MICROSCOPY RESEARCH AND TECHNIQUE 33:32-46 (1996) Lignocellulolysis by Ascomycetes (Fungi) of a Saltmarsh Grass (Smooth Cordgrass) S.Y. NEWELL, D. PORTER,AND W.L. LINGLE Marine Institute, University of Georgia, Sapelo Island, Georgia 31327 (S.Y.N.); Department of Botany, University of Georgia, Athens, Georgia 30602 (D.P., W.L.L.) KEY WORDS Lignin, Lignocellulose, Polyphenolic degradation, Fungi, Transmission electron microscopy, Spartina alterniflora, Phaeosphaeria spartinicola ABSTRACT Lignocellulose (LC) makes up greater than 70% of the mature shoots of the prodigiously photosynthetically productive saltmarsh grass Spartina alternif Zora. Naturally decaying shoots of this cordgrass were examined by transmission electron microscopy (after high-pressure freezing and freeze-substitution) as a means of directly detecting lysis of the LC-rich tissues. Portions of the cordgrass were selected that contained ascomata (sexual reproductive structures) of only one of each of four species of fungi (Kingdom Fungi; Subdivision Ascomycotina): Phaeosphaeria spartinicola and Buergenerula spartinae from leaf blades, Phaeosphaeria spartinae from leaf sheaths, and Passeriniella obiones from naked stems. All four of the ascomycetes were LC-lytic. Phaeosphaeriu spartinicola caused both thinning of LC-rich secondary walls of fiber cells from cell lumina outwards (type 2 soft rot, akin to white rot) and digestion extending from hyphae within longitudinal cavities in the secondary walls (type 1 soft rot). The other three species caused either one or the other type of soft rot. Bacterial erosion of cordgrass cells was found only in the samples of naked stems. Ascomycetous decomposers of standing-dead grasses may have potential for biotechnological applications involving alterations of lignocellulose or toxic polyphenolic substances. 0 1996 Wiley-Liss, Inc. INTRODUCTION Net productivity by the predominant saltmarsh grass (smooth cordgrass, Spartina alterniflora Loisel.) in coastal Georgia (USA) marshes is very high. For example, one estimate of annual production of shoot dry mass is 2,840 g m-' (Schubauer and Hopkinson, 1984). About 70-75% of the dry mass of mature cordgrass shoots consists of lignocellulose (LC) (Hodson et al., 19841, and the lignin component of LC can be a key controlling factor for rates of decomposition of plant material (e.g., Dyer et al., 1990; Robinson, 1990; Taylor et al., 1991). During decay of cordgrass leaves in the natural, standing position, ascomycetous fungi are the principal secondary producers based on decomposition of the leaves. Consider the following example findings: a) living-fungal crop in the dead-cordgrass-leaf system of 106 mg organic mass per gram system organic mass; b) 75% capture of system nitrogen in living-hngal mass; c) standing crop of 1,000 ascomata, minute (approximately 100 km diam) but visible, dark brown sexual reproductive structures immersed in leaf tissue, per square centimeter decaying-leaf abaxial surface; d) ratio of funga1:microalgal:bacterial mass in the decaying-leaf system of 600:4:1 (Leuchtmann and Newell, 1991; Newell, 1993, 1994; Newell and Wasowski, 1995). Background Experimental Testing of Saltmarsh-FungalBreakdown of LC Given that fungi are the principal secondary producers during decay of standing cordgrass and that asco- 0 1996 WILEY-LISS,INC. mycetes clearly can be capable of destruction of LC and mineralization of the phenolic components of the lignin of LC (Haider and Trojanowski, 1980; Nilsson et al., 1989), the findings of one of us (Benner et al., 1984) regarding the capabilities of cordgrass ascomycetes seem counterintuitive. Three species of ascomycetes could not mineralize (to 14C02)physicochemically preexposed cordgrass LC faster than 0.09% per day (neither specifically labeled polysaccharide nor lignin portions). For comparison, a mixed assemblage of bacterial species mineralized polysaccharide and lignin components of the cordgrass LC at rates of 0.9 and 0.5% per day, respectively. Two subsequent sets of investigators obtained results contrary to the first tests of cordgrass LC-lysis by ascomycetes. Torzilli and Andrykovitch (1986) measured 42 day laboratory-microcosm (statically incubated, *NH,NO,) losses of LC and its carbohydrates from living shoots of smooth cordgrass, after oven-drying and grinding, and found that 16-35% of LC was lost but that lignin losses were not detectable by the method used. Bergbauer and Newell (1992) repeated the fungal portion of the 1984 testing (previous paragraph) of fungal capability for cordgrass LC usage, changing some of the incubation conditions. Three key ~~ ~~~ Received January 10,1994; accepted in revised form April 10, 1994. Address reprint requests to Steve Newell, Marine Institute, University of Georgia, Sapelo Island, GA 31327. FUNGAL LYSIS OF LC IN SALTMARSH GRASS changes were as follows: 1) malt and yeast extract (800 mg - 1-I total) were present along with the cordgrass LC (1 g * l-'), simulating the presence of labile carbohydrates and proteins that would be present in senescent leaves (in the 1984 work, only autoclaved marsh sediment was added to the LC); 2) uniformly labeled (14C02)and lignin-specifically (['*CIcinnamic acid) labeled LC were used rather than lignin- and polysaccharide-specifically ([14CClglucose)labeled LC; 3) the incubations were performed statically, not with 90 rpm rotary agitation. Bergbauer and Newell (1992) found that for the principal fungal decomposer of cordgrass leaves (Phaeosphaeria spartinicola Leuchtmann [see Newell and Wasowski, 199511, mineralization of LC polysaccharide took place a t 0.83% per day, mineralization of LC lignin occurred at 0.07% per day, and fungal growth efficiency on LC was about 40%(the last with urea as a nitrogen source and no malt or yeast extract present). Potential causal factors for the disparities between the 1984 and 1992 results for LC lysis by cordgrass ascomycetes include the following: 1) agitation depressed fungal LC-lysis in the 1984 experiments; 2) the glucose radiolabel in the 1984 experiments was incorporated into LC by the cordgrass in positions that favored fungal anabolism of the radiolabel; 3) the presence of non-LC carbohydrates in the 1992 experiments favored lignin alteration and thereby enhanced LC oxidation (see Eriksson et al., 1990; Reid, 1991; Reid and Deschamps, 1991). In all three of the above experiments, it is possible that nitrogen sufficiency (from marsh sediment, green leaves, and other added N) suppressed fungal oxidation of lignin (McCarthy et al., 1984; Reid, 1991), and in the 1984 experiment the physicochemical preexposure of LC may have made it available to bacterial enzymes that would not naturally have been physically capable of contacting the LC (Backa et al., 1993; Flournoy et al., 1993; Robinson, 1990; Srebotnik and Messner, 1990). Background Electron Microscopy of Lignocellulolysis As an alternative to the empirical route for determination of the LC-lytic capabilities of cordgrass ascomycetes, we chose to follow the lead of wood-decay science (Adaskaveg et al., 1991; Blanchette, 1991; Eriksson et al., 1990) and examine cordgrass-ascomycete LC-lysis in as direct a fashion as possible: by electron-microscopically examining the interactions that occur in nature between fungal hyphae and cordgrass LC. Gessner et al. (1972) recorded the colonization of internodal surfaces of smooth cordgrass by Buergenerula spartinae Kohlm. & Gessner via scanning electron microscopy. We excavated a bit deeper (using transmission electron microscopy [TEMI), and display below the results of hyphal activity by four common cordgrass ascomycetes, each an inhabitant of one of three vegetative components of shoots (leaf blades, leaf sheaths, or stems). We focus most strongly on decay caused by P. spartinicola, because we know more about the productivity of this species in the cordgrass-decomposition system than we do for the other three species (Newell, 1993; Newell and Wa- 33 Fig. 1. Diagrammatic representation of fiber cells of a grass, drawn from our TEMs of smooth cordgrass (Spartinu alterniflom). The upper cell (about 10 km across) is undecayed;the lower two cells partially drawn are undergoing fungal lysis. Letter labels in this and all following figures either lie directly upon the structure indicated or are shown with a bar that ends upon the structure. ML, the middlelamellar layer between cells (the whole area between the s, layers of adjacent cells includes the primary walls and is termed the compound middle lamella); CA, a cavity being digested out of compound middlelamellar and secondary-walllayers; CC, the cell-corner portion of the middle lamella; FCW, fungal cell wall; GR, granular or amorphous darkened digesta of fiber-cell material; HS, sheath material formed surrounding hyphae; HY, cross-section of fungal hypha causing LC lysis; L, the cell lumen; PC,a pit canal joining two adjacent cells; RS, reticulate structure resulting from lignocellulolysis (LC lysis); S,-S, the three principal portions of the secondary wall of a fiber cell. sowski, 1995), and we compare the results for P. spartinicola to those for the other species of ascomycetes. We directed our examination of fungal activities to those that occur in the most lignocellulose-rich portions of the shoots, namely the sclerenchymatous fiber cells of the bundle sheaths or subepidermal structural layers (Anderson, 1974; Juniper, 1979). Lignocellulose (LC) is a compact matrix of lignin, hemicellulose, and cellulose macromolecules (see Fig. 1.1 of Eriksson et al., 1990). Readers unfamiliar with lignocellulosic structures may refer to the diagram in Figure 1 for orientation; boldface labels used in Figure 1 and in our electron micrographs are given in the text below also for reference. The highly lignocellulosic fiber (F) struc- 34 S.Y. NEWELL ET AL. Fig. 3. Ascoma (A at its base) of P. sportinicola forming a small through the abaxial epidermis of a leafblade pore at its ostiolar tip (T) of smooth cordgrass, showing partial, tangential sections of ascospores (AS) within and released onto the leaf exterior. Note that the aGacent fiber tissue (F) is not as extensively digested as that of Fig. 2. Bar = 30 wm. Fig. 2. Transverse section through a portion of the adaxial side of a standing decaying leaf blade of smooth cordgrass, from a leaf segment exclusively occupied by the sexual structures (aseomata; Fig. 3) ofPhaeosphaeria sportinicola. Hyphae (HY)are present in cell lumina and within the thick, lignocellulosiesecondary walls (S)of the adaxial fiber-cell band (F). Note that the formerly chloroplast-containing and mesophyll chlorenchyma (MI are outer-bundle-sheath cells (0) nearly completely digested. Scale bar = 20 km. tures of grasses have lignin distributed throughout the thick, multilayered secondary wall (S1 [outermost, thin], S2[central, thick], and S3[innermost, thin]) (Fig. 1).However, lignin is most concentrated in the middle lamella (ML), the thin hemicelluloseAignin layer lying between fiber cells, including the wider cell corners (CC), and in some of the xylem cells (He and Terashima, 1991). In wood, the middle lamellakell corners may have three to four times the concentration of lignin found in the secondary walls (Higuchi, 1980), but in grasses the enrichment may be lower (two times [He and Terashima, 19911). Regardless of the enrichment factor, the greater width of the fiber-cell wall than the middle lamella results in the fiber secondary walls containing most (6040%)of the lignin of LC (Eriksson et al., 1990; Zhai and Lee, 1989). Fungal decomposers of LC have traditionally been categorized as causing white rot (equal loss rates for LC components [“simultaneous” white rotl, or disproportionately high loss of lignin + hemicellulose [“delignification” white rotl), brown rot (disproportionately high loss of hemicellulose + cellulose), or soft rot (variable proportionality of loss among the three components of LC) (Blanchette, 1991; Eriksson et al., 1990). White- and brown-rot fungi situate their hyphae (HY) primarily within fiber-cell lumina (L) and cause erosion of cell walls (white rot only) or depolymerization of macromolecules progressing from the lumina toward the middle lamellae (Fig. 1, bottom right). Soft-rot fungi often degrade LC primarily by forming cavities (CAI within the walls of lignocellulosic cells (type 1 decay) (Fig. 1, bottom left), but they may also cause outward-progressing erosion of secondary fiber-cell walls adjacent to hyphae situated in cell lumina (type 2 decay) (Nilsson et al., 1989). When only type 2 decay is present, differentiation from white rot is not clear (Nilsson et al., 1989). Ascomycetous fungi are generally considered to be agents of soft rot, but some species are capable of causing white rot (Eriksson et al., 1990; Nilsson et al., 1989). Ultrastructural aspects of lignocellulolysis have been documented for species of wood-decomposing fungi (especially Phanerochuete chrysosporium Burds.) FUNGAL LYSIS OF LC IN SALTMARSH GRASS 35 Fig. 4. Hyphae (HY) of P . spartinicola digesting lignocellulosic cluding the triangular cell-corner (CC) portion of the middle lamella. secondary fiber-cell walls (S,) primarily from within cell lumina (L) Note the secondary front (digestive?) (F2) located at a distance from but also directly adjacent to or within the middle lamella (ML),in- the hyphae. Bar = 5 pm. 36 S.Y. NEWELL ET AL. Fig. 5. Hyphae (HY) of P. spartinicola digesting lignocellulosic secondary fiber-cell walls (S,) primarily from within longitudinal cavities lysed out of the walls. Note resistance t o lysis of innermost layer of secondary wall (Q. Bar = 5 pm. FUNGAL LYSIS OF LC IN SALTMARSH GRASS 37 seleau and Ruel, 1992; Kim et al., 1993; Nilsson et al., 1989; Ruel, 1990). Fig. 6. Longitudinal section through fiber cells of a leaf' blade of smooth cordgrass showing hyphae (HY) of P. spartinicolu within the cell lumen (left upper hypha) and digesting a longitudinal cavity within the cell wall and middle lamella (right lower hypha). ML, middle lamella; S,, secondary fiber-cell wall. Bar = 3 km. whose capacity for oxidation of lignin and LC carbohydrates is well known (Daniel et al., 1990; Eriksson et al., 1990; Hale and Eaton, 1985; Nilsson et al., 1989; Ruel, 1990; Ruel et al., 1988). The characteristic TEM appearance of LC of secondary walls of fiber cells under attack by lignocellulolytic fungi is a darkening of wall material and formation of reticulate structure (RS) and/or a crumbling into amorphous bits or granules (GR) (Fig. 1) (Ruel, 1990). The darkened material is believed t o be modified (lysed and spontaneously partially repolymerized) lignin. In monocotyledenous shoots, the low-lignin mesophyll-chlorenchyma tissue (M) adjacent to highly lignocellulosic fiber bundles can be nearly totally destroyed by LC-lytic fungi (e.g., rice: Yu et al., 1990; date palm: Adaskaveg et al., 1991). A sheath (HS) commonly appears surrounding the hyphae (HY) that cause the destruction of the LC. The sheath can contain LC-lytic enzymes and, probably, bits of the modified lignin (Hale and Eaton, 1985; Jo- MATERIALS AND METHODS Standing decaying leaf blades and leaf sheaths of smooth cordgrass (Spartina alternifloru) were collected from marshes of Sapelo Island, Georgia (USA) (see Pomeroy and Wiegert, 1981); naked, fallen stems were collected as portions of drift wrack in the same marshes. Living, green leaf blades were collected and used to prepare the diagram in Figure 1.Portions of the decaying shoot material were selected on the basis of apparent occupancy by single species of ascomycetes. Sole occupancy was determined by direct-microscopically determined exclusive presence of mature ascomata (sexual reproductive structures); in such areas, there is evidence (immunoassay vs. chemical-index assay) that most, if not all, of the hyphae within cordgrass tissue is formed by the sexually reproductive species (Newell, 1994). The four species of ascomycetes selected for study were Phaeosphaeria spartinicola (leaf blades), Buergenerula spartinae-(leaf blades), Phaeosphaeria spurtinae (Ellis & Everhart) Shoemaker & Babcock (leaf sheaths), and Passeriniella obiones (Crouan & Crouan) Hyde & Mouzouras (naked stems) (see Kohlmeyer and Volkmann-Kohlmeyer, 1991). We recognize that evidence of LC lysis in our micrographs could have resulted from microbial activity that took place before thorough pervasion by the fungal species that we were targeting; this caveat would be least likely to apply to the two species involved in decay of leaf blades, since they are the first microbial invaders of the senescent blades (Newell, 1993). Small pieces (1-2 mm2 surface area) were dissected from the shoot tissue and prepared for transmission electron microscopy (TEM) by high-pressure freezing followed by freeze-substitution (Lingle et al., 1992). The pieces were vacuum-infiltrated with 15% dextran prior to placement in planchettes. Void space in the planchettes was filled with dextran solution. The planchettes were inserted into a Balzers HPM 010 high-pressure freezer (see Dahl and Staehelin, 1989) where specimens were sprayed with liquid nitrogen immediately after pressurization to 2,100 bars. After freezing, specimens were collected in liquid N, and transferred to substitution fluid a t -80°C where they were held for 72 h and then gradually (over a 6 h period) warmed to room temperature. The substitution fluid consisted of 2% osmium tetroxide and 0.01% uranyl acetate in HPLC-grade acetone. After several acetone washes, the specimens were infiltrated slowly with either Embed 812 or Quetol651 epoxy resins (EM Sciences, Ft. Washington, PA). Ultrathin sections were cut and poststained with uranyl acetate and Reynolds' lead citrate (Lingle, 1989) prior to examination in a Zeiss EM 902A transmission electron microscope. We prepared the following number of electron micrographs for each species, each from a separate field of view: P. spartinicola, 66;B . spurtinae and P. obwnes, 36 each; P. spartinae, 14. The micrographs presented as figures below were chosen to be representative of the images found for each species. 38 S.Y. NEWELL ET AL. Fig. 7. Transverse section through fiber cells of a leaf blade of smooth cordgrass containing hyphae ( H Y ) of P. spartinicola both within the cell lumen and in lysed cavities in the secondary wall (S.J Note glycogenrosettes (G)and electron-densehyphal-sheath(HS) material and granular wall lysate (GR). Bar = 2Fm. Fig. 8. Leaf-blade xylem of smooth cordgrass containing hyphae = 5 pm. (HY)of P. spartinicola. Bar Fig. 10. Digestion of xylem (X)and phloem (P)in leaf blades of smooth cordgrass by hyphae (HY)of Buergenerulu spartinue. Note that cells of the outer bundle sheath (0) have been fully digested and lost. Bar = 20 pm. RESULTS Fig. 9. Production of both reticulate dark lysate (RS)and dark amorphous lysate (GR) (see Background) as a result of digestion of fiber cells of smooth eordgra~sleaf blade by hyphae (HY)of P. spurtinicolu. Bar = 5 y.m. Phaeosphaeria spartinicola and Cordgrass LC Abaxial portions of cordgrass leaves occupied by mature ascomata of P. spartinicola (see Leuchtmann and Newell, 1991; Newell and Wasowski, 1995) exhibited areas of the former leaf mesophyll chlorenchyma, outer bundle sheath, and phloem in which only thin remnants of plant cell walls and empty fungal hyphae remained (Fig. 2). Fiber tissue was also extensively degraded, with nearly all cells containing fungal hyphae in some sections (Fig. 2). In other sections, fewer fiber cells were as heavily damaged (Fig. 3). Hyphae of P. spartinicola brought about LC lysis from within cell lumina (Figs. 4-9), from positions along compound middle lamellae (Figs. 4-7, 9), and from cavities formed within secondary walls of fiber cells (Figs. 5-7). In some cells, there were two electron-opaque (digestive?) fronts visible within secondary fiber walls, one dark and close to the hyphae, one fainter and often distant from the hyphae (Fig. 4). Hyphae within lumina of xylem cells (which can be the most heavily lignified in grasses [He and Terashima, 19911) caused dissolution of the cell walls, leaving remnant pit membranes in some cases (Fig. 8). Darkened, amorphous, cell-wall lysate often was directly juxtaposed to the fungal hyphal wall, appearing to become a part of the hyphal sheath (Fig. 7). Alongside some fiber cells, the.middle ktmelh appeared be resistant to degradation, and in others it was clearly darkened and al- 40 S.Y.NEWELL ET AL. Fig. 11. Hyphae (HY)of B . spartinae in a fiber cell of smooth cordgrass. Note that digestion of the lignocellulosic secondary wall (S,)and production of granular lysate (GR)apparently extends primarily from within the cell lumen and pit canals (PO. Bar = 5 Fm. tered if not dissolved or mineralized, with fungal hy- walls, P. spartinicola could also produce reticulate phae growing directly within it (Figs. 4-7, 9). In darkened lysate (Fig. 9). The thin S3layer of the fiber addition to the darkened amorphous lysate of fiber walls was often apparently resistant to lysis (Figs. 5, FUNGAL LYSIS OF LC IN SALTMARSH GRASS 41 Fig. 13. Hyphae (HY) of Phaeosphaeria spartinae causing erosional thinning from within lumina of fiber-cell walls of a leaf sheath of smooth cordgrass. Bar = 5 pm. Fig. 12. Hyphae (HY) of B. spartinae in the lumen and wall of an abaxial epidermal cell of a leaf blade of smooth cordgrass. Note the darkening of the epidermal-cell wall just under the plant cuticle (CU) around vacant (V)spaces that appear to have previously contained hyphae (compare the vacant spaces to the intact hypha (IH)a t top center). Bar = 10 pm. 7). Some hyphae within fiber cells contained dense concentrations of glycogen rosettes (Fig. 7) (glycogen is a fungal storage carbohydrate) (see Cooke and Whipps, 1993; Lingle, 1989). Buergenerula spartinae and Cordgrass LC B . spartinae caused xylem dissolution and fiber-wall erosion, with alteration of middle lamellae (Figs. 10, 11). Although we observed hyphae of B . spartinue within the secondary walls of fiber cells, it was most commonly seen to cause secondary-wall digestion from within cell lumina and their pit canals (Fig. 11).Hyphae of B . spartinae grew within the outer portions of epidermal cells or between the cuticle and the wall of the epidermis and appeared to cause the surrounding material to become very electron opaque (Fig. 12; compare epidermal walls at the top of Fig. 2). PhaeosphaeM spartinae and Cordgrass LC The hyphae ofp. spartinae caused thinning to total from cell lumina outwards (l?igs. loss of fiber cell 1 3 9 14); these hyphae were not observed to grow in longitudinal cavities in fiber-cell secondary walls. Mid- Fig. 14. Hyphae (HY) of P. spartinae causing erosional lysis of subepidermal abaxial fiber and adjacent parenchyma cells. Note short bore holes (B)between some cells, sometimes passing directly through cell-corner portions of middle lamellae (lower marked hole). Bar = 20 pm. S.Y. NEWELL ET AL. 42 Fig. 15. Hypha (HY) of P.spartime at the eroding edge of a fiber cell. Note that the hyphal sheath (HS)is continuous with dark cell-wall lysate, under which reticulate wall digesta (RS)can be seen. Bar = 2 pm. dle lamellae were largely unlysed by P. spartinae; this species grew fkom cell to cell via pit connections or through short, transverse bore holes (Fig. 14). The thinning of fiber walls characteristically left a serrate edge to the wall, with a very electron-opaque, thin covering that was often a t least partially continuous with hyphal-sheathing material (Fig. 15). Beneath the dark covering, areas of reticulate-digested secondary wall could be seen (Fig. 15). Hyphae of P. spartinae growing in the outer edges of epidermal-cell walls caused a marked darkening (Fig. 14, upper left) like that of B. spartinae (Fig. 12). the stem, sometimes in close proximity to fungal hyphae (Fig. 181, and in other instances in areas wherein remaining fungal hyphae appeared empty of cytoplasm (not shown). Bacteria outside of the plant cuticle in epiphytic positions were commonly observed for all four types of fungal samples, but bacterial epidermal erosion was found only for the P. obiones-occupied portions of shoots. DISCUSSION All four of the cordgrass ascomycetous decomposers were capable of lysis of lignocellulose (LC). Both type 1 (cell-wall decay via wall-cavity formation) and type 2 Passeriniella obiones and Cordgrass LC (cell-wall erosion from the interior [see Nilsson et al., Both cell lumina and secondary-wall cavities were 19891) soft rot were found. B. spartinae and P. spartinae occupied by hyphae of P. obiones (Figs. 16,17). Hyphae caused only or predominantly type 2 decay. P. obiones within secondary walls commonly caused cavities to caused type 1 only or predominantly, and P. spartinienlarge until only disjunct innermost and outermost cola caused both types 1 and 2. The middle lamella, one layers remained (Fig. 17). Middle lamellae were some- of the portions of cordgrass LC likely to be the most times not apparently changed (Fig. 17), and in other lignin-rich (He and Terashima, 1991), was often clearly cases parts of the middle lamellae were altered (Fig. altered by P. spartinicola, the principal decomposer of 16, top center). Colonies of bacteria were found in leaf blades (Figs. 4, 6, 9) (Newell, 1993). TEM images eroded pits in the very thick-walled epidermal cells of for cordgrass ascomycetes acting upon LC-rich cells FUNGAL LYSIS OF LC IN SALTMARSH GRASS 43 Fig. 16. Hyphae (HY) ofPusserznieZluobiones in the lumen (L) and in a longitudinal cavity in the wall of a subepidermal fiber cell of a naked stem of smooth cordgrass. Note continuity of dark hyphal sheath (HS) with dark wall lysate (GR)and darkening of a portion of the cell corner (CC) of the middle lamella. Bar = 2 q. Fig. 17. Hyphae (HY)of P. obiones as in Fig. 16, but here much of the secondary wall of the cell is digested and collapsed, apparently from broadening of longitudinal cavities (CA) formed as in Fig. 16, between what may be multiple SIB, layers (see Adaskaveg et al., 1991). Bar = 2 pm. were characteristic of lignin-destroying fungi: formation of electron-dense granules or networks as the cellulose/hemicellulose fabric of the secondary walls of fiber cells was unraveled (Figs. 2, 4, 5, 9, ll, 15, 16). Compare the figures of Ruel et al. (1988) for Phanerochaete chrysosporium, a very well-known lignin-oxidizing basidiomycetous fungus, and figures of Nilsson et al. (1989)for Daldinia concentrica [Bolt.:Fr.l Ces. & De Not, a white-rotting xylariacean ascomycete capable of causing 40%losses of lignin. For cordgrass ascomycetes and for LC destruction documented for other fungi, the dark granules or networks often were continuous with fungal extracellular hyphal-sheathing material (e.g., Figs. 7, 15, 16). This continuity suggests that the hyphal sheaths contained a mixture of partially oxidized and partially repolymerized lignin with fungal extracellular polysaccharides and LC-lytic enzymes (Daniel et al., 1990; Faix et al., 1985; Hale and Eaton, 1985; Joseleau and Ruel, 1992; Nicole et al., 1993; Nilsson et al., 1989; Ruel et al., 1988). Because all four of the cordgrass ascomycetes of this report make melanin (polyphenolic, brown, electron-dense pigments; see the melanin deposits in ascospore walls of P. spartinicola in Fig. 3) and because fungal melanin is usually deposited in walls and hyphal sheaths, it may be that melanin plus repolymerized lignin is formed as a conglomerate by the cordgrass ascomycetes (see Bell and Wheeler, 1986; Kohlmeyer and Volkmann-Kohlmeyer, 1991). The hypothesized lignomelanin conglomerates might be measured as lignin by some methods (e.g., acid-detergent methods [Torzilli and Andrykovitch, 19861),resulting in underestimates of lignolysis (Newell, 1993). Nilsson et al. (1989) examined by TEM the patterns of LC lysis in wood for several ascomycetes. In parallel, they chemically measured losses of lignin and LC carbohydrates. They found that 1) there was no difference between type 1 and type 2 soft-rot fungi with respect to losses of lignin and LC carbohydrates (12-44% loss of lignin, 22-91% of glucan, 28-92% of xylan, 4 weeks decay), 2) percentage losses of lignin ranged from x 1 to X 0.5 the percentage losses of total dry mass (see also Abe, 1989) and 3) losses of syringyl lignin units were greater than losses of vanillyl units, leading to decreases in the syringy1:vanillyl (S:V) ratio in the remaining lignin. The data of Bergbauer and Newell (1992) and Haddad et al. (1992) suggest that similar chemical changes to those found by Nilsson et al. (1989) occur for the ascomycete attack on cordgrass depicted herein. Haddad et al. (1992) found that for naturally decaying, standing leaf blades of smooth cordgrass, lignin phenols were lost at a rate greater than that for loss of total organic mass, implying that fungal lysis of cordgrass LC does not result in rising concentrations of fungal-resistant lignin. Bergbauer 44 S.Y. NEWELL ET AL. Fig. 18. Epidermal cells of naked stem of smooth cordgrass, with erosion bacteria (B)in pits in the outer secondary walls and hyphae (HY)of P. obiones in lumina of the same cells. Bar = 5 pm. and Newell (1992) found that cinnamyl and syringyl phenols of extracted cordgrass LC were especially susceptible to alteration by P. spartinicola, with S:V falling (from 0.9 to 0.6) over 45 days incubation. However, Haddad et al. (1992) did not find a decrease in S:V for standing decaying cordgrass leaves, suggesting that selective syringyl alteration may not be as marked in the field as it can be in vitro. Based on our TEM results and the findings of Bergbauer and Newell (1992) and Haddad et al. (1992), we suggest that the higher ligninphenol concentrations for standing-dead than for green smooth cordgrass shoots reported by Alberts et al. (1992) were probably more the result of plant maturation and senescence processes than, as Alberts et al. inferred, the result of inability of cordgrass ascomycetes to cause lignin losses. Some portions of the LC-rich fiber tissues of cordgrass were not heavily invaded and lysed, even in direct juxtaposition with ascomata of the most aggressively LC-lytic species, P. spartinicola (Fig. 3). Patchiness of decay extensiveness is well known for wooddecay fungi (Eriksson et al., 1990). It seems reasonable that it would be advantageous to the fungus to avoid causing near-total loss of structural integrity in the decaying leaf, even though it does decay both chlorenchymal nonstructural and lignified structural tissue, since a large proportion of the fungal propagule-projecting bodies (the ascomata) (see Newell and Wasowski, 1995) might lose the ability to deposit spores on new substrate within the shoot canopy if the leaves were to rapidly crumble to the marsh floor. Our TEM inspection of decaying portions of smoothcordgrass leaf blades and sheaths confirms Newell’s (1993) conclusion, based on epifluorescence-microscopic assay of bacterial biovolume, that bacteria contribute negligibly to microbial standing crop of decaying, standing cordgrass leaves. We found no images of bacteria causing lysis of plant cells for leaf preparations. For naked stems, however, that had fallen to the marsh sediment prior to collection, we did find bacterial colonies that were apparently cell-wall-lytic (Fig. 18). Newell (1993) suggested that tunneling bacteria might be important decomposers of fallen portions of decaying smooth cordgrass, based on unpublished information from the study of Benner et al. (1984) (see Newell, 1993). The bacteria that we found in pits in epidermal cells of cordgrass, though, were typical of erosion bacteria (Singh and Butcher, 1991; see also Barnabas, 1992). These erosion bacteria in cordgrass stems appeared to be embedded in an extracellular matrix (Fig. 18), perhaps as an adaptation to periodic dryness (Roberson et al., 1993). Finally, our TEM findings for cordgrass ascomycetes have implications with respect to fungal biotechnology. All four of the ascomycetes presented images characteristic of LC-digesting fungi, and a t least one of them destroyed LC-rich cells to an extent comparable to white-rot basidiomycete fungi (see Adaskaveg et al., 1991; Eriksson et al., 1990; Ruel, 1990). Thus, one might expect that the cordgrass ascomycetes (and ascomycetes of other standing-dead grass shoots) would be candidates for the types of biotechnological usage to which the LC-degrading basidiomycetes have been put, especially destruction of toxic phenolic substances (including chlorinated molecules), alterations of paper-industry raw materials or wastes, and improvement of ruminant fodder (e.g., Abernethy and Walker, 1993; Akin et al., 1993; Blanchette, 1991; Hammel, 1992; Kirkpatrick et al., 1990; Loske et al., 1990; Moyson and Verachtert, 1991; Ollikka et al., 1993; Yadav and Reddy, 1993). Bergbauer et al. (1992) have preliminarily confirmed this expectation: they discovered that the anamorph (asexual form) of an ascomycete of standing, decaying shoots of a sedge from freshwater wetlands could effectively dechlorinate and degrade polyphenolrich effluents from a pulp-mill bleachery. ACKNOWLEDGMENTS Financial support for this research was received from the National Science Foundation (grant OCE-9115642) and the Sapelo Island Research Foundation. We thank Darrell Casey for expert assistance with preparation of our figures. This is UGMI contribution 747. REFERENCES Abe, Y. (1989) Effect ofmoisture on decay of wood by xylariaceous and diatrypaceousfungi and quantitative changes in the chemical components of decayed woods. Trans. Mycol. Soc. Japan, 30:169-181. FUNGAL LYSIS O F LC IN SALTMARSH GRASS Abernethy, G.A., and Walker, J.R.L. (1993)Degradation of the insecticide Hydramethylnon by Phanerochaete chrysosporium. Biodegradation, 4131-139. Adaskaveg, J.E., Blanchette, R.A., and Gilbertson, R.L. (1991)Decay of date palm wood by white-rot and brown-rot fungi. Can. J. Bot., 69:615-629. Akin, D.E., Sethuraman, A., Morrison, W.H., Martin, S.A., and Eriksson, K.-E.L. (1993)Microbial delignification with white rot fungi improves forage digestibility. Appl. Environ. Microbiol., 59:42744282. Alberts, J.J., Filip, Z., Rice, M.T., Hedges, J.I., and Jacobsen, T.R. (1992)CuO-oxidation products, acid-hydrolyzable monosaccharides and amino acids of humic substances occurring in a salt marsh estuary. Org. Geochem., 18:171-180. Anderson, C.E. (1974)A review of structure in several North Carolina salt marsh plants. In: Ecology of Halophytes. R.J. Reimold and W.H. Queen, eds. Academic Press, New York, pp. 307-344. Backa, S., Gierer, J., Reitberger, T., and Nilsson, T. (1993)Hydroxyl radical activity associated with the growth of white-rot fungi. Holzforschung, 47:181-187. Barnabas, A.D. (1992)Bacteria on and within leaf blade epidermal cells of the seagrass Thulussodendmn ciliutum (Forssk.) Den Hartog. Aquat. Bot., 43:257-266. Bell, A.A., and Wheeler, M.H. (1986)Biosynthesis and functions of fungal melanins. Ann. Rev. Phytopathol., 24:411-451. Benner, R., Newell, S.Y., Maccubbin, A.E., and Hodson, R.E. (1984) Relative contributions of bacteria and fungi to rates of degradation of lignocellulosic detritus in salt-marsh sediments. Appl. Environ. Microbiol., 48:36-40. Bergbauer, M., and Newell, S.Y. (1992)Contribution to lignocellulose degradation and DOC formation from a salt marsh macrophyte by the ascomycete Phaeosphaeria spartinicola. FEMS Microb. Ecol., 86341-348. Bergbauer, M., Eggert, C., and Kalnowski, G. (1992)Biotreatment of pulp mill bleachery effluents with the coelomycetous fungus Stugonospom gigaspom.Biotech. Lett., 14317-322. Blanchette, R.A. (1991)Delignification by wood-decay fungi. Ann. Rev. Phytopathol., 29381-398. Cooke, R.C., and Whipps, J.M. (1993)Ecophysiology of Fungi. Blackwell Scientific, Oxford, 337 pp. Dahl, R., and Staehelin, L.A. (1989)High-pressure freezing for the preservation of biological structure: Theory and practice. J. Electron Microsc. Tech., 13:165-174. Daniel, G., Pettersson, B., Nilsson, T., and Volc, J. (1990)Use of immunogold cytochemistry to detect Mn(I1)-dependent and lignin peroxidases in wood degraded by the white rot fungi Phanerochaete chrysosporium and Lentinula edodes. Can. J. Bot., 68:920-933. -Dyer, M.L., Meentemeyer, V., and Berg, B. (1990)Apparent controls of mass loss rate of leaf litter on a regional scale. Scand. J. For. Res., 5311-323. Eriksson, K.-E.L., Blanchette, R.A., and Ander, P. (1990)Microbial and Enzymatic Degradation of Wood and Wood Components. Springer-Verlag. Berlin, 407 pp. Faix, O.,Mozuch, M.D., and Kirk, T.K.(1985)Degradation of gymnosperm (guaiacyl) vs. angiosperm (syringyl/guaiacyl) lignins by Phanerochaete chrysosporium. Holzforschung, 39:203-208. Flournoy, D.S., Paul, J.A., Kirk, T.K., and Highley, T.L. (1993) Changes in the size and volume of pores in sweetgum wood during simultaneous rot by Phanerochaete chrysosporium Burds. Holzforschung, 47:297-301. Gessner, R.V., Goos, R.D., and Sieburth, J.McN. (1972)The fungal microcosm of the internodes of Spartinu alterniflom. Mar. Biol., 16:269-273. Haddad, R.I., Newell, S.Y., Martens, C.S., and Fallon, R.D. (1992) Early diagenesis of lignin-associated phenolics in the salt marsh grass Spurtinu alterniflom. Geochim Cosmichim Acta, 56:37513764. Haider, K., and Trojanowski, J. (1980)A comparison of the degradation of I4C-labeled DHP and corn stalk lignins by micro- and macrofungi and bacteria. In: Lignin Biodegradation: Microbiology, Chemistry, and Potential Application, Vol. 1. T.K. Kirk, T. Higuchi, and H.-M. Chang, eds. CRC Press, Boca Raton, pp. 111-134. Hale, M.D., and Eaton, R.A. (1985)The ultrastructure of soft rot fungi. 11.Cavity-forming hyphae in wood cell walls. Mycologia, 77: 594-605. Hammel, K.E. (1992)Oxidation of aromatic pollutants by lignin-degrading fungi and their extracellular peroxidases. In: Degradation of Environmental Pollutants by Microorganisms and their Metal- 45 loenzymes. H. Sigel and A. Sigel, eds. Marcel Dekker, New York, pp. 41-60. He, L.,and Terashima, N. (1991)Formation and structure of lignin in monocotyledons. IV. Deposition process and structural diversity of the lignin in the cell wall of sugarcane and rice plant studied by ultraviolet microscopic spectroscopy. Holzforschung, 45191-198. Higuchi, T. (1980)Lignin structure and morphological distribution in plant cell walls. In: Lignin Biodegradation: Microbiology, Chemistry,and Potential Applications, Vol. 1. T.K. Kirk, T. Higuchi, and H.-M. Chang, eds. CRC Press, Boca Raton, pp. 1-19. Hodson, R.E., Christian, R.R., and Maccubbin, A.E. (1984)Lignocellulose and lignin in the salt marsh grass Spartinu alterniflom: Initial concentrations and short-term, post-depositional changes in detrital matter. Mar. Biol., 81:l-7. Joseleau, J.P., and Ruel, K. (1992)Ultrastructural examination of lignin and polysaccharide degradation in wood by white-rot fungi. In: Biotechnology in Pulp and Paper Industry. M. Kuwahara and M. Shimada, eds. Uni Publishers, Tokyo, pp, 195-202. Juniper, B.E. (1979)The structure and chemistry of straw. Agric. Prog., 54:18-27. Kim, Y.S., Goodell, B., and Jellison, J. (1993)Immunogold labelling of extracellular metabolites from the white-rot fungus Tmmetes uersicolor. Holzforschung, 47~25-28. Kirkpatrick, N., Reid, I.D., Ziomek, E.,and Paice, M.G.(1990)Biological bleaching of hardwood krafi pulp using Tmmetes (Coriolus) foam. Appl. uersicolor immobilized in polyurethane . . .. Microbiol. Biotechnol., 33~105-108. Kohlmever, J.. and Volkmann-Kohlmever. B. (1991)Illustrated kev to thifil&entous higher marine h - g i . 'Bot. Mar., 34:l-61. Leuchtmann, A., and Newell, S.Y. (1991)Phaeosphaeria spartinicola, a new species on Spartinu.Mycotaxon, 41:l-7. Lingle, W.L. (1989)Enhanced staining of the Basidiomycete Panellus stypticus prepared for transmission electron microscopy by freeze substitution. Crypt. Bot., 1:236-242. Lingle, W.L., Clay, R.P., and Porter, D. (1992)Ultrastructural analysis of basidiosporogenesis in Panellus stypticus. Can. J. Bot., 7 0 2017-2027. Loske, D., Huttermann, A., Majcherczyk, A., Zadrazil, F., Lorsen, H., and Waldinger, P. (1990)Use of white rot fungi for the clean-up of contaminated sites. In: Advances in Biological Treatment of Lignocellulosic Materials. M.P. Coughlan and M.T. Amaral Collaw, eds. Elsevier, London, pp. 311-321. McCarthy, A.J., Macdonald, MJ., Paterson, A,, and Broda, P. (1984) Degradation of [14Cllignin-labeled wheat lignocellulose by whiterot fungi. J. Gen. Microbiol., 1301023-1030. Moyson, E., and Verachtert, H. (1991)Growth of higher fungi on wheat straw and their impact on the digestibility of the substrate. Appl. Mimbiol. Biotechnol., 36421-424. Newell, S.Y. (1993)Decomposition of shoots of a salt-marsh grass; methodology and dynamics of microbial assemblages. Adv. Microb. E d . , 13:301-326. Newell, S.Y. (1994)Ecomethodology for organo-osmotrophs, prokaryotic unicellular versus eukaryotic mycelial. Microb. Ecol., 28151157. Newell, S.Y., and Wasowski, J. (1995)Sexual productivity of a key saltmarsh microbial secondary producer. Estuaries, 18:241-249. Nicole, M., Chamberland, H., Rioux, D., Lecours, N., Rio, B., Geiger, J.P., and Ouellette, G.B. (1993)A cytochemical study of extracellular sheaths associated with Rigidoporus lignosus during wood decay. Appl. Environ. Microbiol., 592578-2588. Nilsson, T., Daniel, G., Kirk, T.K., and Obst, J.R. (1989)Chemistry and microscopy of wood decay by some higher ascomycetes. Holzforschung, 43:ll-18. Ollikka, P., Alhonmaki, K., Leppanen, V.-M., Glumoff, T., Raijola, T., and Suominen. I. (1993)Decolorizatioh of azo. triohenvl methane. heterocyclic, and polymeric dyes by lignin peroxidase-isoenzymes from Phanerochaete chrysosporium. Appl. Environ. Microbiol., 59 4010-4016. Pomeroy, L.R., and Wiegert, R.W. (1981)The Ecology of a Salt Marsh. Springer-Verlag. Berlin, 271 pp. Reid, I.D. (1991)Nutritional regulation of synthetic lignin (DHP) d e g radation by Phlebiu (Merulius) tremellosa: Effects of nitrogen. Can. J . Bot., 69~156-160. Reid, I.D., and Deschamps, A.M. (1991)Nutritional regulation of synthetic lignin (DHP) degradation by the selective white-rot fungus Phlebia (Merulius) tremellosa: Effects of glucose and other cosubstrates. Can. J. Bot., 69147-155. Roberson, E.B., Chenu, C., and Firestone, M.K. (1993)Microstruc- 46 S.Y.NEWELL ET AL. tural changes in bacterial exopolysaccharides during desiccation. Soil Biol. Biochem., 25:1299-1301. Robinson, J.M. (1990)Lignin, land plants, and fungi: Biological evolution affecting P h a n e m i c oxygen balance. Geology, 15607-610. Ruel, K. (1990)Ultrastructural alterations of wood cell walls during degradation by fungi. In: Advances in Biological Treatment of Lignocellulosic Materials. M.P. Coughlan and M.T. Amaral Collafo, eds. Elsevier, London, pp. 117-128. Ruel, K., Butez, M.-F., Barnoud, F., and Joseleau, J.-P. (1988)Evidence by electron microscopy of lignin recondensation during fungal degradation. Cellulose Chem. Tech., 22287-292. Schubauer, J.P., and Hopkinson, C.S. (1984)Above- and below-ground emergent macrophyte production and turnover in a coastal marsh ecosystem, Georgia. Limnol. Oceanogr., 29:1052-1065. Singh, A.P., and Butcher, J.A. (1991)Bacterial degradation of wood cell walls: A review of degradation patterns. J. Inst. Wood Sci., 12:143-157. Srebotnik, E., and Messner, K. (1990)Enzymatic attack of wood is limited by the inaccessibility of the substrate. In: Biotechnology in Pulp and Paper Manufacture. Applications and Fundamental In- vestigations. T.K. Kirk and H.-M. Chang, eds. Butterworth-Heinemann, Boston, pp. 111-119. Taylor, B.R., Prescott, C.E., Parsons, W.J.F., and Parkinson, D.(1991) Substrate control of litter decomposition in four Rocky Mountain coniferous forests. Can. J. Bot., 69:2242-2250. Torzilli, A.P., and Andrykovitch, G. (1986)Degradation of Spartina lignocellulose by individual and mixed cultures of salt-marshfungi. Can. J. Bot., 642211-2215. Yadav, J.S., and Reddy, C.A. (1993)Degradation of benzene, toluene, ethylbenzene, and xylenes (BTEX) by the lignin-degrading basidiomycete Phanerochoete chrysosporium. Appl. Environ. Microbiol., 59:756-762. Yu, H., Cai, Z.-Y., Huang, X.-Y., and Xu,H. (1990)Biological degradation and delignification of rice straw. In: Biotechnology in Pulp and Paper Manufacture. Applications and Fundamental Investigations. T.K. Kirk and H.-M. Chang, eds. Butterworth-Heinemann, Boston, pp. 59-68. Zhai, H.M., and Lee, Z.Z. (1989)Ultrastructure and topochemistry of delignification in alkaline pulping of wheat straw. J. Wood Chem. Tech., 9387-406.