identifying microorganisms

8/12/2019 Identifying Microorganisms

1/8!""#Nature Publishing Group

P E R S P E C T I V E S

thermodynamically unstable substances.

Perhaps the most ecologically importanttypes of enzymatic reactions are those thatcatalyse oxidation/reduction reactions between

electron donors and electron acceptors.These reactions allow microorganisms to

generate metabolic energy, and to surviveand grow. Microorganisms procreate bycarrying out complex, genetically regulated

sequences of biosynthetic and assimilativeintracellular processes. Each daughter cell

has essentially the same macromolecularand elemental composition as its parent.Therefore, the integrated metabolism of all

nutrients is implicit in microbial growth.The growth and survival of microorganismsdrives the geochemical cycling of the ele-

ments, detoxifies many organic and inorganiccontaminants, makes essential nutrients

present in the biomass of one generationavailable to the next, and maintains the

conditions required by other inhabitants ofthe biosphere1012 (TABLE 1).

This article presents a perspective on pastand current attempts to discover the identity

of microorganisms that are responsible forcatalysing key biogeochemical reactions in insitusoils, sediments and waters. The tradi-tional challenges to reaching this goal arediscussed, as are recent innovations to over-

come these challenges. Insights are sought bycontrasting ways of documenting causalityin medical microbiology Kochs postu-

lates with those of environmental micro-biology.

setting (for example, anaerobic peatlands,

oceanic hydrothermal vents, soil humus anddeep subsurface sediments) features its ownset of resources that can be physiologically

exploited by microorganisms.The free-energy-governed interactions between these resources,

their settings, the microorganisms themselvesand ~3.5 billion years of evolution are prob-

ably the source of the metabolic diversity ofthe microbial world8. Microorganisms arethe primary agents of geochemical change,and the global biomass of prokaryotes is

approximately equal to that of all other(eukaryotic) life forms9.

Their small size, ubiquitous distribution,high specific surface area,potentially high rateof metabolic activity, physiological respon-

siveness, genetic malleability, potentially rapidgrowth rate and unrivalled enzymatic andnutritional diversity cast microorganisms in

the role of recycling agents for the biosphere.Enzymes accelerate reaction rates between

!"#$%&'$ ) *+%,-.+,-$ /0,1-$2,3&%4 $25/6

&37 /&'+ 7&4 23 /0/%4 +&"2$&$ $+%,-.+,-$

$+/ .1,"/6 52'%,,%.&32#5# +&0/ "//3

%/#8,3#2"1/ 9,% 5&23$&2323. $+/ "2,#8+/%/:

;/#82$/ $+/ '%-'2&1 8&%$ $+&$ $+/4 81&4 23 $+/

'4'123. ,9 3-$%2/3$# 23 +&"2$&$# #-'+ #,21#6

#/725/3$# &37 "#$&'1/# $+&$+&0/ $%&72$2,3&114 258/7/7 9-37&5/3$&1

52'%,"2&1 /',1,.4 23?-2%2/# &%/ 3,< 42/1723.

$, $/'+32'&1 &70&3'/5/3$# $+&$ +&0/

258,%$&3$ 8&%&11/1# 23 5/72'&1 52'%,"2,1,.4:

*+/ 8&'/ ,9 3/< 72#',0/%2/# $+&$ 7,'-5/3$

/',1,.2'&1 8%,'/##/# &37 $+/2% '&-#&$20/

&./3$#


2/8!""#Nature Publishing Group44 0) MAY 2005) VOLUME 3 www.nature.com/reviews/micro


medium that is designed to select a small

subset of the initial community. The logicbehind enrichment culturing involves devis-ing growth conditions that allow particular

members of the community to grow andeventually dominate within the mixed popu-

lation that was initially present. For instance,if one is interested in finding aerobic micro-organisms that can grow on benzene (oxidiz-

ing it to CO2

and incorporating the substratecarbon into new cells), then the enrichmentmedium would contain benzene as the sole

carbon and energy source,and oxygen as theelectron acceptor. A 1 g soil inoculum can

contain 40,000 species7,14, although only asmall percentage of these would be expectedto grow on benzene.After a 12 week incuba-

tion, benzene degraders would become dom-inant. Then,by plating small volumes of the

enriched populations onto benzene growth

medium solidified with agar, individualcolonies of benzene degraders can be picked,

further purified,isolated and characterizedusing appropriate physiological,biochemicaland/or genetic procedures.

It is important to note that naturallyoccurring microbial communities used asinocula typically consist of uncharacterized,

highly diverse populations (see above), whichusually are morphologically non-distinct

rods and cocci. Each cell has the geneticpotential to carry out a multitude of meta-bolic processes,although conditional regula-

tion can severely limit gene expression inthe natural environment. Furthermore, dor-mancy (or very slow growth) is the norm for

most cells in nature, because all habitats arenutrient limited15. Therefore, the presence in

Enrichment culturing from nature

Some of the earliest and most influentialinvestigations in the history of environmentalmicrobiology relied on enrichment culturing

strategies1,13 to identify and isolate individualmicrobial cultures capable of carrying out

novel metabolic processes, such as growth onammonia as an energy source, fixation ofatmospheric nitrogen into cell protein,and

the use of unusual (perhaps pollutant)organic compounds as carbon and energysources or final electron acceptors. FIGURE 2

provides an integrated overview of the proce-dures used in environmental microbiology to

conduct such inquiries, and how to interpretthe resultant data.

Enrichment culturing uses a sample of a

naturally occurring microbial community asan inoculum for laboratory-prepared growth

*&"1/ A ) Examples of physiological processes catalysed by microorganisms in biosphere habitats

9,"$%-- :)'(,% "/ &,"$%-- ;7&.$)* 0)1.')'

K ># PA

,-%.+/#"."0-%&

M43$+/$2' ,%.&32' ',58,-37# ;/',58,#2$2,36 J>K9,%5&$2,3 !11 PK6PX

G/$%,1/-5 +47%,'&%",3# ;/',58,#2$2,36 J>K9,%5&$2,3 !11 PN@-/1 &772$20/# CE*YZD ;/',58,#2$2,36 J>

K9,%5&$2,3 M16 MK9,%5&$2,3 M16 M

KO &37 \S

X"/',5/ \

K. M# RN6RP

;/32$%292'&$2,3 \>X

O 2# -#/7 &3 /1/'$%,3 &''/8$,% &37 ',30/%$/7 $, \K. M16 M< RR6RT

23*453# ()(*+

MK,L27&$2,3 MKO &37 MW "/',5/ M>NKO >#6 @

NKO %/7-'$2,3 M>

NKO 2# -#/7 &3 /1/'$%,3 &''/8$,% &37 ',30/%$/7 $, MW &37 MKO >#6 M

KO6 32$%2$/F \>

XO 6 32$%&$/F >#6 ,'/&3 #/725/3$#F

>




NATURE REVIEWS ) MICROBIOLOGY VOLUME 3 ) MAY 2005 ) 44 1

of microbiological information become sim-plified (that is, farther removed from the field

study site), the risk of obtaining ecologicallyquestionable information increases. If inves-tigators take the path of cultivation-based

inquiry,not only might ecologically insignifi-cant organisms be isolated,but the laboratory

conditions selected for subsequent testingmight also fail to activate ecologically relevantgenes (FIG.2). During the past two decades,

great efforts have been made to developmethods for non-cultivation-based inquiry

characterize, understand and duplicate fieldconditions in the laboratory undermines the

acceptance of laboratory measurements per-formed on field samples as valid surrogates fortrue, in situfield processes (FIG. 2).

When measurements are performed in

situwithin a field study site (such as soil, a

lake or an ocean), the complexity of the sys-tem is high, and the information obtained(such as geochemical parameters and field

observations) is directly applicable to thesystem under study. However, as the sources

an environmental sample of a particular

organism that is capable of a particularprocess cannot be taken as evidence that the

process is occurring in situ8,16.

Striving for ecological significance

Implicit in enrichment procedures is the abilityof microorganisms to respond and changewhen subjected to environmental perturba-

tions. The nature of microbial responsivenessduring enrichment culturing is clear: resusci-tation from dormancy and growth of (often)

minor populations during incubation periodslasting daysyears.But even if relatively brief

incubations preclude shifts in populationdynamics owing to growth and death,microorganisms still respond to environ-

mental change. For instance, intricate bio-chemical signalling pathways allow cells tosense and respond to key nutrients (for

example, light,oxygen,other electron accep-

tors and carbon sources17), stress (such asacid, oxidative damage or inhibitory sub-stances18), and cell-to-cell signalling molecules(quorum sensing pheromones19). The time

frames for these responses range fromnanoseconds (light) to milliseconds (oxygen,

toxicity) to minutes (enzyme synthesis) orhours (sporulation).

This remarkable propensity of populations

within naturally occurring microbial com-munities to change is a blessing for micro-biologists practising enrichment culture.

However, it is a major impediment for thoseseeking to interpret physiological and ecologi-

cal measurements performed on laboratory-incubated environmental samples such aswater, soils or sediments. The validity of mea-

surements conducted on microbial commu-nities removed from their original field set-ting is uncertain, because we cannot be sure

that conditions imposed on the nativemicroorganisms (post-sampling and incuba-

tion) have not quantitatively or qualitativelyaltered these populations and their physiolog-ical reactions. Potentially misleading bottle

effectsare implicit in all measurements per-formed on sampled microbial communi-

ties20,21. This situation has been likened to theHeisenberg Uncertainty Principle in quantumchemistry, which formally recognizes the

mutual exclusivity of simultaneous determi-nation of the position and momentum of anelectron8. When one begins in a field site and

strives to dissect site-derived samples, thecloser microorganisms in the community are

examined, the more likely the resultant infor-mation is to suffer from artefacts imposed bythe sampling and/or measurement proce-

dures. The investigators inability to obtaindisturbance-free field samples and to fully

Systemcomplexity

and probabilityof ecologicalrelevance ofinformation

Enquiry pathand sources of

information

Informationand/or

database

Ecologicalvalidation

Biogeochemicalprocess or

disease

IVKP

Lipids

Cell components:

DNA

RNA

Protein

Physiologicalassays

Metabolites

Enzyme structureand function

Taxonomy

Genetics

Sample

Field-derivedcell images

HighCultivation

approaches

Selective

medium

Selectivemedium

Non-cultivationapproaches

Fixedsample

Extractedsample

Low

NH, BG

MMiP

BM

P, M, T, BM,MB, B, OMICS

MR

M, BM, B, T, P,MB, OMICS, G

BM, MB, T,B, OMICS

Culture

Field study site

@2.-%/ K ) ="5%* /", '0% 6%#%,)'."# )#5 .#'%,&,%')'."# "/ %#3.,"#4%#')* 4.$,"1."*"6.$)*

.#/",4)'."#> 2.'0 %4&0)-.- "# /.%*5 ,%*%3)#$% )#5 %$"*"6.$)* 3)*.5)'."# "/ 5)')8 J,1-53 A

8%,027/# & #'&1/ 9,% /0&1-&$23. $+/ 12=/14 /',1,.2'&1 %/1/0&3'/ ,9 239,%5&$2,3 23 $+/ ,$+/% $+%// ',1-53#:

_2$+ /&'+ #-''/##20/ 5/$+,7,1,.2'&1 #$/8 &EBJM6 ./3,52'#6 8%,$/,52'#6 5/$&",1,52'#6 $%&3#'%28$,52'#6 &37 #, ,3F

G6 8+4#2,1,.2'&1 '+&%&'$/%2Q&$2,3F *6 $&L,3,52' '+&%&'$/%2Q&$2,3:




geochemical processes in field habitats are noteasy to discern, and because such processesgenerally proceed regardless of intervention.

Culturability is probably the other mainfactor that has allowed medical micro-

biologists to flourish while environmentalmicrobiologists have perhaps fallen out ofstep.Culturability is a direct reflection of two

interacting issues: the relative ratio of target tonon-target organisms in the initial inoculum;

and an ability to accurately simulate thenative habitat in media. When Robert Kochembarked down the cultivation-based path

(FIG. 2), his initial field sample (blood from adiseased sheep) was essentially a monoculturecontaining a large number of regular, rod-

shaped,colorless, immotile structures26 thatwere microscopically discernible.Compare this

to the vast,confusing zoo of candidates (forexample,40,000 species and 109 cells per gramof soil) that confronts a soil microbiologist.

Furthermore, Koch found that the blood-borne bacilli readily reproduced on solidmedia containing nutrient gelatin or boiled

potato26. Easy culturability is not a given inmedical microbiology (for example,Treponemapallidum (syphilis) and Mycobacterium leprae(leprosy) cannot yet be grown in vitro28).However, the uniform, stable,globally distrib-

uted nutrient conditions of the human bodyare undeniably easy to mimic in growth

(for example, the use of 16S rRNA in investi-gations of microbial diversity). The resultantmolecular, biomarker and genomic infor-

mation has been revolutionary in terms ofthe insights that have been attained5,2224.

The non-cultivation-based procedures havesucceeded in generating ecologically signifi-cant information. However, both cultiva-

tion-based and non-cultivation-basedenquiries are imperfect and biased25. It isfor this reason that ecological validation is

necessary (FIG.2).

Environmental Kochs postulates?

In 1884, Robert Koch26 developed funda-mental criteria for proving that a particular

microorganism (Bacillus anthracis) wasresponsible for a particular process

(anthrax disease) in a particular habitat(sheep). This generalized four-step guide-line, known as Kochs postulates, is as fol-

lows: (i) the microorganism should befound in all cases of the disease in question,and the microorganisms distribution in the

body should be in accordance with thelesions observed; (ii) the microorganism

should be grown in pure culture in vitro(oroutside the body of the host) for severalgenerations; (iii) when such a pure culture is

inoculated into susceptible animal species,the typical disease must result; and (iv) the

microorganism must again be isolatedfrom the lesions of such experimentallyproduced disease.

Kochs postulates have been the goldstandard in medical microbiology for estab-

lishing causality, and have survived intact tothe present, with minor modifications thataccommodate recent molecular biological

techniques27,28. However,for microbiologistsconcerned with ecological processes, linkinga microorganisms identity to its activity in

its habitat has, with several exceptions,proven difficult.Below,I suggest why medical

microbiologists have so far been more suc-cessful than environmental microbiologistsin identifying causative agents.

TABLE 2 compares and contrasts, for medicaland environmental microbiology, four key

factors that influence the determination ofcausality: the complexity of the habitat andits inhabitants,the process of interest, identi-

fying a potential causative agent, and linkingthis agent to the process of interest in thefield. As stated in TABLE 2, human disease is

readily recognized in the field and has anenormous detrimental impact.Therefore, the

impetus for understanding and intervening isalso enormous. By contrast, the impetus fordiscovery and management of ecologically

important biogeochemical reactions hasbeen less pressing perhaps because bio-

*&"1/ K ) Contrasts between information on causality in medical and environmental microbiology

=%5.$)* 4.$,"1."*"67 ?#3.,"#4%#')* 4.$,"1."*"67

9#"-08 %: 5"$-0"08 803.-+.

S-5&3# &%/ .1,"&114 72#$%2"-$/7 &37 /0,1-$2,3&%214 #$&"1/ M,21#6 #/725/3$# &37


5/8




microbiology.However, Kochs postulates are

only applicable in limited contexts because theactive microorganisms must be cultivated,andmust initially be present in low numbers or be

absent from the inoculated habitat.The next six entries in TABLE 3a illustrate

the foundations and later developments inmicroscopy-based attempts to link identity toactivity without using Kochs postulates.

Microscopy and microautoradiography wereinitially used to see which cells in mixed micro-bial communities incorporated radiolabelled

substrates.Later, microautoradiography wascombined with cell-specific probing: fluores-cent antibodies targeting cell-surface antigens

of cultivated bacteria or fluorescent oligo-nucleotides targeting sequences of taxo-

nomically revealing rRNA, often derived fromuncultivated microorganisms.Recent efforts

ecological relevance, then the microorgan-

isms that are eventually isolated are morelikely to be those that are active in nature.Second, as analyses of field-fixed samples

deliver increasingly sophisticated informa-tion about expressed genes and proteins

used by microorganisms in their native habi-tats, inferences can be made about in situphysiological conditions, carbon substrates

and nutritional needs. Such informationcan, in turn, guide the design of media sothat new microorganisms can be cultivated.

Last, the several paths of information flowfor validating data shown in FIG. 2 need to be

more widely used. These validation paths are:following Kochs postulates by inoculatingfield sites,use of pure-culture-derived omics

biomarkers to guide analyses performed onextracted samples, and the use of microscopy

and biomarker probes to confirm the field

relevance of information from both pure cul-tures and extracted samples.

Selected examples of past and current

investigations aimed at linking identity ofmicroorganisms to their field activity are

shown in TABLES 3a,b. The entries were chosento be representative of the types of strategies,techniques, challenges and breakthroughs that

have occurred in environmental microbiologyover the past several decades. The emphasis ison identifying microorganisms and being sure

that they catalyse biogeochemical reactionsin

situ in real-world field sites containing soil,sediment or water.The first two entries (sym-

biotic nitrogen fixation and biodegradation oftrichloroethene in contaminated groundwater)

reveal that Kochs postulates can be powerfuland insightful when used in environmental

*&"1/ X" ) Selected examples of efforts to identify microorganisms responsible for field biogeochemical processes

9,"$%-- =.$,"",6)#.-4 @%''.#6 @',)'%67 !"44%#'),7 '/&3




NATURE REVIEWS ) MICROBIOLOGY VOLUME 3 ) MAY 2005 ) 44 5

information generated by cultivation- and

non-cultivation-based procedures.By strength-ening and extending the model built onKochs postulates, future inquiries will surely

accelerate the progress in linking ecologicallyimportant microorganisms to their activity in

real-world habitats.

Eugene L.Madsen is at the Departmentof Microbiology,Cornell University,

Ithaca, New York,14853,USA.e-mail: [email protected]

doi:10.1038/nrmicro1151A: Y/2f/%23'=6 E: _: !3+k-9-3.#0/%#-'+/ 52$ ^%/5"&=$/%2l3:

5+423,-&-,22 =B C,D2+3#%-%8#+ G&%$ BB6 A6 XXORA CAUUUD:Z3.12#+ $%&3#1&$2,3 23 0%/ H: o *+23.#$&76 *: @: G%,=&%4,$2'

720/%#2$4 p 5&.32$-7/6 743&52'#6 &37 ',3$%,1123. 9&'$,%#:J.#+4.+ CFD6 AWRNOAWRR CKWWKD:

AP: S/32#6 j: C/7D J'3;#;,- ,4/ *%3(,4.6 %=


8/8

44 6) MAY 2005) VOLUME 3 www.nature.com/reviews/micro


TV: @12/%5&3#6 J: Y: o M'+527$6 Z: H: !-$,%&72,.%&8+4 &37

255-3,91-,%/#'/3'/ ',5"23/7 9,% &-$,/',1,.2'&1 #$-74,9 #23.1/ '/11 &'$202$4 '/&3: 1,2'3+ IBH6

AWKTOAWXA CKWWND:

!'=3,

identifying microorganisms

Documents