Structural Determination of a Key Exopolysaccharide in Mixed Culture

Oct 29, 2010 - Nuclear magnetic resonance (NMR) techniques were used to elucidate the structure of an exopolysaccharide material previously revealed t...
0 downloads 11 Views 678KB Size
Download PDF
Environ. Sci. Technol. 2010, 44, 8964–8970

Structural Determination of a Key Exopolysaccharide in Mixed Culture Aerobic Sludge Granules Using NMR Spectroscopy THOMAS SEVIOUR,† LYNETTE K. LAMBERT,‡ M A I T E P I J U A N , § A N D Z H I G U O Y U A N * ,† The University of Queensland, Advanced Water Management Centre (AWMC), St. Lucia, QLD 4072, Australia, The University of Queensland, Centre for Advanced Imaging, St. Lucia, QLD 4072, Australia, and Catalan Institute for Water Research (ICRA), Technological Park of University of Girona, 17003, Spain

Received August 4, 2010. Revised manuscript received October 12, 2010. Accepted October 14, 2010.

Nuclear magnetic resonance (NMR) techniques were used to elucidate the structure of an exopolysaccharide material previously revealed to be important in formation of aerobic granules. The 1D NMR spectral data acquired showed that this gel-forming polysaccharide was a major component of granular EPS, while 1D and 2D NMR spectra showed it consisted of eight sugar residues. These were assigned as R-galactose, R-rhamnose, 2-acetoamido-2-deoxy-R-galactopyranuronic acid, β-mannose, β-galactose, β-glucuronate, β-glucosamine, and N-acetyl β-galactosamine. With the exception of 2-acetoamido-2-deoxy-R-galactopyranuronic acid, a highly unusual sugar, their presence was confirmed with highperformance anion-exchange chromatography with pulsed amperometric detection (HPAEC-PAD). Carbon and proton shifts were assigned for each sugar. Heteronuclear multiple bond correlation (HMBC) and nuclear Overhauser enhancement spectroscopy (NOESY) were used to identify linkage sites between individual sugar residues. This gel-forming exopolysaccharide appeared to be a highly complex single heteropolysaccharide with a repeat sequence of R-galactose, β-mannose, β-glucosamine, N-acetyl-β-galactosamine, and 2-acetoamido2-deoxy-R-galactopyranuronic acid. It has a disaccharide branch of β-galactose and β-glucuronic acid attached to 2-acetoamido-2-deoxy-R-galactopyranuronic acid and an R-rhamnose branch attached to R-galactose.

Introduction Aerobic sludge granules are dense microbial community aggregates with the potential to provide higher intensity secondary wastewater treatment than conventional floccular aggregates (1). They share the same functionality regarding their ability to perform carbon, nitrogen, and phosphorus removal (2). However, the mechanisms for their formation * Corresponding author phone: +61-7-33654730; fax +61-733654726; e-mail: [email protected]. † Advanced Water Management Centre. ‡ Centre for Advanced Imaging. § Catalan Institute for Water Research (ICRA). 8964

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 23, 2010

are still poorly understood. To ensure more reliable operation, recent research has focused on answering how and why they form (3-5). Particular attention has been paid to the role of the extracellular polymeric substances (EPS) (6, 7) in granulation. Their importance in granule formation and stability has been attributed to certain EPS properties including hydrophobicity and porosity (8, 9). Several other EPS structural features have also been cited as being important in granulation, although the literature is often contradictory to their precise roles. These include polysaccharide overproduction (10), the existence of a protein core (11), and an outer EPS layer with a polysaccharide backbone (12). To clarify the role of EPS, Seviour et al. (13, 14) characterized their granules treating abattoir wastewater as hydrogels and demonstrated that the possession of a gel-forming EPS distinguished granules from flocs. They then fractionated granule EPS and showed that only the high molecular weight (MW) exopolysaccharide fraction displayed analogous gellike behavior (15). In the absence of any detailed structural characterization it was not clear whether it was a single or complex of several different polysaccharides. Lin et al. (16) attempted chemical characterization of the gel-forming exopolysaccharide component in their granules treating domestic wastewater. They described it as alginate-like, based on its propensity to form gels in the presence of calcium chloride, a high abundance of uronic acid residues, and its FT-IR spectrum. A detailed structure of any gel-forming granule exopolysaccharide is, however, yet to be elucidated. Nuclear magnetic resonance (NMR) spectroscopy can resolve repeat unit structures of regular and heterogeneous polysaccharides using a combination of 1D and 2D NMR techniques, including correlation spectroscopy (COSY), total correlation spectroscopy (TOCSY), heteronuclear single quantum coherence (HSQC), heteronuclear multiple bond correlation (HMBC), and nuclear Overhauser enhancement spectroscopy (NOESY) (17). NMR has been applied to study exopolysaccharides produced by a bacterium isolated from wastewater aggregates (18) however not from a mixed microbial aggregate and not resulting in a full description of the chemical structure of the repeating unit/s. The aim of this paper was to resolve the complete structural organization of the gel-forming exopolysaccharide identified by Seviour et al. (15). The application of 1D and 2D NMR spectroscopy shows it consists of a single, complex heteropolysaccharide. Relating chemical structure to its established function makes an important and novel contribution to our current understanding of the role of EPS in granule structure.

Materials and Methods Reactor Operation and Microbial Community in Granules. The aerobic sludge granules used in this study were sampled from a lab-scale sequencing batch reactor (SBR) treating abattoir wastewater in an enhanced biological phosphorus (EBPR) process configuration. The influent had average chemical oxygen demand (COD), total nitrogen (N), and total phosphorus (P) concentrations of approximately 600, 230, and 35 mg/L, respectively. Details of the SBR operation and sludge characteristics can be found in Yilmaz et al. (19) and Lemaire et al. (6). For more details of SBR operation and sludge sampling refer to the Supporting Information. EPS Sample Preparation. Recovered biomass aggregates were washed with chloroform/methanol (2:1, 1:1, 1:2 v/v) and freeze dried (FTS Thermal Systems, NY). The EPS was solubilized as described by Seviour et al. (14). Lyophilized 10.1021/es102658s

 2010 American Chemical Society

Published on Web 10/29/2010

products were suspended in 0.1 M NaOH, incubated at 4 °C (2 h), and centrifuged (15 min, 12 000 rpm). For the crude EPS preparation (Preparation 1) the centrate was dialyzed (molecular weight cutoff [MWCO] 3500) against Milli-Q water and freeze dried. To enrich for exopolysaccharides (Preparation 2), perchloric acid (70%) was added to the centrate for protein removal. Polysaccharides were captured by fractional precipitation with cetylpyridinium chloride (CPC) and methanol, recovered by centrifugation, dialyzed, and freeze dried as described by Seviour et al. (14). To isolate the high-MW exopolysaccharide fraction (Preparation 3), Preparation 2 was redissolved in 0.025 M NaOH (10 mg mL-1) and processed by gel permeation chromatography (GPC) (15). The highMW exopolysaccharide fraction was dialyzed against Milli-Q and freeze dried. Nuclear Magnetic Resonance Spectroscopy. Exopolysaccharide material (3-16 mg) was lyophilized and dissolved in 1000 µL of 0.1 M NaOH in D2O. NMR spectra were recorded on a Bruker Avance 750 spectrometer operating at 749.28 MHz for 1H and 188.42 MHz for 13C, using a 5 mm tripleresonance inverse probe equipped with a z gradient. An additional 13C spectrum was acquired on an Avance 500 spectrometer using a 5 mm broad-band probe. Chemical shifts are reported in ppm referenced to trimethylsilyl propionate (TSP). All spectra were acquired using Bruker standard pulse sequences. One-dimensional (1D) TOCSY spectra were obtained with mixing times 30-120 ms using a 10 kHz spin-locking field (20). Two-dimensional (2D) homonuclear double-quantum-filtered-COSY, gradientselected-COSY, and NOESY (20, 21) with a mixing time of 200 ms were obtained. 2D 1H, 13C heteronuclear experiments included the following spectra: multiplicity-edited HSQC (22) and HSQC-TOCSY (20) with a 120 ms mixing time and spinlocking field of 8.3 kHz. HMBC spectra (21) were acquired with an evolution delay of 6 and 8 Hz. For the HSQC-NOESY spectrum (21), a 200 ms mixing time was used. All spectra were recorded at 333 K because of increased resolution, except for the 1D TOCSY spectrum for the anomeric proton with a proton shift corresponding to that of water at 333 K, which was undertaken at 313 K.

FIGURE 1. 1D 1H NMR spectra of EPS Preparation 1 (10 mg/mL) (A), Preparation 2 (16 mg/mL) (B), and Preparation 3 (3 mg/mL) (C) in D2O at 333 K.

Results and Discussion Profiling the NMR Spectra of the Exopolysaccharide. The 1D 1H NMR spectra of the crude EPS (Preparation 1), EPS enriched for exopolysaccharide by fractional precipitation (Preparation 2), and the high-MW exopolysaccharide isolate (Preparation 3) are presented in Figure 1A, 1B, and 1C, respectively. The profile of the spectrum of the high-MW exopolysaccharide isolate is typical of polysaccharides including gellan, galactomannan, poly N-acetyl glucosamine, and heparin sulfate (23-26). Ignoring the water peak, which dominates with δH 4.42 at 60 °C (where δH is the 1H chemical shift in ppm), this high-MW exopolysaccharide signal is also clearly a major component of the crude EPS NMR spectrum (Figure 1A). The broad peaks in the 1H chemical shift range 3.0-0.9 ppm in the 1D spectrum of the crude EPS are indicative of saturated primary, secondary, and tertiary hydrogens and therefore consistent with the presence of high-MW lipid and protein (27). Their removal in Preparations 2 and 3 verifies the process of selective enrichment of the high-MW exopolysaccharide. The profiles of the polysaccharide regions of the spectra remain unchanged during the purification process, showing that this high-MW exopolysaccharide was the dominant exopolysaccharide in the crude EPS. This highMW exopolysaccharide, shown to be exclusively responsible for the characteristic gel-forming properties of the granular EPS (15), therefore appears to be overproduced in the granule’s EPS. Given that no marked difference was found between the spectra of Preparation 2 and Preparation 3,

FIGURE 2. HSQC spectrum of EPS Preparation 2 in D2O at 333 K (16 mg/mL) showing characteristic polysaccharide 13C and 1H shifts. Preparation 2 was used in subsequent 2D experiments requiring higher concentrations for structural determination of the gel-forming high-MW exopolysaccharide. In the HSQC spectrum at 333 K (Figure 2) there are 7 resolved cross peaks with δH and δC characteristic of anomeric positions (17) (δC is 13C chemical shift in ppm). Refer to Figure SI-1, Supporting Information, for the full 13C spectrum. The HSQC undertaken at 313 K revealed an additional anomeric cross peak at δH 4.44, δC 107.5. Therefore, the gel-forming exopolysaccharide comprises a minimum of eight major spin systems pertaining to monosaccharide residues. From the integration of the anomeric protons in the 1D spectrum, there were approximately equal quantities of each monosaccharide. In the HSQC a cross peak appears at δH 1.32, δC 17.0 characteristic of the methyl group of a deoxy sugar such as rhamnose. A cross peak at δH 2.71, δC 59.0 is in the region of the H2 and C2 resonances of nonacetylated hexosamines VOL. 44, NO. 23, 2010 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

8965

FIGURE 3. 1D TOCSY spectra for each of the eight spin systems identified for EPS Preparation 2 (16 mg/mL) in D2O, including r-rhamnose (A), β-glucosamine (B), r-galactose (C), 2-acetoamido-2-deoxy-r-galactopyranuronic acid (D), N-acetyl-β-galactosamine (E), β-mannose (F), β-glucuronic acid (G), and β-galactose (H). The irradiated peak for each subspectrum is identified with an asterisk (*). The peak identified by the dagger (†) is due to lactate impurity. Note that spectra A and B and spectra C-H are presented on different axes. All spectra were collected at 333 K with the exception of β-glucosamine (313 K). (e.g., glucosamine or galactosamine). The two cross peaks at δH 4.29, δC 51.8 and δH 4.01, δC 54.1 are consistent with the H2 and C2 resonances of acetylated hexosamines, suggesting that two hexosamines may be N-acetylated. This is supported further by the presence of a signal at δH 2.04, δC 22.9, which indicates an N-acetyl methyl group (18, 24). Identifying Monosaccharide Residues Through NMR Analysis. The eight major sugar residues of the gel-forming exopolysaccharide were identified by NMR analysis as R-galactose, R-rhamnose, 2-acetoamido-2-deoxy-R-galactopyranuronic acid, β-mannose, β-galactose, β-glucuronate, β-glucosamine, and N-acetyl-β-galactosamine. Details of the assignment are provided below. The presence of the neutral sugars rhamnose, mannose, and galactose and the acidic sugar glucuronic acid was confirmed by high-performance anion-exchange chromatography with pulsed amperometric detection (HPAEC-PAD). For more details on HPAEC-PAD, refer to the Supporting Information. Galactosamine and glucosamine were confirmed with Waters AccQTag Ultra chemistry (Waters). A subspectrum for each sugar residue was generated by 1D TOCSY (Figure 3). For more details of 1D TOCSY, refer to the Supporting Information. From the sizes of the coupling constants estimated in the 1D TOCSY spectra in conjunction with 1H,1H-COSY (Figures SI-3 and SI-4, Supporting Information) it was possible to identify the base sugar unit for each spin system. One-bond CH couplings (1JC,H) were also determined from a 13C-coupled HSQC experiment to confirm 8966

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 23, 2010

the stereochemistry at the anomeric carbons (refer to Table SI-1, Supporting Information, for one-bond anomeric CH couplings). 1JC1,H1 values of ∼170 Hz indicate equatorial protons at C1 (i.e., R) and 1JC1,H1 ≈ 160 Hz axial protons (i.e., β) (17). By overlaying the HSQC and HSQC-TOCSY spectra, assignment of the protonated carbons was achieved (for more details refer to the Supporting Information). Refer to Table 1 for a full list of 13C and 1H assignments as well as coupling constants where they could be determined. However, only R-rhamnose was able to be identified, including full 1H peak assignment, using just the 1H,1H-COSY and 1D TOCSY spectra (Figure 3A) (refer to the Supporting Information for a full description). For all other sugar residues a combination of techniques in addition to these spectra was required. With β-glucosamine, for example, the signals for the two H6 protons in the 1D TOCSY spectrum (Figure 3B) are weak, and so HSQC-TOCSY (discussed later, Figure SI-5, Supporting Information) was also required to confirm the H6 shifts. For β-mannose (Figure 3F) HSQC-TOCSY and HSQC spectra were also required to identify chemical shifts at positions 3-5. The HMBC spectrum (Figure SI-6A, discussed later) was then required to assign position 6 (refer to the Supporting Information for more information). The HMBC identified the β-glucuronic acid residue as carboxylated at C5, and to differentiate between β-glucuronic and β-galacturonic acid, the 1H shifts were compared to published shifts for the two residues (28) (refer to the Supporting Information). For the remaining sugars it was possible to identify signals attributable to H1 to H4 from their 1D TOCSY spectra. However, H4 signals for each sugar were narrow, indicating small couplings to H3 and H5. TOCSY transfer through H4 is inefficient in galactose-based sugars because H4 is equatorial. It was thus possible to identify the sugars as galactose-based. However, for full assignment HMBC and HSQC-NOESY (spectrum not shown) experiments were required to connect H4 with H5, with additional assistance from HSQC-NOESY to connect H5 with H6. For more details on the assignment of R-galactose, β-galactose, N-acetyl-βgalactosamine, and 2-acetoamido-2-deoxy-R-galactopyranuronic acid refer to the Supporting Information. Intermonosaccharide Connections. The 2D HMBC experiments construct a 2D correlation map between the chemical shifts of long-range coupled 1H and 13C. The correlations are observed over two and three bonds. Hence, intraresidue three-bond correlations can be observed between the anomeric carbon and the proton at the linkage point of the adjacent sugar and correspondingly between the anomeric proton and the carbon at the linkage point. By analyzing the HMBC spectrum it was thus possible to determine the complete sequence of sugar residues in the gel-forming exopolysaccharide, with only a repeat connection unable to be observed. Table 2 summarizes the intermonosaccharide correlations determined from the HMBC experiments (see Figure SI-6A, Supporting Information, for the HMBC spectrum showing cross peaks with the monosaccharide anomeric protons). The remaining cross peaks observed in the HMBC are due to intrasugar couplings. Although no correlation was observed to β-GlcN C1, there is a very clear cross peak from β-GlcN H1 to β-GalNAc C3, strongly indicating a β-GlcN-(1f3)-β-GalNAc linkage. In the case of the anomeric of R-GalANAc, however, no HMBC correlation was detected to either C1 or H1. A NOESY spectrum of the exopolysaccharide was therefore used to identify a possible repeat connection through R-GalANAc (see Figure SI-6B, Supporting Information). NOESY provides through-space correlations between protons that are positioned near each other and are not necessarily connected

TABLE 1. 13C and 1H Chemical Shifts and Coupling Constants for Sugar Residues Present in Exopolysaccharide from Aerobic Sludge Granules sugar R-galactose (R-Gal)

R-rhamnose (R-Rha)

2-acetoamido-2-deoxy-R-galactopyranuronic acid (R-GalANAc)

N-acetyl β-galactosamine (β-GalNAc)

β-mannose (β-Man)

β-glucuronic acid (β-GlcA)

β-glucosamine (β-GlcN)

β-galactose (β-Gal)

a

proton shift (ppm) H1 H2 H3 H4 H5 H6 H1 H2 H3 H4 H5 H6 H1 H2 H3 H4 H5

5.66 3.95 4.04 4.10 3.96 3.73 5.03 4.06 3.78 3.49 3.88 1.32 4.97 4.30 4.13 4.66 4.71

H1 H2 H3 H4 H5 H6 H1 H2 H3 H4 H5 H6 H1 H2 H3 H4 H5

4.84 4.02 3.84 4.11 3.59 3.65/3.84a 4.75 4.01 3.90 3.89 3.55 3.79/3.93 4.69 3.44 3.54 3.54 3.73

H1 H2 H3 H4 H5 H6 H1 H2 H3 H4 H5 H6

4.44 2.71 3.53 3.66 3.53 3.74/3.86 4.60 3.73 3.80 4.22 3.72 3.80/3.83

carbon shift (ppm) C1 C2 C3 C4 C5 C6 C1 C2 C3 C4 C5 C6 C1 C2 C3 C4 C5 C6 C1 C2 C3 C4 C5 C6 C1 C2 C3 C4 C5 C6 C1 C2 C3 C4 C5 C6 C1 C2 C3 C4 C5 C6 C1 C2 C3 C4 C5 C6

100.1 77.3 71.1 80.9 74.9 63.1 104.8 72.7 73.1 74.8 72.1 19.5 101.2 51.8 80.6 80.3 75.0 178.1 104.1 54.2 83.6 71.1 77.1 64.0 102.9 73.7 76.8 74.0 77.6 63.9 106.5 76.2 78.4 74.6 78.7 178.3 107.5 59.0 76.9 81.6 77.4 63.2 107.4 73.1 85.7 71.1 77.5 64.0

coupling constant (Hz) JH1,H2 JH2,H3 JH3,H4

JH2,H3 JH3,H4 JH4,H5 JH5,H6

4.0 11.0 2.7

3.5 9.69 0.6 6.3

JH1,H2 JH2,H3 JH3,H4

4.5 11.0 3.6

JH1,H2 JH2,H3 JH3,H4

8.3 9.5 3.0

JH1,H2 JH2,H3 JH4,H5

8.1 8.7 7.8

JH1,H2 JH2,H3 JH3,H4

8.1 8.7 9.4

JH1,H2 JH2,H3 JH3,H4

8.8 8.8 3.3

Tentative assignment.

through bonds (17). A HSQC-NOESY spectrum was also acquired to aid assignment of overlapping H1 cross peaks in the NOESY spectrum. Refer to Table 2 for a summary of the observed NOESY cross peaks with the monosaccharide anomeric protons. These provide further evidence for the linkages proposed from the HMBC spectrum (Figure SI-6A, Supporting Information). Therefore, based on interconnections between all monosaccharides, the gel-forming exopolysaccharide component of these aerobic sludge granules is a single compound. Given that it is high MW (15), it is likely that the monomer sequence is repeated. From the HMBC alone it was not possible to identify a repeat connection. Linkages through C1 were obtained for all sugars except R-GalANAc. In the HSQC-NOESY experiment only two intersugar cross peaks were identified to R-GalANAc H1 to R-Gal H4 and H6. In the more sensitive NOESY spectrum, weak and medium intensity inter-residue cross

peaks were identified from R-GalANAc H1 to R-Gal H1 and H3, with another medium intensity cross peak to δH 3.95. It was not possible to distinguish this as being due to R-Gal H2, H5, or both (δH 3.95 and 3.96, respectively). Strong NOESY cross peaks were also observed to δH 4.1 and δH 3.7, which were attributed from the HSQC-NOESY spectrum to R-Gal H4 and H6. On the basis of intersugar NOESY cross peaks with R-GalANAc protons (refer to Table SI-2, Supporting Information) the other possible repeat connection is to R-Rha. However, there are no NOESY cross peaks between R-GalANAc H1 and any R-Rha protons. Coupled with the NOESY cross peaks between R-Gal and R-GalANAc, this suggests that the probable repeat connection is between R-GalANAc and R-Gal. Additionally, 13C shifts for a connected sugar are moved downfield by 4-10 ppm at anomeric and linked positions relative to an equivalent free monosaccharide, whereas the change is much smaller at unmodified positions (17). Compared to a free residue (28, 29), in this polysaccharide VOL. 44, NO. 23, 2010 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

8967

FIGURE 4. Description (A) and illustration (B) of the proposed structure of the repeat unit of exopolysaccharide from aerobic sludge granules. r-GalANAc is 2-acetoamido-2-deoxy-r-galactopyranuronic acid.

TABLE 2. Intersugar Residue Correlations Observed from HMBC and NOESY Spectraa anomeric HMBC cross peaks H1 R-Gal

C3 β-Man

C1

NOESY cross peaks to H1

H3 β-Man

H1 β-Man H2 β-Man H3 β-Man H4 β-Man H5 β-Man H5 R-GalANAc H1 R-Rha H3 R-Rha β-GlcA C3 β-Gal H3 β-Gal H3 β-Gal H4 β-Gal β-GlcN C3 β-GalNAc H2 β-GalNAc H3 β-GalNAc H4 R-Gal/H4 β-GalNAc β-GalNAc C4 R-GalNAc A H4 R-GalNAc A H4 R-GalANAc H5 R-GalANAc H1 β-Gal β-Man C4 β-GlcN H4 β-Glc H4 β-GlcN R-Rha C2 R-Gal H2 R-Gal H1 R-Gal H2/H5 R-Gal H3 R-Gal H4 R-GalANAc H5 R-GalANAc R-GalANAc H1 R-Gal H2/H5 R-Gal H3 R-Gal H4bR-Gal/H4 β-GalNAc H2 β-Gal/H6bR-Gal β-Gal C3 R-GalANAc H3 R-GalNAc A H3 R-GalANAc a / indicates overlap of cross peaks in the NOESY spectrum. b Indicates where an overlapping NOESY cross peak has been resolved from the HSQC-NOESY.

δC1, δC2, and δC4 for R-Gal have been moved downfield by ∼7, 8, and 10 ppm, respectively, with other R-Gal13C shifts moved by less than 4 ppm. This is consistent with R-Gal being linked through positions 1 and 2 (linkages already described from HMBC) and position 4. All 13C shifts of R-Rha on the other hand are within 3 ppm of those for a free residue except for the linked position C1 (28, 30). R-Rha therefore likely exists as a branch through position 1 to R-Gal C2 and could be close spatially with R-GalANAc, which would explain the 8968

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 23, 2010

NOESY cross peaks between H1 of R-Rha and R-GalANAc (i.e., to H4 and H5). The likely repeat connection, however, is through position 4 of R-Gal (i.e., R-GalANAc-(1f4)-R-Gal). The proposed structure of the repeat unit of the gelforming exopolysaccharide in aerobic sludge granules is given in Figure 4. The absolute stereochemistry was not determined for these sugars. R-Rha has been drawn in the L-configuration however, as it occurs in nature almost exclusively in the L-form. All other sugar residues have been drawn in the D form (31). Implications of Findings. EPS have been shown to be important to the structure and formation of aerobic sludge granules (7, 9). However, information regarding their exact structural role and the contribution of individual EPS is limited (10-12). Seviour et al. (13, 14) demonstrated that aerobic sludge granules were physical hydrogels and that they were distinguished from flocs by the expression of a gel-forming EPS. Seviour et al. (15) fractionated the gel forming by GPC and showed that it comprised three major fractions including an exclusively high-MW polysaccharide component. Only this high-MW polysaccharide fraction displayed the gel-forming behavior characteristic of the granules and their EPS, indicating that it was structurally important (15). Here the full structure of this exopolysaccharide is described, showing the gel-forming properties of granule EPS to be attributable to a single exopolysaccharide. A popular approach to granule EPS characterization studies is to identify the distribution and role of general macromolecular groups with fluorescently labeled probes targeting particular functional chemical groups or following changes brought about after exposing them to degradative enzymes targeting suspected polymeric substrates like R-polysaccharides, proteins, or nucleic acids (11, 12). Another approach is to attribute an importance to a general macromolecular group by colorimetric assays (10, 11). The first approach depends on the specificity of the selected probes and enzymes, and the other depends on the choice of suitable colorimetric standards. Both approaches are limited by a lack of understanding of EPS composition and structure, which potentially distorts the conclusions drawn. This is the first time the full structure of a macromolecule from aerobic sludge granules with an established structural function has been described.

Lin et al. (16) reported the only other attempt to structurally characterize a granular EPS, yet they described it only in terms of its similarities to another known gel-forming exopolysaccharide (i.e., alginate). This would appear to differ from the structure of the exopolysaccharide revealed in this study. NMR structural analysis of their exopolysaccharide would be required to confirm the differences; however, there are notable similarities. The gel-forming exopolysaccharide in this study also contains charged residues, including uronic acids. Similar to alginate, these charged residues could therefore form active sites for intermolecular cross-linking (32). It remains to be established whether the exopolysaccharide performs a similar role in other sludge granules or only the granules analyzed in the study. The NMR spectrum of EPS from floccular sludge (data not shown) indicates that this exopolysaccharide is not synthesized in floccular sludge. This is evident from the absence of any of the anomeric peaks that are diagnostic in identifying the monosaccharides comprising the gel-forming exopolysaccharide. However, process configuration and operation, functionality of the sludge granules, and the microbial population (6, 19) are all typical of wastewater treatment processes performing biological nutrient removal, suggesting that this gel-forming exopolysaccharide is unlikely to be unique in its distribution.

Acknowledgments This work was funded by the Environmental Biotechnology Cooperative Research Centre (EBCRC) Pty Ltd., Australia. T.S. is an Australian Postgraduate Award and EBCRC Scholarship recipient. Sugar analysis was facilitated by access to the Australian Proteome Analysis Facility supported under the Australian Government’s National Collaborative Research Infrastructure Strategy (NCRIS). We would also like to thank Prof. Robert Seviour, La Trobe University, for helpful discussions and valuable input.

Note Added after ASAP Publication This paper was published ASAP on October 29, 2010. There were minor text updates in Materials and Methods and Results and Discussion sections. The revised paper was reposted on November 9, 2010.

Supporting Information Available Technical information regarding structural determination, including one-bond anomeric CH couplings (Table SI-1), details of intersugar NOESY cross peaks (Table SI-2), and cross-referenced NMR spectra, i.e., 13C (Figure SI-1), COSY (Figure SI-3,4), HSQC-TOCSY (Figure SI-5), HMBC and NOESY (Figures SI-6 and SI-7); there are also additional details regarding reactor operation and monosaccharide analyses including HPAEC-PAD chromatograms (Figure SI-2). This material is available free of charge via the Internet at http:// pubs.acs.org.

Literature Cited (1) Liu, Y.; Tay, J.-H. State of the art of biogranulation technology for wastewater treatment. Biotechnol. Adv. 2004, 22 (7), 533– 563. (2) de Kreuk, M. K.; Heijnen, J. J.; van Loosdrecht, M. C. M. Simultaneous COD, nitrogen, and phosphate removal by aerobic granular sludge. Biotechnol. Bioeng. 2005, 90 (6), 761–769. (3) Wang, Z.-W.; Li, Y.; Zhou, J.-Q.; Liu, Y. The influence of shortterm starvation on aerobic granules. Proc. Biochem. 2006, 41 (12), 2373–2378. (4) Liu, Y.; Tay, J.-H. The essential role of hydrodynamic shear force in the formation of biofilm and granular sludge. Water Res. 2002, 36 (7), 1653–1665. (5) Ren, T.-T.; Liu, L.; Sheng, G.-P.; Liu, X.-W.; Yu, H.-Q.; Zhang, M.-C.; Zhu, J.-R. Calcium spatial distribution in aerobic granules and its effects on granule structure, strength and bioactivity. Water Res. 2008, 42 (13), 3343–3352.

(6) Lemaire, R.; Webb, R. I.; Yuan, Z. Micro-scale observations of the structure of aerobic microbial granules used for the treatment of nutrient-rich industrial wastewater. ISME J 2008, 2 (5), 528–541. (7) Liu, Y.-Q.; Liu, Y.; Tay, J.-H. The effects of extracellular polymeric substances on the formation and stability of biogranules. Appl. Microbiol. Biotechnol. 2004, 65 (2), 143–148. (8) Adav, S. S.; Chang, C.-H.; Lee, D.-J. Hydraulic characteristics of aerobic granules using size exclusion chromatography. Biotechnol. Bioeng. 2008, 99 (4), 791–799. (9) Tay; Tay, J. H.; Liu; Liu, Q. S.; Liu; Liu, Y. The effects of shear force on the formation, structure and metabolism of aerobic granules. Appl. Microbiol. Biotechnol. 2001, 57 (1), 227–233. (10) Tay, J. H.; Liu, Q. S.; Liu, Y. The role of cellular polysaccharides in the formation and stability of aerobic granules. Lett. Appl. Microbiol. 2001, 33 (3), 222–226. (11) McSwain, B. S.; Irvine, R. L.; Hausner, M.; Wilderer, P. A. Composition and Distribution of Extracellular Polymeric Substances in Aerobic Flocs and Granular Sludge. Appl. Environ. Microbiol. 2005, 71 (2), 1051–1057. (12) Adav, S. S.; Lee, D.-J.; Tay, J.-H. Extracellular polymeric substances and structural stability of aerobic granule. Water Res. 2008, 42 (6-7), 1644–1650. (13) Seviour, T.; Pijuan, M.; Nicholson, T.; Keller, J.; Yuan, Z. Understanding the properties of aerobic sludge granules as hydrogels. Biotechnol. Bioeng. 2009, 102 (5), 1483–1493. (14) Seviour, T.; Pijuan, M.; Nicholson, T.; Keller, J.; Yuan, Z. Gelforming exopolysaccharides explain basic differences between structures of aerobic sludge granules and floccular sludges. Water Res. 2009, 43 (18), 4469–4478. (15) Seviour, T.; Donose, B. C.; Pijuan, M.; Yuan, Z. Purification and Conformational Analysis of a Key Exopolysaccharide Component of Mixed Culture Aerobic Sludge Granules. Environ. Sci. Technol. 2010, 44 (12), 4729–4734. (16) Lin, Y.; de Kreuk, M.; van Loosdrecht, M. C. M.; Adin, A. Characterization of alginate-like exopolysaccharides isolated from aerobic granular sludge in pilot-plant. Water Res. 2010, 44 (11), 3355–3364. (17) Bubb, W. A. NMR spectroscopy in the study of carbohydrates: Characterizing the structural complexity. Concepts Magn. Reson., Part A 2003, 19A (1), 1–19. (18) Allen, M. S.; Welch, K. T.; Prebyl, B. S.; Baker, D. C.; Meyers, A. J.; Sayler, G. S. Analysis and glycosyl composition of the exopolysaccharide isolated from the floc-forming wastewater bacterium Thauera sp. MZ1T. Environ. Microbiol. 2004, 6 (8), 780–790. (19) Yilmaz, G.; Lemaire, R.; Keller, J.; Yuan, Z. Simultaneous nitrification, denitrification, and phosphorus removal from nutrient-rich industrial wastewater using granular sludge. Biotechnol. Bioeng. 2008, 100 (3), 529–541. (20) Claridge, T. D. W. High-resolution NMR techniques in organic chemistry; Pergamon: Amsterdam, 1999; Vol. 19. (21) Braun, S., Kalinowski, H.-O., Berger, S. 150 and More Basic NMR Experiments: a Practical Course; Wiley-VCH: Weinheim, 1998. (22) Parella, T.; Sanchez-Ferrando, F.; Virgili, A. Improved sensitivity in gradient -based 1D and 2D multiplicity-edited HSQC Experiments. J. Magn. Reson. 1997, 126, 274–277. (23) Fialho, A. M.; Martins, L. O.; Donval, M.-L.; Leitao, J. H.; Ridout, M. J.; Jay, A. J.; Morris, V. J.; Sa-Correia, I. Structures and Properties of Gellan Polymers Produced by Sphingomonas paucimobilis ATCC 31461 from Lactose Compared with Those Produced from Glucose and from Cheese Whey. Appl. Environ. Microbiol. 1999, 65 (6), 2485–2491. (24) Mack, D.; Fischer, W.; Krokotsch, A.; Leopold, K.; Hartmann, R.; Egge, H.; Laufs, R. The intercellular adhesin involved in biofilm accumulation of Staphylococcus epidermidis is a linear beta1,6-linked glucosaminoglycan: purification and structural analysis. J. Bacteriol. 1996, 178 (1), 175–183. (25) Vieira, ´I. G. P.; Mendes, F. N. P.; Galla˜o, M. I.; de Brito, E. S. NMR study of galactomannans from the seeds of mesquite tree (Prosopis juliflora (Sw) DC). Food Chem. 2007, 101 (1), 70–73. (26) Warda, M.; Toida, T.; Zhang, F.; Sun, P.; Munoz, E.; Xie, J.; Linhardt, R. Isolation and characterization of heparan sulfate from various murine tissues. Glycoconjugate J. 2006, 23 (7), 555– 563. (27) McMurry, J. Structure Determination: Nuclear Magnetic Resonance Spectroscopy. In Organic Chemistry, 6th ed.; McMurry, J., Ed.; Thomson Brooks Cole: Cornell University, 2004; pp 424-463. (28) Stenutz, R.; Jansson, P.-E.; Widmalm, G. A practical guide to structural analysis of carbohydrates. http://www.casperold. organ.su.se/sop/ (accessed 2010). VOL. 44, NO. 23, 2010 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

8969

(29) Roger, O.; Kervarec, N.; Ratiskol, J.; Colliec-Jouault, S.; Chevolot, L. Structural studies of the main exopolysaccharide produced by the deep-sea bacterium Alteromonas infernus. Carbohydr. Res. 2004, 339 (14), 2371–2380. (30) Ahrazem, O.; Prieto, A.; Leal, J. A.; Inmaculada Gime´nez-Abia´n, M.; Jime´nez-Barbero, J.; Bernabe´, M. Fungal cell wall polysaccharides isolated from Discula destructiva spp. Carbohydr. Res. 2007, 342 (8), 1138–1143.

8970

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 23, 2010

(31) Nelson, D. L.; Cox, M. M. Carbohydrates and Glycobiology. In Lehninger Principles of Biochemistry, 3rd ed.; Nelson, D. L., Cox, M. M., Eds.; Worth Publishers: New York, 2000; pp 293-321. (32) Morris, V. J. Gelation of Polysaccharides. In Functional Properties of Food Macromolecules; Mitchell, J. R., Ledward, D. A., Eds.; Elsevier Applied Science Publishers: London, 1986.

ES102658S