Bioresource Technology 174 (2014) 321–327
Contents lists available at ScienceDirect
Bioresource Technology journal homepage: www.elsevier.com/locate/biortech
Short Communication
Rapid fingerprinting of methanogenic communities by high-resolution melting analysis Jaai Kim, Changsoo Lee ⇑ School of Urban and Environmental Engineering, Ulsan National Institute of Science and Technology (UNIST), 50 UNIST-gil, Eonyang-eup, Ulju-gun, Ulsan 689-798, Republic of Korea
h i g h l i g h t s
g r a p h i c a l a b s t r a c t
A new community fingerprinting
method based on HRM analysis was developed. Melting peak plots provided robust fingerprints for community structure comparison. The new method offers an alternative to monitor variations in community structure.
a r t i c l e
i n f o
Article history: Received 1 September 2014 Received in revised form 7 October 2014 Accepted 9 October 2014 Available online 16 October 2014 Keywords: Denaturing gradient gel electrophoresis High-resolution melting analysis Methanogenic community Microbial community structure Molecular fingerprinting
a b s t r a c t Characterizing microbial community structure using molecular techniques is becoming a popular approach in studies of waste/wastewater treatment processes. A rapid and robust tool to analyze microbial communities is required for efficient process monitoring and control. In this study, a new community fingerprinting method based on high-resolution melting (HRM) analysis was developed and applied to compare methanogenic community structures of five different anaerobic digesters. The new method produced robust community clustering and ordination results comparable to the results from the commonly used denaturing gradient gel electrophoresis (DGGE) performed in parallel. This method transforms melting peak plots (MPs) of community DNA samples generated by HRM analysis to molecular fingerprints and estimates the relationships between the communities based on the fingerprints. The MP-based fingerprinting would provide a good alternative to monitor variations in microbial community structure especially when handling large sample numbers due to its high-throughput capacity and short analysis time. Ó 2014 Elsevier Ltd. All rights reserved.
1. Introduction Biological waste/wastewater treatment is one of the most important and practical applications of biotechnology. The functioning of a biological process basically relies on the collective activity of the microorganisms involved. Microbial community composition is therefore a crucial factor affecting the system ⇑ Corresponding author. Tel.: +82 52 217 2822; fax: +82 52 217 2819. E-mail address: [email protected] (C. Lee). http://dx.doi.org/10.1016/j.biortech.2014.10.037 0960-8524/Ó 2014 Elsevier Ltd. All rights reserved.
performance (Wagner et al., 2002). Although extensive efforts have been made to study the microbial ecology of various environmental bioprocesses, methodological limitations in detecting unculturable populations led to biased and incomplete views of microbial community structure in early studies (Wagner et al., 1993). In virtue of the rapid development of culture-independent molecular techniques in the last two decades, it has become possible to take a closer look at the ‘true’ microbial community structure in complex environmental systems. Although there are still several limitations to overcome, advanced molecular techniques have provided
322
J. Kim, C. Lee / Bioresource Technology 174 (2014) 321–327
exciting insights into the ecophysiology of diverse waste/ wastewater treatment communities in recent studies. An important ecological question in biological waste/ wastewater treatment processes is how microbial community structure changes over time or in response to operating conditions. Molecular fingerprinting techniques, such as denaturing gradient gel electrophoresis (DGGE), terminal restriction fragment length polymorphism (T-RFLP), and single-strand conformation polymorphism (SSCP), have widely been applied to address this question due to the ability to analyze and compare multiple samples simultaneously with high reproducibility (Smalla et al., 2007). These electrophoresis-based techniques separate marker gene
fragments amplified via polymerase chain reaction (PCR) from total DNA samples based on differences in nucleotide sequence, and therefore generate unique separation patterns (i.e., molecular fingerprints) for different microbial communities. DGGE is probably the most commonly used fingerprinting method to study environmental microbial communities including those in waste/ wastewater treatment processes (Marzorati et al., 2008). This technique provides a separation pattern of discrete bands, which each represent a unique population or phylotype (Loisel et al., 2006), on a polyacrylamide gel for each sample analyzed. Each banding profile therefore reflects the structure of a microbial community in the studied environment. The obtained profiles can be
Fig. 1. PCR amplification and HRM analysis of archaeal 16S rRNA genes from the digester samples. Amplification curves (A), melting curves (B), and melting peak plots (C).
J. Kim, C. Lee / Bioresource Technology 174 (2014) 321–327
used to statistically compare different community structures using, for example, clustering and multivariate analyses (Boon et al., 2002; Gilbride et al., 2006; Lee et al., 2008). However, DGGE has several drawbacks such as co-migration of different sequences in a band, potential of gel-to-gel variation, and labor-intensive and time-consuming experimental procedure (Hjelmsø et al., 2014; Kim et al., 2002; Loisel et al., 2006). A viable and reliable alternative to DGGE fingerprinting, is therefore desirable. Recently, alternative assays based on denaturing highperformance liquid chromatography (Kim et al., 2002), flow cytometry (De Roy et al., 2012), and high-resolution melting (Hjelmsø et al., 2014) were demonstrated in microbial ecology studies. High-resolution melting (HRM) is a post-PCR analysis used to differentiate DNA fragments with small sequence differences, for example, in genotyping, mutation scanning, and polymorphism detection (Lin et al., 1997). This technique is based on the use of highly sensitive DNA dye which can separate sequences with even only one nucleotide variation. HRM-based microbial community analysis was first demonstrated only in a very recent study which compared the bacterial community structures in different soil samples based on their melting curve shapes (Hjelmsø et al., 2014). This study introduces an advanced HRM-based fingerprinting strategy which analyzes microbial community structure based on melting peak profiles. The newly developed method was applied to methanogenic community fingerprinting of five lab- and fieldscale anaerobic digesters. The analysis results were compared with the DGGE results performed in parallel for verification.
323
region from 82.0 to 89.5 °C (Hjelmsø et al., 2014), and melting peak plots were generated by taking the negative first derivative of the melting curves ( dRn/dT). HRM data were acquired and analyzed using Quantstudio 12K Flex Software ver. 1.2 (Applied Biosystems). Each DNA sample was analyzed in triplicate. 2.3. DGGE analysis Amplification of archaeal 16S rRNA genes was carried out using the same PCR primers and thermal program as for HRM analysis. For improved resolution of amplicons on a gel, a 40-base GC clamp was added to ARC787F primer (Muyzer et al., 1993). PCR mixtures (50 lL each) were prepared using a PCR premix (AccuPower PCR PreMix, Bioneer, Daejeon, Korea). The obtained PCR products were electrophorized on an 8% (w/v) polyacrylamide gel (denaturant gradient, 35–65%) using a D-code system (Bio-Rad, Hercules, CA) as previously described with minor modifications (Kim et al., 2013). The DGGE gel was then stained with SYBR Safe dye (Molecular Probe, Eugene, OR) and scanned under blue light to visualize the banding patterns. 2.4. Statistical analyses of molecular fingerprints Fluorescence levels (Rn) and dissociation rates ( dRn/dT) measured in each melting curve and melting peak plot were normalized against the average value of the first five Rn readings, i.e., the fluorescence level corresponding to the total DNA
2. Methods 2.1. DNA samples Mixed liquor was collected from five anaerobic digesters under different operating conditions: three 5-L reactors treating whey permeate at 35 °C (RW1), 45 °C (RW2), and 55 °C (RW3), a 2-L reactor treating sewage sludge at 35 °C (RS), and a 7000-m3 field-scale digester treating sewage sludge (FS). Total DNA was extracted from each reactor sample using an automated nucleic acid extractor (Exiprogen, Bioneer, Daejeon, Korea) according to the manufacturer’s instructions. One milliliter of each sample was spun down in a 1.5-mL tube at 13,000 g for 3 min, and the pellet was washed with distilled water by repeated resuspending (in 1 mL), pelleting (13,000 g for 1 min), and decanting (900 lL) to eliminate cell debris and impurities. A 200-lL portion of the biomass suspension was loaded on the extractor with the ExiProgen Bacteria Genomic DNA Kit (Bioneer). The purified DNA was eluted in 200 lL of elution buffer and stored at 20 °C until use. 2.2. HRM analysis PCR amplification and HRM analysis were conducted using a Quantstudio 12K Flex system (Applied Biosystems, Singapore). Archaeal 16S rRNA genes were amplified using ARC787F and ARC1059R primers (Yu et al., 2005) in a two-step thermal program: predenaturation for 10 min at 95 °C followed by 40 cycles of 15 s at 95 °C and 1 min at 60 °C. Each PCR mixture was prepared in a final volume of 20 lL: 1 lL of template DNA, 1 lL of each primer (final concentration, 500 nM), 7 lL of PCR-grade water, and 10 lL of 2 MeltDoctor HRM Master Mix (Applied Biosystems). The resulting amplicons were then denatured at 95 °C for 10 s and renatured at 60 °C for 1 min, and their melting profiles were monitored by measuring changes in fluorescence level with increasing temperature from 60 to 95 °C at a rate of 0.015 °C/s (485 signal acquisitions per sample). Melting curves were constructed by plotting the normalized reporter signal (Rn) versus temperature within a melting
Fig. 2. Archaeal 16S rRNA gene DGGE profiles analyzed from the digester samples.
324
J. Kim, C. Lee / Bioresource Technology 174 (2014) 321–327
amplicons. A matrix was generated by taking the calculated relative Rn (Rnr) values at an interval of 0.2 °C (i.e., separation resolution) from all melting curves. A matrix describing melting peak profiles was also constructed in the same manner based on the relative dRn/dT values. The DGGE image was analyzed using TotalLab 1D image-processing software (TotalLab, Newcastle, UK) for determining the normalized intensity (rolling ball background subtraction, radius = 50) and position of individual bands in each lane. The relative intensity of each band to the total band intensity in each lane was calculated, and the weighted data set for all lanes were converted into a matrix. To comparatively analyze the relationships among the methanogen communities studied, clustering analysis with unweighted pair group method with arithmetic means (UPGMA) and non-metric multidimensional scaling (NMS) were performed on the matrices generated above using PAST software ver. 3.01 and PC-ORD software ver. 5.0 (MjM Software, Gleneden, OR, USA), respectively. Both statistical analyses were carried out based on Bray–Curtis distance measure, a generally recommended measure for community data (McCune and Grace, 2002).
3. Results and discussion The PCR amplification curves of the archaeal 16S rRNA gene from five different community DNA samples are presented in Fig. 1A. All PCR runs show a crossing point before 30 cycles with a clear exponential amplification profile, which is recommended for good HRM analysis (Kim et al., 2002). The melting curves (MCs) and melting peak plot (MPs) within the melting region are shown in Fig. 1B and C, respectively. The samples from different
sources show clearly distinguishable profiles in both graphs, indicating that the methanogenic communities tested were different in melting characteristics and thus in community structure. The MC and MP data were normalized against the initial fluorescence signals in each run. The relative values obtained were then used to construct the matrices for subsequent statistical analyses (refer to Section 2). Fig. 2 shows the archaeal DGGE fingerprints of five different community DNA samples. The five samples showed different banding patterns from one another in terms of the position and intensity of individual bands. This likely reflects the differences in archaeal community structure among the reactors sampled, although visual inspection of a DGGE gel cannot give a quantitative comparison of fingerprints. The digitized DGGE fingerprints were transformed into a matrix of relative intensity of individual bands for further analysis of the relationships between the archaeal communities studied. Three cluster trees constructed respectively based on the data matrices prepared from the MCs, MPs, and DGGE profiles (DPs) are shown in Fig. 3. The MP and DP cluster trees show very similar clustering patterns. Two mesophilic methanogenic communities treating sewage sludge (i.e., RS and FS) are closely clustered and relatively distantly located from the others treating whey permeate (i.e., RW1–3), likely reflecting the influence of different substrates. This corresponds to the appearance of several relatively low-GC sequences (i.e., the bands in the upper part of the gel) only in the whey permeate-feeding communities, with the other two communities being relatively simple with less bands, in the DGGE analysis (Fig. 2). Among the communities treating whey permeate, mesophilic RW1 is distantly related to thermophilic RW2 and RW3 which are closely clustered together. This may reflect the effect of operating temperature. Interestingly, RW1 shows a distinct MP
Fig. 3. Cluster trees of the sampled methanogenic communities constructed based on the MC (A), MP (B), and DP (C) fingerprints.
J. Kim, C. Lee / Bioresource Technology 174 (2014) 321–327
325
Fig. 4. NMS ordination plots of the sampled methanogenic communities created based on the MC (A), MP (B), and DP (C) fingerprints.
profile with a clear peak around 83.5 °C, not observed in RW2 and 3 (Fig. 1C), according with the occurrence of a low-GC band (i.e., a lower melting temperature) with strong intensity exclusively in the RW1 lane of the DGGE gel (Fig. 2). On the other hand, the
MC cluster tree displays a markedly different pattern from the other trees, with poor grouping of replicate samples. This reproducibility problem of MC fingerprints is more clearly seen in RS and RW3 samples distantly spread in the tree, possibly leading to
326
J. Kim, C. Lee / Bioresource Technology 174 (2014) 321–327
inaccurate clustering results. These suggest that the community clustering based on the MP fingerprints produced more robust and accurate results than did the MC-based one. The NMS ordination plots of the first two dimensions are shown in Fig. 4. NMS reduces a fingerprint profile generated from a microbial community (i.e., a DGGE lane, an MC, or an MP in this study) into one point in a low-dimensional space so that communities with similar structures are located close together in the space (Fromin et al., 2002). Therefore, variations in community structure among different samples can be visualized in an ordination plot. All NMS plots constructed in this study showed very high explainability (>93%) of the variance among the analyzed archaeal community structures (i.e., the cumulative r2 for the ordination axes > 0.93), with the stress and instability being sufficiently low for statistically meaningful ordination analysis (McCune and Grace, 2002). The NMS results are in good agreement with the clustering analysis results. The NMS plots based on the MP and DP fingerprints produced similar ordination patterns with clear spatial differentiation of the methanogenic communities from different digesters. The MP-based NMS shows a robust clustering of replicate samples, whereas the community profiles are not clearly differentiated (or clustered) according to their sources by the MC-based NMS. This suggests that the MP fingerprints were more appropriate than the MC fingerprints for reliable NMS ordination of methanogenic community structures in the tested anaerobic digester samples. The overall experimental results demonstrated that HRM analysis can produce molecular fingerprints, based on MP profiles, for comparing microbial community structures. Although HRM analysis does not offer options for microbial identification or phylogenetic affiliation of the community members, it can be useful for rapid scanning of a large number of samples for community structure due to its high-throughput capacity and short assay time (Hjelmsø et al., 2014). The newly proposed MP-based community fingerprinting method in this study can be an efficient alternative to DGGE for, for example, monitoring temporal and spatial variations in methanogenic (and likely also other microbial) community structures in waste/wastewater treatment environments, including anaerobic digesters. In such analyses, good differentiation between different community structures is crucial for reliable results. Theoretically, amplified DNA fragments with different melting temperatures shall appear as separate melting peaks or bands in HRM or DGGE analysis, respectively. However, similar DNA sequences from close relatives may show similar melting (or denaturing) characteristics and not be distinguished as separate melting peaks by HRM analysis, and DGGE also has similar resolution issues due to co-migrating and overlapping bands (Vallaeys et al., 1997). Particularly in an MP fingerprint, poor separation of melting peaks can result in shouldering (or peak broadening), which makes the fingerprint difficult to interpret (Caux-Moncoutier et al., 2011). This problem can be more pronounced when dealing with complex environmental communities generally composed of a wide variety of microorganisms (Figs. 1 and 2). To overcome the limitation, in this study, the methanogenic community structures of five different digesters were compared based on the variations in their MPs within a defined melting region, rather than by analyzing the plots on a peak-by-peak basis. The similarity between the MPs was estimated in quantitative terms by Bray–Curtis distance measure, and the MP-based community fingerprinting was successful in explaining the relationships between the methanogenic community structures studied (Figs. 3 and 4). Although high-throughput sequencing, e.g., pyrosequencing, is increasingly employed in microbial ecology studies due to its ability to generate a large database providing a detailed description of community structure and sequence information at the same time, it is still very costly and time-consuming to build a sizable sequence data collection for reliable community analysis
(Bolzonella et al., 2005). A fast, easy, and inexpensive tool to characterize microbial community structure is therefore needed, particularly in studies of waste/wastewater systems, and the new fingerprinting method proposed in this study would be suitable for such use. 4. Conclusions HRM-based community fingerprinting methods, based on the MC or MP profiles of 16S rRNA gene fragments, were developed and tested for community clustering in comparison to the widely used DGGE analysis. The MP-based fingerprinting method demonstrated a robust and reliable differentiation of methanogenic communities from different anaerobic digesters. This new method is therefore suggested to have high potential as an alternative to DGGE for analyzing structural differences between microbial communities. It provides an attractive tool particularly when handling large sample numbers because it offers high-throughput and rapid analysis. In contrast, the MC-based fingerprinting failed to show robust community clustering performance. Acknowledgements This research was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Science ICT and Future Planning (2014R1A1A1002329) and also by Korea Ministry of Environment (MOE) through a Waste-to-Energy Human Resource Development Project. The authors thank Dr. Jae-Won Jung (Life Technologies Korea LLC) for technical support and helpful discussion on HRM analysis. References Bolzonella, D., Pavan, P., Battistoni, P., Cecchi, F., 2005. Mesophilic anaerobic digestion of waste activated sludge: influence of the solid retention time in the wastewater treatment process. Proc. Biochem. 40, 1453–1460. Boon, N., Windt, W.D., Verstraete, W., Top, E.M., 2002. Evaluation of nested PCR-DGGE (denaturing gradient gel electrophoresis) with group-specific 16S rRNA primers for the analysis of bacterial communities from different wastewater treatment plants. FEMS Microbiol. Ecol. 39, 101–112. Caux-Moncoutier, V., Castéra, L., Tirapo, C., Michaux, D., Rémon, M.-A., Laugé, A., Rouleau, E., De Pauw, A., Buecher, B., Gauthier-Villars, M., Viovy, J.-L., StoppaLyonnet, D., Houdayer, C., 2011. EMMA, a cost- and time-effective diagnostic method for simultaneous detection of point mutations and large-scale genomic rearrangements: application to BRCA1 and BRCA2 in 1,525 patients. Human Mutat. 32, 325–334. De Roy, K., Clement, L., Thas, O., Wang, Y., Boon, N., 2012. Flow cytometry for fast microbial community fingerprinting. Water Res. 46, 907–919. Fromin, N., Hamelin, J., Tarnawski, S., Roesti, D., Jourdain-Miserez, K., Forestier, N., Teyssier-Cuvelle, S., Gillet, F., Aragno, M., 2002. Statistical analysis of denaturing gel electrophoresis (DGE) fingerprinting patterns. Environ. Microbiol. 4, 634– 643. Gilbride, K.A., Frigon, D., Cesnik, A., Gawat, J., Fulthorpe, R.R., 2006. Effect of chemical and physical parameters on a pulp mill biotreatment bacterial community. Water Res. 40, 775–787. Hjelmsø, M.H., Hansen, L.H., Bælum, J., Feld, L., Holben, W.E., Jacobsen, C.S., 2014. High resolution melt analysis for rapid comparison of bacterial community composition. Appl. Environ. Microbiol. 80, 3568–3575. Kim, J., Lee, S., Lee, C., 2013. Comparative study of changes in reaction profile and microbial community structure in two anaerobic repeated-batch reactors started up with different seed sludges. Bioresour. Technol. 129, 495–505. Kim, M., Ahn, Y.-H., Speece, R.E., 2002. Comparative process stability and efficiency of anaerobic digestion; mesophilic vs. thermophilic. Water Res. 36, 4369–4385. Lee, C., Kim, J., Do, H., Hwang, S., 2008. Monitoring thiocyanate-degrading microbial community in relation to changes in process performance in mixed culture systems near washout. Water Res. 42, 1254–1262. Lin, J.-G., Chang, C.-N., Chang, S.-C., 1997. Enhancement of anaerobic digestion of waste activated sludge by alkaline solubilization. Bioresour. Technol. 62, 85–90. Loisel, P., Harmand, J., Zemb, O., Latrille, E., Lobry, C., Delgenès, J.-P., Godon, J.-J., 2006. Denaturing gradient electrophoresis (DGE) and single-strand conformation polymorphism (SSCP) molecular fingerprintings revisited by simulation and used as a tool to measure microbial diversity. Environ. Microbiol. 8, 720–731.
J. Kim, C. Lee / Bioresource Technology 174 (2014) 321–327 Marzorati, M., Wittebolle, L., Boon, N., Daffonchio, D., Verstraete, W., 2008. How to get more out of molecular fingerprints: practical tools for microbial ecology. Environ. Microbiol. 10, 1571–1581. McCune, B., Grace, J.B., 2002. Analysis of Ecological Communities. MjM Software Design, Glenden Beach, OR, USA. Muyzer, G., Waal, E.C., Uitterlinden, A.G., 1993. Profiling of complex microbial populations by denaturing gradient gel electrophoresis analysis of polymerase chain reaction-amplified genes coding for 16S rRNA. Appl. Environ. Microbiol. 59, 695–700. Smalla, K., Oros-Sichler, M., Milling, A., Heuer, H., Baumgarte, S., Becker, R., Neuber, G., Kropf, S., Ulrich, A., Tebbe, C.C., 2007. Bacterial diversity of soils assessed by DGGE, T-RFLP and SSCP fingerprints of PCR-amplified 16S rRNA gene fragments: do the different methods provide similar results? J. Microbiol. Methods 69, 470–479.
327
Vallaeys, T., Topp, E., Muyzer, G., Macheret, V., Laguerre, G., Rigaud, A., Soulas, G., 1997. Evaluation of denaturing gradient gel electrophoresis in the detection of 16S rDNA sequence variation in rhizobia and methanotrophs. FEMS Microbiol. Ecol. 24, 279–285. Wagner, M., Amann, R., Lemmer, H., Schleifer, K.H., 1993. Probing activated sludge with oligonucleotides specific for proteobacteria: inadequacy of culturedependent methods for describing microbial community structure. Appl. Environ. Microbiol. 59, 1520–1525. Wagner, M., Loy, A., Nogueira, R., Purkhold, U., Lee, N., Daims, H., 2002. Microbial community composition and function in wastewater treatment plants. Antonie Van Leeuwenhoek 81, 665–680. Yu, Y., Lee, C., Kim, J., Hwang, S., 2005. Group-specific primer and probe sets to detect methanogenic communities using quantitative real-time polymerase chain reaction. Biotechnol. Bioeng. 89, 670–679.
Contents lists available at ScienceDirect
Bioresource Technology journal homepage: www.elsevier.com/locate/biortech
Short Communication
Rapid fingerprinting of methanogenic communities by high-resolution melting analysis Jaai Kim, Changsoo Lee ⇑ School of Urban and Environmental Engineering, Ulsan National Institute of Science and Technology (UNIST), 50 UNIST-gil, Eonyang-eup, Ulju-gun, Ulsan 689-798, Republic of Korea
h i g h l i g h t s
g r a p h i c a l a b s t r a c t
A new community fingerprinting
method based on HRM analysis was developed. Melting peak plots provided robust fingerprints for community structure comparison. The new method offers an alternative to monitor variations in community structure.
a r t i c l e
i n f o
Article history: Received 1 September 2014 Received in revised form 7 October 2014 Accepted 9 October 2014 Available online 16 October 2014 Keywords: Denaturing gradient gel electrophoresis High-resolution melting analysis Methanogenic community Microbial community structure Molecular fingerprinting
a b s t r a c t Characterizing microbial community structure using molecular techniques is becoming a popular approach in studies of waste/wastewater treatment processes. A rapid and robust tool to analyze microbial communities is required for efficient process monitoring and control. In this study, a new community fingerprinting method based on high-resolution melting (HRM) analysis was developed and applied to compare methanogenic community structures of five different anaerobic digesters. The new method produced robust community clustering and ordination results comparable to the results from the commonly used denaturing gradient gel electrophoresis (DGGE) performed in parallel. This method transforms melting peak plots (MPs) of community DNA samples generated by HRM analysis to molecular fingerprints and estimates the relationships between the communities based on the fingerprints. The MP-based fingerprinting would provide a good alternative to monitor variations in microbial community structure especially when handling large sample numbers due to its high-throughput capacity and short analysis time. Ó 2014 Elsevier Ltd. All rights reserved.
1. Introduction Biological waste/wastewater treatment is one of the most important and practical applications of biotechnology. The functioning of a biological process basically relies on the collective activity of the microorganisms involved. Microbial community composition is therefore a crucial factor affecting the system ⇑ Corresponding author. Tel.: +82 52 217 2822; fax: +82 52 217 2819. E-mail address: [email protected] (C. Lee). http://dx.doi.org/10.1016/j.biortech.2014.10.037 0960-8524/Ó 2014 Elsevier Ltd. All rights reserved.
performance (Wagner et al., 2002). Although extensive efforts have been made to study the microbial ecology of various environmental bioprocesses, methodological limitations in detecting unculturable populations led to biased and incomplete views of microbial community structure in early studies (Wagner et al., 1993). In virtue of the rapid development of culture-independent molecular techniques in the last two decades, it has become possible to take a closer look at the ‘true’ microbial community structure in complex environmental systems. Although there are still several limitations to overcome, advanced molecular techniques have provided
322
J. Kim, C. Lee / Bioresource Technology 174 (2014) 321–327
exciting insights into the ecophysiology of diverse waste/ wastewater treatment communities in recent studies. An important ecological question in biological waste/ wastewater treatment processes is how microbial community structure changes over time or in response to operating conditions. Molecular fingerprinting techniques, such as denaturing gradient gel electrophoresis (DGGE), terminal restriction fragment length polymorphism (T-RFLP), and single-strand conformation polymorphism (SSCP), have widely been applied to address this question due to the ability to analyze and compare multiple samples simultaneously with high reproducibility (Smalla et al., 2007). These electrophoresis-based techniques separate marker gene
fragments amplified via polymerase chain reaction (PCR) from total DNA samples based on differences in nucleotide sequence, and therefore generate unique separation patterns (i.e., molecular fingerprints) for different microbial communities. DGGE is probably the most commonly used fingerprinting method to study environmental microbial communities including those in waste/ wastewater treatment processes (Marzorati et al., 2008). This technique provides a separation pattern of discrete bands, which each represent a unique population or phylotype (Loisel et al., 2006), on a polyacrylamide gel for each sample analyzed. Each banding profile therefore reflects the structure of a microbial community in the studied environment. The obtained profiles can be
Fig. 1. PCR amplification and HRM analysis of archaeal 16S rRNA genes from the digester samples. Amplification curves (A), melting curves (B), and melting peak plots (C).
J. Kim, C. Lee / Bioresource Technology 174 (2014) 321–327
used to statistically compare different community structures using, for example, clustering and multivariate analyses (Boon et al., 2002; Gilbride et al., 2006; Lee et al., 2008). However, DGGE has several drawbacks such as co-migration of different sequences in a band, potential of gel-to-gel variation, and labor-intensive and time-consuming experimental procedure (Hjelmsø et al., 2014; Kim et al., 2002; Loisel et al., 2006). A viable and reliable alternative to DGGE fingerprinting, is therefore desirable. Recently, alternative assays based on denaturing highperformance liquid chromatography (Kim et al., 2002), flow cytometry (De Roy et al., 2012), and high-resolution melting (Hjelmsø et al., 2014) were demonstrated in microbial ecology studies. High-resolution melting (HRM) is a post-PCR analysis used to differentiate DNA fragments with small sequence differences, for example, in genotyping, mutation scanning, and polymorphism detection (Lin et al., 1997). This technique is based on the use of highly sensitive DNA dye which can separate sequences with even only one nucleotide variation. HRM-based microbial community analysis was first demonstrated only in a very recent study which compared the bacterial community structures in different soil samples based on their melting curve shapes (Hjelmsø et al., 2014). This study introduces an advanced HRM-based fingerprinting strategy which analyzes microbial community structure based on melting peak profiles. The newly developed method was applied to methanogenic community fingerprinting of five lab- and fieldscale anaerobic digesters. The analysis results were compared with the DGGE results performed in parallel for verification.
323
region from 82.0 to 89.5 °C (Hjelmsø et al., 2014), and melting peak plots were generated by taking the negative first derivative of the melting curves ( dRn/dT). HRM data were acquired and analyzed using Quantstudio 12K Flex Software ver. 1.2 (Applied Biosystems). Each DNA sample was analyzed in triplicate. 2.3. DGGE analysis Amplification of archaeal 16S rRNA genes was carried out using the same PCR primers and thermal program as for HRM analysis. For improved resolution of amplicons on a gel, a 40-base GC clamp was added to ARC787F primer (Muyzer et al., 1993). PCR mixtures (50 lL each) were prepared using a PCR premix (AccuPower PCR PreMix, Bioneer, Daejeon, Korea). The obtained PCR products were electrophorized on an 8% (w/v) polyacrylamide gel (denaturant gradient, 35–65%) using a D-code system (Bio-Rad, Hercules, CA) as previously described with minor modifications (Kim et al., 2013). The DGGE gel was then stained with SYBR Safe dye (Molecular Probe, Eugene, OR) and scanned under blue light to visualize the banding patterns. 2.4. Statistical analyses of molecular fingerprints Fluorescence levels (Rn) and dissociation rates ( dRn/dT) measured in each melting curve and melting peak plot were normalized against the average value of the first five Rn readings, i.e., the fluorescence level corresponding to the total DNA
2. Methods 2.1. DNA samples Mixed liquor was collected from five anaerobic digesters under different operating conditions: three 5-L reactors treating whey permeate at 35 °C (RW1), 45 °C (RW2), and 55 °C (RW3), a 2-L reactor treating sewage sludge at 35 °C (RS), and a 7000-m3 field-scale digester treating sewage sludge (FS). Total DNA was extracted from each reactor sample using an automated nucleic acid extractor (Exiprogen, Bioneer, Daejeon, Korea) according to the manufacturer’s instructions. One milliliter of each sample was spun down in a 1.5-mL tube at 13,000 g for 3 min, and the pellet was washed with distilled water by repeated resuspending (in 1 mL), pelleting (13,000 g for 1 min), and decanting (900 lL) to eliminate cell debris and impurities. A 200-lL portion of the biomass suspension was loaded on the extractor with the ExiProgen Bacteria Genomic DNA Kit (Bioneer). The purified DNA was eluted in 200 lL of elution buffer and stored at 20 °C until use. 2.2. HRM analysis PCR amplification and HRM analysis were conducted using a Quantstudio 12K Flex system (Applied Biosystems, Singapore). Archaeal 16S rRNA genes were amplified using ARC787F and ARC1059R primers (Yu et al., 2005) in a two-step thermal program: predenaturation for 10 min at 95 °C followed by 40 cycles of 15 s at 95 °C and 1 min at 60 °C. Each PCR mixture was prepared in a final volume of 20 lL: 1 lL of template DNA, 1 lL of each primer (final concentration, 500 nM), 7 lL of PCR-grade water, and 10 lL of 2 MeltDoctor HRM Master Mix (Applied Biosystems). The resulting amplicons were then denatured at 95 °C for 10 s and renatured at 60 °C for 1 min, and their melting profiles were monitored by measuring changes in fluorescence level with increasing temperature from 60 to 95 °C at a rate of 0.015 °C/s (485 signal acquisitions per sample). Melting curves were constructed by plotting the normalized reporter signal (Rn) versus temperature within a melting
Fig. 2. Archaeal 16S rRNA gene DGGE profiles analyzed from the digester samples.
324
J. Kim, C. Lee / Bioresource Technology 174 (2014) 321–327
amplicons. A matrix was generated by taking the calculated relative Rn (Rnr) values at an interval of 0.2 °C (i.e., separation resolution) from all melting curves. A matrix describing melting peak profiles was also constructed in the same manner based on the relative dRn/dT values. The DGGE image was analyzed using TotalLab 1D image-processing software (TotalLab, Newcastle, UK) for determining the normalized intensity (rolling ball background subtraction, radius = 50) and position of individual bands in each lane. The relative intensity of each band to the total band intensity in each lane was calculated, and the weighted data set for all lanes were converted into a matrix. To comparatively analyze the relationships among the methanogen communities studied, clustering analysis with unweighted pair group method with arithmetic means (UPGMA) and non-metric multidimensional scaling (NMS) were performed on the matrices generated above using PAST software ver. 3.01 and PC-ORD software ver. 5.0 (MjM Software, Gleneden, OR, USA), respectively. Both statistical analyses were carried out based on Bray–Curtis distance measure, a generally recommended measure for community data (McCune and Grace, 2002).
3. Results and discussion The PCR amplification curves of the archaeal 16S rRNA gene from five different community DNA samples are presented in Fig. 1A. All PCR runs show a crossing point before 30 cycles with a clear exponential amplification profile, which is recommended for good HRM analysis (Kim et al., 2002). The melting curves (MCs) and melting peak plot (MPs) within the melting region are shown in Fig. 1B and C, respectively. The samples from different
sources show clearly distinguishable profiles in both graphs, indicating that the methanogenic communities tested were different in melting characteristics and thus in community structure. The MC and MP data were normalized against the initial fluorescence signals in each run. The relative values obtained were then used to construct the matrices for subsequent statistical analyses (refer to Section 2). Fig. 2 shows the archaeal DGGE fingerprints of five different community DNA samples. The five samples showed different banding patterns from one another in terms of the position and intensity of individual bands. This likely reflects the differences in archaeal community structure among the reactors sampled, although visual inspection of a DGGE gel cannot give a quantitative comparison of fingerprints. The digitized DGGE fingerprints were transformed into a matrix of relative intensity of individual bands for further analysis of the relationships between the archaeal communities studied. Three cluster trees constructed respectively based on the data matrices prepared from the MCs, MPs, and DGGE profiles (DPs) are shown in Fig. 3. The MP and DP cluster trees show very similar clustering patterns. Two mesophilic methanogenic communities treating sewage sludge (i.e., RS and FS) are closely clustered and relatively distantly located from the others treating whey permeate (i.e., RW1–3), likely reflecting the influence of different substrates. This corresponds to the appearance of several relatively low-GC sequences (i.e., the bands in the upper part of the gel) only in the whey permeate-feeding communities, with the other two communities being relatively simple with less bands, in the DGGE analysis (Fig. 2). Among the communities treating whey permeate, mesophilic RW1 is distantly related to thermophilic RW2 and RW3 which are closely clustered together. This may reflect the effect of operating temperature. Interestingly, RW1 shows a distinct MP
Fig. 3. Cluster trees of the sampled methanogenic communities constructed based on the MC (A), MP (B), and DP (C) fingerprints.
J. Kim, C. Lee / Bioresource Technology 174 (2014) 321–327
325
Fig. 4. NMS ordination plots of the sampled methanogenic communities created based on the MC (A), MP (B), and DP (C) fingerprints.
profile with a clear peak around 83.5 °C, not observed in RW2 and 3 (Fig. 1C), according with the occurrence of a low-GC band (i.e., a lower melting temperature) with strong intensity exclusively in the RW1 lane of the DGGE gel (Fig. 2). On the other hand, the
MC cluster tree displays a markedly different pattern from the other trees, with poor grouping of replicate samples. This reproducibility problem of MC fingerprints is more clearly seen in RS and RW3 samples distantly spread in the tree, possibly leading to
326
J. Kim, C. Lee / Bioresource Technology 174 (2014) 321–327
inaccurate clustering results. These suggest that the community clustering based on the MP fingerprints produced more robust and accurate results than did the MC-based one. The NMS ordination plots of the first two dimensions are shown in Fig. 4. NMS reduces a fingerprint profile generated from a microbial community (i.e., a DGGE lane, an MC, or an MP in this study) into one point in a low-dimensional space so that communities with similar structures are located close together in the space (Fromin et al., 2002). Therefore, variations in community structure among different samples can be visualized in an ordination plot. All NMS plots constructed in this study showed very high explainability (>93%) of the variance among the analyzed archaeal community structures (i.e., the cumulative r2 for the ordination axes > 0.93), with the stress and instability being sufficiently low for statistically meaningful ordination analysis (McCune and Grace, 2002). The NMS results are in good agreement with the clustering analysis results. The NMS plots based on the MP and DP fingerprints produced similar ordination patterns with clear spatial differentiation of the methanogenic communities from different digesters. The MP-based NMS shows a robust clustering of replicate samples, whereas the community profiles are not clearly differentiated (or clustered) according to their sources by the MC-based NMS. This suggests that the MP fingerprints were more appropriate than the MC fingerprints for reliable NMS ordination of methanogenic community structures in the tested anaerobic digester samples. The overall experimental results demonstrated that HRM analysis can produce molecular fingerprints, based on MP profiles, for comparing microbial community structures. Although HRM analysis does not offer options for microbial identification or phylogenetic affiliation of the community members, it can be useful for rapid scanning of a large number of samples for community structure due to its high-throughput capacity and short assay time (Hjelmsø et al., 2014). The newly proposed MP-based community fingerprinting method in this study can be an efficient alternative to DGGE for, for example, monitoring temporal and spatial variations in methanogenic (and likely also other microbial) community structures in waste/wastewater treatment environments, including anaerobic digesters. In such analyses, good differentiation between different community structures is crucial for reliable results. Theoretically, amplified DNA fragments with different melting temperatures shall appear as separate melting peaks or bands in HRM or DGGE analysis, respectively. However, similar DNA sequences from close relatives may show similar melting (or denaturing) characteristics and not be distinguished as separate melting peaks by HRM analysis, and DGGE also has similar resolution issues due to co-migrating and overlapping bands (Vallaeys et al., 1997). Particularly in an MP fingerprint, poor separation of melting peaks can result in shouldering (or peak broadening), which makes the fingerprint difficult to interpret (Caux-Moncoutier et al., 2011). This problem can be more pronounced when dealing with complex environmental communities generally composed of a wide variety of microorganisms (Figs. 1 and 2). To overcome the limitation, in this study, the methanogenic community structures of five different digesters were compared based on the variations in their MPs within a defined melting region, rather than by analyzing the plots on a peak-by-peak basis. The similarity between the MPs was estimated in quantitative terms by Bray–Curtis distance measure, and the MP-based community fingerprinting was successful in explaining the relationships between the methanogenic community structures studied (Figs. 3 and 4). Although high-throughput sequencing, e.g., pyrosequencing, is increasingly employed in microbial ecology studies due to its ability to generate a large database providing a detailed description of community structure and sequence information at the same time, it is still very costly and time-consuming to build a sizable sequence data collection for reliable community analysis
(Bolzonella et al., 2005). A fast, easy, and inexpensive tool to characterize microbial community structure is therefore needed, particularly in studies of waste/wastewater systems, and the new fingerprinting method proposed in this study would be suitable for such use. 4. Conclusions HRM-based community fingerprinting methods, based on the MC or MP profiles of 16S rRNA gene fragments, were developed and tested for community clustering in comparison to the widely used DGGE analysis. The MP-based fingerprinting method demonstrated a robust and reliable differentiation of methanogenic communities from different anaerobic digesters. This new method is therefore suggested to have high potential as an alternative to DGGE for analyzing structural differences between microbial communities. It provides an attractive tool particularly when handling large sample numbers because it offers high-throughput and rapid analysis. In contrast, the MC-based fingerprinting failed to show robust community clustering performance. Acknowledgements This research was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Science ICT and Future Planning (2014R1A1A1002329) and also by Korea Ministry of Environment (MOE) through a Waste-to-Energy Human Resource Development Project. The authors thank Dr. Jae-Won Jung (Life Technologies Korea LLC) for technical support and helpful discussion on HRM analysis. References Bolzonella, D., Pavan, P., Battistoni, P., Cecchi, F., 2005. Mesophilic anaerobic digestion of waste activated sludge: influence of the solid retention time in the wastewater treatment process. Proc. Biochem. 40, 1453–1460. Boon, N., Windt, W.D., Verstraete, W., Top, E.M., 2002. Evaluation of nested PCR-DGGE (denaturing gradient gel electrophoresis) with group-specific 16S rRNA primers for the analysis of bacterial communities from different wastewater treatment plants. FEMS Microbiol. Ecol. 39, 101–112. Caux-Moncoutier, V., Castéra, L., Tirapo, C., Michaux, D., Rémon, M.-A., Laugé, A., Rouleau, E., De Pauw, A., Buecher, B., Gauthier-Villars, M., Viovy, J.-L., StoppaLyonnet, D., Houdayer, C., 2011. EMMA, a cost- and time-effective diagnostic method for simultaneous detection of point mutations and large-scale genomic rearrangements: application to BRCA1 and BRCA2 in 1,525 patients. Human Mutat. 32, 325–334. De Roy, K., Clement, L., Thas, O., Wang, Y., Boon, N., 2012. Flow cytometry for fast microbial community fingerprinting. Water Res. 46, 907–919. Fromin, N., Hamelin, J., Tarnawski, S., Roesti, D., Jourdain-Miserez, K., Forestier, N., Teyssier-Cuvelle, S., Gillet, F., Aragno, M., 2002. Statistical analysis of denaturing gel electrophoresis (DGE) fingerprinting patterns. Environ. Microbiol. 4, 634– 643. Gilbride, K.A., Frigon, D., Cesnik, A., Gawat, J., Fulthorpe, R.R., 2006. Effect of chemical and physical parameters on a pulp mill biotreatment bacterial community. Water Res. 40, 775–787. Hjelmsø, M.H., Hansen, L.H., Bælum, J., Feld, L., Holben, W.E., Jacobsen, C.S., 2014. High resolution melt analysis for rapid comparison of bacterial community composition. Appl. Environ. Microbiol. 80, 3568–3575. Kim, J., Lee, S., Lee, C., 2013. Comparative study of changes in reaction profile and microbial community structure in two anaerobic repeated-batch reactors started up with different seed sludges. Bioresour. Technol. 129, 495–505. Kim, M., Ahn, Y.-H., Speece, R.E., 2002. Comparative process stability and efficiency of anaerobic digestion; mesophilic vs. thermophilic. Water Res. 36, 4369–4385. Lee, C., Kim, J., Do, H., Hwang, S., 2008. Monitoring thiocyanate-degrading microbial community in relation to changes in process performance in mixed culture systems near washout. Water Res. 42, 1254–1262. Lin, J.-G., Chang, C.-N., Chang, S.-C., 1997. Enhancement of anaerobic digestion of waste activated sludge by alkaline solubilization. Bioresour. Technol. 62, 85–90. Loisel, P., Harmand, J., Zemb, O., Latrille, E., Lobry, C., Delgenès, J.-P., Godon, J.-J., 2006. Denaturing gradient electrophoresis (DGE) and single-strand conformation polymorphism (SSCP) molecular fingerprintings revisited by simulation and used as a tool to measure microbial diversity. Environ. Microbiol. 8, 720–731.
J. Kim, C. Lee / Bioresource Technology 174 (2014) 321–327 Marzorati, M., Wittebolle, L., Boon, N., Daffonchio, D., Verstraete, W., 2008. How to get more out of molecular fingerprints: practical tools for microbial ecology. Environ. Microbiol. 10, 1571–1581. McCune, B., Grace, J.B., 2002. Analysis of Ecological Communities. MjM Software Design, Glenden Beach, OR, USA. Muyzer, G., Waal, E.C., Uitterlinden, A.G., 1993. Profiling of complex microbial populations by denaturing gradient gel electrophoresis analysis of polymerase chain reaction-amplified genes coding for 16S rRNA. Appl. Environ. Microbiol. 59, 695–700. Smalla, K., Oros-Sichler, M., Milling, A., Heuer, H., Baumgarte, S., Becker, R., Neuber, G., Kropf, S., Ulrich, A., Tebbe, C.C., 2007. Bacterial diversity of soils assessed by DGGE, T-RFLP and SSCP fingerprints of PCR-amplified 16S rRNA gene fragments: do the different methods provide similar results? J. Microbiol. Methods 69, 470–479.
327
Vallaeys, T., Topp, E., Muyzer, G., Macheret, V., Laguerre, G., Rigaud, A., Soulas, G., 1997. Evaluation of denaturing gradient gel electrophoresis in the detection of 16S rDNA sequence variation in rhizobia and methanotrophs. FEMS Microbiol. Ecol. 24, 279–285. Wagner, M., Amann, R., Lemmer, H., Schleifer, K.H., 1993. Probing activated sludge with oligonucleotides specific for proteobacteria: inadequacy of culturedependent methods for describing microbial community structure. Appl. Environ. Microbiol. 59, 1520–1525. Wagner, M., Loy, A., Nogueira, R., Purkhold, U., Lee, N., Daims, H., 2002. Microbial community composition and function in wastewater treatment plants. Antonie Van Leeuwenhoek 81, 665–680. Yu, Y., Lee, C., Kim, J., Hwang, S., 2005. Group-specific primer and probe sets to detect methanogenic communities using quantitative real-time polymerase chain reaction. Biotechnol. Bioeng. 89, 670–679.
