Tải bản đầy đủ - 0 (trang)
Chapter 8. Molecular Methods in Geomicrobiology

Chapter 8. Molecular Methods in Geomicrobiology

Tải bản đầy đủ - 0trang



species-level differentiation of microorganisms is commonly assigned a 97% identify cutoff; this

is based on a weak correlation of whole genome DNA–DNA hybridization results with 16S rRNA

sequence similarity (Stackebrandt and Goebel, 1994). In whole genome hybridization, strains that

are >70% similar are considered to be part of the same species and it was shown that above this

cutoff no 16S rRNA sequence similarity <97% could be detected. Because any 16S rRNA sequence

of a new organism once established is entered into a databank (e.g., Ribosomal Database Project II:

http://rdp.cme.msu.edu/or GenBank: http://www.ncbi.nlm.nih.gov/Genbank/), it is possible to compare 16S rRNA sequences from a new isolate with those in the database to determine if it has been

previously reported and to identify its closest relatives. Although 16S rRNA sequences have proven

to be a fairly robust phylogenetic marker, we refer to the reader to a review by Olsen and Woese that

discusses the importance of comparing results from multiple molecules, both as a method for testing the overall reliability of the organismal phylogeny and as a method for more broadly exploring

the history of the genome (Olsen and Woese, 1993). Because sequencing whole microbial genomes

is a routine now, it is also possible to consider phylogenetic relationships based on whole genomes.

This is sometimes referred to as phylogenomics and we refer the reader to a review by Delsuc et al.

(2005) that discusses this approach.

Probes based on specific 16S rRNA nucleotide sequences can be applied directly to intact cells that

have been treated to make them permeable to the probes and allow hybridization between probes and

the corresponding nucleotide sequence in the ribosomal 16S rRNA. When such probes are labeled

with a fluorescent dye or with a radioactive isotope, cells that have reacted with the labeled probe

can be readily located by fluorescence microscopy or autoradiography, respectively (Giovannoni

et al., 1988; DeLong et al., 1989; Amann et al., 1990, 1992, 1995; Tsien et al., 1990; Ward et al., 1990;

Braun-Howland et al., 1992,1993; Jurtshuk et al., 1992; Schrenk et al., 1998; Orphan et al., 2001;

Michaelis et al., 2002). This fluorescence microscopy technique is called fluorescence in situ hybridization (FISH). The 16S rRNA probes can be group-, genus-, and species-specific. Caveats of this

method are that it requires genes to be actively transcribed (the metabolic rates of some important

geomicrobial organisms may be so slow that insufficient rRNA is present to detect), it can only be

used with sequences that are already known, and there is a chance that related but nontarget genes

will interfere with meaningful identification. One approach to circumvent the first of these issues

is to amplify the signal using catalyzed reporter deposition (CARD-FISH) (Pernthaler et al., 2002);

this technique has been used to identify organisms in biofilm communities in soil and marine environments (Fazi et al., 2005; Ferrari et al., 2006). Alternatively, chromosomal painting can be used

to directly target genomic DNA with fluorescently labeled nucleic acid probes to identify specific

sequences without the need for gene expression (Lanoil and Giovannoni, 1997).

Beyond probing for organisms with identifying sequences that are already known, various

culture-independent molecular approaches can be used to detect (new) organisms in the environment, and estimate the diversity of a microbial community. By using the polymerase chain reaction

(PCR) technique (Guyer and Koshland, 1989) with primers targeting 16S rRNA genes (16S rDNA),

it is possible to amplify 16S rRNA genes in the DNA extracted from the community. The resultant

mixture of 16S rDNA must then be resolved. One approach employs cloning. The 16S rDNA in

each clone is reamplified by PCR and the nucleotide sequence of the rRNA is determined. This

is sometimes referred to as making a clone library. Another approach to resolving the 16S rDNA

mixture derived from amplification of the DNA extract is to use denaturing gradient gel electrophoresis (DGGE), which can separate fragments of the same length that differ by only one or two

nucleotides. Attempts can then be made to hybridize rDNA bands in the electrophoretic patterns

with oligonucleotide probes from known organisms and thus identify the organisms corresponding

to the band that successfully hybridized with a probe. The bands from the electrophoresis gels can

also be eluted, purified if needed, reamplified, and their nucleotide sequence determined (Ward,

1992; Muyzer et al., 1993; Stahl, 1997; Ward et al., 1998; Madigan et al., 2000). Such sequences

can then be compared to sequences in an appropriate database to determine if they were previously reported. The second major technique for fingerprinting microbial communities is terminal

CRC_7906_Ch008.indd 140

11/10/2008 6:54:23 PM

Molecular Methods in Geomicrobiology


restriction fragment length polymorphism (t-RFLP) (Liu et al., 1997; Clement et al., 1998). t-RFLP

relies on the discrimination of sequences by restriction enzymes, which fractionate DNA molecules

according to the location of specific restriction sites. The resulting fragments are separated by size.

Individual sequences are differentiated because one of the PCR primers carries a fluorescent label

so that the fragment adjacent to this primer can be detected when it migrates past a laser detector on

a DNA sequencer. Although a convenient way to compare the diversity of different samples, t-RFLP

is a poor method to accurately resolve phylogenetic differences (Dubar et al., 2001). Increasingly,

DNA microarrays are being used to rapidly assess the microbial diversity of an environmental

sample (Wu et al., 2006).

An important caveat in the estimation of microbial diversity by 16S-based approaches is that 16S

rRNA genes can occur in multiple, and in some cases, divergent, copies per genome (Klappenbach

et al., 2001). Other potential complications to accurate diversity assessment through PCR-based 16S

techniques include potential biases in nucleic acid recovery due to differences in the ability of cells

to lyse based on differential membrane durability; sorption of cells onto particles or burial within

a complex matrix making them inaccessible; and co-extraction of contaminants that may hamper

downstream processing, shearing of DNA leading to chimeric products, and errors in the PCR reaction itself (Acinas et al., 2005). For a more complete discussion of these issues, we refer the reader

to von Wintzingerode et al. (1997).


Although molecular approaches have revolutionized our ability to detect organisms in nature,

ultimately, if we are to understand how they operate in any detail (see Section 8.4), we must be

able to study them in culture. For many years, the great plate count anomaly (i.e., the failure of

the majority of cells from nature to form colonies on plates, despite being visualized by microscopy as viable) vexed microbiologists interested in studying dominant organisms from the environment (Eilers et al., 2000). In recent years, however, a number of important advances have been

made toward this end. These include using diffusion chambers in media that simulate the composition of the natural environment (Kaeberlein et al., 2002), high-throughput procedures for isolating

cell cultures through the dilution of natural microbial communities into very low nutrient media

(Rappe et al., 2002), and encapsulating cells in gel microdroplets for massively parallel cultivation

under low nutrient flux conditions, followed by flow cytometry to detect microdroplets containing

microcolonies (Zengler et al., 2002). Two key concepts underlying the success of these methods are

(1) the composition of designed media must approximate the composition of the natural environment as closely as possible, and (2) organisms often need to be cultured in combination with others,

as cross-feeding between cells may be essential for their growth and not readily simulated in the

laboratory. Increasing attention on the part of geochemists to characterize the composition of the

environments in which microorganisms live is helping geomicrobiologists design more appropriate

media for their cultivation (Svensson et al., 2004).




Although the use of molecular signatures to identify specific organisms in the environment has

revolutionized microbial ecology, these molecules can only tell us who is there but cannot inform

us about their metabolic activities. To infer what geomicrobial organisms are doing in any particular locale, the classic techniques described in Chapter 7 have great value. In addition, several

techniques have been developed in recent years that permit concomitant molecular identification

of geomicrobial organisms and their metabolic activity (Figure 8.1). Some of these techniques are

even operable at the single-cell level. In general, they all rely on three steps: (1) identifying the

CRC_7906_Ch008.indd 141

11/10/2008 6:54:23 PM









Mass spectrometer CsCl gradient








Methanol-oxidizing community

in soil




FIGURE 8.1 An illustration of some of the techniques described in Section 8.3. Starting with an environmental sample (e.g., a methanol oxidizing soil community), a labeled substrate (labeled either with stable

isotopes or radioisotopes) is provided to the cells within the sample. Cells that can metabolize this substrate

incorporate the label into DNA, RNA, or other cellular constituents. Depending on the method used, cells

that incorporate this label are identified using MAR or secondary iron mass spectrometry (SIMS), and their

phylogenetic identify is confirmed using fluorescent probes that bind to their DNA or RNA (FISH). Before

attempting FISH-MAR or FISH-SIMS, SIP can be used to identify the organisms that take up the substrate

of interest. See text for details.

geomicrobial organism(s) of interest using a molecular label; (2) separating this population from

other organisms in the sample; and (3) analyzing this population with respect to its isotopic composition and its genomic content (to infer potential metabolic functions).



The first attempts to combine molecular identification (using 16S rRNA) with isotopic analysis of

single cells were made in 1999 by two different groups (Lee et al., 1999; Ouverney and Fuhrman,

1999). Both employed FISH (described in Section 8.2) with microautoradiography (MAR), and

enabled direct microscopic observation of whether particular cells had consumed a radioactively

labeled substrate under specific incubation conditions or not. In the past several years, this technique has been perfected and employed in many different contexts, from single cells to biofilms

(for a recent review, see Wagner et al., 2006). Recently, a quantitative FISH-MAR technique was

developed and used to measure the substrate affinity, Ks, of uncultured target organisms in activated

sludge (Nielsen et al., 2003). FISH-MAR only measures assimilation of the radiolabeled substrates

into macromolecules; unincorporated labeled compounds are not retained by the FISH fixation

process. Nevertheless, FISH-MAR is a very sensitive technique because radiotracers are incorporated into alll macromolecules, not just nucleic acids. An advantageous consequence of this is that

FISH-MAR requires relatively short incubation times (generally a few hours), which minimizes the

possibility of substrate breakdown and cross-feeding of secondary organisms.

CRC_7906_Ch008.indd 142

11/10/2008 6:54:23 PM

Molecular Methods in Geomicrobiology


The main limitations of FISH-MAR for geomicrobial studies are (1) it depends on a priori knowledge about the phylogenetic affiliation of the studied organism(s); (2) the number of populations that

can be specifically detected in a single FISH experiment is limited by the number of available fluorophores that can be applied simultaneously (currently <10); (3) some geomicrobial samples may

not be well suited for FISH analysis, as sediment/soil particles might obscure the resident bacteria;

and (4) the radioactive substrates of geomicrobial interest may not be available. The first, second,

and third limitations can be circumvented by molecular techniques that cast a broader net, such

as stable isotope probing (SIP) or the isotope array (described later); the fourth limitation may be

circumvented in some cases by simultaneously incubating the sample with unlabeled complex substrate and labeled 14CO2. This technique, called HetCO2-MAR, takes advantage of the fact that most

heterotrophs assimilate CO2 during biosynthesis in various carboxylation reactions. In addition,

HetCO2-MAR may permit better differentiation between substrate incorporation without growth

(e.g., into storage compounds) and actual growth than traditional FISH-MAR because growth is

thought to be required for active carboxylation reactions. An important caveat to this approach,

however, is that autotrophs can rapidly consume the labeled 14CO2 substrate, so if heterotrophs are

the target, specific autotrophic inhibitors must be added to avoid false positives. Another potential

limitation of FISH-MAR is that it requires active cellular incorporation of the substrate (important

organisms could be missed if populations of interest are metabolically quiescent at the time of sampling). The more general caveats to FISH discussed in Section 8.2 earlier, also apply.

A different single-cell approach that directly couples isotopic and phylogenetic analysis is FISHsecondary ion mass spectrometry (FISH-SIMS) (Orphan et al., 2001). In this technique, fluorescent

rRNA-targeted probes are applied to identify the microbial cells of interest in an environmental

sample (FISH). The precise location of hybridized cells, which are fixed on a glass or silicon slide

round, is mapped, and then the slide is transferred to an ion microprobe (e.g., SIMS, nanoSIMS,

time-of-flight-SIMS [ToF-SIMS]) to measure their stable isotopic composition. The SIMS works

by using a focused primary ion beam (typically Cs+) to sputter the surface of a specimen, thereby

ejecting secondary ions that are measured with a mass spectrometer to determine the elemental, isotopic, or molecular composition of the specimen. SIMS is a highly sensitive surface analysis technique, able to detect elements in the parts per billion range from samples with a primary beam spot

size ranging from tens of nanometers to tens of micrometers. Depending upon the ion microprobe

instrument used, data may be acquired with a static primary beam, resulting in either an average

isotopic value for the specimen, or if analyzing a cell aggregation or biofilm, time-dependent isotopic profiles may be obtained. Alternatively, ion imaging using nanoSIMS or multi-isotope imaging

mass spectrometry (MIMS) also shows promise, enabling the visualization of biologically relevant

ions (elemental or isotopic) within microorganisms, structured microbial assemblages, or tissues

(Lechene et al., 2007; Li et al., 2008). ToF-SIMS coupled to a gallium beam has also been used

successfully for micron-scale ion imaging (Cliff et al., 2002) and, recently, has been extended to

the characterization of archaeal lipid biomarkers in environmental samples with a spatial resolution

approaching 5 µm (Thiel et al., 2007).

Although, in principle, it is possible to measure a variety of biologically interesting stable isotopes and elements with SIMS, including isotopes of C, N, H, O, S, Ca, and Fe, to date, FISH-SIMS

has only been used to measure natural abundance δ13C in single cells collected from the environment. Like FISH-MAR, this method is compatible with stable isotope labeling experiments and

has been used successfully with both 13C and 15N labeling. For example, Orphan et al. (2001) used

FISH-SIMS to measure the δ13C profiles from the periphery to the interior of individual syntrophic

consortia from marine sediments at methane seeps. These studies revealed extremely low values

of δ13C in the inner cores of layered cell aggregates, identified as belonging to an uncultured group

of archaea, known as the ANME-2. Surrounding these cells, were sulfate-reducing bacteria of the

Desulfosarcina-Desulfococcus clade. This work provided strong evidence of metabolic syntrophy

between these organisms catalyzing the anaerobic oxidation of methane. Although the light 13C

signal affiliated with methane-oxidization is diagnostic and readily interpreted by SIMS analysis,

CRC_7906_Ch008.indd 143

11/10/2008 6:54:24 PM



other forms of C metabolism may be more difficult to interpret without the use of labeled substrate

additions. Additional variables to be considered for natural abundance stable isotope measurements

in single cells include the isotopic effect of exogenous carbon added during cell fixation and during

the hybridization with the fluorescently labeled oligonucleotide probe.

The major limitation of single-cell isotopic techniques is that they can only provide information

about the metabolism of substrates that can be labeled and traced isotopically. This handicaps our

ability to infer an integrated picture of metabolism for a single cell. Toward this end, two new PCRbased methods have been developed that permit the isolation of the genomic content of single cells

from complex environmental samples. The first such technique is termed microfluidic digital PCR,

and couples 16S rDNA identification of single, uncultivated cells with the simultaneous detection of

other genes in these cells. Cells are diluted and separated from a complex mixture using a microfluidic device. This separation step is critical, as single, partitioned cells then serve as templates for

individual multiplex PCR reactions. Although to date, this technique has been used to link the presence of a particular metabolic gene of interest with single bacterial species residing in the hindguts

of wood-feeding termites (Ottesen et al., 2006), in principle, it can be applied to link any number of

genes to 16S rDNA–identified single cells. This technique operates independently of the physiological state of the cell at the time of harvest, gene expression or position on the genome. Moreover, the

use of single-cell PCR avoids PCR artifacts such as amplification biases and unresolvable chimeric

products (described in Section 8.2) (Ottesen et al., 2006). Subsequent to the termite hindgut study,

similar microfluidic techniques were used to isolate and identify organisms from the human subgingival crevice. In this work, genomes were amplified from individual cells of interest, permitting the

sequencing and assembly of >1000 of their genes (Marcy et al., 2007). As genome-sequencing methods improve for single cells (Hutchinson and Venter, 2006), we can anticipate even better genomic

recovery in the future. The strength of this approach is that it allows specific organisms that have the

potential to catalyze interesting reactions (as measured by the presence of particular genes encoding the function[s] of interest) to be identified from complex environmental samples, assuming no

a priori knowledge of which organisms are involved. In this respect, it is similar to the isotope array

or SIP, only with microfluidic digital PCR, the cell as a distinct informational entity is preserved.

Another technique that has been developed recently (Pernthaler et al., 2008) is called magnetoFISH.

H Magneto-FISH combines 16S rRNA–targeted FISH and a tyramide-based amplification reaction (CARD-FISH) (Pernthaler et al., 2002), with immunomagnetic capture of hybridized cells using

paramagnetic beads coated with an antibody targeting the fluorochrome applied in the CARD-FISH

procedure. The selective capture of target organisms from complex environmental samples, combined

with metagenomic analysis and PCR surveys of diagnostic metabolic genes, can provide information

regarding the metabolic potential of uncultured microorganisms/microbial assemblages recovered

directly from the environment. Pernthaler et al. (2008) used this method to purify a specific lineage

of syntrophic anaerobic methane oxidizing archaea from marine sediment, along with their associated partner bacteria. Subsequent metagenome sequencing, PCR, and microscopy of these captured

consortia revealed unexpected diversity in the associated bacteria, and yielded new insights into the

versatility and metabolism of this syntrophic association. A unique advantage of the magneto-FISH

method is that it enables the recovery of the complete genetic content of uncultured target organisms,

and allows native partnerships to be identified by physical coassociation. A caveat is that spurious

microorganisms may potentially be copurified along with the true consortia. Although physical association alone cannot be used as exclusive proof of a syntrophic relationship, inferred associations and

metabolic potential revealed by magneto-FISH and metagenomics identify new areas for follow-up




So far, we have discussed single-cell techniques that employ fluorescent probes to identify organisms

phylogenetically, providing a handle for subsequent isotopic or genomic analysis to infer function.

CRC_7906_Ch008.indd 144

11/10/2008 6:54:24 PM

Molecular Methods in Geomicrobiology


A large number of fluorescent probes also exist that can potentially be used to measure other parameters of interest including pH, membrane potential, enzyme activity, reactive oxygen and nitrogen

species, toxins, and viability. A commercial supplier of such probes is Molecular Probes®, which

specializes in producing fluorescent reagents to measure a variety of attributes of living cells. For

example, the popular LIVE/DEAD stain (or BacLight kit) is used to measure the viability of cells

in environmental samples, and respiratory activity can be visualized using the dye 5-cyano-2,3ditolyl tetrazolium chloride (CTC). We note, however, that often these probes are developed in

yeast or Escherichia coli, and their application to geomicrobial organisms has not been calibrated.

This means one must be cautious when using these probes, as conventional interpretations of what

they signify may not apply (e.g., see Teal et al., 2006 and for a discussion of their use in biofilm

systems, see a recent review by Stewart and Franklin, 2008). Nevertheless, the application of fluorescent probes developed in other systems to geomicrobiology has the potential to greatly expand

our appreciation of cell biological processes in diverse organisms (see Section 8.4). For a probe to

be successful, however, it must be able to localize to a proper place in the cell (or microbial community), where it can measure the attribute of interest. A need exists for probe development to make

cell biological studies of this kind possible for diverse geomicrobial organisms. In the future, it is

possible that endogenous metabolites that exhibit a diagnostic and measurable fluorescent signal,

may even be detected directly inside living cells.



Until now, this section has dealt with techniques that are performed at the single-cell level. Although

this can be advantageous, there are times when a coarser degree of resolution is more helpful in

approaching a problem in geomicrobiology. For example, if one is interested in a particular metabolic process, but does not know which organisms are involved, methods that employ FISH are

not as helpful as techniques that do not require a priori knowledge of identifying sequences. Two

primary methods exist for this purpose: SIP and the isotope array.

SIP was first applied to the analysis of PLFAs. Because groups of microorganisms often have

diagnostic PLFA molecules, it is possible to identify those that are enriched in naturally abundant

stable isotopes such as 13C or 15N. If one knows the source of the label, one may conclude that the

organism (identified by its PLFA molecules) metabolized the substrate. Petsch et al. used this technique to make a case for the involvement of a complex microbial community in the weathering of

organic material from late Devonian black shales (Petsch et al., 2003). However, because nothing

is known about PLFA patterns for microorganisms for which there are no cultured representatives,

PLFA molecules are not as useful biomarkers as nucleic acids. Accordingly, most SIP studies today

employ isotopically labeled DNA or RNA as the phylogenetic hook. Because the buoyant density of

DNA varies with its guanine and cytosine (GC) content, the incorporation of a heavy isotope (e.g.,


C) into DNA enhances the density of labeled DNA compared to unlabeled (e.g., 12C) DNA. A band

of heavy DNA can thus be separated from the unlabeled population, cloned and sequenced to identify organisms involved in the metabolism of the labeled substrate using PCR-primers targeting 16S

rDNA. Alternatively, PCR can amplify other genes that might play a role in catalyzing the metabolism of the substrate. DNA-SIP was first made use of to identify organisms involved in methanol

consumption in forest soil (Radajewski et al., 2000). These initial studies involved artificially high

substrate concentrations and relatively long incubation times (∼40 days), which raised the issue

that such exposure might also result in the detection of organisms that used labeled intermediates

or by-products generated by the primary consumer organisms. If one is interested in identifying

(syn)trophic interactions, however, this may be useful. RNA-based SIP can potentially mitigate this

problem. Regardless of whether primary consumers or downstream cross-feeders are the target, one

powerful attribute of SIP is that 13C-labeled DNA can be cloned to generate a metagenomic library.

Unlink magneto-FISH, which enriches for genomes of interest based on 16S phylogeny, DNASIP can enrich for genomes based on the incorporation of specific substrates into the DNA itself.

CRC_7906_Ch008.indd 145

11/10/2008 6:54:24 PM



For thorough recent reviews of DNA and RNA SIP, with more complete discussions of their advantages and limitations see (Dumont and Murrell, 2005) and (Whiteley et al., 2006).

One potential drawback to SIP is that it requires high levels of 13C incorporation (>50%) to

enable separation of the nucleic acids of interest using standard techniques, which makes it insensitive to slowly growing organisms (or organisms using multiple substrates). A new technique that

permits precise isotopic analysis of nanogram amounts of samples has recently been developed

(Sessions et al., 2005) that may be useful in this context. Known as spooling wire microcombustion (SWiM), this technique has been used to analyze both rRNA, specifically captured to target

particular phylogenetic groups (Pearson et al., 2008) and whole cells sorted by flow cytometry (i.e.,

fluorescence-activated cell sorting [FACS]) (Eek et al., 2006). The method is compatible both with

substrates containing a natural 13C abundance, and with artificially 13C-enriched tracers. In brief,

a single droplet of liquid sample is placed onto a continuously, slowly spooling wire that is carried into a combustion reactor where nonvolatile organic matter is quantitatively oxidized to CO2,

H2O, and NOx. CO2 is then carried in a helium gas stream to an isotope ratio mass spectrometer

(IRMS) for measurement of the 13C/12C ratio. Although remarkably sensitive, an important practical

limitation of this FACS-SWiM marriage is the difficulty of separating and accumulating sufficient

biomass (∼107 bacterial cells or 104 eukaryotic cells, Eek et al., 2006) to make the measurement,

particularly for cells that are not naturally autofluorescent. To detect these cells, exogenous amplified fluorophores (such as those used in CARD-FISH reactions) must be applied, but this appears

to add too much exogenous carbon to the cell to yield an accurate δ13C value. The development of

brighter fluorescent labels for whole cells that do not add appreciable carbon is needed to make

FACS-SWiM more broadly applicable.

An alternative to using stable isotopes to identify organisms involved in the uptake of particular substrates is to use radioisotopes. This was described earlier for single cells (FISH-MAR), but

microarrays can also be used to screen microbial communities. In the isotope array, radiolabeled

RNA (typically containing 14C) is extracted from a sample, labeled with a fluorophore and probed

against an oligonucleotide microarray that targets 16S rRNA of different microorganisms. In this

manner, organisms that can take up the label can be rapidly identified, without a priori knowledge

of the community’s composition (Adamczyk et al., 2003). The isotope array is a high-throughput

method and requires only a short incubation time. But unlike SIP, the isotope array can only detect

organisms whose sequences are represented on the array, so in this regard, it is more limited. As

noted previously in the discussion of limitations to FISH-MAR, organisms that are metabolically

quiescent (or simply growing slowly) at the time of sampling would be missed, and the technique is

limited to substrates for which there is a radiolabel.


Although isotopic labeling can be a powerful way to link specific metabolic capabilities with particular organisms in the environment, it is limited to metabolisms for which there are isotopic tracers.

In the absence of these, one can use genomes to infer potential metabolic capabilities or the ability

to catalyze geochemically significant reactions. We emphasize the word potential, as genomes, at

best, can only hint at what might be possible for a given organism. This type of analysis (often called

metagenomics; Riesenfeld et al., 2004; DeLong, 2007) relies heavily upon computational analyses

to compare different sequences to each other and to identify motifs in the DNA or the gene products

that are predicted to have a specific function. Hypotheses can be generated about what types of reactions a given protein might catalyze, or the conditions under which the gene that encodes it might be

expressed; sometimes, genomic analysis can even be used to make predictions about the behavior

of entire microbial communities (Tyson et al., 2004; DeLong et al., 2006). Ultimately, these hypotheses must be tested through classical genetic and biochemical analyses to prove that the connection

between the presence of a particular gene and an inferred metabolism/predicted geochemical effect

is actually causal as opposed to correlative (see Section 8.4 and Martinez et al., 2007).

CRC_7906_Ch008.indd 146

11/10/2008 6:54:24 PM

Molecular Methods in Geomicrobiology


Various methods exist that enable genomic information to be acquired from specific organisms

in the environment (e.g., magneto-FISH, digital PCR). In cases where the cellular host has been

lost (e.g., SIP or simply DNA-purified in bulk from the environment without isotopic selection

and cloned into cosmid, fosmidd or bacterial artificial chromosome [BAC] libraries), metagenomic

information can still be assigned to specific organismal entities if sufficient DNA is cloned to permit the assembly of putative whole genomes. We underscore the word putative, as it remains very

challenging, despite advances in bioinformatics, to reconstruct original genomes from environmental DNA. Despite these challenges, environmental genomic data can be extremely powerful. Not

only does it permit gene expression in communities to be monitored in situ (Ram et al., 2005), but

it allows the tracking of specific genotypes and their close relatives in their natural environment

(Rich et al., 2008) and may guide the design of media to enable the isolation/study of particular

geomicrobial organisms (Tyson et al., 2005; Sabehi et al., 2005). For a more complete discussion of the opportunities and challenges of environmental genomics in the context of geomicrobial

problems, see Banfield et al. (2005).


If one is interested in the expression of a particular metabolic gene or its gene product, specific

probes can be designed to detect it. For example, PCR primers designed to target archaeal amoA

(the gene encoding ammonia monooxygenase alpha-subunit) revealed the widespread presence of

ammonia-oxidizing archaea in marine water columns and sediments, suggesting that these organisms may play a significant role in the global nitrogen cycle (Francis et al., 2005). Another example

is the use of PCR primers targeting the arrA gene, which encodes one of the subunits of the respiratory arsenate reductase common to most arsenate-respiring bacteria, to detect the expression of

this gene in arsenic-contaminated sediments (Malasarn et al., 2004). In these studies, bulk mRNA

was extracted from sediments and reverse transcribed to DNA prior to PCR analysis. Although

these studies looked for the expression of genes in the environment qualitatively, it is possible to

quantify gene expression using real-time PCR or quantitative PCR (q-PCR) (Walker, 2002). Such

an approach was used to quantify niche partitioning among Prochlorococcus ecotypes along oceanscale environmental gradients (Johnson et al., 2006).

A caveat to PCR-based approaches is that, in the case of proteins, posttranscriptional controls

may affect the abundance (integrated production and decay) of the gene products of interest; accordingly, detection of a DNA or RNA sequence in a given environment cannot prove that its product is

present in the environment. A better check for this is to use antibodies targeted against the protein

of interest. One example of where this has been used is in the area of (per)chlorate reduction, where

immunoprobes were raised against the enzyme chlorite dismutase (CD) (O’Connor and Coates,

2002). CD is highly conserved among the dominant (per)chlorate-reducing bacteria in the environment, thus CD appears to be a good target for a probe specific to this process. If one seeks spatial resolution of gene/protein expression within microbial communities, FISH can be used, with

the caveat that signal amplification is usually needed to detect low-abundance mRNA or protein

(Stewart and Franklin, 2008).




As discussed in Sections 8.2 and 8.3, molecular biology has found wide application in geomicrobiology by helping investigators determine which microorganisms are present in a given environment

and what their (potential) metabolic activities are in that environment. Underlying many of these

techniques are assumptions about the biological function of specific molecules in the cell (e.g.,

rDNA is a good phylogenetic marker because it is known to be an essential, highly conserved,

and slowly evolving constituent of the replication machinery common to all life; measuring the

CRC_7906_Ch008.indd 147

11/10/2008 6:54:24 PM



expression of specific metabolic genes in the environment presupposes we know which genes to

look for, and what their gene products do). How does one elucidate these genes in the first place,

and determine their cellular function? This is the province of classical genetics, biochemistry, and

cell biology—fields whose tools have been developed and honed over many decades in a variety of

model systems ranging from Escherichia coli to yeast to fruit-flies to mice.

Because geomicrobially significant organisms have often been assumed to be difficult to tame

in the laboratory (although this may simply reflect a lack of trying rather than an inherent problem), less is currently understood about how they catalyze geobiologically significant processes.

But this is changing rapidly. The past decade has seen heightened interest in exploring the genetics,

biochemistry, and cell biology of geomicrobial organisms. Owing to advances in culturing, the

plummeting costs of sequencing microbial genomes, and the entrance into the field of students with

training in genetics, biochemistry, and cell biology, geomicrobiology is now primed for exciting

discoveries on the mechanistic level. Why does mechanism matter? Because it permits a variety of

interesting questions to be asked, including What genes are required for a process of interest? What

are the products of these genes? Where do they reside in the cell? How do they function (e.g., what

molecular partners—cofactors, other proteins—are required)? If they are enzymes, what are their

catalytic rates? And under what conditions are they expressed? Ultimately, the answers to these type

of questions will permit geomicrobiologists to return to the environment to search for the presence

and expression of particular genes/proteins once their function is known (e.g., Karkhoff-Schweizer

et al., 1995; Malasarn et al., 2004; Sabehi et al., 2005).


There are many ways one can use genetics to study living cells. Here, we restrict our discussion to

the genetic approaches that are routinely applied in geomicrobiology: the construction of mutants,

the expression of DNA in a foreign host, and the labeling of specific genes to visualize their expression or the cellular location of their products.

Mutagenesis is used to identify genes essential for catalyzing a process of interest. Various methods for mutagenesis exist, offering the potential to eliminate/delete genes entirely, introduce point

mutations into specific genes, or introduce genes into an organism. The effects of these different

types of mutations can be far ranging, from altering the amino acid composition of a protein and

thereby affecting its substrate specificity, to changing the regulation of an entire network of genes,

to eliminating the ability to make a set of proteins. The latter type of mutagenesis represents a

loss-of-function approach, and is commonly used to identify genes essential for a given function.

Examples of where this has been used in geomicrobiology include work on phototrophic iron oxidation (Jiao et al., 2005; Jiao and Newman, 2007), manganese oxidation (van Waasbergen et al., 1996),

iron reduction (Coppi et al., 2001; DiChristina et al., 2002; Myers and Myers, 2002), arsenate reduction (Saltikov and Newman, 2003; Murphy and Saltikov, 2007), magnetosome formation (Komeili

et al., 2004; Scheffel et al., 2008), and methanogenesis (Pritchett and Metcalf, 2005). Many other

examples are provided elsewhere in this book.

Mutagenesis can either be random or directed. To make random mutations, transposons are

often used. Transposons are mobile genetic elements that (usually) insert in an unbiased fashion

into the genome of the organism of interest. (Note:


some transposons such as Tn7,

7 insert in a neutral

chromosomal site, and thus can be used to introduce genes in single-copy—see Koch et al., 2001.)

They disrupt the sequence of the gene into which they land, thereby destroying the gene product

and rendering it dysfunctional. Commonly, transposons carry a constitutively expressed antibiotic

resistance marker, so that their insertion may be selected by plating cells on medium containing the

antibiotic. Mating between a donor strain (that carries the transposon) and a recipient strain (the one

to be mutagenized) is often used for transposon delivery. Transposon libraries must be screened

to identify the subset of mutants that have a defect in the phenotype (i.e., property or function) of

interest. Although chemical mutagens or UV can also be used to make random mutations throughout

CRC_7906_Ch008.indd 148

11/10/2008 6:54:25 PM

Molecular Methods in Geomicrobiology

1. An organism possessing a

function of interest is isolated


2. Random chromosomal insertion of a

transposon achieved via mating = mutagenesis






3. Mutant library is screened

4. Sequencing with

primers specific to the

transposon to identify

the insertion site

5. A wild-type copy of the gene is provided

to the mutant = complementation; wildtype phenotype is rescued

Mutant loses a gene

necessary for the function

of interest because the

transposon interrupted it

Any other gene

interrupted (provided it is

not essential) will exhibit

the wild-type phenotype

FIGURE 8.2 Steps involved in transposon mutagenesis (i.e., a loss-of-function assay). As an example, Fe(II)

oxidation to Fe(III) is provided as the phenotype of interest. The blue fragment represents the part of the chromosome that contains the gene(s) that encodes the function of interest. The yellow fragment represents the

transposon. Qualitatively, bottles turn a rusty brown color when iron oxidation occurs. See text for details.

the genome, the advantage of transposons is that their site of insertion can be readily mapped using

molecular techniques, permitting identification of the genes into which they insert. For many applications in geomicrobiology, this is sufficient to determine which genes control a process of interest

(Figure 8.2). If one seeks to learn more about the importance of particular residues within a given

gene, however, transposons are not appropriate; in this case, site-directed mutagenesis (a controlled

technique for generating mutations, including point mutations) or chemical mutagenesis (the addition of chemicals that promote modifications of DNA at the scale of single residues) must be used.

For example, these more highly resolved mutagenic techniques would be appropriate if one wanted

to understand the function of a specific protein in greater detail (e.g., which sites are necessary for

it to bind its substrate, which are required for the binding of cofactors, and which control its ability

to associate with other molecules in the cell).

After mutagenesis is performed, mutants are identified either through a selection or a screen.

A selection permits only those mutants that have the desired properties to grow, whereas a screen

requires characterizing the behavior of thousands of mutants to identify those that have the phenotype of interest. Depending on the manner in which one has identified candidate mutants, secondary

screens may be required to narrow the pool of candidates down to only those that are interesting. For

example, if one performs a screen to find genes that control various steps in a biochemical reaction

and if the assay for mutant identification involves looking at the rate at which a reaction proceeds,

then false mutants could be identified by the screen that are simply slow growers but which do not

have a specific defect in the reaction of interest. These mutants could be sorted out by measuring

the growth rate of all candidates and only continuing to study those whose growth were normal with

respect to the wild-type (i.e., parent) strain. Once interesting mutants are identified, the nature of

the mutation must be determined through sequencing and genetic verification. Often this includes

a complementation experiment, where a wild-type copy of the gene of interest is put back into the

mutagenized strain (often on a plasmid), to demonstrate that it can restore the original phenotype.

CRC_7906_Ch008.indd 149

11/10/2008 6:54:25 PM



In the event that the wild-type phenotype is not rescued by this experiment, it means that the mutation responsible for the phenotype of interest lies elsewhere on the chromosome. Sequence analysis

can help generate hypotheses to explain why the mutant behaves the way it does (see the discussion

on bioinformatics in Section 8.4.2), and thus infer what affects the process of interest. To test these

hypotheses, however, physiological, biochemical, or cell biological experiments are required.

However, not every organism is amenable to mutagenesis. For a practical discussion of this, see

Newman and Gralnick (2005). An alternative strategy to identify genes from these organisms (or

from environmental genomic sequences for which an organismal host may or may not be known)

that are necessary and sufficient to catalyze a particular reaction is to clone and express them in a

foreign host (i.e., heterologous complementation). This is a gain-of-function strategy, in contrast

to the loss-of-function mutagenesis approach described above (Figure 8.3). Toward this end, the

host must have the metabolic machinery necessary to process the gene product(s) one wishes to

express. For example, if one seeks to express genes encoding multiheme cytochromes, such as those

found in Geobacterr and Shewanella, in a genetically tractable foreign host such as Escherichia coli,

E. coli must be capable of processing these cytochromes in addition to its own cytochromes; this may

be accomplished by engineering E. coli to express extra versions of the machinery for cytochrome

Non-tractable organism



1. Extraction of genomic


2. DNA is cut into smaller

fragments using restriction


3. DNA fragments are cloned into

a vector and transformed into a

genetically tractable host on a vector

4. Screen host to discover which

strain carries the gene/function of


5. DNA fragment containing gene(s) of interest is sequenced

FIGURE 8.3 Steps involved in heterologous complementation (i.e., a gain-of-function assay). As an example, Fe(II) oxidation to Fe(III) is provided as the phenotype of interest. The red, blue, and green fragments

symbolize different parts of the chromosome. The green fragment represents the one that contains the gene(s)

that encodes the function of interest. Qualitatively, bottles turn a rusty brown color when iron oxidation

occurs. See text for details.

CRC_7906_Ch008.indd 150

11/10/2008 6:54:25 PM

Tài liệu bạn tìm kiếm đã sẵn sàng tải về

Chapter 8. Molecular Methods in Geomicrobiology

Tải bản đầy đủ ngay(0 tr)