Covalent live-cell labeling of proteins using a photoreactive fluorogen

Article abstract of DOI:10.1016/bs.mie.2020.04.019

Fluorescence microscopy has dramatically advanced our understanding of the processes that drive biological systems by enabling the imaging and tracking of biomolecules of interest inside of living cells. In particular, proteins of interest can be genetica

Full text of DOI:10.1016/bs.mie.2020.04.019

ARTICLE IN PRESS
Covalent live-cell labeling
of proteins using a
photoreactive fluorogen
Tewoderos M. Ayele†, Steve D. Knutson†, Jennifer M. Heemstra∗
Department of Chemistry, Emory University, Atlanta, GA, United States
*Corresponding author: e-mail address: jen.heemstra@emory.edu
Contents
1. Introduction
2. Chemical synthesis of malachite green diazirine
2.1 Rationale
2
6
6
2.2 Equipment
7
2.3 Chemicals
8
2.4 Protocol—Chemical synthesis of MG-diazirine
3. Preparation of FAP fusion vectors
3.1 Rationale
3.2 Equipment
3.3 Chemicals
3.4 Protocol—Preparation of mCer3-FAP plasmid vector
4. Live cell imaging with malachite green diazirine
4.1 Rationale
4.2 Equipment
4.3 Chemicals
4.4 Protocol—Live cell imaging with MG-diazirine
5. Summary
References
8
13
13
14
15
15
16
16
16
17
17
22
22
Abstract
Fluorescence microscopy has dramatically advanced our understanding of the pro-
cesses that drive biological systems by enabling the imaging and tracking of bio-
molecules of interest inside of living cells. In particular, proteins of interest can be
genetically tagged with fluorescent proteins or labeled with small molecule fluorophore
probes to enable visualization. However, both of these methods are generally limited in
signal-to-background resolution and options are limited for achieving temporal control
†
These authors contributed equally.
#
Methods in Enzymology
ISSN 0076-6879
https://doi.org/10.1016/bs.mie.2020.04.019
2020 Elsevier Inc.
All rights reserved.
1

ARTICLE IN PRESS
2
Tewoderos M. Ayele et al.
over labeling. Photoreactive “fluorogenic” dyes can overcome these limitations and
enable user-defined crosslinking with low background fluorescence. In this chapter,
we discuss current approaches for live cell protein labeling with particular emphasis
on the novel use of photoreactive fluorogenic dyes for protein imaging. We further
describe in detail the synthesis and characterization of a fluorogenic malachite green
probe functionalized with a photoreactive diazirine crosslinker and illustrate how to
apply this probe toward covalent photoaffinity labeling and imaging of target proteins
in live cells.
1. Introduction
Fluorescence microscopy has significantly advanced our understand-
ing of cell biology by enabling direct observation of the complex and
dynamic activities of myriad biomolecules within the cell (Stephens &
Allan, 2003). In particular, nucleic acids and proteins can be fluorescently
tagged using a number of methods to enable visualization and tracking.
The ability to observe the location and movement of biomolecules has
in turn provided significant insights into the interactions of these mole-
cules and their contributions to overall cellular structure and function
(Giepmans, Adams, Ellisman, & Tsien, 2006).
To enable imaging of proteins in live cells, an early strategy was devel-
oped in which a protein of interest (POI) is genetically fused in frame to a
fluorescent protein (FP) (Toseland, 2013). To obtain this gene fusion, the
coding DNA sequence for the POI is amplified, isolated, and ligated to
the gene of an appropriate FP, such as green fluorescent protein (GFP)
(Chalfie, Tu, Euskirchen, Ward, & Prasher, 1994) or one of the many
other color variants (Fig. 1A). This fusion construct is then inserted into
an appropriate mammalian expression plasmid vector and introduced into
cells, which subsequently express the fluorescent fusion product to enable
visualization. This technique has been widely adopted and is quite useful,
as all components needed for visualization are genetically encodable and
produced inside of the target cells. Further, molecular cloning techniques
are relatively cheap and straightforward, and various fusion protein designs
and iterations can be quickly constructed and inserted into appropriate
plasmid vectors using economical reagents (Giepmans et al., 2006). These
constructs can also be easily introduced into most cell types using standard
transient transfection techniques or, when required, stably integrated
into cellular genomes using lentiviral or CRISPR-based strategies

ARTICLE IN PRESS
Covalent live-cell labeling of proteins
3
Fig. 1 Methods for genetic tagging and fluorescent visualization of target proteins.
(A) The protein of interest (POI) can be genetically fused to an intrinsically fluorescent
protein (FP), or (B) attached to a peptide or protein tag that is recognized by reactive
fluorophores. (C) The POI can also be fused to a fluorogen activating protein (FAP),
which enhances the fluorescent signal of specific fluorogen molecules. Adapted with
permission from Ayele, T. M., Knutson, S. D., Ellipilli, S., Hwang, H., & Heemstra, J. M.
(2019). Fluorogenic photoaffinity labeling of proteins in living cells. Bioconjugate
Chemistry, 30(5), 1309–1313. Copyright (2019) American Chemical Society.
(Lackner et al., 2015). However, the constitutive fluorescence of these FP
fusions presents a large limitation in signal-to-noise resolution. To achieve
sufficient fluorescent signal inside this cell, strong expression promoters are
typically required in mammalian plasmid vectors, including cytomegalovirus
(CMV) and elongation factor-1a (EF-1a). While these promoters are often
necessary to achieve sufficient protein levels for many POI fusions to enable
robust visualization, this technique can also be problematic in that it results
in the uncontrolled production of an unnaturally high copy number of the
fusion protein. This can generate an overall diffuse signal throughout the cell
and may not recapitulate natural expression levels or subcellular localization
(Deer & Allison, 2004; Qin et al., 2010).
To overcome these limitations, a number of alternative protein labeling
techniques have been developed that utilize chemical conjugation strate-
gies to functionalize POI with fluorescent organic dyes. These approaches
enhance the selective reaction of the fluorescent probe by expressing a
POI fused with a genetically encodable peptide or protein motif that has high
affinity to the fluorescent dye (Fig. 1B). An early example of this approach is
the tetracysteine-biarsenical system (Griffin, Adams, & Tsien, 1998). The
membrane-permeable fluorophore ligands used for this technique, ReAsH
and FIAsH, contain dithioarsolane groups to enable strong non-covalent che-
lation with the Cys-Cys-X-X-Cys-Cys peptide motif fused to POI, where
X can be any amino acid. This system is elegant in its simplicity but presents

ARTICLE IN PRESS
4
Tewoderos M. Ayele et al.
limitations for broad applicability due to high background signal, potential
arsenic toxicity, and elevated photobleaching properties of the fluorescent
ligands (Marks & Nolan, 2006; Stroffekova, Proenza, & Beam, 2001).
This inspired the development of small-molecule labeling systems that
instead take advantage of enzymatic recognition to drive selectivity.
These approaches include the SNAP/Clip-tag (Gautier et al., 2008), LAP
(Ferna´ndez-Sua´rez et al., 2007), Halo-tag (Los et al., 2008), and coiled-coil
tag (Reinhardt, Lotze, Mo€rl, Beck-Sickinger, & Seitz, 2015). The advan-
tages of these methods include substantially improved control over the
intensity and timing of fluorescent labeling and the ability to harness
the diverse palette of small-molecule fluorophores. However, covalent pro-
tein labeling techniques that target specific amino acids or short peptide
sequences still lack high labeling selectivity in complex biological systems
where other proteins or molecules are likely to have similar reactive groups.
For example, protein labeling systems that rely on cysteine residue func-
tionalization lose their selectivity in environments that contain high concen-
trations of glutathione or other biological molecules that have nucleophilic
thiol groups. Additionally, these techniques still rely on intrinsically fluores-
cent molecules, which produce high background signal and hence require
extensive washing steps to remove unreacted dye from the cytosol.
In order to address these challenges, several next-generation fluorophore
molecules have been developed that are conditionally fluorescent and
undergo a dramatic increase in emission when particular biochemical or
physical conditions are met (Bruchez, 2015). One class of these probes
are fluorogenic molecules, which can adopt somewhat similar planar archi-
tectures compared to traditional organic fluorophores yet display the impor-
tant inclusion of a freely rotating bond, which prevents the molecule
from spending a significant amount of time in the planar conformation.
This results in low fluorescence in solution, with significant enhancement
of signal upon rotational restriction imparted by temperature or viscosity
changes. Alternatively, this signal enhancement can be achieved through
binding to a nucleic acid strand or protein if the binding mode restricts
the fluorogenic molecule in a planar conformation. In practice, visualization
of POIs has been achieved by combining a fluorogenic small molecule and a
“fluorogen activating protein” (FAP) that has been engineered to tightly
bind to the dye molecule and restrict its overall rotational movement.
Similar to previous genetic tagging methods, the POI can then be visualized
by in-frame genetic fusion with the FAP domain (Fig. 1D), generating high
signal-to-background ratio without the requirement of washing away
unbound fluorogen (Chen et al., 2014; Telmer et al., 2015).

ARTICLE IN PRESS
Covalent live-cell labeling of proteins
5
Fig. 2 Photoreactive malachite green enables covalent fluorogenic labeling of proteins.
Chemical structures of (A) malachite green (Qin et al., 2010) and (B) malachite green
diazirine (MG-diazirine), illustrating arginine spacer and diazirine photo-crosslinking
group. (C) Genetic fusion of a POI with FAP combined with MG-diazirine enables fluo-
rogenic localization and UV crosslinking of MG to target proteins. Adapted with permis-
sion from Ayele, T. M., Knutson, S. D., Ellipilli, S., Hwang, H., & Heemstra, J. M. (2019).
Fluorogenic photoaffinity labeling of proteins in living cells. Bioconjugate Chemistry,
30(5), 1309–1313. Copyright (2019) American Chemical Society.
Malachite green (Qin et al., 2010) is a well-characterized and widely
adopted small molecule fluorogenic dye, originally developed and used as
an industrial pigment (Shukla & Mathur, 1995) and has since found
extensive use in bioimaging applications. MG is weakly fluorescent in solu-
tion and displays low quantum yields due to free rotational movement
of two of the aromatic rings in the triphenylmethane structure (Fig. 2A).
However, when this rotation is restricted, fluorescence is greatly enhanced.
This advantageous property was exploited for RNA imaging purposes,
where researchers used directed evolution to evolve an RNA aptamer
motif that can bind to MG and generate a fluorescent signal. By fusing
this sequence to an RNA transcript of interest, this technique enables visu-
alization of RNA molecules and real-time tracking of their movement
within different cellular compartments (Babendure, Adams, & Tsien,
2003; Yerramilli & Kim, 2018). Similarly, researchers sought to apply this
property toward protein labeling and imaging, and utilized yeast-display
to select for single-chain antibody FAPs that exhibit strong binding affinity
toward MG (Szent-Gyorgyi et al., 2008). While this approach circumvents
many of the background resolution problems previously described, the
fluorogen-FAP interaction is non-covalent, and thus the signal generated

ARTICLE IN PRESS
6
Tewoderos M. Ayele et al.
is directly limited by the strength of this binding interaction. Moreover, the
MG dye can become unbound from the FAP and diffuse out of cells over
time, resulting in shorter visualization windows and a lack of precise
temporal control over when signal can be produced.
To address these limitations, we derivatized the MG scaffold with both
an arginine spacer and a photoreactive diazirine group (Fig. 2B) (Ayele,
Knutson, Ellipilli, Hwang, & Heemstra, 2019). The arginine moiety aids
in solubility of the molecule by imparting charge, and appending this amino
acid has also been shown to improve cell permeability and uptake of small
molecule and peptide payloads (Brock, 2014; Rothbard et al., 2002). To
provide a photoreactive group on the molecule, we attached a diazirine
moiety proximal to the arginine linker. Diazirine forms a highly reactive
carbene upon irradiation with ultraviolet light, which then reacts to form
covalent attachments with neighboring biomolecules through insertion at
CdH, NdH, or OdH bonds (Dubinsky, Krom, & Meijler, 2012). The
derivatized MG-diazirine fluorogen retains the ability to bind with its
cognate antibody-chain FAP (Szent-Gyorgyi et al., 2008) and produces
significant fluorescence enhancement activity, and the appended diazirine
enables permanent covalent attachment to the protein via UV irradiation
(Fig. 2C). Together, this attachment results in stable fluorogenic labeling
of target proteins and allows user-defined timing of this labeling event. In
this chapter we describe in detail a protocol for the chemical synthesis, puri-
fication, and characterization of the MG-diazirine small molecule. We addi-
tionally illustrate how to introduce this compound into cells and visualize a
POI-FAP fusion using photoaffinity labeling and fluorescence microscopy.
2. Chemical synthesis of malachite green diazirine
2.1 Rationale
Malachite green (Qin et al., 2010) is among the most widely used fluorogenic
probes for biological applications, and binds to an engineered fluorogen acti-
vating protein (FAP) with high affinity (Szent-Gyorgyi et al., 2008). This
FAP domain can be genetically fused to proteins of interest to enable fluo-
rogenic live-cell labeling. However, this probe-protein interaction is
non-covalent, and is thus limited by diffusion-based signal losses and a lack
of temporal control in labeling. Using a procedure modified from previous
synthetic routes (Xing et al., 2009), this protocol describes a route to syn-
thesize and derivatize a malachite green fluorogen having a photoreactive
diazirine crosslinker for covalent labeling of FAP-POI fusions (Scheme 1).

ARTICLE IN PRESS
Covalent live-cell labeling of proteins
7
Scheme 1 Synthesis of MG-diazirine. Adapted with permission from Ayele, T. M.,
Knutson, S. D., Ellipilli, S., Hwang, H., & Heemstra, J. M. (2019). Fluorogenic photoaffinity
labeling of proteins in living cells. Bioconjugate Chemistry, 30(5), 1309–1313. Copyright
(2019) American Chemical Society.
2.2 Equipment
(a) Rotary evaporator equipped with vacuum system
(b) Magnetic stirrer with heating module
(c) Stir bars
(d) Glass funnel
(e) #1 Whatman Paper
(f ) Latex balloons
(g) Glass chromatography column
(h) 5mL Round bottom flask
(i) 25mL round bottom flask
(j) 2mL glass vial
(k) 25mL glass separatory funnel

ARTICLE IN PRESS
8
Tewoderos M. Ayele et al.
(l) Preparative TLC plates (Sigma-Aldrich) and chamber
(m) Hamilton syringes
(n) NMR tubes
2.3 Chemicals
(a) p-Nitrobenzaldehyde (Sigma-Aldrich)
(b) Zinc chloride (Chem-Impex International, Inc.)
(c) N,N-Dimethylaniline (Sigma-Aldrich)
(d) Acetone (Fisher Scientific)
(e) Ethyl acetate (Fisher Scientific)
(f ) Hexane (Fisher Scientific)
(g) Methanol (Fisher Scientific)
(h) Tetrahydrofuran (Fisher Scientific)
(i) Palladium on carbon (Pd/C 5%, w/w) (Sigma-Aldrich)
(j) Hydrogen gas (Nexair)
(k) Celite (Sigma-Aldrich)
(l) Pyridine (Chem-Impex International, Inc.)
(m) N,N0-Diisopropylcarbodiimide (Chem-Impex International, Inc.)
(n) Fmoc-Arg(Boc)2-OH (Chem-Impex International, Inc.)
(o) Dichloromethane (Fisher Scientific)
(p) Water
(q) Anhydrous sodium sulfate (Sigma-Aldrich)
(r) Dimethylformamide (Sigma-Aldrich)
(s) Piperidine (Chem-Impex International, Inc.)
(t) SDA (NHS-Diazirine, or succinimidyl 4,40-azipentanoate) (Thermo
Fisher Scientific)
(u) Acetic acid (Fisher Scientific)
(v) Chloroform (Fisher Scientific)
(w) Trifluoroacetic acid (Sigma-Aldrich)
(x) Hydrochloric acid (Fisher Scientific)
2.4 Protocol—Chemical synthesis of MG-diazirine
2.4.1 p-nitro-malachite green (compound 1)
(a) Dissolve p-nitrobenzaldehyde (250mg, 1.65mmol) and zinc chloride
(451mg, 3.31mmol) in N,N-dimethylaniline (0.503mL, 3.97mmol)
in a 5mL round bottom flask.
(b) Stir this solution while refluxing at 100 °C for 5h.
(c) Cool the reaction mixture to room temperature and add 5mL of
acetone.

ARTICLE IN PRESS
Covalent live-cell labeling of proteins
9
(d) Filter off the undissolved zinc salt using Whatman filter paper and a glass
funnel. Repeat this washing step with acetone two more times to
remove the maximum amount of zinc salt.
(e) Concentrate the solution under reduced pressure and purify the resulting
crude oil by flash column chromatography (silica, 1:10 ethyl acetate/
hexane). Note: To simplify the purification of the crude oil by flash
chromatography, remove as much of the excess N,N-dimethylaniline
as possible under reduced pressure before loading onto the silica column.
(f ) Remove solvent from collected and pooled column fractions in vacuo to
yield a yellow solid, p-nitro-leucomalachite green (compound 1).
2.4.2 p-amino-leucomalachite green (compound 2)
(a) Dissolve p-nitro-leucomalachite green (1) (100mg, 0.267mmol) in 9mL
of methanol/tetrahydrofuran (1/2, v/v) in a 25mL round bottom flask.
(b) To this solution add 10mg of Pd/C (5%, w/w). Note: When working
with Pd/C and hydrogen gas, proper precautions need to be taken to
prevent ignition of flammable solvents.
(c) Cap the round bottom flask with a rubber septum and start stirring the
mixture.
(d) With a needle and vacuum line, evacuate the reaction flask until bub-
bling is observed. Stop the vacuum and carefully insert the hydrogen
balloon to backfill the flask. Remove the hydrogen balloon. Repeat this
process three more times.
(e) Let the reaction stir at room temperature for 3h.
(f ) Filter the crude solution through a celite and wash the celite with 5mL
of methanol three times.
(g) Concentrate the pooled filtrate under reduced pressure to yield a light
blue solid (2).
(h) To assess yield and purity of the collected intermediate product, perform
1H NMR at this step. A representative spectrum is shown in Fig. 3.
2.4.3 (9H-Fluoren-9-yl)methyl(1-((4-(bis(4-(dimethylamino)phenyl)
methyl)phenyl)amino)-5-N,N-diboc-guanidino-1-oxopentan-2-yl)
carbamate (compound 3)
(a) To a 2mL glass vial containing a micro stir bar, add 2
(50.0mg, 0.148mmol) to 500μL of pyridine. To this solution, add
N,N0-diisopropylcarbodiimide (70.0μL, 0.445mmol) and Fmoc-
Arg(Boc)2-OH (106mg, 0.178mmol). Stir at room temperature
for 12h.

ARTICLE IN PRESS
10
Tewoderos M. Ayele et al.
N
NH2
N
13.0
12.0
11.0
10.0
9.0
8.0
7.0
6.0
5.0
4.0
3.0
2.0
1.0
0.
f1 (ppm)
Fig. 3 Representative 1H NMR spectrum of compound 2. Adapted with permission from
Ayele, T. M., Knutson, S. D., Ellipilli, S., Hwang, H., & Heemstra, J. M. (2019). Fluorogenic pho-
toaffinity labeling of proteins in living cells. Bioconjugate Chemistry, 30(5), 1309–1313.
Copyright (2019) American Chemical Society.
(b) Concentrate the reaction under reduced pressure, and dissolve the
crude product in 5mL of dichloromethane.
(c) Transfer the solution to a 25mL separatory funnel containing water
(10mL) and shake vigorously.
(d) Allow the two layers to separate.
(e) Carefully drain the bottom organic layer into an Erlenmeyer flask, and
collect the top aqueous layer into a separate suitable waste container.
(f ) Transfer the collected organic layer from the previous step into a
new 25mL separatory funnel containing water (10mL) and repeat steps
(c)–(e) two additional times.
(g) Remove residual water from the combined organic layers by directly
adding anhydrous sodium sulfate.
(h) Carefully decant organic layer into a new flask and concentrate under
reduced pressure.
(i) Purify the crude product using flash column chromatography (silica).
Load the crude product onto the silica with a minimal amount of

ARTICLE IN PRESS
Covalent live-cell labeling of proteins
11
Fig. 4 Representative 1H NMR spectrum of compound 3. Adapted with permission from
Ayele, T. M., Knutson, S. D., Ellipilli, S., Hwang, H., & Heemstra, J. M. (2019). Fluorogenic pho-
toaffinity labeling of proteins in living cells. Bioconjugate Chemistry, 30(5), 1309–1313.
Copyright (2019) American Chemical Society.
dichloromethane using a Pasteur pipette and elute with 5%, 10%, 15%,
35%, and 50% ethyl acetate in hexane to yield a dark green solid 3.
(j) To assess yield and purity of the collected product, perform 1H NMR at
this step. A representative spectrum is shown in Fig. 4.
2.4.4 (Z)-N-(4-((4-(5-(2,3-Bis(tert-butoxycarbonyl)guanidino)-2-
(3-(3-methyl-3H-diazirin-3-yl)propanamido)pentanamido)
phenyl)(4-(dimethylamino)phenyl)methylene)cyclohexa-
2,5-dien-1-ylidene)-N-methylmethanaminium (compound 4)
(a) Note: Diazirine is photoreactive. Use proper precautions to minimize
ultraviolet light exposure in all steps during this stage of the procedure.
(b) Dissolve compound 3 (30.0mg, 0.0325mmol) in 500μL of
dimethylformamide in a 2mL glass vial.
(c) To this solution, add 500μL of 20% piperidine in N,N-
dimethylformamide and stir this solution for 1h at room temperature.
(d) Purify this product using preparative TLC (silica, 2/100 methanol/
dichloromethane) to yield a dark blue solid.
(e) Dissolve the obtained solid in 500μL of N,N-dimethylformamide.

ARTICLE IN PRESS
12
Tewoderos M. Ayele et al.
(f ) To this solution, add succinimidyl 4,40-azipentanoate (22.0mg,
0.0975mmol) and stir the solution at room temperature for 5h.
(g) Concentrate the reaction mixture under reduced pressure to obtain a
crude product of the diazirine functionalized intermediate.
(h) To this solid, add 100μL of 30% acetic acid in chloroform, and reflux
the solution at 60 °C for 4h.
(i) Dry and concentrate the reaction under reduced pressure and purify com-
pound 4 using preparative TLC (silica, 1:10 methanol/dichloromethane).
(j) To assess yield and purity of the collected product, perform 1H NMR at
this step. A representative spectrum is shown in Fig. 5.
2.4.5 N-(4-((4-(Dimethylamino)phenyl)(4-(5-
guanidino-2-(3-(3-methyl-3H-diazirin-3-yl)propanamido)
pentanamido)phenyl)methylene)cyclohexa-2,5-dien-1-ylidene)-
N-methylmethanaminium (compound 5, MG-diazirine)
(a) Note: The diazirine functionalized intermediate at this stage is photo-
reactive. Use proper precautions to minimize ultraviolet light exposure.
Fig. 5 Representative 1H NMR spectrum of compound 4. Adapted with permission from
Ayele, T. M., Knutson, S. D., Ellipilli, S., Hwang, H., & Heemstra, J. M. (2019). Fluorogenic pho-
toaffinity labeling of proteins in living cells. Bioconjugate Chemistry, 30(5), 1309–1313.
Copyright (2019) American Chemical Society.

ARTICLE IN PRESS
Covalent live-cell labeling of proteins
13
(b) In a 5mL round bottom flask, dissolve compound 4 (20.0mg,
0.0246mmol) in 2mL of 50% trifluoroacetic acid/dichloromethane.
(c) Stir the reaction for 4h at room temperature to allow for Boc-
deprotection. Remove the solvent in vacuo.
(d) To remove residual trifluoroacetic acid, redissolve the solid in 10% aq.
hydrochloric acid and concentrate reaction mixture under reduced
pressure to obtain product 5 as a dark blue solid (MG-diazirine.)
1
(e) To assess yield and purity of the collected final product, perform H
NMR at this step. A representative spectrum is shown in Fig. 6.
(f ) Store final product as a solid at room temperature. Protect from light.
3. Preparation of FAP fusion vectors
3.1 Rationale
This procedure outlines how to prepare a plasmid for transfection into
HeLa cells for subsequent live cell imaging with the MG-diazirine probe.
Specifically, we utilized a commercially available pcDNA vector from
Addgene (Fig. 7), which contains the FAP protein fused to a fluorescent
1
Fig. 6 Representative H NMR spectrum of compound 5 (MG-diazirine). Adapted with
permission from Ayele, T. M., Knutson, S. D., Ellipilli, S., Hwang, H., & Heemstra, J. M.
(2019). Fluorogenic photoaffinity labeling of proteins in living cells. Bioconjugate
Chemistry, 30(5), 1309–1313. Copyright (2019) American Chemical Society.

ARTICLE IN PRESS
14
Tewoderos M. Ayele et al.
Fig. 7 Plasmid map of mCer3-FAP control expression vector highlighting important
components for cloning and expression.
mCerulean3 (mCer3) reporter protein (Telmer et al., 2015). The fluores-
cence reporter protein served as a transfection control in the initial validation
of this approach, but use of a fluorescent protein is not required for imaging
applications. The vector also contains both bacterial replication features and
appropriate promoters for mammalian expression. These vectors are typi-
cally supplied as agar stabs containing pre-transformed E. coli cells harboring
the plasmid of interest. We outline how to replicate and extract this plasmid
from bacteria for eventual transfection into HeLa cells for expression. This
vector also contains numerous restriction enzyme sites surrounding the
mCer3 gene (Fig. 7), allowing users to clone and insert virtually any POI
into this backbone and fuse to FAP.
3.2 Equipment
(a) Autoclave
(b) 100mmꢀ15mm sterile petri dishes with lids (VWR)
(c) 5mL culture tubes, plastic, with Caps (VWR)
(d) Sterile pipet tips/toothpicks
(e) Inoculating loop (optional)

ARTICLE IN PRESS
Covalent live-cell labeling of proteins
15
(f ) stationary incubator capable of reaching 37 °C
(g) 15mL conical tubes with caps (VWR)
(h) shaker incubator capable of reaching 37 °C and 250rpm
(i) centrifuge that can accommodate 15mL conical tubes and reach
>6800x g
(j) microcentrifuge that can accommodate 1.5mL tubes and reach
>10,000x g
(k) UV–Vis spectrophotometer
3.3 Chemicals
(a) Plasmid vector pcDNA3.1-KozATG-dL5-2XG4S-mCer3 (Addgene
73207)
(b) LB agar (Miller), granulated (EMD Millipore)
(c) LB broth (Miller), granulated (EMD Millipore)
(d) Ampicillin, sodium salt (Sigma Alrich)
(e) QIAprep Spin Miniprep Kit (Qiagen)
(f ) Ultrapure water
3.4 Protocol—Preparation of mCer3-FAP plasmid vector
(a) Prepare luria broth (LB) agar (Miller) plates supplemented with ampi-
cillin by first combining 37g of the powdered media mixture for every
1L of ultrapure water. Mix to dissolve completely, and sterilize in a
suitable container by autoclaving.
(b) Separately, prepare liquid LB broth (Miller) media by combining 25g
of the powdered media mixture for every 1L of ultrapure water. Mix to
dissolve completely, and sterilize in a suitable container by autoclaving.
(c) Place hot LB broth and agar mixture in a water bath equilibrated to
50 °C. While media is equilibrating, prepare a 100mg/mL ampicillin
(amp) stock solution in ultrapure water.
(d) When LB broth and agar is cool to the touch, dilute amp stock 1000ꢀ
in the LB agar medium to a final concentration of 100μg/mL.
(e) While LB agar is still a molten liquid, carefully but quickly pour
ꢁ20–25mL into 100mmꢀ15mm sterile petri dishes. Cover loosely
with lids and allow to solidify at room temperature.
(f ) Addgene vectors typically arrive as agar stabs which contain E. coli cells
transformed with the plasmid of interest. To grow more bacteria
through replication, insert a sterile pipet tip, toothpick, or inoculating
loop into the stab, and streak onto a LB+amp agar plate from step (e).

ARTICLE IN PRESS
16
Tewoderos M. Ayele et al.
Use aseptic technique and serially streak while rotating the plate
to obtain single colonies. Incubate plate(s) upside down at 37 °C
overnight.
(g) Inspect plates. If successful, the streak plate(s) should display robust
bacterial growth and many distinct colonies. To store plates long term,
wrap plates in parafilm and store at 4 °C. Plates can be stored for at least
1year if kept dry.
(h) Prepare an overnight starter culture by adding 5mL of LB broth + amp
to a desired number of culture tubes. Using a sterile pipet tip or tooth-
pick, “pick” an individual bacterial colony by gently touching it. Drop
the tip/toothpick directly into one culture tube with broth and cap.
Repeat for desired number of tubes. Transfer all tubes to a shaking
incubator in a suitable rack, and incubate overnight at 37 °C with
shaking at 250rpm.
(i) The next day, visually inspect culture tubes. Broth should be noticeably
turbid, indicating successful bacterial growth.
(j) Extract plasmid using the QIAprep Spin Miniprep kit according the
manufacturer’s instructions. Several other similar kits are commercially
available from a number of vendors. Quantify the concentration of
purified plasmid by UV–Vis spectrophotometry, and store at 4 °C until
further use.
4. Live cell imaging with malachite green diazirine
4.1 Rationale
Lastly, we describe how to transfect HeLa cells with these vectors and
image the expressed fusion protein through introduction of MG-diazirine
followed by photo-crosslinking and analysis by fluorescence microscopy.
4.2 Equipment
(a) Laminar flow hood
(b) CO2 incubator
(c) Centrifuge
(d) Water bath (37°C)
(e) Inverted brightfield microscope
(f ) Fluorescent microscope
(g) Hemacytometer or automated cell counter
(h) T25 flasks

ARTICLE IN PRESS
Covalent live-cell labeling of proteins
17
(i) T75 flasks
(j) Sterile 15mL polypropylene conical centrifuge tubes
(k) Sterile 96-well tissue culture plates, with lids
(l) Disposable pipettes
(m) High intensity UV lamp, 100W/ 365nm (Analytik Jena)
4.3 Chemicals
(a) HeLa cells (ATCC CCL-2)
(b) Eagle’s Minimum Essential Medium (EMEM) (Thermo Fisher
Scientific)
(c) FBS (Fetal bovine serum) (Thermo Fisher Scientific)
(d) Penicillin-streptomycin (10,000U/mL) (Thermo Fisher Scientific)
(e) Trypsin-EDTA (0.25%), phenol red (Thermo Fisher Scientific)
(f ) 70% Ethanol
(g) Opti-MEM Reduced Serum Medium (Thermo Fisher Scientific)
(h) Lipofectamine 3000 with P3000 reagent (Thermo Fisher Scientific)
(i) Dimethylsulfoxide (DMSO) (Fisher Scientific)
4.4 Protocol—Live cell imaging with MG-diazirine
4.4.1 HeLa cell culture and maintenance
Note: Mammalian cell culture requires use of appropriate biosafety control
measures and may require institutional training and safety approval.
Note: All steps must be performed using sterile aseptic technique in a
laminar flow hood. All media should be purchased sterile or filtered through
0.2μm filters. All materials need to be autoclaved or sprayed down liberally
with 70% ethanol prior to bringing inside the laminar flow hood.
Note: All media and liquids should be prewarmed to 37°C before com-
ing into contact with cells.
(a) Prepare “complete” EMEM media by combining the following.
Mix well. Scale as needed.
i. EMEM base media: 450mL
ii. FBS: 50mL
iii. Pen/Strep (10,000U/mL): 5mL
(b) HeLa cell stocks are stored in liquid nitrogen in cryovials. Begin a starter
culture by immersing the frozen cryovial in a 37°C water bath until
fully thawed. Take care to only submerge the bottom half of the vial.
(c) When thawed, aseptically transfer contents of the vial (ꢁ1–2mL) to a
15mL conical tube containing 10mL of EMEM complete media
prewarmed to 37°C.

ARTICLE IN PRESS
18
Tewoderos M. Ayele et al.
(d) Pellet cells by spinning at 500x g for at least 5min. Carefully remove
supernatant and resuspend gently in 10mL of fresh EMEM
complete media.
(e) Transfer to a T25 flask and incubate at 37°C, 5% CO2.
(f ) The next day, use brightfield microscope to confirm adherence of cells
and the absence of floating or dead cells.
(g) Continue to incubate flask at 37°C, 5% CO2. Replace media 2–3 times
per week, or if media has begun to turn yellow-orange.
(h) When cells have reached >75% confluency, remove media in flask
and rinse with ꢁ2–3mL of 0.25% Trypsin/EDTA. Rock back and
forth gently and discard. Replace with 2mL of fresh 0.25% Trypsin/
EDTA and repeat. Remove 1.5mL of this solution and return cells
to 37°C incubator for 5–20min. Monitor cell detachment by visualiz-
ing under the microscope.
(i) When fully detached, add 9.5mL fresh EMEM complete media and
rinse cells from the bottom of the flask. Pipet up and down gently to
fully resuspend and distribute cells.
(j) Split cells 1:2 into a larger T75 flask by combining 5mL of the
resuspended mixture from step (g) with 20mL of fresh media. Mix
gently by pipetting up and down and incubate flask horizontally at
37°C, 5% CO2. Replace media 2–3 times per week, or if media has
begun to turn yellow-orange.
(k) If needed, subcultivate as outlined in steps (f )–(h) using splitting ratios
between 1:2 and 1:10 where appropriate. Note: HeLa cells should be
passaged 2–3 times after seeding a flask from cryostorage before they
are suitable for transfection. Do not exceed confluency over ꢁ95%
in flasks. Do not culture cells beyond 25 rounds of passage.
4.4.2 Plasmid transfection
(a) At this stage, it is valuable to perform an optimization experiment to
identify suitable conditions for transfection and confirm FAP-fusion
expression inside cells. Additionally, this experiment further confirms
successful synthesis of the MG-diazirine ligand.
(b) After a suitable number of passages, HeLa cells are ready for transfec-
tion at confluencies anywhere between 50% and 75% in a T75 flask.
Begin plating by removing media in flask and rinsing with ꢁ2–3mL of
prewarmed 0.25% Trypsin/EDTA. Rock back and forth gently and
discard. Replace with 2mL of fresh 0.25% Trypsin/EDTA and repeat.

ARTICLE IN PRESS
Covalent live-cell labeling of proteins
19
Remove 1.5mL of this solution and return cells to 37°C incubator
for 5–20min. Monitor cell detachment by visualizing under the
microscope.
(c) When fully detached, add 9.5mL fresh EMEM complete media
and rinse cells from the bottom of the flask. Pipet up and down gently
to fully resuspend and distribute cells.
(d) Take a small sample of this suspension and count cells using a hema-
cytometer or automated cell counter.
(e) Dilute cells to approximately 1ꢀ105 cells/mL using EMEM complete
media. Prepare at least 12mL of this suspension for each full 96-well
plate. This can be scaled up or down as needed. Mix solution gently
but thoroughly using a pipet or inversion to evenly distribute cells.
(f ) Transfer 100μL of this suspension to each desired well of a 96-well
plate. Add 100μL EMEM media to any empty wells. Incubate plates
at 37°C, 5% CO2 overnight.
(g) The next day, confirm adherence of cells and the absence of floating or
dead cells.
(h) Monitor cells daily until they reach confluency of 70%–90%.
(i) Remove media from wells and replace with 100μL pre-warmed
Opti-MEM reduced serum medium.
(j) Prepare transfection solution A by combining the following. Amounts
correspond to enough transfection material for 1 well. Scale up as
needed.
i. Opti-MEM reduced serum medium: 5μL
ii. Lipofectamine 3000: 0.15μL
(k) Prepare transfection solution B by combining the following. Amounts
correspond to enough transfection material for 1 well. Scale up as
needed.
i. Opti-MEM reduced serum medium: 10μL
ii. pcDNA FAP Plasmid Vector: 200ng
iii. P3000 Reagent: 0.4μL
(l) Combine solutions by adding 5μL each of A and B. Mix well by
flicking the tube. Let incubate at room temperature for at least 30min.
(m) Add 10μL of this combined A/B mixture to each desired well of the
96-well plate. Note: It is strongly advised to also prepare control wells
without DNA added as well as a “no transfection” control well.
(n) Incubate cells at 37°C, 5% CO2 for 12h. After this period, remove
Opti-MEM from each well and replace with 100μL fresh EMEM
complete media. Incubate cells at 37°C, 5% CO2 for an additional 6h.

ARTICLE IN PRESS
20
Tewoderos M. Ayele et al.
(o) Prepare a 50mM MG-diazirine stock solution (ꢁ30.5mg/mL)
in DMSO.
(p) Prepare a 50μM working solution of MG-diazirine in EMEM com-
plete media by diluting the 50mM stock 1:1000.
(q) Remove media from each well and add 100μL of this working solu-
tion to each appropriate well. Let cells incubate for 15min at 37°C.
(r) Remove MG-diazirine solution from each well and briefly wash three
times with 100μL fresh EMEM complete media.
(s) Proceed to imaging. We previously used a Leica DMi8 confocal
fluorescence microscope with a 10ꢀ objective to obtain our images.
However, most standard fluorescent microscopes should be suitable
for imaging. The most important consideration is the available exci-
tation lasers and emission filters in the microscope. If using the
mCer3-FAP plasmid, the required spectral considerations for mCer3
are excitation maximum at 433nm, emission at 475nm and FAP/
MG-diazirine excitation maximum at 640nm and emission at 668nm.
(t) Obtain images for each appropriate channel using the microscope.
Images should show a clear distinction when comparing between
cells that have been transfected with the reporter mCer3-FAP plasmid
versus non-transfected controls. Representative images are shown in
Fig. 8, illustrating signal from both mCer3 and FAP-MG diazirine.
This experiment thus confirms successful synthesis of MG-diazirine
and functional introduction into cells. Additionally, the images
obtained from this experiment also provide a good overall measure
of plasmid transfection efficiency in HeLa cells, allowing users to opti-
mize conditions toward maximum percentage of expressing cells.
4.4.3 UV photo-crosslinking and imaging
(a) At this stage, it is valuable to perform a “wash out” experiment to
further confirm proper experimental set up and establish successful
photo-crosslinking conditions.
(b) UV light irradiation can produce large amounts of heat. To avoid
thermal stress on cells, place the 96-well plate in a styrofoam container
lined with ice while irradiating.
(c) Position the UV light source above the cells, and irradiate plate for
5min. We have previously found this time to be sufficient for most
cross-linking applications. However, light sources from different
vendors can vary significantly in light output and efficiency, so it is
advisable to optimize UV irradiation time before proceeding.

ARTICLE IN PRESS
Covalent live-cell labeling of proteins
21
Fig. 8 Representative fluorescence microscopy images of transfected HeLa cells incu-
bated with MG-diazirine. Transfection with the reporter mCer3-FAP plasmid should
result in both mCer3 and MG-diazirine fluorescent signal to confirm successful trans-
fection and expression of the construct, as well as functional MG-diazirine binding
and fluorogenic activity. Cells not exposed to the vector should produce no detectable
background fluorescence.
Fig. 9 Representative fluorescence microscopy images in transfected cells with or
without UV irradiation. Increasing wash time should produce a steady loss of fluores-
cence in cells without UV treatment, while maintenance of signal is indicative of
successful photo-crosslinking of MG-diazirine with FAP. Adapted with permission from
Ayele, T. M., Knutson, S. D., Ellipilli, S., Hwang, H., & Heemstra, J. M. (2019). Fluorogenic pho-
toaffinity labeling of proteins in living cells. Bioconjugate Chemistry, 30(5), 1309–1313.
Copyright (2019) American Chemical Society.

ARTICLE IN PRESS
22
Tewoderos M. Ayele et al.
(d) Proceed to fluorescence microscopy and obtain initial “time 0” images.
At this point, cell samples that have and have not undergone UV irra-
diation should both produce robust MG/FAP signal.
(e) Remove media from wells and replace with 100μL fresh prewarmed
EMEM complete media. Let cells incubate at 37°C for 10min.
(f ) Repeat step (d) and annotate images for “10min.”
(g) Repeat step (e) at 20, 30 and 40min, obtaining images at each time-
point. Images should result in a retained signal in UV treated cells,
but a predictable loss of signal in non-irradiated samples (Fig. 9).
5. Summary
Visualizing proteins inside of living cells is a powerful tool for under-
standing their dynamic biological roles and behavior. Genetic tagging
of proteins with fluorescent protein domains or covalent labeling with
small-molecule fluorophores are useful strategies for imaging and tracking
of proteins by fluorescence microscopy. A promising alternative approach
employs conditionally active fluorogenic dyes coupled with engineered fluor-
ogen activating proteins to achieve significantly higher signal-to-background
ratios and greater ease of use. We recognized that this approach could be fur-
ther extended to enable time-resolved control of labeling through photo-
crosslinking of the fluorogen to the FAP receptor. In this chapter, we outline
the basic techniques for synthesizing a malachite green fluorogen having a
photo-reactive diazirine functional group, tagging proteins of interest with
fluorogen activating proteins, and visualizing proteins inside of live cells using
our covalent labeling approach.
References
Ayele, T. M., Knutson, S. D., Ellipilli, S., Hwang, H., & Heemstra, J. M. (2019).
Fluorogenic photoaffinity labeling of proteins in living cells. Bioconjugate Chemistry,
30(5), 1309–1313.
Babendure, J. R., Adams, S. R., & Tsien, R. Y. (2003). Aptamers switch on fluorescence of
triphenylmethane dyes. Journal of the American Chemical Society, 125(48), 14716–14717.
Brock, R. (2014). The uptake of arginine-rich cell-penetrating peptides: Putting the puzzle
together. Bioconjugate Chemistry, 25(5), 863–868.
Bruchez, M. P. (2015). Dark dyes–bright complexes: Fluorogenic protein labeling. Current
Opinion in Chemical Biology, 27, 18–23.
Chalfie, M., Tu, Y., Euskirchen, G., Ward, W. W., & Prasher, D. C. (1994). Green fluo-
rescent protein as a marker for gene expression. Science, 263(5148), 802–805.
Chen, Y., Clouthier, C. M., Tsao, K., Strmiskova, M., Lachance, H., & Keillor, J. W. (2014).
Coumarin-based fluorogenic probes for no-wash protein labeling. Angewandte Chemie
International Edition, 53(50), 13785–13788.

ARTICLE IN PRESS
Covalent live-cell labeling of proteins
23
Deer, J. R., & Allison, D. S. (2004). High-level expression of proteins in mammalian
cells using transcription regulatory sequences from the Chinese hamster EF-1α gene.
Biotechnology Progress, 20(3), 880–889.
Dubinsky, L., Krom, B. P., & Meijler, M. M. (2012). Diazirine based photoaffinity labeling.
Bioorganic & Medicinal Chemistry, 20(2), 554–570.
Ferna´ndez-Sua´rez, M., Baruah, H., Mart´ınez-Herna´ndez, L., Xie, K. T., Baskin, J. M.,
Bertozzi, C. R., et al. (2007). Redirecting lipoic acid ligase for cell surface protein label-
ing with small-molecule probes. Nature Biotechnology, 25(12), 1483.
Gautier, A., Juillerat, A., Heinis, C., Corr^ea, I. R., Jr., Kindermann, M., Beaufils, F., et al.
(2008). An engineered protein tag for multiprotein labeling in living cells. Chemistry &
Biology, 15(2), 128–136.
Giepmans, B. N., Adams, S. R., Ellisman, M. H., & Tsien, R. Y. (2006). The fluorescent
toolbox for assessing protein location and function. Science, 312(5771), 217–224.
Griffin, B. A., Adams, S. R., & Tsien, R. Y. (1998). Specific covalent labeling of recombinant
protein molecules inside live cells. Science, 281(5374), 269–272.
Lackner, D. H., Carrꢀe, A., Guzzardo, P. M., Banning, C., Mangena, R., Henley, T., et al.
(2015). A generic strategy for CRISPR-Cas9-mediated gene tagging. Nature
Communications, 6, 10237.
Los, G. V., Encell, L. P., McDougall, M. G., Hartzell, D. D., Karassina, N., Zimprich, C.,
et al. (2008). HaloTag: A novel protein labeling technology for cell imaging and protein
analysis. ACS Chemical Biology, 3(6), 373–382.
Marks, K. M., & Nolan, G. P. (2006). Chemical labeling strategies for cell biology. Nature
Methods, 3(8), 591.
Qin, J. Y., Zhang, L., Clift, K. L., Hulur, I., Xiang, A. P., Ren, B.-Z., et al. (2010).
Systematic comparison of constitutive promoters and the doxycycline-inducible pro-
moter. PLoS One, 5(5), e10611.
Reinhardt, U., Lotze, J., Mo€rl, K., Beck-Sickinger, A. G., & Seitz, O. (2015). Rapid cova-
lent fluorescence labeling of membrane proteins on live cells via coiled-coil templated
acyl transfer. Bioconjugate Chemistry, 26(10), 2106–2117.
Rothbard, J. B., Kreider, E., VanDeusen, C. L., Wright, L., Wylie, B. L., & Wender, P. A.
(2002). Arginine-rich molecular transporters for drug delivery: Role of backbone
spacing in cellular uptake. Journal of Medicinal Chemistry, 45(17), 3612–3618.
Shukla, S., & Mathur, M. R. (1995). Low-temperature ultrasonic dyeing of silk. Journal of the
Society of Dyers and Colourists, 111(11), 342–345.
Stephens, D. J., & Allan, V. J. (2003). Light microscopy techniques for live cell imaging.
Science, 300(5616), 82–86.
Stroffekova, K., Proenza, C., & Beam, K. G. (2001). The protein-labeling reagent FLASH-
EDT 2 binds not only to CCXXCC motifs but also non-specifically to endogenous
cysteine-rich proteins. Pflu€gers Archiv, 442(6), 859–866.
Szent-Gyorgyi, C., Schmidt, B. F., Creeger, Y., Fisher, G. W., Zakel, K. L., Adler, S., et al.
(2008). Fluorogen-activating single-chain antibodies for imaging cell surface proteins.
Nature Biotechnology, 26(2), 235.
Telmer, C. A., Verma, R., Teng, H., Andreko, S., Law, L., & Bruchez, M. P. (2015). Rapid,
specific, no-wash, far-red fluorogen activation in subcellular compartments by targeted
fluorogen activating proteins. ACS Chemical Biology, 10(5), 1239–1246.
Toseland, C. P. (2013). Fluorescent labeling and modification of proteins. Journal of Chemical
Biology, 6(3), 85–95.
Xing, W., He, L., Yang, H., Sun, C., Li, D., Yang, X., et al. (2009). Development of a sen-
sitive and group-specific polyclonal antibody-based enzyme-linked immunosorbent
assay (ELISA) for detection of malachite green and leucomalachite green in water and
fish samples. Journal of the Science of Food and Agriculture, 89(13), 2165–2173.
Yerramilli, V. S., & Kim, K. H. (2018). Labeling RNAs in live cells using malachite green
aptamer scaffolds as fluorescent probes. ACS Synthetic Biology, 7(3), 758–766.