This invention relates to microbial production of cellulose. More particularly, this invention relates to production of microbial cellulose by facultatively anaerobic microorganisms.
It is well-known that naturally-occurring strains of the Gram-negative microorganism Acetobacter xylinum, also known as Gluconoacetobacter xylinum (see Yamada et al., 1998, Int. J. Syst. Bacteriol. 48:327-328), are able to produce and secrete significant quantities of cellulose when grown under small-scale laboratory culture conditions wherein each microbial cell produces a single strand of cellulose commonly referred to as a strand or a fibril. Each fibril comprises multiple inter-twisted cellulose chains or microfibrils. The biochemical basis, genetics and regulation of cellulose biosynthesis in Acetobacter xylinum have been extensively studied, reported, and reviewed. Acetobacter spp. are obligate aerobic microorganisms, i.e., they have a strict requirement for O2 for respiration which drives their metabolism, growth and cellulose production. When grown in standing i.e., non-shaken/agitated, liquid cultures, Acetobacter xylinum produces pellicles comprising disorganized layers of long intertwined cellulose strands at the interfaces between the air and liquid media. As the extent and thickness of the cellulose-containing pellicle layers increase in such standing cultures, they increasingly impede then stop O2 availability from the headspace above the pellicles to the Acetobacter cells underlying the pellicles thus limiting and stopping cellulose production. Although cellulose produced by Acetobacter spp. in standing liquid cultures is chemically similar to cellulose produced from wood pulp, the major difference is that the cross-sectional diameter of Acetobacter spp. cellulose fibrils is usually about 2 orders of magnitude smaller than cellulose fibrils from wood. Typically, the cross-sectional dimensions of microfibrils produced by Acetobacter spp. is about 1.6 nm 5.8 nm, and they are twisted together to form fibrils (i.e., strands) having cross-sectional dimensions of about 3.2 nm×133 mm.
Cellulose production can be increased by culturing Acetobacter spp. in agitated liquid media wherein O2 availability to individual Acetobacter spp. cells is increased through dissolved O2 continually dispersed within and throughout the liquid media. Cellulose produced by Acetobacter spp. grown in such culture conditions is localized in multiple pellets circulating throughout the media. U.S. Pat. No. 4,863,565 and related U.S. Pat. Nos. 5,079,162, 5,144,162, 5,871,978, and 6,329,192 disclose that the macroscopic structure of cellulose in pellets produced by Acetobacter spp. cultured in agitated liquid media is characterized by a three-dimensional reticulated lattice structure that is significantly different from the layered cellulose macrostructure produced in pellicles from standing liquid cultures. The reticulated cellulose structure from liquid cultures is characterized by elongated strands of cellulose interconnected by shorter cellulosic branches or filaments having cross-section diameters of 0.1μto 0.2μ, thereby forming grid-like patterns extending in three dimensions. The formation of the shorter cellulosic branches or filaments is apparently caused by one or more cellulose microfibrils separating out from the main fibril produced by an Acetobacter spp. cell as a result of the constant culture agitation. The shorter cellulosic branches interconnect and comingle with cellulose fibrils produced by other Acetobacter spp. cells thereby giving rise to the grid-like lattice structure. It also appears that the rates of agitation of liquid cultures significantly affect (a) the physical properties of the cellulose fibrils, strands, branches and filaments formed by Acetobacter spp., and (b) the degree of interconnecting and comingling that occurs; a low rate of agitation will result in the formation of larger cellulose-containing pellets while increasingly higher rates of agitation produce increasingly smaller cellulose pellets.
There are numerous problems encountered in attempting to scale cellulose production by Acetobacter spp. in large volumes of liquid media. For example, it appears that naturally occurring strains of cellulose-producing Acetobacter spp. are unstable when cultured in shaken or agitated liquid cultures and commonly spontaneously mutate into cellulose non-producing variants thereby limiting Acetobacter spp. cellulose production potential. As liquid culture volumes are increased, increasingly larger impellers and faster rates of impeller speeds are necessary to produce and maintain the levels of dissolved O2 required to sustain Acetobacter spp. respiration, metabolism and cellulose production. U.S. Pat. No. 4,863,565 and related U.S. Pat. Nos. 5,079,162, 5,144,162, 5,871,978, and 6,329,192, and 6,329,192 teach that shear forces in liquid media caused by high impeller speeds significantly reduce the sizes of the three-dimensional reticulated cellulose structures produced by Acetobacter spp. thereby substantially degrading the properties of the cellulose product and its commercial usefulness. Yet another problem commonly associated with cellulose production by Acetobacter spp. in both standing and agitated cultures is the propensity of these microorganisms to convert glucose to gluconic acid and/or keto-gluconic acid thereby significantly dropping the pH of the media resulting in cessation of cellulose production. Furthermore, the conversion of glucose to gluconic and keto-gluconic acids decreases glucose availability for cellulose production.
Strategies developed to address cellulose production problems associated with Acetobacter spp. include: (1) creating mutants with reduced propensity for converting glucose into acids (e.g., U.S. Pat. No. 5,079,162) or alternatively, with modified carbohydrate and/or amino acid metabolism thereby increasing rates of cellulose production (e.g., U.S. Pat. Nos. 5,962,278, 6,110,712 and 6,140,105), (2) adding cell-division inhibitors to modify and perhaps improve the physical structure and properties of cellulose produced in agitated liquid cultures (e.g., U.S. Pat. Nos. 6,060,289 and 6,627,419), (3) increasing the availability of dissolved O2 in large-volume vessels by combining two different-shaped impellers to concurrently aerate and agitate liquid media (e.g., U.S. Pat. No. 6,013,490), (4) increasing O2 availability in liquid cultures contained within vessels by increasing the amount of aeration introduced into the vessel, thereby reducing the partial pressure of CO2 while increasing the partial pressure of O2 (e.g., U.S. Pat. No. 6,017,740), and (5) post-harvest processing methods for Acetobacter spp. cellulose produced in agitated liquid cultures to improve its physical properties (e.g., U.S. Pat. No. 6,153,413). However, such strategies are complicated, costly and still have the challenge of providing sufficient O2 to enable optimal metabolism and cellulose production by Acetobacter spp. in large-volume liquid cultures.
It is well-known that other genera of obligate aerobic Gram-negative microorganisms are able to produce small amounts of cellulose from various carbon substrates under carefully controlled conditions. Such obligate aerobic cellulose-producing microorganisms include Pseudomonas sp., Alcaligenes sp., Achromobacter sp., Aerobacter sp., Azotobacter sp., Agrobacterium sp., and Rhizobium sp. isolated from sewage samples (Deinema et. al., 1971, Arch. Mikrobiol. 78:42-57), Rhizobium sp. isolated from leguminous plants (Napoli et al., 1975, Appl. Microbiol. 30:123-131), and Agrobacterium tumefaciens (Mathysse et al., 1995, J. Bacteriol. 177:1069-1075). However, the amounts of cellulose produced by these microbial genera are small relative to their carbon substrate inputs, and also, when compared to cellulose production by Acetobacter spp. Deinema et al. show in their
More recently, cellulose production and involvement in biofilm formation have been demonstrated in facultative anaerobic Gram negative bacteria including Escherichia coli, Klebsiella pneumoniae and Salmonella enterica (Nobles et al., 2001, Plant Physiol. 127:529-542). These species produce minute amounts of cellulose and are not expected to be of value for large-scale production.
The exemplary embodiments of the present invention, at least in preferred forms, are directed to the production of bacterial cellulose by facultatively anaerobic microorganisms.
According to one specific embodiment of the present invention, there is provided a biologically pure culture of a cellulose-producing strain of Proteus sp. as exemplified by Proteus myxofaciens. In a preferred form, there is provided a biologically pure culture of cellulose-producing Proteus myxofaciens strain IDAC 071005-01. However, the embodiments of the present invention include cellulose production with Proteus myxofaciens strain ATCC 19692.
According to another preferred embodiment of the present invention, there is provided a method for production of bacterial cellulose wherein a cellulose-producing strain of Proteus sp. is cultured in a liquid medium suitable for culturing facultatively anaerobic microorganisms, separating bacterial cellulose produced in said liquid medium from the liquid medium, washing said separated bacterial cellulose and drying said bacterial cellulose.
In a preferred form, the liquid medium is provided with a carbohydrate substrate containing therein at least one sugar selected from the group consisting of glucose, sucrose, fructose, lactose, xylose, and rhamnose.
In another preferred form, the liquid medium is provided with a pH selected from the range of 5 to 10, more preferably from the range of 6 to 9, and most preferably in the range of 7 to 8.5.
In yet another preferred form, the liquid medium is provided with a buffer. The buffer preferably comprises at least one of a sodium chloride buffer or a phosphate buffer, and more preferably, comprises a mixture of sodium chloride and phosphate buffers.
According to one aspect of the present invention, the method for producing bacterial cellulose comprises culturing a cellulose-producing strain of Proteus sp. in a liquid medium suitable for culturing facultatively anaerobic microorganisms under aerobic conditions. In a preferred form, the cellulose-producing strain is Proteus myxofaciens IDAC 071005-01. In another preferred form, the cellulose-producing strain is Proteus myxofaciens strain ATCC 19692.
According to another aspect of the present invention, the method for producing bacterial cellulose comprises culturing a cellulose-producing strain of Proteus sp. in a liquid medium suitable for culturing facultative anaerobic microorganisms under anaerobic conditions. In a preferred form, the cellulose-producing strain is Proteus myxofaciens IDAC 071005-01. In another preferred form, the cellulose-producing strain is Proteus myxofaciens strain ATCC 19692.
According to further preferred embodiment of the present invention, there is provided a bacterial cellulose product produced by a cellulose-producing strain of a Proteus sp. In a preferred form, the cellulose-producing strain is Proteus myxofaciens IDAC 071005-01. In another preferred form, the cellulose-producing strain is Proteus myxofaciens strain ATCC
Samples of Proteus myxofaciens strain PARC-59 were deposited under the terms of the Budapest Treaty at the INTERNATIONAL DEPOSITORY AUTHORITY OF CANADA (IDAC) of 1015 Arlington Street, Winnipeg, Manitoba, R3E,3R2, Canada (Telephone: 204-789-6030; Facsimile: 204-789-2018). The deposit was made on Oct. 7, 2005 and was assigned accession number 071005-01.
The present invention will be described in conjunction with reference to the following drawings, in which:
Facultatively anaerobic microorganisms can proliferate and produce extracellular products both in the presence and the absence of oxygen. When oxygen is present in their culture media, obligate aerobes and facultative anaerobes will utilize the oxygen to oxidize carbohydrates thereby producing energy in the form of ATP along with substrates required for subsequent cellular metabolism, growth and productivity. Oxygen serves as the terminal electron acceptor in the electron transport chain by which ATP is produced. When oxygen is limiting or absent from their culture media, obligate aerobic microorganisms such as Acetobacter spp. are unable to oxidize carbohydrates resulting in the cessation of energy production, subsequent metabolism and growth. However, facultatively anaerobic microorganisms are able to utilize a wide variety of other compounds as alternative terminal electron acceptors for the electron transport chain in a process known as anaerobic respiration. Examples of compounds that may be utilized by facultative anaerobes for anaerobic respiration include nitrogenous compounds such as nitrates and nitrites, sulfur compounds such as sulfates, sulfites, sulfur dioxide, and elemental sulfur, carbon dioxide, iron compounds, and manganese compounds. Consequently, compared to obligate aerobic microorganisms, facultative anaerobes are easier to produce and maintain in large-volume fermentation production systems, and do not require the same complexity of culture media and equipment configurations.
According to Bergey's Manual of Systematic Bacteriology, Vol. 1 (1968), facultative anaerobic Gram-negative microorganisms generally belong to one of three families, i.e., Family 1—Enterobacteriaceae, Family II—Vibrionaceae, and Family III—Pasteurellaceae. The Family Enterobacteriaceae is made up of 13 genera which are widely distributed throughout the world. They are commonly found in soil and water systems, and also are present as normal internal and external flora in humans and animals. Their host range includes insects, avians, mammals, as well as plants including fruits, vegetables, grains, flowers and trees. Many species of Enterobacteriaceae are pathogens of animals, avian sp., mammals or plants and consequently, this Family has been widely studied. In general terms, Enterobacteriaceae are Gram-negative rod-shaped bacteria that are 0.3-1.0 μm×1.0-6.0 μm. Most grow well on simple media containing a monosaccharide as a carbon source at temperatures in the range of 22° C. to 35° C. Escherichia coli is the type species for Enterobacteriaceae and is considered to be the most thoroughly studied of all species of bacteria because of its ease of isolation, ease of cultivation, rapid generation time, and its ability to be genetically manipulated. As a result, a wide variety of commercial fermentation processes, systems and culture media are known for the production of extracellular compounds by genetically engineered E. coli for use in pharmaceutical and industrial applications.
The Genus Proteus is classified within the Enterobacteriaceae Family and currently consists of 5 named species, i.e., P. mirabilis, P. penneri, P. vulgaris, P. hauseri, and P. myxofaciens and three unnamed genomospecies i.e., Proteus genomospecies 4, 5, and 6. All of the known Proteus sp. are ubiquitous in terrestrial environments, and some species, e.g., P. vulgaris and P. mirabilis, are common components of the normal intestinal flora in humans and animals. All Proteus sp. and strains are characterized by their propensity for: (a) swarming, i.e., movement of cells in periodic cycles of migration and consolidation, when cultured on solid agar media, and (b) production of extracellular slime.
Proteus myxofaciens is a unique but very rare specie isolated from living and dead gypsy moth larvae by Cosenza and Podgwaite who disclosed and characterized this species in 1966 in Antonie Leeuwenhoek J. Microbiol. Serol. Vol. 32 pages 187-191. They deposited their P. myxofaciens isolate with the American Type Culture Collection (ATCC), P.O. Box 1549, Manassas, Va. 20108, U.S.A., under ascension number 19692. Samples of this microorganism may be purchased directly from the ATCC catalogue. Since that time, the existence and properties of the P. myxofaciens ATCC 19692 strain disclosed and made available by Cosenza and Podgewaite have been referred to in numerous publications, but there have been no reports of further work with and/or characterization of this specie nor have there been any reports of isolations of additional strains of P. myxofaciens.
The inventor of the present invention has surprisingly identified a microbial strain (referred to as “PARC-59”) isolated from root and leaf surfaces of field-grown carrots, as a Proteus myxofaciens strain based on comparisons of the morphological, biochemical and genetic profiles of the inventor's PARC-59 isolate with the profiles of P. myxofaciens strain ATCC 19692. Subsequently, the inventor discovered that the P. myxofaciens PARC-59 strain produces copious amounts of cellulose when cultured in liquid media commonly used for propagation and maintenance of microbial cultures, both under aerobic and anaerobic conditions. Accordingly, disclosed herein is a novel Proteus myxofaciens cellulosic strain “PARC-59”. Furthermore, the inventor surprisingly discovered that P. myxofaciens strain ATCC 19692 also produces significant amounts of cellulose in liquid media under both aerobic and anaerobic conditions. Accordingly, disclosed herein is the use of Proteus myxofaciens strains for production of cellulose in liquid media under aerobic or anaerobic conditions.
Those skilled in the art will understand that the cellulosic i.e., cellulose-producing, facultative anaerobic strains of P. myxofaciens disclosed herein can be cultured in a wide variety of simple and complex liquid media provided with various carbon and nitrogen sources and other mineral and/or vitamin supplements useful for culturing facultative anaerobic microorganisms. Suitable carbon sources include monosaccharides such as glucose, fructose, and galactose, disaccharides such as sucrose and maltose, and complex undefined carbohydrate sources such as molasses, either solely or in selected mixtures thereof. Suitable nitrogen sources include purified amino acids, nitrates, ammonium salts, and complex undefined nitrogen-rich materials such as yeast extract, malt extract, casein and corn steep liquors, either solely or in selected mixtures thereof.
The cellulosic strains of P. myxofaciens of the present invention may be grown in liquid media in any vessel or reactor that can be controllably provided with a growth media contained within the vessel or reactor. Because the cellulosic strains of the present invention are facultative anaerobes, it isn't necessary to provide high-speed agitation to generate turbulent flows of liquid media in order to maintain target levels of dissolved O2. Accordingly, those skilled in the fermentation arts will understand how to combine and configure various types of fermenter vessels and reactors with a variety of equipment including impellers and/or fluid circulation pumps to provide single-batch, repeated-batch, or continuous fermentation systems with suitable replenishment of liquid media therein for cellulose production by the cellulosic facultative anaerobic microorganisms disclosed herein. The cellulose produced in such liquid cultures may be harvested by the same equipment and methods used in the art to harvest particulate or precipitated extracellular products or alternatively, cellular debris from other types of large-volume microbial fermentations. An example of a preferred harvesting system is continuous centrifugation of the liquid culture through a stacked-disk multichamber centrifuge wherein a large effective surface area is provided for collecting thereon attached microorganisms including extracellular products such as the cellulose produced by the facultative anaerobic of the present invention. The harvested cellulose should then be washed with water and pressed or centrifuged to remove residual media after which the washed cellulose product of the present invention can be purified by heating to at least 100° C. in an alkaline solution of 1N sodium hydroxide for an appropriate period of time based on the volume of cellulose product being purified. The alkaline solution may be removed by pressing or continual centrifugation and the purified cellulose washed in water, then collected and dried into its final product form.
Since both the ATCC 19692 strain of P. myxofaciens and the IDAC 071005-01 strain, which were obtained from quite different sources, produce cellulose, it appears that this is a basic biological process of P. myxofaciens that will be exhibited by all strains of the microorganism.
The cellulosic facultative anaerobic microorganisms, methods employing said cellulosic facultative anaerobic microorganisms to produce cellulose, and the cellulose product of the present invention are described in more detail in the following examples which are intended to be exemplary of the invention and are not intended to be limiting.
Field carrots grown at the Pacific Agri-Food Research Centre in Summerland, British Columbia, Canada, were harvested from the soil and sampled for assessments of the presence and prevalence of Listeria spp. The APHA method disclosed by Vanderzant et al. (1992, Compendium of Methods for the Microbiological Examination of Foods. 3rd Edition. American Public Health Association, Washington, D.C. pp. 637-663) was adapted for isolation of microorganisms present on the carrots' roots and leaves as follows. Individual root and leaf sections were subjected to primary enrichment by incubation in test tubes containing in UVM Listeria Selective broth, followed by secondary enrichment by transferring to test tubes containing Fraser broth. Subsamples from each test tube containing Fraser broth were streaked onto Petri plates containing Oxford Agar, then incubated at 25° C. for 2 days, after which the plates were observed for the presence of typical Listeria colonies. Presumptive Listeria isolates were purified by repeated streaking onto fresh Petri plates containing Trypticase Soy agar supplemented with 5 g/L yeast extract (TSA-YE). Each purified isolate was given an ID-code number, and then inoculated into test tubes containing Trypticase Soy broth supplemented with 5 g/L yeast extract (TSB-YE), and incubated at 25° C. for 24 h in preparation for further characterizations. One microbial isolate obtained from a sample of carrot leaves and given the ID-code “PARC-59”, produced a solid plug with the consistency of a gel within 24 h after inoculation. The plug physically displaced the liquid culture medium in the test tube after the 24 h incubation period. Another PARC-59 culture was streaked onto Petri plates containing TSA-YE agar, from which, it was inoculated into a test tube containing TSB-YE with similar results, i.e., a solid gel plug was produced within a 24-h incubation at room temperature, and displaced the liquid medium in the test tube.
Gram staining, oxidase and catalase tests were performed with reagents from Difco (Detroit) on 24-h PARC-59 cultures grown at 25° C. on TSB-YE agar plates. The Gram reaction was verified by the KOH method. The isolate was identified as Proteus myxofaciens with the BIOLOG automated microbial identification system (Hayward, Calif.) using GN (Gram-negative) microwell plates and procedures described by the manufacturer. Proteus myxofaciens ATCC 19692, the type strain for this species, was also inoculated in BIOLOG microplates and the carbon substrates utilized or oxidized by the two strains are shown in Table 1. Similarity indices obtained in the BIOLOG identification system were 0.859 and 0.755 for the ATCC 19692 strain and the Summerland PARC-59 isolate, respectively. Positive identification is considered acceptable when the similarity index exceeds 0.5. Accordingly, PARC-59 was confirmed as a P. myxofaciens strain. Proteus myxofaciens strain PARC-59 is referred to herein after as Proteus myxofaciens strain IDAC 071005-01
Proteus myxofaciens ATCC 19692 and PARC-59 in BIOLOG
Individual test tubes containing 10-mL of Trypticase Soy Broth (TSB), TAB amended with yeast extract (Ig/L)(TSB-YE), APT broth (APTB), Tryptose broth (TB) and Nutrient broth (NB) were inoculated with a loopful P. myxofaciens ATCC 19692 or IDAC 071005-01. One tube of each strain was incubated at 6° C., 20° C., 25° C., 30° C., 35° C. and 40° C. for 48 h and then examined for evidence of microbial growth and gel formation. As shown in Table 2, the relative amounts of gel produced by the two P. myxofaciens strains during the incubation period varied with medium composition and temperature of incubation. Gel production was unhindered by vessel geometry and was achieved throughout deep test tubes, suggesting that synthesis of the substance was not subject to oxygen limitation. Gel formation was most extensive at 20° C. and 25° C. by both strains, but each strain produced smaller plugs comprising less solid, more viscous, buoyant material at higher incubation temperatures i.e., temperatures above 25° C. Growth by both strains was observed at 6° C. and at 40° C., but without secretion of gel. More gel was produced in APT broth and TSB-YE than any of the other media. Nutrient broth did not support gel production by either strain (Table 2) suggesting that availability of a carbohydrate substrate was essential for gel production, since NB was the only growth medium that did not contain glucose.
P. myxofaciens
P. myxofaciens
The TSB-YE medium was further investigated because both strains produced gels in this medium at temperatures up to 25° C. Since the gels produced by both strains resisted boiling in 1N NaOH, a method based on this property was developed for purification and quantification as shown below:
Clear, translucent, turgid gels swollen with water were obtained by this procedure. Upon freeze-drying the gel collapsed into a thin, white film that could be torn similar to paper. Furthermore, the dried film did not swell when immersed in water for several days at room temperature.
The effect of sugar type and quantity on the amount of purified, freeze dried gel obtained by cultivation of both strains in supplemented TSB-YE is shown in Table 3.
Gel yields increased with sugar concentrations and were highest in TSB-YE supplemented with sucrose, followed by glucose and fructose. P. myxofaciens IDAC 071005-01 produced much greater amounts of gel in each treatment than did ATCC 19692, regardless of sugar substrate.
The ability of P. myxofaciens to convert different types of sugars into cellulose was assessed by the following method. The following sugars were assessed in this study: glucose, sucrose, fructose, lactose, xylose, and rhamnose. The conversion of each sugar was tested in TSB-YE liquid medium at a concentration of 10 g/L. P. myxofaciens strains IDAC 071005-01 and ATCC 19692 maintained on TSA-YE agar, were inoculated into Erlenmeyer flasks containing 100-ml volumes of TSB-YE amended with the individual sugars (3 replicates/sugar/strain). The inoculated Erlenmeyer flasks were then incubated at 25° C. for 48 h without agitation. Excess medium was removed from the resulting mass of cells and cellulose by application of mechanical pressure. The mass was boiled in 1 N NaOH for 2 h, and then washed three times with distilled water. The purified and washed cellulose samples were freeze-dried prior to dry weight determinations. Both P. myxofaciens strains converted each sugar tested into cellulose (
The effects of increasing sugar concentrations in liquid media on cellulose production were assessed by inoculating P. myxofaciens IDAC 071005-01 into Erlenmeyer flasks containing 100-mL volumes of TSB-YE amended with one of the following concentrations of glucose: 5.0 g/L, 10.0 g/L, 12.5 g/L, 15.0 g/L and 17.5 g/L. The inoculated Erlenmeyer flasks were incubated at 25° C. for 48 h. Cellulose was then harvested from each flask after boiling in 1 N NaOH as described above. The purified and washed cellulose samples were freeze-dried prior to dry weight determinations. The data in
The effects of pH on cellulose production by P. myxofaciens were examined by culturing strains IDAC 071005-01 and ATCC 19692 in 100-mL volumes TSB-YE liquid medium amended with 10 g/L glucose adjusted to pH values of 6.0, 7.0, 8.0, and 8.5 under conditions described above. The data in
The effects of various liquid media supplements including vitamins, minerals, enzyme co-factors, and buffers such as NaCl and phosphate, were individually assessed in 100-mL volumes of TSB-YE liquid medium amended with 10 g/L glucose, under conditions as described above. Addition of omission of various vitamins, minerals, and enzyme co-factors did not have any effects on the cellulose production by P. myxofaciens strains IDAC 071005-01 or ATCC 19692. However, the data in
Proteus myxofaciens strain IDAC 071005-01 was cultured in 100-mL of TSB-YE liquid medium amended with 10 g/L glucose adjusted to pH 8.5 under conditions described above. Samples of gel produced in the medium were removed with a razor blade, then washed and sputter coated with gold and examined in a JEOL model 840 A scanning electron microscope (Japan) at an accelerating voltage of 35 kV. The micrographs revealed a matrix of highly reticulated fibres with embedded bacterial cells (
Acetobacter xylinum strain PARC 017 (isolated from apple cider vinegar) was cultured in static culture for 14 days at 25° C. using the medium and method described by Hestrin and Schramm in Biochemistry Journal Vol. 58 on pages 345-352 (1954). A sample of the pellicle produced on the surface of the culture was removed and prepared for microscopic examination as described above. A micrograph of the bacterial cellulose produced by Acetobacter xylinum strain PARC 017 is shown in
The crystalline structures of the bacterial celluloses produced by Proteus myxofaciens stain IDAC 071005-01 and Acetobacter xylinum strain PARC 017 were measured on a Bruker D8 Discover X-Ray diffraction unit equipped with a GADDS detector. The intensities of the samples were collected over the range from 2θ=5 to 40. The wide-angle diffractometer was used in transmission mode, and the measurements were performed with CuKα1 radiation)(λ=1.54° fit with a 0.5 mm collimator and the scattered photon collected by a GADDS detector. The source was set to 0° while the detector was set to 17°. The raw data were collected and fit mathematically using the method of Vonk described in Journal of Applied Crystallography Vol. 6 on pages 148-152 (1973). The degree of crystallinity of the bacterial cellulose samples was expressed in terms of the crystallinity index (CrI) as defined by Segal et al. in Textile Research Journal Vol. 29 on pages 786-794 (1959). The X-ray defraction pattern of the bacterial cellulose produced by Acetobacter xylinum strain PARC 017 is shown in