ISSN 0003-2654
Analyst
0003-2654(2013)138:7;1-C
www.rsc.org/analyst Volume 138 | Number 7 | 7 April 2013 | Pages 1911–2200
Vo
lum
e 1
38
|
Nu
mb
er 7
|
20
13
Analyst
P
ag
es
19
11
–2
20
0
www.rsc.org/analystRegistered Charity Number 207890
Showcasing research of biological applications of spectroscopic imaging from the laboratory of Professor Sergei Kazarian at the Department of Chemical Engineering, Imperial College London, United Kingdom.
Title: ATR-FTIR spectroscopic imaging: recent advances and
applications to biological systems
Infrared spectroscopic imaging is an emerging technology for
studies of dynamic processes in biological samples such aorta,
skin and live cells, demonstrating its great potential for medical
diagnosis and biomedical research.
As featured in:
See Sergei G. Kazarian and
K. L. Andrew Chan,
Analyst, 2013, 138, 1940–1951.
HOT ARTICLEIngela Lanekoff , Julia Laskin et al.Spatially resolved analysis of glycolipids and metabolites in living Synechococcus sp. PCC 7002using nanospray desorption electrospray ionization
Analyst
PAPER
Dow
nloa
ded
on 1
2 M
arch
201
3Pu
blis
hed
on 3
0 Ja
nuar
y 20
13 o
n ht
tp://
pubs
.rsc
.org
| do
i:10.
1039
/C3A
N36
716A
View Article OnlineView Journal | View Issue
aChemical and Materials Sciences Division, P
Box 999, K8-88 Richland, WA 99352, USA.
[email protected] Sciences Division, Pacic Nor
Washington 99352, USA
† Electronic supplementary informa10.1039/c3an36716a
Cite this: Analyst, 2013, 138, 1971
Received 19th November 2012Accepted 29th January 2013
DOI: 10.1039/c3an36716a
www.rsc.org/analyst
This journal is ª The Royal Society of
Spatially resolved analysis of glycolipids andmetabolites in living Synechococcus sp. PCC 7002 usingnanospray desorption electrospray ionization†
Ingela Lanekoff,*a Oleg Geydebrekht,b Grigoriy E. Pinchuk,b Allan E. Konopkab
and Julia Laskin*a
Microorganisms release a diversity of organic compounds that couple interspecies metabolism, enable
communication, or provide benefits to other microbes. Increased knowledge of microbial metabolite
production will contribute to understanding of the dynamic microbial world and can potentially lead to
new developments in drug discovery, biofuel production, and clinical research. Nanospray desorption
electrospray ionization (nano-DESI) is an ambient ionization technique that enables detailed chemical
characterization of molecules from a specific location on a surface without special sample pretreatment.
Due to its ambient nature, living bacterial colonies growing on agar plates can be rapidly analyzed
without affecting the viability of the colony. In this study we demonstrate for the first time the utility of
nano-DESI for spatial profiling of chemical gradients generated by microbial communities on agar
plates. We found that despite the high salt content of the agar used in this study (�350 mM), nano-
DESI analysis enables detailed characterization of metabolites produced by the Synechococcus sp. PCC
7002 colonies. High resolution mass spectrometry and MS/MS analysis of the living Synechococcus sp.
PCC 7002 colonies allowed us to detect metabolites and lipids on the colony and on the surrounding
agar, and confirm their identities. High sensitivity of nano-DESI enabled identification of several
glycolipids that have not been previously reported by extracting the cells using conventional methods.
Spatial profiling demonstrated that a majority of lipids and metabolites were localized on the colony
while sucrose and glucosylglycerol, an osmoprotective compound produced by cyanobacteria, were
secreted onto agar. Furthermore, we demonstrated that the chemical gradients of sucrose and
glucosylglycerol on agar depend on the age of the colony. The methodology presented in this study will
facilitate future studies focused on molecular-level characterization of interactions between bacterial
colonies.
Introduction
Microbial communities play important roles in almost all envi-ronments on earth.1 They comprise almost half of all life onearth2 and their primary role in the biosphere is as catalysts ofbiogeochemical cycles.3 In addition, some microbes haveimportant effects on specic plants or animals, as pathogens orsymbionts. The diversity among bacteria is enormous andknowledge of bioengineering and metabolic pathways inbacteria has led to benecial industrial applications in produc-tion of pharmaceuticals,4 fermented foods5 and biofuels6,7 as
acic Northwest National Laboratory, PO
E-mail: [email protected]; Julia.
thwest National Laboratory, Richland,
tion (ESI) available. See DOI:
Chemistry 2013
well as inwastewater treatment.8Cyanobacteria are important tothe evolution of modern life on earth as their photosyntheticproduction of O2 was responsible for the transition from areducing to the oxygenic atmosphere of today.9,10Cyanobacterialassemblages oen dominate the top layers of microbial mats11
and aremajor primary producers in the ocean.12 Inmats, severalbacterial species comprise microbial consortia where theexchange of molecules between species couples metabolism orenables communication.11,13–16 Single bacterial cells alsocommunicate by chemical signals within colonies or planktonicpopulations, a process knownasquorumsensing,17 and secrete arange of bioactive compounds such as antibiotics, siderophores,etc.15,16,18 Increased knowledge about bacterial metabolites givesinsight into the dynamic world of microbial communities thatcan be benecial in future industrial applications and forunderstanding biogeochemical cycles on earth.
Chemical characterization of metabolites in bacterial colo-nies is typically performed using nuclear magnetic resonance
Analyst, 2013, 138, 1971–1978 | 1971
Analyst Paper
Dow
nloa
ded
on 1
2 M
arch
201
3Pu
blis
hed
on 3
0 Ja
nuar
y 20
13 o
n ht
tp://
pubs
.rsc
.org
| do
i:10.
1039
/C3A
N36
716A
View Article Online
(NMR) and gas and liquid chromatography mass spectrometry(GC/MS and LC/MS, respectively).19–22 Ambient surface ioniza-tion techniques including desorption electrospray ionization(DESI), low temperature plasma (LTP), and direct analysis inreal time (DART) have been also used for the analysis ofbacterial samples.23–27 However, these studies did not provideinformation on the spatial distribution of metabolites withinthe sample. Mass spectrometry imaging (MSI) is widely used forobtaining spatial distributions of lipids, proteins and metabo-lites in tissue sections,28–31 and has recently been used forspatial proling of metabolites produced by co-culturedmicrobial communities.32 For example, matrix assisted laserdesorption (MALDI) imaging has been used for understandingchemical interactions between bacterial colonies.33,34 Thedistribution of surfactins, secreted by Bacillus subtilis, wasexamined by imprinting the biolm grown on agar onto asilicon shard prior to analysis and imaging using time of ightsecondary ion mass spectrometry (TOF-SIMS).35 Laser desorp-tion postionization (LDPI) MS enabled detection of antibioticcompounds in biolms grown on indium thin oxide slides.36 Inaddition, LDPI-MS has been used for imaging peptides inBacillus subtilis aer transferring biolms to aluminum platesfor peptide derivatization prior to analysis.37 DESI-MS wasutilized for detection of secondary metabolites, localizedbetween two bacterial systems,38 and characterization of lipidproles of several bacterial samples.39 Additionally, severalendogenous molecules have been detected in lyophilized cya-nobacteria using laser ablation electrospray ionization (LAESI)MS.40 However, spatially resolved characterization of microbialcommunities using these techniques requires some degree ofsample preparation (e.g. imprinting, drying and/or matrixapplication) that is destructive to the colony.
Nanospray desorption electrospray ionization (nano-DESI)41
has been recently used for metabolic proling of living bacterialcolonies.42,43Nano-DESI is an ambient ionization technique thatenables chemical analysis of complex organic and biologicalsamples from substrates without sample preparation.41,42,44–46
Similar to liquid microjunction surface sample probe (LMJ-SSP)47 and liquid extraction surface analysis (LESA),48 nano-DESI relies on localized liquid extraction of analytes from thesurface. In nano-DESI, analytes are desorbed into a solventbrought to the sample surface by a primary capillary. The des-orbed analytes are then immediately transferred into a self-aspirating secondary capillary and nano-electrosprayed into themass spectrometer.41 The liquid bridge between the two capil-laries is small (ranging between 10 and 200 mm) which enablesspatially resolved analysis of biological samples.44 Gentleextraction of the analyte molecules from substrates enablesanalysis of living microbial colonies grown on agar. It has beendemonstrated that nano-DESI analysis does not have a signi-cant effect on the viability and the phenotype of living bacterialcolonies.42 In this study, we demonstrate for the rst time theuse of spatially resolved nano-DESI MS for detection of chem-ical gradients of lipids and metabolites in living Synechococcussp. PCC 7002 colonies and the surrounding agar. This capabilityis important for understanding interspecies interactionsbetween microbial colonies on agar plates.
1972 | Analyst, 2013, 138, 1971–1978
Synechococcus sp. PCC 7002 is a marine cyanobacterium thatis typically cultivated on media with a high salt content. Itrequires only light, inorganic mineral nutrients and carbondioxide to produce hundreds of organic compounds whichpotentially could be used in metabolic exchange, nutrientsacquisition, or cell-to-cell signaling. Previously, several meta-bolites in Synechococcus sp. PCC 7002 suspensions have beenidentied using LC-MS,14,49–51 LC-DAD,52 FAB-MS,53GC-FID54 andabsorbance spectrometry.55 However, high salt content of theagar is known to suppress the ion signal in mass spectrometryanalysis, hence limiting the ability of data acquisition.19,56 It hasbeen demonstrated that ambient surface ionization techniquesenable chemical analysis of samples with high salt content.57,58
Here we use nano-DESI, for spatially resolved detection andstructural characterization of metabolites and lipids from livingSynechococcus sp. PCC 7002 colonies and the surrounding high-salt agar. For this initial proof-of-principle study, the colony wasdeliberately grown to generate one-dimensional chemicalgradients. Despite the high salt content of the agar, we were ableto detect numerous lipids and metabolites of Synechococcus sp.PCC 7002 on the colony and detect molecules secreted onto thesurrounding agar. We demonstrate for the rst time the evolu-tion of chemical gradients generated by the microbial commu-nity on the agar plate as a function of the age of the colony. Wealso show rapid detection andMS/MS characterization of severalpreviously identiedglycolipids and report on several glycolipidsfrom the living Synechococcus sp. PCC 7002 colony that have notbeen observed using conventional cell extraction approaches.
ExperimentalBacterial culture
The cyanobacterium Synechococcus sp. PCC 7002 (formerlyknown as Agmenellum quadruplicatum strain PR-6)59,60 wascultivated on modied A+ nobel agar (added to nal concen-tration 1.5%) of the following composition: NaCl (0.3 M), KCl(8 mM), NH4Cl (17 mM), MgSO4$7H2O (20 mM), KH2PO4 (0.37mM), CaCl2 (1.8 mM), Na2EDTA (79 mM), vitamin B12 (3 pM),FeCl3$6H2O (14.4 mM), H3BO3 (0.55 mM), MnCl2$4H2O (22 mM),ZnCl (2.3 mM), Na2MoO4$2H2O (0.18 mM), CuSO4$5H2O (12 pM),CoCl2$6H2O (51 pM),61 supplemented with bicarbonate (5 mM).Buffering was achieved by growing cultures in air supplementedwith 3% of CO2.
Synechococcus sp. PCC 7002 colonies were grown in Petridishes (100 � 15 mm) containing 20 ml of solid medium incu-bated in a light chamber at 28 �Cand 80% relativehumidity for 2–6 days. The samples were subsequently taken out to roomtemperature (21 �C) and incubated under a light bulb (Philipsdaylight deluxe, FC8T9/DX) for 1–4 days. The light bulb producedenough light to keep the bacteria alive without noticeable prop-agation. The time course study was performed on bacterialcolonies kept in the light chamber for 48, 72 and 144 hours. Eachdata point in the time course experiment was acquired from adifferent sample.Hence, this experimentwasperformedonmanybiological replicas of the same organism grown under the sameconditions. For each data point in this experiment, we acquired2–3 technical replicas to ensure the reproducibility of the data.
This journal is ª The Royal Society of Chemistry 2013
Fig. 1 Typical spectra obtained from (A) A+ agar and (B) living Synechococcus sp.PCC 7002 colonies on A+ agar.
Paper Analyst
Dow
nloa
ded
on 1
2 M
arch
201
3Pu
blis
hed
on 3
0 Ja
nuar
y 20
13 o
n ht
tp://
pubs
.rsc
.org
| do
i:10.
1039
/C3A
N36
716A
View Article Online
Nano-DESI analysis
The analysis was performed on a piece of agar cut out of thePetri dish and placed on a microscope glass slide. No furthersample preparation was performed prior to analysis. A custommade nano-DESI source62 was mounted onto an XL LTQ/Orbi-trap mass spectrometer (Thermo Electron, Bremen, Germany).The primary and secondary capillaries were made of 50 �183 mm (ID � OD) fused silica capillaries (Polymicro Technol-ogies, L.L.C., Phoenix). The nano-DESI solvent was composed of10% water (HPLC water, Fisher Scientic) and 90% methanol(Fisher Scientic) and was delivered through the primarycapillary at a ow rate of 0.7–1.2 ml min�1. Ions were producedat the heated capillary inlet of the mass spectrometer, held at250 �C, by applying a high voltage of 2.8–3.5 kV to the primarycapillary. The custom made sample stage and holder wasattached to a motorized XYZ stage (Newport Corp., Irvine, CA),controlled by a custom-designed Labview soware described indetail elsewhere.62 A glass slide with a sample was placed on theholder. Spatially resolved chemical information was obtainedby moving the sample at 0.02 mm s�1 under the nano-DESIprobe. In these experiments, the analysis started on the agaraway from the colony and moved toward the colony. The massspectrometer was operated in the positive mode with massresolution of 60 000 (m/Dm) at m/z 400; MS/MS experimentswere performed by varying the collision energy between 0 and40 units (the setting specied by the manufacturer) for obtain-ing optimal fragmentation. XCalibur soware (Thermo Scien-tic) was used for data analysis and the extracted ionchronograms (time dependent signal for selected m/z) weretransferred into an Excel spreadsheet where they were normal-ized to both the total ion current and the respective maximumintensity and plotted in Origin with a rolling average of 5.Standard glucosylglycerol (Toronto Research Chemicals Inc.)was used to conrm the identity of the endogenous metabolite.
Results and discussionMetabolites and lipids in living Synechococcus sp. PCC 7002colony
A+ medium, widely used to grow Synechococcus sp. PCC 7002,contains 0.3 M NaCl and �50 mM of other inorganic salts. As aresult, themass spectrumof the solidmediumdetected by nano-DESI is very complex (Fig. 1A) with amajority of abundant peakscorresponding to clusters of zinc, copper and chloride, asdetermined by the isotopic patterns. These peaks are notobserved when the nano-DESI probe is localized on the livingSynechococcus sp. PCC 7002 colony (Fig. 1B). Themost abundantmetabolite peaks in the spectrum of the colony are tentativelyassigned as glucosylglycerol ([M + Na]+, m/z 277.0888) and chlo-rophyll a ([M + Na]+, m/z 915.5235). Glucosylglycerol is anosmoprotective compound known to be produced by cyanobac-teria growing in hypertonic milieu, such as seawater.10,63–65
Chlorophyll is one of the major photosynthetic pigments ofcyanobacteria essential for the conversion of sunlight intochemical energy.53 Found at lower abundance is the amino acidarginine ([M + H]+, m/z 175.1187) and sucrose ([M + Na]+, m/z
This journal is ª The Royal Society of Chemistry 2013
365.1046); the lattermay also serve to protect cyanobacteria fromosmosis due to high salt concentration.66–68 [M + Na]+ peak ofsucrose at m/z 365.1046 was assigned by comparing the corre-spondingMS/MS spectrumwith the database spectrum reportedin METLIN (http://metlin.scripps.edu/). To assign glucosylgly-cerol, MS/MS was performed on both the metabolite and acommercially available standard spotted onto A+ agar. Unfor-tunately, neither the metabolite nor the standard produced anyMS/MS fragments to further conrm the assignment. Othermetabolites were assigned based on accurate mass MS/MS,database searches69 (METLIN and LIPID MAPS, http://www.lipidmaps.org/) and comparison with published data.49,50,55
The main lipid peaks in the spectrum (Fig. 1B) correspond tosodium adducts of monogalactosyldiacylglycerol (MGDG) anddigalactosyldiacylglycerol (DGDG) species; these lipids are alsofound as potassium adducts at a lower abundance. Structures ofMGDG and DGDG are shown in Scheme 1A and B, respectively.In agreement with our results – MGDG and DGDG are the mostabundant lipids together with sulfoquinocosyldiacylglycerol(SQDG) and phosphatidylglycerol (PG) in cyanobacteria,including Synechococcus sp. PCC 7002.51,70 Their structures arehighly conserved and they have been shown to play an impor-tant role in photosynthesis.71,72 Oxygenic photosynthesis occursin thylakoid membranes and it has been suggested that theselipids are vital for maintaining the structure and function ofphotosystem II.72
Fig. 2A and B show the distribution of MGDG and DGDGspecies, respectively, observed in the nano-DESI spectrum of
Analyst, 2013, 138, 1971–1978 | 1973
Scheme 1 (A) Structure of MGDG (18:2/16:1) and (B) structure of DGDG (18:2/16:1).
Analyst Paper
Dow
nloa
ded
on 1
2 M
arch
201
3Pu
blis
hed
on 3
0 Ja
nuar
y 20
13 o
n ht
tp://
pubs
.rsc
.org
| do
i:10.
1039
/C3A
N36
716A
View Article Online
Synechococcus sp. PCC 7002. The number of carbons and doublebonds of the fatty acid tail groups in these species are repre-sented by the numbers, e.g. (34:3). For identication, nano-DESIMS/MS experiments were performed while sampling directlyfrom the living Synechococcus sp. PCC 7002 colonies. Fig. 2Cshows anMS/MS spectrum of MGDG (34:3) atm/z 775.5307. Thespectrum contains fragments corresponding to neutral losses ofC18:2, C18:3, C16:0 and C16:1 fatty acids suggesting that thepeak at m/z 775.5307 is comprised of two MGDG species; onecontaining C18:2 and C16:1 fatty acids and another one con-taining C18:3 and C16:0 fatty acids. Even though the position ofthese fatty acids on the glycerol backbone cannot be determined
Fig. 2 (A) Mass spectrum in the m/z 770–785 range showing the most abundantmost abundant [DGDG + Na]+ species. (C) MS/MS spectrum of m/z 775.5307 [MGspectrum of m/z 937.5834 [DGDG (34:3) + Na]+, the blue lines and text show the n
1974 | Analyst, 2013, 138, 1971–1978
based on the MS/MS data, previous studies have shown that theC18 fatty acid is primarily found in the sn-1 position.51,53,73,74
This tentatively assigns the species at m/z 775.5307 as a mixtureof MGDG (18:2/16:1) and MGDG (18:3/16:0). The more abun-dant neutral losses of C18:2 and C16:1, as compared to theneutral losses of C18:3 and C16:0, suggest that the MGDG (18:2/16:1) (Scheme 1A) is the predominant species at m/z 775.5307.
Fig. 2D shows the MS/MS spectrum obtained for m/z937.5834. The neutral losses here correspond to C18:2, C18:3,C16:0 and C16:1 fatty acids and an additional neutral loss ofC6H10O5 corresponding to galactosyl. The neutral loss ofgalactosyl yields m/z 775.5314, assigned as MGDG (34:3), con-rming the assignment ofm/z 937.5834 as a DGDG species. Theratio of the neutral losses of the fatty acids is almost equal,hence the species at m/z 937.5834 can be assigned as a 1 : 1mixture of DGDG (18:2/16:1) (Scheme 1B) and DGDG (18:3/16:0).
Table 1 lists abundant MGDG [M + Na]+ and DGDG [M + Na]+
species assigned by MS/MS analysis, all of them containing atotal of 34 carbons in the fatty acid tails. Less abundant lipidsfound in the spectrum in Fig. 1B were tentatively assigned basedon the accurate mass. All 29 MGDG [M + Na]+, DGDG [M + Na]+
and SQDG [M + 2Na � H]+ species detected on the livingSynechococcus sp. PCC 7002 colonies are summarized in Table 2.To the best of our knowledge, 18 of these intact glycolipids arebeing reported for Synechococcus sp. PCC 7002 for the rst timein this study. Previously, GC-FID has been employed in detec-tion of fatty acid methyl esters derived from Synechococcus sp.PCC 700254 and other cyanobacteria.73,74 These reports support
[MGDG + Na]+ species. (B) Mass spectrum in the m/z 932–947 range showing theDG (34:3) + Na]+, the blue lines and text show the neutral losses and (D) MS/MSeutral losses.
This journal is ª The Royal Society of Chemistry 2013
Table 1 Abundant [MGDG +Na]+ and [DGDG + Na]+ species assigned byMS/MS
m/z Species as assigned by MSMS Ratio
773.5162 MGDG (18:3/16:1)a
775.5313 MGDG (18:2/16:1)a : MGDG (18:3/16:0)a 2 : 1777.5469 MGDG (18:1/16:1)a : MGDG (18:2/16:0)a 2 : 1779.5625 MGDG (18:1/16:0)a
935.5692 DGDG (18:3/16:1)a
937.5854 DGDG (18:2/16:1) : DGDG (18:3/16:0)a 1 : 1939.6012 DGDG (18:1/16:1) : DGDG (18:2/16:0) 1 : 1941.6167 DGDG (18:1/16:0)
a Previously reported in Synechococcus sp. PCC 7002.51
Paper Analyst
Dow
nloa
ded
on 1
2 M
arch
201
3Pu
blis
hed
on 3
0 Ja
nuar
y 20
13 o
n ht
tp://
pubs
.rsc
.org
| do
i:10.
1039
/C3A
N36
716A
View Article Online
the existence of the assigned fatty acid tail groups in Table 1,and the probable fatty acids tail groups in Table 2. SixteenMGDG species of different m/z are listed in Table 2. Only eightof these are found with the DGDG or SQDG head group, sug-gesting a higher diversity of the MGDG lipids. Alternatively, thenarrower distribution of DGDG and SQDG could be attributedto the lower ionization efficiency of these lipids that makesthem difficult to detect using nano-DESI analysis. For example,in this study, we did not detect SQDG (34:4) and also SQDG(34:0) species previously been reported by Montero et al.51 Sincethe relative abundances of the unique MGDG species that do
Table 2 Summary of all detected [MGDG + Na]+, [DGDG + Na]+ and [SQDG +2Na � H]+ species
m/z Assigned speciesNormalizedabundance
693.4545 MGDG (28:2) 0.06695.4699 MGDG (28:1) 0.1721.4862 MGDG (30:2) 0.3723.5010 MGDG (30:1) 0.7725.5170 MGDG (30:0) 0.4747.5010 MGDG (32:3) 0.6749.5167 MGDG (32:2) 8.8751.5327 MGDG (32:1) 7.1773.5162 MGDG (34:4)a 93.7775.5313 MGDG (34:3)a 93.6777.5469 MGDG (34:2)a 100779.5625 MGDG (34:1)a 77.2799.5320 MGDG (36:5) 1.1801.5478 MGDG (36:4) 4.0803.5640 MGDG (36:3) 1.0805.5793 MGDG (36:2) 2.7837.4764 SQDG (32:1)a 0.7839.4941 SQDG (32:0)a 0.9861.4761 SQDG (34:3)a 1.1863.4924 SQDG (34:2)a 0.6865.5072 SQDG (34:1)a 3.5893.5404 SQDG (36:1) 1.3911.5695 DGDG (32:2) 0.4913.5856 DGDG (32:1) 0.5935.5692 DGDG (34:4)a 9.2937.5854 DGDG (34:3)a 8.0939.6012 DGDG (34:2) 5.5941.6167 DGDG (34:1) 4.6963.6007 DGDG (36:4) 0.1
a Previously reported in Synechococcus sp. PCC 7002.51
This journal is ª The Royal Society of Chemistry 2013
not have the corresponding DGDG and SQDG analogs is low(below 1.1%) there is a possibility that similar DGDG and SQDGspecies are below the limit of detection. However, our results arein agreement with previous ndings according to which theratio of MGDG : DGDG : SQDG species in Synechococcus sp. PCC7002 is �100 : 18 : 26 (mol%).54
Spatially resolved nano-DESI analysis
Bacterial colonies produce metabolites to use within or secretefrom the colonies. In this study, metabolites diffusing from thecolony onto agar were detected by scanning the sample underthe nano-DESI probe while recording mass spectra. Fig. 3Ashows a photograph of the nano-DESI probe taken duringanalysis. Spatially resolved chemical analysis was performed bymoving the sample under the nano-DESI probe from the agar tothe colony (3), in the direction shown by the arrow in Fig. 3A.The signal variation as a function of distance from the colonywas obtained by plotting extracted ion chronograms (time-dependent signal for a selected m/z range without chromato-graphic separation) for selected m/z corresponding to analytesof interest. We note that in this study the colony was cultured ina long line to generate one-dimensional gradients on the agarplate. This geometry eliminated the need for two-dimensionalimaging of the sample while providing ample space for exam-ining the technical reproducibility of the data. Data reproduc-ibility was conrmed through acquisition of multiple line scanson the same agar plate.
Fig. 3B shows the spatially resolved data obtained for [MGDG(34:3) + Na]+, [DGDG (34:3) + Na]+ and [SQDG (32:1) + 2 Na�H]+
on living Synechococcus sp. PCC 7002 colonies. In this experi-ment, the probe was rst placed on the A+ agar and subse-quently moved toward and onto the colony. A drastic increase inthe signal is observed for all lipids as soon as the nano-DESIprobe hits the colony, which is marked by the dashed black linein Fig. 3B. The extracted ion chronograms for [arginine + H]+,[glucosylglycerol + Na]+, [sucrose + Na]+ and [chlorophyll a +Na]+ are shown in Fig. 3C. Arginine and sucrose show remark-ably similar patterns on the colony with a continuous increasein signal. In contrast, the traces obtained for chlorophyll a andglucosylglycerol through the Synechococcus sp. PCC 7002 colo-nies show very different behaviors. Specically, the signal ofchlorophyll a increases further into the colony while glucosyl-glycerol shows a rapid increase at the edge of the colony fol-lowed by decay further into the colony. Chlorophyll isassociated with internal membranes and its detection can beattributed to either dead cells or cells that lyse during analysis.
The ability to acquire spatially resolved data from livingbacterial colonies is important for understanding bacterialextracellular signaling and metabolic interaction as well asmetabolic production and nutrition storage within the colony.Molecules secreted onto agar are most likely used for signalingand interactions with other microorganisms and the environ-ment. Nano-DESI is ideally suited for mapping chemicalgradients generated by microbial colonies on agar. This infor-mation is important for understanding interspecies interac-tions or interactions between colonies and mineral surfaces.
Analyst, 2013, 138, 1971–1978 | 1975
Fig. 3 Spatially resolved nano-DESI analysis of a living Synechococcus sp. PCC7002 colony, approx. 3 days old, shown as a function of the distance from thebeginning of the scan. The scan started on agar and moved onto the colony atabout 1.2 mm. (A) A photograph of the nano-DESI probe taken during the scanfrom agar toward the colony. Solvent is brought to the surface through theprimary capillary (1). The analyte molecules are desorbed into the liquid bridge,transferred into the secondary capillary (2), and subsequently ionized by nano-electrospray at the mass spectrometer inlet. Bacterial colonies are green (3) andthe arrow points in the direction of analysis. (B) Signal of [MGDG (34:3) + Na]+
(black), [DGDG (34:3) + Na]+ (red) and [SQDG (32:1) + 2 Na � H]+ (blue) as afunction of position. (C) Signal of arginine (purple), glucosylglycerol (green),sucrose (blue) and chlorophyll a (pink) as a function of position. All traces havebeen normalized to both the total ion current, respective maximum signal and areplotted with a rolling average of 5. The vertical dashed black line in (B) and (C)shows an approximate boundary of the colony.
Fig. 4 Chemical gradient of (A) glucosylglycerol (m/z 277.0893) and (B) sucrose(m/z 365.1051) on A+ agar away from Synechococcus sp. PCC 7002 colonies. Theedges of the colonies are aligned at 0 mm. Green trace: 48 hours old colony, redtrace: 72 hours old colony and blue trace: 144 hours old colony. The inserts showthe molecular structure of glucosylglycerol and sucrose respectively. All traces arenormalized to the total ion current and plotted with a rolling average of 5. Thesucrose intensity is about 4 times less than the intensity of glucosylglycerol.
1976 | Analyst, 2013, 138, 1971–1978
Analyst Paper
Dow
nloa
ded
on 1
2 M
arch
201
3Pu
blis
hed
on 3
0 Ja
nuar
y 20
13 o
n ht
tp://
pubs
.rsc
.org
| do
i:10.
1039
/C3A
N36
716A
View Article Online
To examine to what extent chemical gradients evolve withtime and growth stage, we allowed several Synechococcus sp.PCC 7002 colonies to grow for 1–8 days prior to analysis. Fig. 4shows the results of spatially resolved nano-DESI analysisobtained for 48, 72 and 144 hour-old colonies. Only twometabolites, glucosylglycerol and sucrose, were found both onthe colony and the surrounding agar. We note that the relativeabundance of both glucosylglycerol and sucrose on agar is lessthan 1% of the signal observed on the colony. Extracted ionchronograms of glucosylglycerol [M + Na]+ for the 48, 72 and the144 hours old colonies are shown in Fig. 4A; technical replicatesare shown in Fig. S1 and S2 of the ESI.† The results show that forthe 48 hours old colony, glucosylglycerol was detected only onthe colony. In contrast, an obvious diffusion of glucosylglycerolwas observed for older colonies. Specically, for the 72 hours-old colony the signal of glucosylglycerol extends out at least5 mm away from the colony. This diffusion increases signi-cantly in the 144 hours old colony for which glucosylglycerolsignal is observed even 20 mm away from the colony. Similarresults were obtained for the [M + Na]+ peak of sucrose at m/z365.1051. Fig. 4B shows that sucrose has the same diffusionpattern as glucosylglycerol but is �4 times less abundant.Whatever mechanism caused release of glucosylglycerol andsucrose (whether active excretion or passive cell lysis followed
This journal is ª The Royal Society of Chemistry 2013
Paper Analyst
Dow
nloa
ded
on 1
2 M
arch
201
3Pu
blis
hed
on 3
0 Ja
nuar
y 20
13 o
n ht
tp://
pubs
.rsc
.org
| do
i:10.
1039
/C3A
N36
716A
View Article Online
by diffusion of the metabolites on the surface of agar), theseorganic molecules are a resource that could be catabolized byheterotrophic bacteria as a carbon and energy source. Regard-less of the origin of the observed chemical gradients, the abilityto detect the gradient and observe diffusion of a metabolitefrom living bacteria on media with high salt content is asignicant analytical improvement over the analysis of bulksamples. The spatially resolved nano-DESI analysis of microbialcommunities can aid in further understanding of interaction,signaling and nutrient exchange pathways between bacterialcolonies and within microbial mats. The ambient conditionsduring analysis, in combination with minimal sample prepa-ration, makes nano-DESI an important resource for both quickscreening and more detailed analysis of organic compounds inliving bacterial colonies.
Conclusions
In this study, we demonstrated for the rst time the use ofspatially resolved nano-DESI for the analysis of chemical gradi-ents produced by microbial colonies grown on agar plates.Specically, we used nano-DESI for the analysis of living Syn-echococcus sp. PCC 7002 colonies cultivated on a challenginghigh-salt content A+ agar. Despite the high salt concentration inthe agar, nano-DESI analysis enabled detection of metabolitesproduced and secretedby the Synechococcus sp. PCC7002 colony.High resolution MS andMS/MS were used for identication andstructural characterizationof severalmetabolites andglycolipidsdirectly from the living colonies. Using this approach, we iden-tied several intact glycolipids of Synechococcus sp. PCC 7002that have not been characterized using traditional cell extractionapproaches. Furthermore, spatially resolved nano-DESI enabledcharacterization of chemical gradients of lipids andmetabolitesboth on the colony and on the agar plate. We found that whilemost metabolites were localized on the colony, sucrose andglucosylglycerol – an important osmoprotective molecule –
diffused onto agar. Spatial proling of the chemical gradients ofglucosylglycerol generated by young and aged colonies showedthat diffusion of these metabolites onto agar plates becomesmore pronounced as the colonies age. Although the experimentsperformed in this study were deliberately designed to create one-dimensional gradients on agar, this approach can be readilyexpanded to more complex geometries producing two-dimen-sional maps of molecules on agar plates. Metabolite identica-tion in living colonies is important for understandingmetabolicinteractions in microbial communities, and spatial/temporaldistribution of secreted metabolites. In addition, chemicalcharacterization of bioactive compounds in bacterial coloniescan aid in nding new pharmaceuticals and selection of micro-organisms suitable for use in biofuel production andwastewatertreatment. Future studies will focus on detailed molecular-levelcharacterization of interactions between phototrophic andheterotrophic bacterial colonies grown on agar plates.
Acknowledgements
The research described in this paper is part of the ChemicalImaging Initiative at Pacic Northwest National Laboratory
This journal is ª The Royal Society of Chemistry 2013
(PNNL). It was conducted under the Laboratory DirectedResearch and Development Program at PNNL, a multiprogramnational laboratory operated by Battelle for the U.S. Departmentof Energy (DOE). The work was performed using EMSL, anational scientic user facility sponsored by the DOE's Office ofBiological and Environmental Research and located at PNNL.
References
1 A. Konopka, ISME J., 2009, 3, 1223–1230.2 W. B. Whitman, D. C. Coleman and W. J. Wiebe, Proc. Natl.Acad. Sci. U. S. A., 1998, 95, 6578–6583.
3 A. Konopka, Microbe, 2006, 1, 175–179.4 E. Rodrıguez-Carmona and A. Villaverde, Trends Microbiol.,2010, 18, 423–430.
5 J. E. T. van Hylckama Vlieg, P. Veiga, C. Zhang, M. Derrienand L. Zhao, Curr. Opin. Biotechnol., 2011, 22, 211–219.
6 L. Pilon, H. Berberoglu and R. Kandilian, J. Quant. Spectrosc.Radiat. Transfer, 2011, 112, 2639–2660.
7 A. M. Ruffing, Bioeng. bugs, 2011, 2, 136–149.8 S. W. H. Van Hulle, H. J. P. Vandeweyer, B. D. Meesschaert,P. A. Vanrolleghem, P. Dejans and A. Dumoulin, Chem.Eng. J., 2010, 162, 1–20.
9 P. Cloud, Paleobiology, 1976, 2, 351–387.10 M. Hagemann, FEMS Microbiol. Rev., 2011, 35, 87–123.11 H. W. Paerl, J. L. Pinckney and T. F. Steppe, Environ.
Microbiol., 2000, 2, 11–26.12 D. Sher, J. W. Thompson, N. Kashtan, L. Croal and
S. W. Chisholm, ISME J., 2011, 5, 1125–1132.13 H.W. Paerl and J. L. Pinckney,Microb. Ecol., 1996, 31, 225–247.14 R. Baran, B. P. Bowen and T. R. Northen,Mol. BioSyst., 2011,
7, 3200–3206.15 A. Camilli and B. L. Bassler, Science, 2006, 311, 1113–1116.16 E. A. Shank and R. Kolter, Curr. Opin. Microbiol., 2011, 14,
741–747.17 M. B. Miller and B. L. Bassler, Annu. Rev. Microbiol., 2001, 55,
165–199.18 A. S. Rosado and L. Seldin, World J. Microbiol. Biotechnol.,
1993, 9, 521–528.19 M. Mashego, K. Rumbold, M. De Mey, E. Vandamme,
W. Soetaert and J. Heijnen, Biotechnol. Lett., 2007, 29, 1–16.20 E. Coucheney, T. J. Daniell, C. Chenu and N. Nunan, J.
Microbiol. Methods, 2008, 75, 491–500.21 L. Wu, M. R. Mashego, J. C. van Dam, A. M. Proell, J. L. Vinke,
C. Ras, W. A. van Winden, W. M. van Gulik and J. J. Heijnen,Anal. Biochem., 2005, 336, 164–171.
22 W. K. Erbil, M. S. Price, D. E. Wemmer and M. A. Marletta,Proc. Natl. Acad. Sci. U. S. A., 2009, 106, 19753–19760.
23 Y. Song, N. Talaty, K. Datsenko, B. L. Wanner andR. G. Cooks, Analyst, 2009, 134, 838–841.
24 M. A. Meetani, Y.-S. Shin, S. Zhang, R. Mayer and F. Basile, J.Mass Spectrom., 2007, 42, 1186–1193.
25 J. I. Zhang, A. B. Costa, W. A. Tao and R. G. Cooks, Analyst,2011, 136, 3091–3097.
26 C. Y. Pierce, J. R. Barr, R. B. Cody, R. F. Massung,A. R. Wooltt, H. Moura, H. A. Thompson andF. M. Fernandez, Chem. Commun., 2007, 807–809.
Analyst, 2013, 138, 1971–1978 | 1977
Analyst Paper
Dow
nloa
ded
on 1
2 M
arch
201
3Pu
blis
hed
on 3
0 Ja
nuar
y 20
13 o
n ht
tp://
pubs
.rsc
.org
| do
i:10.
1039
/C3A
N36
716A
View Article Online
27 A. U. Jackson, S. R. Werner, N. Talaty, Y. Song, K. Campbell,R. G. Cooks and J. A. Morgan, Anal. Biochem., 2008, 375, 272–281.
28 T. Greer, R. Sturm and L. Li, J. Proteomics, 2011, 74, 2617–2631.
29 L. S. Eberlin, C. R. Ferreira, A. L. Dill, D. R. Ifa andR. G. Cooks, Biochim. Biophys. Acta, Mol. Cell Biol. Lipids,2011, 1811, 946–960.
30 M. K. Passarelli and N. Winograd, Biochim. Biophys. Acta,Mol. Cell Biol. Lipids, 2011, 1811, 976–990.
31 E. H. Seeley and R. M. Caprioli, Trends Biotechnol., 2011, 29,136–143.
32 J. D. Watrous and P. C. Dorrestein, Nat. Rev. Microbiol., 2011,9, 683–694.
33 D. J. Gonzalez, N. M. Haste, A. Hollands, T. C. Fleming,M. Hamby, K. Pogliano, V. Nizet and P. C. Dorrestein,Microbiology, 2011, 157, 2485–2492.
34 Y.-L. Yang, Y. Xu, R. D. Kersten, W.-T. Liu, M. J. Meehan,B. S. Moore, N. Bandeira and P. C. Dorrestein, Angew.Chem., Int. Ed., 2011, 50, 5839–5842.
35 D. Debois, K. Hamze, V. Guerineau, J.-P. Le Caer,I. B. Holland, P. Lopes, J. Ouazzani, S. J. Seror, A. Brunelleand O. Laprevote, Proteomics, 2008, 8, 3682–3691.
36 G. L. Gasper, L. K. Takahashi, J. Zhou, M. Ahmed, J. F. Mooreand L. Hanley, Anal. Chem., 2010, 82, 7472–7478.
37 P. D. Edirisinghe, J. F. Moore, K. A. Skinner-Nemec,C. Lindberg, C. S. Giometti, I. V. Veryovkin, J. E. Hunt,M. J. Pellin and L. Hanley, Anal. Chem., 2007, 79, 508–514.
38 J. Watrous, N. Hendricks, M. Meehan and P. C. Dorrestein,Anal. Chem., 2010, 82, 1598–1600.
39 J. I. Zhang, N. Talaty, A. B. Costa, Y. Xia, W. A. Tao, R. Bell,J. H. Callahan and R. G. Cooks, Int. J. Mass Spectrom.,2011, 301, 37–44.
40 G. Parsiegla, B. Shrestha, F. Carriere and A. Vertes, Anal.Chem., 2012, 84, 34–38.
41 P. J. Roach, J. Laskin and A. Laskin, Analyst, 2010, 135, 2233–2236.
42 J. Watrous, P. Roach, T. Alexandrov, B. S. Heath, J. Y. Yang,R. D. Kersten, M. van der Voort, K. Pogliano, H. Gross,J. M. Raaijmakers, B. S. Moore, J. Laskin, N. Bandeira andP. C. Dorrestein, Proc. Natl. Acad. Sci. U. S. A., 2012, 109,E1743–E1752.
43 C. M. Rath, T. Alexandrov, S. K. Higginbottom, J. Song,M. E. Milla, M. A. Fischbach, J. L. Sonnenburg andP. C. Dorrestein, Anal. Chem., 2012, 84, 9259–9267.
44 J. Laskin, B. S. Heath, P. J. Roach, L. Cazares andO. J. Semmes, Anal. Chem., 2012, 84, 141–148.
45 P. A. Eckert, P. J. Roach, A. Laskin and J. Laskin, Anal. Chem.,2012, 84, 1517–1525.
46 P. J. Roach, J. Laskin and A. Laskin, Anal. Chem., 2010, 82,7979–7986.
47 G. J. Van Berkel, V. Kertesz and R. C. King, Anal. Chem., 2009,81, 7096–7101.
48 G. J. Van Berkel, V. Kertesz, K. A. Koeplinger, M. Vavrek andA.-N. T. Kong, J. Mass Spectrom., 2008, 43, 500–508.
1978 | Analyst, 2013, 138, 1971–1978
49 R. Baran, B. P. Bowen, N. J. Bouskill, E. L. Brodie,S. M. Yannone and T. R. Northen, Anal. Chem., 2010, 82,9034–9042.
50 N. B. Bennette, J. F. Eng and G. C. Dismukes, Anal. Chem.,2011, 83, 3808–3816.
51 O. Montero, J. Maria Porta, J. Porta, G. Martinez andL. M. Lubian, Phycol. Res., 2011, 59, 147–155.
52 J. E. Graham and D. A. Bryant, J. Bacteriol., 2008, 190, 7966–7974.
53 Y. H. Kim, J. S. Choi, J. S. Yoo, Y. M. Park andM. S. Kim, Anal.Biochem., 1999, 267, 260–270.
54 T. Sakamoto, S. Higashi, H. Wada, N. Murata andD. A. Bryant, FEMS Microbiol. Lett., 1997, 152, 313–320.
55 M. D. Macias-Sanchez, C. Mantell, M. Rodriguez, E. M. de laOssa, L. M. Lubian and O. Montero, J. Supercrit. Fluids, 2007,39, 323–329.
56 D. J. Gonzalez, Y. Xu, Y.-L. Yang, E. Esquenazi, W.-T. Liu,A. Edlund, T. Duong, L. Du, I. Molnar, W. H. Gerwick,P. R. Jensen, M. Fischbach, C.-C. Liaw, P. Straight, V. Nizetand P. C. Dorrestein, J. Proteomics, 2012, 75, 5069–5076.
57 A. U. Jackson, N. Talaty, R. G. Cooks and G. J. Van Berkel, J.Am. Soc. Mass Spectrom., 2007, 18, 2218–2225.
58 G. Huang, G. Li and R. G. Cooks, Angew. Chem., Int. Ed., 2011,50, 9907–9910.
59 C. Van Baalen, Bot. Mar., 1962, 4, 129–139.60 R. Rippka, J. Deruelles, J. B. Waterbury, M. Herdman and
R. Y. Stanier, J. Gen. Microbiol., 1979, 111, 1–61.61 M. Ludwig and D. A. Bryant, Front. Microbiol., 2011, 2, 41.62 I. Lanekoff, B. Heath, A. Liuy, M. Thomas, J. Carson and
J. Laskin, Anal. Chem., 2012, 84, 8351–8356.63 A. Ferjani, L. Mustardy, R. Sulpice, K. Marin, I. Suzuki,
M. Hagemann and N. Murata, Plant Physiol., 2003, 131,1628–1637.
64 K. Marin, J. Huckauf, S. Fulda and M. Hagemann, J.Bacteriol., 2002, 184, 2870–2877.
65 S. Klaehn and M. Hagemann, Environ. Microbiol., 2011, 13,551–562.
66 J. E. Lunn, G. D. Price and R. T. Furbank, Plant Mol. Biol.,1999, 40, 297–305.
67 E. Blumwald, R. J. Mehlhorn and L. Packer, Proc. Natl. Acad.Sci. U. S. A., 1983, 80, 2599–2602.
68 A. C. Cumino, M. Perez-Cenci, L. E. Giarrocco andG. L. Salerno, FEBS Lett., 2010, 584, 4655–4660.
69 J. L. Little, A. J. Williams, A. Pshenichnov and V. Tkachenko,J. Am. Soc. Mass Spectrom., 2012, 23, 179–185.
70 H. Wada and N. Murata, Structure and function ofglycoglycerolipids in plants and bacteria, ed. S. Paul-Andreand M. Norio, Springer, Netherlands, 2004, vol. 6, pp. 65–81.
71 G. Holzl and P. Dormann, Prog. Lipid Res., 2007, 46, 225–243.72 N. Mizusawa and H. Wada, Biochim. Biophys. Acta, Bioenerg.,
2012, 1817, 194–208.73 N. Sato, N. Murata, Y. Miura and N. Ueta, Biochim. Biophys.
Acta, 1979, 572, 19–28.74 H. D. Zepke, E. Heinz, A. Radunz, M. Linscheid and R. Pesch,
Arch. Microbiol., 1978, 119, 157–162.
This journal is ª The Royal Society of Chemistry 2013
