Human Metabolome-Derived Cofactors are Required for the Antibacterial Activity of Siderocalin in ...

23 Pages·2016·3.37 MB·English

JBC Papers in Press. Published on October 25, 2016 as Manuscript M116.759183
The latest version is at http://www.jbc.org/cgi/doi/10.1074/jbc.M116.759183
Human metabolites are antibacterial siderocalin cofactors
Human Metabolome-Derived Cofactors are Required for the Antibacterial Activity of Siderocalin in
Urine*

Robin R. Shields-Cutler†‡2, Jan R. Crowley§, Connelly D. Miller†‡, Ann E. Stapleton¥, Weidong
Cui£, and Jeffrey P. Henderson†‡1

†Division of Infectious Diseases, Department of Medicine, Washington University School of Medicine,
St. Louis, MO.
‡Center for Women’s Infectious Diseases Research, Washington University School of Medicine, St.
Louis, MO.
§Department of Internal Medicine, Washington University School of Medicine, St. Louis, MO.
¥Department of Medicine, Division of Allergy and Infectious Diseases, University of Washington, Seattle,
Washington.
£Department of Chemistry, Washington University in St. Louis, St. Louis, MO.

*Running title: Human metabolites are antibacterial siderocalin cofactors
D
o
w
n
To whom correspondence should be addressed: Jeffrey P. Henderson, M.D., Ph.D., Division of Infectious lo
a
Diseases, Washington University School of Medicine, 660 S. Euclid, St. Louis, MO 63110, USA, Tel: de
d
[phone]; Fax: [phone]; E-mail: [email protected] fro
m
h
Keywords: infectious disease; host-pathogen interaction; siderophore; iron; mass spectrometry (MS); ttp
://w
Escherichia coli (E. coli); urinary tract infection; siderocalin; lipocalin 2; NGAL w
w
.jb
c
.o
rg
ABSTRACT that SCN may preferentially combine different b/
y
In human urinary tract infections (UTIs), host ligands to coordinate iron, suggesting that g
u
e
cells release the antimicrobial protein siderocalin availability of specific ligand combinations affects st o
(SCN; also known as lipocalin-2, neutrophil in vivo SCN antibacterial activity. These results n N
gelatinase-associated lipocalin/NGAL, or 24p3) support a mechanistic link between the human ov
e
m
into the urinary tract. By binding to ferric-catechol urinary metabolome and innate immune function. b
e
complexes, SCN can sequester iron, a growth- r 1
9
limiting nutrient for most bacterial pathogens. , 20
1
Recent evidence links the antibacterial activity of INTRODUCTION 8
SCN in human urine to iron sequestration and Iron is an essential micronutrient for nearly all
metabolomic variation between individuals. To microorganisms and its acquisition is a critical
determine whether these metabolomic associations virulence-associated activity for numerous
correspond to functional iron(III)-binding SCN medically-important pathogens (1, 2). Humans
ligands, we devised a biophysical protein binding possess elaborate systems that limit iron’s
screen to identify SCN ligands through direct accessibility to pathogens (1, 3, 4). Among these
analysis of human urine. This screen revealed a innate immune defenses is siderocalin (SCN3; also
series of physiologic unconjugated urinary known as lipocalin-2/Lcn2 or neutrophil gelatinase
catechols that were able to function as SCN associated lipocalin/NGAL), a ~21 kDa soluble
ligands, of which pyrogallol in particular was protein of the lipocalin family notable for a
positively associated with high urinary SCN characteristic eight-stranded antiparallel β-barrel
activity. In a purified, defined culture system, fold that defines a binding site, or calyx (5).
these physiologic SCN ligands were sufficient to Compared to other lipocalin proteins, SCN is
activate SCN antibacterial activity against distinguished by a shallow, positively charged
Escherichia coli. In the presence of multiple SCN calyx with three distinct binding pockets (5–7).
ligands, native mass spectrometry demonstrated
1
Copyright 2016 by The American Society for Biochemistry and Molecular Biology, Inc.
Human metabolites are antibacterial siderocalin cofactors
The SCN calyx does not chelate Fe(III) alone but UTIs and seek to identify physiologically relevant
instead as a complex with catechol ligands, each SCN ligands through direct analysis of human
of which binds the protein through one of three urine. A biophysical screen, intrinsic tryptophan
calyx pockets (5, 8). The resulting SCN-ferric fluorescence quenching, and mass spectrometric
catechol complex shields iron in a form that is analyses identified a series of endogenous SCN
poorly accessible to bacteria. ligands in human urine, of which unconjugated
pyrogallol and caffeic acid were associated with
The prototypical SCN ligand is enterobactin, a high urinary SCN activity. Although defined
secreted microbial iron chelator (siderophore) that media conditions containing a single SCN ligand
chelates ferric ions with extraordinary affinity were sufficient to activate SCN antibacterial
using three interconnected catechols (5, 9–11). In activity, native mass spectrometry revealed that
culture medium, SCN can inhibit growth of SCN can use a specific ligand combination to
urinary pathogenic Escherichia coli that use chelate Fe(III) when multiple ligands are
enterobactin as their sole siderophore, likely available. Together, these studies identify free
through sequestration of Fe(III)-enterobactin (5, catechol metabolites in humans that potentiate
12, 13). However, when these same E. coli strains innate antibacterial immunity by acting as Fe(III)-
are grown in human urine, enterobactin becomes chelating siderocalin cofactors.
D
necessary to resist SCN inhibitory activity, ow
n
suggesting that urinary constituents change the RESULTS lo
a
interaction between SCN and enterobactin (14). Enterobactin and SCN coexist during clinical E. de
d
Because SCN reaches high levels during human coli UTI. To determine whether enterobactin and fro
m
urinary tract infections, these interactions may SCN are simultaneously present in urine during h
influence infection pathogenesis (14–16). human urinary tract infections, we analyzed urine ttp://w
specimens from uncomplicated E. coli cystitis w
w
Human urinary effects on SCN antibacterial patients with stable isotope dilution LC-MS/MS .jb
c
activity exhibit marked individual differences (14). (Fig 1a) and ELISA. Linear enterobactin was .o
rg
In urine from “restrictive” subjects, SCN inhibits detectable in 10 of 15 cystitis urine specimens but b/
y
bacterial growth in an enterobactin-sensitive was undetectable in healthy donors without g
u
manner. In urine from “permissive” subjects, urinary symptoms or bacteriuria (Fig. 1b). SCN es
t o
neither SCN nor bacterial enterobactin production was detected in all cystitis specimens and was n
N
exert prominent effects on E. coli growth. positively associated with enterobactin (Fig. 1c). ov
e
Compared to permissive urine, restrictive urine is To our knowledge this is the first report to directly m
b
e
associated with a higher pH and distinctive demonstrate bacterial enterobactin production r 1
9
aromatic metabolites. These urinary metabolites during a human infection. Moreover, the presence , 2
0
are catechols and other aromatic compounds that of siderocalin in the same specimens raises the 18
have been conjugated to sulfate, most likely in the possibility that enterobactin interactions with this
liver. Although catechols can act as Fe(III)- protein are physiologically relevant.
chelating SCN ligands, their sulfated forms
generally do not, suggesting that these aryl sulfates Differential scanning fluorimetry (DSF) detects
do not directly interact with SCN. Nevertheless, urinary SCN ligands. Previous data indicate that
greater SCN activity at pH above 6.5 is consistent urinary SCN uses urinary aryl cofactors to chelate
with pH optima for ferric catechol complex iron in a stable complex (8, 14, 18), suggesting
formation and binding within the calyx (8, 17). We that a binding assay could be used to identify these
therefore hypothesized that the SCN-activity- putative SCN-binding urinary molecules. While
associated metabolites are biomarkers for intrinsic tryptophan fluorescence quenching (FQ)
unconjugated SCN ligands, which form stable has been used extensively to examine SCN
ferric complexes in the conditions of restrictive binding (5, 8, 19), its susceptibility to
urine (14). spectroscopic interferences prevent its use in
chemically complex, urine-derived specimens. We
In this study, we show that SCN and enterobactin therefore developed a different screening assay
are present simultaneously during human E. coli using a combined liquid chromatography-
2
Human metabolites are antibacterial siderocalin cofactors
differential scanning fluorimetry (LC-DSF) ligands, we assessed the 13 commercially
approach. DSF identifies chromatographic available candidates (including both 3- and 4-
fractions containing SCN ligands through the methyl isomers of methylcatechol) using DSF
thermal stabilization they confer to SCN during (Fig. S2) and FQ (Fig. 4) binding assays. Of these
the melting transition. Because DSF measures metabolites, six changed the SCN T , and five of
M
protein denaturation with a high quantum yield those also exhibited a FQ binding signature (Fig.
fluorescent dye (SYPRO® Orange) and analyzes 4). Only 3,4-dihydroxyhydrocinnamic acid
the melting curve’s first derivative, interferences exhibited discrepant results, with activity in the
from the chemically complex urinary background DSF, but not the FQ assay. The five metabolites
are minimized (20). With DSF, the prototypical with both DSF and FQ binding signals—
SCN ligands enterobactin and 2,3- pyrogallol, caffeic acid, 3-methylcatechol, 4-
dihydroxybenzoic acid (2,3-DHBA) exhibit methylcatechol, and propyl gallate—all contain a
altered melting transitions while the non-binding catechol moiety, consistent with the chemical
2,3-DHBA isomer 2,5-dihydroxybenzoic acid group’s Fe(III) chelation, SCN binding (8, 18, 22),
(gentisic acid) exhibited no binding signal (Fig. and urinary SCN activity (14) associations. These
2a) (21). Enterobactin binding appears to prevent results identify a population of iron-chelating SCN
SCN from melting in our assay, consistent with ligands in human urine and suggest that qualitative
D
previous reports describing high SCN-enterobactin and quantitative differences among these may o
w
n
complex stability (5). 2,3-DHBA did not alter the distinguish restrictive and permissive urine. Of lo
a
melting transitions of the non-binding SCN calyx note, none of these metabolites are sulfate de
d
mutant, where two critical lysine residues have conjugates, which were previously associated with fro
m
been mutated to alanines (SCN K125A/K134A; restrictive urine (14). Together, these findings are h
Fig. 2b,c) (8). DSF positivity corresponded to the consistent with catechol sulfates as biomarkers of ttp
://w
analogous FQ signal with the same reagents (Fig. urinary SCN ligand content. w
w
2d-f). These results support DSF as a feasible .jb
screening tool for urinary SCN ligands. Restrictive urines possess higher SCN ligand c.o
rg
concentrations. To determine whether restrictive b/
y
SCN ligands are present in uninfected human urine contains higher SCN ligand concentrations, g
u
urine. We hypothesized that urine supporting we used GC-MS to quantify unconjugated es
t o
greater SCN antimicrobial activity would have pyrogallol, caffeic acid, 3,4- n
N
higher SCN ligand content. To test this, we dihydroxyhydrocinnamic acid, 4-methylcatechol, ov
e
applied LC-DSF to human urine specimens (Fig. as well as unsubstituted catechol (1,2- m
b
e
3a). Two pooled urine specimens were prepared dihydroxybenzene). Propyl gallate and 3- r 1
9
and screened (Table 1): restrictive urine from four methylcatechol were not reliably detected in these , 2
0
subjects that support high SCN antimicrobial specimens. In a cohort containing 8 highly 18
activity and permissive urine from four subjects restrictive urine specimens representing the top
that support minimal SCN activity. Fractions from 20% of SCN activity and 9 permissive specimens,
both urine pools increased SCN T values in the restrictive urines possessed significantly higher
M
presence of ferric ions (Fig. 3b,c; Fig. S1). The pyrogallol (p = 0.0006) and caffeic acid (p = 0.02)
restrictive urine pool exhibited more T -shifting concentrations than permissive urines (Fig. 5a,b).
M
fractions and greater ΔT values than the Catechol, 3,4-dihydroxyhydrocinnamic acid and 4-
M
permissive urine pool (Fig. 3b vs. Fig 3c). methylcatechol were detectable in all samples but
Metabolomic profiling of active fractions by GC- not significantly different (p > 0.05) between
MS revealed multiple urinary SCN ligand restrictive and permissive groups (Fig. 5c-e).
candidates. Where possible, we confirmed the Pyrogallol concentrations were positively
metabolite identities assigned by spectral matching associated (p < 0.0001) with SCN activity, as
through comparison with commercially available defined previously (14) (Fig 6a,b); free urinary
reference compounds. Metabolites that were pyrogallol concentrations were also positively
enriched in DSF-positive fractions relative to associated (p < 0.0001) with the relative
adjacent DSF-negative fractions are listed in Table abundance of urinary pyrogallol sulfate, a SCN
2. To determine which metabolites are SCN activity correlate previously discovered through
3
Human metabolites are antibacterial siderocalin cofactors
metabolomic profiling (Fig. 6c) (14). These 8). This peak had a 356 Da mass shift,
findings suggest that, although SCN ligands may corresponding to a 1:1:1 Fe(III):pyrogallol:caffeic
be ubiquitous in human urine, specific SCN acid complex. Although the SCN calyx possesses
ligands such as pyrogallol may be critical three binding pockets, the observed 2:1 complexes
determinants of SCN activity in the complex are consistent with the previously proposed
urinary environment. stepwise ligand addition and high affinity of 2:1
ferric complexes (which still possess a delocalized
SCN ligands potentiate antibacterial activity in a negative charge) within the calyx (8, 29). To our
defined medium. To determine whether pyrogallol, knowledge, this is the first demonstration that
caffeic acid, or catechol are sufficient to support SCN can use heterogeneous ligand combinations
SCN activity, we supplemented M63 minimal to chelate Fe(III). Together, these results further
media (23, 24) with concentrations of each support a subgroup of urinary catechols as ferric
compound that approximate levels found in human ion-chelating SCN ligands and suggest that
urine and measured E. coli growth. In media only, preferred ligand combinations may affect SCN
SCN inhibited neither UTI89 nor its enterobactin- antibacterial activity in urine.
deficient mutant, UTI89ΔentB (Fig. 7). Addition
of SCN ligands significantly inhibited growth of DISCUSSION
D
both strains, with UTI89ΔentB inhibited to a Together, these results support a mechanistic ow
n
greater degree (Fig. 7). A non-binding metabolite connection between human metabolomic lo
a
control, t-ferulic acid (3-O-methylated caffeic composition and siderocalin’s antibacterial de
d
acid), did not potentiate SCN activity, nor did activity. This connection emerges from combined fro
m
catechols when added to a nutritionally rich biophysical and mass spectrometric analyses that h
medium (Luria-Bertani broth, LB) where iron resolved physiologically relevant SCN ligands ttp://w
availability does not limit growth (data not from the highly complex and diverse human w
w
shown). These results support a direct role for urinary metabolome. These ligands were sufficient .jb
c
urinary SCN ligand interactions in urinary SCN to potentiate SCN activity in defined media, .o
rg
antibacterial activity. experimentally recapitulating their general b/
y
relationship with SCN activity in urine. Together, g
u
Iron/ligand/SCN complex detection with native these results suggest that the previously-observed es
t o
mass spectrometry. Because human urine contains urinary antimicrobial metabolomic profiles (14) n
N
multiple SCN ligands, we hypothesized that each are associated with free catechols that act as iron- ov
e
SCN molecule may use a combination of different binding SCN cofactors. Among these catechols, m
b
e
ligands in the three calyx pockets to chelate a pyrogallol is highly associated with urinary SCN r 1
9
ferric ion. To investigate this possibility, we used activity and may combine with other catechols to , 2
0
non-denaturing, native mass spectrometry to sequester iron within the SCN calyx. 18
resolve calyx occupancy in the presence of
multiple SCN ligands (25–28); a similar approach In this study, both enterobactin and SCN were
previously confirmed enterobactin binding to SCN observed to coexist during human UTIs (Fig. 1).
(26). Using FTICR mass spectrometry, apo-SCN Although SCN is able to form a tight complex
appeared as peaks at m/z 2585.4 and m/z 2298.2, with Fe(III)-enterobactin, enterobactin appears to
matching the [SCN+8H]8+ and [SCN+9H]9+ charge paradoxically counteract the antibacterial activity
state ions with an observed mass of 20,675.0 ± 0.3 of SCN in human urine, where it can provide iron
Da (calculated mass from sequence = 20,675 Da; for bacterial use (14). This unexpected finding
Fig. 8). Following equilibration with Fe(III) and indicates that additional SCN interactions are
ligands, new peaks corresponding to calyx relevant in the urinary context, including the
occupancy by a 1:2 Fe(III):pyrogallol or formation of Fe(III)-catechol-SCN complexes in
Fe(III):caffeic acid complex were detected. No human urine as suggested by the present work.
new peaks were observed with the non-binder Specifically, urine can possess multiple SCN
salicylic acid (Fig. 8) (21). When SCN was ligands, the relative amounts of these ligands are
exposed to a mixture of pyrogallol, caffeic acid, associated with SCN antibacterial activity, and
and catechol, a single new peak was evident (Fig. SCN may favor specific ligand combinations when
4
Human metabolites are antibacterial siderocalin cofactors
forming Fe(III) complexes. It is possible that there specimen, using a combination of classic
is a hierarchy of Fe(III)-catechol-SCN complexes chromatographic fractionation, biophysical
with varying ability to withhold iron from screens, and mass spectrometry. In this study, GC-
microbes. Once enterobactin has competitively MS allowed us to accurately detect and quantify
removed the iron from the Fe(III)-catechol-SCN many small metabolites, however, it is possible
complex, it is possible that residual occupancy of there are additional ligands not readily detectible
the calyx by catechols or other metabolites by this technique. Advances in native mass
prevents SCN from sequestering ferric spectrometry allowed us to address binding
enterobactin, although this has not yet been degeneracy—the use of several distinct catechol
demonstrated. substituents—between individual SCN complexes.
Our chromatographic screen may therefore have
SCN ligands in human urine (pyrogallol, caffeic missed SCN ligands that bind Fe(III) only in
acid, catechol, and methylcatechol) have varied combination with other, chemically distinct,
and complex origins. They likely derive from ligands. In future studies, native mass
plant-based polyphenolic precursors and may spectrometry may thus be an underused
require metabolic processing by the intestinal complement to crystallographic analyses in
microbiome prior to intestinal absorption (30–34). illuminating a protein’s complete physiologic
D
Following absorption, these compounds may be binding behavior. ow
n
conjugated to sulfate or glucuronate in the liver, lo
a
which would interfere with their ability to chelate These results suggest new strategies for UTI de
d
ferric ions and serve as SCN ligands (8, 18, 35). prevention or treatment. Therapeutically fro
m
Indeed, while sulfated catechols were initially manipulating urinary composition through diet h
associated with restrictive urine (14), this study and/or intestinal microbiome-based interventions ttp://w
identified only unconjugated catechols as urinary may enhance SCN activity in the urinary tract w
w
SCN ligands, consistent with previous reports on (37). Elevating urinary pH would be expected to .jb
c
SCN’s binding capacities (8, 18). Although enhance the efficacy of these interventions, as .o
rg
neurotransmitters such as dopamine and would administration of enterobactin biosynthesis b/
y
epinephrine are widely familiar human catechols, inhibitors. Evaluating these interventions will g
u
these molecules (and their related 3,4- require both more sophisticated experimental es
t o
dihydroxycatechol metabolites) do not readily disease systems and non-toxic interventions n
N
bind SCN (8) and were not detected in the present amenable to direct clinical evaluation. The latter ov
e
study. Indeed, catechols with substituents that are may be accomplished through standardized dietary m
b
e
para to one of the catechol alcohols appear to be interventions studies in conjunction with r 1
9
poor SCN binders (18, 21, 36). Nevertheless, the metabolomic profiling. , 2
0
native MS results raise the possibility that these 18
poor binders might be effective SCN ligands in EXPERIMENTAL PROCEDURES
combination with other catechols. Mixed-ligand Uncomplicated E. coli cystitis specimens. Urinary
complexes also may partially explain why some specimens were from a previously described study
urine specimens support little SCN antimicrobial (14). The Institutional Review Board of the
activity but contain some catechols; it may be University of Washington approved all study
certain combinations, in addition to favorable pH protocols. All patients provided written informed
(14, 17), that enable SCN to restrict bacterial consent for the sample collection and subsequent
growth. A more detailed description of SCN analyses. Clean-catch midstream urine specimens
binding will be necessary to resolve these were obtained from female patients with acute
possibilities. uncomplicated cystitis between 2008 and 2012 at
the University of Washington Hall Health Primary
Identifying new ligands and substrates for proteins Care Center. Women were eligible if they were
and enzymes is a recurring challenge in aged 18-49 years, in good general health, and had
biochemistry. Metabolomic associations motivated fewer than 7 days of typical symptoms of acute
the approach described here, which sought new cystitis defined using previously described criteria
ligands in urine, a notoriously complex biological of dysuria, urinary frequency or urinary urgency
5
Human metabolites are antibacterial siderocalin cofactors
(38). SigmaFAST™ protease inhibitor solution equivalent volume of PBS. After 20 hours, viable
(Sigma) was added at 1/10th volume to freshly bacterial concentrations were determined by cfu
voided urines before clinical centrifugation, and enumeration. Where indicated, media was
the supernatant was frozen at −80°C. supplemented with candidate metabolites at 5 µM.
Uropathogens in midstream urine were identified
using standard methods. Urine specimens analyzed Urine fractionation. Healthy donor urines were
here were collected during subjects’ initial visits collected as approved by the Washington
with a diagnosis of E. coli acute uncomplicated University School of Medicine Institutional
cystitis, with ≥105 cfu/mL of a beta-hemolytic Review Board, and described previously (14). To
isolate. Urinary SCN was quantified by ELISA as fractionate and concentrate urinary small
previously published (14). molecules, we used a multi-step chromatographic
scheme. Four urines of either high or low SCN
antimicrobial activity were pooled (2 mL each)
Enterobactin quantification. Urinary enterobactin
and applied to Chrom P SPE cartridges (Supelco).
was detected and identified using LC-MS/MS.
The flow through, water wash, and elutions with
Briefly, human E. coli cystitis urine samples were
50% and then 100% HPLC-grade methanol
spiked 1:10 with 13C-labeled internal standard (23,
(Sigma) were all collected and lyophilized to
24), then filtered and diluted 1:1 with water before D
LC-MS/MS analysis with a Shimadzu Prominence dryness. Each of the four preparations was then ow
n
UFLC-coupled AB Sciex 4000 QTRAP mass rehydrated in HPLC-grade water (Sigma) and loa
fractionated over a phenylhexyl Eternity column de
spectrometer. Chromatography was conducted d
with a gradient of 0.1% formic acid (Fluka, (250 x 4.6 mm, 5 µm; Supelco) using a from
S0.i1g%m a)f otrom i9c0 %ac iadc,e tuosniintrgi lea n(E MAsDc enMtiisl lipEoxrper)e s+s wchartoemr:matoetghraanpohly gsraydstieemnt o(Bn ioa -RBaiodL).o gAicll DfruaocFtiloonws http://w
phenyl-hexyl column (100 × 2.1 mm, 2.7 µm; (~290 total) were lyophilized to dryness and stored w
w
Supelco, Sigma). Negative electrospray ionization at -80°C until analysis. .jb
c
(ESI) derived [M-H]- precursor-product ions were .org
m/z 686 > 222 for linear enterobactin and m/z 716 Differential scanning fluorimetry. To screen for b/
y
> 232 for the corresponding 13C isotopologue SCN binding in urinary fractions, we measured gu
30 thermal denaturation through differential scanning es
internal standard. Peak area ratios were averaged t o
fluorimetry (DSF), essentially as described (20, n
for triplicate analyses. N
39). Lyophilized urinary HPLC fractions were ov
reconstituted in 100 µL HPLC-grade water thus em
b
Protein purification. Human SCN protein (the yielding a roughly 80-fold concentration from the er 1
kind gift of Roland Strong) was expressed on a original bulk urine. Fractions (10 µL) were added 9, 2
pGEX4T vector in BL21 E. coli as previously to a final concentration of 3 µM SCN in PBS with 018
described (5, 14). The human SCN calyx mutant
a 5X concentration of SYPRO Orange dye
K125A/K134A was expressed on a pET22b vector
(Sigma) and either 3.2 µg/mL FeCl or an equal
3
in BL21 E. coli and purified as described (8, 14).
volume of water, and allowed to equilibrate for 20
Protein purity was monitored by mass
minutes. Using a CFX96 real-time PCR machine
spectrometry and SDS-PAGE.
(Bio Rad), reactions were then heated in 96-well

quantitative PCR plates (Bio Rad) from 22°C to
Media growth assay. Defined media used to test
98°C at 1°C per minute, with plate reading at each
metabolite supplementation were M63 minimal
increment. Melt transitions were monitored on the
salts media supplemented with 0.2% glycerol and
HEX channel and displayed as the derivative of
10 µg/mL niacin, and serum-free RPMI (Gibco).
relative fluorescence units with respect to
SCN activity assays were performed essentially as
temperature (dRFU/dT). Melting temperature (T )
M
described previously (13, 14). Briefly, media was
values were calculated from the raw amplification
inoculated with the uncomplicated cystitis isolate
curves using a Boltzmann regression in Prism v.
UTI89 or isogenic siderophore mutants (23) at 103
7.0a (20). Fractions were also analyzed in the
cfu/mL, from 4-6 hour cultures in LB broth.
absence of protein as a negative control, and SCN
Cultures were grown with 1.5 µM SCN or an
was incubated with ferric catechol (30 µM) as a
6
Human metabolites are antibacterial siderocalin cofactors
positive control, and negative controls included minute; the injector and transfer line temperatures
protein plus iron and fractions without protein were 250°C. Fraction spectra were matched to the
present. Interesting fractions were identified by a NIST11 Mass Spectral Library for chemical
melting temperature (T ) shift. identification using AMDIS, and metabolites
M
elevated in binding fractions compared to non-
Preliminary compounds identified by GC-MS binding fractions were identified as potential
were also confirmed by DSF. These reactions were ligand candidates and analyzed further by DSF
carried out as above, but with a ferric complex and FQ.
(FeL ) concentration of 30 µM. This saturating
3
ligand concentration is used to ensure all protein is To quantify SCN-binding metabolites by GC-MS,
in the bound form for the melt analysis, as we produced calibration curves for each
established (20). Chemicals were purchased from metabolite relative to the exometabolite 4-
Sigma Aldrich and prepared with FeCl or without fluorosalicylic acid (4FSA) (14). Several
3
(equal volume PBS) at a 1:3 ratio of iron to ligand metabolites could not be quantified because no
(Fe(L) ). SCN melt curves were performed as standard was available, or the standard was not
3
above, and a positive DSF result was considered a reliably detectable by GC-MS. Urine specimens
T shift or loss of the melt signal. (100 µL) were spiked with 200 picomoles of
M D
4FSA and treated with 2.4 units of urease (Sigma) ow
n
Fluorescence quenching. We confirmed for 40 minutes before quenching with one volume lo
a
d
metabolite-SCN interactions by measuring binding of methanol. The reaction was dried under e
d
of DSF-positive candidates by intrinsic tryptophan nitrogen and derivatized as described above. GC- fro
m
fluorescence quenching, essentially as described MS was performed in SIM mode based on the two h
(5, 8). SCN protein (500 nM) was mixed with a fragment ions for each candidate that were used ttp://w
dilution series of ligand in the presence and for standard curves. Both fragment ions were w
w
absence of iron in PBS with 2% DMSO in black monitored to assure correct identification. Molar .jb
c
96-well fluorescence plates (Corning). After one- concentrations were determined using the standard .o
rg
hour equilibration, tryptophan fluorescence was curve slope to correct the integrated peak area b/
y
monitored at excitation and emission wavelengths ratio of 4FSA and the analytes of interest. g
u
of 281 nm and 340 nm, respectively, on a TECAN est o
Infinite 200 microplate reader at room Native mass spectrometry. SCN protein (30 µM) n N
temperature. Binding was monitored up to 48 was prepared in 50 mM ammonium acetate ov
e
hours later with no loss of signal or notable change (Sigma), pH 6.95. Where indicated, FeCl3 (50 µM) mbe
in fluorescence quenching patterns (data not and SCN ligand compounds (60 µM for individual r 1
9
shown). compounds; 100 µM total for mixed compounds) , 2
0
were added, equilibrated at room temperature for 2 18
Gas chromatography-mass spectrometry. hours, and stored at 4°C until analysis. For native
Fractions identified by DSF as containing SCN- electrospray ionization, a 10 µL sample was
binding character were profiled by gas injected through a nanoES spray capillary
chromatography-mass spectrometry (GC-MS). (Thermo Scientific) interfaced with a Bruker
Briefly, N-methyl-N-trimethylsilyl Solarix 12T FTICR mass spectrometer (Bruker
trifluoroacetamide (MSTFA)-derivatized samples Daltonics, Bremen, Germany). The capillary
were analyzed on an Agilent 7890A gas voltage was 0.9-1.3 kV. The drying gas
chromatograph interfaced to an Agilent 5975C temperature was 30°C; its flow was 1 L/min. The
mass spectrometer operated in the electron ISCID voltage was set to 70 V to help desolvation.
ionization (EI) mode; the source temperature, The ion-funnel RF amplitude was 300 Vpp, and
electron energy and emission current were 230°C, the ion-funnel voltages were 200 V (funnel 1) and
70eV and 300µA, respectively. GC 18 V (funnel 2). RF frequencies in all ion-
chromatography was performed with an HP-5MS transmission regions were set to the lowest
column (30 m, 0.25mm i.d., 0.25µm film coating; available value: multipole 1 (2 MHz), quadrupole
P.J. Cobert, St. Louis, MO) with a linear (1.4 MHz) and transfer line (1 MHz). Ions
temperature gradient of 80-300°C at 10°C per accumulated for 500 ms in the RF-hexapole ion
7
Human metabolites are antibacterial siderocalin cofactors
trap before transmission to the infinity ICR trap.
The time-of-flight was 1.3 ms for the protein- Statistical analyses. Data were analyzed in Prism
ligand ions. The source region pressure was 2.3 v. 6.0d (GraphPad). For parametric and non-
mbar; the quadrupole region pressure was 4.4x10-6 parametric two-sample unpaired comparisons we
mbar, and the trap-chamber pressure was 1.3x10-9 used t-test and Mann-Whitney U test, respectively.
mbar. MS parameters were slightly modified for To compare multiple groups versus control or each
each individual sample to obtain an optimized other, we used one-way ANOVA. Where
signal. One to several hundred scans was averaged appropriate, post-tests were used to correct for
for each spectrum. External calibration was multiple comparisons.
performed with ESI of cesium perfluoroheptanoic
acetate to a maximum of m/z 8500.

Acknowledgements: The authors thank Roland Strong and Jonathan Barasch for generously sharing
siderocalin constructs, and Lindsey Steinberg and Anne Robinson for careful reading of the manuscript.

Conflicts of interest: The authors declare that they have no conflicts of interest with the contents of this
D
article.4 ow
n
lo
a
d
Author contributions: RS-C and JPH designed the studies and wrote the paper. JRC and RS-C e
d
performed and analyzed GC-MS experiments. CDM and RS-C performed and analyzed DSF and FQ on fro
m
SCN. AES designed and conducted the study from which cystitis urines were acquired. WC performed h
native MS experiments. RS-C performed all other experiments, analyzed data and results, and prepared ttp://w
the figures. All authors reviewed the results and approved the final version of the manuscript. w
w
.jb
c
.o
rg
b/
y
g
u
e
s
t o
n
N
o
v
e
m
b
e
r 1
9
, 2
0
1
8
8
Human metabolites are antibacterial siderocalin cofactors
REFERENCES
1. Cassat, J. E., and Skaar, E. P. (2013) Iron in infection and immunity. Cell Host Microbe. 13, 509–
519
2. Hood, M. I., and Skaar, E. P. (2012) Nutritional immunity: transition metals at the pathogen-host
interface. Nat. Rev. Microbiol. 10, 525–537
3. Cherayil, B. J. (2011) The role of iron in the immune response to bacterial infection. Immunol. Res.
50, 1–9
4. Becker, K. W., and Skaar, E. P. (2014) Metal limitation and toxicity at the interface between host
and pathogen. FEMS Microbiol. Rev. 38, 1235–1249
5. Goetz, D. H., Holmes, M. A., Borregaard, N., Bluhm, M. E., Raymond, K. N., and Strong, R. K.
(2002) The neutrophil lipocalin NGAL is a bacteriostatic agent that interferes with siderophore-
mediated iron acquisition. Mol. Cell. 10, 1033–1043
6. Sia, A. K., Allred, B. E., and Raymond, K. N. (2013) Siderocalins: Siderophore binding proteins
evolved for primary pathogen host defense. Curr. Opin. Chem. Biol. 17, 150–157
7. Clifton, M. C., Corrent, C., and Strong, R. K. (2009) Siderocalins: siderophore-binding proteins of
the innate immune system. Biometals Int. J. Role Met. Ions Biol. Biochem. Med. 22, 557–564
8. Bao, G., Clifton, M., Hoette, T. M., Mori, K., Deng, S.-X., Qiu, A., Viltard, M., Williams, D.,
D
Paragas, N., Leete, T., Kulkarni, R., Li, X., Lee, B., Kalandadze, A., Ratner, A. J., Pizarro, J. C., o
w
n
Schmidt-Ott, K. M., Landry, D. W., Raymond, K. N., Strong, R. K., and Barasch, J. (2010) Iron lo
a
traffics in circulation bound to a siderocalin (Ngal)-catechol complex. Nat. Chem. Biol. 6, 602–609 de
d
9. Raymond, K. N., Dertz, E. A., and Kim, S. S. (2003) Enterobactin: an archetype for microbial iron fro
m
transport. Proc. Natl. Acad. Sci. U. S. A. 100, 3584–3588 h
10. Miethke, M., and Marahiel, M. A. (2007) Siderophore-based iron acquisition and pathogen control. ttp
://w
Microbiol. Mol. Biol. Rev. 71, 413–451 w
w
11. Abergel, R. J., Clifton, M. C., Pizarro, J. C., Warner, J. A., Shuh, D. K., Strong, R. K., and .jb
Raymond, K. N. (2008) The siderocalin/enterobactin interaction: a link between mammalian c.o
rg
immunity and bacterial iron transport. J. Am. Chem. Soc. 130, 11524–11534 b/
y
12. Flo, T. H., Smith, K. D., Sato, S., Rodriguez, D. J., Holmes, M. A., Strong, R. K., Akira, S., and g
u
Aderem, A. (2004) Lipocalin 2 mediates an innate immune response to bacterial infection by es
t o
sequestrating iron. Nature. 432, 917–921 n
N
13. Bachman, M. A., Oyler, J. E., Burns, S. H., Caza, M., Lépine, F., Dozois, C. M., and Weiser, J. N. o
v
e
(2011) Klebsiella pneumoniae yersiniabactin promotes respiratory tract infection through evasion of m
b
e
lipocalin 2. Infect. Immun. 79, 3309–3316 r 1
9
14. Shields-Cutler, R. R., Crowley, J. R., Hung, C. S., Stapleton, A. E., Aldrich, C. C., Marschall, J., , 2
0
and Henderson, J. P. (2015) Human urinary composition controls siderocalin’s antibacterial activity. 18
J. Biol. Chem. 10.1074/jbc.M115.645812
15. Paragas, N., Kulkarni, R., Werth, M., Schmidt-Ott, K. M., Forster, C., Deng, R., Zhang, Q., Singer,
E., Klose, A. D., Shen, T. H., Francis, K. P., Ray, S., Vijayakumar, S., Seward, S., Bovino, M. E.,
Xu, K., Takabe, Y., Amaral, F. E., Mohan, S., Wax, R., Corbin, K., Sanna-Cherchi, S., Mori, K.,
Johnson, L., Nickolas, T., D’Agati, V., Lin, C.-S., Qiu, A., Al-Awqati, Q., Ratner, A. J., and
Barasch, J. (2014) α-Intercalated cells defend the urinary system from bacterial infection. J. Clin.
Invest. 124, 2963–2976
16. Steigedal, M., Marstad, A., Haug, M., Damås, J. K., Strong, R. K., Roberts, P. L., Himpsl, S. D.,
Stapleton, A., Hooton, T. M., Mobley, H. L. T., Hawn, T. R., and Flo, T. H. (2014) Lipocalin 2
imparts selective pressure on bacterial growth in the bladder and is elevated in women with urinary
tract infection. J. Immunol. 193, 6081–6089
17. Abergel, R. J., Warner, J. A., Shuh, D. K., and Raymond, K. N. (2006) Enterobactin protonation and
iron release:  Structural characterization of the salicylate coordination shift in ferric enterobactin. J.
Am. Chem. Soc. 128, 8920–8931
2
Human metabolites are antibacterial siderocalin cofactors
18. Bao, G.-H., Barasch, J., Xu, J., Wang, W., Hu, F.-L., and Deng, S.-X. (2015) Purification and
structural characterization of “simple catechol”, the NGAL-siderocalin siderophore in human urine.
RSC Adv. 5, 28527–28535
19. Hoette, T. M., Clifton, M. C., Zawadzka, A. M., Holmes, M. A., Strong, R. K., and Raymond, K. N.
(2011) Immune interference in Mycobacterium tuberculosis intracellular iron acquisition through
siderocalin recognition of carboxymycobactins. ACS Chem. Biol. 6, 1327–1331
20. Niesen, F. H., Berglund, H., and Vedadi, M. (2007) The use of differential scanning fluorimetry to
detect ligand interactions that promote protein stability. Nat. Protoc. 2, 2212–2221
21. Correnti, C., Richardson, V., Sia, A. K., Bandaranayake, A. D., Ruiz, M., Suryo Rahmanto, Y.,
Kovačević, Ž., Clifton, M. C., Holmes, M. A., Kaiser, B. K., Barasch, J., Raymond, K. N.,
Richardson, D. R., and Strong, R. K. (2012) Siderocalin/Lcn2/NGAL/24p3 does not drive apoptosis
through gentisic acid mediated iron withdrawal in hematopoietic cell lines. PloS One. 7, e43696
22. Gómez-Casado, C., Roth-Walter, F., Jensen-Jarolim, E., Díaz-Perales, A., and Pacios, L. F. (2013)
Modeling iron-catecholates binding to NGAL protein. J. Mol. Graph. Model. 45C, 111–121
23. Henderson, J. P., Crowley, J. R., Pinkner, J. S., Walker, J. N., Tsukayama, P., Stamm, W. E.,
Hooton, T. M., and Hultgren, S. J. (2009) Quantitative metabolomics reveals an epigenetic blueprint
for iron acquisition in uropathogenic Escherichia coli. PLoS Pathog. 5, e1000305
D
24. Chaturvedi, K. S., Hung, C. S., Crowley, J. R., Stapleton, A. E., and Henderson, J. P. (2012) The o
w
n
siderophore yersiniabactin binds copper to protect pathogens during infection. Nat. Chem. Biol. 8, lo
a
731–736 de
d
25. Cui, W., Rohrs, H. W., and Gross, M. L. (2011) Top-down mass spectrometry: recent fro
m
developments, applications and perspectives. The Analyst. 136, 3854–3864 h
26. Doneanu, C. E., Strong, R. K., and Howald, W. N. (2004) Characterization of a noncovalent ttp
://w
lipocalin complex by liquid chromatography/electrospray ionization mass spectrometry. J. Biomol. w
w
Tech. JBT. 15, 208–212 .jb
27. Heck, A. J. R. (2008) Native mass spectrometry: a bridge between interactomics and structural c.o
rg
biology. Nat. Methods. 5, 927–933 b/
y
28. Ruotolo, B. T., and Robinson, C. V. (2006) Aspects of native proteins are retained in vacuum. Curr. g
u
Opin. Chem. Biol. 10, 402–408 es
t o
29. Correnti, C., and Strong, R. K. (2012) Mammalian siderophores, siderophore-binding lipocalins, n
N
and the labile iron pool. J. Biol. Chem. 287, 13524–13531 o
v
e
30. Gonthier, M.-P., Cheynier, V., Donovan, J. L., Manach, C., Morand, C., Mila, I., Lapierre, C., m
b
e
Rémésy, C., and Scalbert, A. (2003) Microbial aromatic acid metabolites formed in the gut account r 1
9
for a major fraction of the polyphenols excreted in urine of rats fed red wine polyphenols. J. Nutr. , 2
0
133, 461–467 18
31. Li, M., Wang, B., Zhang, M., Rantalainen, M., Wang, S., Zhou, H., Zhang, Y., Shen, J., Pang, X.,
Zhang, M., Wei, H., Chen, Y., Lu, H., Zuo, J., Su, M., Qiu, Y., Jia, W., Xiao, C., Smith, L. M.,
Yang, S., Holmes, E., Tang, H., Zhao, G., Nicholson, J. K., Li, L., and Zhao, L. (2008) Symbiotic
gut microbes modulate human metabolic phenotypes. Proc. Natl. Acad. Sci. U. S. A. 105, 2117–
2122
32. van Duynhoven, J., Vaughan, E. E., Jacobs, D. M., Kemperman, R. A., van Velzen, E. J. J., Gross,
G., Roger, L. C., Possemiers, S., Smilde, A. K., Doré, J., Westerhuis, J. A., and Van de Wiele, T.
(2011) Metabolic fate of polyphenols in the human superorganism. Proc. Natl. Acad. Sci. U. S. A.
108 Suppl 1, 4531–4538
33. Booth, A. N., Masri, M. S., Robbins, D. J., Emerson, O. H., Jones, F. T., and De Eds, F. (1959) The
metabolic fate of gallic acid and related compounds. J. Biol. Chem. 234, 3014–3016
34. Wikoff, W. R., Anfora, A. T., Liu, J., Schultz, P. G., Lesley, S. A., Peters, E. C., and Siuzdak, G.
(2009) Metabolomics analysis reveals large effects of gut microflora on mammalian blood
metabolites. Proc. Natl. Acad. Sci. 106, 3698–3703
35. Falany, C. N. (1997) Enzymology of human cytosolic sulfotransferases. FASEB J. 11, 206–216
3