A Fluorescent Transport Assay Enables Studying AmpG Permeases Involved in Peptidoglycan Recycling and Antibiotic Resistance

Article abstract of DOI:10.1021/acschembio.6b00552

Inducible AmpC β-lactamases deactivate a broad-spectrum of β-lactam antibiotics and afford antibiotic resistance in many Gram-negative bacteria. The disturbance of peptidoglycan recycling caused by β-lactam antibiotics leads to accumulation of GlcNAc-1,6-anhydroMurNAc-peptides, which are transported by AmpG to the cytoplasm where they are processed into AmpC inducers. AmpG transporters are poorly understood; however, their loss restores susceptibility toward β-lactam antibiotics, highlighting AmpG as a potential target for resistance-attenuating therapeutics. We prepare a GlcNAc-1,6-anhydroMurNAc-fluorophore conjugate and, using live E. coli spheroplasts, quantitatively analyze its transport by AmpG and inhibition of this process by a competing substrate. Further, we use this transport assay to evaluate the function of two AmpG homologues from Pseudomonas aeruginosa and show that P. aeruginosa AmpG (Pa-AmpG) but not AmpP (Pa-AmpP) transports this probe substrate. We corroborate these results by AmpC induction assays with Pa-AmpG and Pa-AmpP. This fluorescent AmpG probe and spheroplast-based transport assay will enable improved understanding of PG recycling and of permeases from the major facilitator superfamily of transport proteins and may aid in identification of AmpG antagonists that combat AmpC-mediated resistance toward β-lactam antibiotics.

Full text of DOI:10.1021/acschembio.6b00552

Subscriber access provided by CORNELL UNIVERSITY LIBRARY
Article
A Fluorescent Transport Assay Enables Studying AmpG Permeases
Involved in Peptidoglycan Recycling and Antibiotic Resistance.
G. Evan Perley-Robertson, Anuj K. Yadav, Judith L. Winogrodzki,
Keith A Stubbs, Brian L. Mark, and David J. Vocadlo
ACS Chem. Biol., Just Accepted Manuscript • DOI: 10.1021/acschembio.6b00552 • Publication Date (Web): 21 Jul 2016
Downloaded from http://pubs.acs.org on July 27, 2016
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 free 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 accessible to all
readers and 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.
ACS Chemical Biology 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 36
ACS Chemical Biology
1
2
3
4
5
6
7
8
9
A Fluorescent Transport Assay Enables Studying AmpG Permeases
Involved in Peptidoglycan Recycling and Antibiotic Resistance.
†,#
†,#
‡
G. Evan PerleyꢀRobertson, Anuj K. Yadav, Judith L. Winogrodzki, Keith A.
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
§
‡
†,
Stubbs, Brian L. Mark, and David J. Vocadlo *
†
5
Department of Chemistry, Simon Fraser University, Burnaby, British Columbia, V5A
1
S6, Canada
‡
Department of Microbiology, University of Manitoba, Winnipeg, Manitoba, R3T 2N2,
Canada
§
School of Chemistry and Biochemistry, University of Western Australia, Crawley,
1
0
Western Australia 6009, Australia
ABSTRACT
Inducible AmpC βꢀlactamases deactivate a broadꢀspectrum of βꢀlactam
antibiotics and afford antibiotic resistance in many Gramꢀnegative bacteria. The
disturbance of peptidoglycan recycling caused by βꢀlactam antibiotics leads to
accumulation of GlcNAcꢀ1,6ꢀanhydroMurNAcꢀpeptides, which are transported by AmpG
to the cytoplasm where they are processed into AmpC inducers. AmpG transporters are
poorly understood, however, their loss restores susceptibility toward βꢀlactam
antibiotics, highlighting AmpG as a potential target for resistanceꢀattenuating
therapeutics. We prepare a GlcNAcꢀ1,6ꢀanhydroMurNAcꢀfluorophore conjugate and,
using live E. coli spheroplasts, quantitatively analyze its transport by AmpG and
1
5
2
0
1
ACS Paragon Plus Environment

ACS Chemical Biology
Page 2 of 36
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
inhibition of this process by a competing substrate. Further, we use this transport assay
to evaluate the function of two AmpG homologues from Pseudomonas aeruginosa and
show that P. aeruginosa AmpG (PaꢀAmpG) but not AmpP (PaꢀAmpP) transports this
probe substrate. We corroborate these results by AmpC induction assays with Paꢀ
AmpG and PaꢀAmpP. This fluorescent AmpG probe and spheroplastꢀbased transport
assay will enable improved understanding of PG recycling and of permeases from the
major facilitator superfamily of transport proteins, and may aid in identification of AmpG
antagonists that combat AmpCꢀmediated resistance toward βꢀlactam antibiotics.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
5
30
INTRODUCTION
Antibiotic drug resistance is a growing problem of major societal concern.(1)
Resistance to βꢀlactams, which constitute more than 50% of the antibiotics in clinical
use, is particularly disconcerting.(2) βꢀlactams block biosynthesis of peptidoglycan (PG),
a critical structural component of the bacterial cell wall that confers cellular rigidity and
prevents osmotic lysis.(3) Due in large part to overreliance on and misuse of βꢀlactams,
clinically relevant bacteria are acquiring a variety of βꢀlactamases – enzymes that
hydrolytically deactivate βꢀlactams.(4) The chromosomallyꢀencoded group 1, class C, βꢀ
lactamase AmpC(5, 6) which confers resistance against penicillins and
cephalosporins,(7) is particularly problematic and has been identified in clinically
relevant strains of Pseudomonas aeruginosa and Enterobacteriaceae.(6, 8) Indeed,
AmpCꢀbearing P. aeruginosa causes lifeꢀthreatening nosocomial infections in the
immunocompromised, as well as chronic respiratory infections in patients with cystic
3
5
4
0
2
ACS Paragon Plus Environment

Page 3 of 36
ACS Chemical Biology
1
2
3
4
5
fibrosis.(9, 10) Although βꢀlactamase inhibitors exist, many only poorly inhibit AmpC.(6)
To block AmpCꢀmediated resistance to βꢀlactams, efforts have been made to develop
AmpCꢀtargeting inhibitors.(11, 12) Unfortunately, βꢀlactamases such as AmpC can
develop mutational resistance toward these inhibitors,(7, 13, 14) and so new
approaches to combatting βꢀlactam antibiotic resistance are needed. One promising
strategy for preventing AmpCꢀmediated resistance involves preventing its induction – a
process that is intricately linked to the Gramꢀnegative PG recycling pathway and is
stimulated by βꢀlactams in most AmpCꢀproducing bacteria (Figure 1).
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
5
0
The PG layer is a meshꢀlike heteropolymer of glycan strands composed of
alternating βꢀ1,4ꢀlinked Nꢀacetylglucosamine (GlcNAc) and Nꢀacetylmuramic acid
(MurNAc) residues that are crossꢀlinked by shortꢀstem peptides.(15) PG recycling
5
5
enables cell growth and division while maintaining the structural integrity of the cell
wall.(16) Beginning this process, autoꢀlysins(17) act on the PG layer to generate a pool
of GlcNAcꢀ1,6ꢀanhydroMurNAc disaccharides bearing stem peptides of three to five
amino acids that are linked to the lactyl group of MurNAc (GlcNAcꢀanhMurNAcꢀ
peptides, Figure 1).(16) These species are then transported from the periplasm to the
cytoplasm by the inner membrane permease AmpG,(18) where the βꢀglucosaminidase
NagZ removes GlcNAc to produce anhMurNAcꢀpeptides.(19, 20) The anhMurNAcꢀ
peptides are metabolized further through multiple routes to yield UDPꢀMurNAcꢀ
pentapeptide, a precursor of PG biosynthesis.(16, 21) Binding of PG recycling
intermediates to the LysRꢀtype transcriptional regulator AmpR is thought to control
AmpC expression, with the anhMurNAcꢀpeptides(22, 23) acting as AmpC inducers while
UDPꢀMurNAcꢀpentapeptide acts as an AmpC repressor.(22) In the absence of βꢀ
60
6
5
3
ACS Paragon Plus Environment

ACS Chemical Biology
Page 4 of 36
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
lactams, UDPꢀMurNAcꢀpentapeptide binds AmpR, repressing ampC transcription.
However, exposure to βꢀlactams disturbs PG recycling and results in accumulation of
anhMurNAcꢀpeptides in the cytoplasm, one of which binds AmpR and converts it into an
activator of ampC transcription.(22, 23) The resulting AmpC protein deactivates βꢀ
lactams to restore PG homeostasis, which in turn brings AmpC expression back to
basal levels.(22)
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
7
0
Deletion of the gene encoding NagZ attenuates AmpC induction in P.
aeruginosa.(24) Moreover, inhibitors targeting NagZ have indicated that a smallꢀ
75
molecule therapeutic approach to combatting AmpCꢀmediated resistance is
tractable.(25, 26) Compared to NagZ, the highly conserved inner membrane permease
AmpG – a member of the major facilitator superfamily of secondary transporters(27) –
has received little attention even through it plays a critical role in AmpC induction.(28,
2
9) Indeed, genetic inactivation of AmpG in P. aeruginosa reverses AmpCꢀmediated
80
resistance, even in mutated strains that exhibit constitutive hyperproduction of
AmpC.(30) Given its ubiquitous presence in Gramꢀnegative bacteria,(28, 31-33) its
central role in AmpC induction,(28-31) and its unusual sugarꢀconjugate transport
activity,(18) an improved understanding of AmpG would provide new insights into PG
metabolism and could lead to the discovery of therapeutics that attenuate βꢀlactam
antibiotic resistance. However, a convenient method to examine the kinetics of AmpGꢀ
mediated transport does not exist, making the characterization of AmpG transporters
and the identification of AmpG inhibitors problematic.
8
5
4
ACS Paragon Plus Environment

Page 5 of 36
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
Figure 1. Induction of AmpC βꢀlactamase is stimulated by the impact of βꢀlactam
antibiotics on the Gramꢀnegative PG recycling pathway. In the presence of βꢀlactam
antibiotics PG fragments accumulate, leading to a buildup of AmpCꢀinducer molecules
(boxed in green). These bind to transcriptional regulator AmpR and convert it from a
repressed state, bound to AmpCꢀrepressor molecules (boxed in red), into an activator of
ampC transcription.
9
0
9
5
One difficulty in studying AmpG transporters is that recombinant transmembrane
proteins can be difficult to express due to, among other problems, poor folding
efficiency, toxicity at elevated expression levels, and a propensity for mislocalization in
cells.(34) Secondly, designing assays for transport proteins, which simply translocate
substrates, is more challenging than for enzymes, which form and break chemical
1
00
5
ACS Paragon Plus Environment

ACS Chemical Biology
Page 6 of 36
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
bonds. Assays of transport proteins require substrate capture vessels that allow
transported substrate to be retained for quantification while extravesicular substrate is
washed away.(35, 36) Such assays generally use isotopically labeled substrates to
monitor transport, though potentially more convenient is to use fluorescently labeled
substrates. Problematic, however, is that the structural perturbations associated with
fluorophore installation may not be tolerated by the target transporter. Indeed, the
difficulty of studying transport proteins using fluorescently labeled substrates is
evidenced by the fact that only one fluorescenceꢀbased assay of a sugar transporter
has been described.(37) However the probe, 2ꢀ(Nꢀ(7ꢀnitrobenzꢀ2ꢀoxaꢀ1,3ꢀdiazolꢀ4ꢀ
yl)amino)ꢀ2ꢀdeoxyglucose (NBDꢀGlc), has proven invaluable for monitoring glucose
transport in bacteria(37) and eukaryotic cells.(38)
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
105
1
10
Here we describe a convenient and quantitative assay of AmpG, enabled by the
synthesis of a fluorescent analogue of the natural substrate and the use of E. coli
spheroplasts as substrate capture vessels. We show that the assay can be used to
measure the kinetics of AmpGꢀmediated transport and to identify compounds that inhibit
this process. Moreover, we demonstrate the utility of the assay by examining the
transport capabilities of two P. aeruginosa (Pa) proteins, PaꢀAmpG and PaꢀAmpP; and
establish that PaꢀAmpG, but not PaꢀAmpP, transports PG fragments containing
1
15
–
GlcNAcꢀanhMurNAc. We corroborate this finding by AmpC induction assays in ampG
120
E. coli complimented with either PaꢀAmpG or PaꢀAmpP.
6
ACS Paragon Plus Environment

Page 7 of 36
ACS Chemical Biology
1
2
3
4
5
6
7
8
9
RESULTS AND DISCUSSION
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
125
Design and Synthesis of Probe 9. Inspired by previous work on glucose
transporters,(37) we envisioned that a fluorescently labeled AmpG substrate could be
used to examine the kinetics of AmpGꢀmediated transport. To design such a substrate,
we considered qualitative studies showing that radiolabeled GlcNAcꢀanhMurNAcꢀ
peptides, obtained through partial purification of bacterial cell wall fragments, do not
require the pendant peptide in order to be transported by AmpG.(18) We therefore
reasoned that replacing the stem peptide with a chromophore, while retaining the core
disaccharide structure, would yield a fluorescent analogue of the GlcNAcꢀanhMurNAcꢀ
peptides without markedly compromising recognition and transport by AmpG
permeases. To limit the steric demands of this change, we opted to use the low
molecular weight fluorophore Alexa Fluor 350 (AF350). Moreover, AF350 exhibits high
photostability, quantum yield, and water solubility, as well as pH independent excitation
and emission spectra, qualities that make it suitable for cellꢀbased assays. With these
considerations in mind, we carried out the synthesis of probe 9 (GlcNAcꢀanhMurNAcꢀ
AF350, Scheme 1).
1
30
1
35
1
40
Using an established route with minor modifications, we synthesized the
disaccharide GlcNAcꢀanhMurNAc through a convergent approach beginning with
commercially available
Dꢀglucal 1 and 1,3,4,6ꢀtetraꢀOꢀacetylꢀ2ꢀaminoꢀ2ꢀdeoxyꢀβꢀ
Dꢀ
glucopyranose hydrochloride 5.
Dꢀglucal 1 was first converted to the 1,6ꢀanhydrosugar,
and an azide was installed at the 2ꢀposition to obtain azidosugar 2.(39) Protecting group
7
ACS Paragon Plus Environment

ACS Chemical Biology
Page 8 of 36
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
45
manipulation allowed selective benzylation in three steps to generate azidosugar 3.(40)
Subsequent installation of a methyl lactate substituent at the 3ꢀposition via direct
displacement to effect stereochemical inversion, followed by debenzylation, acetylation
of the resultant alcohol, and conversion of the azide to an acetamide, furnished glycosyl
acceptor 4.(41) The glycosyl bromide donor 6 was obtained in two steps by first
protecting
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
50
Scheme 1. Synthesis of GlcNAcꢀanhMurNAcꢀAF350 (9). (a) (Bu Sn) O, 3 Å MS,
3
2
MeCN, reflux, 3 hr; then I , RT, 1.5 hr, 84%; (b) NaN , DMF:H O (9:1), 120 °C, 3 hr,
2
3
2
1
55
81%; (c) TBSCl, imidazole, DMF, RT, 82%; (d) BnBr, BaO, Ba(OH) ꢁ8H O, DMF, RT,
2 2
8
ACS Paragon Plus Environment

Page 9 of 36
ACS Chemical Biology
1
2
3
4
5
6
7
8
9
2
1
.5 hr, 90%; (e) TBAF, THF, 90%; (f) (S)ꢀMeCH(OTf)CO Me, NaH, DCM, 70%; (g) 1)
2
0% Pd/C, H ; 2) Ac O, Py; 3) 10% Pd/C, H , pressure, 80%; (h) 1) Phthalic anhydride,
2
2
2
TEA; 2) Ac O, Py, 68%; (i) 33% HBr in CH COOH; 85% (j) AgOTf, 4 Å MS, DCM, –40
2
3
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
to 0 °C, 85%; (k) 1) N H , EtOH, reflux; 2) Ac O, Py; 3) K CO , MeOH; 75%; (l)
2
4
2
2
3
160
LiOHꢁH O, THF:Dioxane:H O (4:2:1), 99%; (m) NH (CH ) NHCbz, HBTU, DIPEA, DMF,
2 2 2 2 2
5
0%; (n) 1) 10% Pd/C, H , MeOH; 2) AF350, DIPEA, DMF, 40%.
2
the amine of 1,3,4,6ꢀtetraꢀOꢀacetylꢀ2ꢀaminoꢀ2ꢀdeoxyꢀβꢀDꢀglucopyranose hydrochloride 5
as a phthalimide and then treating the product with 33% hydrobromic acid in acetic
acid.(42) Coupling of glycosyl acceptor 4 and glycosyl donor 6 using KoenigsꢀKnorr
conditions provided the protected disaccharide which, upon removal of the phthalimide
and subsequent acetylation of the resultant amine, followed by hydrolysis of the methyl
ester, furnished the disaccharide GlcNAcꢀanhMurNAc 7.(41) To facilitate appending a
suitable fluorophore and to ensure the fluorophore did not abut the disaccharide, we
installed a spacer moiety on disaccharide 7 using benzyl 2ꢀaminoethylcarbamate to
yield the Cbzꢀprotected aminoethyl amide 8. Finally, we removed the carboxybenzyl
group using catalytic Pd/C under hydrogen and directly coupled the resultant crude
amine with the fluorophore (AF350) to yield the desired GlcNAcꢀanhMurNAcꢀAF350
conjugate (probe 9).
165
170
1
75
Transport Assay Development. With probe 9 in hand, we developed a
convenient assay to quantify AmpGꢀmediated transport. Transport assays using whole
cells have been developed for different classes of compounds, including but not limited
to sugars,(37) nucleotides,(43) and amino acids.(44) In these cases, the natural
9
ACS Paragon Plus Environment

ACS Chemical Biology
Page 10 of 36
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
substrate is scavenged from the extracellular environment. For AmpG, however, the
substrates are generated within the periplasmic space by action of endogenous autoꢀ
lysins,(17) making it unlikely that probe 9 can be transported directly from the media into
intact bacterial cells. Indeed, we observed that wildꢀtype E. coli BW25113 (EcBW)(45)
were unable to take up probe 9 even after several hours (Supplementary Figure 1). In
previous work, a freezeꢀthaw protocol had been used to expose E. coli AmpG to
biosyntheticallyꢀderived radiolabeled substrates. This approach enabled qualitative
insights into AmpG specificity, but gave low signalꢀtoꢀnoise and high variability.(18) To
develop a more convenient and quantitative method of studying AmpG, we turned to an
approach used in electrophysiology studies of ion channels,(46) which is to generate
spheroplasts – bacteria that have had their outer membrane and cell wall largely
removed but which maintain an intact inner membrane.
1
1
1
80
85
90
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
We prepared spheroplasts from EcBW, and from an isogenic strain lacking
–
AmpG (ampG , EcBW∆ampG). This preparation involved elongation of the cells by
exposure to cephalexin, treatment with lysozyme to digest the cell wall and cause the
elongated cells to balloon outward, and washing to remove enzymes released from
ruptured cells. Spheroplasts can also be prepared without the use of cephalexin,(47)
which simply leads to smaller spheroplasts. To verify that the spheroplasts generated
from the EcBW and EcBW∆ampG strains were comparably viable we used NBDꢀ
Glc.(37) After incubation, excess NBDꢀGlc was removed by repeated cycles of dilution,
gentle centrifugation, and careful aspiration of the supernatant. Given that the absence
of a cell wall makes the spheroplasts more delicate than whole bacterial cells, low
centrifugal and shear forces were used for pelleting and aspiration, respectively, to
195
2
00
1
0
ACS Paragon Plus Environment

Page 11 of 36
ACS Chemical Biology
1
2
3
4
5
6
7
8
9
minimize the likelihood that the spheroplasts would rupture. Three cycles of washing
was sufficient to eliminate greater than 99.5% of extravesicular NBDꢀGlc, putting the
supernatant fluorescence from the last wash on par with the background fluorescence
of the diluting solution (Supplementary Figure 2). Quantitation of spheroplast
fluorescence after incubation with NBDꢀGlc indicated a signal tenꢀfold higher than
background for both the EcBW and EcBW∆ampG strains (Figure 2). Using fluorescence
microscopy, we observed that for both strains NBDꢀGlc was localized within the
spheroplasts, excluding the possibility that fluorescence originated from debrisꢀ or
membraneꢀassociated NBDꢀGlc (Figure 2). We noted that some of the elongated E. coli,
which make up a portion the postꢀspheroplasting population, also exhibited significant
fluorescence. This is not unexpected since NBDꢀGlc can pass through the cell wall and
outer membrane.(37) A fraction of the spheroplasts and the elongated E. coli exhibited
no fluorescence. Given that that NBDꢀGlc is used as a measure of cell viability, these
cells likely ruptured during the spheroplasting process. Cumulatively, these
fluorescence data demonstrate the comparable viability of spheroplasts prepared from
the EcBW and EcBW∆ampG strains, and support their use as substrate capture vessels
to study transport reactions.
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
2
2
2
05
10
15
We next examined AmpGꢀmediated transport of probe 9 using EcBW and
EcBW∆ampG spheroplasts. As with NBDꢀGlc, we found that three washes was
sufficient to remove the vast majority of extravesicular probe 9 (Supplementary Figure
2
20
2
). We saw that only EcBW spheroplasts became fluorescent on exposure to probe 9
Figure 2). No such increase in fluorescence was seen for EcBW∆ampG spheroplasts,
indicating that probe 9 uptake is AmpGꢀmediated and that nonspecific uptake of
(
1
1
ACS Paragon Plus Environment

ACS Chemical Biology
Page 12 of 36
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
2
2
2
25
30
35
metabolized probe 9, if any, is negligible. Parallel examination of NBDꢀGlc uptake
confirmed comparable viability of the spheroplast preparations. Examination of the postꢀ
assay spheroplasts by fluorescence microscopy revealed probe 9 localized within the
EcBW spheroplasts, whereas no fluorescence was observed within the EcBW∆ampG
spheroplasts (Figure 2). Notably, image acquisition was facile for probe 9 in comparison
to NBDꢀGlc, as the high photostability of the AF350 fluorophore prevented
photobleaching during imaging, underscoring the utility of probe 9 in quantitative
fluorescence microscopy. We also note that the proportion of viable spheroplasts is
relatively consistent since the signal stemming from the transport assay varies by no
more than 20% among batches and is always at least 10ꢀfold higher than background.
Interestingly, some of the elongated EcBW cells were also capable of internalizing
probe 9, suggesting that during spheroplasting their cell walls were made permeable
while still retaining sufficient structural integrity to maintain the native rodꢀshape of the
bacterium. Together, these quantitative fluorescence measurements and fluorescence
images indicate that internalization of probe 9 is AmpGꢀmediated.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
40
1
2
ACS Paragon Plus Environment

Page 13 of 36
ACS Chemical Biology
1
2
3
4
5
6
7
8
9
–
Figure 2. Wildꢀtype (EcBW) spheroplasts, but not isogenic ampG (EcBW∆ampG)
spheroplasts, are able to internalize probe 9. A: Fluorescence of EcBW and
EcBW∆ampG spheroplasts after incubation in the presence (+) or absence (ꢀ) of either
NBDꢀGlc (430 ꢂM) or probe 9 (50 ꢂM). The spheroplasts uptake NBDꢀGlc, indicating
they are comparably viable. Error bars reflect the standard error of the mean.
Excitation/emission wavelengths were 350/450 nm for probe 9 and 475/535 nm for
NBDꢀGlc. B: DIC (top), fluorescence (middle), and overlay (bottom) images of EcBW
(right) and EcBW∆ampG (left) spheroplasts after incubation with NBDꢀGlc; or probe 9
(C). The spheroplast preparation was diluted 10ꢀfold prior to imaging. Scale bars
indicate 20 ꢂm. Excitation wavelengths of 365 nm and 470 nm were used for probe 9
and NBDꢀGlc, respectively.
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
2
45
2
50
Kinetic Measurements of Transport. We next assessed the kinetics governing
the AmpGꢀmediated transport of probe 9. Since MichaelisꢀMenten kinetics have been
successfully used to describe membrane transport proteins,(48, 49) we applied this
model to AmpGꢀmediated transport. Given the high concentration of probe 9 outside the
spheroplasts, we made the assumption that loss of probe 9 from the spheroplasts
during the uptake phase of the assay was negligible. To minimize potential export of
probe 9 during washing steps (though no mechanisms for active export of PG recycling
fragments are known), manipulations were performed at 4 °C and samples were kept on
ice.(36) Given the polarity of probe 9 and the fact that it bears a negatively charged
sulfonic acid moiety, we expected that passive diffusion of probe 9 out of the
spheroplasts would be negligible due to both electrostatic repulsion and limited solubility
2
55
2
60
1
3
ACS Paragon Plus Environment

ACS Chemical Biology
Page 14 of 36
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
2
2
2
65
70
75
within the phospholipid bilayer. To establish the Michaelis constant (K ) for AmpGꢀ
M
mediated transport of probe 9, we also needed to ensure that the assay does not
proceed beyond steadyꢀstate conditions, thus allowing initial rates of internalization to
be determined. We therefore evaluated the timeꢀdependent uptake of probe 9 using 50
ꢂM of the probe and EcBW spheroplasts. We found that the rate of internalization
remained relatively constant over the first 90 minutes of the assay (Supplementary
Figure 3). We opted to use a 1ꢀhour incubation for the kinetic experiments, as this time
point is within the linear region for uptake of probe 9 while being long enough to
minimize errors associated with small time variations in sample handling. As is required
for accurate MichaelisꢀMenten kinetic analysis, we verified that the substrate
concentration does not change appreciably during the course of the assay by
comparing the fluorescence of the spheroplasts after one hour of uptake (~1200 RFU,
Figure 2) to that of the first supernatant (~27000 RFU, Supplementary Figure 2). In this
way we estimated that only 4% of probe 9 was internalized. Moreover, we noted that the
amount of probe 9 within the supernatant does not change appreciably with time, and
thus the amount internalized is negligible when compared to the total concentration in
the assay (Supplementary Figure 3). Using these conditions we measured initial rates of
probe 9 internalization by EcBW spheroplasts, and fit the data to the MichaelisꢀMenten
equation to obtain an apparent KM (KM,app) of 340 ± 50 ꢂM (Figure 3). We also
determined the KM,app from a LineweaverꢀBurk plot, and found a similar value
(Supplementary Figure 3).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
280
2
85
Having established that the uptake assay can quantify the kinetics of AmpGꢀ
mediated transport of probe 9, we next tested whether it could be applied to assess the
1
4
ACS Paragon Plus Environment

Page 15 of 36
ACS Chemical Biology
1
2
3
4
5
6
7
8
binding of competitive ligands. Previous work demonstrated that the disaccharide
GlcNAcꢀanhMurNAc is transported by AmpG whereas the monosaccharide anhMurNAc
is not.(18) Thus, we reasoned that GlcNAcꢀanhMurNAc should interfere with the
transport of probe 9 by acting as a competitive substrate for AmpG. Conversely, the
monosaccharides GlcNAc and anhMurNAc should have no impact on the ability of
AmpG to transport probe 9. To test this idea, we incubated EcBW spheroplasts with 50
ꢂM of probe 9 for one hour in the presence of varying concentrations of GlcNAcꢀ
anhMurNAc, GlcNAc, or anhMurNAc, and quantified the uptake of probe 9. A doseꢀ
response curve showed a decreasing initial rate of probe 9 internalization with
increasing GlcNAcꢀanhMurNAc concentration, indicating that GlcNAcꢀanhMurNAc does
indeed compete with probe 9 for AmpGꢀmediated transport (Figure 3). As expected, the
rate of probe 9 internalization was unaffected by the presence of either GlcNAc or
anhMurNAc (Supplementary Figure 4). From the doseꢀresponse curve, we obtained an
IC for GlcNAcꢀanhMurNAc of 78 ± 1 ꢂM. Using this IC along with the concentration
2
90
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
295
3
3
3
00
05
10
50
50
of probe 9 used in the assay and the KM,app for probe 9 with respect to AmpG, we
calculated an apparent K (Ki,app) of 70 ± 10 ꢂM using the ChengꢀPrusoff equation.(50)
i
Since the binding of GlcNAcꢀanhMurNAcꢀpeptides to AmpG has been proposed to
depend primarily on the disaccharide moiety, one would expect that both probe 9 and
GlcNAcꢀanhMurNAc should have similar affinities for AmpG. Indeed, this appears to be
the case as the apparent Michaelis constant for the probe 9ꢀAmpG complex (KM,app
=
3
40 ± 50 ꢂM) is similar to the apparent inhibition constant of the GlcNAcꢀanhMurNAcꢀ
AmpG complex (Ki,app = 70 ± 10 ꢂM). The 5ꢀfold difference may be attributable to the
pendant fluorophore of probe 9.
1
5
ACS Paragon Plus Environment

ACS Chemical Biology
Page 16 of 36
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 3. Initial rates of probe 9 uptake by EcBW spheroplasts can be fit to the
MichaelisꢀMenten equation, and the uptake can be inhibited by the competing
substrate GlcNAcꢀanhMurNAc. A: A MichaelisꢀMenten fit of the initial rates of uptake
3
15
for varying probe 9 concentrations, from which an apparent K of 340 ꢂM ± 50 ꢂM was
M
calculated. B: A doseꢀresponse curve indicating inhibition of AmpGꢀmediated uptake of
probe 9 (50 ꢂM) by the competing substrate GlcNAcꢀanhMurNAc, from which an IC of
50
7
8 ± 1 ꢂM was determined.
1
6
ACS Paragon Plus Environment

Page 17 of 36
ACS Chemical Biology
1
2
3
3
20
Transport of Probe 9 by P. aeruginosa AmpG Homologues. Two P.
aeruginosa (Pa) proteins, Pa4393 (PaꢀAmpG), and Pa4218 (PaꢀAmpP) have recently
been shown to have 45% and 41% sequence identity, respectively, to E. coli AmpG.(28)
Moreover, analysis of their topology shows 14 and 10 transmembrane helices for Paꢀ
AmpG and PaꢀAmpP respectively, lending support to their putative roles as membrane
localized sugar transporters.(51) Despite their similarity to E. coli AmpG and their
topological resemblance to transport proteins, PaꢀAmpG and PaꢀAmpP share only 19%
sequence identity, suggesting they may not have redundant functions.(28) Previous
studies agree that PaꢀAmpG contributes to maintaining intrinsic βꢀlactam resistance in
P. aeruginosa, however the role of PaꢀAmpP is less clear, with contradictory reports in
the literature regarding its involvment in AmpC expression.(28, 30, 51) To define the
roles of PaꢀAmpG and PaꢀAmpP in mediating transport of peptidoglycan fragments and
in AmpCꢀmediated resistance to βꢀlactam antibiotics, we performed AmpG transport
assays using probe 9 and spheroplasts transformed with these P. aeruginosa
membrane proteins.
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
25
330
3
35
We cloned PaꢀAmpG and PaꢀAmpP fused to V5 and His tags into the pBAD
6
vector (pBADꢀPaꢀAmpGꢀV5HisꢀKm and pBADꢀPaꢀAmpPꢀV5HisꢀKm) to enable inducible
expression in E. coli. For reasons unknown, EcBW transformed with either of these two
plasmids would not induce AmpC when coꢀtransformed with a plasmidꢀborne ampR-
ampC operon. However, E. coli strain MG1655 (EcMG) accommodates this coꢀ
expression system and allowed us to evaluate the function of PaꢀAmpG and PaꢀAmpP
using both probe 9 and AmpC induction assays within a single E. coli strain (see below).
3
40
–
As a host for the uptake assay with probe 9 we used an ampG strain (EcMG∆ampG).
1
7
ACS Paragon Plus Environment

ACS Chemical Biology
Page 18 of 36
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Both PaꢀAmpG and PaꢀAmpP expressed well in EcMG∆ampG using the pBAD system
(Supplementary Figure 5). Spheroplasting of EcMG∆ampG was carried out similarly to
3
45
the EcBW strains (see Methods). Cellular fractionation of the spheroplasts revealed that
the inner membrane fraction was enriched in PaꢀAmpG and PaꢀAmpP, indicating proper
targeting of the proteins under spheroplasting conditions (Supplementary Figure 5).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Knowing the P. aeruginosa transporters are properly expressed in spheroplasts,
we next carried out the AmpG transport assay as was done for the EcBW strains.
Regardless of whether the EcMG∆ampG spheroplasts were transformed with an empty
vector (pBAD322ꢀV5HisꢀKm), or one expressing PaꢀAmpG or PaꢀAmpP, NBDꢀGlc was
internalized to the same extent indicating comparable viability (Figure 4). However, only
the spheroplasts expressing PaꢀAmpG were able to internalize probe 9; negligible probe
3
3
3
50
55
60
9
internalization was observed for spheroplasts expressing PaꢀAmpP or those
transformed with empty vector (Figure 4). The inability of PaꢀAmpP to transport probe 9
does not stem from poor expression, since it is more highly expressed in spheroplasts
than is PaꢀAmpG (Supplementary Figure 5). These results were corroborated by
fluorescence microscopy, which shows that EcMG∆ampG spheroplasts bearing any of
the three vectors became fluorescent after incubation with NBDꢀGlc (Supplementary
Figure 6), whereas only the spheroplasts expressing PaꢀAmpG were fluorescent after
incubation with probe 9 (Figure 4). Again, analysis of the washing procedure indicated
that a negligible amount of either NBDꢀGlc or probe 9 was found in the supernatant after
three washes (Supplementary Figure 2). These results indicate that only PaꢀAmpG, but
not PaꢀAmpP, can transport probe 9, suggesting that PaꢀAmpG, but not PaꢀAmpP,
1
8
ACS Paragon Plus Environment

Page 19 of 36
ACS Chemical Biology
1
2
3
3
65
codes for a functional transporter of the disaccharide GlcNAcꢀanhMurNAc in P.
aeruginosa.
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
–
Figure 4. Spheroplasts of ampG E. coli (EcMG∆ampG) transformed with Paꢀ
3
3
3
70
75
80
AmpG, but not with PaꢀAmpP or an empty vector, are able to internalize probe 9.
A: Fluorescence of EcMG∆ampG spheroplasts carrying an empty vector, or expressing
V5His tagged PaꢀAmpG or PaꢀAmpP, after incubation in the presence (+) or absence (ꢀ)
of either NBDꢀGlc (430 ꢂM) or probe 9 (50 ꢂM). The spheroplasts uptake NBDꢀGlc,
indicating they are comparably viable. Error bars reflect the standard error of the mean.
Excitation/emission wavelengths were 350/450 nm for probe 9 and 475/535 nm for
NBDꢀGlc. B: DIC (top), fluorescence (middle), and overlay (bottom) images of
EcMG∆ampG spheroplasts transformed with the empty vector (right), or plasmids
allowing the expression of PaꢀAmpG (center), or PaꢀAmpP (left), after incubation with
probe 9. The spheroplast preparation was diluted 10ꢀfold prior to imaging. Scale bars
indicate 20 ꢂm. The excitation wavelength used was 365 nm.
1
9
ACS Paragon Plus Environment

ACS Chemical Biology
Page 20 of 36
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
We corroborated the results of the uptake assay using an AmpC induction
experiment that examined the ability of untagged PaꢀAmpG (pBADꢀPaꢀAmpGꢀKm) and
PaꢀAmpP (pBADꢀPaꢀAmpGꢀKm) to complement EcMG∆ampG. EcMG and
EcMG∆ampG were transformed with a plasmid harboring the ampR-ampC operon from
C. freundii to enable monitoring of AmpC induction.(25) As controls, EcMG and
EcMG∆ampG were coꢀtransformed with empty vector (pBAD322ꢀKm), and after
exposure to the AmpCꢀinducer
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
85
3
90
2
0
ACS Paragon Plus Environment

Page 21 of 36
ACS Chemical Biology
1
2
3
4
5
6
7
8
9
–
Figure 5. P. aeruginosa PaꢀAmpG but not PaꢀAmpP is able to complement ampG
E. coli (EcMG∆ampG). All E. coli strains were transformed with a plasmid bearing the
C. freundii ampRꢀampC operon. When transformed with empty vector, wildꢀtype E. coli
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
–
(
EcMG, gray), but not an isogenic ampG strain (EcMG∆ampG, white), is capable of
395
expressing AmpC when treated with cefoxitin. A: Transformation of EcMG∆ampG with
untagged PaꢀAmpG (black) rescues AmpC expression in a manner dependent upon the
amount of PaꢀAmpG expressed. B: Transformation of EcMG∆ampG with untagged Paꢀ
AmpP (black) is unable to rescue AmpC expression regardless of the amount of Paꢀ
AmpP expressed. AmpC activity given in nanomoles of nitrocefin hydrolyzed per minute
per mg protein. Error bars represent standard deviations.
4
00
cefoxitin EcMG exhibited dramatically higher AmpC activity than did EcMG∆ampG
(
Figure 5). This observation confirmed that E. coli AmpG is required for AmpC induction.
Coꢀtransformation of EcMG∆ampG (which lacks E. coli AmpG) with the plasmid
expressing PaꢀAmpG resulted in induction of AmpC in a manner dependent on the
amount of PaꢀAmpG expressed – heightened expression of PaꢀAmpG from the pBAD
promotor lead to heightened levels of AmpC activity (Figure 5). Conversely, coꢀ
transformation of EcMG∆ampG with the plasmid expressing PaꢀAmpP resulted in no
AmpC induction, regardless of the amount of PaꢀAmpP expressed (Figure 5). These
data indicate that PaꢀAmpP is not involved in AmpC induction. Coupled with the results
of the transport assay (Figure 4), we suggest that this is due to PaꢀAmpP being unable
to transport peptidoglycan fragments akin to probe 9 that are the precursors to AmpCꢀ
inducer molecules.
405
4
10
2
1
ACS Paragon Plus Environment

ACS Chemical Biology
Page 22 of 36
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Conclusions. Chemical synthesis of a GlcNAcꢀ1,6ꢀanhydroMurNAcꢀfluorophore
conjugate (probe 9), in conjunction with adaptation of methods used in
electrophysiology to generate live E. coli spheroplasts,(46) enabled us to develop a
quantitative fluorescenceꢀbased transport assay of AmpG proteins. AmpG transporters
are ubiquitous among Gramꢀnegative bacteria(28, 29, 31-33) and play a critical role in
PG metabolism and in providing resistance against βꢀlactam antibiotics.(29, 52)
Surprisingly, this assay, to our knowledge, is only the second fluorescent transport
assay developed using a carbohydrateꢀbased substrate,(37) and the first for a
periplasmic transporter. Importantly the assay is convenient, allowing facile kinetic
analysis of AmpG without the complex preparations(49) or radiation safety concerns(48)
associated with quantitative transport assays of other permeases.
4
15
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
4
20
4
25
We expect that this assay can be used generally, in conjunction with engineered
bacterial strains, to assess the activity of AmpG homologues from other bacteria. We
demonstrated this possibility by showing that recombinant P. aeruginosa AmpG (Paꢀ
AmpG), but not PaꢀAmpP, transports GlcNAcꢀanhMurNAcꢀpeptides that lead to
induction of AmpC βꢀlactamase, a finding supported by complementation experiments
–
430
using ampG E. coli harboring an ampRꢀAmpC operon. These data clarify previous
conflicting observations surrounding the activity of PaꢀAmpP, which were based solely
on bacteriological studies.(28, 51)
We further expect that adaptation of spheroplasting procedures, as we have
done for different E. coli strains, will enable examination of AmpG transport directly in
other bacterial species, rather than using recombinant methods. This will afford a more
accurate understanding of AmpG function and its role in PG recycling, as endogenous
4
35
2
2
ACS Paragon Plus Environment

Page 23 of 36
ACS Chemical Biology
1
2
3
4
5
6
7
8
9
molecular effectors will remain in place. Of interest would be examining PG recycling
within Gramꢀpositive bacteria, which appear to lack AmpG homologues but for which
PG recycling plays a crucial role in the bacterial life cycle and has recently been linked
to βꢀlactam antibiotic resistance.(52-55) In particular, with strainꢀspecific optimization to
the spheroplasting procedures, this assay could be used to determine whether Gramꢀ
positive organisms transport GlcNAcꢀanhMurNAc, and could thereby uncover Gramꢀ
positive proteins that are functionally related to AmpG. Moreover, this assay lays out a
blueprint to interrogate other periplasmic transporters using a spheroplastꢀbased
approach so long as a fluorescently labelled substrate is synthetically tractable and
tolerated by the target.
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
4
40
4
45
Given that this transport assay enables monitoring of AmpG inhibition, we
anticipate that with minor adjustments such as reducing the sample handling steps by
employing a rapidꢀfiltration method,(48) the assay could prove useful in high throughput
screening to identify compounds that could serve as AmpG inhibitors. Moreover,
improved knowledge of AmpGꢀsubstrate interactions could be accessed through this
assay, and would aid in the rational design of such inhibitors. Indeed, analogues of
GlcNAcꢀanhMurNAc may themselves prove to be potent inhibitors of AmpG
transporters. Accordingly, this spheroplastꢀbased transport assay should prove useful in
understanding fundamental aspects regarding the regulation of this important class of
inner membrane transporters, and also aid in the identification and characterization of
compounds to combat AmpCꢀmediated βꢀlactam antibiotic resistance within a range of
problematic bacteria including Pseudomonas aeruginosa.
450
4
55
2
3
ACS Paragon Plus Environment

ACS Chemical Biology
Page 24 of 36
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
4
60 METHODS
Synthetic Procedures for Key Chemical Compounds. Compound 8. To a
mixture of 7 (40 mg, 0.0836 mmol) and benzyl 2ꢀaminoethylcarbamate (19.5 mg, 0.100
mmol) in DMF (2 mL) was added HBTU (47.6 mg, 0.126 mmol) and DIPEA (43.7 ꢂL,
0.251 mmol). This mixture was stirred overnight at room temperature, concentrated by
rotary evaporation, and purified by silicaꢀgel column chromatography (6:3:1
EtOAc:MeOH:H2O) to give 8 (27 mg, 0.0418 mmol, 50%). See the supporting
information for detailed characterization of the compound.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
4
65
GlcNAc-anhMurNAc-AF350 9. To a solution of 8 (12 mg, 0.018 mmol) in MeOH
(5 mL) was added 10% Pd/C (10 mol%), and H2 was bubbled into the reaction at room
470
temperature. After 4 hr, the reaction mixture was filtered and the filtrate was
concentrated by rotary evaporation. To this residue (10 mg, 0.019 mmol) in anhydrous
DMF (2 mL) at 0 °C was added AF350 (5 mg, 0.012 mmol) and DIPEA (3.2 ꢂL, 0.018
mmol). The solution was allowed to slowly warm to room temperature and was stirred
overnight, after which concentration by rotary evaporation and purification by silicaꢀgel
4
75
column chromatography (7.4:1.6:1 EtOAc:MeOH:H O) gave 9 (5.9 mg, 0.00732 mmol,
2
40%). See the supporting information for detailed characterization of the compound.
Preparation of Spheroplasts. The protocol was adapted from a previously
published procedure(46). An alternative procedure which does not use antibiotic can
also be used.(47) For the BW25113 strains, cultures of EcBW or EcBWꢃampG were
prepared in lysogeny broth (LB) and grown aerobically overnight at 37 °C. The next
4
80
o
day, the cells were inoculated, 1/100, into 400 mL of LB and allowed to grow at 37 C,
2
4
ACS Paragon Plus Environment

Page 25 of 36
ACS Chemical Biology
1
2
3
4
5
6
7
8
9
until the OD600 reached ~ 0.5–0.6. To 1 L of LB, preꢀheated to 42 °C, was added: 120
ꢀ1
ꢀ1
mL of OD600 0.5–0.6 culture, and 7.2 mL of 2 mgꢁmL (EcBW) or 1 mgꢁmL
(EcBW∆ampG) cephalexin. The cells were further incubated at 42 °C until they had
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
4
4
4
85
90
95
elongated to 25 – 50 ꢂm as determined by differential interference contrast (DIC)
microscopy using an Axioskop 2 Plus microscope. The cultures were then centrifuged
at 3000 rcf at 4 °C for 10 min and the resulting pellets resuspended in 100 mL of 0.8 M
sucrose. For spheroplasting, the following reagents were added in the indicated order:
ꢀ1
6 mL of 0.2 M Tris pH 7.2, 4.8 mL of 1 mgꢁmL lysozyme, 2 mL deionized water, and 6
mL of 0.025 M EDTA. Once spheroplast formation was complete, as monitored by DIC
microscopy, 40 mL of iceꢀcold stop solution (87.5% 0.8 M sucrose, 9.5% deionized
water, 2% 1 M MgCl , 1% 1 M Tris pH 7.2 (v/v)) was added, followed by 280 mL of
2
dilution solution (98% 0.8 M sucrose, 1% 1 M MgCl , 1% 1 M Tris pH 7.2 (v/v)). The
2
spheroplast suspensions were then centrifuged at 2000 rcf at 4 °C for 25 min, washed
once with 5 mL of dilution solution, and then resuspended to homogeneity in dilution
ꢀ1
solution to a concentration of 0.4–0.5 gꢁmL based on the mass of the pellet (containing
both active and inactive spheroplasts). In the case of MG1655, overnight cultures of
EcMGꢃampG transformed with either pBAD322ꢀV5HisꢀKm, pBAD322ꢀPaꢀAmpGꢀV5Hisꢀ
ꢀ1
Km, or pBAD322ꢀPaꢀAmpPꢀV5HisꢀKm, were carried out in the presence of 35 ꢂgꢁmL
500
kanamycin. The next day, the cells were inoculated, 1/50, into 500 mL LB and allowed
to grow at 28 °C until the OD600 reached ~ 0.6–0.7. To induce expression, the cells
were grown an additional 4 hr at 28 °C in the presence of 0.2% (w/v) Lꢀarabinose.
During the last hour of induction cephalexin was added to a final concentration of 6
ꢀ1
ꢂgꢁmL , after which the cells were harvested by centrifugation and spheroplasts were
2
5
ACS Paragon Plus Environment

ACS Chemical Biology
Page 26 of 36
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
5
05
prepared essentially as described above. The spheroplast stocks were stored at 4 °C
and used within 48 hours.
Uptake Assay for Probe 9 and NBDꢀGlc. Experiments were carried out in
duplicate using EcBW and EcBW∆ampG spheroplasts, or with spheroplasts derived
from EcMG∆ampG transformed with the indicated expression plasmid. Briefly, 50 ꢂL of
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ꢀ1
ꢀ1
510
0.4 gꢁmL (EcBW) or 100 ꢂL of 0.4 gꢁmL (EcMG) spheroplasts were added to
microcentrifuge tubes and the volume was brought up to 250 ꢂL with dilution solution.
The assay was then initiated by the addition of 100 ꢂL of 175 ꢂM probe 9, or of 1.5 mM
NBDꢀGlc. Blanks were prepared in which dilution solution was added in place of the
fluorescent compound. The tubes were mixed by inversion several times, then
incubated for 1 hr (EcBW) or 2 hr (EcMG) at 25 °C in the dark using a digital heatblock.
Final reaction conditions were: 350 ꢂL; 50 ꢂM probe 9 or 430 ꢂM NBDꢀGlc, and 0.06
5
15
ꢀ1
ꢀ1
gꢁmL (EcBW) or 0.12 gꢁmL (EcMG) spheroplasts. Postꢀassay spheroplast washing
and fluorescence quantitation was performed as described below.
MichaelisꢀMenten Kinetic Evaluation of the Uptake of Probe 9. The
experiment was carried out in singlicate using spheroplasts derived from EcBW. In
microcentrifuge tubes, a series of probe 9 concentrations was obtained via threeꢀfold
serial dilution, from a 900 ꢂM stock, into a final volume of 100 ꢂL of dilution solution.
5
20
ꢀ1
Then the assay was initiated by the addition of 50 ꢂL of 0.5 gꢁmL spheroplasts. The
tubes were inverted several times to mix and were incubated for 1 hr at 25 °C in the
dark using a digital heatblock. Final reaction conditions were: 150 ꢂL; 600 ꢂM to 0 ꢂM
5
25
ꢀ1
probe 9, and 0.17 gꢁmL spheroplasts. Postꢀassay spheroplast washing and
fluorescence quantitation was performed as described below.
2
6
ACS Paragon Plus Environment

Page 27 of 36
ACS Chemical Biology
1
2
3
4
5
6
7
8
Competitive Inhibition of Probe 9 Uptake by GlcNAcꢀanhMurNAc. The
experiment used EcBW spheroplasts, and was carried out in singlicate with GlcNAc or
anhMurNAc, or in triplicate with GlcNAcꢀanhMurNAc. In microcentrifuge tubes, a range
of concentrations was initially prepared by fourꢀfold dilution of GlcNAc or anhMurNAc, or
threeꢀfold dilution of GlcNAcꢀanhMurNAc, in 100 ꢂL of dilution solution. Then, 50 ꢂL of
5
30
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
00 ꢂM probe 9 in dilution solution was added and the competitive assay was initiated
ꢀ1
by the addition of 50 ꢂL of 0.52 gꢁmL EcBW spheroplasts. Blanks contained dilution
solution in place of probe 9. The tubes were then mixed by inversion and were
incubated for 1 hr at 25 °C in the dark using a digital heatblock. Final reaction conditions
were: 200 ꢂL; 1 mM to 0 mM GlcNAc or anhMurNAc, or 1.5 mM to 0 mM GlcNAcꢀ
535
ꢀ1
anhMurNAc; 50 ꢂM probe 9; 0.13 gꢁmL of spheroplasts. Postꢀassay spheroplast
washing and fluorescence quantitation was performed as described below.
5
5
5
40
45
50
Spheroplast Washing and Fluorescence Quantitation. After incubation of
spheroplasts with labeled substrate, each tube was placed on ice and 1 mL of iceꢀcold
dilution solution was added in the same order in which the experiment in question was
initiated. The tubes were then centrifuged at 2000 rcf at 4 °C for 20 min, and the
supernatants were withdrawn using a syringe equipped with an 18 gauged needle and
stored on ice for later fluorescence testing. The syringe was subsequently washed
twice with water after finishing with one set of tubes (i.e. blank, NBDꢀGlc, or probe 9).
Dilution, centrifugation, and supernatant withdrawal was repeated two additional times
to wash away remaining extracellular fluorogenic substrate for EcBW, or three
additional times for EcMG, and the final washed pellets were vortexed to homogeneity
in 200 ꢂL of dilution solution. To a black 96ꢀmicrowell plate was added 150 ꢂL of each
2
7
ACS Paragon Plus Environment

ACS Chemical Biology
Page 28 of 36
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
spheroplast suspensions and the supernatant. Fluorescence was quantified at 25 °C
with excitation/emission wavelengths of 350/450 nm for probe 9, and 475/535 nm for
NBDꢀGlc. Data was acquired using a Synergy 4 microplate reader, reported in relative
fluorescence units (RFU), and was analyzed using Gen5 and Graphpad software.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
5
5
5
55
60
65
βꢀlactamase Assays. EcMG∆ampG bearing the plasmidꢀborne Citrobacter
freundii ampR-ampC operon(25) transformed with either pBADꢀPaꢀAmpGꢀKm or pBADꢀ
ꢀ1
PaꢀAmpPꢀKm were grown at 37 °C with vigorous aeration in M9CA containing 1 ꢂgꢁmL
ꢀ1
ꢀ1
ꢀ1
thiamine, 50 ꢂgꢁmL uracil, 35 ꢂgꢁmL chloramphenicol, 70 ꢂgꢁmL kanamycin and
0.5% (v/v) glycerol until the OD600 reached approximately 0.4. Lꢀarabinose was then
added at the specified concentration and the cells were grown at 23 °C for 1.5 hr. To
ꢀ1
induce ampR-ampC expression, the cultures were supplemented with 16 ꢂgꢁmL
cefoxitin and were grown for an additional 1.5 hr at 23 °C. Crude sonic lysates were
then prepared as described previously(25, 56) and βꢀlactamase specific activity was
determined using a continuous assay procedure that follows the linear rate of liberation
of 2,4ꢀdinitrophenolate from nitrocefin,(57) as determined by measuring the absorption
at 486 nm. The experimental mixture contained 100 ꢂL of PBS and 1 ꢂL of an
appropriately diluted lysate. The reaction was then initiated by the addition of 100 ꢂM
nitrocefin and monitored at room temperature for 5 min. The βꢀlactamase specific
activity is expressed in nanomoles of nitrocefin hydrolyzed per min per mg protein. Wildꢀ
type EcMG and EcMG∆ampG transformed with “empty” pBAD322ꢀKm(58) served as
positive and negative controls. Protein concentrations of the supernatants were
quantified by the Bradford assay using bovine gamma globulin as the standard.
570
2
8
ACS Paragon Plus Environment

Page 29 of 36
ACS Chemical Biology
1
2
3
4
5
6
ASSOCIATED CONTENT
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
575
Supporting Information
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Supporting Information Available: This material is available free of charge via the
Internet.
Additional materials and methods, including microbiological, genetic, and microscopy
techniques; six figures showing: postꢀassay supernatant fluorescence, Western blots,
fluorescence microscopy, and additional kinetic data; NMR spectra for 8 and 9.
580
AUTHOR INFORMATION
Corresponding Author
*
Eꢀmail: dvocadlo@sfu.ca
5
85
Author Contributions
#
These authors contributed equally.
ACKNOWLEDGEMENTS
This work was supported by Canadian Institutes of Health Research (CIHR) Grant
MOPꢀ97818 and Cystic Fibrosis Canada grants (D.J.V. and B.L.M.). B.L.M. holds a
Manitoba Research Chair. D.J.V. thanks the Canada Research Chairs program for
support as a Tier I CRC in Chemical Glycobiology. K.A.S. thanks the Australian
Research Council.
590
2
9
ACS Paragon Plus Environment