Upload File

high throughput phenotyping of e. coli growth and...

  • Home
  • Documents
  • High Throughput Phenotyping of E. coli Growth and ...sites.bu.edu/britereu/files/2019/07/Graham-second-poster.pdf · Nature Biotechnology, n.d. Print. 5. Zilberstein, Dan, et al
1
Template ID: bluegreenwave Size: 36x48 References Conclusions Intracellular metabolism has a direct influence on extracellular pH. Examining this relaDonship more closely may yield important insight into metabolism and improve metabolic modeling. These insights may improve our understanding of human associated microbes that affect health via manipulaDng extracellular pH such as those in the oral and skin microbiomes 1,2 . Despite the importance of this relaDonship, it has yet to be fully characterized in the well studied model organism E. coli K12. We hypothesized that high throughput phenotyping of E. coli growth and extracellular pH will reveal more about this relaDonship. Using the Biolog PM1 96-well plate, we characterized the growth and extracellular pH when E. coli was cultured aerobically on 96 different carbon sources. Notably we were able to group the results from media with similar chemical composiDons and thus idenDfy trends across and within media. We noDced that an inverse relaDonship exists between growth rate and pH among various sugars we tested such as glucose, whereas growth on organic acids such as aceDc acid yielded a direct relaDonship. Previously studied metabolic models yield similar results 3 , however the lack of high-throughput data leaves much leU to be characterized and analyzed. Future efforts will uDlize high throughput phenotyping to study a larger number of organisms under a more diverse set of condiDons such as anaerobic culDvaDon, and growth under different limiDng resources. In conjuncDon with exisDng metabolic models, this data will improve our understanding of the important relaDonship between intracellular metabolism and extracellular pH. Abstract Methods Results High Throughput Phenotyping of E. coli Growth and Extracellular pH Graham Kulig 1 , David Bernstein 2 , Alan Pacheco 3 , Meghan Thommes 2 , Daniel Segre 2,3,4 1 Elon University; Boston University BRITE BioinformaDcs REU Program, Summer 2016, 2 Boston University, Biomedical Engineering Department, 3 Boston University, BioinformaDcs Program, 4 Boston University, Biology Department Indicator I.F. I.F. + Indicator 1. 2. 3. 4. 5. 1. Combine inoculaDng fluid (IF) and pH indicator (a mixture of Bromocresol Purple and Phenol Red). 2. Add 100uL indicator + IF to every well, let equilibrate. 3. Culture a fresh plate of E. coli from frozen stock. 4. Inoculate cells in test tube, dilute to OD600 of 0.3, add 20 uL to every well. 5. Insert plate into spectrophotometer, set to read for 90 hours, taking measurements every 0.5 hour at 430, 570, and 800 nm. • In order to obtain accurate pH measurements and not just esDmates, we removed the plate 3 Dmes during the experiment (and once at the beginning and end) to measure pH with a pH meter. Introduction Currently, there is not enough informaDon regarding the relaDonship between cellular metabolism and extracellular pH. New informaDon can be used to beger understand the effect of microbes on the human body by updaDng current metabolic models. High throughput phenotyping (HTP) is one of the most powerful tools currently aiding in microbe characterizaDon and therefore metabolic model construcDon 4 . In order to obtain HTP, the Biolog Phenotypic Microarray plate with 96 different carbon sources was used. As one of the most well characterized bacterial models, and known for its relaDvely simple lab reproducibility, Escherichia coli K12 was chosen as a suitable microbe. Furthermore, E. coli has the ability to respire with or without oxygen and has a wide pH range for survival, making the characterizaDon of E. coli a very versaDle tool for comparisons to many other microbes. Studying the hydrogen ion flux as E. coli metabolizes has painted a clearer picture of how an intracellular funcDon affects the extracellular environment of a microbe. • We observed that, experimentally and on a model basis, intracellular metabolism and extracellular pH have an inverse relaDonship when E. coli is introduced to sugars such as glucose (Fig. 2a). We observed the opposite effect when E. coli was grown on acidic media such as fumaric acid (Fig 2b). As growth rate increased, so too did the pH, revealing a direct relaDonship. Some media displayed both direct and inverse relaDonships, such as Galactonic acid-g-Lactone (Fig. 2c). The exponenDal growth phase for Galactonic acid-g-Lactone was characterized by a drop in pH (inverse) as it fed on Lactone, and then a rise in pH as E. coli switched metabolites to Galactonic acid, resulDng in an increasing pH (direct). SDll, other media such as Uridine grew for all 90 hours and changed pH by a measure of only ~ 0.2 (Fig. 2d). • InteresDngly enough E. coli prefers an opDmal pH of ~ 7 to proliferate 5 , yet when grown on a variety of carbon sources, E. coli alters its pH in a myriad of ways as opposed to remaining at a stable, uniform pH. Going forward, we plan to further invesDgate the metabolic constraints that underlie this phenomenon. This data will be used to conDnue to improve metabolic models such that they will more accurately predict intracellular metabolism in relaDon to extracellular pH. Future HTP will include the characterizaDon of other microbes that have an effect on human health. Acknowledgements 1. Lambers, H., Piessens, S., Bloem, A., Pronk, H. and Finkel, P. (2006), Natural skin surface pH is on average below 5, which is beneficial for its resident flora. InternaDonal Journal of CosmeDc Science, 28: 359–370. doi:10.1111/j.1467-2494.2006.00344.x 2. Loesche, W J. “Role of Streptococcus Mutans in Human Dental Decay.”Microbiological Reviews 50.4 (1986): 353–380. Print. 3. Adadi, Roi et al. “PredicDon of Microbial Growth Rate versus Biomass Yield by a Metabolic Network with KineDc Parameters.” Ed. Nathan D. Price. PLoS Computa>onal Biology 8.7 (2012): e1002575. PMC. Web. 19 July 2016. 4. Christopher S Henry, Maghew DeJongh, Aaron A Best, Paul M Frybarger, Ben Linsay, and Rick L Stevens. High-throughput Genera>on, Op>miza>on and Analysis of Genome-scale Metabolic Models. Rep. N.p.: Nature Biotechnology, n.d. Print. 5. Zilberstein, Dan, et al. "Escherichia coli intracellular pH, membrane potenDal, and cell growth." Journal of bacteriology 158.1 (1984): 246-252. This work was funded by the Boston University BioinformaDcs BRITE REU program. Specifically, I’m very gracious for Gary Benson’s instrucDon over the BRITE 2016 summer program. A very special thank you is extended to David Bernstein, Alan Pacheco, and Meghan Thommes for their invaluable guidance throughout this process. Fig 2: E.coli pH and absorbance at OD800 nm over 90 hours. A) Glucose growth corresponded with a drop in extracellular pH B) Fumaric acid growth corresponded with an increase in extracellular pH C) Galactonic Acid-g-Lactone exhibited interesDng pH dynamics with an iniDal decrease followed by an increase D) Uridine growth corresponded with minimal pH change Spectrophotometer Fig 1: A) Graph of pH indicator absorpDon spectrum. • The 430 and 570 nm readings were used to measure pH using the pH indicator (dye). This is based on the pH indicator dye: the media transiDons from yellow to purple as the pH increases. Therefore as the pH increases, the purple absorbance is decreasing and the yellow absorbance is increasing. So the raDo of yellow to purple absorbance will increase as the pH increases. • 800 nm reading was used to measure bacterial growth as the absorbance at this wavelength was not affected by pH changes, and thus is directly proporDonal to biomass (growth) B) Photograph of plate following experiment • Yellow wells resulted from carbon sources that drove E. coli to lower the pH while purple wells resulted from carbon sources that did the inverse. A B A B C D Fig 3: The pH esDmate based on the pH indicator dye (red) is ploged against actual pH measurements (light blue). The pH indicator dye was a relaDvely accurate tool for predicDng pH. A regression line (black) shows the predicted pH response based on esDmate data. EsFmate Measurement Regression

Upload: others

Post on 05-Jul-2020

1 views

Category:

Documents


0 download

Report

TRANSCRIPT

Page 1: High Throughput Phenotyping of E. coli Growth and ...sites.bu.edu/britereu/files/2019/07/Graham-second-poster.pdf · Nature Biotechnology, n.d. Print. 5. Zilberstein, Dan, et al

TemplateID:bluegreenwaveSize:36x48

References

Conclusions Intracellularmetabolismhas adirect influenceonextracellularpH. Examining this relaDonshipmore closelymay yield important insight intometabolismand improve metabolicmodeling.TheseinsightsmayimproveourunderstandingofhumanassociatedmicrobesthataffecthealthviamanipulaDngextracellularpHsuchasthoseintheoralandskinmicrobiomes1,2.Despitethe importanceof this relaDonship, it has yet to be fully characterized in thewell studiedmodelorganismE.coliK12.WehypothesizedthathighthroughputphenotypingofE.coligrowthandextracellularpHwillrevealmoreaboutthisrelaDonship.UsingtheBiologPM196-wellplate,we characterized the growth and extracellular pH when E. coli was cultured aerobically on 96different carbon sources.Notablywewere able to group the results frommediawith similar chemical composiDons and thus idenDfy trends across andwithinmedia.WenoDced that aninverserelaDonshipexistsbetweengrowthrateandpHamongvarioussugarswetestedsuchas glucose, whereas growth on organic acids such as aceDc acid yielded a direct relaDonship.Previouslystudiedmetabolicmodelsyieldsimilarresults3,howeverthelackofhigh-throughput data leaves much leU to be characterized and analyzed. Future efforts will uDlize high throughput phenotyping to study a larger number of organisms under amore diverse set of condiDons such as anaerobic culDvaDon, and growth under different limiDng resources. In conjuncDonwith exisDngmetabolicmodels, this data will improve our understanding of theimportantrelaDonshipbetweenintracellularmetabolismandextracellularpH.

Abstract

Methods

Results

HighThroughputPhenotypingofE.coliGrowthandExtracellularpHGrahamKulig1,DavidBernstein2,AlanPacheco3,MeghanThommes2,DanielSegre2,3,4

1ElonUniversity;BostonUniversityBRITEBioinformaDcsREUProgram,Summer2016,2BostonUniversity,BiomedicalEngineeringDepartment,3BostonUniversity,BioinformaDcsProgram,4BostonUniversity,BiologyDepartment

IndicatorI.F.

I.F.+Indicator

1.

2.

3.

4.

5.1. Combine inoculaDng fluid (IF) and pH indicator (amixture ofBromocresolPurpleandPhenolRed).2.Add100uLindicator+IFtoeverywell,letequilibrate.3.CultureafreshplateofE.colifromfrozenstock.4.Inoculatecellsintesttube,dilutetoOD600of0.3,add 20 uL to everywell.5. Insert plate into spectrophotometer,set toread for90hours, takingmeasurementsevery0.5hourat430,570,and800nm.• In order to obtain accurate pH measurements and not justesDmates,we removed theplate 3Dmesduring the experiment(and once at the beginning and end) tomeasure pH with a pHmeter.

Introduction Currently, there is not enough informaDon regarding the relaDonship between cellularmetabolism and extracellular pH. New informaDon can be used to beger understand theeffect of microbes on the human body by updaDng current metabolic models. Highthroughputphenotyping(HTP)isoneofthemostpowerfultoolscurrentlyaidinginmicrobecharacterizaDon and thereforemetabolicmodel construcDon4. In order to obtain HTP, theBiologPhenotypicMicroarrayplatewith96differentcarbonsourceswasused.Asoneofthemost well characterized bacterial models, and known for its relaDvely simple labreproducibility,Escherichia coli K12was chosenasa suitablemicrobe. Furthermore, E. colihastheabilitytorespirewithorwithoutoxygenandhasawidepHrangeforsurvival,makingthecharacterizaDonofE.coliaveryversaDletoolforcomparisonstomanyothermicrobes.StudyingthehydrogenionfluxasE.colimetabolizeshaspaintedaclearerpictureofhowanintracellularfuncDonaffectstheextracellularenvironmentofamicrobe.

•  Weobservedthat,experimentallyandonamodelbasis,intracellularmetabolismand extracellular pH have an inverse relaDonship when E. coli is introduced tosugarssuchasglucose(Fig.2a).WeobservedtheoppositeeffectwhenE.coliwasgrownonacidicmediasuchasfumaricacid(Fig2b).Asgrowthrateincreased,sotoodidthepH,revealingadirectrelaDonship.Somemediadisplayedbothdirectand inverse relaDonships, such as Galactonic acid-g-Lactone (Fig. 2c). TheexponenDal growth phase for Galactonic acid-g-Lactonewas characterized by adropinpH(inverse)asitfedonLactone,andthenariseinpHasE.coliswitchedmetabolites toGalactonic acid, resulDng in an increasingpH (direct). SDll, othermediasuchasUridinegrewforall90hoursandchangedpHbyameasureofonly~0.2(Fig.2d).

•  InteresDnglyenoughE.coliprefersanopDmalpHof~7toproliferate5,yetwhengrownonavarietyofcarbonsources,E.coli altersitspHinamyriadofwaysasopposedtoremainingatastable,uniformpH.Goingforward,weplantofurtherinvesDgate themetabolic constraints that underlie this phenomenon. This datawill beused to conDnue to improvemetabolicmodels such that theywillmoreaccuratelypredict intracellularmetabolisminrelaDontoextracellularpH.FutureHTP will include the characterizaDon of other microbes that have an effect onhumanhealth.

Acknowledgements

1.  Lambers, H., Piessens, S., Bloem, A., Pronk, H. and Finkel, P. (2006), Natural skin surface pH is onaveragebelow5,whichisbeneficialforitsresidentflora.InternaDonalJournalofCosmeDcScience,28:359–370.doi:10.1111/j.1467-2494.2006.00344.x

2.  Loesche,W J. “Role of StreptococcusMutans in Human Dental Decay.”Microbiological Reviews 50.4(1986):353–380.Print.

3.  Adadi, Roi et al. “PredicDonofMicrobialGrowthRate versus Biomass Yield by aMetabolicNetworkwithKineDcParameters.”Ed.NathanD.Price.PLoSComputa>onalBiology8.7(2012):e1002575.PMC.Web.19July2016.

4.  ChristopherSHenry,MaghewDeJongh,AaronABest,PaulMFrybarger,BenLinsay,andRickLStevens.High-throughputGenera>on,Op>miza>onandAnalysisofGenome-scaleMetabolicModels.Rep.N.p.:NatureBiotechnology,n.d.Print.

5.  Zilberstein,Dan,etal."EscherichiacoliintracellularpH,membranepotenDal,andcellgrowth."Journalofbacteriology158.1(1984):246-252.

This work was funded by the Boston University BioinformaDcs BRITE REU program.Specifically, I’mverygracious forGaryBenson’s instrucDonover theBRITE2016summerprogram. A very special thank you is extended to David Bernstein, Alan Pacheco, andMeghanThommesfortheirinvaluableguidancethroughoutthisprocess.

Fig2:E.colipHandabsorbanceatOD800nmover90hours.A)GlucosegrowthcorrespondedwithadropinextracellularpHB)FumaricacidgrowthcorrespondedwithanincreaseinextracellularpHC)GalactonicAcid-g-LactoneexhibitedinteresDngpHdynamicswithaniniDaldecreasefollowedbyanincreaseD)UridinegrowthcorrespondedwithminimalpHchange

Spectrophotometer

Fig1:A)GraphofpHindicatorabsorpDonspectrum.•  The430and570nmreadingswereusedtomeasurepHusingthepHindicator(dye).ThisisbasedonthepHindicatordye:themediatransiDonsfromyellowtopurpleasthepHincreases.ThereforeasthepHincreases,thepurpleabsorbanceisdecreasingandtheyellowabsorbanceisincreasing.SotheraDoofyellowtopurpleabsorbancewillincreaseasthepHincreases.

•  800nmreadingwasusedtomeasurebacterialgrowthastheabsorbanceatthiswavelengthwasnotaffectedbypHchanges,andthusisdirectlyproporDonaltobiomass(growth)

B)Photographofplatefollowingexperiment•  Yellowwells resulted from carbon sources that droveE. coli to lower thepHwhile purplewellsresultedfromcarbonsourcesthatdidtheinverse.

A B

A B

C D

Fig3:ThepHesDmatebasedonthepHindicatordye(red)isplogedagainstactualpHmeasurements(lightblue).ThepHindicatordyewasarelaDvelyaccuratetoolforpredicDngpH.Aregressionline(black)showsthepredictedpHresponsebasedonesDmatedata.

EsFmate

Measurement

Regression

Remote Sensing for Phenotyping - WordPress.com

TLI 2012: Drought phenotyping for legumes

FLOWCHART of HDL Phenotyping Process

European initiatives on phenotyping platforms

Case Study: pea phenotyping

and Phenotyping Research - BFA

Primer - sites.bu.edu

Center for Computational Molecular Phenotyping

High-throughput hyperdimensional vertebrate phenotyping

3g Cassavabase workshop: manage phenotyping

Next-generation Phenotyping Using Interoperable Big · PDF fileNext-generation Phenotyping Using Interoperable Big Data ... Vaccine safety surveillance ... Next-generation Phenotyping

HIGH-THROUGHPUT PHENOTYPING

INTUSSUSCEPTION - sites.bu.edu

CORDIS i FP6: 4 koraka do uspjeha Fabio ZILBERSTEIN

Next-generation phenotyping

Evaluation Considerations for EHR-Based Phenotyping

Genomic Selection & Precision Phenotyping

IMPACT - sites.bu.edu

Akshat Kumar & Shlomo Zilberstein. MAP Estimation for graphical …dechter/courses/ics-295/winter... · 2011. 1. 30. · Akshat Kumar & Shlomo Zilberstein. MAP Estimation for graphical

(Forward Scattering Phenotyping) using BARDOT

European Plant Phenotyping Network · PDF fileJoint research activities Sensor development Good Phenotyping Practics IT for phenotyping Massonnet et al., 2010 Fiorani et al., 2012

Reverse Phenotyping - cragenomica.es

Latent Space Phenotyping: Automatic Image-Based Phenotyping … · 2020. 11. 19. · based phenotyping, we propose the Latent Space Phenotyp-ing (LSP), a novel imageanalysis technique

Musa base phenotyping workflow demo

ab leary 93 - sites.bu.edu

Drug Metabolism and Reaction-Phenotyping

SQL Explore 2012 - Michael Zilberstein: ColumnStore

Effective Wolfdog Phenotyping - Dark Forest Kennelsthedarkforestgroup.com/files/Effective_Wolfdog... ·  · 2014-10-31Effective Wolfdog Phenotyping By: ... figures will determine

Combining high-throughput phenotyping and genome-wide ... · demonstrate that high-throughput phenotyping has the potential to replace traditional phenotyping and serves as a novel

Book Phenotyping and Genotyping NTM

TPRI competition conference July 2018 - sites.bu.edu

PlantScreen™ Phenotyping Systems

RADIOGRAPHIC PHENOTYPING OF COPD

Annals of Epidemiology - sites.bu.edu

Adoption Delay in a Standards War - sites.bu.edu