO 2

ACS Chemical Biology Page 30 of 30 N

F O

Masked compound: Dye, Indicator, or Drug

Engineered enzyme in specific cells (red)

ACS Paragon Plus Environment

No enzyme (blue) Cell-specific unmasking of compound (green)