APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Dec. 1983, p. 1409-14160099-2240/83/121409-08$02.00/0Copyright C 1983, American Society for Microbiology

Vol. 46, No. 6

Microbial Decomposition of Wood in Streams: Distribution ofMicroflora and Factors Affecting [14C]Lignocellulose

MineralizationtNICHOLAS G. AUMEN,' PETER J. BOTTOMLEY, 1•2* G. MILTON WARD,' AND STAN V. GREGORY4Departments of Microbiology,' Soil Science,' and Fisheries and Wildlife,' Oregon State University, Corvallis,

Oregon, 97331-3804, and Department of Biology, University of Alabama, University, Alabama, 354863

Received 11 August 1983/Accepted 20 September 1983

The distribution and lignocellulolytic activity of the microbial community wasdetermined on a large log of Douglas fir (Pseudotsuga menziesit) in a PacificNorthwest stream. Scanning electron microscopy, plate counts, and degradationof [ 14C]lignocelluloses prepared from Douglas fir and incubated with samples ofwood taken from the surface and within the log revealed that most of the microbialcolonization and lignocellulose-degrading activity occurred on the surface. La-beled lignocellulose and surface wood samples were incubated in vitro withnutrient supplements to determine potential limiting factors of C4Cflignocellulosedegradation. Incubations carried out in a nitrogenless mineral salts and traceelements solution were no more favorable to degradation than those carried out indistilled water alone. Incubations supplemented with either (NH4)2SO4 or organicnitrogen sources showed large increases in the rates of mineralization overincubations with mineral salts and trace elements alone, with the greatest effectbeing observed from an addition of (NH 4)2SO4. Subsequent incubations with(NH4)2SO4 , KNO3, and NH4NO3 revealed that KNO3 was the most favorable forlignin degradation, whereas all three supplements were equally favorable forcellulose degradation. Supplementation with glucose repressed both lignin andcellulose mineralization. The results reported in this study indicate that nitrogenlimitation of wood decomposition may exist in streams of the Pacific Northwest.The radiotracer technique was shown to be a sensitive and useful tool forassessing relative patterns of lignocellulose decay and microbial activity in wood,along with the importance of thoroughly characterizing the experimental systembefore its general acceptance.

The large biomass of old-growth coniferousforests of the Pacific Northwest can result insubstantial accumulations of logs on the forestfloor and in the streams which drain the sur-rounding watershed (17). Standing crops ofwood in Cascade Mountain stream channels canexceed 40 kg m-2 (21). The determination of therole large woody debris plays in ecosystems isessential, especially in view of current manage-ment strategies for logging operations whichmay significantly decrease the amounts of debrisavailable for habitat formation and biologicalprocessing (36; J. D. Hall and C. O. Baker,1975, A workshop on logging debris in streams,Oregon State University, Corvallis, and F. J.Swanson and G. W. Lienkaemper, 1978, U.S.Department of Agriculture Forest Service Gen-eral Technical Report PNW-69).

+ Oregon State University Agricultural Experiment StationTechnical Paper no. 6915. Riparian Contribution no. 12.

A major component of this woody debris islignocellulose, which represents more than one-half of the total carbon present in wood. Recent-ly developed radiotracer methodologies thatspecifically label the lignocellulose fraction ofplant tissue have led to ever increasing amountsof information on the microorganisms, biochem-ical processes, and environmental factors whichaffect the degradation of lignocellulose (9).

In recent years, much emphasis has beenplaced on studies involving specific microorga-nisms in pure culture (4-6, 11, 19, 22, 23, 30).Although this is of undeniable importance, suchinformation cannot be extrapolated easily tonatural systems. Research that uses radiotracermethods with environmental samples has beenlimited to soil and sediment systems (6-8, 14,18, 25). Large woody debris differs from soilsand sediments in that the substrate has a verysmall surface area-to-volume ratio, is structural-ly recalcitrant, and sometimes occurs in flowing-

1409

APPL. ENVIRON. MICROBIOL.1410 AUMEN ET AL.

water environments which are frequently low innutrients.

This paper describes the development and useof a radiotracer assay to determine the charac-teristics of lignocellulose degradation in wood.The assay was used to determine patterns ofmicrobial decomposition in large logs of Douglasfir (Pseudotsuga menziesii) which have beenlying for known periods of time in a streamchannel of an old-growth forest.

MATERIALS AND METHODSPreparation and characterization of [ 14C]lignocellu-

loses. Douglas fir lignocelluloses labeled specifically ineither the lignin or the cellulose fraction were preparedby the method of Crawford (9). Freshly cut ends ofDouglas fir branches, each ca. 1 m in length, wereimmersed in aqueous solutions of either 50 ILCi of 1.-[U_lec ]phenylalanine (10 mCi mmo1 -1 ) or 50 ILCi of D-[U-'4C]glucose (3 mCi mmor 1) to initiate the labelingof lignin and cellulose, respectively. Radiochemicalswere obtained from Amersham Corp., ArlingtonHeights, Ill. Just before uptake of the radioactivesolution was completed, additional water was added asneeded to keep the cut ends immersed, and -the stemswere allowed to metabolize the radiolabels underconstant illumination until wilting began to occur (ca. 1week). The cambial tissue was stripped from the mainstem and dried at 50°C for 2 weeks. The labeled tissuewas ground to pass a (no. 40) mesh screen andsubjected to a sequence of hot water, ethanol-ben-zene, and ethanol extractions to remove undesiredlabeled plant constituents, as previously described indetail by Crawford (9). The extractive-free tissue wasdried at 50°C and stored in a desiccator at roomtemperature.

The specific activities of the extractive-free [ 14C-lignin]- and [ HC-cellulose]lignocelluloses were deter-mined by combustion of 10-mg samples in a model 306Packard Tri-Carb oxidizer (Packard Instrument Co.,Inc., Rockville, Md.), and the resulting 14CO2 wastrapped and counted by standard liquid scintillationtechniques.

Distribution of HC within the labeled material wasdetermined by the modified Kiason procedure of Ef-fland (13). Triplicate 200-mg portions of each labeledsubstrate were digested for 1 h with 72% (wt/wt)H2SO4. The digest was diluted (1:28), heated at 120°Cand at a pressure of 103 kPa for 1 h, and then filteredthrough tared, frilled glass crucibles. The residue wasdried overnight at 105°C and weighed to determine theinitial lignin content of the woody tissue. The specificradioactivity was determined with the Packard oxidiz-er as described above. The Kiason filtrate was adjust-ed to pH 4.5 with CaCO3 and stored at 4°C forsubsequent carbohydrate determinations. A portion ofthe filtrate was used for determination of its radioac-tivity by pipetting 0.1-m1 samples onto 5.5-cm-diame-ter Whatman no. 1 filter papers which were combust-ed, and the 14CO2 was quantified as described above.

The distribution of radiolabel in wood sugars wasdetermined by TAPPI method T250 pm-75 (The Tech-nical Association of the Pulp and Paper Industry,Atlanta, Ga.), with the following modifications. Por-tions (100 ml) of the Klason filtrates were lyophilized

to concentrate the filtrate and then rehydrated with 5ml of distilled water. Incorporation of '°C into woodsugars was determined by descending paper chroma-tography with a butanol-pyridine-water (10:3:3, vol/vol) solvent system. Solutions of glucose, mannose,and xylose were used as standard markers. The sec-tions of the chromatograms corresponding to glucose,mannose, and xylose were removed and analyzed for' 4C activity by oxidation.

The total nitrogen remaining in the radiolabeledlignocelluloses was estimated by micro-Kjeldahl anal-ysis. Radioactivity associated with the protein fractionwas estimated by incubating 10-mg samples of thelabeled substrates with 6 U of Type XIV bacterialprotease (Sigma Chemical Co., St. Louis, Mo.) in 0.05M phosphate buffer, pH 7.5, at 37°C for 2 h. Sampleswere analyzed in quintuplicate, and changes in specificradioactivity of the residue were determined by sam-ple oxidation before and after protease incubation.

Study site and sample collection. Wood samples usedas inocula for [ HC]lignocellulose degradation experi-ments were obtained from Mack Creek, a third-orderstream in an old-growth section of the H. J. AndrewsExperimental Ecological Reserve, located at an eleva-tion of 830 m in the Cascade Mountain Range, Oreg.Surface scrapings were obtained with a carpentryplane, and cores were taken with a 12-mm-diameterincrement borer (Forestry Suppliers, Inc., Jackson,Miss.) from bark-free, stream-wetted portions of anold-growth Douglas fir log that fell into the stream in1977. Samples were placed in sterile Whirlpak bags,stored on ice, transported to the laboratory, andprocessed within 48 h. Wood cores were split in thelaboratory with a sterile knife and subsampled to avoidcontamination that may have been carried down fromthe surface by use of the increment borer.

Sample preparation. Wood samples were homoge-nized in sterile distilled water for 8 min at a setting of30 on a VirTis model 45 homogenizer (VirTis Co.,Inc., Gardiner, N.Y.). This duration of homogeniza-tion had been previously determined to yield thehighest numbers of microorganisms from wood sam-ples by use of standard microbiological plate counts ondiluted cryptic soy agar (data not shown). Experimen-tal treatments were prepared by placing 1-ml portionsof homogenate into 60-m1 glass serum bottles contain-ing 20 ml of an incubation medium and fitted with glasscapillary (2-mm-diameter) bubbler tubes and sleeve-type rubber stoppers. The treatments were bubbledeither continuously or once per day for 15 min withfiltered, humidified, CO2-free air. Incubations werecarried out at room temperature (21 ± 2°C). Outflowgas from the incubation bottles was passed through 8%(wt/vol) NaOH to absorb 14CO2 , radioactivity wasdetermined by acidification of the alkali with concen-trated H2SO4 , and the 14CO2 was trapped on filterpaper soaked with p-phenylethylamine (free base,Sigma). Filter papers were then placed in 15 ml ofliquid scintillation cocktail and their radioactivity wascounted on a Beckman model LS8000 liquid scintilla-tion counter (Beckman Instruments, Inc., Fullerton,Calif.). Counting efficiency was determined by using aquench series and the external standard technique,with data corrected for efficiency and background.The liquid scintillation cocktail consisted of Spectra-fluor (Amersham), methanol, and toluene (16:100:125,vol/vol). The removal of 14CO2 from the alkali traps

Substrate Sp act(dpm/mg) In Klason In Klason

residue filtrate Recovered After proteasedigestion

% Original radioactivity

ECOMPOSITION OF WOOD IN STREAMS 1411VOL. 46, 1983 MICROBIAL D

was shown to be complete by using known quantitiesof Na2'4CO3.

Spatial distribution and [ 14C]lignocellulose decompo-sition experiments. The spatial distribution of the mi-crobial community in the study log was determined byobtaining surface scrapings and core samples from adepth of 25 cm in July and September 1981. Platecounts were performed on samples of homogenate byspreading 0.1 ml of serial dilutions (3 x 10-2 to 3 x10-4 in distilled water) of wood homogenate on thesurface of one-tenth-strength tryptic soy agar plates.Incubation was carried out at 30°C for 5 to 7 days,under both aerobic and anaerobic conditions. Anaero-bic incubation conditions were achieved by incubatingplates inside GasPak anaerobic jars (BBL Microbiolo-gy Systems, Cockeysville, Md.) placed in a 30°Cincubator. Preliminary studies had revealed that dilut-ed tryptic soy agar yielded a higher recovery ofmicroorganisms from the homogenate than did full-strength media.

Wood homogenates from the surface scrapings andcore samples were also incubated in the presence of 10kdpm of either ["C-lignin]- or ["C-cellulose]lignocel-lulose for 27 days in distilled water, with the evolved"CO2 trapped and quantified at ca. 7-day intervals.Portions of the sample obtained before incubationwere prepared for scanning electron microscopy(SEM) by fixation with 0.33% (vol/vol) glutaraldehydesolution, dehydrated in a graded ethanol series, anddried in a model DCP-1 critical-point dryer (DentonVacuum, Inc., Cherry Hill, N.J.). Samples were thenmounted on stubs, coated with 10 nm of gold-palladi-um, and observed on an ETEC Autoscan scanningelectron microscope.

The effects of various amounts of added V 4C-ligninl-lignocellulose on ' 4CO2 evolution were determined byincubating 0.026-g (dry weight) portions of wood in-oculum (collected in October 1982) with 10, 20, 30, and40 kdpm of labeled substrate in distilled water.

The effects of various medium supplements onL.Ilignocellulose mineralization were examined by

incubating 0.013 and 0.025 g (dry weight) of woodinoculum (collected in February and April 1983, re-spectively) in media supplemented with organic andinorganic components. All incubation treatments con-tained either 20 kdpm of [ 14C-lignin]lignocellulose or10 kdpm of ["C-cellulosellignocellulose as the labeledsubstrate. The various incubation media (20 ml) usedin the experiments were distilled water alone andmineral salts solution alone and in combination withone of the following amendments (final concentrationsin grams per liter): (NH4)2SO4 , 3.0; yeast extract, 9.0;glucose, 1.8; KNO 3 , 2.3; or NH4NO3 , 0.9. The mineralsalts solution was composed of the following compo-nents (grams per liter of distilled water): CaCl 2 , 0.6;MgSO4 • 7H 20, 0.2; NaCI, 0.1; ferric citrate, 0.1;

K 2HPO4 • 3H 20, 0.5; KH 2PO4 , 0.188; and 10 ml of atrace elements solution containing (milligrams perliter) H 3B03 , 143; MnSO4 • 4H20, 102; ZnSO4 • 7H20,22; CuSO4 • 5H 2O, 8; CoCl 2 • 4H 20, 10; andNa2MoO4 • 2H 20, 5. Controls for all experiments con-sisted of distilled water and the appropriate labeledlignocellulose. Incubations were conducted in quintu-plicate for 3 to 4 weeks at room temperature, and the'4CO2 evolved was determined at weekly intervals.Results were plotted as accumulated kilodisintegra-tions per minute per gram (dry weight) of woodinoculum, and the standard error of the mean wasdetermined for values obtained at each sampling time.

RESULTS‘..]lignocellulose characterization. The spe-

cific activities and distribution of ' 4C within thelabeled lignocellulose are summarized in Table 1and correspond favorably to values reported inthe literature for Douglas fir [ 14C]lignocellu-loses, as do the results of the Klason analysis(11). Chromatographic analysis of the Klasonfiltrate of the [ 14C-cellulose]lignocellulose dem-onstrated that all of the radioactivity in thehydrolysate was found in the three sugars ana-lyzed (glucose, mannose, and xylose), with 82%residing in the glucose fraction. In contrast, only42% of the total radioactivity of the [14C]ligninKlason filtrate could be accounted for in thethree sugars, with 22% found in the glucosefraction (presumably from the acid hydrolysis ofcontaminating labeled cellulose). Similar pat-terns of label distribution within the Klasonfiltrate are reported by Maccubbin and Hodsonfor pine lignocelluloses (25) and by Crawfordand Crawford (10) for lignocelluloses from sev-eral different sources. The 58% of the radioactiv-ity in the [ 14C]lignin Klason filtrate that is unac-counted for by the carbohydrate analysis ispresumably in the form of either acid-solublelignin or amino acids liberated by acid hydroly-sis. In fact, it has been noted that a seriousdrawback of the Klason analysis is that consid-erable 72% H2SO4-soluble lignin is present inmany lignocelluloses (9, 11). Protease digestionof the labeled lignocelluloses used in this studyresulted in a 5% loss of radioactivity whencompared with undigested substrate. A value fortotal protein content of 3.5%, assuming thatprotein = N x 6.25, was obtained from micro-Kjeldahl analysis.

TABLE 1. Distribution of ' 4C within [14C]lignocellulose

ugnin]lignocellulose 1.211 76.5 27.6 104.1 94.5[14C-cellulose]lignocellulose 560 28.8 72.1 100.9 95.0

10

E33

010

7

YCCO

O

OIC)

AE

276 17Incubation time (days)

1412 AUMEN ET AL.

•

14 C- LIGNIN andCELLULOSEMINERALIZATION

FIG. 1. 1 14C]lignocellulose degradation in surfacescrapings and core samples (25-cm depth) from awetted log. Symbols: CI, surface [ 14C]cellulose: 0,core [ 14C]cellulose; *, surface [ HC]lignin; •, core,140,tilignin. Each point is the mean of two replicates.Bars, Standard errors of the means.

Spatial distribution and [ 14C]lignocellulose de-composition experiments. A series of samplestaken to investigate the spatial distribution of themicrobial community on decaying wood showedthat most of the colonization and activity oc-curred on the outer surface of the log. Platecounts revealed 1.44 x 107 CFU/g of wood fromthe aerobically incubated surface samples and2.22 x 106 CFU/g of wood from the anaerobical-ly incubated surface samples. Wood samplesobtained from a depth of 25 cm in the log yieldedno CFU under either aerobic or anaerobic condi-tions.

Evolution of 14CO2 from the 14C incubationexperiments demonstrated that the greatest lig-nocellulose mineralization activity was in thesurface sample incubated with [I4C]cellulose(Fig. 1). Decomposition of the [ I4C]cellulose inthe surface samples was more than four timesthat of the interior sample, with a similar rela-tionship observed for the [ 14C]lignin treatments.The SEM study also revealed a colonizationpattern consistent with the results reportedabove (Fig. 2). One can see evidence of microbi-al colonization on the exposed side of the sur-face scraping, including individual bacterialcells, actinomycete-like filaments, and possiblyfungal hyphae (Fig. 2A). There is very littlemicrobial colonization apparent on the under-side of surface scrapings or on the sample ob-

APPL. ENVIRON. M1CROBIOL.

tained from a depth of 25 cm (Fig. 2B and C).Preliminary experimentation indicated that in

the case of the [ 14C-lignin]lignocellulose, cautionmust be exercised in the selection of the appro-

FIG. 2. SEM of wood samples from the study log.(A) Exposed side of a surface scraping, (B) undersideof a surface scraping, and (C) core sample from a 25-cm depth. Bars, 10 Rm.

7Incubation

14time (days)

21

Ea.

12

Cr 16 MINERALIZATION

12

8

a

E

14C- LIGNIN

MICROBIAL DECOMPOSITION OF WOOD IN STREAMS 1413VOL. 46, 1983

FIG. 3. Effects of various amounts of r 4C-ligninl-lignocellulose on mineralization rates with a fixedquantity of wood inoculum. Symbols: 0, 10-kdpmaddition; *, 20-kdpm addition; ^, 30-kdpm addition;*, 40-kdpm addition. Each point is the mean of fivereplicates. Bars, Standard errors of the means.

priate amounts of radiolabel added to incubationtreatments. Generally accepted amounts (10 to30 kdpm per treatment) may result in suboptimalmineralization rates, depending on the type andamount of inoculum added. Results of an experi-ment illustrating the effect of various amounts oflabeled substrate with a fixed quantity of inocu-lum are presented in Fig. 3. Increasing rates of[ 14C-lignin]lignocellulose mineralization wereevident with increasing "C substrate additions.Significant differences were observed betweenthe 10-kdpm addition and the other three treat-ments.

Supplementation of the incubation treatmentswith either inorganic or organic nitrogen sourcesresulted in [ 14C]lignocelluloses labeled in thecellulose fraction being mineralized to '4CO2 atrates three- to ninefold faster than those of the[14C]lignin fraction (Fig. 4). Organic and inor-ganic nitrogen supplements affected the rates ofmineralization of the [ 14C]lignin more than[ 14C]cellulose mineralization rates when com-pared with samples incubated in either distilledwater or nitrogenless mineral salts solution. Glu-cose had a substantial inhibitory effect on[ 14C]cellulose mineralization and less of an ef-fect on [ 14C]lignin mineralization. Incubationwith the nitrogenless mineral salts solution wasonly slightly more favorable than with distilledwater for [ 14C]lignin mineralization and resultedin no difference for the [ 14C]cellulose treatment.The greatest stimulation of degradation was dueto the addition of (NH4)2SO4 , which increasedthe [ 14C]lignin and [ 14C]cellulose mineralizationrates by factors of 12 and 5, respectively, abovethat of the distilled water control.

Incubation of a different set of wood sampleswith [ 14C]lignocelluloses in the presence of three

different forms of inorganic nitrogen,(NH4)2SO4 , KNO3 , and NH4NO3 , again demon-strated substantial increases in the rates of min-eralization (Fig. 5). [ 14C]cellulose was mineral-ized to 14CO2 at rates four to six times those of[ 14C]lignin mineralization. Here, in contrast tothe previous experiment, rates of [14C]cellulosemineralization were enhanced more than thelignin rates by medium supplements when com-pared with incubation in mineral salts solutionalone. Of particular interest is the observationthat the KNO3 addition was the most favorablefor [ 14C]lignin decomposition, whereas all threeinorganic nitrogen supplements were equallyfavorable to [ 14C]cellulose decomposition. Addi-tion of NH4NO3 resulted in [ 14C]lignin mineral-ization rates similar to those of the (NH4)2SO4treatment.

DISCUSSIONDegradation of [ RC]lignocelluloses has been

shown by this study to be a sensitive and useful

3 48 14

C - CELLULOSEMINERALIZATION

S 36

"67a 24

FIG. 4. Effects of medium supplements on [ 14C]lig-nocellulose mineralization. Symbols: *, distilled wa-ter alone; •, mineral salts solution; *. salts plusglucose; 0, salts plus yeast extract; *. salts plus(NH 4),SO4 . Each point is the mean of five replicates.Bars, Standard errors of the means.

40

E30 I4C— CELLULOSE

8F.

ic 20

0

6 ioa.

o0••0

MINERALIZATION

c.) MINERALIZATION

: 80

I4C-UGNIN

1

14 217

APPL. E NVIRON. M ICROBIOL.1414 AUMEN ET AL.

Incubation time (days)FIG. 5. Effects of various inorganic nitrogen spe-

cies on ["C]lignocellulose mineralization. Symbols:, mineral salts solution; *, salts plus (NH 4)2SO4 ; 0,

salts plus NH4NO3 ; A, salts plus KNO 3 . Each point isthe mean of five replicates. Bars, Standard errors ofthe means.

tool for assessing relative patterns of lignocellu-lose decay and microbial activity in wood. Ap-plication of this technique should prove particu-larly attractive to those interested in studyingwood decay when it is compared with conven-tional weight loss studies. Measurements ofwood decomposition by substrate weight lossmay require many months of substrate incuba-tion and provide no information on decay ratesof specific chemical components.

Before the radiotracer technique is adoptedfor use in a particular experimental system,however, it is essential that a thorough chemicalcharacterization of the ["C]lignocellulose beobtained. For example, degradation of smallamounts of contaminating

degradation protein

residing in the lignin label could result in nonlig-nin-derived "C being evolved, leading to anoverestimate of lignin-degrading capability. Thequantities of "C-labeled protein contained in thelignin label used here were low ( ....55%) and werewell within ranges reported by other investiga-

tors using Douglas fir ["C]lignocelluloses (5,11). The results of Klason and carbohydrateanalyses of the labeled material and the ratiosbetween lignin and cellulose mineralization rateswere also similar to other values reported in theliterature and indicate that evolution of "CO2from the treatments represents lignocellulosemineralization (5, 7, 10, 11, 25).

Rates of in vivo ("Cllignin decompositionwere maximized so that the effects of physicaland chemical manipulations could be morereadily measured. This was first accomplishedby varying the amount of "C label with a fixedinoculum size (Fig. 3). Even though the 30- and40-kdpm additions were slightly more favorablethan the 20-kdpm addition by week 4 of incuba-tion, the latter amount was selected to conservelabeled substrate and to restrict incubations to amaximum of 3 weeks. The appropriate level of"C substrate addition should be determined forall applications of the described technique and,surprisingly, has only recently been consideredin the literature (3).

Degradation of ["C]lignocelluloses, platecounts, and SEM all demonstrated that microbi-al colonization on the log was mainly a surfacephenomenon. This supports suggestions in theliterature that microbial activity is generallyrestricted to the surface of decomposing wood inaquatic environments (2, 12, 32, 36). The ob-served pattern of surface-related microbial ac-tivity probably reflects the lack of gallery-form-ing insect activity in aquatic wood, a factorconsidered a potentially important microbial dis-tribution mechanism in logs decomposing interrestrial environments (2, 12). Waterloggingand the absence of tunneling insects may limitthe access of oxygen into the log, which is arequirement for significant microbial breakdownof natural lignin (9, 38). Oxygen limitation wouldalso restrict the growth of the hyphae of aerobiclignin-degrading fungi into the inner parts of thewood (24). Restriction of microbial activity tothe surface of aquatic wood results in extremelyslow rates of decomposition given the smallsurface area-to-volume ratio of large logs andthe recalcitrant nature of lignocellulose.

Though significant ["C]lignocellulose decayoccurred during incubation in distilled water,rates of breakdown were greatly enhanced bysupplementation of media with inorganic nitro-gen (Fig. 4). The observation that NO3 - Nenrichment enhanced ["C]lignin degradationmore so than did (NH 4)2SO4 or NH4NO3 isparticularly worthy of note. Preliminary experi-mental evidence obtained by us suggests thatnitrate ammonification may be occurring in theKNO3-enriched incubation treatment (data notpublished). This observation warrants furtherstudy to determine whether it is related to the

VOL. 46, 1983 MICROBIAL DECOMPOSITION OF WOOD IN STREAMS 1415

favorable effect of NO3- N addition on lignindegradation. When NO3- N is supplied in com-bination with NH4 + N, however, [ 14

mineralization rates are similar to those of the(NH4)2SO4 treatment alone, suggesting that aclassical repression of NO3 - metabolism byNH4 + may be in effect (35).

In contrast to the stimulatory effects of nitro-gen supplementation, the repression resultingfrom glucose addition suggests that the microbi-al :community present in decaying wood willutilize more favorable carbon and energysources when they are available. To knowwhether or not this repressive effect is directlyori the ligninolytic community per se or indirect-ly-on other members of the community requiresfurther study. Repression of lignocellulose min-eralization by simple sugars has been notedbefore, in both environmental samples and pureculture work (1, 34).

There are few published results of studiesconcerning the effect of nutrient amendments on[ 14C]lignocellulose decomposition in naturalsamples, and none exist for wood substrates tothe best of our knowledge. Positive correlationshave been observed between synthetic [ 14C]lig-nin degradation rates and NO 3- N concentra-tions in sediments, and additions of nitrogen toarctic lake sediments enhanced [14C]cellulosemineralization and had no effect on [14C]lignindecay (14, 18). Pure culture work on the physiol-ogy of white-rot fungi and lignin-degrading ac-tinomycetes also show contrasting effects ofnitrogen amendments (4, 22, 24, 30). In nonra-dioactively labeled decomposition studies, how-ever, several investigators report that added Ncan stimulate wood decomposition by fungi (15,20, 33).

The enhancement of lignocellulose degrada-tion by mineral nitrogen supplementation report-ed here suggests that nitrogen limitation mayexist in decomposing wood in Pacific Northweststreams. Nitrogen concentrations in wood areusually low, with faster decay rates correlatingwith higher nitrogen content of the woody tissue(28). Even in the later stages of decompositionwhen nitrogen content has been shown to in-crease, it may not be in a form available for useby microorganisms (29, 31). These observationsindicate that the microbiota on decaying woodshould respond favorably to an external sourceof N. Streamwater sources of nitrogen are low inlotic ecosystems of the Pacific Northwest wherenitrogen limitation has been observed (37). Thiscould be responsible for the extremely slowbiological processing rates of wood in thesenatural environments. Obviously, further stud-ies in situ will need to be carried out.

Other nutritionally important mineral ele-ments do not appear to be limiting to the degra-

dation process, as evidenced by the lack ofsubstantial increase in 14C mineralization ratesin the presence of a nitrogenless complete min-eral salts and trace elements source (Fig. 4).Correlative studies in other stream ecosystems,however, have suggested that decay rates ofwood may respond to increased phosphorusconcentrations in streamwater (27) and to theinitial quality of the substrate (16, 26).

Results from this study have demonstrated theapplicability of the [ 14C]lignocellulose techniquein assessing the activity of lignocellulolytic mi-crobial communities in decaying wood. Nowthat optimal conditions have been establishedfor laboratory mineralization studies, this tech-nique can be used in conjuction with othermethodologies to further characterize the micro-bial communities involved in wood decay in thestream environment and the physicochemicaland nutritional factors that affect these rates ofactivity.

ACKNOWLEDGMENTS

We thank R. Kepler for preparation of the figures and C.Dahm for his helpful suggestions and criticisms. We areindebted to D. L. Crawford, University of Idaho, Moscow,who kindly supplied I HCIlignocelluloses for preliminary ex-perimentation and provided much useful information on theradioisotopic technique. We acknowledge the Department ofBiology at the University of Alabama for use of the SEMfacilities.

This work was supported in part by National ScienceFoundation grants DEB80-04652. DEB80-22634, and DEB81-12455 and the Oregon Agricultural Experiment Station.

LITERATURE CITEDAnder, P., and K.-E. Eriksson. 1975. Influence of carbo-hydrates on lignin degradation by the white-rot fungusSporotrichum pulverulentum. Sven. Papperstidn. 78:642-652.Anderson, N. H., J. R. SedeU, L. M. Roberts, and F. J.Triska. 1978. The role of aquatic invertebrates in process-ing of wood debris in coniferous forest streams. Am. Midi.Nat. 100:164-82.Baker, K. H. 1983. Effects of selected assay parameterson measurement of lignocellulose mineralization with aradiolabeled substrate. Appl. Environ. Microbiol.45:1129-1131.Barder, M. J., and D. L. Crawford. 1981. Effects ofcarbon and nitrogen supplementation on lignin and cellu-lose decomposition by a Streptomyces. Can. J. Microbiol.27:859-863.Crawford, D. L. 1978. Lignocellulose decomposition byselected Streptomyces strains. Appl. Environ. Microbiol.35:1041-1045.Crawford, D. L., and R. L. Crawford. 1976. Microbialdegradation of lignocellulose: the lignin component. Appl.Environ. Microbiol. 31:714-717.Crawford, D. L., R. L. Crawford, and A. L. Pometto III.1977. Preparation of specifically labeled "C-(lignin)- and"C-(cellulose)-lignocelluloses and their decomposition bythe microflora of soil. Appl. Environ. Microbiol. 33:1247-1251.Crawford, D. L., S. Floyd, A. L. Pometto III, and R. L.Crawford. 1977. Degradation of natural and Kraft ligninsby the microflora of soil and water. Can. J. Microbiol.23:434-440.

9. Crawford, R. L. 1981. Lignin biodegradation and transfor-mation. John Wiley & Sons. Inc.. New York.

1416 AUMEN ET AL. APPL. ENVIRON. MICROBIOL.

Crawford, R. L., and D. L. Crawford. 1978. Radioactivemethods for the study of lignin biodegradation. Dev. Ind.Microbiol. 19:35-49.Crawford, R. L., L. E. Robinson, and A. M. Cbeh. 1980.I T-labeled Iignins as substrates for the study of ligninbiodegradation and transformation, p. 61-76. In T. K.Kirk, T. Higuchi, and H.-M. Chang (ed.).. Lignin biodeg-radation: microbiology, chemistry, and potential applica-tions, vol. 1. CRC Press. Boca Raton, Fla-Dudley, T., and N. H. Anderson. 1982. A 'survey ofinvertebrates associated with wood debris in aquatichabitats. Melanderia 39:1-21.Mind, M. J. 1977. Modified procedure to determineacid-insoluble lignin in wood and pulp. Tappi 60:143-144.Federle, T. W., and J. R. Vestal. 1980. Lignocellulosemineralization by arctic lake sediments in response tonutrient manipulation. Appl. Environ. Microbiol. 40:32-39.Findlay, W. P. K. 1934. Studies on the physiology of thewood-destroying fungi. I. The effect of nitrogen contentupon the rate of decay of timber. Ann. Bot. (London)48:109-117.Fogel, R., and K. Cromack, Jr. 1977. Effect of habitat andsubstrate quality on Douglas fir litter decomposition inwestern Oregon. Can. J. Bot. 55:1632-1640.Franklin, J. F., and R. W. Waring. 1979. Distinctivefeatures of the Northwestern coniferous forest: develop-ment. structure and function, p. 59-86. In R. H. Waring(ed.). Forests: fresh perspectives from ecosystem analy-sis. Oregon State University Press, Corvallis.Hackett, W. F., W. J. Connors, T. K. Kirk, and J. G.Zeikus. 1977. Microbial decomposition of synthetic IT-labeled lignins in nature: lignin biodegradation in a varietyof natural materials. Appl. Environ. Microbiol. 33:43-51.Haider, K., and J. Trojanowsld. 1975. Decomposition ofspecifically IT-labeled phenols and dehydropolymers ofconiferyl alcohol as models for lignin degradation by softand white rot fungi. Arch. Microbiol. 105:33-41.Hungate, R. E. 1940. Nitrogen content of sound anddecayed coniferous woods and its relation to loss inweight during decay. Bot. Gaz. (Chicago) 102:382-392.Keller, E. A., and F. J. Swanson. 1979. Effects of largeorganic material on channel form and fluvial processes.Earth Surf. Processes 4:361-380.Keyser, P., T. K. Kirk, and J. G. Zeikus. 1978. Ligninoly-tic enzyme system of Phanerochaete chrysosporium: syn-thesized in the absence of lignin in response to nitrogenstarvation. J. Bacteriol. 135:790-797.Kirk, T. K., W. J. Connors, R. D. Bleam, W. F. Hackett,and J. G. Zeikus. 1975. Preparation and microbial decom-position of synthetic VTIlignins. Proc. Natl. Acad. Sci.U.S.A. 72:2515-2519.

Kirk, T. K., E. Schultz, W. J. Connors, L. F. Lorenz, andJ. G. Zelkus. 1978. Influence of culture parameters onlignin metabolism by Phanerochaete chrysosporium.Arch. Microbiol. 117:277-285.Maccubbin, A. E., and R. E. Hodson. 1980. Mineraliza-tion of detrital lignocelluloses by salt marsh sedimentmicroflora. Appl. Environ. Microbiol. 40:735-740.Melillo, J. M., J. D. Aber, and J. F. Muratore. 1982.Nitrogen and lignin control of hardwood leaf litter decom-position dynamics. Ecology 63:621-626.Melillo, J. M., R. J. Naimoli, J. D. Aber, and K. N. Eshle-man. 1983. The influence of substrate quality and streamsize on wood decomposition dynamics. Oecologia (Berlin).58:281-285.Merrill, W., and E. B. Cowling. 1966. Role of nitrogen inwood deterioration. IV. Relationship of natural variationin nitrogen content of wood to its susceptibility to decay.Phytopathology 56:1324-1325.Merrill, W., and E. B. Cowling. 1966. Role of nitrogen inwood deterioration: amounts and distribution of nitrogenin tree stems. Can. J. Bot. 44:1555-1579.Reid, I. D. 1983. Effects of nitrogen sources on celluloseand synthetic lignin degradation by Phanerochaete chry-sosporium. Appl. Environ. Microbiol. 45:838-842.Rice, D. L. 1982. The detritus nitrogen problem: newobservations and perspectives from organic geochemis-try. Mar. Ecol. Prog. Ser. 9:153-162.Savory, J. G. 1954. Damage to wood caused by microor-ganisms. J. Appl. Bacteriol. 17:213-218.Schmitz, H., and F. Kaufert. 1936. The effect of certainnitrogenous compounds on the rate of decay of wood.Am. J. Bot. 23:635-638.Skowronika, T. 1977. The influence of some carbohy-drates on the cellulolytic activity of actinomycetes of thegenus Streptomyces. Pol. J. Soil Sci. 10:41-49.Stouthamer, A. H. 1976. Biochemistry and genetics ofnitrate reductase in bacteria. Adv. Microb. Physiol.14:315-375.Trbka, F. J., and K. Cromack, Jr. 1979. The role of wooddebris in forests and streams. p. 171-190. In R. H. Waring(ed.), Forests: fresh 'perspectives from ecosystem analy-sis. Oregon State University Press, Corvallis.Triska, F. J., J. R. Sedell, and S. V. Gregory. 1982. Conif-erous forest streams, p. 292-332. In R. L. Edmonds (ed.),Analysis of coniferous forest ecosystems in the westernUnited States. Hutchinson Ross Publishing Co., Strouds-burg, Pa.

38. Zeikus, J. G. 1980. Fate of lignin and related aromaticsubstrates in anaerobic environments, p. 101-109. InT. K. Kirk, T. Higuchi, and H.-M. Chang (ed.), Ligninbiodegradation: microbiology, chemistry, and potentialapplications, vol. 1. CRC Press, Boca Raton, Fla.