OUP user menu

Molecular phylogenetic diversity of the microbial community associated with a high-temperature petroleum reservoir at an offshore oilfield

Hui Li , Shi-Zhong Yang , Bo-Zhong Mu , Zhao-Feng Rong , Jie Zhang
DOI: http://dx.doi.org/10.1111/j.1574-6941.2006.00266.x 74-84 First published online: 1 April 2007


The microbial community and its diversity in production water from a high-temperature, water-flooded petroleum reservoir of an offshore oilfield in China were characterized by 16S rRNA gene sequence analysis. The bacterial and archaeal 16S rRNA gene clone libraries were constructed from the community DNA and, using sequence analysis, 388 bacterial and 220 archaeal randomly selected clones were clustered with 60 and 28 phylotypes, respectively. The results showed that the 16S rRNA genes of bacterial clones belonged to the divisions Firmicutes, Thermotogae, Nitrospiraeand Proteobacteria, whereas the archaeal library was dominated by methanogen-like rRNA genes (Methanothermobacter, Methanobacter, Methanobrevibacterand Methanococcus), with a lower percentage of clones belonging to Thermoprotei. Thermophilic microorganisms were found in the production water, as well as mesophilic microorganisms such as Pseudomonasand Acinetobacter-like clones. The thermophilic microorganisms may be common inhabitants of geothermally heated specialized subsurface environments, which have been isolated previously from a number of high-temperature petroleum reservoirs worldwide. The mesophilic microorganisms were probably introduced into the reservoir as it was being exploited. The results of this work provide further insight into the composition of microbial communities of high-temperature petroleum reservoirs at offshore oilfields.

  • petroleum reservoir
  • 16S rRNA gene diversity
  • microbial community
  • clone library


The deep subsurface petroleum reservoir is generally an extreme environment characterized by high temperature, high pressure, high salinity and anoxic conditions with multiphase fluids of oil, gas and water. Over the past decade, much attention has been paid to the microbial habitats associated with subterranean petroleum-rich strata, because of their scientific significance and industrial applications, such as microbially enhanced oil recovery (Amy & Haldeman, 1997; Krumholz et al., 1997; Magot et al., 2000; Head et al., 2003; Van Hamme et al., 2003; Aitken et al., 2004; Liu et al., 2005). Many physiological types have been recognized from a number of geographically distant oil reservoirs by traditional approaches. These include sulphate reducers (Rueter et al., 1994), sulphidogens (L'Haridon et al., 1995), fermentative microorganisms (Grassia et al., 1996), methanogens (Nilsen & Torsvik, 1996), manganese and iron reducers (Greene et al., 1997; Slobodkin et al., 1999) and acetogens (Davydova-Charakhch'yan et al., 1993). Culture-based methods have long been implemented to improve our understanding of petroleum microbiology and to develop applications, such as oil spill remediation (Prince et al., 1999; Swannell et al., 1999), biofiltration of volatile hydrocarbons (Ergas et al., 1999), microbially enhanced oil recovery (Banat et al., 2000) and oil or fuel upgrading through desulphurization (Setti et al., 1999) and denitrogenation (Benedik et al., 1998).

Nevertheless, our current knowledge of the microbial diversity in such subsurface ecosystems is still limited. The application of molecular techniques, in particular the analysis of the 16S rRNA genes retrieved, has been shown to be effective for the characterization of complex microbial assemblages in environmental samples (Marchesi et al., 2001; Kemp & Aller, 2004). In petroleum microbiology, an increasing application of culture-independent techniques has allowed a more complete characterization of microbial communities in subsurface oil reservoirs. Molecular methods based on reverse sample genome probing, dot blot DNA hybridization with functional gene probes and 16S rRNA gene sequence analysis have been applied to identify the sulphate-reducing bacterial populations inhabiting a low-temperature, water-flooded well in western Canada (Voordouw et al., 1992, 1996). Recently, both 16S rRNA gene sequence phylogenetic analysis and enrichment culture techniques have been used to characterize the thermophilic microbial assemblages in the Miocene Monterey Formation, a prominent high-temperature, oil-bearing formation in California (Orphan et al., 2000). In addition, parallel measurements by culture-based enrichments, 16S rRNA gene sequence analysis and oligonucleotide matrix array hybridization methods have been performed to investigate the key thermophilic bacteria and archaea in a continental high-temperature oil reservoir in western Siberia, Russia (Bonch-Osmolovskaya et al., 2003). A comparable study of the microbial community in a low-temperature, low-salinity, biodegraded petroleum reservoir of the Western Canadian Sedimentary Basin has been reported. This study employed a multidisciplinary approach including chemical and geochemical analyses, biodegradation studies, and culture-based and 16S rRNA gene sequence analyses (Grabowski et al., 2005). Recently, the bacterial community in a high-temperature, water-flooded petroleum reservoir of a continental oilfield in China was studied by the rRNA gene approach (Li et al., 2006).

In this study, the bacterial and archaeal communities and their diversity in a high-temperature, water-flooded petroleum reservoir of the Qinghuang Unit in China were analysed by 16S rRNA gene sequence analysis. The Qinghuang Unit has been operated in primary production since October 2001. Water flooding of the reservoir began in December 2003 and, since then, it has been continuously flooded with recycled production water.

Materials and methods

Sample collection and nucleic acid extraction

The samples of production water were collected directly in sterile steel screw-cap bottles from a sampling valve in the pipeline of the well head in October 2005. The sandy oil-bearing horizon in the target reservoir is at about 1100–1300 m, and the temperature is about 65°C in situ. The salinity of the formation water is about 4226 mg L−1, and its pH is 7.1–7.6. The bottles, which were filled completely with production water (oil–water mixture), were sealed to prevent contamination, immediately transported to the laboratory and stored at 4°C before cell concentration. The oil in the water samples was removed by heating the samples to 70°C for 15 min and by phase separation in a 2-L sterile separatory funnel. Microbial biomass was collected from c. 2 L of the water phase by centrifugation at 15 000 gat 4°C. Total community DNA was extracted from the resulting cell pellets by lysozyme–proteinase K–sodium dodecyl sulphate (SDS) treatment, followed by standard phenol–chloroform extractions (Murray et al., 1998). Nucleic acids were purified with a DNA purification kit (V-gene, China).

Construction of 16S rRNA gene libraries

The DNA from the production water was used to construct bacterial and archaeal libraries. Bacterial and archaeal small-subunit rRNA genes were amplified by PCR using two sets of primers: 8f (5′-AGAGTTTGATYMTGGCTCAG-3′) and 1492r (5′-CGGTTACCTTGTTACGACTT-3′) (Lane, 1991), and 109f (5′-ACKGCTCAGTAACACGT-3′) and 1041r (5′-GGCCATGCACCWCCTCTC-3′) (Reysenbach & Pace, 1995; Grosskopf et al., 1998), respectively. PCR was performed with a ‘reconditioning PCR’ programme (Thompson et al., 2000), and six reactions were carried out for each sample. The PCR products from the same sample were pooled to minimize PCR bias (Polz & Cavanaugh, 1998). The pooled PCR products were purified with a PCR production purification kit (V-gene). The purified amplicons were cloned with a pMD19-T vector kit (Takara, Japan) according to the manufacturer's instructions.

Sequencing and phylogenetic analysis

The plasmid DNA was isolated from selected clones using an AxyPrep-96 Plasmid Kit (Axygen). The rRNA gene inserts were sequenced on an automated ABI 377 sequencer (Dye-Terminator Cycle Sequencing Ready Reaction FS Kit, PE Applied Biosystems, USA) using M13 universal sequencing primers. The resulting sequences (c. 450 bp) were determined for orientation, and then, in a preliminary step, were compared with each other and with sequences in the GenBank database of the National Center for Biotechnology Information (NCBI) (http://www.ncbi.nlm.nih.gov) using the blastnetwork service (Altschul et al., 1997) to determine their approximate phylogenetic affiliations. Sequences that differed by less than 3% were considered to belong to the same phylotype (Stackebrandt & Goebel, 1994). Finally, one to three sequences per phylotype were fully sequenced using M13 universal sequencing primers and the 16S rRNA gene primers 907r and 1392r (Lane, 1991). The sequences (about 1450 bp of bacterial and 950 bp of archaeal rRNA genes) were examined for possible chimeric artefacts using the programs chimera-check(Cole et al., 2003) in the Ribosomal Database Project II (http://rdp.cme.msu.edu/) and Bellerophon (Huber et al., 2004). The sequences without chimeras were initially submitted to the fastaprogram (version 3) (Pearson & Lipman, 1988) to determine their closest phylogenetic relatives. Sequences were aligned to their nearest neighbour using clustal x(Thompson et al., 1997). Phylogenetic trees were constructed on the basis of the Kimura two-parameter model (Kimura, 1980) and the neighbour-joining algorithm (Saitou & Nei, 1987) using the phylippackage (Felsenstein, 2005). Bootstrap analysis with 1000 replicates was applied to assign confidence levels to the nodes of the trees. The consense program was used to generate consense trees, and the treeview program was used to draw the trees (Page, 1996).

Statistical analysis

A rarefaction analysis (Heck et al., 1975), coverage (Good, 1953) and the Shannon index (Wang et al., 2005) were applied to estimate the representation of the phylotypes and to characterize the microbial diversity of water samples. The rarefaction curves were produced with the software analytic rarefaction1.3, which is available online at http://www.uga.edu/∼strata/software/index.html. The coverage of clone libraries was calculated from the equation Embedded Image, where Piis the proportion of each phylotype and irelative to the total number of phylotypes. The Shannon–Wiener index is a general diversity index that is positively correlated with species richness and evenness, and is more sensitive to changes in the abundance of rare species (Chao, 1987).

Nucleotide sequence accession numbers

The GenBank accession numbers for the rRNA gene sequences are as follows: DQ675023DQ675082 for the Qinghuang Oilfield Bacterial (QHO-B) library and DQ785482DQ785509 for the Qinghuang Oilfield Archaeal (QHO-A) library.


The microbial diversity of production water from the petroleum reservoir at the offshore oilfield was examined by calculating the diversity measures of the libraries and by analysing the phylogenetic distribution of 16S rRNA gene clones in bacterial and archaeal libraries.

Diversity measures

Four hundred and thirty-two and 288 white colonies were selected for the bacterial and archaeal libraries, respectively. PCR analysis of the 720 clones using universal plasmid primers revealed 608 positive clones containing the inserts of 16S rRNA gene fragments of the corresponding sizes. The clone libraries obtained were designated QHO-B and QHO-A; they comprised 388 and 220 clones containing the 16S rRNA gene inserts of bacteria and archaea, respectively. In total, 88 phylotypes were identified, with 60 and 28 detected from the bacterial and archaeal libraries, respectively. Using Good's formula, the coverages of archaeal and bacterial libraries were 92.3% and 90.7%, respectively, which indicated that the major part of the diversity in the libraries was detected. However, the asymptote of the rarefaction curves (Fig. 1) suggested that the diversity in the libraries was not sufficient and that there was a need for further sampling. The rarefaction curves also suggested that the sequence population was less diverse for archaeal community but more diverse for bacterial community. This conclusion was further supported by the Shannon–Wiener index, which was 2.33 for archaeal library and 2.90 for bacterial library.


Rarefaction curves generated for 16S rRNA genes in bacterial (△) and archaeal (▲) libraries. Error bars indicate 95% confidence intervals. Clones were grouped into phylotypes at a level of sequence similarity of ≥97%.

Phylogenetic affiliation of bacterial 16S rRNA gene sequences

To determine the bacterial diversity retrieved in the petroleum reservoir, a bacterial 16S rRNA gene library was constructed with the community DNA extracted from production water. Three hundred and eighty-eight clones were partially sequenced (around 450 bp) and categorized on the basis of their sequence similarity (97% identity). At least one clone per phylotype was completely sequenced (around 1500 bp) and the corresponding type sequences were used to construct phylogenetic trees. Of the total of 60 phylotypes, most sequence types had relatively high levels of similarity with their closest counterparts in public databases, but 10 phylotypes showed less than 97% sequence similarity to their nearest database entries and may belong to hitherto unknown phylotypes (Stackebrandt & Goebel, 1994). Comparative analysis of retrieved sequences showed that the phylotypes were representatives of the following divisions: Proteobacteria, Firmicutes, Nitrospirae, Thermotogaeand Sphingobacteria.


One hundred and seventy-five clones (45% of the total clones from the bacterial clone library) were placed into Proteobacteria(Table 1, Fig. 2). Within Alphaproteobacteria, 15 clones represented by QHO-B5 were identified (>98%) as Sinorhizobiumsp. R-25078, a denitrifying strain cultured from activated sludge (Heylen et al., 2006). Eighty clones fell into Gammaproteobacteria(Table 1, Fig. 2). In this group, 38 and 21 clones were closely related (>98%) to Hydrocarboniphaga effusaAP103 and Hydrocarboniphaga effusaAP102, respectively. These species were active in alkane and aromatic hydrocarbon degradation (Palleroni et al., 2004). Fifteen clones were clustered within Betaproteobacteria. In this group, 12 clones were closely related (>98%) to Comamonassp. D22, a strain used in hydrogen oxidizing–denitrifying bioreactors (Smith et al., 2005), and three clones were distantly related (82.7%) to Rhodocyclussp. HOD 5, a strain isolated from a deep borehole (Nazina et al., 2000). Eighteen clones were clustered within Epsilonproteobacteria, and were all closely related (>97%) to the unclassified bacterium APG130A, an Alvinella pompejanasymbiont isolated from a deep-sea hydrothermal vent annelid (Campbell & Cary, 2001). The remaining Proteobacteria-related phylotypes all had less than three clones, which suggested that they were not major components of the bacterial community.

View this table:

Distribution of dominant sequence types from the bacterial and archaeal 16S rRNA gene libraries

Domain, phylumType sequence*Closest cultivated speciesSimilarity (%)Number of clones (%)
FirmicutesQHO-B5Thermoanaerobacter keratinophilus9932 (9.1)
QHO-B52Thermoanaerobacter subterraneusSL99943 (11.1)
QHO-B53Thermovenabulum ferriorganovorum9911 (2.8)
QHO-B46Geobacillus subterraneus34T985 (1.3)
NitrospiraeQHO-B54Nitrospirasp. SRI-9961 (<1)
QHO-B55Thermodesulfovibriosp. TGE-P19914 (3.6)
ThermotogaeQHO-B56Fervidobacterium islandicumAW-1991 (<1)
QHO-B58Thermotogasp. KOL69921 (5.4)
QHO-B59Thermotoga thermarum9956 (14.4)
QHO-B60Thermotogalessp. SRI-19721 (5.4)
ProteobacteriaQHO-B5Sinorhizobiumsp. R-250789815 (3.9)
QHO-B6Rhodocyclussp. HOD 5833 (<1)
QHO-B7Comamonassp. D229812 (3.1)
QHO-B35Hydrocarboniphaga effusaAP1029921 (5.4)
QHO-B44Bacterium APG130A9718 (4.6)
SphingobacteriaQHO-B49Sphingobacteriumsp. MG2913 (<1)
EuryarchaeotaQHO-A15Methanobacterium formicicumFcam9936 (9.3)
QHO-A7Methanothermobacter thermautotrophicusGC-19958 (15)
QHO-A11Methanobacterium subterraneumC2BIS9923 (5.9)
QHO-A25Methanococcus maripaludisS2938 (2.1)
QHO-A2Methanobrevibactersp. Mc30941 (<1)
CrenarchaeotaQHO-A5Candidatus Nitrosopumilus maritimus821 (<1)
  • * Only partial 16S rRNA gene phylotypes are listed.

  • Percentage of 16S rRNA gene similarity to its closest relative.

  • Proportion of clones in the libraries.


Phylogenetic tree of the Proteobacteria16S rRNA gene phylotypes of the Qinghuang Oilfield Bacterial (QHO-B) sequence types (shown in bold) and closely related sequences from the EMBL database. Putative divisions are listed to the right. To save space, only the major clades of Proteobacteria-related clones (data not shown) were used to construct the phylogenetic tree. The topology shown was obtained with the neighbour-joining method. Bootstrap values (n=1000 replicates) of ≥50% are reported. The scale bar represents the number of changes per nucleotide position. Clostridium formicaceticum(X77836) and Clostridium aminobutyricum(X76161) were used as outgroups.


Ninety-five clones (24% of the total clones from the bacterial clone library) were affiliated with Firmicutes, and most clones in this phylum were closely related to the genera Thermoanaerobacterand Thermovenabulum, which are thermophilic microorganisms (Table 1, Fig. 3). Phylotype QHO-B51, representing 32 clones, was closely related (99%) to Thermoanaerobacter keratinophilus, a thermophilic anaerobic bacterium with keratinolytic activity (Riessen & Antranikian, 2001). Phylotype QHO-B52, representing 43 clones, was closely related (99%) to Thermoanaerobacter subterraneusSL9, a thermophilic anaerobe isolated from an oil reservoir in southwest France (Fardeau et al., 2004). Another thermophilic phylotype QHO-B53, displaying 11 clones, was closely related (99%) to Thermovenabulum ferriorganovorum(Zavarzina et al., 2002). Other clones were clustered with the genus Geobacillus, of which phylotype QHO-B46, with only five clones, was closely related (98%) to Geobacillus subterraneus34T, a moderately thermophilic, hydrocarbon-oxidizing strain isolated from formation water of an oilfield in Russia (Nazina et al., 2001).


Phylogenetic tree of the Firmicutes, Nitrospiraand Thermotogae16S rRNA gene phylotypes of the Qinghuang Oilfield Bacterial (QHO-B) sequence types (shown in bold) and closely related sequences from the EMBL database. Putative divisions are listed to the right. The topology shown was obtained with the neighbour-joining method. Bootstrap values (n=1000 replicates) of ≥50% are reported. The scale bar represents the number of changes per nucleotide position. Thermotoga maritima(M21774) and Desulfurobacterium thermolithotrophum(AJ001049) were used as outgroups.


One hundred clones (26% of the total clones from the bacterial clone library) were placed into Thermotogae(Table 1, Fig. 3). Fifty-six clones represented by phylotype QHO-B59 were closely related (99%) to Thermotoga thermarum, a hyperthermophilic strain isolated from the Kubiki oil reservoir in Japan (Takahata et al., 2001). Phylotype QHO-B60, including 21 clones, was closely related (97%) to Thermotogalessp. SRI-1, a strain detected in a high-sulphide, hot spring microbial mat (Skirnisdottir et al., 2000). Other clones in this group were closely related (>98% sequence identity) to previously cultured bacteria, including Thermotogasp. KOL6, Fervidobacterium islandicumand Fervidobacteriumsp. CBS-1.

Other phyla

The remaining 18 clones, falling into two phyla, made up less than 1% of the total bacterial sequences. Fifteen and three clones were assigned to Nitrospiraeand Sphingobacteriaphyla, respectively (Table 1, Fig. 3). Within the Nitrospiraephylum, 14 clones, represented by phylotype QHO-B55, were closely related (99%) to Thermodesulfovibriosp. TGE-P1, a gram-negative, thermophilic, sulphate-reducing strain isolated from thermophilic, anoxic, wastewater treatment processes (AB021302). Another clone in this group, represented by phylotype QHO-B54, was distantly related (96%) to Nitrospirasp. SRI-9 (AF255603). Clones in the Sphingobacteriaphylum, represented by phylotype QHO-B49, were distantly related (<91%) to Sphingobacteriumsp. MG2 (AY556417).

Phylogenetic affiliation of archaeal 16S rRNA gene sequences

To determine the archaeal diversity retrieved from the petroleum reservoir, an archaeal 16S rRNA gene library was constructed with the community DNA extracted from the production water. Two hundred and twenty clones were partially sequenced (around 450 bp) and categorized on the basis of their sequence similarity (97% identity). At least one clone per phylotype was sequenced (around 950 bp), and the corresponding type sequences were used to construct phylogenetic trees. Of the 28 phylotypes in total, half-sequence types had relatively high levels of similarity with their closest counterparts in public databases (Table 1, Fig. 4), whereas 14 phylotypes (50%) showed less than 96% sequence similarity to their nearest database entries and may belong to hitherto unknown phylotypes (Stackebrandt & Goebel, 1994).


Phylogenetic tree of the methanogen 16S rRNA gene phylotypes of the Qinghuang Oilfield Archaeal (QHO-A) sequence types (shown in bold) and closely related sequences from the EMBL database. Putative divisions are listed to the right. The topology shown was obtained with the neighbour-joining method. Bootstrap values (n=1000 replicates) of ≥50% are reported. The scale bar represents the number of changes per nucleotide position. Thermoproteus tenax(M35966) and Desulfurococcus mobilis(X06188) were used as outgroups.

Comparative analysis of the retrieved sequences showed that most sequences were representatives of Methanobacteria. Only one sequence represented by Thermoproteiwas distantly related (82%) to CandidatusNitrosopumilus maritimus’, an autotrophic ammonia-oxidizing marine archaeon (Konneke et al., 2005). Methanogens were phylogenetically related to representatives of the genera Methanothermobacter, Methanobrevibacter, Methanococcusand Methanothermobacter. The phylotype QHO-A15, representing 36 clones, was closely related (>99%) to Methanobacterium formicicumFcam, a culturable methanogen isolated from ricefield soils (Joulian et al., 1998). The most abundant sequence type in this group, QHO-A7, displaying 58 clones, was closely related (>99%) to Methanothermobacter thermautotrophicusGC-1, which was similar to the phylotype found in the Dagang oilfield (Nazina et al., 2006). The phylotype QHO-A11, including 23 clones, was closely related (>99%) to Methanobacterium subterraneumC2BIS, an alkaliphilic, eurythermic, halotolerant methanogen isolated from deep granitic groundwater (Kotelnikova et al., 1998). Within Methanococcus, most sequences were distantly related (<93%) to Methanococcus maripaludisS2.


In this study, molecular analysis was used to characterize the microbial diversity in production water from a high-temperature, water-flooded petroleum reservoir at the Qinghuang offshore oilfield. Although many efforts have been made to define the microbial diversity of petroleum reservoirs in recent years (Orphan et al., 2000; Bonch-Osmolovskaya et al., 2003; Grabowski et al., 2005; Li et al., 2006), this study still revealed many 16S rRNA gene sequences that had not been reported previously from such extreme environments. The data from this study are particularly important in that 24 phylotypes showed less than 97% sequence similarity to any database entry. Compared with other molecular techniques, 16S rRNA gene sequence analysis has the great advantage that the generation of sequence data can be used to design group-specific probes and primers for further studies, such as real-time PCR and FISH, which have not yet been used to characterize the microbial diversity in petroleum reservoirs. We conclude that the level of effort expended in this study was not sufficient to exhaustively sample microbial diversity, but that it was successful in characterizing a large fraction of the oilfield system.

The phylotypes belong mainly to methanogenic archaea, fermentative bacteria and archaea, and sulphate-reducing bacteria, indicating that the phylotype richness is low in the high-temperature petroleum reservoir, as shown previously (Li et al., 2006). The apparently low diversity may be attributed to the extreme conditions of the subsurface petroleum reservoir environment. Indeed, it is widely recognized that microbial diversity is often low in hyperthermal environments, as emphasized by studies in a hot spring where only one or a few bacterial taxa were found (Reysenbach et al., 2000). Nevertheless, geothermally heated oil reservoirs, in which liquid hydrocarbons are the prevailing organic matter, are unique ecological niches for thermophilic anaerobic hydrocarbon degraders and methanogens. The bacterial community of a high-temperature petroleum reservoir have been studied, and have shown that some clones have a number of ecologically beneficial features adapting them to the habitat (Li et al., 2006). The thermophilic bacteria and archaea found were present in the genera Thermoanaerobacter, Thermodesulfovibrio, Thermotoga, Methanothermobacterand Methanococcus. In this group, Thermotoga-related clones were the second most abundant in the QHO-B library, and displayed a large amount of 16S rRNA gene sequence variability. As Thermotoga-related 16S rRNA genes have been detected from multiple production well environments (Orphan et al., 2000; Bonch-Osmolovskaya et al., 2003; Li et al., 2006), it is obvious that the individual population of this group resides in different reservoirs. Within the archaeal library, 16S rRNA gene types belonged to the genera Methanothermobacter, Methanobacter, Methanobrevibacterand Methanococcus, which is consistent with the results of studies of methanogens from high-temperature oilfields obtained by other authors (Mueller & Nielsen, 1996; Nilsen & Torsvik, 1996). The prevalence of methanogen phylotypes in the association studied led us to conclude that thermophilic methanogens may dominate the subsurface ecosystem in these oil reservoirs.

Although this petroleum reservoir system was characterized by a high temperature (65 °C), many phylotypes related to mesophilic microorganisms, such as Desulfothiovibrio, Pseudomonasand Acinetobacter, were detected. It should be noted that bacterial 16S rRNA genes closely related to Hydrocarboniphaga effusa, which is capable of degrading alkanes and aromatic hydrocarbons, were found in an oil reservoir for the first time. This oil reservoir is an open system, as it is continuously flooded with water. Together with the tons of unsterilized water recycled from injection wells to production wells, large numbers of microorganisms would be introduced into the petroleum reservoir. Some may reside in the cooler portions of the reservoir or along the walls or openings of production well tubing, which may result in the detection of mesophilic microorganisms in production water. In addition, the intensive use of molecular techniques for the investigation of microbial ecosystems may introduce biases that may increase or decrease the frequency of a particular phylotype found in the 16S rRNA gene library, but this may not necessarily represent the true proportion of the corresponding species in the original environment (Suzuki et al., 1998). PCR and DNA extraction are two acknowledged sources of methodological bias (Suzuki & Giovannoni, 1996; von Wintzingerode et al., 1997). However, they are the primary sources of information available to assess the phylogenetic richness and complexity of microbial communities. The analysis of this information provides insight into the likely complexity of microbial communities in petroleum reservoirs and inspiration for future studies.


This work was supported by the National Natural Science Foundation of China (50374038, 50574040), the Ministry of Education of China (20030251002) and Shanghai Municipal Science and Technology Commission (045407017).


  • Editor: Michael Wagner


View Abstract