The potential of bacteria isolated from ruminal contents of seaweed-eating North Ronaldsay sheep to hydrolyse seaweed components and produce methane by anaerobic digestion in vitro

The production of methane biofuel from seaweeds is limited by the hydrolysis of polysaccharides. The rumen microbiota of seaweed-eating North Ronaldsay sheep was studied for polysaccharidic bacterial isolates degrading brown-seaweed polysaccharides. Only nine isolates out of 65 utilized > 90% of the polysaccharide they were isolated on. The nine isolates (eight Prevotella spp. and one Clostridium butyricum) utilized whole Laminaria hyperborea extract and a range of seaweed polysaccharides, including alginate (seven out of nine isolates), laminarin and carboxymethylcellulose (eight out of nine isolates); while two out of nine isolates additionally hydrolysed fucoidan to some extent. Crude enzyme extracts from three of the isolates studied further had diverse glycosidases and polysaccharidase activities; particularly against laminarin and alginate (two isolates were shown to have alginate lyase activity) and notably fucoidan and carageenan (one isolate). In serial culture rumen microbiota hydrolysed a range of seaweed polysaccharides (fucoidan to a notably lesser degree) and homogenates of L. hyperborea, mixed Fucus spp. and Ascophyllum nodosum to produce methane and acetate. The rumen microbiota and isolates represent potential adjunct organisms or enzymes which may improve hydrolysis of seaweed components and thus improve the efficiency of seaweed anaerobic digestion for methane biofuel production.


Introduction
Seaweed is an attractive biomass for microbial biofuel production. It does not require irrigation with valuable water resources, it generally has no lignin (which is present in terrestrial plants and is not only difficult to breakdown but reduces degradation, and therefore digestibility, of plant cell wall cellulose complexes), and there is little if any competition for its use as a food source, unlike many terrestrial crops. It can be collected as beach cast seaweed, which is regarded as a pollutant of leisure beaches, or it can be harvested sustainably from natural seaweed beds. It could also be farmed, and it grows rapidly; for example 60 tons dry weight per hectare per year of Laminaria japonicum were reported to be produced in China (Morand et al., 1991). From the 1970s through to 1990s various groups were involved in assessing the suitability of seaweed (particularly Macrocystis pyrifera) as a biomass for anaerobic digestion (AD) to produce methane biofuel (see review by Forro, 1987). The overall result of these studies indicated that seaweeds were generally a suitable biomass for AD.
In the Northern Atlantic Laminaria, Fucus and Ascophyllum species are particularly common brown algae found growing on the litterol and sub-litterol seashore. They are also beach cast in large quantities along the seashore, particularly in western Scotland during the winter (Morand et al., 1991). Animal manure is often utilized as an inoculum for anaerobic digesters but various reports have suggested that specific bacteria capable of fermenting marine phycocolloids may accelerate and increase biogas production from seaweeds (Morand et al., 1991). Orpin and colleagues (1985) studied the ruminal microbiota of the North Ronaldsay breed of sheep, the male adults of which survive almost entirely on a seaweed diet on the remote island of North Ronaldsay in the Orkney Islands, Scotland. They noted that there were important differences in the speciation and metabolic activity of seaweed-fed North Ronaldsay sheep rumen microbiota compared with grass-fed sheep (particularly, laminarin, mannitol, fucoidan, mannuronic and guluronic acids were far more readily hydrolysed) and concluded that the microbiota adapted to efficiently digest seaweeds. Even in humans, the ability to digest seaweeds has recently been associated with alterations in gut flora; with gut bacteria acquiring seaweed degrading genes from marine species (Hehemann et al., 2010).
The efficient hydrolysis of seaweed polysaccharides, particularly alginates, is seen as a rate limiting step for efficient AD to proceed (Morand et al., 1991;Horn, 2000). With a long-term aim of developing and characterizing an optimal inoculum with which to degrade seaweeds by AD, we aimed to examine whether bacterial isolates from the rumen microbiota of North Ronaldsay sheep could produce enzymes to effectively hydrolyse seaweed polysaccharides by the concerted actions of polysaccharide depolymerases and glycoside hydrolases. Small scale serial culture studies with the whole rumen microbiota also determined if seaweed polysaccharides and whole seaweed homogenates could be effectively digested to produce methane and acetate.

Isolation of polysaccharide degraders
Pre-enrichment in liquid media was preferred to direct isolation of rumen microbiota as the ruminal contents had been frozen and this may have impacted on the viability of the microbial populations. Growth was observed in enrichment cultures with all of the substrates tested and subsequent counts (log 10 cfu ml -1 enrichment culture) on agar plates containing various substrates were: alginate (10.6-10.9), laminarin (10.5-10.7), fucoidan (7.8-8.6) and cellulose (6.7-8.5).
The enrichment cultures resulted in the isolation of 65 colonies on media containing either laminarin (20), alginate (18), fucoidan (11), cellulosic derivatives (12) or Laminaria hyperborea extract (4). These were selected as representative isolates on the basis of distinct morphology. The 65 isolates were subsequently tested for polymer utilization in broth culture and only nine degraded > 90% of their recovery polysaccharide. Isolates enriched and recovered on L. hyperborea extract were found unable to degrade individual polysaccharide substrates and in all probability were members of the microbial population that utilized non-polymeric cellular constituents such as the storage polysaccharide mannitol and the products released by the degradative population, and exemplified the rationale for not using seaweed extract as the only primary enrichment isolation substrate for degraders.

Range of polysaccharide utilization
The ability of the nine isolates to degrade a range of seaweed polysaccharides was then assessed. All nine isolates were able to effectively utilize laminarin while seven utilized alginic acid (Table 1). Only strains L7 and L10 exhibited any substantive activity against fucoidan. However, all strains were capable of utilizing carbohydrate constituents present in the seaweed extract and eight exhibited some activity against carboxymethylcellulose (Table 1).
All the isolates grew on the L. hyperborea extract broth (Table 1) but growth (increased absorbance at 595 nm) only occurred with isolate C8 on the mixed Fucus extract and by four strains (A9, A12, A14 and L12) on the Ascophyllum nodosum extract (data not shown). Addition of glucose stimulated increased growth of all the isolates in the presence of the three seaweed extracts indicating a lack of readily assimilated substrates rather than growth inhibition by substances in the extract. Measurement of residual carbohydrates confirmed that all isolates, excepting L8, had utilized constituents in the L. hyperborea extract, whereas evidence of carbohydrate removal from the other two extracts by any of the isolates was inconclusive. All isolates formed acetate when grown on laminarin (Table 1) and thus have the potential to degrade seaweed polysaccharides and produce a key metabolite for acetoclastic methanogens.

Selection and identification of three most active isolates
All of the nine isolates were obligately anaerobic and all but L10 and L12 were Gram-negative rods. Isolate L10 withstood heating at 80°C for 15 min and was a presumptive Clostridium sp. Eight of the nine isolates were identified to at least Genus level (isolate A12 was not identifiable) by having a high degree of homology (97-100%) with 16S rRNA genes lodged in the NCBI nucleotide database. The species identification also matched morphological and biochemical studies which gave the same presumptive Genus/speciation assignation. A11 and L7 were Prevotella species (100% homology) and L10 was a Clostridium butyricum. These are common rumen bacteria, and other isolates (A9, A14, L8 and C8) were similarly Prevotella spp. Isolate L12 was Clostridium botulinum.
Three isolates chosen for further study were: isolate L7, a Prevotella sp., was consistently the most effective of the polysaccharide degraders that could also utilize fucoidan (18.6% degradation); L10, a C. butyricum, was an effective laminarin degrader that also utilized fucoidan (20.3% degradation); and A11, a Prevotella sp. initially recovered on alginic acid, had degradative profiles similar to several of the other isolates, but was chosen for its acetogenic potential.

Glycosidase and polysaccharidase activities of isolates
The potential of isolates L7, L10 and A11 to utilize, or be involved in the degradation of other polysaccharide classes (e.g. galactans, mannans, xylans etc.) was determined by monitoring the range of their polysaccharide depolymerase (polysaccharidase) and glycoside hydrolase (glycosidase) activities.
Isolates L7 and A11 were found to have wide-ranging glycosidase activities ( Table 2). The activity profile was not markedly affected by the growth substrate indicating that growth on seaweed extract and individual seaweed polysaccharides resulted in the expression of a wide range of activities. There was little or no detectable b-Lfucosidase, b-D-glucuronidase, b-D-galacturonidase, b-Dmannosidase and a-D-xylosidase activities (results not shown). The glycosidase profile of L10 was considerably more restricted, although higher activities were detected after growth on seaweed extract. Polysaccharidases active against a-glucans (amylopectin), b-glucans [laminarin (b1-3), cello-oligosaccharides (b1-4)] and alginate were detected in all the isolates (Table 3). There was an indication that the growth substrate did affect detectable activity profiles. There was no evidence for the formation of galactanase or a fucoidanase by L7 or A11 under the growth conditions examined although low levels were apparently present in L10.
Alginic acid depolymerization may take place by glycanase-type reactions in which the glycosidic bond is cleaved directly to generate oligomeric fragments or alternatively the chain cleavage may occur via a lyase type reaction in which degradation products with unsaturated reducing ends are produced and detected by their absorbance at 230 nm (Haugen et al., 1990). Absorbance increased concomitantly with alginate breakdown. The increase in absorbance (per mg protein after incubation for 90 min) for enzyme preparations recovered from cells grown on the isolation polymer (A11 alginate, L7 L10 laminarin) or L. hyperborea extract was 1.28/1.33, 0.24/ 1.73 and 0.19/0.19 with enzyme from A11, L7 and L10 respectively. The activity in A11 was expressed similarly when the organism was grown on either alginate or the alginatecontaining seaweed extract. However, there was a pronounced difference in the activity of strain L7 after growth on laminarin or the Laminaria extract, where growth in the latter resulted in enzyme induction and increased activity. Only background activities were detected in the nonalginolytic strain L10.

Sequential culture studies
A range of serial cultures of whole North Ronaldsay sheep rumen microbiota were able to utilize seaweed constituents and whole seaweed homogenates as the primary carbohydrate source. The cultures were established by successive serial culture at 7-day intervals over a 6-week period. The isolated polymers were, with the exception of fucoidan, extensively utilized ( Fig. 1A and B) and carbohydrate components were also removed from the seaweed homogenates. The efficiency of carbohydrate removal was highest with L. hyperborea and lowest from the Fucus spp. supplemented culture.
After incubation methane and carbon dioxide was detected in all the serial cultures and with many substrates these were present in proportions normally encountered in the rumen ecosystem, i.e. 30% methane, 65% carbon dioxide.
Stable methanogenic cultures were established with structural polymers and storage polymers including mannitol (Fig. 2) confirming that the ability to catabolize a wide range of seaweed carbohydrates was present in the initial ruminal population. Although fucoidan was catabolized less effectively than other polysaccharides its utilization was also associated with consistent methane formation.
The serial cultures established on seaweed homogenates were also methanogenic, but the proportion of methane detected in the headspace fermentation gases tended to be lower than observed when individual extracted carbohydrates were used (Fig. 2). The methane proportion stabilized at approximately 10% in cultures growing on A. nodosum but was of the order of 20-25% in the cultures established on L. hyperborea and Fucus species (Fig. 2).
It was evident (Fig. 3) that all cultures were acetogenic and accumulated acetate during each 7-day culture period. Acetate levels were consistent over the period but were substrate-dependent as fucoidan, which was not extensively utilized, was the least acetogenic. However, seaweed homogenates were as effective substrates for acetate formation as individual polysaccharides (Fig. 3).

Discussion
In this study the structural integrity of the cellulosic substrate (filter paper) was lost after enrichment culture indicating that cellulolytic activity was present. It is possible that cellulolytic bacteria are only present in low numbers in the North Ronaldsay sheep rumen and this may account for the inability of Orpin and colleagues (1985) to isolate these directly. Orpin and colleagues (1985) also found only 13% of the culturable bacteria from seaweedfed sheep grew on alginate, 71% on laminarin, 13% on fucoidan and 99% on mannitol. These were however significantly higher percentages than those from pasture-fed animals (2%, 32%, 0% and 0% respectively). Additionally, Orpin and colleagues (1985) found substantial numbers of seaweed-fed rumen bacteria utilized mannuronic and guluronic acids, which are likely to be released on hydrolysis of alginate in the rumen.
In the present study the majority of the 65 isolates recovered from enrichment cultures were found to be unable to degrade polymeric substrates. Many other anaerobic bacteria are reported to grow only when part of a consortium in which mutualistic metabolic exchanges occur (Stams and Plugge, 2009;Hillesland and Stahl, 2010;Miller et al., 2010;Summer et al., 2011) and many of these isolates may require mutualistic interactions for growth on polymeric substrates.
It was probable, however, that these isolates were effective scavengers of polysaccharide degradation prod- ucts and soluble sugars and were therefore likely to be important in the microbial consortia in the removal of soluble sugars and generation of methanogenic precursors. It would be worthwhile to screen these non-degrader strains for their capacity to produce acetate. The ability of nine selected polysaccharide degrading isolates to degrade a range of seaweed polysaccharides was assessed in broth culture. There was only very limited growth in fucoidan and seaweed-extract containing media. This agrees with the findings of Orpin and colleagues (1985), who found few isolates utilized fucoidan and none in pasture fed animals. Isolates that degrade this sulfated polysaccharide (such as L7 and L10) are then potentially important.
The wide ranging glycosidase activity profiles of isolates L7 and A11 are compatible with them being strains of Prevotella (formerly Bacteroides) ruminicola as studies by Williams and colleagues (1984) demonstrated similarly diverse activities in the Type strain of this ruminal species.
The three most active isolates demonstrated polysaccharidase activities against a-glucans, b-glucans, cellooligosaccharides and alginate and there was evidence that activity was induced when growth was on a complex substrate (L. hyperborea extract). The possibility exists that other enzymes or higher activities may be detected when the organisms encounter the appropriate polysaccharidic compounds as exemplified by the effect of laminarin on strains L7 and L10.
In the serial culture studies on the whole rumen microbiota only about 40% of the available carbohydrate was removed in the fucoidan supplemented cultures. Although the utilization of this particular polymer was less extensive, the usage was consistent over the period confirming that the ability to degrade this sulfated polymer does exist in the ruminal population and can be maintained in serial culture. However, no single fucoidan degrading isolates were able to be selected on agar plates. Novel strategies may therefore be required to isolate the bacterial species involved in fucoidan breakdown. The failure to isolate the species involved is probably associated with their relative ability to compete in the isolation procedures with organisms in the secondary scavenging population or may indicate dependent syntrophy among two or more species. The utilization of the seaweed storage carbohydrate mannitol was particularly stable over the period of serial culture. The presence of mannitol utilizing bacteria in the rumen is more specific to animals consuming seaweed as many of the major species of ruminal bacteria do not utilize this sugar alcohol (Stewart et al., 1997). Although the identity of the bacteria involved has yet to be established, Orpin and colleagues (1985) observed a much higher proportion of mannitol utilizers in the ruminal con-tents of North Ronaldsay sheep consuming seaweed than in animals of the same breed fed grass.
That the cultures actively produced methane over the course of the trial confirmed that frozen rumen contents from seaweed-eating sheep are potentially suitable inocula for AD systems and further investigation to optimize the process could result in increased methane yields. Although the proportion of methane produced from the different substrates during serial culture was similar it cannot be assumed that the species profiles were the same. Additional studies are required to identify the microbiological composition of sequential cultures, which is likely to be substrate-dependent.
All serial cultures were acetogenic and accumulated acetate during each 7-day culture period. Acetate is a substrate for acetoclastic methanogenesis.
In conclusion these rumen microbial consortia, established by serial culture, have the potential to be inocula for Week 1 Week 2 Week 3 Week 4 Week 5 Week 6 Fig. 2. Methane accumulation as monitored at the end of the 7-day consecutive incubation period for each of the serial cultures. The proportion of methane as a percentage of the headspace gas is indicated for various enrichment substrates at the end of week 1 ( ) through to the end of week 6 (ᮀ). Week 1 Week 2 Week 3 Week 4 Week 5 Fig. 3. Acetate accumulation as monitored at the end of the 7-day consecutive incubation period for each of the serial cultures. The acetate level is indicated for various enrichment substrates at the end of week 1 ( ) through to the end of week 5 (ᮀ).
AD to enhance the utilization of specific seaweed varieties. Alternatively, the individual bacterial isolates recovered (such as L7, L10 and A11) may offer sources of enzymes or may act as adjunct cultures for the efficient AD of seaweeds in renewable biofuel production.

Collection of rumen contents
Three North Ronaldsay seaweed-eating, mature male sheep were transported to the abattoir at Kirkwall, Orkney. The rumen contents were recovered immediately after slaughter and were combined in a heavy duty polythene bag surrounded with ice packs for transfer to the laboratory in less than 24 h at below 10°C. The pooled rumen contents were divided into completely filled 100 ml screw-capped containers and stored at -20°C.
The resultant PCR amplicons were sequenced at the Dundee Sequencing Laboratory (College of Life Sciences, University of Dundee, Scotland) and the sequences derived were BLAST searched for homology with 16S rRNA gene library sequences at http://www.ncbi.nlm.nih.gov.

Serial culture of whole rumen microbiota
Cultures (5 ml volumes) were in single seaweed polysaccharide or whole-seaweed homogenate supplemented basal medium initially inoculated with homogenized ruminal contents that had been stored prior to use at -20°C. After incubation at 39°C for 7-day a follow-on serial culture was set up using a 10% inoculum from the established culture. The substrates used (2 mg ml -1 ) were alginate, laminarin, fucoidan, carrageenan, cellulose (filter paper), amylopectin, xylan and mannitol. Cultures were also set up using L. hyperborea, A. nodosum and Fucus spp. homogenates at 10 mg ml -1 . The residual carbohydrate at the end of each 7-day serial culture period was determined by the method of Dubois and colleagues (1956).

Analytical techniques
Headspace gases in serial cultures were analysed by gas chromatography using a Perkin Elmer 8500 Gas Chromatograph fitted with a methanizer and flame ionization detection. A 6 feet ¥ 1/8 inches column packed with 80/100 carbosphere was used with a helium carrier gas flow rate of 20 ml min -1 (Speck and Burke, UK). The separation was run isothermally with an oven and injector temperature of 150°C; the detector temperature was maintained at 300°C. Peak area was calculated with a Hewlett Packard HP3396 series II in-line integrator.
Cell-free culture supernatants were analysed for the presence of acetate using an enzyme assay kit (R-Biopharm). The method was adapted for use at the 96-well microtitre plate scale by reducing reagent/sample size by a factor of 10. The analytical method and calculation used were as described by the manufacturer. A standard curve comprising 0-2.96 mg of acetate in the assay system was constructed for each assay. The basal medium contained approximately 1.7-1.8 mg ml -1 acetate and due allowance was made for this endogenous component in determining assay results.