Available online at www.sciencedirect.com

Bioresource Technology 99 (2008) 6556–6564

Microbial pretreatment of cotton stalks by solid state cultivationof Phanerochaete chrysosporium

Jian Shi, Mari S. Chinn *, Ratna R. Sharma-Shivappa

Department of Biological and Agricultural Engineering, Campus Box 7625, North Carolina State University, Raleigh, NC 27695-7625, USA

Received 7 May 2007; received in revised form 30 October 2007; accepted 3 November 2007Available online 31 January 2008

Abstract

White rot fungi degrade lignin and have biotechnological applications in conversion of lignocellulose to valuable products.Pretreatment is an important processing step to increase the accessibility of cellulosic material in plant biomass, impacting efficiencyof subsequent hydrolysis and fermentation. This study investigated microbial pretreatment of cotton stalks by solid state cultivation(SSC) using Phanerochaete chrysosporium to facilitate the conversion into ethanol. The effects of substrate moisture content (M.C.;65%, 75% and 80% wet-basis), inorganic salt concentration (no salts, modified salts without Mn2+, modified salts with Mn2+) and culturetime (0–14 days) on lignin degradation (LD), solids recovery (SR) and availability of carbohydrates (AOC) were examined. Moisturecontent significantly affected lignin degradation, with 75% and 80% M.C. degrading approximately 6% more lignin than 65% M.C. after14 days. Within the same moisture content, treatments supplemented with salts were not statistically different than those without salts forLD and AOC. Within the 14 day pretreatment, additional time resulted in greater lignin degradation, but indicated a decrease in SR andAOC. Considering cost, solid state cultivation at 75% M.C. without salts was the most preferable pretreatment resulting in 27.6% lignindegradation, 71.1% solids recovery and 41.6% availability of carbohydrates over a period of 14 days. Microbial pretreatment by solidstate cultivation has the potential to be a low cost, environmentally friendly alternative to chemical approaches. Moisture relationshipswill be significant to the design of an effective microbial pretreatment process using SSC technology.� 2007 Elsevier Ltd. All rights reserved.

Keywords: Cotton stalk; Pretreatment; Phanerochaete chrysosporium; Bioethanol; Lignin

1. Introduction

Currently, corn is the primary raw material for ethanolproduction in the United States (NCGA, 2005). However,lignocellulosic biomass has the potential to provide a moreeconomical feedstock as a result of its widespread availabil-ity, sustainable production and low starting value (ORNL,2005). Conversion of lignocellulosics to ethanol employsthree major steps including (1) pretreatment to breakdownthe lignin and open the crystalline structure of cellulose, (2)hydrolysis with a combination of enzymes to reduce cellu-lose to glucose and (3) microbial fermentation of glucose to

0960-8524/$ - see front matter � 2007 Elsevier Ltd. All rights reserved.

doi:10.1016/j.biortech.2007.11.069

* Corresponding author. Tel.: +1 919 515 6744; fax: +1 919 515 7760.E-mail address: mari_chinn@ncsu.edu (M.S. Chinn).

ethanol (Sun and Cheng, 2002). Although challenges existin the optimization of all the steps, pretreatment remainsto be one of the main barriers preventing commercial suc-cess and makes up one third of the total production costsduring ethanol production (NREL, 2000).

Existing pretreatment methods have largely been devel-oped on the basis of physicochemical technologies such asmicrowave, ionizing radiation, steam explosion, diluteacid, alkali, and oxidation or varied combinations (Moiseret al., 2005). However, typical physical and chemical pre-treatments require high-energy (steam or electricity) and/or corrosion-resistant, high-pressure reactors, whichincrease the need for specialty equipment and cost of pre-treatment. Furthermore, chemical pretreatments can bedetrimental to subsequent enzymatic hydrolysis and micro-bial fermentation apart from producing acidic or alkaline

J. Shi et al. / Bioresource Technology 99 (2008) 6556–6564 6557

waste water which needs pre-disposal treatment to ensureenvironmental safety (Keller et al., 2003).

Microbial pretreatment employs microorganisms andtheir enzyme systems to breakdown lignin present in ligno-cellulosic biomass. This environmentally friendly approachhas recently received increased attention (Hadar et al.,1993; Camarero et al., 1994; Sawada et al., 1995; Kelleret al., 2003) and has potential advantages over the prevail-ing physicochemical pretreatment technologies due toreduced energy and material costs, simplified processesand equipment, and use of biologically based catalysts.Several basidiomycetes such as Phanerochaete chrysospori-

um, Ceriporiopsis subvermispora, Phlebia subserialis andPleurotus ostreatus are capable of efficiently metabolizinglignin in a variety of lignocellulosic materials (Hatakka,1983; Sawada et al., 1995; Keller et al., 2003; Taniguchiet al., 2005). P. chrysosporium, is a white rot fungus thatproduces unique extracellular oxidative enzymes thatdegrade lignin (Jager et al., 1985). The physiology of P.

chrysosporium has been widely investigated because of itsnon-specific nature and exceptional oxidation potentialfor lignin degradation. The degradation efficiency and min-imal use of cellulosic polymers relative to other white rotfungi make P. chrysosporium a suitable candidate for pre-treatment and lignin degradation (Blanchette, 1991). How-ever, the efficiency of biodegradation varies withcultivation methods and environmental conditions (Reddyand D’Souza, 1994).

Solid state cultivation (SSC) is defined as the growth ofmicroorganisms on solid materials in the absence offree liquid (Pandey, 1992) and the technology has beeninvestigated for the development of bioprocesses, such asbioremediation and biodegradation of hazardous com-pounds, biological detoxification of agro-industrial resi-dues, biotransformation of crops and crop-residues fornutritional enrichment, biopulping, and production ofvalue-added products (Pandey et al., 2000). Compared toliquid cultivation, where microorganisms are submergedin free liquid, SSC possesses advantages such as simpleroperation, lower raw material and input costs and reducedenvironmental impact from waste water. Moreover, solidstate cultivation provides environmental conditions underwhich white rot fungi grow in nature, thus making it idealfor P. chrysosporium growth (Datta et al., 1991). Consider-ing these merits, applying SSC technology to pretreat lig-nocellulosic biomass may lead to better efficiency andreduce pretreatment costs.

Efforts have been made to apply solid state cultivationto produce ligninase or degrade lignocellulosic biomass(Kerem et al., 1992; Tengerdy and Szakacs, 2003; Coutoand Sanroman, 2005; Ganesh et al., 2006). Pilot scale fun-gal pretreatment facilities have been developed and testedfor biopulping of wood chips (Akhtar et al., 1998; Hatakkaet al., 2003). However, very limited resources report theapplication of SSC of P. chrysosporium for pretreatmentof lignocellulosic materials, especially to meet the needfor generation of a sugar platform for ethanol production

(Lee, 1997). The feasibility of this environmentally benignpretreatment process is still questioned, mainly due to theextremely long treatment time as well as the difficulty inselectively degrading lignin (Hatakka, 1983). Researchhas shown that delignification ability of P. chrysosporium

could be improved by optimizing nutrient supplementsand cultivation methods to enhance ligninolytic enzymeproduction (Reddy and D’Souza, 1994). Moisture content,substrate characteristics, nutrient supplements, aerationand culture duration are important variables for perfor-mance of fungi in SSC, impacting growth and metaboliteformation (Krishna, 2005; Asgher et al., 2006). However,these processing parameters have not been fully explored.Aeration and oxygen concentration are highly importantfor lignin degradation by fungal strains, especially in liquidcultivation systems (Leisola et al., 1983; Miura et al., 2004).Kerem et al. (1992) conducted a comparative study of P.

chrysosporium and P. ostreatus on their lignin degradingcapabilities in SSC of cotton stalks (65% moisture, up to30 days), examining the influence of oxygen on perfor-mance. Although the partial pressure of oxygen was nota limiting factor for their system, the effects of other keySSC parameters on degradation of lignocellulosic sub-strates remain undefined. Identifying the most favorablecultivation conditions for microbial pretreatment in SSCwill help establish the processing needs for effective applica-tion of this technology.

This study examines the microbial pretreatment of cot-ton stalks by solid state cultivation using P. chrysosporium.The effects of initial substrate moisture content (65%, 75%and 80% wet-basis), inorganic salt concentration (no salts,modified salts without Mn2+, modified salts with Mn2+)and culture time (0–14 days) on lignin degradation, solidsrecovery and availability of carbohydrates were investi-gated to evaluate process effectiveness.

2. Methods

2.1. Strain and inoculation

The fungal strain, P. chrysosporium (ATCC 24725), wasobtained from the Forest Products Laboratory of USDAForest Service (Madison, WI) and maintained as a frozenculture (�80 �C) in 30% glycerol. Propagation of the organ-ism for SSC was performed on Potato Dextrose Agar (PDA)plates for 2 days at 39 �C (Kirk et al., 1978). Spore suspen-sions were prepared by washing the agar surface with10 ml of sodium acetate buffer (50 mM, pH 4.5). Sporecounts were determined with a hemacytometer (Hausser Sci-entific, Horsham, PA) and the final spore inoculum had aconcentration of 5 � 106 spores/ml. The SSC cultures con-tained 5 � 106 spores/g air dry cotton stalk substrate.

2.2. Biomass preparation

Cotton stalks (shredded and baled) were harvested inearly October 2003 from Cunningham Research Station

6558 J. Shi et al. / Bioresource Technology 99 (2008) 6556–6564

(Kinston, NC). The stalks with initial moisture content(M.C.) of 7% were ground to 1 mm particle size forcomposition analysis. The feedstock for pretreatment wasground to 3 mm by a Thomas Wiley Laboratory Mill(Model No. 4, Thomas Scientific, Swedesboro, NJ) andstored in air tight containers at room temperature untiluse for pretreatment.

2.3. Experimental design and statistical analysis

All treatment combinations were completed in tripli-cate. The main and interaction effects of moisture con-tent, salt concentration and time factors (3 � 3 � 6factorial design) on lignin degradation, solids recoveryand carbohydrates were evaluated using PROC GLMin SAS 9.1 software (SAS� Inc., Cary, NC) for analysisof variance and significance tests. A 95% confidence levelwas used for statistical analysis and Tukey simultaneoustests were performed to assess statistical differencesbetween treatments.

2.4. Solid state cultivation pretreatment

Three inorganic salt formulations (no salts (SD), modi-fied NREL salts (SDS), modified NREL salts plus Mn2+

(SDM)) and three initial moisture contents (65%, 75%and 80% wet-basis) were examined. Modified NREL Salts(Keller et al., 2003) included (per gram air dry cottonstalks) 4 mg NaNO3, 10 mg KC1, 14 mg MgSO4 � 7H2O,0.14 mg FeSO4 � 7H2O, 40 mg KH2PO4, and 0.02 mg thia-mine in acetate buffer (20 mM, pH 4.5). Modified NRELsalts were supplemented with 18 mg MnSO4 � H2O to pre-pare the modified NREL salts + Mn2+ formulation(Brown et al., 1990). Treatments without salt addition weresupplemented with 20 mM acetate buffer (pH 4.5) to createthe appropriate substrate moisture content factor level.Microbial control flasks, without fungal inoculation, wereset up along with the pretreatment studies to quantify theeffects of substrate preparation, soaking and agitation.

Pretreatment cultures were grown in 250 ml Erlenmeyerflasks capped by silicone stoppers with inlet and exit linesconnected to 0.2 lm filters (Acro� 37 TF, Pall Co., NY).Flasks containing 3 g cotton stalk samples were autoclavedfor 20 min (121 �C, 124 kPa) then cooled prior to additionof salt formulations (autoclaved separately) and 3 ml sporesuspension (5 � 106 spores/g air dry cotton stalk). Initialsubstrate moisture content was adjusted to the respectivefactor level through the addition of salt solutions (4.9,8.1 and 10.9 ml for 65%, 75% and 80% moisture content,respectively) and inoculum spore suspension. Pretreat-ments were carried out in an air convection incubator at39 �C and destructively sampled at predetermined timeintervals (0, 4, 6, 8, 10, and 14 days). Flasks were flushedwith pure oxygen (125 ml/min) for 10 min every 3 daysstarting from day 0 (Kirk et al., 1978; Tien and Kirk,1988). Between flushing events, the flasks were closed byclamping off inlet and exhaust tubing lines.

2.5. Analysis methods

The total solids, acid soluble lignin, and acid insolublelignin content of untreated cotton stalks were determinedby NREL Laboratory Analytical Procedures, LAP 001,LAP 003, LAP 004 (Ehrman, 1994,1996; Templeton andEhrman, 1994). Ash content, extractives, and holocellulose(combination of hemicellulose and cellulose), were deter-mined for the untreated stalks by the gravimetric methodsdeveloped by Han and Rowell (1997).

At each time interval, enzymes from pretreatment cul-tures were harvested by soaking in sodium acetate buffer(30 ml, pH 4.5) and incubated at 39 �C (30 min). The pre-treated substrate suspension was then filtered throughpre-ignited and weighed crucibles (30 M Kimax, KimbleGlass Inc, Vineland, NJ). Supernatants were stored at�80 �C for analysis of free cellulase activities. Solid frac-tions were dried in a convection oven (105 �C, 12 h) andthe dry solids were weighed and tested for lignin contentaccording to LAP 003 and LAP 004. The treatment filtratefrom lignin analysis was used for quantification of xylose,arabinose, glucose and galactose sugars by HPLC (DionexCorp., Sunnyvale, CA) equipped with a CarboPacTM PA10(4 � 250 mm) anion exchange column, and a pulsed amper-ometric detector (Silverstein et al., 2007). All samples wereneutralized to pH 5–6, centrifuged (5000g, 10 min at 20 �C)and filtered through 0.45 lm Milipore filters before analy-sis (Silverstein et al., 2007).

Lignin degradation is defined as the percentage of totallignin reduced after pretreatment was calculated using Eq.(1).

LD ¼ ð1� w � ðaþ bÞw0 � ða0 þ b0Þ

Þ � 100% ð1Þ

where w is the dry weight of the treated sample (g); w0 is theinitial dry weight of the untreated sample (g); a, b are the %of acid-soluble and acid-insoluble lignin of the treated sam-ple and a0, b0 are the % of acid-soluble and acid-insolublelignin of the untreated sample.

Solids recovery was calculated as a percentage of thetotal solids recovered after pretreatment based on the ini-tial sample (dry weight). Availability of carbohydrateswas estimated as a percentage (w/w) of the total glucanand xylan in the solids recovered. Arabinan and galactanwere ignored due to their very low presence.

Cellulase activity (IU/ml liquid in SSC) was tested onWhatman Grade No. 1 filter paper using a modifiedDNS assay (Adney and Baker, 1996; Chinn, 2003). Oneinternational unit (IU) of cellulase activity is defined asthe amount of enzyme that releases 1 lmol of glucose perminute at pH 4.8 and 55 �C.

3. Results

The analysis of variance (ANOVA) table for main andinteraction effects of moisture content (M.C.), salt formula-tion (salt) and time factors (Time) on lignin degradation

J. Shi et al. / Bioresource Technology 99 (2008) 6556–6564 6559

(LD), solids recovery (SR) and availability of carbohy-drates (AOC) is shown in Table 1. The main effect of mois-ture content, salt formulation and time were statisticallysignificant for lignin degradation and solid recovery(P < 0.05). Similarly, moisture content and time were sig-nificant main effects for availability of carbohydrates, yetstatistical differences for salt formulation were not signifi-cant. The interaction of all factors, except salt-time, wassignificant for solid recovery. The interaction betweenM.C. and time was significant for LD and AOC suggestingthat for each moisture level, the cultivation time for pre-treatment influences the effectiveness of P. chrysosporium

to achieve greater lignin degradation and preserve carbohy-drates (P < 0.05). Interaction between salt formulation andtime did not impact LD significantly. Depending on thecultivation time, salts had a significant impact on perfor-mance and quantity of lignin removed across all moisturelevels. Differences between treatment combinations werefurther examined for LD, SR, and AOC.

3.1. Lignin degradation

The effects of the three initial pretreatment moisture lev-els and three salt concentrations on percent lignin degradedover time are presented in Fig. 1. There was a lag period(0–4 days) before significant lignin degradation wasobserved (P < 0.05). Beyond day 4, lignin was continuouslydegraded following linear trends at an average rate of 2.2%per day for all treatments. For 65% M.C., lignin degrada-tion rates slowed down after day 10 and the final lignindegradation on day 14 was 19.8% on average for all saltformulations (Fig. 1a).

Pretreatments with 75% and 80% M.C. (Figs. 1b and c)showed similar trends within 10 days but higher ligninreduction occurred after 14 day cultivation when comparedwith the 65% M.C. treatments. A noticeable difference wasthat after day 10, pretreatments with 75% and 80% M.C.continued to degrade lignin whereas those at 65% moisturetreatment slowed down. For all three M.C.s, longer pre-treatment time (14 day) resulted in significantly higher lig-nin degradation beyond 10 days. Comparisons of moisturecontent and salt formulation combinations for the 14 dayculture period evaluated using Tukey’s studentized range

Table 1ANOVA of main effects and interactions of initial moisture content (M.C.), ssolids recovery (SR) and carbohydrate availability (AOC) over the 14 day per

Source D.F. Lignin degradation Sol

MSq F value P-value MS

M.C. 2 132.3 20.78 <.0001 11Salt 2 20.2 3.18 0.0456Time 5 1574.2 247.23 <.0001 268M.C. � salt 4 24.2 3.80 0.0063M.C. � time 10 36.7 5.76 <.0001 1Salt � time 10 9.3 1.46 0.1648M.C. � salt � time 20 5.7 0.89 0.5962

M.C. is moisture content; D.F. is degrees of freedom; MSq is mean square.

(HSD) are presented in Table 2. The lignin degradationobserved in the 75% and 80% M.C. pretreatments for allsalt formulations was significantly higher than the 65%M.C. pretreatment (P < 0.05).

Although the main effects of salts are significant for alldata, the addition of salts did not enhance the lignin degra-dation in cotton stalks by P. chrysosporium in solid statecultivation after 14 days’ incubation (Fig. 1). The treat-ments with salts showed similar performance as those with-out salts over time and across all moisture contents. Asshown in Table 2, the 65% moisture pretreatment withmodified NREL salts resulted in 18.4% lignin degradation,which was not significantly different from the treatmentwith modified NREL salts + Mn2+ (P < 0.05). Treatmentswithout any supplemental salts degraded 20.0% lignin atthe end of day 14, which was statistically the same as theother salt treatments. The results observed for the 75%and 80% M.C.s were similar; indicating that extent of lig-nin degradation in cotton stalks does not require saltsupplementation.

3.2. Solids recovery and carbohydrates availability

The solid recovery and availability of carbohydratesafter 14 days pretreatment for the initial moisture contentand salt concentration combinations are shown in Table2. It was shown that initial moisture content and salt sup-plementation may lead to significant differences in solidrecovery. Treatment at 75% M.C., supplemented withmodified NREL salts (SDS) gave significantly higher solidrecovery at 73.1% than the other two salt levels. However,for 65% M.C, no significant difference was observed acrossthe three salt levels. At 80% M.C., supplementing modifiedNREL salts + Mn2+ (SDM) resulted in high solid recovery(72.2%), followed by SDS (70.2%) and SD (68.8%). In gen-eral, treatment with 75% M.C. gave higher solid recoverythan the other two M.C. levels, however, on average alltreatments resulted in approximately 71.4% solid recovery.

For availability of carbohydrates, pretreatments with75% initial M.C. gave on average 42.5% carbohydratesavailability which was significantly greater than 65% and80% initial M.C. (P < 0.05). Within each moisture content,no significant difference was observed between the three

alt formulation (salt) and time factors (time) on lignin degradation (LD),iod

id recovery Availability of carbohydrates

q F value P-value MSq F value P-value

7.4 160.18 <.0001 2128.8 59.21 <.00014.1 5.53 0.0052 0.4 0.01 0.98844.3 3662.17 <.0001 1012.1 28.15 <.00018.2 11.16 <.0001 15.9 0.44 0.77745.1 20.63 <.0001 264.3 7.35 <.00010.7 0.93 0.5133 36.0 1.0 0.44831.4 1.91 0.0188 27.1 0.75 0.7611

0 2 4 6 8 10 12 140

5

10

15

20

25

30 SD65 SDS65 SDM65

a

Lign

in D

egra

datio

n (%

)

Time (days)0 2 4 6 8 10 12 14

0

5

10

15

20

25

30b

SD75 SDS75 SDM75

Lign

in D

egra

datio

n (%

)

Time (days)

0 2 4 6 8 10 12 140

5

10

15

20

25

30 SD80 SDS80 SDM80

c

Lign

in D

egra

datio

n (%

)

Time (days)

Fig. 1. Lignin degradation during the solid state cultivation at (a) 65%, (b) 75%, (c) 80% moisture content (wet basis, j no salts, SD; d modified NRELsalts, SDS; and N modified NREL salts plus Mn2+, SDM).

Table 2Statistical comparisons for lignin degradation, solid recovery and availability of carbohydrates (day 14) with three moisture contents and three saltsformulations

M.C. Percentage of lignin degradation Solid recovery Availability of carbohydrates

65% 75% 80% 65% 75% 80% 65% 75% 80%

SD a20.0y a27.6x a26.7x a72.2x b71.1x c68.8y a25.8y a41.6x a23.8y

SDS a18.4y a26.6x a27.1x a71.7y a73.1x b70.8y a30.0y a43.6x a20.0z

SDM a20.4z a27.7x a24.5y a70.8x b71.9x a72.2x a30.4y a42.4x a19.5z

Letters (a, b, c) on the left of the data represent the levels of significance within three salt formulations (among columns); letters (x, y, z) on the right of thedata represent the levels of significance within three moisture contents (among rows); P-values generated by Tukey’s studentized range (HSD) test.

6560 J. Shi et al. / Bioresource Technology 99 (2008) 6556–6564

salt levels. This indicates that the addition of salts may notbe helpful in preserving carbohydrates.

The treatments resulting in higher availability of carbo-hydrates, higher solid recovery, as well as higher lignin deg-radation could be considered preferred pretreatments. By astatistical analysis, the lignin degradation for pretreatmentswith 75% M.C. was not statistically different than 80%

M.C. treatments, and both were more effective than 65%M.C. However, 75% M.C. exceeded 80% M.C. (P < 0.05)in both solid recovery and availability of carbohydrates.Within 75% M.C., use of salts did not provide betterperformance for lignin degradation and availability of car-bohydrates. Although, supplementing modified NRELsalts resulted in slightly higher solid recovery, considering

J. Shi et al. / Bioresource Technology 99 (2008) 6556–6564 6561

cost effectiveness and performance efficiency, solid statecultivation at 75% initial M.C. without salt supplementa-tion would be a more preferable microbial pretreatmentapproach.

3.3. Enzyme activities

Cellulase activities were monitored during the 14 daycultivations (Fig. 2). Activity increased over time as a resultof lignin breakdown and release of cellulose. The 65%M.C. treatments on average produced the highest levelsof activity in comparison to the other moisture treatments(P < 0.05). However, when compared with a high titer cel-lulase producing fungus such as Trichoderma reesei whichproduced cellulase yields of 250 to 430 IU/g of cellulose

0 42 1086 12 14

0 42 6

0.0

0.2

0.4

0.6

0.8

1.0

0.0

0.2

0.4

0.6

0.8

1.0

a

Cel

lula

se A

ctiv

ities

(IU

/ml l

iqui

d in

SS

C)

Time (day)

SD65 SDS65 SDM65

c

Cel

lula

se A

ctiv

ities

(IU

/ml l

iqui

d in

SS

C)

Tim

SD80 SDS80 SDM80

Fig. 2. Cellulase activities during 14 days pretreatment. (a) with 65% initial mmodified NREL salts plus Mn2+, SDM65; (b) with 75% initial moisture contensalts plus Mn2+, SDM75; (c) with 80% initial moisture content, j: no salts SDSDM80.

(Chahal, 1985), all the pretreatment cultures maintainedlower cellulase activities, below 1.0 IU/ml liquid (or2.5 IU/g initial substrate) in SSC. Adding Mn2+ signifi-cantly increased cellulase activities for all initial moisturecontents (P < 0.05).

4. Discussion

Reducing the lignin content of biomass helps to exposethe highly-ordered crystalline structure of cellulose andfacilitates substrate access by hydrolytic enzymes (Sunand Cheng, 2002). The effectiveness of a pretreatmentmethod can be reflected by the percentage of lignin degra-dation with simultaneous conservation of cellulose poly-mers. P. chrysosporium can decompose lignin effectively

0 42 1086 12 14

108 12 14

0.0

0.2

0.4

0.6

0.8

1.0b

Cel

lula

se A

ctiv

ities

(IU

/ml l

iqui

d in

SS

C)

Time (day)

SD75 SDS75 SDM75

e (day)

oisture content, j: no salts SD65; d: modified NREL salts, SDS65; N:t, j: no salts SD75; d: modified NREL salts, SDS75; N: modified NREL80; d: modified NREL salts, SDS80; N: modified NREL salts plus Mn2+,

6562 J. Shi et al. / Bioresource Technology 99 (2008) 6556–6564

by excreting unique extra-cellular peroxidases (LiP andMnP) as secondary metabolites under nutrient starvation(Tien and Kirk, 1988). Therefore, in this microbial pre-treatment process, lignin degradation relies on fungalgrowth and metabolite formation during the 14 day solidstate cultivation on cotton stalks. There was an initial lagperiod before significant lignin degradation occurred, mostlikely because the fungus needed to produce a combinationof enzymes to facilitate degradation and use the cottonstalk as a carbon source (Hatakka and Uusi-Rauva,1983). However, between 4 and 14 days, the lignin contentdecreased fairly consistently. For the 75% and 80% mois-ture content treatments, additional culture time may havebeen beneficial to overall lignin degradation as degradationrates did not decrease. Yet the recovery of solids and car-bohydrates for hydrolysis would have been reduced.

Moisture content is significant to microbial performancein SSC systems and affects growth as well as product for-mation (Lonsane et al., 1985; Mitchell et al., 2000). Sub-strate moisture content in SSC ranges from 30% to 80%on a wet weight basis, depending upon the properties ofthe substrates and the microbial species used (Raimbault,1998). Water in SSC systems functions as a solvent fortransport of nutrients, metabolites and other solutes as wellas helps maintain stable cellular and molecular structures(Gervais and Molin, 2003). An accepted means of report-ing water relationships with solid substrates for microbialcultures is through the thermodynamic property of water,water activity (aw), which describes its availability forchemical reactions (Grajek and Gervais, 1987; Krishna,2005). However, other research groups have found thequantity and kinetic properties of free water to be moreinfluential to the diffusion of nutrients and metabolitesand overall microbial activity (Troller and Stinson, 1981;Oriol et al., 1988; Chinn et al., 2007). Moisture contentwas a key factor that affected the lignin degrading perfor-mance of P. chrysosporium. The level and rate of lignindegradation was reduced at 65% moisture content in com-parison to the higher moisture levels. Although this lowermoisture content supported active fungal growth, it maynot fully support the metabolic functions of P. chrysospori-

um. Oriol et al. (1988) showed that even when the initialmoisture content is sufficient in early stages, water contentwill gradually accumulate inside the fungal cells during cul-tivation. The residual water outside the cells is continu-ously lost, eventually limiting the water availability orwater activity of the substrate matrix. This limitation inavailable water could be one possible explanation for thehampered lignin degradation in the 65% treatments afterjust 10 days. The 75% and 80% moisture content treat-ments demonstrated nearly 28% lignin degradation after14 days, indicating that the water levels were adequate.Asgher et al. (2006) found similar results in their study ofP. chrysosporium in SSC on corn cobs with moisture con-tent ranging between 40% and 90%. They reported a para-bolic relationship between ligninase activity and moisturecontent, with 70% as the optimum level. It may be possible

to increase the initial moisture content of the cotton stalk(within its water holding capacity) to improve perfor-mance, however higher moisture contents limit oxygentransfer, often inhibiting aerobic SSC cultures, and increasesusceptibility to bacterial contamination (Ramesh andLonsane, 1990; Asgher et al., 2006). For P. chrysosporium,it is important to provide environmental conditions thatsupport diffusion of oxygen into the liquid and solid phasesof the SSC system for increased ligninolytic activity (Leiso-la et al., 1983; Kerem et al., 1992).

In addition to initial moisture content, supplementingnutrients is another key factor that affects fungal growthand metabolic activities in SSC, especially in cases wherethe microorganism requires certain nutrients crucial to sup-port metabolism (Chinn et al., 2007; Kachlishvili et al.,2006). Most studies suggest that addition of nutrients, espe-cially Mn2+ ions, can enhance the performance of P. chry-

sosporium in pure artificial media by stimulatingproduction of lignin-degrading enzymes responsible forthe lignin break-down processes (Kirk et al., 1978; Brownet al., 1990). In this study, at 65% moisture content, therewas an initial stimulating effect of Mn2+ during the first10 days of the incubation, which was in accord with thehighly significant salt main effect and M.C. and salt inter-action for lignin degradation in ANOVA (Table 1). Never-theless, results in this study indicate that supplementingnutrients such as NREL salts and Mn2+ ions does notenhance lignin degradation in solid state cultivation on cot-ton stalks after 14 days’ pretreatment. Considering the useof cotton stalk, instead of pure lignin, as a carbon source,the natural composition of inorganic salts in the stalks maybe sufficient to meet critical nutrient requirements of P.

chrysosporium (Kerem et al., 1992; Nieto-Fernandezet al., 1998). The supplementation of extra nutrient wasnot necessary. The effectiveness of SSC pretreatments,without salt addition, on fungal growth and lignin degra-dation supports the recognized advantages of SSC systemsin using solid substrates that provide necessary nutrientsand offer simplified medium preparation (Pandey et al.,2000). These results demonstrate the feasibility and poten-tial of SSC microbial pretreatment to reduce input costs.

Besides delignification, solids recovery and availabilityof carbohydrates are important criteria for evaluating pre-treatment performance. Pretreatments which result inhigher solids recovery are expected to provide more cellu-lose during the hydrolysis step. Although lignin degrada-tion is critical, higher cellulose content in recovered solidseventually provides higher availability of carbohydratesfor ethanol fermentation. A higher selectivity, ratio of lig-nin degradation to cellulose reduction (change in glucancontent after pretreatment), implies increased delignifica-tion with preservation of cellulose (Camarero et al., 1994;Keller et al., 2003). It was anticipated that inorganic saltsupplements, especially Mn2+, would improve ligninaseproduction resulting in selective lignin degradation.However, the addition of salts in this study neitherimproved the delignification efficiency nor the preservation

J. Shi et al. / Bioresource Technology 99 (2008) 6556–6564 6563

of cellulose. This may be related to the composition of thecotton stalks and their ability to provide sufficient nutrientsfor meeting the requirements of P. chrysosporium (Keremet al., 1992). In this study, M.C. levels had the greatesteffect on maintaining the balance between high solidsrecovery, availability of carbohydrates and lignin degrada-tion. It appeared that 80% M.C. was a poor choice forAOC such that it resulted in less than 20% carbohydrates,followed by 65% M.C. This was most likely related to theinfluence of water content on fungal metabolism of cottonstalks. The moisture content levels of 65% and 80% maynot be optimal for ligninolytic enzyme production and lig-nin degradation, but are more supportive of fungal growthand sugar metabolism (Lonsane et al., 1985). It is suggestedthat 75% M.C. was still the best overall choice, whichachieved a selectivity of 0.82 (data not shown) comparableto previous reports (Blanchette, 1991; Sawada et al., 1995).Higher selectivity for lignin degradation can be achieved bychoosing different fungal strains such as P. ostreatus and C.

subvermispora or genetically modifying P. chrysosporium toproduce less cellulase activity (Kirk et al., 1986; Keremet al., 1992). However, the use of these alternative strainsgenerally requires longer pretreatment time (Taniguchiet al., 2005).

Higher cellulase activities typically improve the abilityof the fungus to degrade cellulose to available sugars. Itcan also lead to higher cellulose consumption, which lowersoverall availability of carbohydrates. Our results showedhighest cellulase activity at the end of 14 day pretreatment,which may indicate that more cellulose was accessible tothe fungus through decomposition of the lignin structure.Another interesting phenomenon was the higher cellulaseactivity associated with Mn2+ addition. Although the ini-tial purpose of supplementing Mn2+ was to improve lignin-ase activities, thus increasing lignin degradation. Besidesincreasing cellulase activity, Mn2+ supplementation is nota likely consideration for microbial pretreatment sinceMn2+ addition did not significantly improve process per-formance in lignin degradation by SSC.

In conclusion, initial moisture content significantlyaffected lignin degradation, with 75% and 80% M.C., pro-viding better results than 65% M.C. and longer pretreat-ment time (up to 14 days) resulted in higher lignindegradation. Within the same initial moisture content, sup-plementing with nutrients did not enhance lignin degrada-tion. Considering cost, solid state cultivation (14 days) with75% initial moisture content without salt supplements wasthe most preferable pretreatment. Microbial pretreatmentby solid state cultivation of P. chrysosporium was able todegrade lignin in cotton stalks and has the potential tobe an energy-saving, low cost, simple, and environmentfriendly approach which can extensively reduce the severityof chemical pretreatments. However, further investigationson delignification kinetics and hydrolysis of the pretreatedmaterials are needed for process improvement. There isalso a need to prevent further metabolism of the solid sub-strate by our ligninolytic fungus and potential colonization

by cellulolytic microorganisms in between pretreatmentand saccharification processing steps. On a large scale,microbial pretreatment can parallel processing techniquesused in mushroom production to maintain cellulose avail-ability. One practice that could be useful is peak heatingor bulk pasteurization once pretreatment is complete.Challenges with scale up still need to be examined to fullyassess the utility of microbial pretreatment in lignocellu-losic conversion processes.

Acknowledgements

The authors would like to thank Dr. Michael D. Boyette(Dept of Biological and Agricultural Engineering, NCSU)and Dr. Ralph A. Dean (Center for Integrated Fungal Re-search, NCSU) for their guidance and useful discussions.

References

Adney, B., Baker, J., 1996. Measurement of cellulase activities. LAP-006NREL Analytical Procedure, National Renewable Energy Laboratory,Golden, CO.

Akhtar, M., Scott, G.M., Swaney, R.E., Kirk, T.K., 1998. Overview ofbio-mechanical and biochemical pulping research. In: Enzyme Appli-cation in Fiber Proceeding. American Chemical Society, John Wiley &Sons, Inc, New York, NY, pp. 2–16, Chapter 2.

Asgher, M., Asad, M.J., Legge, R.L., 2006. Enhanced lignin peroxidasesynthesis by Phanerochaete chrysosporium in solid state bioprocessingof a lignocellulosic substrate. W. J. Microbiol. Biotechnol. 22, 449–453.

Blanchette, R.A., 1991. Delignification by wood-decay fungi. Annu. Rev.Phytopathol. 29, 381–398.

Brown, J.A., Glenn, J.K., Gold, M.H., 1990. Manganese regulatesexpression of manganese peroxidase by Phanerochaete chrysosporium.J. Bacteriol. 172, 3125–3130.

Camarero, S., Galletti, G.C., Martinez, A.T., 1994. Preferential degrada-tion of phenolic lignin units by two white rot fungi. Appl. Environ.Microbiol. 60, 4509–4516.

Chahal, D.S., 1985. Solid-state fermentation with Trichoderma reesei forcellulase production. Appl. Environ. Microbiol. 49, 205–210.

Chinn, M.S., 2003. Solid substrate cultivation of anaerobic thermophilicbacteria for the production of cellulolytic enzymes. Ph.D. dissertation.Lexington, KY: University of Kentucky, Department Biosystems andAgricultural Engineering.

Chinn, M.S., Nokes, S.E., Strobel, H.J., 2007. Influence of processconditions on end product formation from Clostridium thermocellum

27405 in solid substrate cultivation on paper pulp sludge. Bioresour.Technol. 98, 2184–2193.

Couto, S.R., Sanroman, M.A., 2005. Application of solid-state fermen-tation to ligninolytic enzyme production. J. Biochem. Eng. 22, 211–219.

Datta, A., Bettermann, A., Kirk, T.K., 1991. Identification of a specificmanganese peroxidase among ligninolytic enzymes secreted by Phan-

erochaete chrysosporium during wood decay. Appl. Environ. Microb.57, 1453–1460.

Ehrman, T., 1994. Method for determination of total solids in biomass. In:Laboratory Analytical Procedures No. 001. National RenewableEnergy Laboratory, Golden, CO.

Ehrman, T., 1996. Method for determination of acid-soluble lignin inbiomass. In: Laboratory Analytical Procedures No. 004. NationalRenewable Energy Laboratory Golden, CO.

Ganesh, K.A., Sekaran, G., Krishnamoorthy, S., 2006. Solid statefermentation of Achras zapota lignocellulose by Phanerochaete chry-

sosporium. Bioresour. Technol. 97, 1521–1528.

6564 J. Shi et al. / Bioresource Technology 99 (2008) 6556–6564

Gervais, P., Molin, P., 2003. The role of water in solid-state fermentation.J. Biochem. Eng. 13, 85–101.

Grajek, W., Gervais, P., 1987. Influence of water activity on the enzymebiosynthesis and enzyme activities produced by Trichoderma viride TSin solid-state fermentation. Enzyme Microbiol. Technol. 9, 658–662.

Hatakka, A.I., 1983. Pretreatment of wheat straw by white-rot fungi forenzymatic saccharification of cellulose. Eur. J. Appl. Microbiol.Biotechnol. 18, 350–357.

Hatakka, A.I., Uusi-Rauva, A.K., 1983. Degradation of 14C labeledpoplar wood lignin by selected white-rot fungi. Eur. J. Appl.Microbiol. Biotechnol. 17, 235–242.

Hatakka, A., Maijala, P., Hakala, T., Hauhio, L., Ellmen, J., 2003. A newwhite-rot fungus and its use in the pretreatment of wood. Finnishpatent, FI 112248.

Hadar, Y., Kerem, Z., Gorodecki, B., 1993. Biodegradation of lignocel-lulosic agricultural waste by Pleurotus ostreatus. J. Biotechnol. 30,133–139.

Han, J., Rowell, J., 1997. Chemical composition of fibers. In: Rowell, R.,Young, R., Rowell, J. (Eds.), Paper Composites from Agro-BasedResources. CRC Lewis Publisher, New York, pp. 3–134.

Jager, A., Croan, S., Kirk, T.K., 1985. Production of ligninases anddegradation of lignin in agitated submerged cultures of Phanerochaete

chrysosporium. Appl. Environ. Microbiol. 50, 1274–1278.Kachlishvili, E., Penninckx, M.J., Tsiklauri, N., 2006. Effect of nitrogen

source on lignocellulolytic enzyme production by white-rot basidio-mycetes under solid-state cultivation. W. J. Microbiol. Biotechnol. 22,391–397.

Keller, F.A., Hamilton, J.E., Nguyen, Q.A., 2003. Microbial pretreatmentof biomass potential for reducing severity of thermo-chemical biomasspretreatment. Appl. Biochem. Biotechnol. 105, 27–41.

Kerem, Z., Friesem, D., Hadar, Y., 1992. Lignocellulose degradationduring solid-state fermentation: Pleurotus ostreatus versus Phanero-

chaete chrysosporium. Appl. Environ. Microbiol. 58, 1121–1127.Kirk, T.K., Schultz, E., Connors, W.J., Zeikus, J.G., 1978. Influence of

culture parameters on lignin metabolism by Phanerochaete chrysospo-

rium. Arch. Microbiol. 117, 277–285.Kirk, T.K., Tien, M., Johnsrud, S.C., Eriksson, K.E., 1986. Lignin

degrading activity of Phanerochaete chrysosporium Burds: Comparisonof cellulase-negative and other strains. Enzyme Microb. Technol. 8,75–80.

Krishna, C., 2005. Solid-state fermentation systems – an overview. Crit.Rev. Biotechnol. 25, 1–30.

Lee, J., 1997. Biological conversion of lignocellulosic biomass to ethanol.J. Biotechnol. 56, 1–24.

Leisola, M., Ulmer, D., Fiechter, A., 1983. Problem of oxygen transferduring degradation of lignin by Phanerochaete chrysosporium. Eur. J.Appl. Microbiol. Biotechnol. 17, 113–116.

Lonsane, B.K., Ghildyal, N.P., Budiatman, S., Ramakrishna, S.V., 1985.Engineering aspects of solid state fermentation. Enzyme Microb.Technol. 7, 258–265.

Mitchell, D.A., Berovic, M., Krieger, N., 2000. Biochemical engineeringaspects of solid state bioprocessing. Adv. Biochem. Eng. Biotechnol.68, 61–138.

Miura, D., Tanaka, H., Wariishi, H., 2004. Metabolomic differentialdisplay analysis of the white-rot basidiomycete Phanerochaete chry-

sosporium grown under air and 100% oxygen. FEMS Microbiol. Lett.234, 111–116.

Moiser, N., Wyman, C., Dale, B., Elander, R., Lee, Y.Y., Holtzapple, M.,Ladisch, M., 2005. Features of promising technologies for pretreat-ment of lignocellulosic biomass. Bioresour. Technol. 96, 673–686.

NCGA (National Corn Growers Association), 2005. Ethanol & publicpolicy, Available through: http://www.ncga.com/ethanol/main/provi-sions.htm, Accessed on Feb 22nd, 2006.

Nieto-Fernandez, F.E., Lee, J.J., Koestler, R.J., 1998. Accessing biode-terioration in wood using ATP photometry Part III: Estimation of thefungal biomass of Phanerochaete chrysosporium in decayed wood usingATP and energy charge measurements. Int. Biodeterior. Biodeg. 41,35–39.

NREL (National Renewable Energy Laboratories), 2000. Technicalreport: Determining the cost of producing ethanol from corn starchand lignocellulosic feedstocks, NREL/TP-580-28893.

Oriol, E., Raimbault, M., Roussos, S., Viniegra-Gonzales, G., 1988.Water and water activity in the solid-state fermentation of cassavastarch by Aspergillus niger. Appl. Microbiol. Biotechnol. 27, 498–503.

ORNL (Oak Ridge National Laboratory), 2005. Biomass as feedstock fora bioenergy and bioproducts industry: the feasibility of a billion-tonannual supply, Available through: http://feedstockreview.ornl.gov/pdf/billion_ton_vision.pdf.

Pandey, A., 1992. Recent process developments in solid-state fermenta-tion. Process Biochem. 27, 109–117.

Pandey, A., Carlos, R.S., David, M., 2000. New developments in solidstate fermentation: I-bioprocesses and products. Process Biochem. 35,1153–1169.

Raimbault, M., 1998. General and microbiological aspects of solidsubstrate fermentation. Electro. J. Biotechnol. 1, 174–188.

Ramesh, M.V., Lonsane, B.K., 1990. Critical importance of moisturecontent of the medium for alpha-amylase production by Bacillus

licheniformis M27 in solid state fermentation system. Appl. Microbiol.Biotechnol. 33, 501–505.

Reddy, C.A., D’Souza, T.M., 1994. Physiology and molecular biology ofthe lignin peroxidases of Phanerochaete chrysosporium. FEMS Micro-biol. Rev. 13, 137–152.

Silverstein, R.A., Chen, Y., Sharma-Shivappa, R.R., Boyette, M.D.,Osborne, J., 2007. A comparison of chemical pretreatment methods forimproving saccharification of cotton stalks. Bioresour. Technol. 98,3000–3011.

Sun, Y., Cheng, J., 2002. Hydrolysis of lignocellulosic materials forethanol production: a review. Bioresour. Technol. 83, 1–11.

Sawada, T., Nakamura, Y., Kobayashi, F., Kuwahara, M., Watanabe, T.,1995. Effects of fungal pretreatment and steam explosion on enzymaticsaccharification of plant biomass. Biotechnol. Bioeng. 48, 719–724.

Taniguchi, M., Suzuki, H., Watanabe, D., Sakai, K., Hoshino, K.,Tanaka, T., 2005. Evaluation of pretreatment with Pleurotus ostreatus

for enzymatic hydrolysis of rice straw. J. Biosci. Bioeng. 100, 637–643.Templeton, D., Ehrman, T., 1994. Determination of acid-insoluble lignin

in biomass. In: Laboratory Analytical Procedures No. 003. NationalRenewable Energy Laboratory Golden, CO.

Tengerdy, R.P., Szakacs, G., 2003. Bioconversion of lignocellulose in solidsubstrate fermentation. J. Biochem. Eng. 13, 169–179.

Tien, M., Kirk, T.K., 1988. Lignin peroxidase of Phanerochaete chrysos-

porium. Methods Enzymol. 161, 238–249.Troller, J.A., Stinson, J.V., 1981. Moisture requirements for growth and

metabolite production by lactic acid bacteria. Appl. Environ. Microb.42, 682–687.