Detecting protein interactions in live cells via complementation of a hydrolysis-deficient β-lactamase

Article abstract of DOI:10.1039/c0cc01998d

Protein-fragment complementation assay using a hydrolysis-deficient β-lactamase provides a robust tool for studying protein-protein interactions.

Full text of DOI:10.1039/c0cc01998d

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Detecting protein interactions in live cells via complementation
of a hydrolysis-deficient b-lactamasewz
Jiangsong Jiang,a Junxiang Zhangb and Shuwei Li*bc
Received 21st June 2010, Accepted 19th July 2010
DOI: 10.1039/c0cc01998d
Protein–fragment complementation assay using a hydrolysis-
deficient b-lactamase provides a robust tool for studying
protein–protein interactions.
demonstrates many advantages, including small size
(B30 kD), no homologs in mammalian cells, high enzymatic
activity, and availability of cell permeable substrates for
fluorescent detection.12 This enzyme catalyzes the degradation
of its substrates, b-lactam compounds, through a two-step
mechanism. First, a b-lactam structure forms a stable covalent
bond with Ser70. A water molecule activated by Glu166 then
attacks this enzyme-substrate intermediate to turn over the
enzyme. A point mutation at this position, such as E166N or
E166A, impairs b-lactamase activity so that the enzyme will be
trapped at acyl-enzyme intermediate due to its inability to
remove the substrate from Ser70.13 Consequentially, reconsti-
tuted Lac-E166N can leave a permanent marker on the
N-terminal fragment of b-lactamase when treated with a
b-lactam substrate, even after both fragments dissociate. If a
substrate containing an affinity tag is employed, the N-terminal
fragment along with its fusion protein can be enriched by
affinity purification and identified directly by mass spectro-
metry (MS), while a fluorescent substrate can make it compa-
tible with fluorescent imaging and FACS screening.
Protein–protein interactions play a critical role in virtually all
biological processes, including cell growth, differentiation, and
communication. To understand interactions among various
proteins, therefore, is a central theme of modern biology
research. Over the past decades, numerous methods, such as
yeast two-hybrid,1 protein-fragment complementation assay
(PCA),2 and tandem affinity purification,3 have been developed
in order to decipher intricate protein interaction networks.
Ideally, protein interactions should be determined in their
local compartments and physiological conditions, to prevent
misidentification of binding events that are biologically irrelevant.
PCA, in which nonfunctional N- and C-terminal fragments of
a reporter protein fused with two interacting proteins can
reconstitute its activity to generate a readout signal for detec-
tion, is one of such approaches that can probe protein inter-
actions in their native environments. A variety of PCA
techniques, each of which differs in the reporter protein and
detection method,2,4 are currently available, offering a versa-
tile set of choices for broad applications. For example, a green
fluorescent protein (GFP) reporter can provide an optical
signal for detecting protein interactions by fluorescent micro-
scopy or fluorescent-activated cell sorting (FACS),5 while a
dihydrofolate reductase (DHFR) based PCA allows binding
proteins to be recognized through a cell-survival selection.6
Nevertheless, although PCA is a convenient tool for identifying
unknown interacting proteins from a large library in bacteria7
or yeast,8 performing the same task in mammalian cells
remains technically challenging as it usually requires virus
transfection of cDNA libraries9 and screening strategies with
single cell resolution.10
To test whether Lac-E166N can be used as a reporter for
PCA applications, we first carried out an in vitro experi-
ment using a pair of known interacting proteins (Fig. 2a).
The N-terminal fragment of b-lactamase (residues 23–196)
carrying E166N mutant was fused with the SH3 domain from
yeast protein Sho1p (residues 302–367) through a flexible
linker (GGGGS)3 to construct NLacN-SH3.14 The C-terminal
fragment of b-lactamase (residues 198–286) was linked with a
SH3 domain binding peptide (QQIVNKPLPPLPVAGSS,
Kd = 800 nM) to form LacC-PPLP. We then incubated these
protein fragments, either alone or in pair, with a fluorescent
penicillin derivative (Flu-AP), and analyzed the reaction
mixtures by SDS-PAGE. NLacN-SH3 became fluorescent
only in the presence of LacC-PPLP, suggesting NLacN-SH3
was covalently modified by Flu-AP at Ser70 and the split
fragments of this catalytically impaired Lac-E166N could
refold into a functional structure upon association.
Herein, we describe a novel PCA based on a hydrolysis-
deficient b-lactamase mutant (Lac-E166N),11 in hope of
providing a simple protocol for the identification of associa-
ting proteins in live cells (Fig. 1). Wild type b-lactamase is a
serine hydrolase that confers penicillin resistance in bacteria.
It has already been widely used as a PCA reporter, which
a Department of Cell Biology and Molecular Genetics,
University of Maryland, College Park, MD 20742, USA
b Institute of Bioscience and Biotechnology Research,
University of Maryland, Rockville, MD 20850, USA.
E-mail: sli@umd.edu
c Department of Chemistry and Biochemistry, University of Maryland,
College Park, MD 20742, USA
w This article is part of the ‘Emerging Investigators’ themed issue for
ChemComm.
Fig. 1 PCA using Lac-E166N. The catalytic mechanism of wild type
b-lactamase and its hydrolysis-deficient mutant E166N is illustrated.
P-prey protein; B-bait protein; T-tag.
z Electronic supplementary information (ESI) available: Experiment
procedures. See DOI: 10.1039/c0cc01998d
ꢀc
This journal is The Royal Society of Chemistry 2011
182 | Chem. Commun., 2011, 47, 182–184

View Article Online
between two b-lactamase fragments induced by dimerization of
leucine zipper domain.
Because CCF2-AM leaves a fluorescent tag on the Ser70 of
NLacN after reaction, this PCA strategy would also allow us
to track the movement of a protein after it interacts with a
partner. To confirm this, we inserted a nuclear localization
signal before the leucine zipper domain so both fragments
(NLS-Zip-NLacN and NLS-Zip-LacC) should travel to nuclei.
When these fragments were expressed in HeLa cells, we found
blue fluorescent NLS-Zip-NLacN was indeed concentrated
in nuclei (Fig. 2c). At this moment, however, we could
not determine whether NLS-Zip-NLacN interacted with
NLS-Zip-LacC in nuclei or moved into nuclei after binding
with NLS-Zip-LacC in cytoplasm.
Compared to wild type b-lactamase based PCA, one
advantage of using hydrolysis-deficient Lac-E166N as a
reporter is that proteins interacting with a bait could be
identified with MS directly (Fig. 3a). To demonstrate this
idea, we synthesized a novel cell permeable prodrug substrate
of b-lactamase, CP-AM, which contains a terminal alkyne that
can react with azide bioorthogonally. This compound was
added into a HEK293T cell culture expressing Zip-NLacN
and Zip-LacC. Once entering cells, CP-AM became active
after its protective AM group was removed by nonspecific
cellular esterases and formed a stable covalent bond with
Ser70 in the Zip-NLacN. Cells were then collected, lysed,
and treated with solid support beads coated with azide.
Fig.
2 Lac-E166N as a PCA reporter in vitro and in vivo.
(a) NLacN-SH3 and LacC-PPLP were incubated with Flu-AP either
alone or in pair and analyzed by SDS-PAGE. Coomassie blue staining
(left); Fluorescent imaging (right). (b) HeLa cells expressing
Zip-NLacN alone (left) or both Zip-NLacN and Zip-LacC (right)
were treated with CCF2-AM. (c) HeLa cells expressing NLS-Zip-
NLacN alone (left) or both NLS-Zip-NLacN and NLS-Zip-LacC
(right) were treated with CCF2-AM.The parentheses indicate both
fragments were expressed by the same plasmid. I-DIC (differential
interference contrast); II-green channel; III-blue channel; IV-overlay
of I, II, and III.
Next, we examined if Lac-E166N fragments can recover its
function inside live cells. In order to compensate low transfec-
tion efficiency in mammalian cells, we constructed one plasmid
that expresses both NLacN and LacC fragments fused to the
C-terminus of a leucine zipper domain adopted from yeast
GCN4 protein (Zip-NLacN and Zip-LacC). This leucine
zipper domain can form a homodimer and is often used as a
model system for testing protein interactions.15 After HeLa
cells were transfected with this plasmid, a cell permeable
b-lactamase substrate, CCF2-AM (AM = acetoxymethyl),
was added to detect the activity of reconstituted Lac-E166N
(Fig. 2b). CCF2-AM is a fluorescence resonance energy
transfer (FRET) based molecule, which emits green light
(518 nm) at its intact form, yet becomes blue fluorescent
(447 nm) upon enzymatic degradation.16 We observed that
about 60% HeLa cells transfected with this plasmid exhibited
blue fluorescence. In contrast, none of cells displayed blue
fluorescence when only Zip-NLacN fragment was expressed,
indicating blue fluorescence resulted from the interaction
Fig. 3 Discovery of protein interactions using MS in Lac-E166N
based PCA. (a) Scheme of how interacting proteins are enriched and
identified. (b) A representative peptide from enriched Zip-NLacN was
identified by LC-MS/MS and Mascot. The sequence of this peptide
and its b-/y- ions are marked. (c) Peptide sequence of Zip-NLacN.
Tryptic peptides identified by LC-MS/MS and Mascot are coloured in
red and Ser70 is coloured in green. The sequence in the box is from
leucine zipper domain.
ꢀc
This journal is The Royal Society of Chemistry 2011
Chem. Commun., 2011, 47, 182–184 | 183

View Article Online
In the presence of Cu(I) catalyst, labeled Zip-NLacN was
immobilized covalently on beads, washed thoroughly, then
digested by trypsin. The resulting peptides were analyzed by
LC-MS/MS and Mascot for protein identification. We found
peptides from both b-lactamase and leucine zipper domain
(Fig. 3b and c) with minimum background from other endo-
genous proteins, suggesting this affinity enrichment protocol
was highly efficient.
6 S. Shibasaki, K. Sakata, J. Ishii, A. Kondo and M. Ueda, Appl.
Microbiol. Biotechnol., 2008, 80, 735–743.
7 J. H. Park, J. H. Back, S. H. Hahm, H. Y. Shim, M. J. Park,
S. I. Ko and Y. S. Han, J. Microbiol. Biotechnol., 2007, 17,
1607–1615.
8 K. Tarassov, V. Messier, C. R. Landry, S. Radinovic,
M. M. Molina, I. Shames, Y. Malitskaya, J. Vogel, H. Bussey
and S. W. Michnick, Science, 2008, 320, 1465–1470; U. Stelzl,
U. Worm, M. Lalowski, C. Haenig, F. H. Brembeck, H. Goehler,
M. Stroedicke, M. Zenkner, A. Schoenherr, S. Koeppen, J. Timm,
S. Mintzlaff, C. Abraham, N. Bock, S. Kietzmann, A. Goedde,
E. Toksoz, A. Droege, S. Krobitsch, B. Korn, W. Birchmeier,
H. Lehrach and E. E. Wanker, Cell, 2005, 122, 957–968; J. F. Rual,
K. Venkatesan, T. Hao, T. Hirozane-Kishikawa, A. Dricot, N. Li,
G. F. Berriz, F. D. Gibbons, M. Dreze, N. Ayivi-Guedehoussou,
N. Klitgord, C. Simon, M. Boxem, S. Milstein, J. Rosenberg,
D. S. Goldberg, L. V. Zhang, S. L. Wong, G. Franklin, S. Li,
J. S. Albala, J. Lim, C. Fraughton, E. Llamosas, S. Cevik, C. Bex,
P. Lamesch, R. S. Sikorski, J. Vandenhaute, H. Y. Zoghbi,
A. Smolyar, S. Bosak, R. Sequerra, L. Doucette-Stamm,
M. E. Cusick, D. E. Hill, F. P. Roth and M. Vidal, Nature,
2005, 437, 1173–1178.
By introducing a single point mutation into b-lactamase, we
were able to establish a new PCA system that can be applied
on either fluorescent imagining or affinity enrichment of
interacting proteins by using different substrates. Although
this method is not as sensitive as wild type b-lactamase based
PCA since the readout signal is not enzymatically amplified, it
would allow protein interactions to be detected at their local
sites, providing valuable information on their biological func-
tions. In addition, this approach would potentially enable us
to discover unknown binding proteins from a large cDNA
library. Finally, because protein interacting events can leave a
permeant tag on the NLacN fragment even after protein
complexes dissociate, it would be of great interest to explore
whether this technique can detect transient and low-affinity
protein–protein interactions.
9 Z. Ding, J. Liang, Y. Lu, Q. Yu, Z. Songyang, S. Y. Lin and
G. B. Mills, Proc. Natl. Acad. Sci. U. S. A., 2006, 103,
15014–15019.
10 E. Kiss-Toth, F. M. Guesdon, D. H. Wyllie, E. E.
Qwarnstrom and S. K. Dower, J. Immunol. Methods, 2000, 239,
125–135.
11 G. Minasov, X. Wang and B. K. Shoichet, J. Am. Chem. Soc.,
2002, 124, 5333–5340.
12 T. Wehrman, B. Kleaveland, J. H. Her, R. F. Balint and
H. M. Blau, Proc. Natl. Acad. Sci. U. S. A., 2002, 99,
3469–3474; A. Galarneau, M. Primeau, L. E. Trudeau and
S. W. Michnick, Nat. Biotechnol., 2002, 20, 619–622.
13 H. Adachi, T. Ohta and H. Matsuzawa, J. Biol. Chem., 1991, 266,
3186–3191.
14 J. A. Marles, S. Dahesh, J. Haynes, B. J. Andrews and
A. R. Davidson, Mol. Cell, 2004, 14, 813–823.
15 P. B. Harbury, T. Zhang, P. S. Kim and T. Alber, Science, 1993,
262, 1401–1407.
Notes and references
1 J. R. Parrish, K. D. Gulyas and R. L. Finley, Jr., Curr. Opin.
Biotechnol., 2006, 17, 387–393; S. Fields and O.-k. Song, Nature,
1989, 340, 245–246.
2 M. Morell, S. Ventura and F. X. Aviles, FEBS Lett., 2009, 583,
1684–1691.
3 T. Burckstummer, K. L. Bennett, A. Preradovic, G. Schutze,
O. Hantschel, G. Superti-Furga and A. Bauch, Nat. Methods,
2006, 3, 1013–1019.
4 I. Remy and S. W. Michnick, Nat. Methods, 2006, 3, 977–979;
P. A. Vidi and V. J. Watts, Mol. Pharmacol., 2009, 75, 733–739.
5 I. Remy and S. W. Michnick, Methods Enzymol., 2004, 32, 381–388.
16 G. Zlokarnik, P. A. Negulescu, T. E. Knapp, L. Mere, N. Burres,
L. Feng, M. Whitney, K. Roemer and R. Y. Tsien, Science, 1998,
279, 84–88.
ꢀc
This journal is The Royal Society of Chemistry 2011
184 | Chem. Commun., 2011, 47, 182–184