Genomics Tools in Environmental Impact Assessment - American

Genomics Tools in Environmental Impact Assessment - American...

0 downloads 31 Views 407KB Size

GENOMICS Tools in Environmental Impact Assessment


Subtle impacts caused by industrial development that may have long-term effects can be monitored at the microbial level using selected tools.

ustainable development is based

scoping, focusing, impact assessment, im-

on a paradigm that minimum

pact mitigation, impact evaluation, and

adverse environmental effects

monitoring and feedback—that can lead to

ensure long-term benefits. Envi-

specific delineated objectives of the EIA (3–8).

ronmental impact assessment

EIAs have historically been seen as a con-

(EIA) has gained attention worldwide as a

straint to development, but this view is grad-

way to provide intergenerational equity

ually disappearing (11). Initially, developers

(1–3). In an EIA, the consequences of de-

ignored the ecosystem while preparing for

fined actions are identified, quantified,

an EIA because resilience in a natural sys-

evaluated, and predicted with a broad

tem would take care of perturbation in early

remit to socioeconomic developmental ac-

stages of the developmental process (12, 13).

tivities (3–5). New approaches to EIA have

Because adverse environmental effects are

evolved, and methodologies are now clas-

not immediately apparent, the promoters

sified and addressed systematically at dif-

and the end users have initially ignored the

ferent levels of strategic decision making (6,

legislative obligation to creating a sustain-

7 ). If the objectives of the EIA are focused

able process (14).


on only ecosystems and their components,

However, emerging interactive tools for

the process can be considered an ecolog-

evaluating the status of natural resources

ical impact assessment (EcIA) (8). Essentially,

could help change this scenario (4, 8).

such studies characterize the carrying ca-

Genomics tools provide an option for an-

pacity of the associated biogeochemical

alyzing basic resources at the nucleic acid

cycles of a selected system and estimate the

level of an organism or population of

impact from developmental activities (9).

microorganisms (15, 16). In this article, we

An intrusion of any developmental activi-

discuss the application of selected ge-

ty always changes the course of evolution

nomics tools for an EIA, so that a pertur-

of a dynamically stabilized system, such as

bation in an ecosystem due to industrial

a microbial community (10). Therefore, we

development could be analyzed at the

can identify common procedural steps—

basic resource level.

© 2003 American Chemical Society



Genomics tools aid in monitoring and assessment Applying genomics tools for the monitoring and assessment process can help evaluate the environmental impact of industrial activity, including causes of pollution and mitigation and cleanup processes.

Baseline data

Genomics tools

Environmental Environmental impact Impact

Short-term effects

Long-term effects


Soil Industry Water

Restoration of baseline

Bioremediation and bioaugmentation

Reinstatement with shifted baseline with no/very low ecological penalties

Mitigation measures

Genomics tools

Information flow

Microbial communities are key driving forces through which essential ecological services are rendered to higher microorganisms of that ecosystem (9, 17, 18). The ecological services are part of the biogeochemical cycle, which is catalyzed by the biochemical potential of microbial communities (12). Genomics tools offer new ways to understand the overall performance of microbial communities (19– 21). Using tools such as DNA microarrays or real-time polymerase chain reaction (PCR) analyses, scientists can estimate the catabolic activities of microorganisms and, on correlation with organic loading, understand the survival dynamics of a community (22, 23). However, the genetic information needed to network these events is still insufficient because of the lack of knowledge for unculturable bacteria (24). Nevertheless, to get more insight into the metabolic network, scientists can interface genomics tools with the rules derived from nonequilibrium thermodynamics to model the survival of microbial communities in the environment (25). These tools provide information on the DNA content, which could be used for a monitoring and assessment strategy in an EIA process. Figure 1 shows how genomics tools can be applied to monitoring and the assessment and how they can help evaluate mitigation processes.

The recommendations of any EIA process depend on the details provided through the main techniques used in scoping: baseline studies, checklists of stressors and receptors and their spatial and temporal distributions based on conceptual models, and finally the design of a network diagram (1–4, 8). These activities could be reframed, better understood, and more appropriately applied by using genomics tools as shown in the network diagram in Figure 2. Initially, the aim of the scoping process is to establish the physicochemical status of the site and the forces that are disturbing the environment. These activities quantitatively assess stress and consist mostly of analyzing different kinds of organic and inorganic loadings. The next step is identifying receptor components and networking them as part of a series of interactive or parallel events. These events are grouped in different layers of microbial communities representing a collection of receptor microorganisms that are sensitive to the immediate effects of the stress. Depending on a microbial layer’s inherent resilience, a qualitatively specific—but possibly not quantifiable—threshold for the stress may exist, through which the effects can pass along to the subsequent receptors until they affect the higher microbes of that community. For example, if any toxic chemical that inhibits the survival


of nitrogen-fixing bacteria enters the system, the biogeochemical cycle that maintains the interspecies flow of nitrogen as substrate will become imbalanced. This result suggests that the microbiological components of the basic resources are immediate receptors for the disturbances. If the disturbances are temporary, the microbial components initially will absorb the stresses according to their assimilatory capacity. In this scenario, the resources are mostly restored; otherwise, the cumulative effect of stress will be transferred to the next layer, which is eukaryotic flora and fauna. The second aspect of this scoping process is to evaluate the most sensitive indicators of biogeochemical cycles and how they have been supported by the available gene pool, which requires a priori knowledge.

Definitions In an EIA, protocol with genomic tools can be envisaged in which the biological components of the basic resources are further reduced and expressed in their abstract form, such as DNA. Microbial ecosystems in these basic resources can be inventoried with genomics tools as shown in Figure 3. It has been proposed that by using molecular tools, scientists can generate a fingerprint by focusing on the total microbial capacity and its diversity. The DNA fingerprint could then be assigned an indicator DNA signature (IDS), which would allow disturbances at the genetic level to be monitored. The status of energy flow and carbon transport is another crucial feature that shows the performance of these systems. Therefore, an important factor is the keystone biochemical processes (KBP), which involve maintaining the levels of carbon, nitrogen, and phosphorus in any ecosystem mediated via microorganisms.

Options for assigning IDS The IDS can be developed with different criteria to represent the available DNA pool in various ways. The process of determining IDS begins with PCR (26). Using thermostable DNA polymerase under the defined reaction conditions, sub-strings of in vitro DNA are amplified. The amplicons or products are generated within the limits of hybridized primers and are separated and analyzed using electrophoresis. The limiting step in this analysis could be the extraction of the total DNA pool, but methods have been reported that overcome this problem (27–30). Randomly amplified polymeric DNA (RAPD) analysis is the most abstract way of representing the available total DNA pool of different eukaryotic and prokaryotic life forms in any ecosystem for an IDS (31). RAPD offers high repeatability and speed of analysis. The selected primer randomly hybridizes to inverted repeats on a template DNA pool to amplify the target strings by PCR. In these cases, the primers could be designed using different pattern search tools that are commercially available. We have used the subsequences 5′AAGAGCCCGT-3′ and 5´-CCCGTCAGCA-3´ to assess diversity, both at total DNA levels and in different isolates (32). With different dinucleotide composition, the primers also could be designed randomly, screened,

and selected for a specific problem. The amplicons of varying sizes in the reaction are generated on the basis of the frequency and the distance of occurrences of these repeats (33–35). Amplified ribosomal DNA restriction analysis (ARDRA) targets only the eukaryotic population in any heterogeneous template DNA sample. The key is the 16S ribosomal DNA (rDNA) gene, which is conserved on an evolutionary scale but is still diverse enough to identify and classify eubacteria. Hence, ARDRA has been used for assessment of eukaryotic diversity. Universal primers have been reported that could amplify the phylogenetically diverse 16S rDNA strings from a heterogeneous DNA template population. Primer pairs used for the analysis include 5´AGAGTTTGATCMTGGCTCAG-3´ and 5´-TACGGYTACCTTGTTACGACTT-3´, which yield approximately 1450 base pair products (35, 36). The closely related eubacteria can be discriminated on the basis of their dinucleotide composition (33, 34). It is quite possible that these discriminating dinucleotides create a few palindromes, which can be applied as diagnostic tools in differentiating the 16S rDNA amplicons. Based on this principle, ARDRA has been used as a diagnostic tool (35). Here, the 16S rDNA amplicons were subjected to restriction digestion using enzymes that hydrolyze frequently occurring palindromic sequences. The digested products were resolved by electrophoresis. For example, this kind of study of microbial diversity has been used to measure the temporal contamination levels in a groundwater flow path in a coaltar-waste-contaminated aquifer. Alternatively, researchers could define the dynamics FIGURE 2

Generic conceptual network This conceptual diagram illustrates scoping of basic resources and generating information to assess disturbances at the DNA level. Biogeochemical cycles • C:N:P Ratio • Biodegradative operon • Micronutrients • Missing metabolic potentials

Scoping of basic resources (targeting at molecular levels)

Receptors involved • Bacterial community • Eukaryotic microorganisms • Symbiotic/ nonsymbiotic associations

Air Generic conceptual network Quantitative assessment of stress • Area affected • Quantity of loading • Pattern of loading • Inorganic loading • Organic loading



Monitoring selected coordinates over time after mapping

Tracking based on a priori knowledge • Specific loci • Mobile DNA element • DNA signatures due to reinstated baseline



An array of genomics tools This scheme illustrates the different types of analytical genomics tools that can be used for evaluating disturbances in the total gene pool and its performance. Time series analysis

Evaluation and prediction of disturbances

Organic and inorganic chemical analysis

IDS for representative microbial community by RAPD

Sample preparation for monitoring and assessment

Genomics tools

Stress assessment based on chemical analysis

Disturbances Baseline

ARDRA analysis on timescale

Monitored site

Bandsharing index

Selection of indicator bacterial species

Keystone biochemical processes

Biogeochemical cycles

Microbial community

Indicator/novel/umbrella/ uncultivable species Tracking based on a priori knowledge


Analysis for • CO2 production • CO2 fixation • CH4 production • Nitrogen levels (NH4+, NO2–, NO3–) • Phosphates

Catabolic operons associated with baseline data or shifting with time

of eubacterial diversity by analyzing the ARDRA clone libraries (37 ). In addition, novel or rare microorganisms could be identified on the basis of unique sequences and digestion patterns. For example, the caffeinemetabolizing Stenotrophomonas maltophilia has been associated with a naphthalene-degrading population in agricultural soil (37). Similarly, we also have isolated S. maltophilia strain HPC-252 (GenBank Accession No. AY348312) from the wastewater treatment plant of an oil refinery. This observation suggests that these kinds of isolates can be used as indicator species to monitor hydrocarbon contaminated sites because they are colonized with microbial flora thriving on hydrocarbons.

Housekeeping and analysis for KBP The microscopic environment of basic resources has biotic and abiotic components like any other natural ecosystem. A typical carbon:nitrogen:phosphorus ratio can help assign the density of the active biotic component in a niche (38). These kinds of analyses can be performed by using precision techniques, such as 14Clabeled sources (39), or by determining the relationship of available nitrogen in soil and the liberated CO2 (40). It has been proposed that an index for physiological parameters (IPP) could be established for a given system as shown in Figure 3. Parameters such as CO2 fixation, CO2 generation, and CH4 production can indicate biological activity, whereas the levels of ions 360 A ■ ENVIRONMENTAL SCIENCE & TECHNOLOGY / OCTOBER 1, 2003

Microarray analysis


PCR/multiplex or real-time PCR

NO 2– , NO 3– , NH 4+, and free PO 43– can measure the modulation in such activities. For example, microbial flora adapt to a contaminated site over a period of time, which can remove the stress on basic resources arising from contamination. However, the microbes also demand a balance of nutrients to mineralize the available excess organic loading.

Interactive predictions using IDS and KBP Similarity index. RAPD markers have been described as a simple method to detect polymorphism (41). The polymorphism is detected as the presence or absence of bands; the latter may be caused either by the failure to prime a site, possibly because of nucleotide sequence differences, or by insertions or deletions in the fragment between the two priming sites. Amplicon band patterns are often used to determine how a microbial community is evolving on a time scale or, at different sampling points, how these communities resemble each other in a given ecosystem. Accordingly, a similarity index measure SI is defined as SI =

2Sxy Sx + Sy

where Sxy is the number of bands that are shared by sample x and y, while Sx and Sy are the number of bands for sample x and y, respectively. The similarity between the samples is expressed as a band-sharing

x 100


where wi is the numerical weight associated with each of the parameters, indicating their relative importance as an indicator of activity-dependent stress/ modulation, and pi is defined as pi =

Evaluation and prediction of impact based on index for a model scenario (a) A system at status quo compared with (b) a system just beginning to face a perturbing situation. IPP stands for the index for physiological parameters, BSIR stands for band-sharing index based on RAPD patterns, and BSIA stands for band-sharing index based on ARDRA patterns. (a)

pi1 pi 0

which is the concentration of ith parameter, with suffix “0” indicating the base concentration, while pi1 is the concentration of the ith parameter, with suffix “1” indicating the present concentration, which is to be compared with the reference. Such an index computed at different time points would indicate physiological imbalances incurred due to various driving forces, as shown by the simulated data in Figure 4.

120 110 100 90 80 70


Impact assessment based on indices


1 2 3 4 5 6 7 8 9 10


(b) 120 110 Index

The trends of indices over time and their impacts on each other can also be studied. For example, two possible scenarios are considered in Figure 4. In Figure 4a, development does not disturb the basic resources. The simulated data indicate that with time there is hardly any change in the physiological status, which is supported by the RAPD and ARDRA patterns. The illustration considers the baseline data, which are included in the indices. In ideal conditions, when organic or inorganic loadings intrude into basic resources, it is expected that the microbial community structure will not change immediately. However, the community’s performance as measured by IPP could be affected. If the disturbances change the community’s aerobic respiration rate, the CO2 liberation levels will be affected and consequently the IPP level will decrease

1 2 3 4 5 6 7 8 9 10




Time po in


Microarray analysis could determine the composition of a microbial community by using the 16S rDNA probes or probes derived from operons associated with catabolic capacity (15, 43–45). However, the degree of specificity, sensitivity, quantification, and associated costs of currently available technologies limit the use of microarrays as a regular monitoring tool (46).



Alternative options for further detailing

100 90 80 70




wi pi

without significantly affecting the DNA signature of that community, as shown in Figure 4b. If this scenario induces a change in the DNA signature, it could be an alarming situation in which an immediate mitigation plan should be formulated. However, if the change leads to a drift in the baseline data without real penalties to expected ecological services, then the generated knowledge has to be carefully adapted into the information flow of the EIA process.

Time po in

index (BSI), which is used to generate a matrix. This matrix of genetic distance for all pairs of samples can thus be obtained and used for constructing a dendrogram to illustrate the time-dependent variability of the samples (31, 42). The index can be applied to evaluating the associated variability after the restriction digestion analysis of the amplified ribosomal DNA fragments. In other words, it could even be calculated for ARDRA patterns. Although the ARDRA patterns are used for selection of novel clones, collecting samples for the EIA process can be difficult. Hence, the BSI can be used to quantify the observed changes in the ARDRA pattern. IPP. IPPs would indicate the performance of a microbial community in terms of organic loading and the changes in metabolism with time. Before a site is developed, baseline data are collected as a reference to be compared with the postdevelopmental data. However, for a site already being developed, a reference of baseline data could be established by taking samples from the location with the most developmental activity and comparing them to samples from peripheral areas that are expected to be the least disturbed. A comparison of the data before and after the activity is initiated can be quantified in terms of IPP using simple weighted arithmetic means, such that



A priori knowledge. On the basis of a priori knowledge or history, the indicator gene could be selected. If a problem is caused by a sewage system, then the intrusion vector carries the biological elements of human origin and could be directly used for IDS. We have reported different loci-specific PCR protocols that could be applied as IDS under designated intrusion activity (27–29). For example, the presence of E. coli in drinking water samples suggests contamination or mixing with waste from human origin. Another option is to gauge the effects of different anthropogenic forces that are collectively exerting stress on a community structure, such as organic pollutants. If the genes required to metabolize such unforeseen organic loading are acquired over time by the umbrella species of the community, then the DNA pool could be analyzed for these genes and considered under IDS (21, 47 ). If the analysis requires targeting different loci, then multiplex PCR protocols can be applied in which loci-specific amplicons are simultaneously amplified in the same PCR (48). To support this approach, if an indicator species has already been identified, then programs like PRIMROSE can assist in designing the genus-specific IDS primers (49). With data from the Ribosomal Database Project, PRIMROSE finds probes on the basis of PCR primers for use as phylogenetic and ecological tools to identify and enumerate eubacteria. Similarly, it is expected that under stress, the natural recombination rate will be induced, and the mobile element carrying DNA could be considered for IDS. For example, researchers could use primers derived from transposon elements (or insertion sequences) (50) or catabolic mobile genetic elements (51) for PCR. Indicator/novel species. The exotic species generally get the most attention in EIA or EcIA processes. Hence, sometimes they are selected as indicator species. Extremophiles, bacteria that can survive in extreme environments, have opened new vistas in science (52, 53). An extremophile group of prokaryotes can fill metabolic niches left behind by the oxygen-using, carboneating eukaryotes (54– 56). This metabolic flexibility leads to the presence of the extremophiles in microbial communities at all strata of life and suggests their possible use as indicators of environmental stresses. Soil samples collected from subsurface levels have different types of microbial communities and hence an intruding catabolic locus or a bacterial species in such samples could be used as indicator species (57). Similarly, mass loading for chemical pollutants from the air affects microbial communities residing over vegetation (58), which changes the structure of the microbial community structure from which the IDS is derived. For example, finding hydrocarbon-degrading bacteria on the surface of a leaf can indicate that the surrounding air is contaminated with exhaust gases from fuel engines.

Genomic tools and the EIA process The EIA process aims at creating a benign scenario in which scientists can monitor developmental activity. The described microbial genomics tools can 362 A ■ ENVIRONMENTAL SCIENCE & TECHNOLOGY / OCTOBER 1, 2003

help in scoping and focusing procedures so that a mitigation plan can be drawn up. The basic exercise would involve the same approach routinely used in any EIA process, that is, moving from scoping, to sampling, to mitigation and monitoring. However, EIA without genomic tools could only be confined to basic resources and their management. Environmental genomics tools can provide an “interactive interpretation” of basic resources, going all the way up to higher flora and fauna and even to the inhabitants. Genomic tools consider the existence, interaction, and survival of different gene pools with reference to each other in complex ecosystems (59). Building new centers of industry is the most common major developmental activity, and the pollution from these nucleated developments always ends up migrating to different locations (4, 60). Hence, from inception, industries are encouraged to follow Integrated Pollution Prevention and Control (IPPC) plans. Different countries have various guidelines to carry out a regular EIA study, which typically include monitoring of crucial parameters and updating their IPPC plans. In some cases, the impacts are obvious and easily measurable. However, even with structured IPPC plans, sometimes the approach misses subtle adverse developmental effects, such as chronic low-level exposure to toxic molecules. In these situations, the system is forced to adapt and may suffer long-term consequences. Even in extreme cases, the analysis of basic resources with microbial genomics tools could aid sustainable mitigation plans. As microbial genomics tools evolve, they will improve the understanding of microbial diversity and their associated biochemical compartmentalization will add new dimensions to the proposed EIA approach. The mathematical and statistical analysis will provide finer dimensions not only to develop monitoring tools but also to evaluate the acquired data. For example, by developing more sensitive tools for target-specific monitoring by PCR methods, we have shown that the primers can be designed on the basis of Shannon’s entropy in the mismatched region in 16S rDNA sequences for tracking Pseudomonas (61). Tools such as multidimensional scaling or the equivalent can be adapted for rapid EIA on the basis of IDS (62, 63). Similarly, to analyze the performance of different components in the microbial community structure of a basic resource, scientists can use tools that handle complex systems and help identify the real stressors that cause impact and observed perturbation (64–67). Hemant J. Purohit is a senior scientist; Dhanajay V. Raje, Atya Kapley, and Parsuraman Padmanabhan are scientists; and Rishi N. Singh is a former director at the National Environmental Engineering Research Institute in Nehru Marg Nagpur, India. Address all correspondence to Purohit at [email protected]

References (1) Wilkins, H. Environ. Impact Assess. Rev. 2003, 23, 401–414. (2) Lenzen, M.; et al. Environ. Impact Assess. Rev. 2003, 23,

263–282. (3) Environmental Impact Assessment for Waste Treatment and Disposal Facilities; Petts, J., Eduljee, G., Eds.; Wiley & Sons: New York, 1994. (4) Ecological Impact Assessment; Treweek, J., Ed.; Blackwell Science: Cambridge, MA, 1999. (5) Singleton, R.; Castle, P.; Short, D. Environmental Assessment; Thomas Telford Publishing: London, 1999. (6) Partidario, M. R. Environ. Impact Assess. Rev. 1996, 16, 31–35. (7) Smit, B.; Spaling, H. Environ. Impact Assess. Rev. 1995, 15, 81–106. (8) Applied Ecology and Environmental Management; Newman, E. I., Ed.; Blackwell Science: Oxford, U.K., 2000. (9) Chapelle, F. H. Ground-Water Microbiology and Geochemistry; Wiley & Sons: New York, 2001. (10) Krivov, S.; Ulanowicz, R. E.; Dhahiya, A. Biosystems 2003, 69, 39–54. (11) Annandale, D.; Taplin, R. Environ. Impact Assess. Rev. 2003, 23, 383–397. (12) Macalady, J.; Banfield, J. F. Earth Planet. Sci. Lett. 2003, 209, 1–17. (13) Rondon, M. R.; Goodman, R. M.; Handelsman, J. Trends Biotechnol. 1999, 17, 403–409. (14) Cherp, A. J. Environ. Manage. 2001, 62, 357–374. (15) Newman, D. K.; Banfield, J. F. Science 2002, 296, 1071–1076. (16) Greer, C. W.; et al. Environ. Sci. Technol. 2001, 35, 365A–370A. (17) DeLong, E. F. Curr. Opin. Microbiol. 2002, 5, 520–524. (18) Dahllof, I. Curr. Opin. Biotechnol. 2002, 13, 213–217. (19) Torsvik, V.; Ovreas, L.; Thingsstad, T. F. Science 2002, 296, 1064–1066. (20) Fenchel, T. Science 2002, 296, 1068–1071. (21) Microbial Diversity in Time and Space; Colwell, R. R., Simidu, U., Ohwada, K., Eds.; Plenum Press: New York, 1996. (22) Kitagava, E.; et al. Environ. Sci. Technol. 2002, 36, 3908–3915. (23) Heller, H.; et al. Environ. Sci. Technol. 2002, 36, 3977–3984. (24) Kaeberlein, T.; Lewis, K.; Epstein, S. S. Science 2002, 296, 1127–1129. (25) Vallino, J. J. Biol. Bull. 2003, 204, 174–179. (26) Baker, G. C.; Tow, L. A.; Cowan, D. A. J. Microbiol. Methods 2003, 53, 157–164. (27) Kapley, A.; Purohit, H. J. Environ. Sci. Pollut. Res. 2001, 8, 89–90. (28) Kapley, A.; Lampel, K.; Purohit, H. J. Water Environ. Res. 2001, 73, 461–465. (29) Purohit, H. J.; Kapley, A. Trends Biotechnol. 2002, 20, 325–326. (30) Gabor, E. M.; de Vries, E. J.; Janssen, D. B. FEMS Microbiol. Ecol. 2003, 44, 153–163. (31) Statistical Genomics: Linkage, Mapping and QTL Analysis; Liu, B. H., Ed.; CRC Press: Boca Raton, FL, 1998. (32) Purohit, H. J.; et al. J. Microbiol. Methods 2003, 52, 315–323. (33) Raje, D. V.; Purohit, H. J.; Singh, R. N. J. Comput. Biol. 2002, 9, 819–829. (34) Pesole, G.; Liuni, S.; D’Souza, M. Bioinformatics 2000, 16, 439–450. (35) Bakermans, C.; Madsen, E. L. Microb. Ecol. 2002, 44, 95–104. (36) Cusick, S. M.; O’Sullivan, D. J. Appl. Environ. Microbiol. 2000, 66, 2227–2231. (37) Padmanabhan, P.; et al. Appl. Environ. Microbiol. 2003, 69, 1614–1622. (38) The Industrial Waste Water System Handbook; Stephenson, R. L., Blackburn, J. B., Eds.; Lewis Publishers: New York, 1998. (39) Radajewski, S.; et al. Nature 2000, 403, 646–649. (40) Hu, S.; et al. Nature 2001, 409, 188–190. (41) Parez, T.; Albornoz, J.; Dominguez, A. Mol. Ecol. 1998, 7, 1347–1357. (42) Stenlid, J.; Vasiliauskas, R. Mol. Ecol. 1998, 7, 1265–1274. (43) Torsvik, V.; Ovreas, L. Curr. Opin. Microbiol. 2002, 5, 240–245. (44) Valinsky, L.; et al. Appl. Environ. Microbiol. 2002, 68, 3243–3250. (45) Greene, E. A.; Voordouw, G. J. Microbiol. Methods 2003, 53, 211–219.

(46) Zhou, J.; Thompson, D. K. Curr. Opin. Microbiol. 2002, 13, 204–207. (47) Taylor, P. M.; et al. FEMS Microbiol. Lett. 2002, 216, 61–66. (48) Kapley, A.; Lampel, K.; Purohit, H. J. Biotechnol. Lett. 2000, 22, 1913–1918. (49) Ashelford, K. E.; Weightman, A. J.; Fry, J. C. Nucleic Acids Res. 2002, 30, 3481–3489. (50) Genetics of Populations; Hedrick, P. W., Ed.; Jones and Bartlett Publishers: Boston, MA, 1999. (51) Top, E. M.; Springael, D.; Boon, N. FEMS Microbiol. Ecol. 2002, 42, 199–208. (52) Macelroy, R. D. Biosystems 1974, 6, 74–75. (53) Rothschild, L. J.; Mancinelli, R. L. Nature 2001, 409, 1092–1101. (54) Nealson, K. H.; Conrad, P. G. Philos. Trans. R. Soc. London, Ser. B 1999, 354 (I-1392), 1923–1939. (55) Bond, P. L.; Smriga, S. P.; Banfield, J. F. Appl. Environ. Microbiol. 2000, 66, 3842–3849. (56) Takai, K.; et al. Appl. Environ. Microbiol. 2002, 68, 3046–3054. (57) Ginn, T. R.; et al. Adv.Water Res. 2002, 25 (8–12), 1017–1042. (58) Grantz, D. A.; Garner, J. H. B.; Johnson, D. W. Environ. Int. 2003, 29 (2–3), 213–239. (59) Purohit, H. J.; Singh, R. N. In Business Opportunities in Biotechnology; Tewari, S., Ed. Confederation of Indian Industry: New Delhi, 2001; pp 231–239. (60) Flynn, S. J.; Löffler, F. E.; Tiedje, J. M. Environ. Sci. Technol. 2000, 34, 1056–1061. (61) Purohit, H.; Raje, D.; Kapley, A. BMC Bioinformatics 2003, 4, 19. (62) Parente, E.; Ricciardi, A. J. Microbiol. Methods 2002, 49, 121–134. (63) Werker, A.; Hall, E. Environmetrics 2002, 13, 347–363. (64) Shanker, R.; Purohit, H. J.; Khanna, P. Bioremediation: Principles and Practice, Bioremediation Technologies; Tecnomic Publishing Co. Inc.: Lancaster-Basel, 1998, pp 81–96. (65) Hanski, I. Science 1999, 283, 1858–1860. (66) Weng, G.; Bhalla, U. S.; Iyengar, R. Science 1999, 284, 92–95. (67) Strogatz, S. H. Nature 2001, 410, 268–276.